Issuu on Google+

Empowered lives. Resilient nations.

Energy and Communal Services in Kyrgyzstan and Tajikistan: A Poverty and Social Impact Assessment

Editor: Ben Slay

UNDP Bratislava Regional Centre 2011

This publication, and the research on which it is based, has been sponsored by UNDP’s Regional Bureau for Europe and CIS, and by the UNEP-UNDP Poverty and Environment Initiative. The editor would like to thank Bryan Schell and Zuzana Aschenbrennerova for their editorial assistance, and Balázs Horváth for helpful comments. The views expressed here are not necessarily those of UNDP, the United Nations, or its Member States. UNDP, and the authors of this study, would like to thank the experts, public officials, and citizens of Tajikistan and Kyrgyzstan, without whose support, insights, and expertise this publication would not have been possible.


Table of Contents List of tables, figures, and boxes ………………………………………………………… List of acronyms and abbreviations …………………………………………………….

3 10

Chapter 1: Introduction and executive summary ……………………………………...

13

Chapter 2: Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan ………………………………………………………………

23

Executive summary ……………………………………………………………………….. Recent macroeconomic, socio-economic, and energy trends …………………………… Recent research on poverty, energy, and vulnerability in Kyrgyzstan and Tajikistan ……. Energy tariffs and social policy in Kyrgyzstan in 2010: What can happen when things go wrong ……………………………………………………………………………………… Conclusion …………………………………………………………………………………

24 26 31

Chapter 3: Kyrgyzstan’s Energy Sector ………………………………………………..

56

Executive summary ……………………………………………………………………….. Sectoral overview …………………………………………………………………………. Poverty and household access to energy ………………………………………………….. Social protection and the energy sector …………………………………………………… Conclusions and recommendations ………………………………………………………..

57 59 94 102 111

Chapter 4: Tajikistan’s Energy Sector ………………………………………………….

116

Executive summary ……………………………………………………………………….. Introduction ……………………………………………………………………………….. Sectoral overview …………………………………………………………………………. The energy/poverty nexus ………………………………………………………………… Poverty alleviation and energy sector development ……………………………………… Social assistance to mitigate the impact of rising energy prices ………………………….. Conclusions and recommendations ……………………………………………………….. Appendices ………………………………………………………………………………...

117 120 121 156 168 174 192 196

Chapter 5: Water and Communal Services in Kyrgyzstan ……………………………

202

Executive summary ……………………………………………………………………….. Sectoral overview …………………………………………………………………………. Poverty trends ……………………………………………………………………………... Social policy issues ……………………………………………………………………….. Conclusions ……………………………………………………………………………….. Appendices ………………………………………………………………………………...

203 206 229 241 245 247

Chapter 6: Water and Communal Services in Tajikistan ……………………………..

260

Executive summary ……………………………………………………………………….. Introduction and overview ………………………………………………………………...

261 263

50 55

1


Poverty and Social Impact Assessment

2

Sectoral analysis …………………………………………………………………………... Poverty trends ……………………………………………………………………………... Focus group discussions …………………………………………………………………... Social protection and social policy issues ………………………………………………… Lessons learned and recommendations …………………………………………………… Appendices ………………………………………………………………………………...

264 275 278 281 282 287

Chapter 7: Conclusions …………………………………………………………………...

290

Reflections on responses ………………………………………………………………….. Programming, policy, and research recommendations ……………………………………

291 293

Bibliography ……………………………………………………………………………...

296


List of tables, figures, and boxes Figure 1.1—Energy consumption trends in Tajikistan and Kyrgyzstan (2007-2010) ……… Figure 1.2—Trends in household energy, consumer prices in Kyrgyzstan and Tajikistan (2008-2010) …………………………………………………………………………………. Figure 1.3—Household water tariff trends in Kyrgyzstan (2007-2010; June 2007 = 100) … Figure 1.4—Declining household access to electricity in Tajikistan (2007-2009) …………. Figure 1.5—Shares of households reporting electricity cut-offs (and their frequency) in Kyrgyzstan ………………………………………………………………………………...... Figure 1.6—Shares of household spending on various energy sources in Kyrgyzstan (2009) Figure 1.7—Shares of households reporting use of various energy sources in Tajikistan (2007-2009) …………………………………………………………………………………. Figure 1.8—Shares of household spending devoted to energy in Kyrgyzstan ……………... Figure 1.9—Shares household spending devoted to energy in Tajikistan (2009) ………….. Table 2.1—Macroeconomic trends (2008-2010) …………………………………………… Figure 2.1—Monthly water volume trends at Central Asia’s largest hydropower stations (2009-2011) …………………………………………………………………………………. Map 2.1—Central Asia’s electricity transmission grid ……………………………………... Figure 2.2—Impact of the dissolution of the integrated Central Asian electricity transmission network on Tajikistan’s electricity exports and imports (2009-2010) ………... Figures 2.3, 2.4—Income poverty rates in Tajikistan, Kyrgyzstan, other countries (19992005) ………………………………………………………………………………………… Figure 2.5—Remittance inflows as a share of GDP (2005-2010) ………………………….. Figure 2.6—Share of population without access to improved sanitation services .................. Figure 2.7—Share of population without access to improved water sources ......................... Figure 2.8—Maternal, under-five mortality rates (2005) …………………………………... Table 2.2—Poverty trends in Tajikistan (2007-2009) ……………………………………… Table 2.3—Regional income poverty rates in Tajikistan (2007-2009) ……………………... Table 2.4—Trends in absolute poverty rates in Kyrgyzstan (2006-2009) ………………….. Table 2.5—Fiscal, social policy trends in Kyrgyzstan, Tajikistan (2008-2011) ……………. Figure 2.9—Kyrgyzstan: Increases in external migration, remittance inflows (2010) ……... Table 2.6—Kyrgyzstan’s social protection instruments: Efficiency and effectiveness (2005 data) …………………………………………………………………………………………. Table 2.7—Social policy instruments in Kyrgyzstan (2007-2009) ………………………… Table 2.8—Electricity tariff trends in Kyrgyzstan (per kWh, 2009-2010) …………………. Table 2.9—District heating tariff trends in Kyrgyzstan (per gcal, 2009-2010) …………….. Figure 2.10—Distribution of social benefits in Kyrgyzstan, by household deciles (20062010) ………………………………………………………………………………………… Figure 2.11—Distribution of pension benefits in Kyrgyzstan, by household deciles (20082010) ………………………………………………………………………………………… Figure 2.12—Distribution of categorical benefits in Kyrgyzstan, by household deciles (20082010) ………………………………………………………………………………….

16 16 17 18 18 19 20 20 21 27 28 29 30 31 32 33 33 33 35 36 36 36 39 44 45 51 51 51 52 52 3


Poverty and Social Impact Assessment

Table 2.10—Kyrgyzstan: Trends in energy, water production and supply (2007-2010) …... Table 2.11—Kyrgyzstan: 2010 macroeconomic trends (preliminary data) ………………… Figure 3.1—Effective household electricity tariffs in the former Soviet republics (2007) … Figure 3.2—Kyrgyzstan: Trends in energy production, consumption (2007-2010) ………... Figure 3.3—Kyrgyzstan: Household energy price inflation trends (2007-2010) …………... Box 3.1—Kyrgyzstan: Energy sector privatization chronology ……………………………. Figure 3.4—Kyrgyzstan: Electricity production, consumption, losses, and exports (in million kWh, 2005-2010) …………………………………………………………………… Table 3.1—Kyrgyzstan: End uses of electricity generated (2005-2010) …………………… Figure 3.5—Kyrgyzstan: Financial results for power generation, distribution companies (in million som, 2006-2009) ……………………………………………………………………. Figure 3.6—Kyrgyzstan: Collection rates in the electric power sector (2007-2009) ………. Figure 3.7—Kyrgyzstan: Electricity sector quasi-fiscal deficit (2002-2009) ………………. Figure 3.8—Trends in electricity generation, and in water volumes at the Toktogul hydropower reservoir (2008-2010) …………………………………………………………. Map 3.1—Existing and planned hydropower plants and high-voltage transmission lines …. Figure 3.9—Kyrgyzstan: Thermal power production, consumption, losses, and exports (in thousand gigacalories, 2006-2010) …………………………………………………………. Figure 3.10—Kyrgyzstan: Financial results for thermal power companies (in million som, 2006-2009) ………………………………………………………………………………….. Figure 3.11—Kyrgyzstan: Gas supply, consumption, and losses (in thousand meters3, 20062010) ………………………………………………………………………………….. Figure 3.12—Kyrgyzstan: Gas sector financials ……………………………………………. Figure 3.13—Kyrgyzstan: Gas, consumer price inflation trends (2007-2010) ……………... Figure 3.14—Kyrgyzstan: Fixed assets (by book value) in the gas sector (2006-2009, in million som) ………………………………………………………………………………… Figure 3.15—Kyrgyzstan: Trends in fixed asset depreciation, cash collections (2006-2009) Figure 3.16—Kyrgyzstan: Coal production, consumption, imports and exports (in thousand tons, 2006-2010) …………………………………………………………………………….. Figure 3.17—Kyrgyzstan: Financial results of coal companies (in million som, 2006-2009) Figure 3.18—Kyrgyzstan: Additional costs (above current tariff levels) per kWh associated with electricity generated from decentralized renewables ………………………………….. Table 3.2—Kyrgyzstan: Differences between planned and actual costs of larger power companies (2009-2010) ……………………………………………………………………... Figure 3.19—Kyrgyzstan: Actual versus planned household electricity tariffs (per kWh, 2006-2012) ………………………………………………………………………………….. Figure 3.20—Kyrgyzstan: Cost share trends in the electric power sector (2006-2009) ……. Figure 3.21—Kyrgyzstan:Actual versus planned household thermal power tariffs (per gigacal., 2007-2012) ………………………………………………………………………… Figure 3.22—Kyrgyzstan: Cost share trends in the thermal power sector (2006-2009) ……. Figure 3.23—Kyrgyzstan: Shares of material costs in the thermal power sector (2006-2009) ………………………………………………………………………………………… Table 3.3—Kyrgyzstan: Monthly per-capita household expenditures (in som, 2006-2010) .. 4

53 54 59 60 60 61 63 64 64 65 65 66 68 70 71 72 73 73 74 74 75 76 78 80 81 81 82 82 83 96 96


Figure 3.24—Kyrgyzstan: Trends in household expenditures, energy consumption* (20072009) ………………………………………………………………………………………. Figure 3.25—Kyrgyzstan: Share of household spending absorbed by energy expenditures (2006-2010) …………………………………………………………………………………. Figure 3.26—Kyrgyzstan: Household energy expenditures by energy sources (2006-2010) Figure 3.27—Kyrgyzstan: Household expenditures on energy, by deciles (2006-2010) …... Figure 3.28— Kyrgyzstan: Shares of total household energy expenditures devoted to various energy sources (by household decile, 2009) ………………………………………... Figure 3.29—Kyrgyzstan: Shares of total household energy expenditures devoted to various energy sources (by household location, 2009) ……………………………………... Figure 3.30—Kyrgyzstan: Share of households reporting interruptions in electricity service (2006-2009) …………………………………………………………………………………. Figure 3.31—Kyrgyzstan: Share of households (by income decile) experiencing weekly (or more frequent) interruptions in electricity service (2008-2009) ……………………………. Figure 3.32—Kyrgyzstan: Share of households (by location) experiencing weekly (or more frequent) interruptions in electricity service (2008-2009) ………………………………….. Figure 3.32—Kyrgyzstan: Share of households (by location) experiencing weekly (or more frequent) interruptions in electricity service (2008-2009) ………………………………...... Table 3.4—Kyrgyzstan: Average per-capita monthly energy expenditures, by decile group (in som) ……………………………………………………………………………………... Figure 3.33—Kyrgyzstan: Ratio of monthly pension, social assistance benefits to the national monthly subsistence minimum (2007-2009) ………………………………………. Figure 3.34—Kyrgyzstan: Trends in the distribution of social benefits by household deciles (2006-2010) …………………………………………………………………………. Figure 3.35—Kyrgyzstan: Trends in the distribution of pension benefits by household deciles (2008-2010) …………………………………………………………………………. Figure 3.36—Kyrgyzstan: Trends in the distribution of categorical benefits by household deciles (2008-2010) …………………………………………………………………………. Figure 3.37—Kyrgyzstan: Trends in the distribution of PFMB benefits by household deciles (2008-2010) …………………………………………………………………………. Figure 3.38—Kyrgyzstan: Trends in the distribution of MSB benefits by household deciles (2008-2010) …………………………………………………………………………………. Figure 4.1—Tajikistan: Annual increase in household tariffs ……………………………… Figure 4.2—Tajikistan: Total primary energy supply ………………………………………. Table 4.1—Tajikistan: Energy system production capacity (in megawatts) ……………….. Table 4.2—Tajikistan: Barqi Tojik’s hydropower plants …………………………………... Table 4.3—Tajikistan’s thermal power plants ……………………………………………… Figure 4.3—Tajikistan: Investments in the energy sector (million US$) …………………... Figure 4.4—Tajikistan: Monthly electricity imports and exports …………………………... Table 4.4—Tajikistan: Electricity balance (million kWh) ………………………………….. Figure 4.5—Tajikistan: Shares of main electricity consumption categories (%) …………... Figure 4.6—Tajikistan: Electricity consumption (million kWh) …………………………… Table 4.5—Tajikistan: Total customer payments of Barqi Tojik …………………………...

97 97 97 98 98 99 99 100 100 100 100 103 105 106 106 107 107 120 122 123 126 126 127 128 128 129 129 130 5


Poverty and Social Impact Assessment

Table 4.6—Tajikistan: Electricity tariffs (US¢/kWh without VAT, unless otherwise noted) Table 4.7—Tajikistan: Average electricity tariffs (without VAT) …………………………. Table 4.8—Tajikistan: Planned tariff increase for households (with VAT) ………………... Figure 4.7—Tajikistan: Pamir Energy electricity supply (hours) and losses (%) …………... Table 4.9—Tajikistan: Pamir Energy summary statistics (2010) …………………………... Figure 4.8—Tajikistan: Pamir Energy generation and sales ………………………………... Table 4.10—Tajikistan: Pamir Energy electricity tariffs (US¢/kWh) ……………………… Figure 4.9—Tajikistan: Committed and disbursed (actual) investments in large Barqi Tojik energy projects (million US$) ………………………………………………………………. Table 4.11—Tajikistan's gas balance ……………………………………………………….. Figure 4.10—Tajikistan: Gas consumption by sectors, million m3 …………………………. Table 4.12—Tajikistan: Purchase price of natural gas from Uzbekistan, US$ for 1000 m3 Table 4.13—Tajikistan: Gas transit losses ………………………………………………….. Table 4.14—Tajikistan: Coal production (thousand tons) ………………………………….. Figure 4.11—Retail prices for hard coal in Tajikistan compared to the EU ($/ton) ………... Figure 4.12—Tajikistan: Communal service tariffs and wholesale gas price trends, (2007 = 100) .......................................................................................................................................... Figure 4.13—Tajikistan: Household heating tariffs (Somoni per gCal) ……………………. Table 4.15—Tajikistan: Small hydropower plants (1991-2010) …………………………… Table 4.16—Tajikistan: Ministry of Energy and Industry’s small hydropower plant classification scheme ………………………………………………………………………... Table 4.17—Tajikistan: Costs and benefits of scaling up small hydropower plants for 100,000 vulnerable households ............................................................................................... Box 4.1—Tajikistan: A national trust fund for decentralized renewables and energy efficiency ……………………………………………………………………………………. Figure 4.14—Tajikistan: Share of households reporting power cuts: winter 2007 (%) ……. Figure 4.15—Tajikistan: Share of households reporting power cuts: summer 2007 (%) …... Figure 4.16—Tajikistan: Share of households using electricity, urban vs. rural (%) ............. Figure 4.17—Tajikistan: Average monthly electricity consumption per household, urban vs. rural (kWh) ……………………………………………………………………………… Table 4.18—Tajikistan: Household electricity consumption (2006-2008) ………………..... Figure 4.18—Tajikistan: Share of households using electricity, by quintile (%) ................... Figure 4.19—Tajikistan: Average monthly consumption of electricity per household, by quintile (kWh) ………………………………………………………………………………. Figure 4.20—Tajikistan: Share of households using electricity, by household size (%) ........ Figure 4.21—Tajikistan: Average monthly consumption of electricity per household, by household size (kWh) ……………………………………………………………………… Figure 4.22—Tajikistan: Average monthly electricity consumption, by gender of household head (kWh) ……………………………………………………………………… Figure 4.23—Tajikistan: Share of households using electricity, by gender of household head (%) .................................................................................................................................. Figure 4.24—Tajikistan: Tajikistan: Share of households using electricity by nationality (%) ........................................................................................................................................... 6

130 131 131 132 132 133 133 136 141 141 141 142 143 143 145 145 147 149 150 154 156 156 157 157 158 158 158 159 159 159 159 160


Figure 4.25—Tajikistan: Average monthly electricity consumption per household by nationality (kWh) …………………………………………………………………………… Figure 4.26—Tajikistan: Gas pipeline users by location (%) ………………………………. Figure 4.27—Tajikistan: Gas pipeline users: poor vs. non poor (%) ……………………….. Figure 4.28—Tajikistan: Share of LPG users: poor vs. non-poor (%) ................................... Figure 4.29—Tajikistan: Share of LPG users by location (%) ............................................... Figure 4.30—Tajikistan: Household heating sources (%) ………………………………….. Figure 4.31—Tajikistan: Share of households having and using district (central) heating by location (%) ............................................................................................................................. Figure 4.32—Tajikistan: Share of households having and using district (central) heating: poor/non poor (%) ................................................................................................................... Figure 4.33—Tajikistan: Share of households using LPG for heating, poor/non poor (%) ... Figure 4.34—Tajikistan: Share of households using LPG for heating by location (%) ……. Figure 4.35—Tajikistan: Sources of heating, by location (%) ……………………………… Figure 4.36—Tajikistan: Sources of heating, poor vs. non-poor …………………………… Table 4.19—Tajikistan: Benchmarks used in measuring affordability (% of total household income/expenditure) ………………………………………………………………………… Figure 4.37—Tajikistan: Shares of household expenditures on various energy services/sources, 2009 (%) ………………………………………………………………….. Figure 4.38—Tajikistan: Share of average annual energy consumption in total household consumption, poor vs. non-poor 2009 (%) ………………………………………………….. Figure 4.39—Tajikistan: Average annual energy consumption in total household consumption for the poorest quintile by location, 2009 (%) ………………………………... Figure 4.40—Tajikistan: Average annual energy consumption in total household consumption, by location, 2009 (%) ………………………………………………………... Figure 4.41—Tajikistan: Share of household expenditures on food, 2009 (%) …………….. Figure 4.42—Tajikistan: New debts incurred in past three months, 2008-2010 …………… Figure 4.43—Tajikistan: Top ten constraints on business investments in Tajikistan ………. Figure 4.44—Tajikistan: Potential impact of increased electricity tariffs on affordability (% of household expenditure by quintiles) ……………………………………………………... Figure 4.45—Tajikistan: Potential impact of increased electricity tariffs on average household energy affordability ……………………………………………………………… Figure 4.46—Tajikistan: Social, economic, and environmental impact of energy subsidies (UNEP) ……………………………………………………………………………………… Table 4.20—Tajikistan: Benefits and shortcomings of social assistance mechanisms ……... Table 4.21—Tajikistan: State budget allocations for the electricity and gas assistance programme ………………………………………………………………………………….. Table 4.22—Tajikistan: Number of electricity and gas assistance programme recipient households, by region ……………………………………………………………………….. Table 4.23—Tajikistan: Allowances under the social assistance program for electricity and gas …………………………………………………………………………………………… Table 4.24—Tajikistan: Income threshold to quality for compensation ……………………. Table 4.25—Tajikistan: Budget expenditures on social assistance in 2009 ………………...

160 160 160 161 161 161 162 162 162 162 163 164 164 165 166 166 166 167 167 169 171 172 173 175 176 176 177 178 179 7


Poverty and Social Impact Assessment

Figure 4.47—Tajikistan: Current system of electricity and gas compensation …………….. Figure 4.48—Tajikistan: Distribution of social assistance payments across quintiles of consumption expenditure in 2009 (%) ……………………………………………………… Table 4.26—Tajikistan: Indicator composition of Tajikistan’s proxy means testing pilot …. Table 4.27—Tajikistan: Share of electricity expenditures in total household expenditures, by poverty level, under different tariff scenarios …………………………………………… Table 4.28—Tajikistan: Impact on poverty under different tariff increases ………………... Table 4.29—Tajikistan: Subsidies under different scenarios ……………………………….. Table 4.30—Tajikistan: Tariffs at Pamir Energy and Barqi Tojik net of subsidy, for residential consumers ……………………………………………………………………….. Table 5.1—Kyrgyzstan: Revenue/cost ratios for drinking water supply in smaller cities (January 2010) ………………………………………………………………………………. Table 5.2—Kyrgyzstan: Cash flow statement for Bishkek and Osh vodokanals (in thousands of som) …………………………………………………………………………… Figure 5.1—Kyrgyzstan: Percentage of population in urban areas with access to sewerage (2008) ……………………………………………………………………………………….. Table 5.3—Kyrgyzstan: Urban sewage systems ……………………………………………. Table 5.4—Kyrgyzstan: Revenue/cost ratio for sewage service (2005 data) ………………. Table 5.5—Kyrgyzstan: Legal status of vodokanals (based on registration certificates, 2010) ………………………………………………………………………………………… Figure 5.2—Kyrgyzstan: Tariff setting for vodokanals …………………………………….. Table 5.6—Kyrgyzstan: Residential drinking water and sewage tariffs (per person per month) in urban areas ……………………………………………………………………….. Table 5.7—Kyrgyzstan: Collections rates of urban water supply and sewage tariff revenues in 2009 ………………………………………………………………………………………. Box 5.1—Kyrgyzstan: Private sector water and sanitation service delivery ……………….. Box 5.2—Public private partnerships in communal services: Lessons from Yerevan ........... Table 5.8—Kyrgyzstan: Access to communal services for poor, non-poor households (2007) ……………………………………………………………………………………….. Table 5.9—Kyrgyzstan: Water supply sources, distances, and family size (2009) ................ Table 5.10—Trips and time to fetch water in Issyk-Kul ……………………………………. Figure 5.3—Kyrgyzstan: Enteric infection incidence rates among children under 14 years (per 100,000 people) ………………………………………………………………………... Figure 5.4—Kyrgyzstan: Household water tariff trends (2007-2010; June 2007 = 100) …... Table 5.11—Kyrgyzstan: Shares of household expenditures among the three poorest deciles allocated to water and sanitation and solid waste management services by oblast, 2005-2009 …………………………………………………………………………………… Table 5.12—Kyrgyzstan: Shares of household expenditures in the poorest and the richest deciles allocated to water and sanitation expenditures by oblast, 2005-2009 ………………. Box 5.3—Kyrgyzstan: Communal service tariff increases and affordability in Karakol …... Figure 5.5—Kyrgyzstan: Willingness to pay higher water tariffs ………………………….. Figure 5.6—Kyrgyzstan: Household perceptions of drinking water quality (2008) ………... Box 5.4—Kyrgyzstan: Kyrgyzstan: NGOs and consumer protection .................................... 8

180 180 183 185 186 186 189 209 210 210 211 212 217 219 220 221 226 227 229 230 230 231 232

232 233 234 235 237 239


Table 5.13—Bishkek housing subsidies (as of December 2010) …………………………… Table 6.1—Tajikistan: Annual average inflation rates for consumer prices, communal services, 2007-2010 …………………………………………………………………………. Table 6.2—Tajikistan: Annual GDP growth rates, 2002-2011……………………………… Table 6.3—Water tariffs, Dushanbe (2010) ………………………………………………… Table 6.4—Water meterage costs (Dushanbe, December 2010) …………………………… Table 6.5—Monthly tariffs for solid waste management services (Dushanbe, 2010) ……… Table 6.6—Tajikistan: Focus group discussions ……………………………………………. Table 6.7—Tajikistan: Categories of individuals eligible for discounted communal service payments …………………………………………………………………………………….. Figure 6.1—Annual average inflation rates: Consumer prices, water tariffs (Moldova) … Figure 6.2—Trends in social benefits, communal service subsidies (Moldova) ……………

242 263 263 266 268 271 277 282 283 283

9


Poverty and Social Impact Assessment

List of acronyms and abbreviations A/O ACGRKS ADB AIDS AIMSCEM AKFED AMFI APEC AC BARQI TOJIK CAIPS CAREC CAPP CAPS CASA CASAREM CAUPS CGE CHP DHSE DC DSP&HA ECA EGI EITI ESMAP EU FDI FEC FES FESTI FAO GBAO GTZ GoT GDP GMCL GMI GPW HIV HPP HUS&E 10

The Aiyl Okmotu Automatic Commercial Gas Record Keeping System Asian Development Bank Acquired Immunodeficiency Syndrome Automated Information and Measurement System for Commercial Electricity Metering Aga Khan Foundation for Economic Development Average Monthly Per Capita Family Income Asia Pacific Economic Cooperation Alternating current Barqi Tojik Integrated Power System of Central Asia Central Asia Regional Economic Cooperation Chernobyl Atomic Power Plant Central Asia Power System Central Asia South Asia Central Asia South Asia Regional Electricity Market Central Asian United Power System Computerized General Equilibrium (Model) Combined Heat and Power (Plant) Dushanbe Heat Service Enterprise Direct current The Department of Social Protection and Humanitarian Aid Europe and Central Asia Electricity Governance Initiative Extractive Industries Transparency Initiative Energy Sector Management Assistance Program European Union Foreign Direct Investment Fuel and Energy Complex Fuel and Energy Sector Transparency Initiative in the Fuel and Energy Sector of the Kyrgyz Republic Food and Agriculture Organization (of the United Nations) Gorno Badakhshan Autonomous Oblast Deutsche Gesellschaft fuer Technische Zusammenarbeit Government of Tajikistan Gross Domestic Product Guaranteed Minimum Consumption Level Guaranteed Minimum Income Great Patriotic War Human Immunodeficiency Virus Hydro Power Plant Housing and Utility Services and Energy


HVDC HVTL IDA IDB IEA IEG IMF IDB JEA JICA JV KR kWh LPG LSGs LSS MOEI MOF MLSP MDG MLSD MOU MSBs MSEC MTEF MW MW NEP NGO OBA OJSC PESAC PFMB PSP PPP PPCR PMT PAMS PRSP PSIA QFD RAO UES RES RSPDs SBs SCBA SC SHPP SPD

A high-voltage, direct current High-voltage Transmission Line International Development Association Islamic Development Bank International Energy Association Independent Evaluation Group International Monetary Fund Islamic Development Bank Joint Economic Assessment Japanese Agency of International Cooperation Joint Venture Kyrgyz Republic Kilowatt hour Liquefied petroleum gas Local Self-Governments Living Standards Survey Ministry of Energy and Industry (Tajikistan) Ministry of Finance (Tajikistan) Ministry of Labor and Social Protection (Tajikistan) Millennium Development Goals Ministry of Level and Social Development Memorandum of Understanding Monthly Social Benefits Medical-Social Experts Commission Medium Term Expenditure Framework Megawatt Minimum Wage National Energy Program Non-Governmental Organization Output Based Aid Open joint stock company Public Enterprise Structural Adjustment Credit Monthly Benefit to Poor Families with Children Private Sector Participation Public Private Partnership Pilot Program for Climate Resilience (part of the Strategic Climate Fund (SCF), a multi-donor Trust Fund within the Climate Investment Funds) Proxy Means Targeting Project Analysis Macroeconomic Simulator Poverty Reduction Strategy Paper Poverty and Social Impact Assessment Quasi Fiscal Deficit Unified Energy System (Russian Joint stock company) Renewable energy sources Rayon Social Protection Departments State Benefits Social cost benefit analysis State Company Small hydro power plant Social Protection Bodies 11


Poverty and Social Impact Assessment

SSEA TPP TPES TALCO TJS TSPDs TWh UE UMB UNICEF UNESCO USAID USTDA USB USD UNDP UNEP VOLL WB WHO

12

The State Agency for Social Support under the KRG Thermal power plant Total Primary Energy Supply Tajik Aluminum Company Tajik Somoni (national currency, TJS) Town Social Protection Departments Terawatt-hour (1 terawatt is 10^12 watts) Utility Enterprise Unified Monthly Benefit United Nations Children’s Education Fund United Nations Educational, Scientific and Cultural Organization United States Agency for International Development United States Trade and Development Agency Unified Social Benefit United States Dollar United Nations Development Program United Nations Environmental Programme Value of Lost Load World Bank World Health Organization


Chapter 1: Introduction and executive summary

Author: Ben Slay

13


Poverty and Social Impact Assessment

Overview The importance of poverty and energy issues in Central Asia—particularly in lowincome Tajikistan and Kyrgyzstan—is acknowledged by the international community and governments. However, until 2008, these issues were often treated as separate topics. Poverty reduction tended to focus on removing barriers to sustainable economic growth, reducing income inequality, and modernizing social policies. Energy policies reflected the beliefs that universal access to electricity grids had been secured, but that reforms were needed to attract investment into the energy sector, in order to reduce losses, boost electricity exports, and extend household access to gas and central heating—particularly outside of large cities. Rising incomes and reforms were likewise seen as keys to addressing those development challenges in communal services (i.e., water, sanitation, waste management) that had not been resolved during the Soviet period. The impact of the winter energy crisis of 2007-2008 on Tajikistan, and of the drought of 2008 on Tajikistan and Kyrgyzstan, challenged these beliefs.1 These events precipitated large declines in electric power generation and consumption in both countries (Figure 1.1), significantly increasing national and household energy insecurity, particularly in rural areas. By reducing electricity supplies to water treatment and pumping facilities, as well as schools, hospitals, and other social service installations, these developments also led to cutbacks in these services, highlighting the absence of progress in the communal services sector. The socio-political turmoil that afflicted Kyrgyzstan during April-June 2010— which was driven in part by popular anger over energy tariff increases introduced in January of that year—was a further reminder of the consequences of “compound crisis”. 1

See Johannes F. Linn, “Impending Water Crisis in Central Asia: An Immediate Threat”, 2 June 2008, Brookings Institution, Washington D.C. (http://www.brookings.edu/opinions/2008/0619_central _asia_linn.aspx).

14

These developments underscored how— despite continuing economic growth— millions of people in Tajikistan and Kyrgyzstan—particularly in low-income households, as well as in isolated, mountainous communities—could be deprived of access to basic energy (as well as communal and public) services. They demonstrated the need for a deeper understanding of the links between poverty, energy, and environmental resource (particularly water) management. And they reminded governments and the international community of the importance of developing responses to the poverty/energy/vulnerability nexus that can both provide short-term emergency relief and increase long-term resilience to such threats. In preparation for the third Central Asia Regional Risk Assessment meeting held in Almaty on 14-15 April 2011,2 UNDP sponsored four poverty and social impact assessments3 of current policy and reform trends in the energy and communal service sectors in Tajikistan and Kyrgyzstan. These studies, which are based on the latest household survey (and other official) data, cast new light on household energy and water insecurities in these two countries.4 These studies show that, although some important progress has been in recovering from the compound crisis, many difficulties and challenges remain, particularly for poor households. The assessments, abridged versions of which are presented in chapters 36 of this manuscript, also raise important 2

See http://europeandcis.undp.org/senioreconomist/show/360 41D49-F203-1EE9-BC4D74E018F7066E. 3 For more on poverty and social impact assessments, see http://www.undp.org/poverty/projects_psia.shtml, and http://siteresources.worldbank.org/INTTOPPSISOU/Re sources/PSIA_Guide.pdf. 4 These full-length poverty and social impact assessments of trends in these countries’ energy and communal services sectors, are available at http://europeandcis.undp.org/senioreconomist/show/DF 091745-F203-1EE9-B8E083A9D6DBBC7D.


Chapter 1 - Introduction

questions about the effectiveness of social policy in mitigating poor household energy and water insecurities in these two countries. They also examine the actual and potential impact of the: •

large increases in household water and energy tariffs and prices (Figure 1.2);

changes in household energy consumption patterns, particularly in terms of coal, gas, central heating, firewood, and dung;

government “anti-crisis” responses to energy shortages, particularly in terms of measures to create additional electricity generation and transmission capacity, both via large investment projects and support for the expansion of decentralized renewable energy technologies (e.g., small hydropower); and

the design and implementation of energy efficiency projects.

Three relatively warm and wet winters, combined with the government’s response, helped power supply and consumption in Tajikistan to recover slightly in 2010. However, it still remained well below 2007 levels. In Kyrgyzstan, supply constraints pushed electricity consumption in 2010 to some 17 percent below 2007 levels. On the other hand, declines in electricity (and gas) consumption in Kyrgyzstan were partly offset

by large increases in coal consumption. Moreover, preliminary data from 2011 in Kyrgyzstan point to large improvements in electricity generation and consumption. In the communal services sector, both governments have dramatically raised tariffs. In Tajikistan, communal service tariffs rose by an average annual rate of 50 percent during 2007-2010; in Kyrgyzstan, household water tariffs rose about 87 percent in the same period (Figure 1.3). These increases, should they continue, could raise concerns about the affordability of and access to communal services. In the pages that follow, these issues are examined in an integrated, holistic fashion. Analysis of household budget survey data and financial, investment, and competition policies are combined to assess the current state and future implications of household poverty and vulnerability in Tajikistan and Kyrgyzstan, in terms of access to affordable energy and communal services. In addition to helping governments and international development agencies in Central Asia find the best answers to difficult development challenges, this study seeks to contribute to the growing literature on poverty, energy, and environmental sustainability. The remainder of this chapter briefly summarizes this publication’s key findings and results. A more detailed presentation and analysis of their implications are found in the concluding chapter.

15


Poverty and Social Impact Assessment Figure 1.1—Energy Energy consumption trends in Tajikistan and Kyrgyzstan (2007-2010) 2010)

102 100

180 Tajikistan: Electricity consumption (2007 = 100)

Kyrgyzstan (2007 = 100)

160 140

98

120 100

96

80

94

60

Electricity consumption Coal consumption

92

40 20

90

Gas consumption

0

88 2007

2008

2009

2010

2007

2008

2009

2010

* Generation less exports and losses. *** Production less net exports. Sources: State statistical agencies, UNDP calculations

Figure 1.2—Trends Trends in household energy, consumer prices in Kyrgyzstan and Tajikistan (2008-2010)

30% 2008

25%

63%

2008

2009

22%

2010

2009 2010

41%

14% 7%

8%

21%

21% 8% 7%

Energy Prices

CPI

Kyrgyzstan: Annual average inflation rates.

Energy Prices

CPI

Tajikistan: Annual average inflation rates

Note: For Tajikistan, the data are for household electricity tariffs. Source: State statistical agencies.

Trends in household energy access Official household survey data allow general pictures of sectoral trends to be supplemented with information about household access to, and the affordability of, electricity in Tajikistan and Kyrgyzstan. Although presented according to different methodologies, these data show that household access to electricity services 16

declined after 2007 in both countries. While the share of households reporting connections to the grid apparently increased in Tajikistan during 2007-2009 2009 (Figure 1.4), average household electricity consumption declined by 8.5 percent (from 390 to 357 kilowatt hours) during this time. Moreover, a 10 percent decline was reported for electricity consumption inn rural households (from 364 to 327 kilowatt hours). In Kyrgyzstan, the


Chapter 1 - Introduction

household survey data show a sharp increase in the frequency of electricity cut-offs in 2008,

which moderated somewhat in 2009 (Figure 1.5).

Figure 1.3—Household water tariff trends in Kyrgyzstan (2007-2010; June 2007 = 100)

220

200

180

160

140

120

100

2007

2008

2009

2010

UNDP calculations, based on State Statistical Committee operational data.

These data indicate that high connection rates to electricity grids (above 95 percent in both countries) became less meaningful during this time. This is particularly the case for rural households, where access to reliable electricity services seems most problematic. In Tajikistan, rural areas account for only around 10 percent of electricity consumption; an estimated 1 million people in rural areas spend much of the winter without regular electricity supplies. In Kyrgyzstan, 78 percent of low-income households in 2008 reported weekly (at least) interruptions in electricity service, compared to 72 percent for the country as a whole. Since rural poverty rates in Kyrgyzstan are well above national averages, these data point to special hardships for families in rural (and particularly mountainous) areas.

How are vulnerable households coping with uncertain access to electricity? The data in Figure 1.6 indicate that the poorest households (first and second income deciles) in Kyrgyzstan rely almost exclusively on coal and electricity for heating (as well as for lighting). Low-income households, and households in rural and mountainous areas, have very limited access to gas, central heating, and hot water supply. As such, their access to winter heating is more likely to be affected by interruptions in electricity supplies. By contrast, urban and upper-income households are more likely to have access to gas, central heating, and hot water. They are less likely to be inconvenienced by electricity cut-offs.

17


Poverty and Social Impact Assessment Figure 1.4—Declining Declining household access to electricity in Tajikistan (2007-2009 (2007 2009)

99%

2007

477

2009

2007

98%

2009

419

97%

390

96%

364

357 95%

327 94% 93% National

Dushanbe

National

Villages

Dushanbe

Villages

Average monthly household electricity consumption (kWh)

Share of households using electricity

Source: State statistical agency.

Figure 1.5—Shares Shares of households reporting electricity cut-offs cut offs (and their frequency) in Kyrgyzstan

70%

No interruptions Once a week

Few times yearly Several times weekly

Once a month Daily

60% 50% 40% 30% 20% 10% 0% 2006

2007

2008

2009

Source: State statistical committee.

In Tajikistan, more than half of households surveyed during 2007 and 2009 reported heating with wood and dung (Figure 1.7). Although the share of households reporting use of electricity for heating rose to during this time, coal use remained relatively small. These data call attention to the unintended ecological (deforestation) and health (wood, dung combustion) consequences associated with poor rural households’ energy coping strategies.

18

The different patterns of energy use reported in the two countries’ household survey data are in many respects consistent with the sectoral data (e.g., reflecting the greater use of coal in Kyrgyzstan). In other respects, they are surprising surprising—as with the much lower (albeit growing) share of households in Tajikistan an reporting use of electricity for heating. In general, however, data from both countries point to significant increases in household energy insecurity, linked primarily to reduced access to electricity supplies.


Chapter 1 - Introduction

Figure 1.6—Shares Shares of household spending on various energy sources in Kyrgyzstan (2009)

100% 80% Hot water 60%

Central heat Gas

40%

Electricity Coal

20% 0% 1 decile

National average

2 decile

9 decile

10 decile

Source: State Statistical Committee.

expensive to repair and maintain. Due to deteriorating infrastructure, half the population does not have access to adequate drinking systems and an even higher share is without access to adequate sanitation facilities.

Trends in access to communal ommunal services s Current trends in access to water, sanitation and solid waste management services offer a more complicated picture. In Kyrgyzstan, access to water is not threatened by rising tariffs, but rather by the decapitalization of the water and communal service infrastructure. Most of Kyrgyzstan’s urban and rural water supply and sewer systems were constructed 40-50 50 years ago without due regard to financial sustainability concerns. They are becoming increasingly

Poor accesss to water cannot be addressed within the current financial parameters of many service providers providers, due to tariffs that are set below cost cost-recovery levels and inadequate institutional capacity for effective billing and contracting. Efforts to close the financing ncing gap associated with maintaining the existing water structure

Figure 1.7—Shares of households s reporting use of various energy sources in Tajikistan (2007-2009)

2007

2009

44% 40% Half the population heats with wood and dung . . . with health, deforestation implications

32% 26%

17% 10%

13% 12%

12% 4%

Wood Source: State Statistical Agency

Electricity

Dung

Coal

Other

19


Poverty and Social Impact Assessment

(which which was assessed at $37.5 million in 2009) are also challenged by a household unwillingness to pay more for communal services. This is despite the fact that, that according to official household old survey data, even poor households pay tariffs that are well below affordability thresholds commonly used by the World Bank, UN,, and OECD. Figure 1.8—Shares Shares of household spending devoted to energy in Kyrgyzstan

6.5% 6.3%

6.2%

5.9% 5.4%

In Tajikistan, communal ommunal service tariffs rose by an average annual rate of 50 percent from 2007-2010. As in Kyrgyzstan, communal service providers in Tajikistan operate below cost-recovery recovery levels and current revenue from tariffs cannot cover the costs of operating and upgrading ading a deteriorating communal service infrastructure. The he 2006 Water Sector Development Strategy assessed Tajikistan’s investment needs in this sector at nearly $1 billion. However, unlike the situation in Kyrgyzstan, the vast majority of poor households are not paying tariffs for centrally centr supplied communal services. In addition to underscoring the importance of increasing investments nts in infrastructure expansion, this reduces the impact of communal service tariff increases on poor households.

20

Household energy and communal service affordability The sharp increases in household energy tariffs and prices (Figure Figure 1. 1.2) have raised concerns about the affordability of energy services in these countries, particularly for low-income income households. While definitions de of energy affordability vary, international benchmarks used by the World Bank, UN, and OECD governments imply that devoting 15 15-20 percent of household expenditures to energy (understood as electricity, heat, gas, hot water, etc.) is affordable. The national survey data on shares of household spending devoted to energy in these two countries are somewhat surprising in this regard. The large increases in household energy tariffs introduced in January 2010 are widely viewed as contributing to the April Apr 2010 popular uprising that unseated President Kurmanbek Bakiyev. However, Kyrgyzstan’s survey data for 2006-2010 indicate that shares of household budgets devoted to energy (5.5 (5.56.5 percent) have not been particularly high (Figure 1.8).5 While shares of low-income households’ budgets devoted to energy spending were somewhat above these national averages, they were generally below 10 percent. In Tajikistan, by contrast, the official survey data indicate that shares of household incomes devoted to energy spending are more likely to bump against the affordability threshold (Figure 1.9). This is particularly the case for households in the lowest income quintile, which were spending 16 percent of their budgets on energy in 2009.

5

Immediately upon assuming power in April 2010, the provisional government rescinded most of the household energy tariff increases introduced in January—will keeping in placee the increased wages, pensions, and social assistance payments intended to mitigate the effects of the tariff increases. This accounts for the apparent reduction in the shares of household budgets devoted to energy spending during the first half of 2010 shown in Figure 1.7.


Chapter 1 – Introduction Figure 1.9—Shares Shares household spending devoted to energy in Tajikistan (2009)

18%

10%

Average

17% 11%

16%

Poorest Quintile

10%

9% 5%

Other urban

Rural

National

Dushanbe

Share of household expenditures devoted to energy, 2009 data. Source: State Statistical Agency.

Since these same data indicate that 59 percent of household incomes (63 percent for poor households) were devoted to purchases of foodstuffs, many families in Tajikistan seemed to spend 75-80 percent of their income (more, in the wintertime) on food and energy. en These findings on energy tariffs are repeated for communal service tariffs. Official survey data in Kyrgyzstan indicate that poor households devote only 0.35 percent of their expenditures to water and sanitation, and another 0.1 percent to solid waste. This is well below the suggested 4-55 percent international affordability benchmark. In Tajikistan, many households seem unwilling or unable to pay more than current communal service tariff rates, although the vast majority of poor households are not paying tariffs for centrally supplied communal services at all.

Social policy lacunae The household survey data indicate that social policy has not been very effective in reducing household energy or water insecurity in either country. In Kyrgyzstan, only nly about half of social benefits are received by low-income households. Instead,, the share of social benefits accruing to upper-income upper households more than doubled (from 6 to 13 percent) during 2008-2010. Moreover, the monetization ization of categorical benefits— benefits including household energy subsidies— subsidies introduced in 2010 seems to have made Kyrgyzstan’s social policy framework more

regressive.. The share of these benefits received by low-income income households dropped from 46 percent in 2009 to only 20 percent during the first half of 2010; 10; it had been at 65 percent in 2008. By contrast, the share of categorical benefits accruing to upper upper-income households rose from 11 to 38 percent. In Tajikistan, social protection seems even less able to reduce household energy and water insecurity.. While the current social assistance scheme features compensation for electricity and gas use for around 1.5 million people, these benefits are not well targeted. And at $2 per month, they are not large enough to have a perceptible impact on vulnerable nerable household incomes. The case for linking social assistance to communal service tariffs iffs is weak, as most poor households are not paying tariffs for services like solid waste collection and sanitation services in the first place. With support from the World Bank and the European Union,, the government is now piloting a social policy reform programme, based on a single household income benefit reflecting proxy means testing. However, the new system is unlikely to be rolled out nationally for some years.

What is to be done? Trends in household water and energy insecurity in Tajikistan and Kyrgyzstan suggest a number of recommendations for researchers and policy makers.

21


Poverty and Social Impact Assessment

Governments in both countries have responded to the April 2010 developments in Kyrgyzstan by taking household electricity tariff increases off the table; by mid-2011 year-on-year electricity price inflation rates in both countries had dropped close to zero. However, in light of the financial pressures on energy companies in Tajikistan and Kyrgyzstan, continuing high global and regional energy prices, and the fact that household electricity tariffs in both countries remain low by global and regional standards, further tariff hikes seem a matter of time.

greater emphasis on off-grid energy solutions, particularly small hydropower and other decentralized renewable energy technologies; and

energy efficiency investments.6

Many of these challenges reflect governance problems in the energy and communal service sectors in Tajikistan and Kyrgyzstan. Improvements in sectoral governance could emphasize:

In anticipation of this, consideration of the following steps seems most prudent:

The establishment or strengthening of national policy frameworks for water and communal service provision.

Strengthening the evidentiary basis for monitoring the links between household incomes, energy and communal service prices and access, and social policy.

o In Tajikistan, this would entail the creation of a national body for water, sanitation, and solid waste policy and investment planning.

The possible expansion of lifeline electricity tariffs (i.e., low tariff rates for small amounts of electricity consumed). The Pamir Energy Company, which provides electricity to Eastern Tajikistan’s impoverished Gorno Badakhshan region, employs lifeline tariffs (albeit with donor subsidies)—showing that lifeline electricity tariffs are feasible.

o In Kyrgyzstan, it would require improvement of existing regulatory and coordination structures. •

Since significant numbers of households in these two countries do not have access to reliable electricity services in the winter time, and since the delivery of social assistance for low-income households in these countries is not particularly effective, direct efforts to provide energy and communal services to insecure households should be stepped up. These include:

22

connection subsidies for households without access to electricity, gas, central heating and communal service grids;

6

Framework legislation in both countries to allow municipalities and communities to manage local water and other communal service infrastructure is in place. This makes possible expanded cooperation among municipalities, private companies, and civil society organizations in order to improve service delivery and attract additional investment into this sector. However, these legal changes need to be backed by appropriate financing and capacity development activities, particularly in the communal service sector.

See Zoran Morvaj, Slavica Robic, Alessandra Bravi, and Ben Slay, “Household energy supplies and integrated rural development in Tajikistan” (http://europeandcis.undp.org/senioreconomist/show/C AADD536-F203-1EE9-BDFD7E97AE1D5760).


Chapter 2: Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Author: Ben Slay

23


Executive summary •

Chapter’s focus: This chapter focuses on three sets of recent (2009-2011) developments concerning the poverty/ energy/vulnerability nexus in Central Asia: (1) trends in household energy and water security in Kyrgyzstan and Tajikistan, as reflected in official socio-economic data; (2) recent research on this topic; and (3) the causes and implications of the April 2010 events in Kyrgyzstan concerning these issues, particularly tariff increases and threats to household energy security.

Household energy (and communal service) tariffs in these two countries rose sharply during 2008-2010. These increases can be justified by the need to reach cost-recovery levels and improve these sectors’ investment prospects. However, the 2010 events in Kyrgyzstan show that higher tariffs can be a double-edged sword. In the absence of: (i) improved management within the energy sector (particularly to reduce theft and losses); (ii) the credible threat of disconnection for non-payment; (iii) effective public communications concerning the need for, and low-cost options for responding to, higher tariffs; and (iv) de facto household willingness to pay higher tariffs, energy tariff hikes can be ineffective—or worse. Energy price inflation in both countries seems likely to be quite subdued in 2011, due to social concerns.

24

Universal access to basic energy (and communal) services in Kyrgyzstan and Tajikistan was lost, or had never been attained, even before the winter shortages of recent years. Official household survey data indicate that, while nearly all households in both countries are nominally connected to the national electricity grid, a sharp increase in the frequency of planned and unplanned service interruptions during 2008-2009 effectively broke the link between grid connections and reliable

electricity supplies. Gas, central heating, and hot water supplies are unavailable to rural and many urban households in both countries. In Tajikistan, an estimated 1 million people (mostly in rural areas) spend much of the winter without electricity supplies. In Kyrgyzstan, households have responded to growing energy insecurities by significantly increasing coal consumption; official survey data indicate that nearly half of Tajikistan’s households rely on firewood and dung for winter heating.

Energy seems to be more affordable in Kyrgyzstan than in Tajikistan. Official household data indicate that, whereas only 6-7 percent of household expenditures are devoted to energy spending in Kyrgyzstan (including for poor households), this share is approximately double in Tajikistan. However, poor households and residents of rural and mountainous regions in both countries are much more likely to suffer from electricity cut-offs than well-to-do urban residents. These data also indicate that more than half of Tajikistan’s households rely on firewood and dung for winter heating.

Social policy seems unable to mitigate hardships posed by current and prospective energy price increases. Official survey data indicate that only about half of Kyrgyzstan’s social benefits are received by low-income households; this share actually dropped slightly during 2009-2010 (to 50 percent, down from 52 percent in 2008). By contrast, the share of social benefits accruing to upper-income households more than doubled (from 6 to 13 percent) during 2008-2010. Much of the increased social spending in response to the crisis developments of 2009-2010— significant shares of which were financed by donors—seems to have leaked to relatively wealthy households. While


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

possibility of drought in 2011. It is notable that a surprise cold snap in March 2011 led to the introduction of sharp electricity rationing in Tajikistan. On the other hand, the winter of 2010-2011 saw visible improvements in energy security in Kyrgyzstan, owing to progress in the refilling the Toktogul hydropower reservoir, Central Asia’s largest.

Tajikistan’s social assistance scheme features compensation for electricity and gas use—up to certain limits, for around 240,000 households (1.5 million people) who are identified by the local authorities—these benefits are not well targeted. And at $2 per month, they are not large enough to have a perceptible impact on vulnerable household incomes.

Important progress has been made . . . Since the advent of Tajikistan’s 2007-2008 winter crisis, governments and development organizations in Central Asia have recognized the importance of these issues, and have taken on board many recommendations that were made in UNDP’s 2009 Central Asia Regional Risk Assessment, Framework for Action7 and other relevant documents. The early recovery paradigm has increasingly been used to bridge the gap between emergency/humanitarian responses and development programming in the two countries; programming in support of disaster risk reduction (including early warning/risk management systems), decentralized renewable energy, and energy efficiency and water conservation has expanded. The governments of Tajikistan and Kyrgyzstan have increased investment in their energy and water infrastructures.

. . . But this progress remains generally untested by a return of drought conditions or severe winter weather. At least some of the progress made has been a reflection of above-normal precipitation and relatively warm winters. It is not clear how well this progress would stand up to the return of drought conditions, or severe winter frosts. Early indications point to a growing

Common frameworks for energy sector development shared by governments and the international community in these two countries appear to be lacking. The emphasis on tariff increases and rebalancing, unbundling, and privatization via sales to strategic (often foreign) investors that has underpinned energy sector reforms in many transition economies seems to have run out of steam in Kyrgyzstan and Tajikistan. In Kyrgyzstan, many pre-April 2010 privatization measures have been repealed, and tariff increases are off the table for the present. The emphasis is instead on reducing corruption within the energy sector by strengthening state control. In Tajikistan, tariff increases likewise seem to be off the table at present, while promises to unbundle the electricity sector continue to be largely honored in the breach. Although the government of Iran is financing the construction of the Sangtuda2 hydropower station, prospects for further privatization and foreign investment seem rather uncertain in light of high public debt levels, and the continuing arrears of the Barqi Tojik state energy monopoly to the Sangtuda-1 hydropower station (75 percent of which is owned by Russia’s RAO EES).

7

See http://europeandcis.undp.org/uploads/public1/files/vuln erability/Senior%20Economist%20Web%20site/undp_c entral_asia_regional_risk_assessment(1).pdf; and http://europeandcis.undp.org/uploads/public1/files/vuln erability/Senior%20Economist%20Web%20site/CARR A_Framework_for_Action_June_2009(1).pdf.

25


Poverty and Social Impact Assessment

Recent macroeconomic, socio-economic, and energy trends This paper’s point of departure is the impact of the winter energy crisis of 20072008 on Tajikistan, and of the subsequent drought of 2008 on Tajikistan and Kyrgyzstan.8 These events, the consequences which in Kyrgyzstan were exacerbated by government decisions to sharply increase electricity exports from hydropower stations along the Naryn cascade during 2007-2008:

8

Aggravated perennial intra-state tensions along Central Asia’s waterenergy nexus;9

Ushered in a multi-year episode of heightened national and household energy and water insecurity in Kyrgyzstan and Tajikistan;

Interacted with 2007-2008 global food price trends to sharpen national and household food security concerns in these two countries;

Were then exacerbated by the impact of the global economic crisis, which significantly reduced the remittance inflows that are a key coping

See Johannes F. Linn, “Impending Water Crisis in Central Asia: An Immediate Threat”, 2 June 2008, Brookings Institution, Washington D.C. (http://www.brookings.edu/opinions/2008/0619_central _asia_linn.aspx). 9 Questions about the extent to which the issues discussed here are relevant to other Central Asian (or CIS) countries are not taken up here. The absence of publicly available high-frequency official data on these issues in Uzbekistan and Turkmenistan, and policies in these countries, argue against expanding the geographic focus of this research beyond Kyrgyzstan and Tajikistan. In light of its upper middle-income country status, and its relatively low levels of income poverty and household food and energy insecurity, this research is not extended to Kazakhstan.

26

mechanism for vulnerable households in both countries; and

Contributed to the government of Kyrgyzstan’s decision to dramatically increase energy tariffs in January 2010—which helped precipitate the April 2010 overthrow of President Bakiyev, and set off months of heightened political and economic instability.

The winter crisis and drought, which led to UN-sponsored humanitarian appeals in Tajikistan and Kyrgyzstan during 2008-2009, also revealed that the development progress associated with the rapid GDP growth enjoyed by these low income countries during most of the previous decade had been quite fragile. Sharp reductions in winter electricity and heat supplies for households, as well as for schools, health care institutions, and enterprises, led to significant declines in industrial output and growing problems of access to public services. According to informal estimates, up to 1 million people in Tajikistan now spend the winter without reliable electricity supplies. The official data also report significant increases in prices for energy (and communal) services during this time (see Table 2.1 below), running well above growth in the consumer price index. The emergence of these threats to household welfare, in the form of declining access to energy, water, and public services, combined with high food and energy prices and falling remittances, blurred the distinction between development programming and humanitarian/emergency responses, for governments and donors.10 While poor rural 10

These points are made in more detail in UNDP’s Central Asia Regional Risk Assessment, 2009


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

In developing countries, a traditional manifestation of household energy insecurity is households that have never been connected to the relevant grids. By contrast, energy insecurity during the winters of 2007-2008 in Tajikistan, and then 2008-2009 and 2009-2010 in Tajikistan and Kyrgyzstan, was in large part a story of grids through which electricity, heat, and gas were not flowing. Likewise, food insecurity increased in 2007-2008 and again in

households in Central Asia may traditionally experience food, energy, and water insecurities, developments since the winter crisis have spread these insecurities to other households, particularly in urban areas outside of the capital cities. Hydropower shortages due to drought reduced access to education, social services, and winter heat, particularly for poor urban households that used to have access to these services.

Table 2.1—Macroeconomic trends, 2008-2010

2008 Output trends GDP growth Industrial output (annual change) Gross agricultural output (annual change) Electricity (annual change): - Generation - Apparent consumption11 Price trends (annual) Consumer price inflation rate (average) Energy price inflation rate (average)12 Communal services inflation (average)13 External trends Current account balance (% of GDP) Remittances (annual change) Months of merchandise import coverage

Kyrgyzstan 2009 2010

2008

Tajikistan 2009 2010

8.4% 15% 1%

2.3% -13% 7%

-1.4% 10% -3%

7.9% -6% -21%

3.4% -10% 8%

6.5% 13% 10%

-21% -4%

-6% -3%

0% -6%

-14% -8%

7% 1%

1% 2%

25% 30% 27%

7% 22% 12%

8% 14% 23%

21% 63% 73%

8% 41% 53%

7% 21% 26%

-8% 65% 3.9

2% -15% 6.0

-7% 30% 6.4

-8% 50% 1.7

-6% -32% 2.3

2% 27% 2.4

All data are from state statistical committees and national banks, or are UNDP calculations, except for the current account balances, which are taken from the April 2011 IMF World Economic Outlook database.

(http://europeandcis.undp.org/home/show/7F66C9DAF203-1EE9-B8DFA2E976F01F3D). For more on longer-term development challenges in Central Asia, see UNDP’s Central Asia Human Development Report—Bringing Down Barriers: Regional Cooperation for Human Development and Human Security, 2005 (http://hdr.undp.org/en/reports/regional/europethecis/na me,3281,en.html). 11 Electricity generation less net exports and losses, statistical office data. Losses are not included in the data for Tajikistan. 12 For Kyrgyzstan: reference is to the “electricity, gas, heat and other fuels” component of the consumer price index; for Tajikistan, reference is to the “electricity, gas, and other fuels” component of the consumer price index. 13 For Kyrgyzstan: data are for increases in household water tariffs.

2010-2011 because of global food price trends, rather than drought or other natural disasters (traditional, conventional causes of food insecurity). All this happened despite continuing economic growth during in both countries (or, at least through the first quarter of 2010 in Kyrgyzstan). References to “compound” or “slow onset” crises, and to “fragile states” to describe this situation in Tajikistan and Kyrgyzstan, have become increasingly common since 2008. Fortunately, some more favourable trends took hold in these two countries in 2009-2010, attenuating some of the underlying compound crisis drivers. First among these was the end of the 2008 drought, thanks to two 27


Poverty and Social Impact Assessment

Figure 2.1—Monthly water volume trends at Central Asia’s largest hydropower stations (2009-2011)

50% Toktogul

40%

Nurek

30% 20% 10%

2011 Mar

2010

Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb

-10%

2009 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

0%

-20% -30% -40% -50% UNDP calculations, based on data posted on the CAwater-Info website (http://www.cawater-info.net/analysis/index_e.htm). Each data point shows the average water volume for that month (in 2009 or 2010) relative to the average volume for that month, going back to 1991-1992. Data points along the horizontal axis (zero deviation from monthly averages) correspond to monthly norms.

consecutive relatively warm wet winters. As the data in Figure 2.1 above show, water volumes in the main hydropower reservoirs along the Vakhsh and Naryn cascades (which account for more than 90 percent of electricity generation capacity in Tajikistan and Kyrgyzstan, respectively) underwent dramatic improvements during 2009-2010. This helped electricity generation to stabilize in these two countries in 2010, making possible a recovery in industrial production (see Table 2.1). However, increased electricity exports, combined with relatively mild autumn weather, seem to have further depressed electricity consumption in Kyrgyzstan last year, while electricity consumption in Tajikistan noted only a slight increase. The relatively mild impact of the global economic crisis on these countries, both of which reported slower but positive GDP growth during 2009, was a second favourable development. On the one hand, the growth slowdown, combined with continuing population growth and large declines in remittance inflows, imply that growth in living standards—particularly for vulnerable households—essentially ground to a halt in 2009. In Kyrgyzstan, pressures on households were exacerbated by a large external adjustment: the current account, which posted

28

a deficit of 8 percent of GDP in 2008, swung to a surplus of 2 percent of GDP in 2009. Combined with 3 percent reported GDP growth, this swing meant a significant drop in domestic demand. Most of this decline was apparently born by Kyrgyzstan’s households: a punishing 14 percent decline in household consumption has been officially reported for 2009.14 On the other hand, the continuation of GDP growth in these two countries in 2009 compared favorably with most other economies in the former Soviet space (not to mention in the European Union, the US, and Japan). Also, improvements in hydropower generation capacity, combined with the strong recoveries reported in remittance inflows and higher prices for key exports (gold, aluminum), set the stage for a return to rapid industrial and GDP growth in 2010 in Tajikistan, as well as in Kyrgyzstan—prior to the events of April and June 2010 that sent the economy into a tailspin.15 14

The current account data come from the IMF’s World Economic and Financial Surveys, Regional Economic Outlook: Middle East and Central Asia, April 2011, p. 71. The household consumption figure comes from the state statistical committee. 15 For more on this, see Ben Slay, “Kyrgyzstan: From ‘compound’ to socio-economic crisis?”,


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan Map 2.1—Central Asia’s electricity transmission grid

Source: World Bank, 2008. More recent additions to the transmission infrastructure are not shown.

However, new energy-sector risks appeared in late 2009, in the form of Uzbekistan’s and Kazakhstan’s withdrawal from the integrated Central Asian electricity transmission grid (see Map 2.1 above). Under the integrated transmission system, Tajikistan’s northern and western regions could rely on the automatic dispatch of (and subsequent settlement for) electricity generated in Uzbekistan. The end of these arrangements pushed the Central Asian countries towards bilateral commercial deals in which prices and quantities of electricity trades are determined ex ante. Developments during the winters of 20092010 and 2010-2011 indicate that the transition from a regional to bilateral trading arrangements occurred without catastrophic http://europeandcis.undp.org/senioreconomist/show/DB 1BDF26-F203-1EE9-B9D2AC9FEC7A7FC1.

consequences. In Tajikistan, this was due in part to increases in domestic generation and transmission capacity: with financing provided by China’s ExImBank, Tajikistan completed the construction of the South-North transmission line that wheels power from the hydro stations on the Vakhsh cascade in the south to Sughd oblast in the north. This new capacity allows Tajikistan to by-pass portions of what had been the common regional grid that lie in Uzbekistan. The installation of turbines at the Sangtuda-1 (and then Sangtuda2) hydropower stations provided additional generation capacity. Kyrgyzstan likewise benefitted from new bilateral agreements with Kazakhstan on non-winter electricity exports: official data indicate that Kyrgyzstan’s electricity exports rose by 87 percent in 2009 and 53 percent in 2010. The November 2010 launch of the first generator at the Kambarata2 hydropower plant on the Naryn cascade 29


Poverty and Social Impact Assessment Figure 2.2—Impact of the dissolution of the integrated Central Asian electricity transmission network on Tajikistan’s electricity exports and imports (2009-2010)

1,000

Monthly electricity exports (mln kWh)

800

Monthly electricity imports (mln kWh)

600 400

Dissolution of the integrated Central Asian electricity transmission network

200

2010

2009

Dec

Nov

Oct

Sep

Aug

Jul

Jun

May

Apr

Mar

Feb

Jan

Dec

Nov

Oct

Sep

Aug

Jul

Jun

May

Apr

Mar

Feb

Jan

0

Statistical Office data UNDP calculations.

likewise helped reduce Kyrgyzstan’s energy shortages during the winter of 2010-2011. On the other hand, official data indicate that Tajikistan’s electricity imports and exports dropped sharply in late 2009, after the dissolution of the integrated Central Asian electricity transmission system (see Figure 2.2). This does not bode well for Tajikistan’s ability to import electricity going forward— especially from (or via) Uzbekistan, which held up millions of dollars worth of Tajikistan’s imports over the course of 2010. It is telling that, despite the addition of significant additional capacity for power generation (from Sangtuda-1) and transmission (from the south-north power line), an unexpected cold snap led to the reintroduction of nation-wide electricity rationing in March 2011. Increased electricity imports were not an option. In the longer term, two decades of under-investment have left both countries’ energy sectors facing growing risks of decapitalization and “catastrophic failure”. The Joint Economic Assessment published by the World Bank in mid-2010 assessed the immediate needs of Kyrgyzstan’s energy sector (for the winter of 2010-2011) at $180 million. Out of this amount $124 million was

30

needed for essential and critical repairs for heating plants and district heating systems, and to ensure security in electricity generation.16 A USAID study found that the power sector in late 2010/early 2011 was at a “tipping point”, and that at least $190 million in new investments were needed to reduce risks of catastrophic failure.17 The funding needed for all of Kyrgyzstan’s energy sector investment projects is estimated in the $5-6 billion range, a sum that exceeded 2010 GDP ($4.6 billion). It is not clear when, whether, or from whom these funds could be obtained. Likewise, the total funding gap for the shortterm energy projects listed in Tajikistan’s 2010-2012 Poverty Reduction Strategy Paper is estimated at some $2.2 billion—a sum that does not include medium- to longer-term energy sector investments (e.g., the Roghun hydropower station, the construction costs of which would be as high as $3 billion).

16

For more see, “The Kyrgyz Republic Joint Economic Assessment: Reconciliation, Recovery, and Reconstruction”, ADB, IMF, the World Bank, 2010. 17 “Highlights from the Executive Summaries of the Phase 1 Reports on the Management Diagnostic of NESK, EPP & and the Distribution Companies”, Bishkek, USAID, 15 April 2011.


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Recent research on poverty, energy, and vulnerability in Kyrgyzstan and Tajikistan Until recently, our understanding of income- and non-income income poverty (particularly as concerns access to reliable energy, water, and communal services) in Tajikistan and Kyrgyzstan were constrained by the absence of: (1) internationally comparable income poverty data that are more recent than 2005 (from the World Bank’s POVCALNET database18); (2) national (i.e., not necessaril necessarily internationally comparable) poverty data that are more recent than 2007-2008 2008 (i.e., before the onset of the winter and global economic crises); and (3) data sets and studies

quantifying and exploring post post-2007 links between income poverty on the one hand hand, and falling remittances, higher food and energy prices, and reductions in access to energy, communal, and social services (as well as other vulnerability drivers, like gender, age, family size, rural versus urban location, ethnicity, and the like) on the other. Data from the World Bank’s POVCALNET database indicate that absolute income poverty rates in these countries rose dramatically in the 1990s during the transition

Figures 2.3, 2.4—Income Income poverty rates in Tajikistan, Kyrgyzstan, other countries (1999 (1999-2005)

Share of population living below PPP$2.15/day threshold

Share of population living below PPP$1/day threshold 45%

82%

1999 73%

2002

34%

2002 2005

2005

56%

56%

1999 36%

71%

46%

22%

21% 16% 25% 17%

13% 10% 11%

5% 2%

12% 5% 2%

Tajikistan

Kyrgyzstan Kazakhstan

Turkey

Russia

Tajikistan

1%

Kyrgyzstan Kazakhstan

2% 3% Turkey

2% 0% Russia

Source: World Bank POVCALNET database. 18

This global database (http://go.worldbank.org/NT2A1XUWP0)) seeks to provide comparable equivalized absolute and relative income poverty estimates that are based on consumption expenditure data gathered by household budget (or living standards) surveys, and are made comparable by the use of common prices and purchasing-power-parity parity exchange rates developed under the International Comparison Programme (See http://siteresources.worldbank.org/ICPEXT/Resources/I CP_2011.html).

recession, and then fell sharply during the decade of recovery growth that end ended with the global economic crisis in 2009. While income poverty in most transition economies displays similar trends, they are particularly pronounced in Kyrgyzstan and Tajikistan (see Figuress 3 and 4 above). The large remittance inflows received by these ccountries— particularly Tajikistan (see Figure 2.5 31


Poverty and Social Impact Assessment

below)—made made a major contribution to the rapid pre-2009 2009 growth in household incomes, and therefore to poverty reduction. As labour migration is particularly likely to serve as a coping mechanism for vulnerablee households with unskilled workers, remittances as an antianti poverty device may have significant “self“self targeting” elements.

As of 2005, 90 percent of the population in Kygyrzstan and Tajikistan were living at or below the threshold that the World Bank associated with significant “vulnerability” ty” to poverty. Despite this progress, it is noteworthy that, as of 2005, more than a million people in each country were living at or below $1/day (in purchasing-power-parity parity terms) terms)—the global standard for extreme poverty. The World Bank nk data indicate that, together with Uzbekistan, these countries accounted for three quarters of those living on less than $1/day in the Europe and Central Asia region

in 2005. These data also indicate that, as of 2005, more than half the population in bot both Tajikistan and Kyrgyzstan were living at or below $2.15/day—the the income poverty threshold regarded by the World Bank as most appropriate for the Europe and Central Asia region. It is likewise noteworthy that, as of 2005, 90 percent of the population in Kyrgyzstan rgyzstan and Tajikistan were living at or below the PPP $4.30/day threshold that the World Bank associated with significant “vulnerability” to poverty. Thus, even factoring in the strong growth in household incomes and consumption recorded during 2006-2008,, it is likely that the vast majority of households in Tajikistan and Kyrgyzstan were living below, or close to, internationally determined poverty and vulnerability thresholds, prior to the onset of the global economic crisis in late 2008. Internationally comparable non nonincome poverty data for Kyrgyzstan and Tajikistan—particularly particularly concerning access to energy, water, and communal services—are services scarce and generally pre-date date 2007. The most recent internationally comparable data on access to sanitation itation services and improved water sources would seem to be from 2004

Figure 2.5—Remittance Remittance inflows infl as a share of GDP (2005-2010)

60% 50% 2005 40%

2006 2007

30%

2008 20%

2009 2010

10% 0% Tajikistan

Moldova

Sources: World Bank, IMF, EIU data; UNDP calculations.

Kyrgyzstan


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan Figure 2.6—Share Share of population without access to improved sanitation services

1990

Figure 2.7—Share Share of population without access to improved water sources

1990

2004

2004

41%

51%

49% 40% 41% 33%

22%

28% 28%

23%

23%

24%

18%

6%

Kazakhstan Uzbekistan Kyrgyzstan

Uzbekistan Kyrgyzstan Kazakhstan Tajikistan

Tajikistan

Source: UNDP Human Development Report 2006, pp. 306-307. 306

The 2007 Living Standards Survey19 authored by the State Statistical Agency and UNICEF (on the basis of World Bank data) found that 81 percent of the population of Tajikistan—including including 76 percent of those classified as non-poor poor and close to half of the urban population—relied relied on solid fuels for heating and cooking, even before th the 20072008 winter crisis. By contrast, only 5 percent had central heating in their dwelling. According to this study, piped gas was

(see Figuress 6 and 7 above); they point to significant levels of deprivation, which typically had not seen much improvement after 1990. By contrast, more conventional non-income poverty measures (e.g., maternal, child mortality rates—see Figure 2.8 2. below) place these countries around global averages—although although well below rates reported in some lower middle-income income CIS countries (e.g., Moldova).

Figure 2.8— —Maternal, under-five mortality rates (2005)

170 Maternal mortality (per 1000 live births)

150

140

Under-5 mortality (per 100,000 live births)

73

67

71

44 22

28 16

Moldova

Russia

29 18

Turkey

Kazakhstan

Kyrgyzstan

Tajikistan

Source: UNDP Human Development Report 2007 2007-2008, pp. 262-263. 19

Tajikistan Living Standards Measurement Survey 2007: Indicators at a Glance,, 2009.

33


Poverty and Social Impact Assessment

supplied to only 18 percent of the population; only 59 percent of the population had access to improved water facilities (83 percent in urban areas, and 43 percent of those classified as non-poor), while only 15 percent had access to improved sanitation facilities20 (48 percent in urban areas, and 19 percent of those classified as non-poor).

on trends that have appeared since Tajikistan’s winter energy crisis of 2007-2008. This research includes:

As in Tajikistan, the vast bulk of those judged to be in poverty in Kyrgyzstan live in rural areas. Similar pre-2009 trends are apparent in the official poverty data for Kyrgyzstan (which are calculated on the basis of the annual subsistence minimum, expressed in nominal som). These show income poverty rates dropping from a high of 63 percent in 2000 to 32 percent in 2008 and 2009. As in Tajikistan, poverty rates in rural areas of Kyrgyzstan have consistently been higher than in urban areas; in 2009 these were at 37 and 22 percent, respectively. As in Tajikistan, the vast bulk (76 percent in 2009) of those judged to be in poverty in Kyrgyzstan live in rural areas. Likewise for child poverty in Kyrgyzstan—a recent World Bank study reports that, as of 2005:21

The release in the late 2010 of official 2009 poverty data for Kyrgyzstan, and, for Tajikistan, the World Bank Poverty Summary Note, updating poverty trends in these countries through 2009;

Draft World Bank research on the social safety net in the two countries, including Social Safety Net in the Kyrgyz Republic: Capitalizing on Achievements and Addressing New Challenges (issued in May 2009) and Tajikistan: Delivering Social Assistance to the Poorest Households (draft from 24 November 2010);

UNICEF-Kyrgyzstan’s National Report on Child Poverty and Disparities, issued in April 2010;

The DFID-funded study The Impact of the Global Economic Crisis on Households in Tajikistan: Sociological Research Results,22 authored by the Panorama public foundation and released in Dushanbe in February 2009;

52 percent of under-18 year-olds and 58 percent of under-6 year-olds were living in poverty; 15 percent of under-18 year-olds and 19 percent of under-6 year-olds were living in extreme poverty; and 70 percent of those living in poverty in were members of households with three or more children.

The year 2010 saw the appearance of new research on to these issues, shedding light 20

Classified as flush or poor/flush plumbing, or a latrine, connected to piped sewer or septic systems. 21 On the basis of household budget survey data. Source: Social Safety Net in the Kyrgyz Republic: Capitalizing on Achievements and Addressing New Challenges, World Bank, 20 May 2009, p. 7.

34

22

“Влияние глобального экономического кризиса на домохозяйства в Республике Таджикистан (по результатам социологического исследования)” (“The impact of the global economic crisis on households in the Republic of Tajikistan: Sociological research results”).


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan Table 2.2—Poverty trends in Tajikistan (2007-2009) 23

Poverty rates Income poverty - Urban - Rural Urban households with no migrants Urban households with migrants Rural households with no migrants Rural households with migrants Households with no children Households with two children Households with four children Extreme income poverty - Urban - Rural

2007 53.1% 49.3% 54.4% 40.5% 33.9% 51.6% 47.6% 39.2% 56.6% 61.6% 17.4% 18.9% 16.9%

2009 47.2% 41.8% 49.2% 34.3% 27.7% 40.6% 32.1% 29.6% 54.6% 71.5% 17.5% 17.5% 17.5%

Source: World Bank Poverty Summary Note for Tajikistan, September 2010.

USAID-funded studies on The Curtailment of Electricity Consumption: The Attitude of the Population and the Impact on the Quality of Life (November 2009) and Kyrgyzstan Household Energy Analysis and Proposed Social Protection Measures (November 2008), written by PA Consulting and published in Bishkek; and

Poverty and social impact assessments of trends in the energy and communal service sectors in Tajikistan and Kyrgyzstan.24

Recent income poverty data The World Bank’s Poverty Summary Note indicates that the global financial crisis did not prevent continuing declines in income poverty rates during 2007-2009 in Tajikistan. 23

Measured as per capita consumption, relative to the cost of purchasing a diet of 2,250 calories per capita per day, plus an allowance for non-food consumption. The nominal poverty line was set at 4.56 Somoni per capita per day (in 2007 prices), of which 64% was devoted to food-consumption and 36% to non-food consumption. 24 See http://europeandcis.undp.org/senioreconomist/show/DF 091745-F203-1EE9-B8E083A9D6DBBC7D.

The data in Table 2.2 indicate that poverty rates during this time fell overall (from 53 to 47 percent), and for both urban (49 to 42 percent) and rural (54 to 49 percent) households. Despite the declines in remittances reported during 2009 (see Figure 2.5), the importance of these inflows in reducing poverty seemed to increase during 2007-2009—particularly in rural households, where the presence of migrants produced a 15.5 percentage-point decline in the poverty rate. These data also underscore the continuing importance of child poverty in Tajikistan: members of households with four or more children were more than twice as likely to live in poverty as members of households with no children. While overall income poverty rates in Tajikistan may have fallen during 2007-2009, the same cannot be said for extreme poverty rates, which remained essentially unchanged (at 17.5 percent) during this time. And while poverty rates for households with two or fewer children fell, they rose sharply (from 62 to 72 percent) for households with four or more children. Likewise, the data in Table 2.3 indicate that poverty and extreme poverty rates rose sharply in the Gorno Badakhshan and Khatlon regions during 2007-2009.

35


Poverty and Social Impact Assessment

Table 2.3—Regional income poverty rates in Tajikistan (2007-2009)

Poverty 2007 2009 53.5% 47.2% 49.4% 41.8% 55.0% 49.2% 43% 34% 43% 62% 47% 54% 49% 39% 69% 48%

National - Urban - Rural Dushanbe Gorno Badakhshan Khatlon Regions of Republican Subordination25 Sughd

Extreme poverty 2007 2009 17.4% 17.5% 18.9% 17.5% 16.9% 17.5% 16% 14% 10% 26% 8% 21% 14% 12% 31% 17%

Sources: World Bank Poverty Summary Note for Tajikistan, September 2010; PRSP for 2010-2012, pp. 6-8.26 Table 2.4—Trends in absolute poverty rates in Kyrgyzstan, 2007-2010

Region National - Urban - Rural Batken region Jalalabad region Issyk-Kul region Naryn region Osh region Talas region Chui region Bishkek city

2007 35.0% 23.2% 41.7% 40.4% 53.0% 38.6% 45.2% 46.6% 35.3% 15.0% 5.0%

Poverty rates 2008 2009 31.7% 31.7% 22.6% 21.9% 36.8% 37.1% 20.7% 31.5% 40.1% 36.9% 52.2% 46.1% 42.7% 44.1% 37.5% 38.3% 43.0% 33.0% 15.8% 21.2% 15.2% 13.2%

2010 33.7% 23.6% 39.5% 33.6% 44.7% 38.0% 53.5% 41.9% 42.3% 21.9% 7.9%

2007 6.6% 3.2% 8.5% 9.2% 12.0% 8.1% 12.8% 6.4% 7.9% 1.4% 0.6%

Extreme poverty rates 2008 2009 6.1% 3.1% 3.2% 2.7% 7.7% 3.3% 3.9% 6.0% 9.8% 0.5% 16.9% 6.6% 11.6% 10.0% 4.5% 2.0% 4.6% 2.9% 2.1% 2.4% 2.1% 3.3%

2010 5.3% 4.2% 6.0%

n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

Source: State Statistical Committee, Bishkek, 2009, 2010.

Table 2.5—Fiscal, social policy trends in Kyrgyzstan, Tajikistan (2008-2011)

Share of GDP devoted to: Budget revenues - Grants Budget expenditures - Public employment - Household transfers27 - Pensions Budget balance: - With grants - Without grants Public debt

2008 25.9% 1.9% 26.2% 6.5% 5.0% 4.4%

Kyrgyzstan 2009 2010 27.1% 28.8% 5.1% 2.9% 36.1% 38.1% 7.3% 8.0% 6.8% 9.4% 6.1% 9.0%

2011 30.4% 4.3% 42.3% 11.1% 9.0% 8.6%

1.6% -0.3% 48.5%

-3.9% -9.0% 58.0%

-7.6% -11.9% 52.5%

-6.4% -9.3% 62.6%

2008 20.5% 1.6% 28.0% 4.1% 3.4% n.a.

Tajikistan 2009 2010 20.0% 20.2% 1.6% 1.3% 28.6% 26.6% 4.7% 4.6% 4.1% 3.9% n.a. n.a.

2011 21.6% 1.7% 29.6% 4.9% 4.1% n.a.

-5.9% -7.5% 30.0%

-7.0% -8.6% 34.3%

-6.3% -8.0% 37.4%

-5.3% -6.4% 36.4%

Sources: IMF country reports (Kyrgyzstan—October 2010, June 2011; Tajikistan—December 2010, June 2011); UNDP calculations. 25

This is the broader region to which Dushanbe city belongs. It includes also Tursunzode (where the Tajikistan Aluminum Company—the country’s largest industrial enterprise—is located), and the headwaters of the Vakhsh river, where the hydropower plants that generate more than 90 percent of Tajikistan’s electricity are located. 26 The regional poverty data for 2007 and 2009 may not be fully comparable. 27 For Kyrgyzstan, these data are for Social Fund expenditures.

36


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Trends in national absolute income poverty data for Kyrgyzstan (which are calculated on the basis of the annual subsistence minimum, expressed in nominal som) are shown in Table 2.4. In contrast to Tajikistan, they show progress in poverty reduction stopping in 2009, when the overall income poverty rate apparently remained constant at 32 percent. On the other hand, a decline in the extreme poverty rate was reported for that year (from 6 to 3 percent). This progress in extreme poverty reduction in 2009 is somewhat difficult to reconcile with Kyrgyzstan’s consumer expenditure data: official household survey data indicate that household expenditures in real terms declined by 6 percent in 2009;28 while Kyrgyzstan’s national accounts data report a 14 percent fall in individual consumption for that year. Kyrgyzstan’s national poverty data also come with some interesting regional variations. Poverty rates in 2008 were reported up in much of northern Kyrgyzstan, including especially Bishkek city, and the Chui, Talas, and Issyk-Kul regions. By contrast, poverty rates in 2009 rose in the southern regions of Batken and Osh—perhaps because of these regions’ reliance on remittances (which fell by 15 percent in that year, for the country as a whole). The increase in national poverty levels reported for 2010 is apparent in call of Kyrgyzstan’s regions, except for Bishkek and Issyk-Kul. Particularly large increases in regional poverty rates were noted for Naryn, Talas, and Jalalabad. Due to their relatively large populations and below-average household income levels, about a quarter of Kyrgyzstan’s poor in 2008 lived in Osh, and another 19 percent lived in Jalalabad. The poverty implications of the global economic crisis may have been attenuated by their fiscal and social policy responses. The IMF data in Table 2.5 indicate that both governments increased the share of GDP devoted to household transfers since 2008. In 28

Nominal growth in household expenditures in 2009, divided by the change in the consumer price index in 2009.

Tajikistan, this share rose from 3.4 percent of GDP in 2008 to around 4.0 percent during 2009-2010; in Kyrgyzstan, it rose from 5.0 percent of GDP in 2008 to around 9.0 percent in 2010. These increases were complemented by growth in the share of GDP devoted to public employment expenditures—which, in light of low public-sector wages in these countries, can likewise have important implications for poverty reduction. However, the long-term impact of these increases, in terms of reducing poverty and household food and energy insecurity, may have been less than what the numbers suggest, for three reasons. First, the resources reflected in these shares of GDP may have been too small to significantly reduce poverty or protect vulnerable households from the impact of job losses, reductions in remittances, or rising food and energy prices. This may have particularly been the case for Tajikistan, whose share of GDP devoted to social protection and insurance is among the lowest in Europe and Central Asia. Second, it is not clear that social assistance in these countries is effectively directed to the most vulnerable households—a question that is explored at length below. Third, these spending increases may not be fiscally sustainable. This may particularly be the case for Kyrgyzstan where, despite strong donor support (grants are projected to average around 3 percent of GDP during 2008-2011), the general government budget balance (with grants treated as budget revenues) deteriorated from a surplus 1.6 percent of GDP in 2008 to a deficit of -6.4 percent of GDP in 2010. Kyrgyzstan’s state debt had therefore risen to 63 percent of GDP in 2010; it had been below 49 percent in 2008. As in Tajikistan, the public debt in Kyrgyzstan is limiting the government’s ability to borrow in order to finance energy infrastructure projects.

37


Poverty and Social Impact Assessment

Remittances and poverty The “Panorama” report is based on data collected from a survey administered to 1500 households in Tajikistan in November 2009, combined with the results of 24 focus group discussions. These data indicate that remittances are the largest income source for 25 percent of Tajikistan’s households (30 percent of rural households). Every third household in Tajikistan receives remittances (40 percent in rural areas, 20 percent in urban areas), and has at least one member who is working abroad. The shares of households receiving labour remittances were highest in Gorno Badakhshan (42 percent) and in the RRS (41 percent). These data indicate that reductions in remittance incomes in 2009 were particularly large for female-headed households—presumably because so many male spouses are working abroad.29 Fears of massive crisis-related returns of unemployed migrants to Tajikistan during 2009-2010 seem not to have been born out. While 44 percent of households interviewed in the Panorama survey indicated that labour migrants within their household were forced to return home during the September 2008 – September 2009 period, 14 percent were able to go back to their destination countries (chiefly Russia). This would put the number of returned migrants who remained at home at approximately 30 percent of the total. While these figures would mean that hundreds of thousands of (primarily) young unemployed men remained in Tajikistan as a result of the crisis, they would also be less than what was feared when the crisis was unfolding in 2009.

reported by the National Bank of Tajikistan/IMF during this time. These data suggest a close correspondence between labour migration and remittances in Tajikistan—as opposed to Armenia, for example, where transfers from the diaspora (as opposed to labour migrants) account for significant shares of remittance incomes. They also suggest that—despite the crisis—dollar incomes remitted per migrant worker remained relatively constant during this time.

Fears of massive crisis-related returns of unemployed migrants to Tajikistan during 2009-2010 seem not to have been borne out. If so, then the 27 percent growth in remittance inflows recorded during 2010 (relative to 2009) would suggest significant reductions in the numbers of “redundant” migrant workers languishing in Tajikistan. However, official data published in UNDP Tajikistan’s November 2010 Tajikistan Monthly Risk Monitoring and Warning Report indicate that the numbers of migrant workers leaving the country during the first ten months of the year had dropped by some 20 percent, compared to the same period of 2009.30

It is also interesting to note that the 30 percent reduction in the numbers of labour migrants that the survey found to be going abroad during September 2008 – September 2009 dovetails closely with the 35 percent reduction in remittance inflows into Tajikistan 29

The significant role of male migrants working abroad may explain why income poverty rates among femaleheaded households during the past decade have been lower than in male-headed households.

38

30

Data provided to this report by Russia’s Federal Migration Service point to a similar (18%) decline in the numbers of Tajikistani citizens entering the Russian Federation during the first eight months of 2010.


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Figure 2.9—Kyrgyzstan: Kyrgyzstan: Increases in external migration, remittance inflows (2010 2010)

316%

110%

101% 63% 29%

Remittances -national

Number of migrants -national

Number of migrants -Jalalabad oblast

Number of migrants --Osh oblast

Numbers of migrants -- Osh city

Relative to 2009. Source: State Statistical Committee data, UNDP calculations.

In Kyrgyzstan, official data indicate that the 15 percent decline in remittance inflows reported for 2009 corresponded to a 19 percent reduction in the number of migrant workers going abroad. By contrast, the 30 percent increase in remittances reported for 2010 corresponds to a 63 percent increase in the number of migrant workers reported leaving Kyrgyzstan. As the data in Figure 2.9 show, rates of out-migration migration from Kyrgyzstan’s southern regions (the centre of the June 2010 ethnic violence) were significantly tly above the national average, thanks to a surge of outmigration from these regions during the second half of the year. In contrast to Tajikistan (where national bank data indicate that remittances in 2010 remained well below 2008 levels), 2010 remittance inflows in Kyrgyzstan are reported to have exceeded 2008 levels. The Panorama data suggest that, despite its socio-economic economic importance, migration in Tajikistan rarely involves more than one worker per household. Only about a quarter of respondent households olds (or 8 percent of total households) send more than one member abroad to work. This pattern would seem to correspond to a certain intraintra household division of gender roles, whereby

men go abroad and women remain at home. This contrasts, for example, with Moldova, where an International Organization of Migration study found that 42 percent of Moldova’s migrants in 2006 were women. Such a figure suggests that both parents are working abroad in significant numbers of Moldovan households. The strains on famil families and children that are sometimes associated with migration may therefore be less in Tajikistan than in Moldova.31 The Panorama research also indicates that, while not as important as migration abroad, internal labour migration in Tajikistan is a significant icant coping mechanism. These data suggest that workers from every fourth rural household and every sixth urban household gain employment in Dushanbe and other cities (different from their original place of residence). Informal sector self self-employment likewise ise continues to be an important coping mechanism: sewing, embroidering, baking, and small-scale scale artisanry (typically conducted 31

See ee Mihail Peleah, “The Impact of Migration on Gender Roles in Moldova”, Development and Transition, no. 8, 2007 (http://www.developmentandtransition.net/index.cfm?m http://www.developmentandtransition.net/index.cfm?m odule=ActiveWeb&page=WebPage&DocumentID=664 ).

39


Poverty and Social Impact Assessment

by women) are reported in every fourth urban household. In Dushanbe, nearly 40 percent of respondent households indicated engagement in these activities. The Panorama data note a doubling of the numbers of households with incomes described as insufficient to purchase basic foodstuffs, as well as households unable to procure consumer durables and other more expensive purchases. Over 70 percent of households surveyed indicated that they typically buy only the cheapest foodstuffs available, and rarely eat meat. These data also suggest that household food production— either for own consumption or for cash sale— is not a panacea: only 7 percent of rural households indicated that income from cultivating their own farms was their primary income source. Moreover, 69 percent of the surveyed households receiving incomes from the sale of agricultural produce reported significant reductions in these earnings during the September 2008 – September 2009 period. Perhaps for this reason, only 12 percent of the surveyed households (and only 14 percent in rural areas) reported responding to the crisis by deciding to grow their own food. Difficulties in obtaining and managing land and other inputs on dehkhan farms32 may be responsible for this: only 21 percent of rural households indicated that dehkhan farming provided significant incomes. These data suggest that, despite some deregulation of the agricultural sector, additional land and agricultural reforms may be needed if land cultivation is to significantly increase food security among vulnerable households in rural areas of Tajikistan.

32

Although most land in Tajikistan remains stateowned, rural households are able to rent small land parcels from collective and state farms. In addition to growing cotton on these plots (often times under compulsory circumstances reminiscent of Soviet central planning), rural households use these parcels in order to grow food and tend livestock. These small-scale household agricultural activities, which also bear a certain resemblance to share-cropping, are referred to as dehkhan farms.

40

Household energy security and social policy—Kyrgyzstan The November 2009 USAID-funded study focused on the impact of the winter energy crisis on living standards in Kyrgyzstan. Although based on focus groups (and therefore not necessarily applying representative sampling techniques33), this study offers some important insights into these issues, including the following: Many of these respondents reported interruptions in electricity supplies even prior to the rationing introduced during the winters of 2008-2009 and 2009-2010. This finding suggests that—in Kyrgyzstan, as in Tajikistan—uninterrupted year-round household electricity supplies had been lost (or had never been achieved) even before the “winter crisis” took hold.34 Respondents pointed to poor management in the electricity sector, corruption,35 and excessive electricity exports as the main causes of these shortfalls. They also indicated that the quality and quantity of official information regarding the

33

The November 2009 study was based on interviews conducted earlier in December 2008 with 73 respondents (45 men, 28 women) in Chui, Jalalabad, Osh, and Talas oblasts, for whom electricity was the main energy source (i.e., for heating and cooking). All respondents were either household heads or, if not, were managing the household budget. The sample was in turn stratified into two groups of respondents—those who regularly pay their electricity bills, and those who don’t. 34 “The electricity was cut off without warning, and it was sometimes off for the whole day. The main reasons for previous cut-offs, according to the respondents, were accidental failures of the electric system.” Source: The Curtailment of Electricity Consumption: The Attitude of the Population and the Impact on the Quality of Life, USAID, Bishkek, November 2009, p. 4. 35 Reference here is to corruption among both the companies and households, some of which (according to respondents) resort to bribing meter readers in order to lower their reported electricity consumption (and therefore expenses). Incidents of illegal electricity connections—often with significant attendant health risks—were also reported.


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

causes of the inadequate.36

energy

shortages

were

Predictably, the surveyed households responded to these reductions in electricity supplies, and to higher energy prices, by reducing electricity usage (e.g., less frequent use of appliances,37 installation of makeshift weather stripping) and by increasing purchases of substitute fuels (e.g., coal,38 firewood, dung).39 Increases in hours worked (to raise household incomes to cover higher energy costs) were also reported. Reductions in purchases of food, clothing, health and education services, and other essentials in response to higher electricity tariffs were a common coping mechanism. These survey data indicate that the significance of the income effect associated with higher energy prices is greater than what is suggested in the official consumption expenditure data, according to which spending on energy, housing, and communal services during 20072008 averaged no more than 6 percent of total household budgets. The USAID-funded study on Kyrgyzstan Household Energy Analysis and Proposed Social Protection Measures (November 2008), and the World Bank���s May 2009 Social Safety Net in the Kyrgyz Republic: 36

In response, “The respondents would like to receive official information from the managers . . . and government officials by the means of live television broadcasting and at general meetings in their settlements” (page 7). 37 Respondents reported no use of refrigerators, or personal computers, during the winter. By contrast, virtually no changes in use of television sets and incandescent lights were reported. 38 According to official data, annual coal consumption in Kyrgyzstan rose some 52 percent during 2006-2010, while coal production rose 74 percent. The ecological consequences of burning Kyrgyzstan’s high-sulfur coal, the role of child labour in the coal sector, and perennial dangers from mine safety issues, suggest that these improvements in energy security may be something of a mixed blessing. 39 These options were typically not available to urban residents, living in apartment blocks and relying on central heating and electricity. Few respondents reported use of generators, which were viewed as expensive, inconvenient, and unreliable.

Capitalizing on Achievements and Addressing New Challenges, are based on official household budget survey data, and focus on linkages between energy tariffs/prices, consumption patterns, poverty, and social policy. Both analyze arguments for moving away from categorical benefits and subsidies—both for energy and in general—in favour of the poor family monthly benefit

Reductions in purchases of food, clothing, health and education services, and other essentials in response to higher energy tariffs were a common coping mechanism. (Kyrgyzstan’s only means-tested social protection instrument, paid to families with children40). The USAID study also focused on the implications of the significant energy tariff increases which were subsequently introduced in January 2010. This study argued (correctly, in hindsight) that, in the absence of: (i) improved management within the energy sector (to reduce extensive theft and losses);41 (ii) the credible threat of disconnection for non-payment; (iii) effective public communications concerning the need for, and low-cost options for responding to, higher tariffs; and (iv) consumer acceptance of the higher tariffs (inter alia in the form of growing arrears that reduce collection and broadened the wedge between nominal and effective tariffs), energy tariff increases would be likely to be ineffective.

40

Prior to 2011, this instrument was officially referred to as the (unified) monthly benefit. 41 The 2008 USAID study reports losses in distribution companies in the range of 31-44 percent, as of 2007. A 2010 USAID report notes significant reductions in losses (to around 25 percent) during 2008-2009, but questions the accuracy of these data and the sustainability of the trend they portray. The 2010 report also states that losses in the 7-10 percent range are a “standard benchmark” in other countries. Source: Electricity Loss Reduction Strategy for the Kyrgyz Power Sector, USAID, Bishkek, 25 March 2010 (revised 30 April 2010), pp. 11, 13.

41


Poverty and Social Impact Assessment

Prior to 2006, Kyrgyzstan’s social protection system was based on 38 types of categorical benefits (“privileges”) for 33 classes of beneficiaries. In addition to subsidies for energy and communal services, these benefits covered housing, transport, communications, health care, and vacations. Beneficiaries were eligible for 100, 50, or 25 percent price discounts up to certain administratively determined consumption levels. Differences between market prices/tariff levels and the out-of-pocket costs actually paid by beneficiaries were covered by the central government.

…only 6 percent of categorical benefits went to extremely poor households in 2005; 64 percent went to non-poor households.

monetization of these categorical energy (and other) benefits) may now be large complete.42 The USAID study found that only 6 percent of categorical (energy and other) benefits went to extremely poor households in 2005; 64 percent went to non-poor households. The mis-targeting for urban households was even more extreme: 5 percent of these benefits went to extremely poor households, while 77 percent went to non-poor households. In terms of energy subsidies, the May 2009 World Bank study found that only 16 percent of housing and utility subsidies (most of which were for energy) went to households in the poorest quintile of the income distribution. The USAID study likewise found that:

• •

The USAID report indicates that some 23 percent of households in Kyrgyzstan in 2008 (about 320,000 people) were recipients of various categorical benefits. Around 20 percent of households received energy subsidies (or discounts for privileged consumers), which were not means-tested. These accounted for 54 percent of all subsidies; electricity subsidies in turn accounted for 60 percent of this figure, or some 30 percent of total categorical benefits. Some 55 percent of Kyrgyzstan’s population received electricity subsidies, in the form of discounts for privileged consumers or supplemental incomes for pensioners. While most of these benefits were delivered in kind, the government since 2006 has sought to gradually monetize them. The steep tariff increases for electricity, heat, water, and some other communal services introduced in January 2010 were linked to this monetization process. Since most of these monetized benefits remained in place even after the April 2010 tariff rollbacks, the

42

92 percent of households receiving gas subsidies were non-poor; 88 percent of households receiving heating subsidies were non-poor; and 60 percent of households receiving electricity subsidies were non-poor.

These figures reflect the fact that fuel use is stratified by income levels and location (rural versus urban). The USAID study reports that 95-98 percent of rural households, and 65 percent of poor urban households, were heating with solid fuels in 2005. Solid fuels were also used for cooking in a majority of rural households, and in some 40 percent of poor urban households. Prior to the winter energy crises of 2008-2009 and 2009-2010, the remainder of Kyrgyzstan’s households relied on central and district heating (in urban areas),43 and on electric heat. Gas, like central 42

By contrast, the 2008 USAID study reports that only 33 percent of categorical energy benefits had been modified as of that year. 43 “Central heating is available in principle to around 13-14 percent of the Republic’s households, and is an urban phenomenon concentrated in and around Bishkek, Osh, and Jalalabad. There has been a steady decline in use of central heating among all urban income groups since 1999. Drop in urban demand for central heating by the extremely poor has been particularly dramatic, and we presume that switching to electricity was the alternative.” Source: Kyrgyzstan


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

and district heating, was only an option for urban households;44 the USAID study reports that gas for heating was only available to 6 percent of the rural population in 2005. However, 23 percent of households (primarily in urban areas) were using gas for cooking in 2005; gas penetration had pushed the share of urban non-poor households using electricity for cooking below 40 percent. Since then, however, sharp increases in wholesale and retail gas prices and disputes with UzbekTransGaz (the monopoly supplier) have reduced the attractiveness of gas in Kyrgyzstan.45 These data suggest that only about a third of poor urban households, and only 20 percent of rural households, were relying on electricity for heat in 2005. However, electricity use for heat (and cooking) seemed much more prevalent among—and therefore electricity tariff increases a greater burden for—poor households. The 2008 USAID study notes that “it is the urban poor and extremely poor that use electricity for heating the most”, and that “there has been a dramatic increase in electricity heating in the last 5-7 years”. It also finds that “if a poor household has no other option but [to] use electric heat, its share of expenditures on electricity will climb up to 1522 percent of the total expenditures”.46

Household Energy Analysis and Proposed Social Protection Measures, USAID, Bishkek, November 2008, p. 4. 44 “57 percent of all gas consumers reside in Bishkek” (in 2005), ibid.. 45 Whereas Kyrgyzstan imported gas from Uzbekistan at $100/1000m3 in 2007, the 2010 price has apparently been set at $220/1000m3. This is roughly in the middle range of world gas prices, which during January-May 2010 ranged from $170/1000m3 in the US to $281/1000m3 for Gazprom’s exports to Western Europe. (Source: IMF commodity price data, http://www.imf.org/external/np/res/commod/index.asp.) 46 By contrast, poor households with access to other heating sources were able to keep their electricity expenditures down to 3-5 percent of total incomes during this time. Non-poor urban households pushed expenditures on electricity down to 2 percent (or less) of total incomes. Source: op cit., pp. 5, 7.

The 2011 poverty and social impact assessment commissioned by UNDP found that somewhat lower shares of total household expenditures (6-7 percent) were devoted to energy spending during 2006-2009. These shares are generally within the range of what is viewed as “affordable” by international development agencies. And while energy

USAID research finds that gas for heating was only available to 6 percent of the rural population in 2005. expenditures absorbed a larger share of household budgets in poor families than in high-income families, the share of household spending devoted to energy expenditures in low-income households had converged towards national averages by the first half of 2010. However, this study also found that poor households rely predominantly on electricity and coal for heating purposes, and that poor households, and households in rural and mountainous areas, were much more likely to be affected by planned and unplanned electricity cutoffs than more well-to-do households in urban areas. These data indicate that electricity in Kyrgyzstan has certain “inferior good” properties, in that poorer households are most likely to rely on electricity for heat and cooking. This suggests that “bloc” or “lifeline” electricity tariffs (under which tariffs rise with greater amounts of electricity consumption) could be effective in mitigating the impact of higher tariffs on low-income households. On the other hand, the effectiveness of such tariffbased measures declines when poor households lose access to the electricity grid— as has been the case in both Tajikistan and Kyrgyzstan (at least, in the winter). In any case, both governments in recent years have moved toward unified electricity tariff structures, to reduce their complexity and increase transparency. 43


Poverty and Social Impact Assessment

Table 2.6—Kyrgyzstan’s social protection instruments: Efficiency and effectiveness (2005 data)

49

Share of GDP Coverage of poorest quintile Targeting accuracy50 Benefit generosity51 Poverty impact:52 • Poverty • Extreme poverty Cost efficiency:53 • Poverty • Extreme poverty

Monthly benefit 0.53% 28% 38% 7%

Monthly social benefit 0.21% 13% 33% 8%

Categorical benefits47 0.6% 16% 13% 1%

Pensions48 5.1% 56% 29% 25%

12% 3%

6% 6%

1% 1%

48% 33%

2 5

3 6

6 27

3 10

Source: Social Safety Net in the Kyrgyz Republic: Capitalizing on Achievements and Addressing New Challenges, World Bank, May 2009, p. 11.

Both the World Bank and USAID studies argue that the poor family monthly benefit is more effective in supporting poor households than categorical subsidies, both for energy and in general. According to World Bank calculations, 28 percent of households in the poorest quintile received the poor family monthly benefit in 2005, while only 16 percent received categorical benefits (Table 2.6). These data also indicate that it took 2 som of budget spending to reduce the poverty gap54 by 1 som (or 5 som of budget spending to reduce the extreme poverty gap by 1 som), compared to 6 and 27 som, respectively, of budget spending via categorical benefits. The poor family monthly benefit would therefore 47

Except for “share of GDP”, all data refer to categorical housing and utilities (including energy) subsidies. 48 Old age, disability, survivor pensions. 49 2007 data. 50 Share of benefits captured by poorest quintile. 51 Ratio of benefit to estimated beneficiaries’ consumption expenditures. 52 Relative reduction in the (extreme) poverty gap. 53 Nominal cost (to the state budget) of reducing the (extreme) poverty gap by one som. 54 The poverty gap is defined as the mean shortfall of the total population from the poverty line (where the non-poor have zero shortfall), expressed as a percentage of the poverty line. This measure, which reflects the depth of poverty as well as its incidence, can be described as a measure of the per-capita resources needed to eliminate poverty (or reduce the poor’s shortfall from the poverty line to zero) via perfectly targeted cash transfers.

44

seem to be a more efficient instrument, in terms of poverty-reduction impact per som of budget spending, than categorical benefits. Such comparisons explain why the government has sought (and donors have encouraged) the monetization of categorical benefits in general, and for energy in particular. They also suggest that social policy makers should seek to increase both the size of the poor family monthly benefit payments, as well as the numbers of poor family monthly benefit recipients. However, Table 2.7 shows, the opposite occurred: between 2007 and 2009 the number of poor family monthly benefit recipients dropped by nearly a quarter (from 475,000 to 362,000).55 When deflated by the consumer price index, the poor family monthly benefit’s value dropped by 23 percent in 2008, before recovering slightly (5 percent in 2009). The average poor family monthly benefit payment in 2008 was only 3-4 percent of the national minimum subsistence level. 55

UNICEF reports that, as of 2007 (before these declines in the numbers of recipients began), these benefits were only received by 8 percent of Kyrgyzstan’s households. Despite the country’s high poverty rate, less than 15 percent of Kyrgyzstan’s children were living in households that were receiving poor family monthly benefits. Source: UNICEFKyrgyzstan National Report on Child Poverty and Disparities, April 2010, p. 54.


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Table 2.7—Social policy instruments in Kyrgyzstan (2007-2009)

Pensions (overall): • Number of recipients • Average monthly pension (som) Old age pensions: • Number of recipients • Average monthly pension (som) Disability pensions: • Number of recipients • Average monthly pension (som) Survivors’ pensions: • Number of recipients • Average monthly pension (som) Real growth in monthly pension (annual average) Social fund balance (share of GDP) Poor family monthly benefit: • Number of recipients • Average monthly payment (som) • Real annual growth (deflated by CPI) Monthly social benefit:56 • Number of recipients • Average monthly payment (som) • Real annual growth (deflated by CPI) Relative to national monthly subsistence minimum: • Pension • Unified poor family monthly benefit • Monthly social benefit

2007

2008

2009

524,000 906

529,000 1120

569,000 1460

369,000 987

373,000 1223

412,000 1567

67,000 691

70,000 845

72,000 1082

88,000 320 11% 0.6%

86,000 469 3% 0.7%

85,000 576 9% n.a.

475,000 125 n.a.

434,000 120 -23%

362,000 135 5%

57,000 461 n.a.

59,000 656 14%

61,000 715 2%

32% 4% 16%

31% 3% 18%

45% 4% 22%

Sources: State Statistical Committee data; UNDP calculations.

This reflects the fact that, as the World Bank points out, the poor family monthly benefit “is a last resort program that complements a number of other, some of which are more generous, social protection programs, most notably pensions”.57 Moreover, the vast majority of poor family monthly benefit recipients—84 percent as of April 2010—are concentrated in the poorer southern regions of Osh, Jalalabad, and

Batken. This would mean that only 16 percent of poor family monthly benefit recipients were in Kyrgyzstan’s northern regions—where urbanization rates are higher, temperatures are colder, and heating needs greater.

56

The benefit is provided without means-testing, mostly to beneficiaries whose health or physical disabilities prevent them from working, 57 Source: Social Safety Net in the Kyrgyz Republic: Capitalizing on Achievements and Addressing New Challenges, World Bank, May 2009, p. 10.

45


Poverty and Social Impact Assessment

The UNICEF 2010 child poverty study notes that children are much less likely to live in poverty when at least one household member is receiving a pension. A comparison of the poor family monthly benefit vis-à-vis pension benefits in Kyrgyzstan (see Table 2.6) indicated that, in 2005, pensions were twice as effective as the poor family monthly benefit in reaching households in the poorest quintile (56 percent of these households received pension incomes, compared to 28 percent for the poor family monthly benefit). While not as accurate in targeting these households (29 percent targeting accuracy for pensions, compared to 38 percent for the poor family monthly benefit), pensions were more generous than the poor family monthly benefit, comprising 25 percent of incomes (compared to 7 percent for the poor family monthly benefit) in recipient households. As a result, the overall impact of pensions in reducing the poverty gap in 2005 was four times greater than the poor family monthly benefit’s (48 versus 12 percent), and ten times greater in terms of reducing the extreme poverty gap (33 versus 3 percent). The UNICEF 2010 child poverty study notes that children are much less likely to live in poverty when at least one household member is receiving a pension.58 The pension system’s greater apparent effectiveness was due largely to the fact that the fiscal resources devoted to pensions exceeded those devoted to the poor family monthly benefit by an order of magnitude (5.1 percent of GDP for pensions, versus 0.53 58

According to this research, the poverty rate for children in Kyrgyzstan in households receiving pension benefits in 2007 was only 43 percent, while in households without pension benefits the child poverty rate was 56 percent. For extreme poverty, these figures were 8 and 27 percent, respectively. Source: UNICEFKyrgyzstan, National Report on Child Poverty and Disparities, April 2010, p. 54.

46

percent of GDP for the poor family monthly benefit). Still, these data indicate that pensions’ efficiency as a poverty-reduction instrument was not noticeably worse than the poor family monthly benefit’s; whereas it took 2 som of poor family monthly benefits to reduce the poverty gap by one som in 2005, it only took 3 som to achieve this same result via pension benefits.59 Moreover, while the numbers of poor family monthly benefit recipients dropped by nearly a quarter and the real value of poor family monthly benefit payments collapsed during 2007-2009, the number of pensioners rose from 524,000 to 569,000 (a 9 percent increase), while the average monthly pension payment grew significantly in real terms during (see Table 2.7). The official data show that Kyrgyzstan’s social fund was consistently in surplus during the 2004-2008 period, which could be interpreted to mean that fiscal sustainability concerns need not be an issue. However, 2 billion som (about 1 percent of GDP) were apparently transferred from the state budget to the social fund in 2009; this transfer was to rise to 5 billion som in 2010. Transfers on this scale may not be fiscally sustainable.

In light of its rather specific nature and the relatively small numbers of beneficiaries, the monthly social benefit’s ability to reach vulnerable houselholds facing threats to energy security is necessarily limited. References to the monthly social benefit, which is provided without meanstesting mostly to beneficiaries whose health or physical disabilities prevent them from working, are often encountered in descriptions of social policy in Kyrgyzstan. Monthly social 59

The effectiveness ratio was somewhat less favourable for reductions in the extreme poverty gap: 10 som for pension benefits, as opposed to 5 som for the poor family monthly benefit.


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

benefit payments were increased in January 2010, to mitigate the social impact of the higher energy tariffs then introduced. However, in light of its rather specific nature and the relatively small numbers of beneficiaries (see Table 2.7), the monthly social benefit’s ability to reach vulnerable households facing threats to energy security is necessarily limited.60 These constraints are apparent in the relatively small share of GDP devoted to the monthly social benefit, and its relatively small coverage of Kyrgyzstan’s poorest households. However, the data in Table 2.6 indicate that—at least in 2005—the monthly social benefit was relatively costeffective in terms of reducing poverty. In sum, the case for cancelling categorical household energy subsidies and using the savings to finance increases in the poor family monthly benefit seems quite strong. As is explained in the next section, however, this is not the course of action chosen by policy makers in Kyrgyzstan.

Household energy security and social policy—Tajikistan The poverty and social impact assessment of Tajikistan’s energy sector commissioned by UNDP in 2011 found that the shares of household expenditures devoted to energy purchases averaged 10 percent over the course of 2009; for the poorest quintile this share was 16 percent. These figures are roughly double the relevant shares for Kyrgyzstan, and are closer to what are commonly regarded as the limits of household energy affordability. This study also found that—in contrast to Kyrgyzstan, where poor households rely predominantly on coal and electricity for heat—more than half the 60

UNICEF found that, in 2007, the child poverty rate in households receiving the monthly social benefit was 43 percent, while in households not receiving this benefit the child poverty rate was 46 percent. The comparable figures for extreme poverty were 8 and 14 percent. Source: National Report on Child Poverty and Disparities, issued in April 2010, p. 55.

population in Tajikistan in 2009 was relying on firewood and dung for heating. These data suggest that energy poverty concerns are particularly sharp during the winter—when an estimated one million people (particularly in rural areas) go for weeks without electricity, and when spending on solid fuel for heating can rise to close to half of total household expenditures. Despite comprising nearly three quarters of Tajikistan’s population, households in rural areas during 2008-2010 only accounted for 8-11 percent of the country’s electricity consumption.61 Moreover, the social protection system in Tajikistan seems particularly unable to protect vulnerable households from poverty in general, and rising energy prices in particular. A December 2010 World Bank study found that “Tajikistan’s system of social assistance exerts almost no downward influence on poverty rates”. This is primarily because of its small size: “the [2009] consolidated government budget for social assistance was small, at about 0.2 percent of GDP62—the lowest in . . . Europe and Central Asia.” The study also found that “the budget for social assistance is not well-targeted. Only 23 percent of social assistance payments reached the poorest quintile of the population.” As a result, “social assistance amounted to less than 3 percent of the per capita monthly consumption of the poorest 20 percent (that is, the lowest quintile) of the population, the lowest in Europe and Central Asia”. By contrast, “pensions . . . amounted to more than 15 percent of the per capita monthly 61

Morvaj, Z., et al., “Intermediate Strategy for Renewable Energy Sources”, UNDP-Tajikistan, 2010, p. 8; available (in draft form) at http://europeandcis.undp.org/senioreconomist/show/DF 091745-F203-1EE9-B8E083A9D6DBBC7D. 62 This figure does not include pensions (which averaged about 3 percent of GDP during 2006-2009), or other forms of social assistance, including compensation for household electricity and gas payments (0.17 percent of GDP), or cash transfers to low-income families that are conditional on school attendance by their children (0.07 percent of GDP). Still, even when these benefits are included, Tajikistan’s social protection system seems quite underfunded.

47


Poverty and Social Impact Assessment

consumption of that quintile”. Likewise, whereas “about 20 percent of the poorest quintile of households were covered by social assistance . . . almost 44 percent of the poorest quintile of households benefited from pensions”.63

A December 2010 World Bank study found that “Tajikistan’s system of social assistance exerts almost no downward influence on poverty rates.” The report also points to a series of institutional weaknesses in Tajikistan’s social assistance system, including:

The small size of payments made under Tajikistan two largest social benefit programmes (household electricity and gas compensation, and cash transfers to parents in low-income households that are conditional upon their children’s school attendance); Basing eligibility for receipt of household electricity and gas compensation on previous years’ budget allocations from the Ministry of Finance, rather than on information concerning recipient households’ 64 incomes or expenditures; Inadequate staffing: only one staff member in each district office is compensated for his/her role in maintaining lists of beneficiaries for these subsidies; and

Diversion by local authorities of funding for the cash transfers that are conditional on school attendance by children from low-income households.

In light of these problems, and with support from the World Bank and the European Commission, the government in 2011 introduced a pilot cash transfer scheme based on proxy means testing. The results of the pilot are to be assessed after its December 2012 completion, hopefully to be followed by a national roll out. While experience from similar countries suggests that this is likely the best option for Tajikistan, the implementation of this reform could run at least 4-5 years. As tariff increases are supposed to continue during this time, hardships for poor households could increase during this transition period. In sum, these new additions to the literature on household energy vulnerability suggest a number of conclusions: First, in both Tajikistan and Kyrgyzstan, universal household access to uninterrupted year-round electricity supplies had been lost, or had never been achieved, even before the “winter crisis” took hold. This underscores the importance of survey research to identify the numbers, location, and vulnerability characteristics of these households in the two countries. Second, the household budget survey data in Tajikistan support the view that higher energy tariffs can constitute a significant burden on low-income households— particularly for urban families that rely on electricity for heating and cooking, who have limited options for fuel switching. In Kyrgyzstan, such households would seem to represent at least 400,000 people.65

63

Tajikistan: Delivering Social Assistance to the Poorest Households, the World Bank, 30 December 2010, pp. 7-9. 64 The vast majority of funds under this programme, which constituted only 2 percent of total social spending (0.06 percent of GDP) in 2009, were devoted to purchases of energy-efficient light bulbs.

48

65

According to official data, the urban poverty rate in 2008 was 22.6 percent; 1,827,400 of Kyrgyzstan’s 5.3 million residents were classified as living in urban areas. This corresponds to 413,000 people. In light of the decline in final consumption reported for 2009 and the problems Kyrgyzstan has experienced in 2010, these


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Third, in Tajikistan, labour migration (primarily external, but also internal) has once again been shown to be the key coping mechanism for low-income (and food insecure) households. By contrast, the effectiveness of small-scale agricultural activities for income generation and food production for household consumption seems rather modest. As remittance inflows in 2010 seem to have returned to pre-crisis levels in both countries (particularly in Kyrgyzstan), this bodes well for vulnerable households’ prospects in the near future. Fourth, despite significant increases in social protection payments in 2010 in Kyrgyzstan, there is persuasive evidence that these payments did not effectively reach many of the households who needed them most. Apparently only 16 percent of poor family monthly benefit recipients in April 2010 were in Kyrgyzstan’s northern regions—where urbanization rates are higher, temperatures are colder, and heating needs are greater; and where poor urban households are more likely to rely on electricity for heating (and cooking). The relatively small size of these benefits, combined with the fact that other key social protection instruments (categorical benefits, the monthly social benefit, pensions) are not means-tested, indicates that most of these increased benefits are not reaching energyinsecure households. Small wonder, then, that the April 2010 protests over higher energy tariffs that unseated the Bakiyev government were centered in northern regions like Talas and Naryn.

magnify the impact of these tariff increases on vulnerable households. These results suggest that additional research may be needed to:

identify the numbers, location, and vulnerability characteristics of households whose welfare could be threatened by higher energy tariffs in the two countries;

combine this with compatible survey research on issues of access to, and affordability of, water, sanitation, and other communal services, in the two countries;

better understand the exact roles (particularly pertaining to household energy security) of the various social assistance instruments in the two countries;

monitor progress with social policy reform in Tajikistan, particular pertaining to household energy;

analyze the longer-term sustainability of pension systems in the two countries; and

analyze whether a return to bloc/lifeline tariffs might not prove to be a more effective (possibly temporary) response to the challenges posed by rising energy prices.

Fifth, Tajikistan’s social protection system likewise faces significant challenges in directing assistance toward vulnerable households. Since household electricity (and other energy and communal service) tariffs in Tajikistan are rising at least as quickly as in Kyrgyzstan, the absence of effective social protection measures in Tajikistan could numbers are unlikely to have declined significantly since 2008.

49


Poverty and Social Impact Assessment

50


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

Energy tariffs and social policy in Kyrgyzstan in 2010: What can happen when things go wrong... Non-traditional threats to vulnerable households, in the form of energy and water (and food) insecurity, are important not only in and of themselves, and not only because addressing them often requires unconventional efforts to bridge the gap between development programming and humanitarian response. These threats can themselves be symptoms of deeper, more fundamental political, social, and economic crises. And, they can provide the spark that transforms development shortcomings into humanitarian crises and armed conflict. The April-June 2010 events in Kyrgyzstan are, sadly, a confirmation of such concerns. While these developments have many causes, the dramatic increases in household electricity and heat tariffs introduced in January—which were part of the response to the energy crises of the past two winters, the mitigation of which by Kyrgyzstan’s social protection instruments was quite imperfect—were one of the precipitate causes of the popular uprising that overthrew President Kurmanbek Bakiyev in April. Subsequent developments featured the rollback of many (although not all) of these tariff hikes, the cancellation of efforts to privatize Kyrgyzstan’s electricity distribution companies, and the introduction of the Fuel and Energy Sector Transparency Initiative. The April 2010 events were followed by violent clashes in May-June between ethnic Uzbek and Kyrgyz in the southern regions of Osh and Jalalabad. Fatalities were estimated at 2000, and the numbers of refugees and internally displaced persons at close to 400,000. Significant damage was done to urban infrastructure in the south, and 5-10 percentage points were taken off of 2010 GDP. Enmities were created or deepened among communities that had lived together for

thousands of years, with possibly irreparable consequences for social capital and cohesion. In this respect, the consequences of the compound crisis in Kyrgyzstan were themselves “compounded”. What exactly happened with energy tariffs and policies in Kyrgyzstan in 2010? On 1 January, electricity and district heating tariffs were raised significantly (see Tables 8 and 9). Simultaneously, the government also increased the average pension payment (by 24 percent), public sector wages (by 200 som per month),66 and cash transfers under the poor family monthly benefit and the monthly social benefit programmes (by about 18 and 81 percent, respectively). In addition, categorical beneficiary groups were reduced from 38 to 25, and these benefits—including those for energy—were monetized. Beneficiaries instead became eligible to receive a monthly lump-sum payment, which varied in size between 1000 to 7000 som (depending on the beneficiary class).67 Residents of mountainous areas accounted for almost two thirds of these beneficiaries, which totaled 52,000 (10 percent of the population). People living with disabilities constituted another 20 percent, while the remainder consisted of war veterans (10 percent), law enforcement officials, military personnel, and a few other groups.

66

This constitutes a 3 percent increase, relative to the officially reported economy-wide average wage. 67 By contrast, the average national gross wage during the first quarter of 2010 was 6321 som.

51


Poverty and Social Impact Assessment

Table 2.8—Electrici Electricity tariff trends in Kyrgyzstan (per kWh, 2009-2010) 68

Users Households Industrial enterprises Budget funded institutions Agricultural users Pumping stations Other users

1 July 2009 In som In $ 0.71 $0.017 1.09 $0.026 1.14 $0.027 1.09 $0.026 0.77 $0.018 1.16 $0.028

1 January 2010 In som In $ 1.50 $.034 1.50 $.034 1.50 $.034 1.50 $.034 1.50 $.034 1.50 $.034

1 April 2010 In som In $ 0.7 $0.016 1.327 $0.029 1.327 $0.029 1.327 $0.029 1.327 $0.029 1.327 $0.029

1 July 2010* In som 1.90 1.90 1.90 1.90 1.90 1.90

* Since annulled. Source: Press reports, UNDP calculations. Table 2.9—District District heating tariff trends in Kyrgyzstan (per gcal, 2009-2010) 2009 2010)

2009 In som 254 254 254 254

User class Households Industrial enterprises Budget funded institutions Other users

In $ $5.9 $5.9 $5.9 $5.9

1 January 2010 In som In $ 1050 $23.7 1050 $23.7 1050 $23.7 1050 $23.7

1 April 2010 In som In $ 715 $15.8 929 $20.5 929 $20.5 929 $20.5

1 July 2010* In som 2500 2500 2500 2500

* Since annulled. Source: Press reports, UNDP calculations.

Figure 2.10—Distribution of social ocial benefits* in Kyrgyzstan, by household deciles eciles (2006 (2006-2010)

2006

60% 52%

2007

51%

50% 50%

2008

42% 43%

42%

38%

2009

30%

2010:H1

13%

11% 7% 6% 7%

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 10 (wealthiest)

* Pensions plus poor family monthly benefits, monthly social benefits, categorical benefits, benefits paid by enterprises. Source: State Statistical Committee, UNDP calculations. 68

Households consumed 38 percent of Kyrgyzstan’s electricity production in 2008. Industrial users accounted for 16 percent and net exports 6 percent, while agricultural users accounted for only 1 percent. Other users (the bulk of which were presumably presuma budget-funded funded institutions) consumed 8 percent of Kyrgyzstan’s electricity. The remainder (31 percent) was accounted as “losses”.

The household budget survey data shownn in Figure 2.10 indicate that, despite the years of attempts at improving the targeting of social protection in Kyrgyzstan, only about half of social benefits went to low-income


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

households (i.e., the lowest three deciles of the income distribution) during the first half of 2010.. By contrast, the share of social benefits accruing to upper-income income households more than doubled (from om 6 to 13 percent) during 2008-2010. 2010. Thus, the lion’s share of the increased social policy response to the crisis developments of 2009-2010—much much of which was financed by donors—seems seems to have leaked to relatively wealthy households.

framework. The share of categorical benefits—including including subsidies for energy and communal services—received received by low low-income households dropped from 46 percent in 2009 to only 20 percent during the first half of 2010. (This share hadd been at 65 percent in 2008.) By contrast, the share of categorical benefits accruing to upper-income income households rose from 11 to 38 percent during this time.

Figure 2.11—Distribution on of pension benefits in Kyrgyzstan, by household deciles (2008-2010) (2008

Figure 2.12—Distribution of categorical benefits in Kyrgyzstan, by household deciles (2008-2010)

2008 65%

53% 55% 48%

2009 2010:H1

38%

2008

46%

28% 23%

24% 17% 14%

2009

2010:H1

43% 42%

38%

31% 20% 11% 4%

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 9 (wealthiest)

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 (wealthiest)

Source: State Statistical Committee, UNDP calculations.

Source: State Statistical Committee, UNDP calculations.

Since pensions account for the bulk of social benefits paid out, this result is driven primarily by trends in pension spending. As the data in Figure 2.11 show, upper-income upper households benefitted handsomely from increases in pension spending during 20092009 2010: 10: the share of pension benefits accruing to relatively wealthy households rose from 23 percent in 2008 to 38 percent during the first half of 2010. By contrast, the share of pension benefits accruing to low-income income households dropped from 24 to 14 percent during this time. These data weaken the case for viewing pensions as a poverty-reduction reduction instrument in Kyrgyzstan.

A similar picture is apparent in the trends in monthly social benefit payments: the share of thesee benefits going to low-income low households dropped from m 57 percent in 2008 to only 12 percent during the first half of 2010. These trends strengthen the case for means-testing categorical and monthly social benefit programmes, and further question their rationale as poverty-reduction reduction instruments instruments.

The data in Figure 2.12 12 also indicate that the monetization of categorical benefits introduced in 2010 deepened the regressive character of Kyrgyzstan’s social policy

As suggested by the 2008 and 2009 USAID studies, tudies, the higher tariffs introduced in 2010 made a major contribution to a popular revolution. In sum, these data suggest that Kyrgyzstan’s social protection system has become less able 53


Poverty and Social Impact Assessment Table 2.10—Kyrgyzstan: Trends in energy, water production and supply (2007-2010)

Electricity: - Generation - Apparent consumption* Coal: - Production - Apparent consumption** Thermal power: - Generation - Apparent consumption*** Water collected and distributed

200769

2008

2009

2010

2011:H1

2%

6%

-21% -4%

-6% -3%

0% -6%

44% 26%

23% -2%

24% 27%

23% -5%

-8% 28%

54% 24%

0% -1% 0%

3% 5% -4%

-4% -3% -2%

-2% -1% -3%

7% n.a. 0%

* Generation less net exports less losses. 2011:H1 data do not reflect losses. ** Domestic production less net exports. *** Generation less losses. Source: State Statistical Committee, UNDP calculations.

to direct benefits to the most needy households during the past two years. This has implications for the system’s ability to protect vulnerable households from possible future increases in energy prices as well. The effectiveness of the social protection instruments that were intended to mitigate the impact of these steep tariff increases on vulnerable households proved unequal to the task. As suggested by the 2008 and 2009 USAID studies, when introduced without: (i) improved management within the energy sector (particularly to reduce theft and losses); (ii) the credible threat of disconnection for non-payment; (iii) effective public communications concerning the need for, and low-cost options for responding to, higher tariffs; (iv) consumer acceptance of the higher tariffs (inter alia in the form of growing arrears that reduced collection and broadened the wedge between nominal and effective tariffs); and (v) adequate targeting of social benefits, the higher tariffs made a major contribution to a popular revolution. On 20 April 2010, a provisional government decree reduced electricity, heating and hot water tariffs, reversing some (but not all) of these increases. The tariff increases that were to be introduced on 1 July were cancelled completely. The resulting tariff structure has some important implications, including the following: 54

69

Nominal tariffs went up, in some cases, substantially. Households after April 2010 were once again paying 2009 tariffs for electricity, but they were paying significantly higher tariffs for district heating and water than they did in 2009.70 Despite benefitting from moderate reductions in April 2010, commercial and budget-funded organizations also began paying higher tariffs for electricity and (especially) heat. As a result, household prices for electricity, gas, heat, and other fuels rose 14 percent in 2010, while household water tariffs rose 31 percent.

Increased cross-subsidization of households. The new tariff structure’s two-tier system (with a lower rate for household electricity consumption)

In 2007 electricity accounted for 39 percent of Kyrgyzstan’s total energy balance. Coal accounted for 21 percent and gas for 20 percent; the remainder was made up of petrol (13 percent), diesel (5 percent), and mazut (2 percent). 70 Household water tariffs were raised 37 percent in March. On the other hand, a presidential decree in late June reduced irrigated water tariffs during the summer of 2010. Source: "Временное правительство Кыргызстана обещает значительно снизить тарифы на поставку поливной воды с 1 июля 2010 года" (24.kg - 24.06.2010).


Chapter 2 – Recent developments in the Poverty/Energy/Vulnerability nexus in Kyrgyzstan and Tajikistan

effectively increased the crosssubsidization of household energy consumption. This should not be confused with a return to lifeline tariffs; households continue pay the same rates for electricity irrespective of their consumption levels.

A $150 million shortfall in energy sector revenues accrued, relative to what was anticipated at the start of 2010.71 The possible implication of this shortfall—which does not include revenues lost due to the cancellation or reconsideration of energy sector privatization projects—were underscored by a USAID study commissioned of late 2010/early 2011, which found that at least $190 million in new investments were needed to reduce risks of catastrophic failure in the electric power sector.72

How have energy and water output and supplies been affected? Despite the recovery in water levels along the Naryn cascade and the introduction of additional generation capacity at the Kambarata-2 hydropower plant, official data indicate that electricity generation remained flat in 2010 (see Table 2.10). Significant increases in electricity exports (to Kazakhstan) combined with a relatively mild autumn (reducing the demand for electric heat during the fourth quarter) apparently resulted in a decline in electricity consumption for the third consecutive year. Thermal power generation and consumption (for central and district heating), and water supplies to households, also continued to decline in 2010—despite ample precipitation. Fortunately, this lackluster performance was followed by a significant

turnaround in electricity (and coal) production and consumption during the first half of 2011. It remains to be seen whether this improvement can be sustained. Table 2.11—Kyrgyzstan: 2010 macroeconomic trends (preliminary data)

Change in GDP - Without Kumtor gold production Change in industrial output Change in gross agricultural output Change in construction output Change in retail sales volume Change in volume of tourism services Change in remittance inflows

-1.4% -2.1% 9.8% -2.8% -22.8% -6.8% -7.4% 30%

Source: State Statistical Committee.

These trends occurred against a backdrop of deteriorating macroeconomic indicators, reflecting the knock-on effects of the April-June events (Table 2.11). After posting 16 percent GDP growth in the first quarter of 2010,73 Kyrgyzstan reported a sharp economic downturn in the second and third quarters, before bottoming out in the last three months of the year. Preliminary official data indicate that GDP fell 1.4 percent in 2010, with construction, services (especially trade and tourism) and agriculture particularly hard hit. This decline in GDP made Kyrgyzstan the only country in the former Soviet Union not to have experienced economic growth in 2010. The recession also had important fiscal consequences: despite strong donor support (grants averaged around 3 percent of GDP during 2008-2011), the general government budget balance (with grants treated as budget revenues) deteriorated from a surplus 1.6 percent of GDP in 2008 to a deficit of -6.4 percent of GDP in 2010. Kyrgyzstan’s state debt had therefore risen to 63 percent of GDP in 2010; it had been below 49 percent in 2008. As in Tajikistan, the public debt in Kyrgyzstan

71

Joint Economic Assessment: Reconciliation, Recovery, Reconstruction, The World Bank, July 2010, p. 91. 72 “Highlights from the Executive Summaries of the Phase 1 Reports on the Management Diagnostic of NESK, EPP & and the Distribution Companies”, Bishkek, 15 April 2011.

73

Base effects may explain much of this high growth rate, as the Kumtor gold mining complex (which generates large shares of Kyrgyzstan’s industrial output, GDP, and export revenues) was not working in the first quarter of 2009.

55


Poverty and Social Impact Assessment

is limiting the government’s ability to borrow to finance energy infrastructure projects. Despite the increases in budget spending on social protection and publicsector employment, as well as the 30 percent growth in remittances (to $1.25 billion—an all-time high) reported in 2010, the national poverty rate rose by two full percentage points, from 31.7 to 33.7 percent of the population. This corresponded to an additional 128,000 people being pushed into poverty. Declining output and incomes, the sharp increases in food price inflation that took hold in the second half of the year, as well as the high annualized rates of energy price inflation, seem to have overwhelmed the impact of the increased budget spending and remittances for vulnerable households.

56

Conclusion Since the winter of 2007-2008, the governments of Kyrgyzstan and Tajikistan, and the international community, have clearly done a great deal to reduce national and household energy insecurities, and to strengthen the resiliency of vulnerable communities to the threats they pose. However, it is also clear that much still remains to be done. Moreover, the 2011 events in Kyrgyzstan remind us that unsuccessful efforts in this direction can have the most unfortunate side effects.


Chapter 3: Kyrgyzstan’s Energy Sector

Authors: Rafkat Hasanov, Kemal Izmailov Editor: Ben Slay

57


Poverty and Social Impact Assessment

Executive summary While the uprising that drove President Kurmanbek Bakiyev from power in April 2010 had many causes, energy issues were a prime contributor. Dramatic increases in power and heat tariffs introduced in January 2010, combined with two winters of electricity rationing, years of rapid growth in household energy costs, and concerns about corruption and mismanagement in the energy sector, were key drivers of tensions in Kyrgyzstan. This chapter explores the socio-economic background to these events, and their aftermath, by assessing the poverty and social implications of tariff increases and other policy reforms now being introduced (or considered) in Kyrgyzstan’s energy sector. Even before the developments of 2010, winter energy insecurities were afflicting significant numbers of households. The impact of the severe winter of 2007-2008 and the subsequent drought that reduced water levels in hydropower reservoirs along the Naryn cascade, the depreciation of the country’s power infrastructure, and the absence of effective market reforms in the energy sector, all contributed to these problems. This chapter analyzes recent trends in the electricity, thermal, gas, and coal sectors, as well as prospects for decentralized renewables in Kyrgyzstan. It focuses in particular on the commercial and regulatory characteristics limiting the attainment of full cost recovery tariffs, as well as on prospects for significant short- and medium-term improvements in management within these sectors. It notes that tariffs for electric and thermal power are being pulled in opposite directions by social acceptability and economic feasibility. Whereas natural gas tariffs depend on the price of imported natural gas, the state sets power and heat tariffs. In so doing, the state is guided primarily by social concerns—leaving many energy companies unprofitable, and complicating the provision of social assistance to vulnerable households. 58

These problems are exacerbated by the monopolistic structures found throughout the energy sector (with the partial exception of coal). The absence of competition and market stimuli creates preconditions for inefficiency and corruption. These problems are aggravated by high levels of fixed asset depreciation, particularly in the power and gas sectors. Modernizing energy production, transmission, and distribution in Kyrgyzstan will require billions of dollars in new investments—which may not be forthcoming at current tariff levels, and with the current regulatory environment. Despite having large coal reserves and reporting large increases in coal production during 2006-2009, imports continue to cover between half and two thirds of Kyrgyzstan’s coal needs. Likewise, virtually all of Kyrgyzstan’s natural gas is imported, from a single supplier (UzbekTransGaz). The government responded to the April 2010 events by introducing the Fuel and Energy Sector Transparency Initiative. FESTI represents an attempt to improve management and governance within the sector, by introducing greater measures of public participation and transparency—but without raising tariffs, or further privatizing energy sector assets, or significantly increasing the role of market forces. Compared to past policies, FESTI is an important step forward, especially in terms of reducing corruption. But at the same time, by focusing on reducing corruption in and improving the management of state-owned monopolies—rather than transforming them into market actors capable of modernizing the energy sector—FESTI is also a modest step forward. Official household survey data (stretching into 2010) indicate that the energy crisis that began in the winter of 2008 has decreased poor households’ access to electricity and other energy products and services. These households have also been affected more by interruptions in electricity


Chapter 3 - Kyrgyzstan’s Energy Sector

supplies. Prior to 2010, significant efforts had been invested in reforming Kyrgyzstan’s social protection system, in part to increase its ability to mitigate the impact of higher energy prices and tariffs on poor and vulnerable households. Unfortunately, there is little evidence to indicate that the social protection system provides these households with effective protection against higher energy costs. Instead, household survey data suggest that Kyrgyzstan’s social protection system became more regressive during 2008-2009, with growing shares of social benefits paid out to upper-income households. On the other hand, these data indicate that households— low-income and otherwise—devote relatively small shares of their budgets to energy productions and services.

tariffs for small volumes of household electricity consumption could be offset by higher tariffs for consumption above this level, thereby leaving average tariff levels unchanged. A number of important research questions are identified in this chapter. These pertain to:

Improvements in the quality of household survey and production/sales data regarding the energy sector, in order to remove inconsistencies within and between these data sets;

Developing possible scenarios for the future of Kyrgyzstan’s energy sector;

In light of the above, this chapter makes the following recommendations:

Improving corporate governance in the energy sector;

The relatively small shares of household budgets devoted to energy expenditures, the small likelihood that household electricity tariffs will be increased in the short (and possibly medium) term, and the difficulties in targeting social benefits to poor households—these factors weaken the case for more closely linking social and energy policies. Important changes would instead include:

Identifying appropriate energy saving technologies, and policies and programmes to accelerate their introduction;

Strengthening the role of affordability analyses in regulating energy tariff increases;

Analysis of obstacles to the accelerated development of small hydropower plants and other decentralized renewable energy technologies, with proposed solutions;

Analysis of the costs of electric and thermal power production and tariff setting; and Analysis of the results of the Fuel and Energy Sector Transparency Initiative.

More closely linking the poor family monthly benefit to the guaranteed minimum income, which should itself be closely linked to the minimum subsistence level; Means-testing the monthly social benefit and categorical benefits, to reduce their regressive character; and

Considering the reintroduction of lifeline electricity tariffs. Reductions in

59


Poverty and Social Impact Assessment

Sectoral overview Prior to the energy crisis that began in early 2008, Kyrgyzstan’s energy sector suffered from benign neglect. Market reforms had largely ended in the early years of the decade; state-owned owned monopolies continued to dominate the sector; attempts to attract pr private capital and foreign investment had met with limited success; and effective electricity tariffs were among the lowest in the region (see Figure 3.1). ). The focus was instead on state mobilization for the construction of new large hydro-power stations (such uch as Kambarata-1 Kambarata

generation and consumption (generation less net exports and losses) dropped by some 25 and 13 percent, respectively, during 2007 20072010. Gas consumption dropped by almost two thirds, as prices of gas imported from Uzbekistan istan rose from $100 to $240 per 1000 cubic meters during 2007-2009. 2009. Faced with growing shortages of centrally supplied electricity and unaffordable gas, many households, businesses, and public institutions switched to coal-fired fired boilers. Domestic coal production duction rose some 41 percent during 2007 20072010; apparent consumption (production less net exports) rose 57 percent. In addition to encouraging the “dash to

Figure 3.1—Effective household ousehold electricity tariffs in the former Soviet republics epublics (2007)

$0.09 Nominal tariff (per kWh) x collection rate

$0.07 $0.06 $0.05 $0.04

$0.04

$0.04 $0.02 $0.01

$0.01

Source: EBRD data, UNDP calculations.

and -22 on the Naryn cascade), and adding to electricity transmission capacity (e.g., via the construction of the Datka-Kemin Kemin high voltage power line). The energy crisis that began in 2008 was something of a wakeup call. As a result of cold temperatures and low water levels in the hydropower stations along the Naryn cascade, electricity blackouts increased abruptly, while sharply higher import prices made gas unaffordable for many households and businesses. These trends are apparent in Figure 3.2,, which show that electricity 60

coal”, the government responded to these developments by permitting energy prices and tariffs to risee at rates well above consumer price inflation (see Figure 3.33). It introduced a programme of anti-crisis crisis measures (including rotating blackouts and brownouts), raised electricity collection rates, reduced electricity losses,74 and accelerated the developme development of 74

Commercial and technical losses (in kWh), as well as accounts payable (in som) in the elec electricity sector declined during 2006-2009, 2009, while the ratio of accounts receivable to gross income in the electricity sector also fell.


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.2—Kyrgyzstan: Kyrgyzstan: Trends in energy production, consumption (2007-2010) 2010)

160 Electricity generation 140 Electricity consumption* 120 Thermal generation 100 2007 = 100

Thermal consumption**

80 Coal production 60 Coal consumption*** 40 Gas supply^ 20 Gas consumption~ 0 2007

2008

2009

Source: State Statistical Committee data, UNDP calculations * Generation less exports and losses. ** Generation less losses.

2010 *** Production less net exports. ^ Imports plus domestic production. ~ Gas supply less losses.

Figure 3.3—Kyrgyzstan: Kyrgyzstan: Household energy price inflation trends (2007-2010) 2010)

68%

2007 2008 2009 2010

53% 38% 28%

25% 10%

33% 17%

13%

12%

7% 8% 0%

Consumer prices

32%

1%

0%

Electricity tariffs

Gas tariffs

Heat tariffs

Annual average increases. Source: State Statistical Committee data, UNDP calculations.

small hydropower plants. The government also tried to reinvigorate privatization in the energy sector by putting stakes in six state-owned state energy companies up for sale. Last but not least, in January 2010 it doubled household electricity tariffs (from m $0.017 to $0.034 per

kWh), and quadrupled household thermal power tariffs (from $5.9 to $23.7 per gcal). These tariff hikes fed into the growing popular dissatisfaction with President Kurmanbek Bakiyev, who was driven from 61


Poverty and Social Impact Assessment

power by unrest and demonstrations in April 2010. Following Bakiyev’s departure, these large tariff hikes were scaled back or rescinded, lessening the impact on vulnerable households. Allegations of mismanagement and corruption led to the cancellation of the energy sector privatizations (see Box 3.1) as

well as to the introduction of the Fuel and Energy Sector Transparency Initiative. Nonetheless, despite this more “social” orientation of energy policy, household energy prices in 2010 continued to rise at rates well above consumer price inflation. 75

Box 3.1—Kyrgyzstan: Energy sector privatization chronology

Company

Severelektro Power Distribution Company

Vostokelektro Power Distribution Company

Oshelektro Power Distribution Company Jalalabadelektro Power Distribution Company Bishkek Combined Heat and Power Plant, and the Bishkek District Heating Distribution Company 75

Privatization process

* A single tender for the privatization of the state-owned stakes in Severelektro, the Bishkek Combined Heat and Power Plant, and the Bishkek District Heating Distribution Company was initiated in late 2008, with a starting price of $137 million. The tender was voided due to a lack of bids by the January 2009 deadline. * A second attempt to sell Severelektro separately was voided in July 2009, for the same reason. * A third attempt, without establishing a starting price, succeeded in December 2009, with the Chakan GES generating company announced the winner. Terms of sale included up-front payment of $3 million and capital investments of some $70 million over 10 years. Reportedly, no performance conditions were included in the tendering documents. * Following allegations of corruption and mismanagement and the events of April 2010, the government nationalized Chakan GES. * Two privatization attempts were declared void for lack of bids. The starting price was set at $41 million. * The third attempt, without establishing a starting price, succeeded in February 2010 with the same Chakan GES announced the winner. The terms of sale included payment of about $1.2 million and investments of $30 million over 10 years. * Following allegations of corruption and mismanagement and the events of April 2010, the government nationalized Chakan GES. * Two attempts of privatization were voided due to lack of bids. The initial starting price was set at $42 million. Two attempts of privatization were voided due to lack of bids. The initial starting price was set at $27 million.

* A single tender for the privatization of the state-owned stakes in Severelektro, the Bishkek Combined Heat and Power Plant, and the Bishkek District Heating Distribution Company was initiated in late 2008, with a starting price of $137 million. * The tender was voided due to a lack of bids by the January 2009 deadline. * No further attempts have been made.

Adapted from Joint Economic Assessment: Reconciliation, Recovery, and Reconstruction, World Bank, July 2010, p. 87.

62


Chapter 3 - Kyrgyzstan’s Energy Sector

Electricity Production, consumption, and losses. The vertically integrated Kyrgyzenergo was unbundled in 2001 into a single generation company, a single transmission company, and four distribution companies. All these companies are state-owned;76 competitive pressures are weak. Virtually all of Kyrgyzstan’s power generation assets belong to the “Power Plants” company, including 15 large hydropower plants, two combined power and thermal plants, and dozens of small-scale power producers.77 Total power generation capacity is 3,740 megawatts; the large hydropower plants account for 2,950 megawatts, while the combined power and thermal plants have a total capacity of 730 megawatts. The state-owned “National Power Grid” company manages the electricity transmission infrastructure, while four regional distribution companies (“Severelektro”, “Vostokelektro”, “Jalalabadelektro”, and “Oshelektro”) monopolize the supply of power in the areas for which they are responsible. The “Power Plants” generation company produces 99 percent of Kyrgyzstan’s electricity. Since more than 90 percent of this is generated by hydropower plants on the Naryn cascade with uniform hydrological conditions, market competition in power generation is difficult to imagine for the foreseeable future. Longer term, however, the construction of large coal-fired power plants (such as Kara-Keche), expansion of decentralized renewables, increased industrial co-generation, the introduction of more flexible management of the hydropower plants on the Naryn cascade, and the possible creation of a regional electricity market, could

offer prospects for competition in electricity generation. Prospects for market competition in electricity transmission are even more constrained, owing to significant associated economies of scale. The state-owned “National Power Grid” company, which now performs this function, seems likely to remain a natural monopoly indefinitely.

It is possible to imagine competition among distribution companies – particularly in urban areas that are located close to the borders of one of the existing zones… There is no distribution market competition among electricity distribution companies, each of which operates in its own territory; their territories do not overlap. While some 27 private wholesalers/small distributors were licensed in 2009 to purchase electricity from “Power Plants” and resell it, these wholesalers typically operate lower-voltage lines, and do not provide most users with an effective alternative to the regional distribution companies. It is possible to imagine competition among distribution companies— particularly in urban areas that are located close to the borders of one of the existing distribution zones, and particularly if the wholesalers are able to expand their activities. However, policy since April 2010 has emphasized strengthening state control over distribution (and other electricity) companies, in order to improve management and reduce corruption.

76

As of 2010, the Ministry of State Property held 80.5 percent equity stakes in “Power Plants” (generation), “Power Grid” (transmission), and in the electricity distribution companies. The Social Fund held another 13.2 percent of the shares in these companies. 77 Other power generating companies include Chakan HPP, Koshoi, Kalinin HPP Ltd., and Ark Ltd.

63


Poverty and Social Impact Assessment

Figure 3.4—Kyrgyzstan: Electricity production, consumption, losses, and exports (in million kWh, 2005-2010)

14891

14523

14830

Generation

Consumption

Losses

Exports

11789 11083 11070

7869

7551

7396

7299

7257

6862 4667

4582

4973 3693 2750 2662

2629 2460

2379

1579

552 1034

2005

2006

2007

2008

2009

2010

Source: State Statistical Committee.

It is not surprising that all the companies produced by Kyrgyzelektro’s 2001 unbundling have been included into the state Monopoly Register.78 Their tariffs are therefore subject to close scrutiny by the Antimonopoly Agency. On the other hand, the structural conditions now prevailing in the power sector—under which a single stateowned monopoly has been replaced by eight such companies, whose prospects for attracting the private capital needed for modernization are quite uncertain—now seem particularly unfortunate. Trends in electricity generation and use, and in the distribution of electricity generated, across consumption (and for households and 78

According to Order No. 524 of the State Agency for Anti-Monopoly Regulation, dated 30 December 2009.

64

other users), exports, and losses, are shown in Figure 3.4, and in Table 3.1. These data indicate that electricity losses have declined since 2005 which, along with reductions in the share of electricity going to exports, helped cushion households against the worst of the decline in electricity generation.79 Losses in Kyrgyzstan nonetheless remain quite high by international standards; losses in the 7-10 percent range are considered a “standard

79

Losses exist in all power systems. Technical losses occur in electricity generation and transmission from the point of generation to the point of final consumption; their size is determined by the distance over which the electricity is transmitted and the quality of transmission equipment. Commercial losses represent the difference between the price (tariff) of electricity supplied to the end user and the payment collected for its consumption.


Chapter 3 - Kyrgyzstan’s Energy Sector

Table 3.1—Kyrgyzstan: Kyrgyzstan: End uses of electricity generated (2005-2010)

2005 49% n.a. n.a. 33% 18%

Consumption - Households - Others Losses Exports

2006 51% n.a. n.a. 32% 17%

2007 53% 32% 21% 31% 16%

2008 64% 37% 28% 31% 5%

2009 66% 37% 29% 25% 9%

2010 62% n.a. n.a. 24% 14%

Source: State Statistical Committee, UNDP calculations.

benchmark”.80 Most (70 percent in 2010) of these losses occur at the distribution stage, due primarily to obsolete equipment, an absence of functioning of meters, inaccurate metering of consumed electricity, as well as outright theft. In Kyrgyzstan, neither the generation, transmission, nor distribution companies are responsible for cash management: payments from end-users rs are collected and accumulate in escrow accounts and divided among the companies on the basis of percentages set monthly by the Ministry of Energy’s En Regulatory Department. Barter and offsets are also used in settlements between sector entities, as well as with end-users. users. Trading figures covering the same electricity flows

reported by different utilities differ from each other.81 In light of these se complications, power company finances are likewise rather complicated. In general, these companies show rapid increases in both revenues and costs (in cash-flow flow terms), with cumulative growth in the latter (60 percent) outpacing the former (41 percent) during 2006 2006-2009. (By contrast, cumulative growth in the GDP deflator and producer price index during this time was 46 and 58 percent, respectively.) As a result, the financial results reported by power sector companies deteriorated sharply after 2006 (see Figure 3.5).

Figure 3.5—Kyrgyzstan: Financial results for power generation, distribution companies (in million som, 2006-2009) 2006

13406 Revenues

Costs

Profit (loss) 11192

11725

12067

9395 9603 8538 8359

179

-208 2006

2007

2008

-533

2009 -1339

Source: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010; UNDP calculations. The data da include small-scale enterprises.

80

Source: Electricity Loss Reduction Strategy for the Kyrgyz Power Sector,, USAID, Bishkek, 25 March 2010 (revised 30 April 2010), pp. 11, 13.

81

Reductions in electricity losses can be exaggerated, for example, when companies over--estimate billing and then accounts receivable.

65


Poverty and Social Impact Assessment

Figure 3.6—Kyrgyzstan: Kyrgyzstan: Collection rates in the electric power sector (2007-2009) 2009)

2007

2009

2008

104% 99% 95%

90%

90%

89%

86%

88%

90%

82% 70% 65%

Households

Overall

Industry

Agriculture

Source: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. The data include small small-scale enterprises. Figure 3.7—Kyrgyzstan: Kyrgyzstan: Electricity sector quasi-fiscal deficit (2002-2009)

12.8%

As a share of GDP 7.6%

5.9%

2002

2005

2006

4.9%

2007

3.7%

2008

3.9%

2009

Source: Ministry of Energy.

Available data indicate that power sector companies responded to these trends by trying to tighten their finances. For example, accounts payable in this sector declined in absolute terms during 2006-2009, 2009, despite large increases in costs and revenues. While Whil accounts receivable (more than half of which are doubtful or unrecoverable debts) in the power sector grew during 2006-2009, 2009, relative to sectoral revenues they declined. Households account for about 70 percent of power sector receivables. Collection ratess increased during this time (see Figure 3.6), ), while the power sector’s quasi-fiscal fiscal deficit continued to fall, dropping from nearly 13 percent of GDP to

66

under 4 percent during 2008--2009 (see Figure 3.7).82 These trends are occurring against a backdrop of electricity tariffs that are still quite low by regional standards (see Figure 3.1).

82

The quasi-fiscal deficit is the sum of: (1) “above “abovestandard” electricity losses; (2) deviations from a 100 percent cash collection rate; and (3) the difference between tariffs and long-run run marginal costs (including investment costs), measured on a cash cash-flow basis. The quasi-fiscal deficit eficit therefore reflects the difference between the income needed to fully cover operating and capital costs in the sector versus actual revenues received. According to Kyrgyzstan’s medium medium-term budget framework, this deficit is to drop below 2 percent of GDP by 2012.


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.8—Trends in electricity generation, and in water volumes at the Toktogul hydropower reservoir (20082010)

50%

Toktogul water volume (deviation from multi-year average*)

40% Year-on-year-change in electricity generated (kWh) 30% 20% 10% 0%

2008

2010

2009

-10% -20% -30% -40% -50% * Calculated relative to the average volume for that month in previous years, going back to 1991-1992. A zero value means water volume in that month was at its multi-year average (no deviation from normal). Sources: State Statistical Committee, CA WaterInfo website; UNDP calculations.

However, the data analyzed above suggest that the financial problems facing the power sector are not due solely, or perhaps even largely, to slow revenue growth because of “low tariffs”,83 or the power companies’ unwillingness/inability to collect tariffs or reduce electricity losses. They instead suggest that the key financial problem facing the power sector has been rapid growth in costs. While some of this growth results from large capital outlays for infrastructure investments, it may also reflect the inability of weak market forces or regulatory oversight to contain costs. Hydropower challenges. Electricity generation in Kyrgyzstan is dominated by hydropower, which provides more than 90 percent of total electricity output. Hydropower plants along the Naryn cascade, with installed capacity of 2870 megawatts, account for about 78 percent of Kyrgyzstan’s total generation capacity. The Toktogul power station, with 1200 megawatts of installed capacity, is 83

The reported 41 percent revenue growth combined with a 23 percent reported reduction in kWh sold during 2006-2009 suggests that effective electricity tariffs rose by 85 percent during these years.

Central Asia’s largest hydropower station, and only multi-year hydropower water storage facility. This reliance on hydropower leaves Kyrgyzstan vulnerable to changes in water levels along the Naryn cascade. This was particularly apparent in 2008, when drought conditions helped push water volumes at Toktogul to extremely low levels (see Figure 3.8). The volume of electricity generated fell 21 percent in that year, and another 6 percent in 2009 as releases were limited by the need to restore water levels. However, because Toktogul is a multi-year storage facility, good management of the water in its reservoir should provide some protection against droughts in the future.84

84

According to data on the CA WaterInfo website (http://www.cawater-info.net/analysis/index_e.htm), water inflows into Toktogul in 2007 were also well below historical averages. However, state statistical committee data indicate that electricity generation rose 2 percent that year, while exports and losses absorbed almost half of the electricity produced (see Table 3.1). This has given rise to claims that mismanagement, rather than drought, caused the sharp drops in water levels at Toktogul in 2007-2008, and then electricity shortages during subsequent winters.

67


Poverty and Social Impact Assessment

Investment projects and prospects. According to various estimates, Kyrgyzstan is not using more than 10 percent of its total hydropower capacity, which is assessed at 140 billion kWh by the “National Power Grid” company.85 The combination of abundant water resources and reliance on hydro power poses a dilemma for policy makers in Kyrgyzstan. The construction of new hydro power plants—both along the Naryn cascade and on smaller rivers—is an obvious way to increase capacity for winter power generation, as well as boost exports and promote economic development. This is the vision of the Central Asia and South Asia Regional Energy Market (CASAREM) project, which is supported by the donor community and the government. The Ministry of Energy has conducted feasibility studies for constructing some 47 hydropower plants across the country.

70 million m3. The construction is being financed by the government budget and an earmarked credit from the Russian 86 Federation. Kambarata-1 would be significantly larger and more expensive: installed capacity of 1,900 megawatts (4 units of 475 megawatts each); power generation of 5,088 million kWh; a water reservoir capacity of 4,650 million m3; and construction costs of some $1.7 billion.

However, prospects for virtually all investment projects in the energy sector are constrained by Kyrgyzstan’s relatively low electricity tariffs. In addition to reducing the cash flow power companies need to finance investments within the sector directly, low electricity tariffs reduce the commercial feasibility of energy projects in other sectors. This applies both to coal (the demand for which is determined in part by the financial situation in the power sector) and to decentralized renewables, the payback periods for which shorten as electricity tariffs increase.

The construction of the Kara-Keche coal-fired power station, with a capacity of at least 1200 megawatts, is also under discussion. This plant would generate electricity from coal that would be mined from the Kavaksky lignite basin, where it would be located. In addition to helping to develop the coal industry, Kara-Keche would diversify Kyrgyzstan’s generation capacity away from its near-total reliance on hydropower. However, the sources of financing for this $1.3 billion project have not yet been identified.

The Kambarata-1 and -2 hydropower projects are at present receiving the greatest attention and assistance from the government. Kambarata-2, the costs of which were estimated at some $100 million in 2007, is under active construction—the first unit was installed in 2010, with construction to be completed by 2015. Key parameters include: installed capacity—360 megawatts (three units of 120 megawatts each); power generation— 1,148 million kWh; water reservoir capacity—

Kyrgyzstan’s topography essentially divides the power transmission network into northern and southern parts, which are linked by a high-voltage transmission line running from Toktogul in the south to Frunzenskaya in the north. Because the Naryn cascade is in the south, Kyrgyzstan’s southern regions have a power surplus, while the north is in deficit.

In addition to reducing the cash flow that power companies need to finance investments within the sector directly, low electricity tariffs reduce the commercial feasibility of energy projects in other sectors.

86 85

Source: Small and Medium Sized Hydropower Development Programme, approved by presidential decree N 365, 14 October 2008.

68

The Agreement between the Government of the Kyrgyz Republic and the Government of the Russian Federation as of February 3, 2009 "On Construction of Kambarata HPP-1".


Chapter 3 - Kyrgyzstan’s Energy Sector

Map 3.1—Existing and planned hydropower plants and high-voltage transmission lines

“Kemin-Almaty” High-voltage transmission line

Kara-Keche “Datka-Kemin” highvoltage transmission line ($343 million)

“Datka” transmission line ($229 million)

“Aigultash-Samat” high-voltage transmission line ($12.9 million)

Source: Ministry of Energy.

However, some parts of the south—such as Leilek in the western part of the Batken region—face power shortages (especially in the winter time) because they are under-served by the existing transmission infrastructure. This problem is being addressed by the construction of the 131 kilometer AigultashSamat transmission line, which is to be completed in November of 2011. This project is financed by a $12 million credit from the Islamic Development Bank, as well as $900,000 from the “National Power Grid” transmission company. Although Kyrgyzstan imports virtually no electricity, the Toktogul-Frunzenskaya transmission line runs through Uzbekistan and Kazakhstan (see Map 3.1). Uzbekistan’s and Kazakhstan’s November 2009 decision to withdraw from—and therefore the de facto dissolution of—the integrated Central Asian electricity transmission grid adds new potential uncertainty to power supplies in Kyrgyzstan’s northern regions. This uncertainty is to be addressed by the construction of the $570 million Datka-Kemin transmission line, the construction of which is expected to be co-financed by China’s ExportImport Bank, the Asian Development Bank,

and others. However, this financing was put on hold following the April 2010 developments; it has not yet been secured. A 1998 government resolution requires the distribution companies to install reliable and safe meters at each service point, and to take regular meter readings. Virtually all electricity meters have been transferred to the balance-sheet of the distribution companies, who must also finance the installation of new meters for households. Thus, all responsibility for electricity metering is placed on distribution enterprises. Offering reliable electricity services to growing numbers of migrant households—many of which have an informal character—located in peri-urban areas poses significant problems for distribution companies. This is the case for Severelektro, which serves Bishkek city. As a result, since 2006 a number of projects (financed by the World Bank, KFW, and the Swiss Economic Cooperation Organization) have helped the distribution companies to improve metering, reduce losses, and extend services to new households. Procurement of meters and computer hardware and software (for billing systems) has played a 69


Poverty and Social Impact Assessment

large role in these projects, which have helped reduce losses and boost cash collection rates. Total financing for electricity sector projects, from international organizations and from the state budget, amounts to about $98 million. This is far short of what is necessary: the Joint Economic Assessment published by the World Bank in mid-2010 assessed the immediate needs of the energy sector (i.e., for the winter of 2010-2011) at $180 million. Out of this amount $124 million was to be allocated to essential and critical repairs for heating plants and district heating systems and to ensure security in electricity generation.87 The funding needed for the generation and transmission projects mentioned above (e.g., Kambarata-1 and -2, Kemin-Datka, KaraKeche) is estimated in the $5-6 billion range. It is not clear when, whether, or from whom these funds could be obtained.

Thermal power Two combined thermal and electric power plants are in operation in Bishkek and Osh cities, covering 85 and 35-40 percent of the households in these two locations, respectively. In addition to these cities, central heating systems are in place in the towns of Kyzyl-Kiya (covering 60 percent) and Karakol (covering 26 percent).88 Small-scale thermal generation enterprises operate in many smaller towns; and residents of stand-alone houses often have their own thermal generating equipment (stoves, boilers, heaters, etc.). Thermal power in Bishkek (most of which is generated by the Bishkek Combined Heat and Power Plant) is delivered by the Bishkekteploset’ and Bishkekteploenergo enterprises; in Osh it is supplied by the Osh combined heating and power plant. Thermal power in other urban areas is also produced by 87

For more see, “The Kyrgyz Republic Joint Economic Assessment: Reconciliation, Recovery, and Reconstruction”, ADB, IMF, the World Bank, 2010. 88 Source: Country Development Strategy for 20082011.

70

the Kyrgyzzhilkommunsoyuz enterprise. All of these distribution companies are stateowned: Bishkekteploset’ takes the form of a joint-stock company in which 84 percent of the shares are state-owned; Bishkekteploenergo is a municipal enterprise

There can be little doubt, for example, that the use of coalfired boilers and firewood has limited the strains on vulnerable households presented by large increases in electricity, gas, and thermal power tariffs since 2007. under the Bishkek mayor’s office; while Kyrgyzzhilkommunsoyuz belongs to the Ministry of Energy. There is no competition in distribution: each enterprise has its own nonoverlapping territory of operations. As a result, Bishkekteploset’, Bishkekteploenergo, and the territorial divisions of Kyrgyzzhilkommunsoyuz have been included into the State Monopoly Register; their tariffs are subject to close inspection by the Antimonopoly Agency. While important economies of scale are associated with the creation and maintenance of thermal infrastructure, significant inter-modal competitive forces are present, particularly at the retail level. Consumers unhappy with district heating tariffs or service quality can switch to electric heaters, gas- or coal-fired boilers, or firewood. There can be little doubt, for example, that the ability to install coal-fired boilers (and, in rural areas, burn firewood) has limited the strains on vulnerable households presented by the large increases in electricity, gas, and thermal power tariffs since 2007. Exclusive treatment of thermal power as a natural monopoly whose tariffs need to be kept low could distort competition among fuel sources—competition that can help limit costs and price increases


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.9—Kyrgyzstan: Thermal power production, p consumption, losses, and exports (in thousand gigacalories, 2006-2010)

Generation 2933

3015

2926 2185

2165

2007

Losses 2887

2270

2815 2209

745

761

748

2006

Consumption

2008

2184

678

2009

631

2010

Source: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. The data include small small-scale enterprises.

while also promoting the development of Kyrgyzstan’s energy sector. Production, consumption, and losses. Kyrgyzstan’s thermal power facilities generate about 3 million gigacalories of thermal power annually (see Figure 3.9), ), two thirds of which come from the Bishkek Combined Heat and Power Plant. Kyrgyzzhilkommunsoyuz and the Osh Combined Heat and Power Pow Plant generate about 540,000 and 120,000 gigacalories annually, respectively. As in the electricity sector, losses are endemic in thermal power, generally exceeding 20 percent of thermal output. However, thanks to reductions in losses from 25 to 22 percent perc during 2008-2010, 2010, the 4 percent cumulative decline in consumption reported for these years was less than the 7 percent decline in production. These high losses are due primarily to infrastructure depreciation and made worse by frequent service interruptions, ptions, which lead to high levels of condensation. The depreciation of fixed assets in Kyrgyzstan’s thermal power

sector is estimated at 60 percent.89 Thus, according to the Ministry of Energy, the number of accidents in the OJSC Bishkekteploset’ in the first st month of the 2009-2010 heating season increased by 60 percent as compared to 2007 2007-2008. This is particularly an issue for Bishkektelposet’, where losses increased to 48 percent in 2009. Thermal sector finances. The financials of the enterprises supplying steam and hot water in recent years have been even less favorable thann in the power sector (see Figure 3.10). ). In 2008 some 588 million som ($16.1 million) in losses were reported reported—40 percent of sectoral revenues—before before dropping back to 343 million som ($8 million) in 2009. As is the case in the electricity sector, thermal power companies reported very rapid growth in costs during 2006-2009—88 88 percent, while the sector’s accounts payable also rose by 66 percent. (By way of comparison the industrial producer price index reported a 58 percent cumulative increase during this time; the consumer price index reported a 47 percent cumulative increase; while the GDP deflator 89

Source: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. Data on small-scale scale enterprises are not included.

71


Poverty and Social Impact Assessment

rose 46 percent). While much-needed much investments to refurbish Kyrgyzstan’s decaying thermal hermal infrastructure can no doubt explain much of this increase, they may not be able to explain all of it. The rapid growth in accounts payable also contrasts with the electric power companies’ success in sharply reducing their payables (in real terms) during d 2006-2009. 2009. On the other hand, thermal power companies’ accounts ccounts receivable only grew by 13 percent during 2006-2009 2009 (a decline in real terms). While doubtful or unrecoverable household debts grew rapidly during this time, they comprised less than 10 percent of total receivables in 2009.90

2010, Kyrgyzzhilkommunsoyuz only cove covered about 37 percent of its costs from sales last year; the difference was financed by state budget, growth in accounts payable, fiscal arrears,, and the like. The 2011 state budget earmarks 740 million som ($15 million) for Kyrgyzzhilkommunsoyuz alone. The situation is similar with Bishkekteploenergo, which is subsidized by the Bishkek city budget.91 Investment projects and prospects. Some $110 million in projects to reconstruct and modernize Kyrgyzstan’s central heating system, co-funded funded by the Asian Deve Development Bank, the World Bank, and other international

Figure 3.10—Kyrgyzstan: Financial results for thermal power companies (in million som, 2006 2006-2009)

Revenues

2408

Costs

2042

2065

Profit (loss)

1530 1280 949

2006

-331

1454

1119

2007

2008 -411

2009 -588

-343

Source: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010; and UNDP calculations. The data include small-scale enterprises.

Revenues reported by thermal power companies increased even faster— —by 118 percent—during 2006-2009. 2009. However, because thermal power tariffs only cover about 45 percent of service costs, the bulk of these revenues come not from users, but from state and local budgets. In 2009, state budget subsidies for communal service providers reached 1 billion som ($23 million). For example, despite reporting an improvement in

organizations (as well as the state budget) are currently being implemented. These projects focus on modernizing the Bishkek and Osh combined heat and powerr plants, replacing inefficient boilers, and in installing thermal meters (in apartment buildings). Higher prices for heat and hot water (see Figure 3.3) have led many households to install meters on their own initiative. 91

90

By contrast, doubtful or unrecoverable househ household debts were 57 percent of electric power company receivables in 2009.

72

By contrast, Bishkekteploset’ does not receive budget subsidies. Instead, it is cross-subsidized subsidized from revenues earned via electricity exports by the “Power Plants” transmission company.


Chapter 3 - Kyrgyzstan’s Energy Sector

are present on the retail gas market: households can substitute electricity or coal coalfired boilers (and, in urban areas, district heating) for gas heat; gas-fired fired appliances can likewise be replaced by electric ones. one Strengthening competition among these fuel sources can help hold cost, tariffs and prices down.

Gas virtually all of which is Natural gas—virtually imported from Uzbekistan—accounts accounts for 30 percent of total energy consumption in Kyrgyzstan. Gas import, transportation, distribution, and sales are handled by the Kyrgyzgaz state-owned owned monopoly,92 whose assets include some 800 kilometers of trunk pipelines and some 2300 kilometers retail pipeline infrastructure.

Imports, consumption, and losses. As the data in Figure 3.11 show, gas consumption in Kyrgyzstan dropped precipitously during 2007-2009, 2009, as the price of gas imported from Uzbekistan rose from $100 to $240 per thousand cubic meters. Preliminary data indicate that, while the import price dropped slightly in 2010, imports and consumption continued to decline. The cumulative decline in gas consumption during 2007 2007-2010 seems to have been on the order of 68 percent, with particularly large drops reported in the industrial sector. Although household gas prices “only” rose by 12 percent last year, 2010 was the seventh consecutive year in which ch household gas prices ros rose faster than consumer price inflation.

There are 279,358 gas consumers, most of which (277,344) 277,344) are households (approximately 1 million individuals); the remainder are utilities, industrial, and commercial entities (2,145 users); and budgetbudget financed organizations (129). About 97 percent of household gas users have meters. meters As it is the only major company involved in the purchase, transport, and distribution of gas in Kyrgyzstan, it is not surprising that Kyrgyzgaz has been included in the monopoly register, and its tariffs subjected to scrutiny by the Antimonopoly Agency. However, wever, as with thermal, competitive forces

Figure 3.11—Kyrgyzstan: Gas supply, supply consumption, and losses (in thousand meters3, 2006 2006-2010)

Imports + production

783

769

762 671

644

Consumption

648

Losses

340 294

247 106

2006

97

2007

97

2008

214 80

77

2009

2010*

* 2010 data for consumption, losses are UNDP estimates. Source: State Statistical Committee. 92

As of 2010, the Ministry of State Property held an 87.9 percent equity stakes in Kyrgyzgaz; the Social Fund held another 5.4 percent.

73


Poverty and Social Impact Assessment

Losses in the gas sector are significant: in 2009 they comprised 23 percent of total gas supply (imports plus domestic production), up from 12-14 14 percent in previous years. These losses reflect the debilitated condition condi of Kyrgyzstan’s gas pipelines, inaccurate metering, theft, and other factors. The highest losses are recorded in the country’s southern regions. Gas sector financials. Kyrgyzgaz’s financial performance is determined primarily by trends in wholesale (imported) and retail gas prices, the procurement of which represents some three quarters of its total expenditures. As the data in Figures 3.12 and 3.13 show, Kyrgyzgaz’s finances deteriorated

sharply during 2007-2009, 2009, when the price of gas imported from Uzbekistan rose from $100 to $240 per thousand cubic meters. Although retail gas prices also increased sharply (relative to consumer prices) during this time, they were unable to keep pace with import prices. Exchange-rate rate effects further added to these losses, sses, as these gas purchases are priced in dollars, and the som depreciated in nominal terms vis-à-vis vis the dollar by some 26 percent during 2008-2010. 2010. Fortunately, UzbekTransGaz did not raise its export prices in 2010, allowing domestic gas prices (which rose ose another 12 percent last year) to partially make up the lost ground. This suggests that a lower loss figure will be reported for 2010.

Figure 3.12—Kyrgyzstan: Gas sector financials (2007-2009, 2009, in million som)

Gross income

Figure 3.13—Kyrgyzstan: Gas, consumer price inflation trends (2007--2010)

250 Gas import prices

4404 4532

Costs

Household gas tariffs

Profit (loss)

3789 3320

3070 3045

Consumer price index 200

150 25 2007 = 100

2007

2008

-128

2009 -469

100 2007

2008

2009

Sources: State Statistical Committee, especially Finances of Enterprises of the Kyrgyz Republic, Bishkek, 2010; and UNDP calculations.

74

2010


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.14—Kyrgyzstan: Fixed assets (by book value) in the gas sector (2006-2009, 2009, in million som)

793

Figure 3.15—Kyrgyzstan: Trends in fixed asset depreciation, cash collections (2006 (2006-2009)

100%

733

80% Share of depreciated fixed assets 496

60% Cash collections

393

40%

20%

0% 2006

2007

2008

2009

2006

2007

2008

2009

Sources: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010; and UNDP calculations.

Despite these losses, Kyrgyzgaz and its subsidiaries have managed to make significant investments in Kyrgyzstan’s gas infrastructure. As the data in Figures 3.14 and 3.15 show,, the book value of fixed assets doubled during 2006-2009, 2009, while the share of worn-out out fixed assets dropped from 40 to 28 percent. Kyrgyzgaz’s high rate (at least 90 percent) of cash collections helped finance these capital expenditures. Investment projects cts and prospects. Investment projects worth some $95 million are currently underway in the gas sector—a sector figure that is five times the book value of the gas sector’s fixed assets. Of this sum, $6 million reflects expenditures to repair damage to the gas distribution stribution network in Osh following the June 2010 events.93 These investments are mainly to replace unreliable gas pipelines and construct new ones, and to improve metering at gas distribution stations.

93

For more see, “The Kyrgyz Republic Joint Economic Assessment: Reconciliation, Recovery, and Reconstruction”, ADB,, IMF, the World Bank, 2010.

Coal Kyrgyzstan’s coal industry hosts 30 mining companies, which include both privately owned and joint stock companies with significant state participation. However, the largest coal mining enterprise is the state stateowned Kyrgyzkomur, under the Ministry of Energy. As of January 2008 proven reserves at Kyrgyzstan’s yrgyzstan’s 70 main coal deposits were reported as 1.3 billion tons; prospective reserves are assessed at than 2.2 billion tons, making Kyrgyzstan potentially one of Central Asia’s coal sector leaders. Deposits are concentrated in four coal basins (Southern (Souther Fergana, Northern Fergana, Uzgen, and Kavak) and three smaller coal coal-bearing areas (Alai, Alabuka-Chatyr-Kol, Kol, and Southern Issyk-Kul). Kul). Both underground and strip mining extraction technologies are used. While there is no leading national coal company, transport ransport costs segment the national market, so that a number of coal coal-mining enterprises have quasi-monopolistic monopolistic regional positions. The monopoly register for 2010 therefore included coal-mining mining enterprises from Batken, Naryn, and Osh regions, whose wholesale le prices were regulated by the state. However, these sellers often face 75


Poverty and Social Impact Assessment

monopsonistic coal buyers (e.g., thermal power plants, municipal enterprises), as well as some large intermediary wholesalers, who resell coal to households and other small-scale final users. A more competitive picture therefore emerges, particularly when Kyrgyzstan’s large coal imports are taken into account. Still, it is not clear whether this combination of bilateral monopolies and wholesale price controls necessarily reduces coal prices for consumers, most of who purchase coal on the open market. It may be that the wholesale price controls discourage domestic production in those regions where import competition is feasible—and drive up retail coal prices in those regions in which high transport costs limit the penetration of coal imports.

607,000 tons in 2009, before dropping back to some 558,000 tons last year. The electricity shortages and sky-rocketing gas and heat prices that took hold during these years obviously contributed to this resurgence. These factors helped wholesale coal prices rise by some 150 percent during 2006-2010. However, coal imports, which exceed domestic production, also soared during this time, rising from 728,000 tons in 2007 to 1.1 million tons in 2010 (Figure 3.16). Thermal plants are the largest users of coal: in 2009, for example, the Bishkek Combined Heating and Power Plant consumed 796,000 tons of coal, as well as 24 million cubic meters of natural gas and 69,000 tons of fuel oil. Due to its technological requirements, the Bishkek plant has to use imported coal— accounting for virtually all of Kyrgyzstan’s coal imports. Significant amounts of coal are burned by families living in individual houses, which are not connected to district heating networks.

Coal production and trade. Coal production in Kyrgyzstan, which reached a high of 4.5 million tons in 1979, had dropped to only 321,000 tons in 2006. Output nearly doubled during the next three years, rising to

Figure 3.16—Kyrgyzstan: Coal production, consumption, imports and exports (in thousand tons, 2006-2010)

Consumption

Imports

Production

Exports

1651

1358

1287

1066 1093

802

1100

975

728

694

605

558

492 321

396 58

109 12

30 2006

2007

Source: State Statistical Committee, UNDP calculations.

76

2008

2009

7 2010


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.17—Kyrgyzstan: Financial results of coal companies (in million som, 2006 2006-2009)

463

Gross income

467

464

439

Costs Profit (loss)

204

227

167 135

2006

-32

2007

-23

2008

-0.5

2009

-28

Source: Finances of Enterprises of the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010; and UNDP calculations. The Th data include small-scale enterprises.

Coal sector financials. Despite this turnaround, coal mining appears to be lossloss making (Figure 3.17), ), as cost growth seems to have more than matched growth in revenues. However, by all accounts, the coal sector is plagued by extensive criminal and shadow economy activities, which reduce the utility u of the sector’s financial data.

production from existing mines, to open new ones, and to create the infrastructure needed to get the coal to market. It is not clear that the financing for such investments would be easily forthcoming.

In many respects,, the coal industry seems relatively well placed to take advantage of the opportunities now facing the energy sector. Market forces play a large role in setting retail coal prices; private capital an and investment are playing significant roles; Kyrgyzstan possess considerable coal reserves for exploitation; and the demand for coal could be boosted both by higher electric and thermal power tariffs (if/when they come) and by the prospective construction off large coalcoal fired power plants (e.g., Kara-Keche). Keche). On the other hand, prospects for coal sector expansion now face a number of obstacles, including the following: •

Many any coal mines are increasingly tapped out, especially in light of the low-tech mining processes that are often employed. Further expansion of the sector could require large investment outlays—to to increase

Informal and criminal activities in the coal sector are quite large; property rights can be difficult to protect protect. These uncertainties limit the attractiveness of investment projects in the coal industry.

Many coal companies ((like other mining ining enterprises) experience pressures from the local authorities to address local social problems (e.g., provision of free coal). While not unreasonable in and of themselves, such pressures can deter inves investment in the coal sector.

Allegations of child labour in Kyrgyzstan’s coal mines may also weigh on prospects for the coal industry’s development.94

94

See, for example, Natalia Antelava, “Child labour in Kyrgyz coal mines” (http://news.bbc.co.uk/go/pr/fr/ http://news.bbc.co.uk/go/pr/fr/-

77


Poverty and Social Impact Assessment

Decentralized renewables Kyrgyzstan has good potential for the application of decentralized renewable energy technologies, primarily small hydropower stations on mountain rivers, solar and wind energy, and biogas plants. In comparison to big hydro and hydro-carbons, decentralized renewable projects are relatively inexpensive (in terms of up-front capital costs), and can attract at least some of the financing needed for their construction and maintenance from donors and the communities in which they are located.95 Despite this, according to one source, less than one percent this potential is being utilized.96

in mostly the 2-3 megawatt installed capacity range) with total installed capacity of 277 megawatts. It also calls for the construction of a wind power plant near Balykchi with a rated capacity of 22 megawatts. However, small hydro projects in Kyrgyzstan face a number of important obstacles.

Seasonality: The streams on which many of these facilities are located are more likely (than larger rivers) to freeze in the winter. These facilities can therefore be rendered inoperable during the season when power and heat are in greatest demand, and when central grids are unable to compensate. By contrast, many communities that are connected to the central power grid draw on it during the summer time (when power is relatively abundant), thereby reducing the demand for offgrid power and the commercial viability of small hydro (and other decentralized renewables) projects.

Hydrology: These streams are also more likely (than larger rivers) to undergo significant reductions in water flow during periods of drought, thereby reducing de facto generation capacity.

Capacity: The knowledge base of local companies in constructing and especially maintaining small hydropower stations remains underdeveloped. Communities that construct small power plants (often with donor assistance) are too often unable to repair or maintain them. Neither private- or public-sector agencies in Kyrgyzstan possess at present the technical expertise needed to construct and especially maintain decentralized renewable energy installations.

Communities that construct small power plants are too often unable to repair or maintain them. For these reasons, Kyrgyzstan’s National Energy Programme97 officially recognizes the importance of decentralized renewables; and a programme to develop small and medium-size power plants has been adopted. 98 The law “On Renewable Energy” (of 31 December 2008) provides the overarching legal framework. A special public agency, the Directorate for Small and Medium-Size Power Facilities, was established in 2008. The government’s programme to develop small and medium-size power plants calls for the construction and reconstruction of 43 small and medium-sized facilities (primarily small hydropower plants, /2/hi/asia-pacific/6955202.stm), BBC News, 24 August 2007. 95 Decentralized renewables (and energy efficiency) projects can also attract carbon finance, under the Kyoto Protocol’s clean development mechanism. 96 Omorov, A. “Small hydro for rural development”. 97 “The National Energy Program of the Kyrgyz Republic for 2008-2010 and the Strategy for Development of the Fuel and Energy Sector until 2025”, approved by parliamentary decree on 24 April 2008, N 346-IV. 98 Approved by presidential decree on 14 October 2008, N 365.

78


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.18—Kyrgyzstan: Additional costs (above current tariff tariff levels) per kWh associated with electricity generated from decentralized renewables

$0.17

$0.09 $0.06

$0.06 $0.04

Small hydro

wind

biomass

geothermal

Solar

Source: UNDP-Kyrgyzstan Kyrgyzstan project “Promotion of renewable energy for development in remote regions”, 2011.

In addition to these sector-specific sector challenges, small hydro projects face obstacles that are common to all decentralized renewable technologies in Kyrgyzstan:

Finance: Funding mechanisms able to transparently combine and manage government, private private-sector, community-based, based, and donor funding on both a commercial and grantgrant financing basis have yet to be fully developed.99

Regulatory: The regulatory and technical requirements—and and the capacity to enforce them— —needed to ensure technological standardization of power generated from off-grid grid power producers are not fully in place.

99

Tariffs: As long as tariffs for electricity generated from other sources remain relatively low (e.g., compared to otherr countries), the “National Power Grid” has little reason to purchase “expensive” electricity from generators using decentralized

The renewable energy and energy efficiency trust fund proposed by UNDP in Tajikistan could possibly be considered for adoption in Kyrgyzstan as well.

renewable technologies. Such purchases must ultimately be subsidized, either from the retained earnings of the national transm transmission company, from cross cross-subsidies implicitly paid by other electricity companies or consumers of power generated from sources, from the state budget, or by donors. Estimates of additional costs associated with generating electricity from decentralized renewables enewables are shown in Figure 3.18.. Relative to current household tariff levels (around $0.015/kWh), these additional costs are considerable.100 Thus, despite strong government and donor interest, progress in constructing the requisite legal framework, and ggrowing amounts of resources to finance their development, small hydro and other decentralized renewable energy technologies in Kyrgyzstan remain in their infancy. Generation costs have yet to be brought down 100

However, since decentralized renewables seem unlikely to comprise a significant share of Kyrgyzstan’s Kyrg energy balance on the near future, the subsidies needed to cover these additional costs would likewise be insignificant.

79


Poverty and Social Impact Assessment

to “competitive” levels; and the technical and commercial regulations needed to facilitate third-party access to the national grid (e.g. feed-in tariffs) have not been made operational.

Energy efficiency Compared to many other former Soviet Republics, Kyrgyzstan is a relatively efficient energy user, in terms of ratios of energy consumption to output. These ratios have generally improved since 2008. In light of Kyrgyzstan’s energy poverty, and the shortage pressures that have sharpened since 2008, these trends are not surprising. Nonetheless, energy conservation policy in Kyrgyzstan is on the agenda. The National Energy Programme for 2008-2010101 called for growth in energy consumption to be limited to 0.4-0.5 percent for each percentage point increase in GDP during 2010-2025. Legislation obliges the government to design and implement energy conservation programmes and projects, and to establish a state energy saving fund. An energy efficient building code was adopted in 2009,102 while Bishkek and other municipalities have moved ahead with subnational energy conservation programmes. UNDP, the EBRD, EU, and other donors are working with the government to translate these principles into practice. However, much of this legislation remains declarative in nature; questions of implementation and enforcement (e.g., regarding the new building code) and financing investments in energy efficiency continue to loom large. Proposals to retrofit apartment buildings with energy efficient heating systems and insulation often founder on difficulties apartment owners face in collective decision making. Kyrgyzstan’s low electricity tariffs reduce the financial 101

The National Energy Program for 2008-2010 and the Strategy for the Fuel and Energy Sector Development until 2025 dated April 24, 2008, N 346-IV. 102 Order № 135 of the State Agency for Architecture and Construction, 2 October 2009.

80

attractiveness of many energy efficiency projects.

However, much of this legislation remains declarative in nature; questions of implementation and enforcement and financing investments in energy efficiency continue to loom large.

Energy tariffs and costs Tariff setting: general principles. As virtually all major companies in the energy sector are regulated as natural monopolies, their tariffs must be “justified” on the basis of “approved costs”. This necessarily gives energy prices a cost-plus character. It also gives energy price determination a bureaucratic, administrative character: enterprises must present to the Regulator detailed planned calculations of expenses, in order to show that the requested tariffs are economically appropriate. The Regulator is obliged to validate these calculations in line with established procedures. Costs are divided into production costs and full costs. Production costs include: •

• • • • •

Material costs (e.g., raw materials and supplies, components and semi-finished products, industrial services, fuel and energy); Labor costs; Social insurance costs; Depreciation of fixed assets; Other costs; and Taxes.


Chapter 3 - Kyrgyzstan’s Energy Sector Table 3.2—Kyrgyzstan: Differences between planned and actual costs of larger power companies (2009-2010)

Cost category

2009

2010

Full cost

-5%

-2%

Production cost

-5%

2%

- Material costs

-4%

-11%

- Labor

-11%

16%

- Social insurance

-12%

16%

- Depreciation of fixed assets

5%

20%

- Other costs

14%

14%

- Taxes

-62%

-32%

Investment costs: - Debt service

120%

14%

- Capital investments

-67%

-47%

Source: Ministry of Energy.

Full costs are the sum of production costs and operating expenses. In addition to production costs, these include investment costs, i.e.,: • •

debt service and principal repayments; and capital costs (for (re)construction of plant and equipment).

Cost-plus pricing gives companies incentives to overestimate anticipated production costs. The Regulator, who is constrained by the government’s social obligations, knows this, and therefore seeks to pare back the companies’ requests as much as is economically admissible. However, the Regulator does not always have the capacity and resources needed to validate the calculations presented by the companies. Actual expenses can deviate from those agreed with the Regulator, for three reasons. The first is the appearance of economically justified costs that were not anticipated at the time the tariff was set (e.g., due to higherthan-expected inflation rates). Second, tariff policies may change, necessitating a change in the trajectory of expenses. This happened in Kyrgyzstan’s electric power sector in 2009

and 2010, when planned tariff increases either did not materialize, or were rescinded after their introduction (Figure 3.19). Third, some anticipated expenses (e.g., investment programmes) may not be made. In Kyrgyzstan’s energy sector, production cost over-runs are often offset by cutbacks in capital spending, as is shown in Table 3.2. Whereas expenses in the power sector were reported at 5 percent below approved levels in 2009 and 2 percent below approved levels in 2010, this was primarily due to systematic failures to execute the capital investment plan. Tariff and costs in the electric power sector. As the data in Figure 3.19 show, household electricity tariffs (in dollar terms) are today roughly where they were in 2006. This reflects the fact that an April 2008 decision to increase tariffs during 2009-2012 was not implemented, as well as the April 2010 rescinding of the government’s December 2009 decision to double tariffs in January 2010. As described above, Kyrgyzstan’s relatively low electricity tariffs have been accompanied by rapid growth in costs in the power sector. As the data in Figure 3.20 show, this growth has been driven primarily by materials expenditures (chiefly for “fuels”103) by companies in this sector; the share of raw materials costs in total expenditures rose from 39 percent in 2006 to 54 percent in 2009. While the “fuels”104 component of the industrial producer price index registered an 84 percent increase during 2006-2009, hydropower’s dominance in Kyrgyzstan’s electricity balance should minimize the impact of these rising fuel prices. Nor are the investment projects in this sector a good explanation for rising fuel costs, as capital outlays under these projects would show up under the “other costs” or possibly “depreciation” categories. 103

In 2009 fuel purchases accounted for 62 percent of reported raw materials expenditures. Source: Enterprise Finance in the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. 104 Refined oil products and coking coal.

81


Poverty and Social Impact Assessment

Figure 3.19—Kyrgyzstan: Actual versus planned household electricity tariffs ariffs (per kWh, 2006 2006-2012)

$0.045 Actual tariffs $0.040 Tariffs per April 2008 policy $0.035 Tariffs per December 2009 policy $0.030 $0.025 $0.020 $0.015 $0.010 2006

2007

2008

2009

2010

2011

2012

Note: these figures are average annual household tariff rates in som, multiplied by the average annual som/$ exchange rate. 2011 20112012 calculations use the average exchange rate for 2010; the “actual tariff” projection for 2011-2012 2011 2012 assumes no change in current (2011) tariff levels. Source: The medium term tariff policy of Kyrgyz Republic on electricity for 2008-2012 2008 2012 as of 24 April 2008, government resolution No 164; and the medium term tariff policy of Kyrgyz Republic on electricity and thermal energy for 2010-2012, 2010 2012, 12 Novemb November 2009, Government Resolution No 699.

Figure 3.20—Kyrgyzstan: Kyrgyzstan: Cost share trends in the electric power sector (2006-2009) 2009)

51%

54%

2006

2007

2008

2009

39% 38% 25% 26% 26% 25%

25%

27% 15%

11% 10% 9% 9%

Materials

Labor and social insurance

Depreciation

13%

All other costs

Source: Enterprise Finance in the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. The data cover all enterprises in the electric power industry, including small enterprises.

82


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.21—Kyrgyzstan: Actual versus planned household thermal power tariffs ariffs (per gigacal., 2007-2012) 2007

$60 Actual tariffs $50 Tariffs per April 2008 policy $40 Tariffs per December 2009 policy $30 $20 $10 $0 2007

2008

2009

2010

2011

2012

Note: these figures are average annual household tariff rates in som, multiplied by average annual som/$ exchange rates. 2011 2011-2012 calculations use the average exchange rate for 2010; the “actual tariff” projection for 2011-2012 2012 assumes no change in current (2011) tariff levels. Sources: The medium term tariff policy of Kyrgyz Republic on thermal energy for 2008-2012, 2008 2012, 24 April 2008, government resolution No 165; and the medium term tariff policy on electricity and thermal energy for 2010-2012, 2012, 12 November 2009, government resolution No 699. Figure 3.22—Kyrgyzstan: Kyrgyzstan: Cost share trends in the thermal power sector (2006-2009) 2009)

72% 68% 67% 70%

2006

2007

2008

2009

18% 18% 17% 19% 6% 5% 4% 4%

Materials

Labor and social insurance

Depreciation

8% 9% 9%

5%

All other costs

Source: Enterprise Finance in the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. The data cover all enterprises in the thermal power industry, including small enterprises.

Tariff and costs in the thermal power sector. Household thermal power tariffs (i (in dollar terms) have risen some 46 percent compared to where they were in 2006 ((Figure 3.21).105 However, had the April 2008 decision to increase tariffs during 2009-2012 2009 105

Household thermal power tariffs are split into tariffs for heating and tariffs for hot water. The data shown in Figure 28 are for heating tariffs, but the trends in hot water tariffs are identical.

been implemented, or had the government’s December 2009 decision to quintuple thermal power tariffs in January 2010 not been rescinded in April, these would be much higher than is the case today. A similar trend, in terms of increases in the shares of expenditures devoted to materials purchases, is apparent in the thermal power sector as we well (Figure 3.22). ). However, it is less pronounced.

83


Poverty and Social Impact Assessment

Figure 3.23—Kyrgyzstan: Shares of material costs in the thermal power sector ector (2006 (2006-2009)

68% 67% 70%

72%

2006

2007

51%

2008

54%

54% 54%

51% 43%

39% 38%

Thermal

2009

Electricity

Coal

Source: Enterprise Finance in the Kyrgyz Republic, State Statistical Committee, Bishkek, 2010. The data cover all enterprises in these sectors, including small enterprises.

As the generation of thermal power in Kyrgyzstan necessarily involves the purchase and consumption of fuels (e.g., gas, coal, mazut), one would expect fuel (raw materials) expenditures to play a large role in the sector’s cost structure. The data in Figure 3.22 therefore show the share of raw materials purchases in total costs rising from 68 to 72 percent during 2006-2009. 2009. To be sure, it is difficult to generalize about costs across the thermal power sector: whereas some companies (e.g., Kyrgyzzhilkommunsoyuz) generate their own thermal energy, others (e.g., Bishkekteploset’) purchase thermal power and distribute it to final users. Nonetheless, the data in Figure 3.22 indicating only moderate growth in raw materials costs in the relatively fuel-intensive fuel thermal power sector contrast with the data in Figure 3.20 showing more robust growth in raw materials costs in the less fuel-intensive fuel electric power sector. The character of materials cost trends in electric power stands out even more strongly when contrasted with cost trends in the coal sector, as well as in the thermal power sector (see Figure 3.23). 3.23

84

Legal and regulatory framework The regulation of the energy sector is based on a number of legal acts. These include: •

The 30 October 1996 law “On Energy” (N 56), which allowed “enterprises in the fuel and energy sector to have any organizational-legal legal form of operation and any form of ownership (public, municipal and private)”. However, this law also assumed predominant government overnment management in the energy sector: an authorized public agency is empowered to set economically justified and socially reasonable price priceand tariff-setting setting mechanisms for electric and heat energy as well as natural gas. This framework frequently results in policy--making based primarily on social, rather than economic criteria.

The market principles for energy sector operations were established in the 28 January 1997 law “On Electric Power” (N 8), which called for “creating a competitive environment and the formation of an energy market”, as


Chapter 3 - Kyrgyzstan’s Energy Sector

well as “encouraging development of the private sector and attracting investments”. This law also set forth the rights and obligations of consumers and contractual relations with suppliers. Thus, distribution companies are obliged to ensure safe, reliable and uninterrupted electricity supply; continuous improvements in the quality of services; the timely handling of customer complaints; and to compensate consumers for material damages incurred, if the company should be at fault. Distribution companies are entitled to cut the power supply to users who do not make timely payments for power consumed.

While the privatization of Kyrgyzstan’s existing large hydropower stations and transmission infrastructure is precluded, the way is open for the privatization of other electricity generation and distribution assets.

The 10 December 1997 law “On Consumer Protection” (N 90) protects consumers against supplier noncompliance with their obligations to deliver services. Administrative penalties are provided for the noncompliance.

The Administrative Code of 4 August 1998 (N 114) protects the integrity of engineering systems and facilities against unauthorized use of energy, and non-compliance with the instructions of power enterprises.

The 21 January 2002 law “On the Special Status of the Toktogul Cascade and the National High-Voltage

Transmission Line” (N 7) precludes the privatization of the state-owned “Power Plants” generation and “National Power Grids” transmission companies. However, the Bishkek Combined Heat and Power Plant may undergo privatization. This law also specifies that ownership of the Kambarata-1 and -2 hydropower plants shall be regulated by separate legislation. Thus, while the privatization of Kyrgyzstan’s existing large hydropower stations and transmission infrastructure is precluded, the way is open for the privatization of other electricity generation and distribution assets. •

The 31 December 2008 law “On Renewable Energy” (N 283), which regulates the development and use of decentralized renewable energy technologies. It calls for mechanisms to stimulate the development of these technologies, and to support producers and consumers of decentralized renewables. In particular, the government is to support tariff-setting so as to guarantee an eight-year payback period for decentralized renewable projects. However, while this law created the legal framework for decentralized renewables, the practical framework for its implementation has yet to be fully introduced.

The 3 February 1999 law “On Coal” (N 18) specifies that coal deposits are state property, and makes local governments and local state administrations responsible for allocating land and licensing geological and mining activities. It also seeks to protect the rights of consumers, and calls for the conditions to attract investment and increase output. This law mandates the coal industry’s restructuring, on the basis of

85


Poverty and Social Impact Assessment

both company and external financing, including from the national budget. The energy sector’s regulatory framework revolves around the Ministry of State Property, the Ministry of Energy, and the latter’s State Department for Regulating the Fuel and Energy Sector (“the Regulator”). The Ministry of State Property formally acts as the owner and manager of state-owned power companies. It designs and implements development strategies (inter alia concerning denationalization and privatization), selects and monitors company management, and the like. The Ministry of Energy performs sectoral functions for the design and implementation of policy development, strategic planning, assessment, and forecasting. The Regulator’s tasks include: balancing the interests of energy producers and consumers (including dispute resolution); licensing energy sector operations; and setting energy tariffs.

power producers on the domestic market”.106 Denationalization and privatization in the electrical power sector was to occur during 1997-1999, in four stages (three of which have been accomplished): •

The first stage entailed Kyrgyzenergo’s transformation into a joint-stock company with some 94 percent of its shares being affixed as state property.

During the second stage, Kyrgyzenergo’s auxiliary enterprises (e.g., repair and maintenance companies) and social entities were transferred to local government ownership.

During the third stage, separate generation, transmission, and distribution companies were divested from Kyrgyzenergo. This reflected the desire to unbundle the country’s power system into independent companies according to their functional characteristics, whose activities could then be regulated (at least in part) by market logic. (Reality since then suggests that such expectations may have been exaggerated).

The fourth stage of the reforms envisaged privatization of distribution companies. However, this has not yet come to fruition. Nor, in light of the developments described in Box 2 above, does this seem likely to happen any time soon.

In practice, the Regulator is not independent of the Ministry of Energy, so that this Ministry discharges both managerial and regulatory functions. The Regulator is therefore often guided by conflicting political and economic interests, which can generate inconsistencies in tariff setting and other dimension of energy policy.

Policy reform to date Prior to the start of reforms in Kyrgyzstan’s energy sector, electricity was supplied in the domestic market and for export by a vertically integrated state-owned Kyrgyzenergo, which monopolized the generation, transmission, and distribution of electricity. In April 1997, the government adopted a programme to demonopolize and partially privatize the electricity sector, as part of efforts to “establish a competitive environment among electrical and thermal

Instead, policy since April 2010 has emphasized the restoration and modernization of state control in the energy sector, with a focus on improving management and reducing corruption. On the other hand, Kyrgyzstan’s current legal framework does permit the privatization of the Kambarata hydropower 106

Government resolution of 23 April 1997 “On the Denationalization and Privatization of the Kyrgyz State Holding JSC Kyrgyzenergoholding”.

86


Chapter 3 - Kyrgyzstan’s Energy Sector

stations,107 the Bishkek Combined Heat and Power Plant, and the Bishkek District Heating Distribution Company, should the government so choose (and appropriate buyers be found). By contrast, there have been no such reforms in thermal power sector. The 20092011 Country Development Strategy included some reform measures for Kyrgyzzhilkommunsoyuz, but no progress was made afterwards. An April 2009 programme to restructure and privatize Kyrgyzgaz108 was quickly followed by a draft agreement between the governments of the Kyrgyz Republic and the Russian 109 Federation, according to which Gazprom would acquire 75 per cent plus 1 share of Kyrgyzgaz’s authorized capital. However, Kyrgyzgaz has not been privatized so far.

The new government that took power in late 2010 has yet to develop a well defined programme for the energy sector. Thus, market reforms in the energy sector have effectively been suspended since April 2010. The new government that took power in late 2010 (following parliamentary elections) has yet to develop a well defined programme for the energy sector. In addition to reassuring the public about the permanence of current energy tariffs, efforts have focused on ensuring the country had adequate energy supplies during the winter of 2010-2011, and 107

Under the 3 February 2009 “Agreement on the Construction of Kambarata HPP-1” between the governments of the Kyrgyz Republic and the Russian Federation, a joint-stock company became the owner of the Kambarata-1 hydropower station. 50 percent of the shares in this joint venture belong to Russia’s INTER RAO EES; 50 percent belong to “National Power Grid”. The Russian partner agreed to find the resources to finance the construction of Kambarata-1. 108 Approved by the KRG Resolution No 229 dated 21 April 2000. 109 Approved by the KRG Resolution no 317-p dated 15 June 2009.

on implementing the Fuel and Energy Sector Transparency Initiative. Energy sector transparency has been a major public concern, contributing inter alia to the April 2010 political developments. For example, consumers frequently have difficulty obtaining information about energy company costs or tariff-setting procedures. According to the January 1997 law “On Electric Power”, the Regulator is to officially inform the public about any change in electricity tariffs through mass-media at least one month before the planned introduction of new tariffs. However, this requirement is sometimes honored in the breach. For example, the decree to double electricity tariffs as of 1 January 2010 was promulgated on 30 December 2009. While legislation gives the public the right to information about the energy sector, these laws are not fully implemented due to a lack of implementing regulations and procedures, and because many citizens are unable to prepare a proper request for information. In order to address these issues, the government in May 2004 adopted the principles of the Extractive Industries Transparency Initiative on increased transparency of mining industry operations (EITI).110 An action plan to implement EITI principles was adopted; statistical forms for disclosing of financial information in the mining sector were developed, and relevant reports prepared. However this initiative has focused primarily on disclosure of information about oil and gas companies, not the electric power industry. Limited civil society involvement in implementation of the initiative has further reduced its significance. Following the April 2010 developments, the government introduced the Fuel and Energy Sector Transparency Initiative

110

Government Decree “About Measures to Increase Transparency of Operations in the Mining Industries” issued on May 14, 2004; N 361

87


Poverty and Social Impact Assessment

(FESTI),111 extending transparency and civic engagement requirements to all enterprises in this sector. In particular, FESTI calls for: • •

• •

Creating a public advisory board for the energy sector; More regular, complete disclosure of financial information in the energy sector, based on the use of modern financial information management systems; The expanded use of escrow accounts in the sector; and The introduction of competitive tenders and auctions for electricity export and import, as well as for major energy sector procurements.

Since FESTI’s July 2010 introduction, agreements have been concluded between the Ministry of Energy and all of Kyrgyzstan’s major energy companies (i.e., the “Power Plants” electricity generation company, the “National Power Grid” transmission company, and the “Severelektro”, “Vostokelektro”, “Jalalabadelektro”, and “Oshelektro” electricity distribution companies). These agreements concern key financial and operational indicators, which are monitored (many via escrow accounts) on a monthly basis by the companies’ boards of directors and the Ministry of Energy. The Ministry of Energy has sought to promote “social competition” by releasing comparisons of electricity companies’ collection rates. Civic engagement in the energy sector has since 2009 been led by a working group of experts, supported inter alia by the Electricity Governance Initiative (EGI), which seeks to implement transparency principles in the electricity sector. The EGI programme, which has been applied in a number of developing countries, is based on a set of 68 policy and

regulatory indicators, of which 32 indicators were selected for application in Kyrgyzstan.112 FESTI and its related support initiatives represent an important step forward, in terms of improving public-sector governance and increasing civic engagement in energy. As the April 2010 events preclude for now the

The improvements in regulatory oversight promised by FESTI may need to be accompanied by measures to simplify the superstructure, and to strengthen corporate governance within the energy companies themselves. accelerated introduction of competitive market forces into the energy sector, increased civic engagement in/oversight of the state-owned monopolies that dominate the sector, on the basis of simple, transparent performance indicators, seems most desirable. However, there is no shortage of supervisory bodies already present in the energy sector; relations between the Ministry of State Property (the energy companies’ nominal owner), the Ministry of Energy, its Regulator, and these companies’ boards of directors were sufficiently complicated even before the creation of the Supervisory Board. The improvements in regulatory oversight promised by FESTI may need to be accompanied by measures to simplify this external regulatory superstructure, and to strengthen corporate governance within the energy companies themselves.

112

111

Decree by the President of the Kyrgyz Republic “On Transparency Initiative in the Fuel and Energy Sector of the Kyrgyz Republic” as of July 20, 2010; №49.

88

Electricity governance in Kyrgyzstan: An institutional assessment, N. Abdyrasulova, N. Kravtsov, Public Fund UNISON, 2009. This document is prepared by a team of experts from public organizations of Kyrgyzstan, with support of the World Resource Institute (USA) and Prayas Energy Group (India).


Chapter 3 - Kyrgyzstan’s Energy Sector

Energy sector development—future scenarios While efforts to improve public-sector governance and increase civic engagement in energy are certainly important, the breadth and depth of the problems facing Kyrgyzstan’s energy sector suggest that more ambitious energy reforms may at some point return to the front burner. In this respect, the delineation of four different development scenarios/directions for the electricity sector now seems possible: 1) Restoring vertically integrated energy companies; 2) Rebundling/merging the distribution enterprises, and making the merged entity responsible for reducing losses; 3) Returning to privatizing and marketizing power companies; and 4) Improving management within the electricity sector without changing its basic ownership or organizational characteristics.

To the extent that some of today’s unbundled companies face questions about managerial capacity, these questions would be multiplied many times over in a rebundling scenario. Scenario 1: Restoring the vertically integrated electricity monopoly. Under this scenario, the electricity generation, transmission, and distribution companies would be merged/rebundled, recreating a single operational centre for decision-making and implementation. Such a rebundling could allow for better internal control over the power sector, particularly in terms of contracting and other commercial activities that are internal to the sector but external to the unbundled companies. It also assumes that the greater internal complexity of the consolidated

company would not overwhelm the managers appointed to run it. If such assumptions are correct, then reductions in losses and some improvements in the procurement of raw materials and other supplies could be expected. This could take some of the pressure off of tariffs, and increase the cash flow needed to finance investments in the power sector. However, if this scenario were to be chosen, the rebundled power company would be very large and complicated—arguably more so than during the Soviet period, when the numbers of users (companies and households) were smaller, technologies simpler, and managers (and customers) had fewer choices to make. To the extent that some of today’s unbundled companies (which are smaller and simpler to manage than a vertically integrated monopoly would be) face questions about managerial capacity, these questions would be multiplied many times over in a rebundling scenario. Efficiency gains from reductions in uncertainties due to external contracting and pricing could be overwhelmed by losses due to overcentralization, additional managerial staffing, less managerial and entrepreneurial initiative, lengthier decision-making processes, and the reduced transparency that comes with greater organizational complexity. In this case, reductions in pressures on tariffs and improvements in cash flow could be shortlived. Scenario 2: Rebundling/merging the distribution companies, and making the merged entity responsible for reducing losses. This scenario reflects the view that the distribution companies, which interact directly with final users, are the “weak link” in the electricity chain. Changing managerial and organizational incentives facing these companies could therefore improve the functioning of the whole sector. In particular, this scenario would feature:

89


Poverty and Social Impact Assessment

Keeping the existing model for power generation and transmission;

Merging the four regional distribution enterprises into a single company, which would inter alia be responsible for reducing losses;

Thereby helping to strengthen central state control over distribution (at the expense of the regional authorities).

This approach shares many of the strengths and weaknesses set forth in Scenario 1 above. Because the scale of the renationalization would be smaller, the benefits and costs (risks) of applying this approach would likewise be smaller. Scenario 3: Return to privatization and marketization. This scenario is based on the belief that the state is often a less effective owner than a private entity, and that a regulatory and ownership regime with greater scope for competitive market forces can provide more effective social control over the sector than state direction. This scenario could feature two key elements: •

Privatization of power companies to strategic investors; and

A greater role for private distribution companies in the resale of electricity from the four regional power distributors.

Prospective privatization options would presumably start with the sale of distribution company assets to strategic investors—with the observance of relevant legal norms and expectations about transparency and accountability. Such a scenario could also bring much needed capital and know-how to Kyrgyzstan’s electricity sector, helping to reduce losses, improve financial performance, and modernize the electricity infrastructure. (Such privatizations, as a way to reduce losses and modernize electricity distribution, are common in other 90

transition economies.) Such options could be expanded to generation—both in hydropower (building on the Kambarata-1 model, in which Russia’s RAO EES would become a majority owner in Kyrgyzstan’s largest hydropower station) and in coal-fired plants (e.g., KaraKeche).

Even if the domestic political calculus should change, Kyrgyzstan’s efforts to find appropriate strategic investors could come up empty. However, any privatization scenario faces three key weaknesses—particularly at the present moment. First: this model was attempted with the proposed sales of the “Severelektro” and “Vostokelektro” distribution companies, and it did not withstand public scrutiny. Attempts to once again move in this direction may not, at present, be politically feasible. In principle, this scenario could be pursued later, under conditions of full transparency. If so, this would lead to a second set of commercial and reputational obstacles, linked to the “low” tariff rates, perceptions (accurate or not) of a poor business environment, political uncertainties, and the like. Even if the domestic political calculus should change, Kyrgyzstan’s efforts to find appropriate strategic investors (e.g., multinational energy companies with deep pockets and proven track records in reducing losses) could come up empty. Third, if such privatizations could be effected within the sector’s existing structural framework, the new owners would still be monopoly sellers of electricity. International experience indicates that the gains from replacing a publicly owned monopoly with a privately owned one are often small. It also underscores the importance of efforts to improve the quality of sectoral regulation—


Chapter 3 - Kyrgyzstan’s Energy Sector

and, if possible, introduce competition from other sources (e.g., small-scale independent power wholesalers, industrial co-generation).

Scenario 4 is based on the assumption that opportunities for improving the relationships between and among enterprises and regulators in this sector have not been exhausted. Marketization and the further extension of the private sector power delivery could be explored without recourse to privatization, if the role for private distribution companies in the resale of electricity from the four regional power distributors could be expanded. Here, the state-owned distribution companies could outsource service delivery responsibilities to these companies on the basis of competitive tenders, the terms of which could include investment programmes and measures to reduce losses. These tenders could take the form of management contracts, long-term concessions, or other public-private partnership modalities, building on the experience of Pamir Energy in Tajikistan (including inter alia donor support for lifeline tariffs for vulnerable households). This approach would face a number of obstacles and drawbacks, including the large variability in distribution costs and cash collection in different regions of Kyrgyzstan, and possible resistance within the state-owned distribution companies (and their allies in the regional governments) to ceding control over assets to potential/actual competitors. On the other hand, Pamir Energy’s experience in making these arrangements work in the forbidding logistical and commercial conditions of Tajikistan’s remote Gorno Badakhshan region suggests that this model could be particularly relevant for some of Kyrgyzstan’s more remote regions. This scenario cannot be treated as a viable short-run alternative. It could perhaps

be pursued later, when better electricity sector governance and regulation reduce popular opposition to tariff increases and private investment in the energy sector, when competitive forces within the electricity sector are better able to limit cost increases, when concerns about Kyrgyzstan’s investment climate are addressed, and when Kyrgyzstan’s social protection system is better able to protect vulnerable households from the consequences of higher electricity tariffs. Scenario 4: Improving management within the electricity sector without changing the sector’s basic ownership or organizational characteristics. This is the scenario currently being pursued, as embodied by FESTI. It is based on the assumption that opportunities for improving the relationships between and among enterprises and regulators in this sector have not been exhausted. Measures to increase the quality and quantity of financial information disclosed by the power companies include:113 • •

• •

mandatory income and asset statements by top managers (and their families); stronger roles for public hearings, external supervisory boards, and other institutionalized forms of consumer protection; modernizing billing, tendering, contractual, and other financial management processes (including via the expanded use of escrow accounts); strengthening the institutional capacity of the Regulator; modernizing power metering—at every stage in the production process, not just for final users/households; and increasing the legal sanctions associated with corruption.

However, FESTI has certain limitations—particularly if it is not combined with measures to improve the market (as well 113

Detailed proposals for implementing many of these measures have been implemented by the USAIDfunded Kyrgyz Energy Advisory Services in Bishkek.

91


Poverty and Social Impact Assessment

as governance) environment in which the electricity sector functions. For example, FESTI has led to the appearance of new regulatory bodies (e.g., the Supervisory Board) in a sector in which there is no shortage of supervisory bodies (e.g., the Ministry of State Property, the Ministry of Energy, its Regulator, and the companies’ boards of directors). The improvements in regulatory oversight promised by FESTI may therefore need to be accompanied by measures to simplify this external regulatory superstructure, and to strengthen corporate governance within the energy companies themselves. A similar scenario exercise can be pursued in gas sector development, but with three key differences: (i) Kyrgyzgaz has not (yet) been unbundled; (ii) the power source/fuel in question (gas) comes almost exclusively from import; and (iii) the principle that household tariffs should cover costs of service is more firmly grounded in the sector’s business model. Scenario 1: Privatizing Kyrgyzgaz— with or without unbundling. This would be similar to scenario (3) for the electric power sector set forth above, with the proviso that the strategic investor (Gazprom) seems most interested in purchasing a controlling stake in Kyrgyzgaz as a vertically integrated (unbundled) monopoly. The main benefits of this scenario reflect Gazprom’s financial and technological capacity, which could be used to expand household access to gas services. In addition, supplies from Gazprom could provide an alternative to deliveries from UzbekTransGaz, thereby ending Kyrgyzstan’s reliance on a single gas supplier. On the other hand, privatization without unbundling could limit subsequent possibilities for potential competition and additional investment in the gas distribution infrastructure. Moreover, the current sociopolitical concerns about privatization in the electric power sector would be likely to extend to Kyrgyzgaz’s potential privatization/sale to 92

Gazprom as well. This suggests that such a privatization is unlikely to occur for the foreseeable future—and when it does, the sale should be subjected to the highest levels of scrutiny and accountability. Moreover, unless the sale were to be accompanied by large increases in domestically produced gas, it is not clear why this privatization should be expected to reduce gas tariffs for Kyrgyzstani households. Scenario 2: Improving the management of Kyrgyzgaz without changing the gas sector’s institutional or ownership framework. This approach is analogous to scenario 4 for the electricity companies described above. That is, better regulation and management of, and governance within, Kyrgyzgaz would be emphasized, rather than privatization or promoting market competition. Transparent agreements between Kyrgyzgaz and the Ministry of Energy could be drafted to monitor the company’s performance; an external supervisory committee could be established; etc. The real questions would perhaps concern whether such arrangements would significantly strengthen incentives to increase efficiency, and where the resources needed to finance the expansion of the gas distribution infrastructure would come from.

Supplies from Gazprom could provide an alternative to deliveris from Uzbektransgaz, thereby ending Kyrgyzstan’s reliance on a single gas supplier. Reforms in the thermal power sector can likewise be considered within a scenario development framework. Key points in this sector include: (i) the sector’s deeply unprofitable nature, and extensive depreciation of fixed assets; combined with (ii) significant competition from other energy sources. Consumers unhappy with district heating tariffs or service quality can rely more


Chapter 3 - Kyrgyzstan’s Energy Sector

extensively on electric heaters, install coal- or gas-fired boilers, or burn firewood.

Production and sales data indicate that Kyrgyzstan since 2007 has experienced significant increases in energy prices and reductions in energy production and consumption. These data indicate that electricity generation and consumption (generation less net exports and losses) dropped by some 25 and 13 percent, respectively, during 2007-2010. Gas consumption dropped by almost two thirds, as prices of gas imported from Uzbekistan rose from $100 to $240 per 1000 cubic meters during 2007-2009. Faced with growing shortages of centrally supplied electricity and unaffordable gas, many households, businesses, and public institutions switched to coal-fired boilers. Domestic coal production rose some 41 percent during 2007-2010; apparent consumption (production less net exports) rose 57 percent. On the whole, household energy prices during 2007-2010 rose by 81 percent, compared to a 44 percent increase in the consumer price index.

While Kyrgyzstan’s power, thermal, gas, and coal sectors have many differences, they share some important similarities. The companies in these sectors are mostly state-owned monopolists that generally report negative profitability, high technical and commercial losses, and (with the exception of the gas sector), high or growing depreciation rates (reaching 60 percent for fixed assets in the thermal power sector). Many of these sectors report large accounts receivable, payable, or both. Outside of the coal sector, private capital is a rarity; and foreign investment is almost completely absent. Progress towards the creation of a regional electricity market, which could help attract foreign investment into Kyrgyzstan’s electricity generation and transmission sectors, remains halting at best.

Scenario 1: Privatizing the major companies (Bishkekteploset’, Kyrgyzzhilkommunsoyuz, Bishkekteploenergo). This scenario does not seem realistic without significant tariff increases, possibly combined with explicit subsidies to the private company(s) that would take over service provision responsibilities.

These data indicate that electricity generation and consumption dropped by some 25 and 13 percent, respectively, during 2007-2010. Scenario 2: Improving company management without changing the sector’s institutional or ownership framework. This approach would be similar to scenario (4) for electric power, or scenario (2) for thermal power. Efforts to improve financial management in and regulation of these companies, inter alia by enhancing transparency and promoting civic engagement, could certainly be pursued. However, given the dismal state of the sector’s finances, and the extent to which the companies rely heavily on subsidies (often provided implicitly and non-transparently) by the national and municipal governments, it is difficult to imagine significant improvements in this sector if its basic financial parameters are not changed.

Conclusions

93


Poverty and Social Impact Assessment

94

Energy tariffs and prices generally seem low, both compared to other transition economies and relative to cost-recovery levels. This reflects strong political and social opposition to raising tariffs. However, since 2007 virtually all of these sectors have reported price or tariff increases well in excess of consumer price inflation. But except for the coal industry, declines in production/generation and consumption were the norm during 2008-2010; for some sectors (e.g., gas), these declines are quite large. This combination of sharply higher prices and declining consumption necessarily raises concerns about deterioration in vulnerable households’ access to reliable, affordable energy services. To be sure, household energy consumption is currently being subsidized, from the state budget (which provides up to 75 percent of the revenues received by thermal power producer Kyrgyzzhilkommunsoyuz), from municipal budgets, via crosssubsidies from electricity exports, and from future generations, who will ultimately have to cover the differences between revenues and costs in the energy sector today—in the form of higher tariffs, higher taxes, or poorer services in the future. Since most of these subsidies increase with the quantity of energy consumed, which in turn is linked to income, they are not likely to help protect Kyrgyzstan’s most vulnerable households. To be sure, some important progress has been made since 2007. Collection rates in the electricity sector have improved significantly as the quasi-

fiscal deficit has dropped; the rate of depreciation of fixed assets in the gas sector has fallen; and the basic legal framework for the development of decentralized renewable energy technologies has been introduced. New electricity generation capacity has been brought on line, particularly in terms of the Kambarata-2 hydropower plant. The Fuel and Energy Sector Transparency Initiative is beginning to improve regulatory governance and increase civic engagement. Rising water levels at the Toktogul hydropower station allowed for large increases in electricity generation and consumption during 2011. Still, taken as a whole, the scale of Kyrgyzstan’s energy problems remains daunting. Particularly worrisome is absence of a concept for attracting private capital and know-how into a sector that is dominated by state-owned monopolies. The inability of the coal sector—where large reserves are present, and where state ownership and price controls play relatively minor roles—to attract significant amounts of private investment is not a hopeful sign. •

The deterioration in energy sector financial performance during a time of sharply rising prices can be explained in part by the declines in consumption, but also by rapid growth in costs. While some of this cost growth reflects the need to build new and refurbish depreciated infrastructure, it may also stem for a lack of effective cost controls when it comes to the procurement of materials (especially fuels), particularly in the electric power sector.


Chapter 3 - Kyrgyzstan’s Energy Sector

Poverty and household access to energy This section reviews issues related to poverty and household access to energy during 2006-2010.114 It focuses in particular on household energy consumption during the energy crisis of 2008 and in response to the tariff increase in 2010.

some 21 percent during 2007-2009 (see Figure 3.16), the survey data show a decline during this time.

Whereas supply-side data indicate that apparent electricity consumption declined by 3 percent in 2009, the survey data indicate that electricity consumption grew in 2009.

Data issues The household surveys conducted by the State Statistical Committee have since 2003 produced data on a number of important indicators. These include: • • • • • • • •

Presence of gas/water/electricity meters; Interruptions in gas/water/electricity services; Types of fuel used for heating; Value of types of fuel used for heating; Heating and cooking expenditures; Bills to be paid; Actual payments; and Volume of energy consumption

These data are extremely important, and require careful analysis. However, their interpretation poses two sets of challenges. First, these data do not always correspond to production and sales data provided by energy companies. For example—whereas: •

supply-side data indicate that apparent household electricity consumption (generation less net exports and losses) declined by 3 percent in 2009 (see Figure 3.2), the survey data indicate that electricity consumption grew in 2009; and

Second, these survey data on household expenditures do not always agree with national income accounting data on consumption expenditure. For example, whereas the survey data indicate that household expenditures in real terms declined by 6 percent in 2009,115 Kyrgyzstan’s national accounts data report a 14 percent reduction in individual consumption for that year. Third, these survey data are not always internally consistent. For example, information on billing for household consumption and actual payments for electricity and gas consumed do not match up: the latter consistently exceed the former by some five percent for gas sales and 10 percent for electricity over a 4.5-year period (January 2006 – mid-2010). While such a variance may be within the statistical margin of error, the size of this discrepancy suggests something more systematic may be at work. Possible explanations include: •

Advance payments—household can pay in advance for future energy consumption. However, while such behavior might happen for 1-2

supply-side data indicate that apparent household coal consumption (production less net exports) rose by 115

114

Data for 2010 are limited to the first half of the year.

Nominal growth in household expenditures in 2009, divided by the change in the consumer price index in 2009.

95


Poverty and Social Impact Assessment

quarters, it does not seem plausible for a 4.5-year period. •

Bank transaction fees incurred when settling energy bills. However, such transactions are quite rare outside of Bishkek; and even there the numbers of users of such services are not significant.

Overestimation of actual expenditures by households—perhaps in light of popular sentiments that energy tariffs and expenditures are too high. However, no such variance among billing and payments is reported for coal, district heating, or other forms of energy.

payments for electricity and gas) are larger for low-income than upper-income households. To the extent that they are not a statistical aberration, these discrepancies can be seen as a “corruption tax” on the poor.

Household energy expenditures

Corruption: households may collude with payment collectors to conceal the full magnitude of energy consumption. Such an explanation can be supported not only by the process of elimination, but also by other research on Kyrgyzstan’s energy sector.116 It is also noteworthy that: o this discrepancy amounted to 79 million som ($1.8 million) for gas and 274 million som ($6.4 million) for electricity, in 2009; o the January 2010 increase in electricity tariffs resulted into a spike in this value during the first quarter of 2010— corresponding to an annualized total of 564 million som ($12.3 million); and o as a share expenditures, discrepancies

of

household these (additional

Aggregate household expenditures during 2006-2009 grew by some 21 percent in real terms, with expenditures in poor and extremely poor households growing as fast as, or slightly faster, than average (Table 3.3). However, expenditure growth for households in the poorest income decile, as well as for rural households, lagged behind other groups.

While such a variance may be within the statistical margin of error, the size of this discrepancy suggests something more systematic may be at work. Household energy consumption— measured as nominal growth in household expenditures on energy goods and services117 divided by the change in the energy components of the consumer price index— also rose during this time (Table 3.5), for households in all income deciles, locations, and social groups monitored. Growth in energy consumption seems to have been particularly rapid in poor and extremely poor households, households in urban areas, and in households headed by a single parent. Growth in energy consumption was more moderate in households living in rural and mountainous areas.

116

The USAID report “The Curtailment of Electricity Consumption: Attitude of the Population and Impact on the Quality of Life” (February 2009) found that “respondents still declare that they pay for electricity regularly and always on time”.

96

117

These include expenditures on electricity, central heating, gas, hot water, coal and other energy sources (chiefly firewood, peat, kerosene, mazut, dung, and bottled liquefied gas).


Chapter 3 - Kyrgyzstan’s Energy Sector

Table 3.3—Kyrgyzstan: Monthly per-capita household expenditures (in som, 2006-2010)

2010 (II quarter) 2250 n.a. n.a. 586

Real growth (cumulative, 2006-2009)* 21% 21% 25% 12%

Households National average Poor Extremely poor 1st decile (poorest)

2006 1272 685 485 488

2007 1579 835 567 473

2008 2253 1316 909 745

2009 2264 1225 895 805

2010 (I quarter) 2198 n.a. n.a. 644

2nd decile 9th decile 10th decile (wealthiest)

567 2037 3218

685 2498 4177

1093 3349 5585

1101 3608 5502

920 3424 5924

919 3431 6419

32% 20% 16%

Urban Rural High mountains Mountains

1534 1117 954 839

1970 1356 1125 961

2729 1985 1812 1768

2865 1934 1843 1499

2856 1810 n.a. n.a.

2701 1985 n.a. n.a.

27% 18% 31% 21%

Single parent With many children With disabled members

1374 1134 1180

1732 1404 1547

2500 2005 1937

2547 1998 2097

n.a. n.a. n.a.

n.a. n.a. n.a.

26% 20% 21%

* Nominal expenditure growth during 2006-2009 deflated by growth in the consumer price index for these years. Source: State Statistical Committee, UNDP calculations.

It is noteworthy that all of this growth in energy consumption occurred during 20062008; declines in overall household expenditures, and especially in household energy consumption, were reported for 2009 (Figure 3.24). The data in Figure 3.25 indicate that the shares of household expenditures devoted to energy increased during 2006-2009, before falling back slightly in the first half of 2010. Expenditures on coal and electricity account for most of this, followed by expenditures on gas (Figure 3.26). However, compared to many other countries, these shares of household expenditures devoted to energy consumption (5.5-6.5 percent) are not particularly large.

the last quarter of the year, as household stockpile coal for the winter. Particularly large shares of household budgets were devoted to energy expenditures in the fourth quarter of 2008 (due to large price increases for coal, as well as exceptionally cold winter temperatures, in that year), as well as in the first quarter of 2010 (due to the large tariff increases for electric and thermal power introduced in January of that year). The data in Figure 3.27 show that energy expenditures absorb a larger share of household budgets in poor families than in high-income families. However, they also show the share of household spending devoted to energy expenditures in low-income households converging towards national averages, by the first half of 2010.

Quarterly seasonality is apparent in these data: energy spending typically rises in

97


Poverty and Social Impact Assessment

Figure 3.24—Kyrgyzstan: Trends in household ousehold expenditures, energy consumption* onsumption* (2007-2009) (2007

Figure 3.25—Kyrgyzstan: Share of household ousehold spending absorbed by energy expenditures xpenditures (2006 (2006-2010)

Consumption expenditure Energy consumption

26%

6.5% 6.3%

6.2%

5.9%

17% 14%

13%

5.4%

2007

2008

2009 6% -6%

-15%

2006

2007

2008

2009

2010:H1

* Nominal expenditure growth deflated by growth in the consumer price index (or its energy component) for these years. Source Source: State Statistical Committee, UNDP calculations.

Figure 3.26—Kyrgyzstan: Household energy expenditures by energy sources (2006 (2006-2010)

50% 40% 30% 20% 10% 0% Coal

2006 Electricity

2007 Gas

2008 Central heat

Note: shares sum to 100 percent. Source: State Statistical Committee, UNDP calculations.

98

2009 Hot water

2010:Q1 Other


Chapter 3 - Kyrgyzstan’s Energy Sector

The data in Figures 3.28 and 3.29 indicate that low-income income households, and households living in rural and especially mountainous areas, have very limited access to gas, central heating, and hot water supply. This near-total total reliance on electricity and coal—especially especially in colder mountainous areas—underscores underscores USAID’s 2009 finding that electricity tariff increases can have

disproportionate effects on real incomes in low-income households.118 It also means that such households are more likely to be affected by interruptions in electricity supplies supplies. This is particularly the case for Kyrgyzstan’s poorest (first decile) households, for whom spending on electricity absorbs about half of energy expenditures. By contrast, urban and upperupper income households are more likely to have

Figure 3.27—Kyrgyzstan: Kyrgyzstan: Household expenditures on energy, by deciles (2006--2010)

2006

2007

2008

2009

2010:H1

8% 7% 6% 5% 4% 1st decile (poorest)

2nd decile

National average

9th decile

10th decile (wealthiest)

Source: State Statistical Committee, UNDP calculations.

Figure 3.28—Kyrgyzstan: Shares of total household energy expenditures devoted to various e energy sources (by household decile, 2009)

120% Hot water

100%

Central heat

80%

Gas 60%

Electricity

40%

Coal

20% 0% 1 decile

2 decile

National average

9 decile

10 decile

Source: State Statistical Committee, UNDP calculations. 118

“The The Curtailment of Electricity Consumption: The Attitude of the Populationn and Impact on the Quality of Life,” USAID 2009. p.4.

99


Poverty and Social Impact Assessment

Figure 3.29—Kyrgyzstan: Shares of total household h energy expenditures devoted to various arious energy sources (by household location, 2009)

100% Hot water

80%

Central heat 60%

Gas

40%

Electricity Coal

20% 0% High mountain

Rural

Mountain

National average

Urban

Source: State Statistical Committee, UNDP calculations.

access to gas, central heating, ating, and hot water. Before the winter of 2008-2009, 2009, quasi quasiuniversal household connections to Kyrgyzstan’s electricity grid meant quasi quasiuniversal access to electricity services. However, the combination of severe frosts, unprecedented low water levels in the Naryn cascade hydropower opower reservoirs, and the decapitalization of the country’s electricity transmission and distribution infrastructure led

to electricity rationing and unplanned service interruptions for households and other users. As the data in Figure 3.30 show, the freq frequency of service interruptions increased abruptly in 2008, when nearly half of Kyrgyzstan’s households reported electricity outages on a daily basis and 70 percent reported several such service interruptions per week. While the situation improved in 2009, daily interruptions still prevailed.119

Figure 3.30—Kyrgyzstan: Share of households h reporting interruptions in electricity service ervice (2006 (2006-2009)

70%

No interruptions

Few times yearly

Once a month

Once a week

Several times weekly

Daily

60% 50% 40% 30% 20% 10% 0% 2006

2007

2008

2009

Source: State Statistical Committee, UNDP calculations. 119

Fortunately, reported interruptions in gas, central heating, and hot water supply were much less frequent.

100


Chapter 3 - Kyrgyzstan’s Energy Sector

Figure 3.31—Kyrgyzstan: Share of households ouseholds (by income decile) experiencing weekly (or more m frequent) interruptions in electricity lectricity service (2008-2009) (2008

77% 73% 72% 63% 65%

1st decile (poorest) 2nd decile National average 9th decile 10th decile (wealthiest) 52% 54% 47% 46%

Figure 3.32—Kyrgyzstan: Share of households (by location) experiencing weekly eekly (or more frequent) interruptions in electricity service ervice (2008 (2008-2009)

93% 85%

82%

72% 55%

Urban National average Rural 78% Mountains High mountains 62% 55% 54% 43%

2008

2009

2008

2009

Source: State Statistical Committee, UNDP calculations.

The data in Figures 3.31 and 3.32 indicate that low-income income households, and especially households in rural and mountainous areas, were more likely to be affected by service interruptions in 2008 and

2009. As these households’ are more likely to rely on electricity for heat as well as light lighting, these data underscore the severity of recent winter hardships.

Table 3.4—Kyrgyzstan: Average per-capita per monthly energy expenditures, xpenditures, by decile group (in som)

2006 69 38 25 25 36 109 150

2007 94 52 30 33 51 152 172

2008 143 74 40 58 85 217 266

2009 148 85 72 55 77 216 297

2010:Q1 200 n.a. n.a. 59 61 323 562

2010:Q2 76 n.a. n.a. 21 23 130 197

Real cum. growth* (2006-2009) 26% 31% 68% 28% 23% 15% 15%

Urban Rural High-mountains Mountains

81 61 59 54

55 87 76 72

156 135 137 125

182 130 110 99

329 124 n.a. n.a.

111 56 n.a. n.a.

31% 23% 9% 8%

Single parents With many children With disabled

80 58 70

108 79 93

168 123 143

185 124 149

n.a. n.a. n.a.

n.a. n.a. n.a.

35% 25% 24%

Household National average Poor Extremely poor 1st decile (poorest) 2nd decile 9th decile 10th decile (wealthiest)

* Nominal expenditure growth during 2006-2009 2009 deflated by growth in the energy component of the consumer price index for these years. Source: State Statistical Committee, UNDP calculations.

101


Poverty and Social Impact Assessment

also means that such households are more likely to be affected by interruptions in electricity supplies. By contrast, urban and upper-income households are more likely to have access to gas, central heating, and hot water.

Conclusions •

Growth in energy consumption seems to have been particularly rapid in poor and extremely poor households, households in urban areas, and in households headed by a single parent. Growth in energy consumption was more moderate in households living in rural and mountainous areas.

The share of household expenditures devoted to energy increased during 2006-2009, but compared to many other countries, they are not particularly large (5.5-6.5 percent). While energy expenditures absorb a larger share of household budgets in poor families than in high-income families, the share of household spending devoted to energy expenditures in low-income households had converged towards national averages by the first half of 2010.

102

Both household expenditures in general, and on energy in particular, increased in real terms during 20062009. However, all this growth occurred during 2006-2008; declines in real household expenditures and especially energy consumption were reported in 2009.

Low-income households, and households in rural and mountainous areas, have very limited access to gas, central heating, and hot water supply. They tend to rely almost exclusively on coal and electricity for heating, as well as illumination. This is particularly the case for Kyrgyzstan’s poorest (first decile) households, for whom spending on electricity absorbs about half of energy expenditures. It

Shortages have broken the link between connection to the grid and access to reliable electricity supplies. Until 2008, quasi-universal household connections to Kyrgyzstan’s electricity grid meant quasi-universal access to electricity services. However, the winter energy crisis in that year led to a sharp increase in the frequency of service interruptions: nearly half of Kyrgyzstan’s households reported electricity outages on a daily basis; while 70 percent reported several such service interruptions per week. While the situation improved in 2009, daily interruptions still prevailed.

Low-income households, and especially households in rural and mountainous areas, are most likely to be affected by service interruptions. As these households’ rely on electricity for heat as well as lighting, these data underscore the severity of the recent winter hardships experienced.

Households apparently pay more for electricity and gas than billing information suggests. While this discrepancy can be explained in various ways, corruptive collusion among households and bill collectors may be responsible for this outcome. If so, this “corruption tax” seems to be a particularly heavy burden on lowincome households.


Chapter 3 - Kyrgyzstan’s Energy Sector

Social protection and the energy sector Overview The Soviet social protection system Kyrgyzstan inherited after independence was poorly prepared for the financial and institutional challenges of transition during the 1990s. In order to maintain macroeconomic stability and reduce poverty, many reforms were subsequently launched, in order to increase the financial sustainability of social policies and better target social support to vulnerable groups. Before 1998, state benefits were calculated on the basis of the minimum wage. However, because many public employees are paid the minimum wage, fiscal constraints often constrained increases in the minimum wage, in the social payments linked to it. In order to separate public-sector wage policies from poverty reduction efforts, social assistance payments were decoupled from the minimum wage in 2009, with social assistance payments determined by trends in the guaranteed minimum income. This indicator, which is set by the government on an annual basis, reflects fiscal and macroeconomic variables, as well as minimum subsistence considerations for poor households. Despite these efforts, the results of these social policy reforms have at best been partial. The effectiveness of social protection remains constrained by a lack of uniform policies, and imperfect coordination among local governments, ministries, and social service providers. As a result, many social benefits continue to be received by non-poor households. The tragic events of April-June 2010 both revealed and aggravated problems in the social sector; inter alia by demonstrating the government’s inability to provide effective protection to households perceived as being victims of the large energy tariff increases.

Despite these efforts, the results of these social policy reforms have at best been partial.

Social policy instruments The pension system dominates social insurance provision, as retirement and disability benefits paid from the Social Fund provide at least three quarters of the transfer payments received by households. Reforms conducted over the past two decades have tightened control over insurance contributions and more closely linked pension benefits to Social Fund contributions made by the insured. These measures have strengthened the financial sustainability of the Social Fund and allowed for increases in pension benefits. Despite this progress, the ratio of the average pension to the average salary (the “replacement rate”) in 2009 was only 35 percent. This is not enough to keep many pensioners out of poverty: as the data in Figure 3.33 show, the average monthly pension benefit is well below the subsistence minimum. Moreover, its “pay-as-you-go” character leaves Kyrgyzstan’s pension system vulnerable to demographic changes, migration outflows, and informal employment at home. These reduce contributions to the Social Fund and its ability to provide current and future beneficiaries with adequate pension benefits. By the end of 2010, there were only 1.8 social insurance contributors per pensioner. If farmers and self-employed individuals (who do not make large contributions to the pension system) are excluded from this calculation, this dependency ratio drops to 1.2 contributors per pensioner. This raises important questions

103


Poverty and Social Impact Assessment

about the sustainability of Kyrgyzstan’s pension system.

means-tested monthly benefit to poor families with children.120

In order to address these imbalances, a funded pillar was introduced into the pension system in January 2010. Under this system, 2 percent of contributors’ wages are allocated to individual pension accounts. Funds in these accounts are invested by the Social Fund, in order to accumulate resources that can be used to finance beneficiaries’ subsequent retirement. However, Kyrgyzstan’s relatively shallow financial system and regulatory uncertainties raise questions about the risks of such an approach.

Monthly social benefits. The MSB is available to:

The ratio of the average pension to the average salary in 2009 was only 35 percent. This is not enough to keep many pensioners out of poverty.

In addition to old age benefits, pensions are paid in cases of disabilities and loss of breadwinners. Pension benefits are also supplemented for various social assistance objectives—such as occurred prior to the large increase in electric and thermal power tariffs introduced in January 2010—and for special classes of pensioners (e.g., military men; individuals, who live and work in highmountain or remote areas; mothers with many or disabled children; workers of companies that have been liquidated; etc.).

• • •

people living with disabilities; children who have lost one or both parents; the elderly who have reached retirement age but are not receiving a pension; and others who, due to age or disease, are not able to provide for their own welfare.

Since January 2010, the MSB ranges from 1,000-2,000 som ($21-$42) per month, which is well below the subsistence minimum. This is because the value of the guaranteed minimum income (which determines MSB eligibility) is set below the extreme poverty line.

Monthly benefits to poor families with children, which are paid to families with children, whose incomes are below the guaranteed minimum income level (which is currently 310 som ($7) per month). The size of the PFMB is determined by the difference between the guaranteed monthly income and actual per-capita monthly income in an applicant household. Other benefits paid to poor families with children under this programme include: •

The value of the guaranteed minimum income is set below the extreme poverty line.

Social assistance is provided via categorical benefits, the monthly social benefit (for people living with disabilities and for families who have lost a breadwinner), and the

120

a lump-sum allowance equal to 300 percent of the guaranteed monthly income, paid when a child is born; an allowance for children below age three—100 percent of the guaranteed monthly income; an allowance for twins—100 percent of the guaranteed monthly income upon their birth; an allowance for triplets (and larger births)—150 percent of the guaranteed

Before 2010 this was called the universal monthly benefit.

104


Chapter 3 - Kyrgyzstan’s Energy Sector

monthly income upon the birth of each child. The PFMB is Kyrgyzstan’s only national means tested social assistance benefit. In addition to calculating recipient households’ monetary and in-kind (e.g., from agriculture) incomes, family assets (e.g., durable goods, draft animals) are also taken into account. As a result, it performs reasonably well in terms of targeting accuracy. World Bank research121 finds that, in 2005, the poorest 40 percent of the population received 81 percent of total PFMB benefits paid out; and that close to 75 percent of the PFMB beneficiaries were in this group. However, as the data in Figure 3.35 show, the PFMB is even more modest than the average monthly pension and social benefit. As with the MSB, this is because the value of the guaranteed minimum income (which determines PFMB eligibility) is set below the extreme poverty line.

Categorical benefits. Although it has undergone reforms in the past two decades, social assistance in Kyrgyzstan retains many characteristics of the categorical benefits system inherited from the Soviet time. State benefits were assigned to offset low wages, to recognize distinctive merits of various groups, and to remediate the health consequences of natural and anthropomorphic disasters. Before 2006 there were 38 beneficiary categories, compensation for 29 of which was provided in-kind. During 2006-2007 two additional types of cash benefits were added (for housing, communal services and energy, as well as for public transport. As of today there are 25 categories of beneficiaries and eight types of privileged services in Kyrgyzstan, financed by the national (and sometimes local) budget(s).122

121

Social Safety Net in the Kyrgyz Republic: Capitalizing on Achievements and Addressing New Challenges, World Bank, May 20, 2009. 122 Significant in-kind benefits are also provided by public- (and private-sector) enterprises to their

The monetization of categorical benefits. In January 2010 the number of categorical benefits was reduced from 38 to 25 categories, while the number of privileged services was reduced from 40 to eight and was fully monetized. Benefits providing households with cash compensation to purchase coal and meet their monthly energy bills were also monetized. The size of these cash benefits (which typically range from 1000 to 7000 som per month) are based on the value of benefits provided before January

As is often the case with social reform in transition economies, the monetization of categorical benefits in Kyrgyzstan met with considerable popular resistance. 2010, and are financed both from national and local government budgets. The elimination of 13 benefit categories also helped reduce the number of beneficiaries from some 281,200 to 53,500 (as of 1 October 2010). These include individuals with incomes above the subsistence level, and individuals of active working age not otherwise suffering from disabilities or disadvantages. As is often the case with social reform in transition economies, the monetization of categorical benefits in Kyrgyzstan met with considerable popular resistance. In addition to helping to stoke the popular unrest that led to the overthrow of the Bakiyev government in April 2010, this resistance took the form of refusals of World War II invalids living in Bishkek to collect cash compensations in lieu of the in-kind subsidies they used to receive for housing, energy, and communal services. In the end, Kyrgyzstan seems likely to continue living with a “patchwork” social protection system consisting of “old-style” categorical benefits, social assistance that is employees (e.g., vacation vouchers), and by donors to low-income or vulnerable households (e.g., foodstuffs).

105


Poverty and Social Impact Assessment

targeted at low-income households, s, and social insurance (e.g., pensions) that absorbs the lion’s share of social policy resources and therefore has the largest overall impact in terms of poverty reduction. In any case, until recently pensions have in many respects been Kyrgyzstan’s most mos effective social policy instrument, in terms ter of poverty reduction.. The pension system does not explicitly target low-income income households. However, because it contains the lion’s share of Kyrgyzstan’s social protection resources, and because so many households lds are living close to or below the poverty line, the pension system is the only social policy instrument with the depth and breadth of influence household incomes in a great many families.

recession caused by the April-June April political events. These increases were made possible by large increases in budget support grants provided by international financial institutions and bilateral donors, which helped offset large fiscal deficits in these years. Whether this fiscal and social policy response to Kyrgyzstan’s socio-economic economic tensions is sustainable remains to be seen.

Unfortunately, only about half of Kyrgyzstan’s social benefits were received by low low-income households. Did this social policy response help to

Figure 3.34—Kyrgyzstan: Trends in the distribution d of social benefits* by household deciles eciles (2006-2010) (2006

2006

60% 52%

2007

51%

50% 50%

2008

42% 43%

42%

38%

2009

30%

2010:H1

13%

11% 7% 6% 7%

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 10 (wealthiest)

* Pensions plus PFMB, MSB, categorical benefits, benefits paid by enterprises. Source: State Statistical Committee, UNDP calculations.

Effectiveness of social protection Social protection expenditures expanded dramatically during ring 2008-2010. 2008 While these increases resulted in part from the monetization of categorical benefits that had previously been delivered in-kind, kind, they also reflected the government’s attempts to ameliorate the social impact of the global gl financial crisis (in 2009), and then of the 2010 106

shield Kyrgyzstan’s most vulnerable households from the impact of the unfavourable macroeconomic trends during 2009-2010? Since (according to these data) social benefits (including pensions) only comprise about 10 percent of household income, it would be difficult for social policy instruments to have a profound impact on household welfare. However, if these social benefits its were concentrated in low-income low households, they could have an important effect.


Chapter 3 - Kyrgyzstan’s Energy Sector

Unfortunately, as the data in Figure 3.34 show, only about half of Kyrgyzstan’s social benefits were received by low low-income households (i.e., the lowest three deciles of the t income distribution). This share actually dropped slightly during 2009-2010 2010 (to 50 percent, down from 52 percent in 2008). By contrast, the share of social benefits accruing to upper-income income households more than doubled (from 6 to 13 percent) during 20082008 2010. This increase came primarily at the expense of the middle deciles in the income distribution, whose share of total social benefits paid out dropped to 38 percent (it had been 51 percent in 2007). Thus, the lion’s share of the increased social policy response to the crisis developments of 2009-2010 2010—much of which was financed by donors— —seems to have leaked to relatively wealthy households. Figure 3.35—Kyrgyzstan: Trends in the distribution of pension benefits by household deciles (2008-2010) (2008

2008

increases in pension spending during 2009 20092010: the share of pension benefits accruing to relatively wealthy households rose from 23 percent in 2008 to 38 percent during the first half of 2010. By contrast, the share of pension benefits accruing to low-income income households dropped from 24 to 14 percent during this time. These data weaken the case for viewing pensions as a poverty-reduction reduction instrument in Kyrgyzstan. The data in Figure 3.36 indicate that the simplification and monetization of categorical benefits its introduced in 2010 deepened—or at least made more explicit explicit— the regressive character of Kyrgyzstan’s social policy framework. The share of categorical benefits—including including subsidies for eenergy and communal services—received received by low-income low Figure 3.36—Kyrgyzstan: Trends in the distribution of categorical benefits by household deciles (2008 (2008-2010)

2008

65%

53% 55% 2009

2009

48% 2010:H1

2010:H1 38%

46% 43% 42% 38%

28% 24%

31%

23% 17%

20% 14% 11% 4%

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 9 (wealthiest)

Since pensions account for the bulk of social benefits paid out, this result is driven primarily by trends in pension spending. As the data in Figure 3.35 show, upper-income upper households benefitted handsomely from

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9 9-10 (wealthiest)

households dropped from 46 percent in 2009 to only 20 percent during the first half of 2010. (This share had been at 65 percent in 2008.) By contrast, the share of categorical benefits accruing to upper-income come households rose from 11 to 38 percent during this time. These 107


Poverty and Social Impact Assessment

Figure 3.37—Kyrgyzstan: Trends in the distribution d of PFMB benefits by household deciles (2008-2010) (2008

Figure 3.38—Kyrgyzstan: Trends in the distribution of MSB benefits by household deciles eciles (2008-2010) (2008

2008 64%

2009

56% 49%

2008 66%

2010:H1

2009

61% 57%

2010:H1

47% 40% 33%

30% 22%

22% 17% 13%

12% 4% 4% 3% Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 9 (wealthiest)

trends strengthen the case for means-testing means categorical benefits.

These data weaken the case for viewing pensions as a povertypoverty reduction instrument in Kyrgyzstan. The data in Figure 3.37,, regarding the distribution of transfers under the poor family monthly benefits programme, tell a more hopeful story: they show the share of total benefits accruing to poor families rising from f 49 percent in 2008 to 64 percent during the first half of 2010. They also indicate that a very small share of these benefits leak to upper-income income households. These trends suggest that Kyrgyzstan’s social policy institutions are increasingly able to direct dir means-tested tested benefits to the household that need them the most. 108

Deciles 1-3 (poorest)

Deciles 4-8 (middle class)

Deciles 9-10 (wealthiest)

Unfortunately, trends in the monthly social benefit payments paint a less cheerful picture: the share of these benefits going to low-income income households dropped from 57 percent in 2008 to only ly 12 percent during the first half of 2010 (Figure 3.38). 3.38 Significant increases in leakages to middle middle- and upperincome households seem to have occurred during this time. These trends question the rationale for considering the monthly social benefit to be an instrument for poverty reduction. They also strengthen the case for means-testing testing categorical benefits. In sum, these data suggest that Kyrgyzstan’s social protection system has become less able to direct benefits to the most needy households during the he past two years. This has implications for the system’s ability to protect vulnerable households from possible future increases in energy prices as well.


Chapter 3 - Kyrgyzstan’s Energy Sector

protection system that apparently emerged during 2009-2010 (described above).

Social protection and the energy sector Social and energy policies in Kyrgyzstan have traditionally been linked in four ways: 1. Categorical benefits providing households with cash compensation to purchase coal and meet their monthly energy bills; 2. The use of lifeline electricity tariffs, under which the consumption of small amounts of electricity was billed at lower rates; 3. Decisions about tariff increases, in which government bodies have sought to address social concerns as well as financial sustainability issues; and 4. Decisions to raise pensions, the PFMB and MSB, in order to provide compensation for higher energy tariffs and prices.

In light of the relatively small share of household spending apparently devoted to energy expenditures, the small likelihood that household electricity tariffs will be increased in the short- (and possibly medium) term, and the difficulties in targeting social benefits to poor households, the case for more closely linking social and energy policies is difficult to make. The issue would instead seem to be one of adopting policies to improve the functioning of the energy sector and the social protection system. In this respect, important changes would include:

• By January 2010, links (1) and (2) had essentially been abandoned. Arguments about transparency and administrative simplicity had led to the cancellation of lifeline tariffs and the monetization of categorical benefits related to energy consumption. Instead, the emphasis switched to (3) and (4). The consequences of this switch included the:

More closely linking the PFMB to the guaranteed minimum income, which should itself be more closely linked to the minimum subsistence level; Means-testing the MSB and categorical benefits, to reduce their regressive character; and Consideration of the reintroduction of lifeline electricity tariffs. Reductions in tariffs for small volumes of household electricity consumption could be offset by higher tariffs for consumption above this level, thereby leaving average tariff levels unchanged.

social unrest evoked by the January 2010 energy tariff increases and the monetization of social benefits, followed by the April 2010 rescinding of some (not all) of these tariff increases—without corresponding reductions in social benefits;

The case for more closely linking social and energy policies is difficult to make.

post-April 2010 social contract not to raise household energy tariffs for an unspecified period of time—signaling the ascension of “social” over “economic” criteria in energy sector tariff policy; and

Conclusions

consolidation of the more regressive character of Kyrgyzstan’s social

Two decades of reforming Kyrgyzstan’s social policy framework have failed to produce positive results, both in general and in terms of links with the energy sector in particular. 109


Poverty and Social Impact Assessment

The effectiveness of social protection remains constrained by a lack of uniform policies, and imperfect coordination among local governments, ministries, and social service providers. As a result, many social benefits continue to leak to nonpoor households. The tragic events of April-June 2010 both revealed and aggravated problems in the social sector, inter alia by demonstrating the government’s inability to provide effective protection to households perceived as being victims of the large energy tariff increases.

110

The monetization of categorical benefits in Kyrgyzstan—particularly in 2010—met with considerable popular resistance. Kyrgyzstan seems likely to continue living with a “patchwork” social protection system consisting of “old-style” categorical benefits, social assistance that is targeted at lowincome households, and social insurance (e.g., pensions) that absorbs the lion’s share of social policy resources and therefore has the largest overall impact in terms of poverty reduction. The most recent household survey data indicate that only about half of Kyrgyzstan’s social benefits were received by low-income households (i.e., the lowest three deciles of the income distribution). This share actually dropped slightly during 20092010 (to 50 percent, from 52 percent in 2008). By contrast, the share of social benefits accruing to upper-income households more than doubled (from 6 to 13 percent) during 2008-2010. Much of the increased social spending in response to the crisis developments of 2009-2010—significant shares of which were financed by donors— seems to have leaked to relatively wealthy households. These data suggest that Kyrgyzstan’s social

protection system has become less able to direct benefits to the most needy households during the past two years. This has implications for the system’s ability to protect vulnerable households from possible future increases in energy prices as well.

Since pensions account for the bulk of social benefits paid out, this result is driven primarily by trends in pension spending. Upper-income households seem to have benefitted handsomely from increases in pension spending during 2009-2010: the share of pension benefits accruing to relatively wealthy households rose from 23 percent in 2008 to 38 percent during the first half of 2010. By contrast, the share of pension benefits accruing to lowincome households dropped from 24 to 14 percent during this time. These data weaken the case for viewing pensions as a poverty-reduction instrument in Kyrgyzstan.

The simplification and monetization of categorical benefits introduced in 2010 seems to have deepened the regressive character of Kyrgyzstan’s social policy framework. The share of categorical benefits—including subsidies for energy and communal services— received by low-income households dropped from 46 percent in 2009 to only 20 percent during the first half of 2010. (This share had been at 65 percent in 2008.) By contrast, the share of categorical benefits accruing to upper-income households rose from 11 to 38 percent during this time.

Likewise, the share of the monthly social benefits (accruing primarily to people with disabilities, to households that have lost a breadwinner, and to retirees not receiving old-age pensions) going to low-income households dropped from 57 percent in 2008 to only 12 percent during the first half of


Chapter 3 - Kyrgyzstan’s Energy Sector

2010. These trends question the rationale for considering the monthly social benefit to be an instrument for poverty reduction.

By contrast, the share of the poor family monthly benefits (Kyrgyzstan’s only national means tested social policy instrument) accruing to lowincome families rose from 49 percent in 2008 to 64 percent during the first half of 2010. Likewise, very small shares (3-4 percent) of these benefits leak to upper-income households. These data suggest that Kyrgyzstan’s social policy institutions are increasingly able to direct means-tested benefits to the household that need them the most. Unfortunately, the benefits paid out under this programme are so small as to have only a minimal poverty-reduction impact. These data strengthen the case for transferring resources from other social protection programmes to the PFMB programme, and for means testing categorical benefits and the monthly social benefit. In light of the relatively small share of household spending apparently devoted to energy expenditures, the small likelihood that household electricity tariffs will be increased in the short- (and possibly medium) term, and the difficulties in targeting social

benefits to poor households, the case for more closely linking social and energy policies is difficult to make. The issue would instead seem to be one of adopting policies to improve the functioning of the energy sector and the social protection system. In this respect, important changes would include:

o More closely linking the PFMB to the guaranteed minimum income, which should itself be more closely linked to the minimum subsistence level; o Means-testing the MSB and categorical benefits, to reduce their regressive character; and o Consideration of the reintroduction of lifeline electricity tariffs. Reductions in tariffs for small volumes of household electricity consumption could be offset by higher tariffs for consumption above this level, thereby leaving average tariff levels unchanged.

111


Poverty and Social Impact Assessment

Conclusions and recommendations Conclusions Production and sales data indicate that Kyrgyzstan during 2007-2010 has experienced significant increases in energy prices and reductions in energy production and consumption. These data indicate that electricity generation and consumption (generation less net exports and losses) dropped by some 25 and 13 percent, respectively, during 2007-2010. Gas consumption dropped by almost two thirds, as prices of gas imported from Uzbekistan rose from $100 to $240 per 1000 cubic meters during 2007-2009. Faced with growing shortages of centrally supplied electricity and unaffordable gas, many households, businesses, and public institutions switched to coal-fired boilers. Domestic coal production rose some 41 percent during 2007-2010; apparent consumption (production less net exports) rose 57 percent. On the whole, household energy prices during 2007-2010 rose by 81 percent, compared to a 44 percent increase in the consumer price index during this time. This combination of sharply higher prices and declining consumption raises concerns about deterioration in household access to reliable, affordable energy services. Despite these inflation rates, energy tariffs and prices generally seem low, relative to other transition economies and to costrecovery levels in Kyrgyzstan. This reflects strong socio-political opposition to raising tariffs. Household energy consumption is currently being subsidized, from the state budget (which provides up to 75 percent of the revenues received by thermal power producer Kyrgyzzhilkommunsoyuz), from municipal budgets, via cross-subsidies from electricity exports, and from future generations, who will ultimately have to cover the differences between revenues and costs in the energy sector today—in the form of higher tariffs, higher taxes, or poorer services. Since most of these subsidies increase with the quantity of 112

energy consumed, which in turn is linked to income, they are not likely to help protect Kyrgyzstan’s most vulnerable households.

Future generations will ultimately have to cover the differences between revenues and costs in the energy sector today. Strategies based on tariff rebalancing, unbundling, and privatization via sales to strategic (often foreign) investors that have underpinned energy sector reforms in many transition economies have run aground in Kyrgyzstan. Privatization measures introduced during the Bakiyev period have been repealed, and electricity tariff increases seem to be off the table for the foreseeable future. The emphasis is instead on reducing corruption by strengthening state control and civic engagement. Whether this approach can significantly reduce losses and attract the capital and expertise needed to modernize Kyrgyzstan’s energy sector, and improve thereby national and household energy security, remains to be seen. Kyrgyzstan’s electricity, thermal, gas, and (to a smaller extent) coal sectors seem likely to remain dominated by state-owned monopolies that generally report negative profitability, high technical and commercial losses, and (with the exception of the gas sector), high or growing depreciation rates (reaching 60 percent for fixed assets in the thermal power sector). Many of these sectors report large accounts receivable, payable, or both. Outside of the coal sector, private capital and foreign investment are almost completely absent. Progress towards the creation of a regional electricity market, which could help attract foreign investment into Kyrgyzstan’s


Chapter 3 - Kyrgyzstan’s Energy Sector

electricity generation and transmission sectors, remains halting at best. The deterioration in energy sector financial performance during a time of sharply rising prices can be explained in part by the declines in consumption, but also by rapid growth in costs. While some of this cost growth reflects the need to build new and refurbish depreciated infrastructure, it may also stem from ineffective cost controls when it comes to the procurement of materials (especially fuels), particularly in the electric power sector. Some important progress has been made since the onset of the winter energy crisis in 2007-2008. Collection rates in the electricity sector have improved significantly as the quasi-fiscal deficit has dropped; fixed asset depreciation in the gas sector has fallen; and the basic legal framework for the development of decentralized renewable energy technologies has been introduced. New electricity generation capacity has been brought on line, particularly in terms of the Kambarata-2 hydropower plant. The Fuel and Energy Sector Transparency Initiative is improving regulatory governance and increasing civic engagement. Still, taken as a whole, the scale of Kyrgyzstan’s energy problems remains daunting. Particularly worrisome is absence of a concept for attracting private capital and know-how into a sector that is dominated by state-owned monopolies. The inability of the coal sector—where large reserves are present and where state ownership and price controls play smaller roles—to attract significant amounts of private investment is not a hopeful sign. Official survey data indicate that energy consumption increased during 20062009. However, all this growth occurred during 2006-2008; large declines in energy consumption were reported in 2009. Growth in energy consumption seems to have been particularly rapid in poor and extremely poor

households, households in urban areas, and in households headed by a single parent. Growth in energy consumption was more moderate in households living in rural and mountainous areas. While the shares of household expenditures devoted to energy increased during 2006-2009, compared to many other countries, they are not particularly large (5.56.5 percent). And while energy expenditures absorb a larger share of household budgets in poor families than in high-income families, the share of household spending devoted to energy expenditures in low-income households had converged towards national averages by the first half of 2010. These survey data confirm that lowincome households, and households in rural and mountainous areas, have very limited access to gas, central heating, and hot water supply. They tend to rely almost exclusively on coal and electricity for heating, as well as for light. This is particularly the case for Kyrgyzstan’s poorest households, for whom spending on electricity absorbs about half of energy expenditures. It also means that such households are more likely to be affected by interruptions in electricity supplies. By contrast, urban and upper-income households are more likely to have access to gas, central heating, and hot water.

The Fuel and Energy Sector Transparency Initiative is improving regulatory governance and increasing civic engagement. Shortages have broken the link between connection to the grid and access to reliable electricity supplies. Until 2008, quasiuniversal household connections to Kyrgyzstan’s electricity grid meant quasiuniversal access to electricity services. However, the winter energy crisis in that year 113


Poverty and Social Impact Assessment

led to a sharp increase in the frequency of service interruptions: nearly half of Kyrgyzstan’s households reported electricity outages on a daily basis; while 70 percent reported several such service interruptions per week. While the situation has improved since then, service outages continue. Survey data indicate that low-income households, and especially households in rural and mountainous areas, are most likely to be affected by service interruptions. As these households’ rely on electricity for heat as well as lighting, these data underscore the severity of the recent winter hardships experienced. The survey data also indicate that households pay more for electricity and gas than billing information suggests. While this discrepancy can be explained in various ways, corruptive collusion among households and bill collectors may be responsible for this outcome. If so, this “corruption tax” seems to be a particularly heavy burden on low-income households.

The tragic events of April-June 2010 both revealed and aggravated problems in the social sector. Two decades of reforming Kyrgyzstan’s social policy framework have produced mixed results, both in general and in terms of links with the energy sector. Many social benefits continue to leak to non-poor households. The tragic events of April-June 2010 both revealed and aggravated problems in the social sector, inter alia by demonstrating the government’s inability to provide effective protection to households perceived as victims of the large energy tariff increases. Kyrgyzstan seems likely to continue living with a “patchwork” social protection system consisting of “old-style” categorical benefits, social assistance that is targeted at low-income households, and social insurance (e.g., pensions) that absorbs the lion’s share of social policy resources and therefore has the 114

largest overall impact in terms of poverty reduction. The official survey data indicate that only about half of Kyrgyzstan’s social benefits are received by low-income households. This share actually dropped slightly during 20092010 (to 50 percent, down from 52 percent in 2008). By contrast, the share of social benefits accruing to upper-income households more than doubled (from 6 to 13 percent) during 2008-2010. Much of the increased social spending in response to the crisis developments of 2009-2010—significant shares of which were financed by donors— seems to have leaked to relatively wealthy households. These data suggest that Kyrgyzstan’s social protection system has become less able to direct benefits to the most needy households. This has implications for the system’s ability to protect vulnerable households from possible future increases in energy prices as well. Since pensions account for the bulk of social benefits paid out, this result is driven primarily by trends in pension spending. Upper-income households seem to have benefitted handsomely from increases in pension spending during 2009-2010: the share of pension benefits accruing to relatively wealthy households rose from 23 percent in 2008 to 38 percent during the first half of 2010. By contrast, the share of pension benefits accruing to low-income households dropped from 24 to 14 percent during this time. These data weaken the case for viewing pensions as a poverty-reduction instrument in Kyrgyzstan. The simplification and monetization of categorical benefits introduced in 2010 seems to have deepened the regressive character of Kyrgyzstan’s social policy framework. The share of categorical benefits—including subsidies for energy and communal services— received by low-income households dropped from 46 percent in 2009 to only 20 percent during the first half of 2010. (This share had been at 65 percent in 2008.) By contrast, the


Chapter 3 - Kyrgyzstan’s Energy Sector

share of categorical benefits accruing to upperincome households rose from 11 to 38 percent during this time. Likewise, the share of the monthly social benefits (accruing primarily to people with disabilities, to households that have lost a breadwinner, and to retirees not receiving old-age pensions) going to lowincome households dropped from 57 percent in 2008 to only 12 percent during the first half of 2010. These trends question the rationale for considering the monthly social benefit to be an instrument for poverty reduction. By contrast, the share of the poor family monthly benefits programme (Kyrgyzstan’s only national means tested social policy instrument) accruing to lowincome families rose from 49 percent in 2008 to 64 percent during the first half of 2010. Likewise, very small shares (3-4 percent) of these benefits leak to upper-income households. These data suggest that Kyrgyzstan’s social policy institutions do have the capacity to direct means-tested benefits to the households that need them the most. Unfortunately, the benefits paid out under this programme are so small as to have only a minimal impact on poverty. These data strengthen the case for transferring resources from other social protection programmes to the PFMB programme, and for means testing categorical benefits and the monthly social benefit.

More closely linking the PFMB to the guaranteed minimum income, which should itself be more closely linked to the minimum subsistence level;

Means-testing the MSB and categorical benefits, to reduce their regressive character; and

The possible reintroduction of lifeline electricity tariffs. Reductions in tariffs for small volumes of household electricity consumption could be offset by higher tariffs for consumption above this level, thereby leaving average tariff levels unchanged.

A number of important research questions have been identified in this report. These pertain to:

Improvements in the quality of household survey and production/sales data regarding the energy sector, in order to remove inconsistencies within and between these data sets;

Developing possible scenarios for the future of Kyrgyzstan’s energy sector;

Improving corporate governance in the energy sector;

Identifying appropriate energy saving technologies, and policies and programmes to accelerate their introduction;

Strengthening the role of affordability analyses in regulating energy tariff increases;

Analysis of obstacles to the accelerated development of small hydropower plants and other decentralized renewable energy technologies, with proposed solutions;

Recommendations In light of the relatively small share of household spending apparently devoted to energy expenditures, the small likelihood that household electricity tariffs will be increased in the short- (and possibly medium) term, and the difficulties in targeting social benefits to poor households, the case for more closely linking social and energy policies is difficult to make. The issue would instead seem to be one of adopting policies to improve the functioning of the energy sector and the social protection system. In this respect, important changes would include:

115


Poverty and Social Impact Assessment

116

•

Analysis of the costs of electric and thermal power production and tariff setting; and

•

Analysis of the results of the Fuel and Energy Sector Transparency Initiative, as well as possibly expanding its coverage and channeling its findings to decisionmakers and the public, in order to more effectively advocate for prompt improvements.


Chapter 4: Tajikistan’s Energy Sector

Authors: Lilit V. Melikyan, Hasmik Ghukassyan Editor: Ben Slay

117


Poverty and Social Impact Assessment

Executive summary •

The combined impact of population growth, the decapitalization of the energy infrastructure, interruptions in gas imports from Uzbekistan, and the dissolution of the regional electricity transmission network has had a dramatic impact on the energy sector of Tajikistan. This was particularly apparent during the winter of 2007-2008, when unusually cold weather brought the energy sector to the brink of collapse. The international financial crisis and rising food prices completed the picture (which some experts call a compound crisis), the most pronounced adverse effects of which were felt by the poor and vulnerable. The poor, who in 2007 constituted more than half of the population, suffered most from the power cuts, lack of heating, reductions in remittances, and increasing food prices. Official data indicate that electricity consumption dropped some 8 percent during 2007-2009, before growing slightly in 2010. Meanwhile, the annual increases in household electricity tariffs during 2007-2010 averaged 57 percent; energy price inflation averaged 42 percent annually.

The situation has improved somewhat since then: the income poverty rate dropped to 47 percent in 2009, while growth in remittances and GDP accelerated in 2010. The Sangtuda-1 hydropower plant was brought into operation; the construction of Sangtuda-2 is progressing; the electricity transmission infrastructure has been extended; and significant numbers of small hydropower plants are coming on line.

The energy sector is still in crisis however, with winter power cuts (up to 10-12 hours a day) continuing in much of the country. The dissolution of the Central Asian electricity transmission network, limited and increasingly expensive gas supplies from Uzbekistan, and an underdeveloped

118

coal sector have left Tajikistan almost solely reliant on hydropower generation, which remains insufficient in the winter. An estimated one million people spend much of the winter (six weeks more) without access to reliable electricity supplies. Despite accounting for nearly three quarters of the total population, households in rural areas during 20082010 only accounted for 8-11 percent of the country’s electricity consumption.123

Improving national and household energy security requires the investment of billions of dollars in electricity generation (including in small-scale hydropower plants), transmission, and distribution, in the gas supply network, and in the coal sector. Unless significant foreign capital inflows are attracted, Tajikistan’s energy investment needs are unlikely to be met. The public debt is approaching levels regarded by the IMF as “inadvisable”, due in part to the construction of the Roghun hydropower plant.

While some reforms have taken place in Tajikistan’s energy sector, overall progress has been slow. The main electricity company (Barqi Tojik) is a vertically integrated state owned monopoly, which suffers from high losses, low tariff collection rates, and high arrears; the main debtors are the TALCO aluminum company, the irrigation sector, and households. While electricity tariffs have been increasing during the last few years (as per agreement with the international financial institutions), they remain below cost recovery levels.

Tariffs have been increasing also for natural gas. Household consumption of electricity and natural gas has declined in the recent years, due to a combined effect

123

“Intermediate Strategy for Renewable Energy Sources”, p. 8.


Chapter 4 - Tajikistan’s Energy Sector

of increasing tariffs and supply cuts. Households have responded by consuming more liquefied petroleum gas (the prices of which were subsidized until recently), coal and firewood—with the attendant adverse implications for the environment (due to extensive deforestation) and health (due to poor air quality). However in 2010 the consumption of liquefied petroleum gas, as well as coal, declined due to increasing tariffs and prices.

The average combined energy basket takes around 10 percent of the average household budget and 16 percent for the poorest quintile. For selected groups of the poorest households, using combinations of heating sources, the share of energy expenses in household spending during the winter time reaches up to 55 percent.

The burden of energy expenditures on households’ budgets could become even heavier in years to come due to increasing tariffs and prices, unless Tajikistan’s social protection becomes much more effective than is the case at present. While the current social assistance scheme features compensation for electricity and gas use— up to certain limits, for around 240,000 households (1.5 million people) who are identified by the local authorities—these benefits are not well targeted. And at $2 per month, they are not large enough to have a perceptible impact on household incomes.

With help from the World Bank and the European Union, the government has introduced a pilot cash transfer scheme based on proxy means testing. The results of the pilot are to be assessed after its December 2012 completion, potentially to be followed by a national roll out. While experience from similar countries suggests that this is likely the best option for Tajikistan, the full scale implementation of this reform seems likely to run at least 4-5 years. As tariff increases are set to

continue during this time, hardships for poor households seem likely to increase.

We suggest the possible reintroduction of lifeline electricity tariff regime at Barqi Tojik (one was in place until 2007), as a transitional measure during this period. Simulations based on Tajikistan’s living standards survey database indicate that, if underwritten by relatively small subsidies (around $3.5 million annually, which hopefully could be grant financed by donors), lifeline tariffs could reduce the severity of income poverty by 5 percent, compared to current levels. The effectiveness of this transitional lifeline scheme could be further enhanced if it were coupled with additional categorical targeting—if tariffs were further reduced for large households with children, since this is a good predictor of poverty in Tajikistan.

The experience of Pamir Energy—Central Asia’s first public-private partnership in the energy sector, which manages Tajikistan’s electricity infrastructure in remote, sparsely populated Gorno Badakhshan on a 25-year concession— supports this argument. Pamir Energy operates a lifeline tariff regime, and has performed well in terms of generation, sales, losses and collections.

International experience suggests that subsidies for household purchases of liquefied petroleum gas (which were in place before 2009), and to offset households’ one-off costs of installing or upgrading meters and connecting to electricity and gas grids, can boost access to clean energy and reduce household energy insecurity. Expanded government and donor support for sustainable forestry and firewood harvesting practices may also be worthy of consideration.

The experience of Pamir Energy shows also that improvements in Tajikistan’s energy regulatory framework can yield 119


Poverty and Social Impact Assessment

large benefits in terms of increased household energy security. In particular, the pursuit of similar improvements in the regulatory environment for small scale hydropower (and other decentralized renewables) will accelerate implementation of the national small hydropower programme, which will be important in improving access to affordable electricity, as well as providing new income- and employment generation opportunities for much of the rural population.

120

decentralized renewable energy technologies. While the numbers of small hydropower plants in Tajikistan are growing, many of these operate only partially, and a sustainable model for their operation is yet to emerge. An improved regulatory framework is needed for such a model to develop, along with better managerial capacity for small hydropower plants.

The expansion of small hydropower plants (and where appropriate, other decentralized renewable technologies) seems to offer the best short-term prospects for bringing reliable year-round electricity services to the estimated 1 million people who do not currently have them. Experience from UNDP pilot projects suggests that the construction of 1000 small hydropower plants with average capacity of 100 kilowatts (in the 30-500 kilowatt range) could provide these households with one kilowatt of electricity—sufficient for year-round access to indoor lighting. The dozens of communities that have developed their own small hydropower installations (often with donor assistance) were until recently unable to sell surplus power back to Barqi Tojik’s grid; many do not have access to the grid. While now there is a purchase obligation for the electricity they produce, at the moment of writing this report only one small HPP is connected to the grid. In December 2010 the Ministry of Energy and Industry approved a model contact for small hydropower plants that wish to connect to the grid, as well as a methodology for tariff calculation. While tariffs still have to be agreed between small hydropower generators and Barqi Tojik, this has brought Tajikistan a step closer to the introduction of feed-in tariffs, as well as increased commercial financing of

Other important strategic policy decisions facing Tajikistan’s energy sector include the design and implementation of:

o A national heating strategy, to guide investment decisions in the electricity, gas, coal, and district heating sectors. The design and implementation of such a strategy would also help develop more effective assistance programmes for energy poor households; o An effective national energy efficiency/energy saving programme. Potential energy savings are estimated at 30 percent of current power consumption by the Ministry of Energy and Industry; other estimates place these savings at 60 percent for rural areas. While some basic laws are in place, these are not sufficient to induce the energy conservation investments that Tajikistan needs. The creation of the UNDP proposed National Renewable Energy and Energy Efficiency Trust Fund could be an important step in this respect;124 and o The financing and regulatory dimensions of the national programme on small hydro power and placing the decentralized renewable energy solutions at the heart of rural development efforts; and so on.

124

See “National Programme for Renewable Energy Sources Based Integrated Rural Development: National Scaling Up”, UNDP-Tajikistan, October 2010.


Chapter 4 - Tajikistan’s Energy Sector

Introduction Household ld energy tariffs and prices increased sharply in Tajikistan during 20072007 2010 (see Figure 4.1). ). These increases occurred against a background of significant winter energy shortages, and attendant hardships for the half of the population living below the national poverty line (17.5 percent of which is classified as living in extreme poverty). Inadequate household access to energy services is a significant barrier to attaining the Millennium Development Goals, and sustainable human development, in Tajikistan. These problems reflect the combined impact of population growth, the decapitalization of Tajikistan thermal and electric ic power infrastructure, interruptions in

gas imports from Uzbekistan, the dissolution of the regional electricity transmission network, and slow progress in modernizing the energy sector. Tajikistan’s main energy service provider, Barqi Tojik, faces signi significant challenges in improving service delivery, raising collection rates, and reducing high technical and commercial losses. While foreign investment and grant financing has been attracted into power generation (e.g., in the Sangtuda-1 and -22 hydropower pplants, Pamir Energy, and in dozens of small, mini miniand micro-hydro hydro power plants), and while electricity supplies have improved since the winter of 2007-2008, much uch of the country remains without electricity supply for significant portions of the day during the winter months.

Figure 4.1—Tajikistan: Tajikistan: Annual increase in household tariffs

97% 78%

69%

Electricity

Gas

59% 39%

35%

38% 11%

2007

2008

2009

2010

Source: UNDP calculations, based on State Statistical Agency consumer price data.

121


Poverty and Social Impact Assessment

Sectoral overview Legal framework. The government of Tajikistan has pursued a two-track approach to energy reform. The first track focuses on reforming the domestic energy sector, energy pricing, financial discipline, and institutional setup. The second focuses on creating an export strategy for electricity. A 2007 energy sector risk assessment carried out by the Asian Development Bank (ADB)125 noted that the legal framework is incomplete. The existing legislation does not efficiently enough stimulate competition in the energy sector. However, with the restructuring of Barqi Tojik underway (see below), there is a reasonable chance that the legal framework in the energy sector will further improve. While there is some private investment in gas, coal, and power generation, the rules for market entry remain vague. There is an urgent need to improve the regulatory and legal frameworks to stimulate private investment to meet growing domestic demand for energy and to capture the export potential of the electricity sector.126

Institutional framework. Most of Tajikistan’s energy companies are publicly owned vertically integrated monopolies. Ownership functions are formally exercised by the Committee on State Property. The Fuel and Energy Department in the Office of the President has considerable influence over these companies. Also important are the Ministry of Energy and Industry (which is responsible for energy policy); the Antimonopoly Commission, under the Prime Minister’s office (which sets electricity, natural gas, and district heat tariffs, in consultation with relevant ministries); and the Ministry of Environmental Protection (responsible for environmental regulation).

125

ADB (2007): Technical Assistance for Governance and Capacity Development Initiative; Manila. 126 Ibid.

122

The Asian Development Bank’s energy sector risk assessment noted that:

the management of public funds and procurement is unpredictable, not linked to market needs, and only partially transparent;

while the process for energy tariff setting has become more formal and transparent, significant improvements are still needed;

there is no comprehensive overall sector planning;

sectoral corruption significant, including:

risks

are

o projects are selected on the basis of political priorities, regional considerations, and preferences of external investors, rather than economic criteria; o funding has focused mainly on new projects, rather than upgrading existing investments and improving regulation; o staffing decisions are sometimes based on patronage considerations; and o failure to optimize the use of resources over the longer term, in order to promote short-term gains.


Chapter 4 - Tajikistan’s Energy Sector

Some progress has been made in reforming energy sector governance, particularly in terms of:127

discriminatory access to Barqi Tojik’s transmission grid.

Barqi Tojik’s separation from the Ministry of Energy and Industry in 2006;

The establishment and working of Pamir Energy (a private company owned by the Aga Khan Development Foundation and financed by the International Finance Corporation and the Swiss Development Cooperation) for electricity generation and distribution in the Gorno Badakhshan region, under a 25-year concession;

The construction of the Sangtuda-1 hydropower station, as a joint venture between the government (with a 25 percent ownership stake) and the Russian RAO-UES utility (with a 75 percent ownership stake); and the Sangtuda-2 hydropower station as a standalone build-operate-transfer project with a 12.5 year concession (co-financed by the government of Iran); and

Energy system capacity. During the Soviet period, 60 percent of Tajikistan’s energy needs were met with diesel fuel produced in other Soviet republics. By 2005, this share had dropped to 43 percent. The latest available data for Tajikistan’s total primary energy supply (TPES) are from 2008. During 2005-2008 the share of hydropower rose from 42 to 56 percent, at the expense of oil (see Figure 4.2). This diminishing role of oil and gas imports accelerated in 2009-2010 due to rising prices of gas imported from Uzbekistan, as well as further growth in the role of hydropower. (Tajikistan’s economically feasible hydropower potential is estimated to be 264 billion kWh per year,128 of which only about 6 percent has been harnessed so far.)

The design and implementation of a national programme for small hydropower production, under which some progress has been made in creating a framework for independent power production and non-

Tajikistan’s energy production consists mainly of hydropower plants (4900 megawatts of installed capacity—94 percent of the national total), and of the Dushanbe and Yavan combined heat and electricity power plants (318 megawatts, 6 percent of the national total), which contribute electricity during seasonal peaks and shortages.

Figure 4.2—Tajikistan: Total primary energy supply (2005, 2008)

2008, %

2005, %

oil

1 4

gas

22

43

42

hydro 56

18 coal

14

Source: IEA 127

ADB: “Country Partnership Strategy: Tajikistan, 2010–2014”, Manila.

128

World Energy Council (2007): “Survey of Energy Resources: Tajikistan”.

123


Poverty and Social Impact Assessment

Table 4.1—Tajikistan: Energy system production capacity (in megawatts)

Name of the power plant

Installed

Available

Average annual working

Hydropower (HPPs) and thermal power plants (TPPs) 1

Nurek HPP

3000

3000

2715

2

Sangtuda-1 HPP

670

670

670

3

Baipaza HPP

600

600

600

4

Kayrakkum HPP

126

126

85

5

Vakhsh Cascade, including

285

235

122

5.1

Golovnaya HPP

240

200

120

5.2

Perepadnaya HPP

30

24

16

5.3

Centralnaya HPP

15

15

11

6

Varzob cascade, including

25

14

8.5

6.1

HPP-1

7

4

4

6.2

HPP_2

15

6

4

6.3

HPP_3

3.5

3.5

0

7

Pamir and Khorog HPPs

37

37

37

8

Small scale HPPs

60

56

56

Thermal and diesel power plants 9

Dushanbe TPP

198

198

100

10

Yavan TPP

120

120

60

11

Diesel PPs, including

60

56

56

11.1

Mobile

9

7.5

7.5

5090

5070

4470

Total

Source: Adapted from the Electricity Governance Initiative in Tajikistan: Institutional and Practical Analysis, Final report, July 2010

124


Chapter 4 - Tajikistan’s Energy Sector

an organizational structure that is inconsistent with the scale of operations;

inadequate control over financial, business, and branch operations; and

the absence of a development strategy, a long term plan, treasury, and risk management functions.

Electricity sector In addition to directly accounting for 5 percent of Tajikistan’s GDP, electricity generation is a key input for the production of the country’s largest export – aluminum – and is essential for the electric pumps that are mainstays of Tajikistan’s irrigated agriculture. Barqi Tojik is responsible for generation, transmission, and distribution in the whole of Tajikistan, except in the Gorno Badakhshan Autonomous Oblast (where privately owned Pamir Energy operates most power facilities) and some small hydro facilities. Barqi Tojik is a vertically integrated utility. It is an open joint stock company in which all shares belong to the state, managed by a chairman who is appointed by, and reports to, the President of Tajikistan. While there is no board of directors, there is a supervisory board comprising senior government ministers and chaired by the prime minister. Four deputy chairmen are responsible for specific portfolios (i.e., generation, distribution, transmission, sales, finance, etc.). According to the Asian Development Bank, Barqi Tojik is effectively managed according to geographic (rather than thematic or functional) criteria, which reduces accountability and complicates the management of commercial activities.129 Barqi Tojik—structural evolution. According the Asian Development Bank report, Barqi Tojik’s financial and operational performance suffers from a number of longrunning, fundamental problems.130 These include:

The ADB report recommends strengthening corporate governance arrangements, in order to address these weaknesses, raise investor confidence, and prepare the electricity sector for future liberalization. Barqi Tojik’s first restructuring plan was developed in May 2006, within the framework of the Asian Development Bank’s It technical assistance programme.131 recommended Barqi Tojik’s unbundling in three phases, in order to improve the company’s financial performance and attract private investment. According to the ADB report, this first restructuring plan received ministerial approval in May 2006 but did not receive full governmental approval until September 2008. In a letter written to the ADB at this time, Tajikistan’s then deputy prime minister expressed renewed support for this restructuring plan, and requested the revision of the timetable for implementing the three phases. According to the plan submitted by the government in 2009, Barqi Tojik’s three-phase restructuring is to proceed as follows:132 •

129

Asian Development Bank, “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009. 130 Based on an analysis provided by the BDO Unicon accountancy firm, Management Letter, OSHPC “Barqi Tojik”, issued in June 2008 for the year ending 2007 accounting period.

Phase 1 (2009-2012) focuses on improving corporate governance and financial management. Key emphases include the creation of an organizational structure based on its three main divisions (i.e., electricity generation, transmission, and

131

ADB (2006): “TA 1817-TAJ – Power Sector Restructuring Plan”; Manila. 132 ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009

125


Poverty and Social Impact Assessment

distribution), with each division to have separate accounting; improving financial reporting and transparency; and designing and implementing a financial and operational performance improvement programme. Tariff, legal, and regulatory reforms are to be designed, and a detailed implementation plan developed, for government approval. This phase also foresees the establishment of an independent sectoral regulatory body, to create an appropriate enabling environment for attracting private investment during the subsequent phases.

2012” (this is a revision of two older decrees from 2003 and 2008), inter alia by restructuring natural monopolies and large enterprises, so as to accommodate potential concession arrangements and management contracts, reorganizations, and/or privatizations. The list includes 27 energy companies, including Barqi Tojik and the Dushanbe and Yavan combined heat and power plants.

Phase 2 (2013-2015) envisions—if deemed appropriate—Barqi Tojik’s dissolution into (or divestiture of) separate, legally independent stateowned enterprises for generation, transmission, and distribution. The links between these enterprises are to be commercial and regulatory in nature. In this phase large-scale independent power generation companies are to be allowed to enter the market and given access to the state-owned transmission network, subject to clearly defined agreements. A detailed implementation plan is to be developed for government approval.

Phase 3 (2016-2018): the government will evaluate the results of the first two phases and consider privatizing generation and distribution assets. The transmission network will remain under state ownership; equal access to the grid will be offered to all players. A detailed implementation plan is to be developed, for government approval.

Opinions about Barqi Tojik’s unbundling, concerning overall desirability as well as details, timing, and modalities, vary among policy makers and experts. International experience shows that unbundling vertically integrated electricity utilities can improve sectoral performance by increasing transparency, promoting competition (via lower barriers to entry for independent power generators) and attracting investment. However, experience from Central Asia suggests that such measures need to be adapted to suit the specific conditions of each country.134 For small economies, horizontal unbundling into small entities may not be desirable—particularly if economies of scale or scope are lost, or if the government bodies responsible for energy sector development are unable to effectively coordinate investment projects that cut across generation, transmission, and distribution activities.

In line with the action plan for the first phase, a March 2010 government decree instructs various agencies to start the implementation of the “Strategic programme of privatization of medium and large enterprises and restructuring of natural monopolies and large companies for 2009126

The next step is to design a detailed implementation plan for restructuring these companies, under the guidance of the State Property Committee, the Ministry of Energy and Industry, and Barqi Tojik.133

133

ADB, “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009.

134

World Bank (2010b): Lights Out? The Outlook for Energy in Eastern Europe and the Former Soviet Union, Washington DC.


Chapter 4 - Tajikistan’s Energy Sector

Barqi Tojik’s hydropower assets are described in Table 4.2; Table 4.3 describes the Dushanbe and Yavan combined heat and power plants, which are now being converted from gas to coal. Table 4.2—Tajikistan: Barqi Tojik’s hydropower plants

Installed capacity (megawatts)

Annual production (gigawatts/ hour)

With the exception of Sangtuda-1, Tajikistan’s power plants are old, with an average age of over 30 years. Barqi Tajik transmission assets: Tajikistan effectively has two separate electrical networks: a northern grid in the Sughd region and a southern grid, both linked to Uzbekistan (with the northern grid also linked to Kyrgyzstan) and a separate network in Gorno Badakhshan (which has only select connections to the national grid). The northern and southern grids were connected only in December of 2009.

Name

Type

Nurek*

Water reservoir

3200

11850

Baipaza

Run-ofriver

600

2525

Run-ofriver

670

2970

Name

Golovnaya** Run-ofriver

240

840

Perepadnaya **

Run-ofriver

30

250

Centralnaya* Run-of* river

30

125

Source: The Electricity Governance Initiative in Tajikistan: Institutional and Practical Analysis, Final report, July 2010

Varzob

Run-ofriver

25

205

Kayrakkum

Run-ofriver

125

755

4921

19520

Tajikistan’s transmission network was mostly built during the 1960s and 1970s. Due to the lack of investment in maintenance and shortage of spare parts, this aging transmission infrastructure has undergone significant physical deterioration, leaving the system extremely vulnerable to weather-related or other shocks.

Sangtuda 1

TOTAL

Table 4.3—Tajikistan’s thermal power plants

* Includes expansion of Nurek’s capacity from 3000 to 3200 megawatts. ** Includes expansion of capacity from 255 to 280 megawatts.

Installed capacity (megawatts)

Annual production (gigawatts/hour)

Dushanbe

198

1300

Yavan

120

790

TOTAL

318

2090

Source: Barqi Tojik

127


Poverty and Social Impact Assessment

Figure 4.3—Tajikistan: Investments in the energy sector (million US$) 180

167.4

160 from own resources (BT), million USD

140

125.9

121.1

from external sources, million USD

120 100 80 60 40

26.9

24.3

20 0.0

1.5

0.0

1.6

0.5

6.8

8.9

11.3

7.6

0.9

12.3

9.9

10.3

0 2002

2003

2004

2005

2006

2007

2008

2009

2010

Source: Barqi Tojik Investment Programme 2010

As shown in Figure 4.3, internal investments were at very low levels until 2006, and are still low now. While the total losses decreased in the recent, they are still high by international standards. Tajikistan’s electricity transmission system was constructed during the Soviet period as an integral part of the Central Asian regional electricity transmission system. The trunk transmission and 110-500 kilovolt power lines under this system were managed by Barqi Tojik’s central dispatching unit. The “Energy” Central Co-ordination Centre (located in Tashkent) served as the nerve centre of this integrated regional electricity system, which

128

was also interconnected with Russia’s Unified Electric Systems grid. In late 2009 Uzbekistan decided to withdraw from the Central Asian integrated power transmission network. As a result, Tajikistan is no longer able to automatically import electricity from Uzbekistan (or electricity from Turkmenistan that is transshipped through Uzbekistan). In addition to reflecting tensions between the two countries, these developments reveal different degrees of interest in regional electricity sector cooperation, as well as significant differences in domestic electricity tariffs in the Central Asian countries.


Chapter 4 - Tajikistan’s Energy Sector

Figure 4.4—Tajikistan: Monthly electricity imports and exports 1000 Monthly electricity imports (kWh) 800 Monthly electricity exports (kWh) 600 Dissolution of the integrated Central Asian electricity transmission system

400 200 0

Source: UNDP, based on data from State Statistical Agency

These developments did not change the region’s transmission infrastructure; nor did they preclude continued electricity exports and imports. Instead, they signified the ascension of bilateral (as opposed to multilateral) electricity trade, and a greater emphasis on the construction of national transmission infrastructure, to ensure domestic electricity supplies. However, reliance on more costly national generation and transmission infrastructure drives up power costs; one study estimated this cost differential at 5 percent, due to higher dispatch and fuel costs alone.135 It can also reduce electricity supply security, particularly during winter season peak demand. Longer term, the cessation of common management of the region’s transmission infrastructure seems likely to lead to the construction of by-pass national infrastructure, further boosting costs in the sector.

Tajikistan since late 2009 has therefore seen a precipitate drop in electricity exports and imports (see Figure 4.4). Since (in contrast to the other Central Asian countries) Tajikistan is a net electricity importer, these declines have complicated the country’s recovery from the 2007-2008 winter crisis. Matters would have been worse if Tajikistan had not borrowed $340 million from China’s ExImBank, in order to complete the southnorth transmission line that reduces Tajikistan’s reliance on imports from Uzbekistan. Tajikistan’s electricity balance is described in Table 4.4. The data demonstrate the decline in electricity generation and consumption starting from 2007. The data in Figure 4.ZZ illustrate the large share of electricity devoted to industry and irrigated agriculture.

Table 4.4—Tajikistan’s electricity balance (million KWh)

2003

2004

2005

2006

2007

2008

2009

Generation

16509

16401

17090

16935

17494

16147

16117

Import

4605

4810

4637

5022

4552

6404

6003

Export

4596

4466

4402

4429

4464

5539

5960

Consumption

16518

16835

17325

17528

17582

17012

16160

Source: Statistics Yearbook, Tajikistan 2010 135

Central Asia Power System Study, Mercados, 2010.

129


Poverty and Social Impact Assessment

Figure 4.5—Tajikistan: Shares of main electricity consumption categories (% %)

Figure 4.6—Tajikistan: Tajikistan: Electricity consumption ((million kWh) 10,000

16%

17%

18%

13%

16%

16%

17%

20%

22%

20%

18%

22%

Losses

47%

47%

2007

2008

7802

Households, others

6,000 4,000 2,000

45%

3908

3612

3144

3359

3044.3

2906

3722 3680

0

2006

Industry 2006

8060

7007

Irrigation 46%

8080

8,000

2007

2008

2009

2009 Population

industry

Irrigation

Source: Statistical Yearbook Tajikistan, 2010 and Barqi Tojik

TALCO, the giant aluminum company and Tajikistan’s largest exporter, accounts for around 35 percent of total electricity consumption. This is roughly double the share of electricity consumed by households. While collections have improved compared to 2004 (see Table 4.5), ), they are still low by international ational and even regional norms.

overlook non-payment.)136 Coupled with non non-

TALCO accounts for around 35 percent of total electricity consumption. This is roughly double the share of electricity consumed by households.

Table 4.5—Tajikistan: Total customer payments of Barqi Tojik

Customer payments (percent of billings)

2004

2009

54%

72%

Source: Barqi Tajik.

With tariffs rising towards cost recovery levels, accounts receivable have grown for both electricity and gas in the recent few years. However, for certain sectors, collections are still very low. Particularly low are the collection rates from TALCO, irrigation, communal service providers, and public sector institutions. As of 6 March 201 2011, TALCO’s arrears stood at 151 million somoni ($34 million). According to Asian Development Bank research, household arrears result from the failure to collect bills in time, and poor financial management. management (Electrical bills are collected door-to-door, doo thus leaving bill collectors open to bribes to 130

transparent tariff setting policies, these billing and financial management issues produce a vicious cycle, in which part of the increases in tariffs is necessitated ssitated by these debts—a debts burden shared by the residential sector as a whole and the poor in particular.

136

ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009.


Chapter 4 - Tajikistan’s Energy Sector

Table 4.6—Tajikistan: Electricity tariffs (US¢/kWh without VAT, unless otherwise noted)

Industry

01/ 2006

02/ 2007

07/ 2007

01/ 2008

05/ 2008

01/ 2009

08/ 2009

01/ 2010

0.81

1.05

1.28

1.54

3.19

3.39

4.24

4.87

1.25

1.25

1.27

8.3 with VAT

TALCO Nonindustry

Special Budget

14.9 0.52

0.58

0.70

0.85

1.27

1.35

1.69

1.94

Communal service providers

0.52

0.58

0.70

0.85

1.27

1.35

1.68

1.94

Electric transport

0.15

0.23

0.29

0.35

1.27

0.90

1.12

1.30

Irrigation pumps

Oct-May

0.31

0.38

0.47

0.57

0.87

0.75

0.96

1.30 with VAT

May-Oct

0.15

0.19

0.23

0.28

0.41

0.65

0.75

0.78

1.40

1.24

1.58

2.06 with VAT

Below 250 KW/h

0.58

0.70

Over 250 KW/h

0.99

1.10

0.48

0.67

0.87

1.26

1.30

1.56

Households, average

Average increase

0.49

Source: Barqi Tojik137

Electricity tariff trends are shown in Table 4.6. In 2008 tariff categories were reduced to five customer groups; favorable tariffs for low-usage customers and special summer and winter tariffs for TALCO were eliminated. Over the last several years tariffs have increased sharply, with the latest (20 percent) increase taking place on 1 January 2010.138 Despite these increases, current tariffs are still below long-run cost recovery levels. According to ADB (2006) the marginal cost of supply (loss adjusted) was $0.028/kWh in the

137

US$ to Somoni exchange rates (based on the information from the National Bank of Tajikistan): 2006 2007 2008 2009 2010 3.2733 3.4412 3.4335 4.0361 4.3748

138

Tariffs were not raised in July 2010 and in January 2011, as scheduled.

summer and $0.045/kWh in winter.139 Electricity tariffs are nonetheless increasing in line with agreements reached between the government and the World Bank140 (the initial 2003 agreement was with the Asian Development Bank, when it was incorporated into ADB’s Power Rehabilitation Project as a loan condition). According to these agreements, tariffs are to increase twice a year until they reach 9.67 dirham ($0.025/kWh in constant price terms), which is taken to be the lowest level for cost recovery. Table 4.7 describes the plan for average tariff increases; Table 4.8 shows increases that were planned for households.

139

ADB (2006): “Tajikistan: Power Rehabilitation Phase II, TA Consultant Report; Project Number: 34515”: Prepared by Hydro Electric Corporation.

140

Development Financing Agreement (Energy Loss Reduction Project) between the Republic of Tajikistan and International Development Association, September 08, 2005.

131


Poverty and Social Impact Assessment Table 4.7—Tajikistan: Average electricity tariffs (without VAT)

Year

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

Dirham/KWh

1.62

1.67

2.23

3.96

5.73

7.77

8.45

9.08

9.67

9.67

9.67

9.67

US¢/KWh

0.52

0.51

0.65

1.15

1.42

1.78

2.01

2.25

2.5

2.5

2.5

2.5141

Source: Barqi Tojik

Tariffs were/are to be increased twice a year, according to the agreement with the World Bank. In the summer of 2010, in the aftermath of the financial crisis, the government decided to skip the increases planned for July 2010, January 2011, and July 2011.

government and the international community, helping Tajikistan respond to this challenge.142

Source: Forecast parameters of social and economic development of Tajikistan 2009-2013 * Estimates based on projected exchange rates for 2011, 2012 and 2013

Established in December 2002, Pamir Energy took control of most of Barqi Tojik’s assets in Gorno Badakhshan on the basis of a 25-year concession agreement. The Aga Khan Development Foundation and the International Finance Corporation are Pamir Energy’s private shareholders. The original financing was a 45 percent equity and 55 percent debt mix, on a total project cost of $26.4 million.143 The Swiss Economic Cooperation Office provided $5 million in grants for consulting and subsidies. The Aga Khan Foundation subsequently invested around $4 million between 2003 and 2007 in project support, bringing funding levels to around $31 million.144

Tariff setting principles stipulate that tariffs need to cover general establishment and administrative expenses, repair and maintenance costs, and depreciation, as well as allow for a profit margin: the basis of the latter, is however, unclear. The costs are determined by Barqi Tojik’s Planning Department and, and together with the proposed tariffs are submitted to the Antimonopoly Commission for approval.

142

Table 4.8—Tajikistan: Planned tariff increase for households, with VAT Year

2011

2012

2013

Tariff, dirham per kWh

10.8

11.9

13.8

Tariff, US$ per kWh

$0.026

$0.029

$0.036*

Pamir Energy. Gorno Badakhshan Autonomous Oblast is Tajikistan’s poorest region, sparsely populated and cut off during the winter from the rest of the country. Providing electricity services in this region is a daunting challenge. Pamir Energy represents an innovative partnership between the 141

Estimates based on the assumed exchange rates for 2011 and 2012 by UNDP

132

See “Tajikistan: Reducing Poverty through Private Infrastructure Services. The Pamir Private Power Project, A case study from Reducing Poverty, Sustaining Growth, What Works, What Doesn’t, and Why”, A Global Exchange for Scaling Up Success. Scaling Up Poverty Reduction: A Global Learning Process and Conference, Shanghai, May 25–27, 2004. 143 The Aga Khan Foundation’s equity contribution was $8.2 million, while the International Finance Corporation provided $3.5 million in equity and $4.5 million in debt. The International Development Association provided a $10 million loan through the government of Tajikistan. 144 Pamir Energy Information note prepared for the current report.


Chapter 4 - Tajikistan’s Energy Sector Figure 4.7—Tajikistan: Tajikistan: Pamir Energy electricity supply (hours) and losses osses (%)

hours per day

24 23

20

34%

Ratio of losses to total power generated

24%

19

16 12

23.5

17

21% 16% 12%

12

8 4 0 2006 2007 2008 2009 2010

2006

2007

2008

2009

2010

Source: Pamir Energy

The concession agreement, which initiated Central Asia’s first public-private public partnership implemented in the energy sector, sets out the legal, regulatory, technical, operational, environmental, and financial framework, as well as the tariff schedule. Pamir Energy provides electricity and winter heat to areas that otherwise would rely on chopped brush, dried dung and coal. During 2002-2005, 2005, Pamir Energy focused on the construction of new and the reconstruction of additional generation capacity. Pamir Energy has invested some $30

million in reconstructing old or building new hydro plants, substations, transmission, and distribution lines—investments investments which have increased installed generation capacity in Gorno Badakhshan by 30 percent (to 42 megawatts). Despite many challenges, the original rehabilitation project was largel largely completed on time and on budget. The company has since continued to perform well operationally operationally— thanks, in part to subsidies provided by the venture’s international partners.

Table 4.9—Tajikistan: Tajikistan: Pamir Energy summary statistics (2010)

Indicator

Value

Service Area Population (estimated) Number of households in Gorno Badakhshan Number of consumers serviced by Pamir Energy ** Residential (96%) ** Government organizations (3%) ** Commercial organizations (1%) Number of consumers on main grid Number of consumers on Pamir Energy independent mini-hydro mini grids Number of consumers connected to non- Pamir Energy mini hydro plant run by communities without other power supplies Number of mini-hydro plants/total capacity Number of larger hydro plants/total capacity Installed generation capacity Distribution Network ** 35 kilovolt lines ** 10 kilovolt lines ** 0.4 kilovolt lines ** 35/10 kilovolt substations ** Smaller sub-stations

64,000 km2 213,000 32,500 29,623 28,347 866 410 19,870 9,753 2,877 9/5.2 MW 2/37 MW 42 MW 513 km 1,123 km 2,026 km 16 782

Source: Pamir Energy

133


Poverty and Social Impact Assessment

Figure 4.8—Tajikistan: Tajikistan: Pamir Energy electricity generation and sales

178

168 151

174

163 146

141

144 Generation

114

112

Sales Losses 57 37

37

2006

2007

2008

29

2009

19

2010

In billion kWhs Source: Pamir Energy

A 2007 USAID study145 found that relations with Pamir Energy’s 30,000 customers were difficult at the start. Some consumers sued to try to obtain power at no cost; there were metering disputes and large non-payment payment problems. The authors conclude that the Aga Khan Foundation’s patient, long longterm approach, and the unique public and private nature of the financing, helped sustain the project where other, purely private investors may have given up. After a decline in electricity generation and sales in 2007 the company grew strongly in 2008 and 2009. Some decline was registered in 2009-2010, 2009 but revenues stayed strong, partly due to improved collections and reduced losses (see Figure 4.8 and Table 4.10). The he company has installed some 12,000 meters, so that 84 percent of all consumption is now metered. Another 3,500 meters are to be installed in 2011, raising this

level to 90 percent. Tariffs at Pamir Energy have risen more rapidly than at Barqi Tojik: the original concession agreement stipulated that until 2010 tariffs should be increased twice a year by 15 percent. In January 2010, Pamir Energy (and Barqi Tojik) raised electricity tariffs by 12 percent,, so that Gorno Badakhshan residents pay 0.12 Somoni ($0.02) per kWh of electricity. Administrative and commercial customers pay 0.22 Somoni ($0.05) per kWh. From 2009, Pamir Energy removed the winter and summer tariff structure and implemented one tariff based on different user categories. At present, Pamir Energy’s household electricity tariffs are one third higher than Barqi Tojik’s.

Table 4.10—Tajikistan: Tajikistan: Pamir Energy electricity tariffs (US¢/kWh)

Residential tariffs

2004

2005

2006

2007

2008

Summer

0.68

0.84

1.03

1.12

1.74

Winter

0.95

1.18

1.45

1.57

2.43

Source: Pamir Energy 145

USAID (2007); “Regional Energy Markets Assistance Program (REMAP) for Central Asia”, The United States Energy Association.

134


Chapter 4 - Tajikistan’s Energy Sector

Subsidies have played an important role in financing Pamir Energy’s operations since the company’s inception.146 The region’s low household incomes, the time needed for households to adjust to energy being supplied on commercial principles, and the investments needed to maintain and expand service coverage meant that electricity consumption subsidies would be required for a number of years. Consequently, funds for social protection costs were mobilized at the same time as the funds for capital expenditures. The funds mostly came from donors: partly from the spread between the interest rate at which the International Development Association lent to the government and the rate at which the government lent to the private investor, as well as from $5 million in grant funding provided by the Swiss Economic Cooperation Organization. Pamir Energy’s tariff structure has incorporated early year and lifeline subsidies. The early year subsidy was designed to help households make the transition to commercial tariffs and to compensate Pamir Energy for sub-commercial tariffs in its early years of operation: some $1.6 million in early year subsidies were disbursed before full tariffs were applied in 2008. The lifeline subsidy was designed to address affordability constraints for poor residential customers, who comprise more than 90 percent of total customers in Gorno Badakhshan. However as an adequate mechanism to target this subsidy towards the poorest customers was not available, it has been provided to all residential customers. The lifeline subsidy threshold was determined by the “Electrowatt” (now “Poyry”) consulting company, which assessed minimal electricity needs at 200 kWh per month in winter and 50 kWh per month in summer. For consumption within these limits, the customer pays 0.25US¢/kWh. For consumption beyond this threshold, customers pay the full rate. Hence, for most residential customers, the proportion 146

This section relies heavily on information provided by Pamir Energy in January 2011.

of total consumption that is subsidized through the lifeline component depends on consumption volumes, which vary across the regions due to affordability and availability factors. Customers with high levels of consumption (such as in Khorog, where average annual consumption is some 700 kWh—significantly in excess of the lifeline threshold) have a lower proportion of their consumption subsidised. Pamir Energy has nine small hydropower stations. All areas served by these small hydropower plants (except for Darvaz, which receives energy from Barqi Tojik and from the Shirg hydropower plant) have limited power supply, i.e. they do not fully meet demand in the areas they serve. More than 4,000 customers, generally those off the main grid in areas like Murghab and Vanj (constituting around 15 percent of the Gorno Badakhshan customer base) have limited access to power, with average winter consumption well below the threshold level. These customers have all of their consumption subsidized, and pay only the lifeline rate. The maximum total annual payment for customers under the lifeline threshold is $3.38 per annum, although this amount in Somoni terms increased during 2009-2010, with the somoni’s depreciation.

More than 4,000 customers have limited access to power, with average winter consumption well below the threshold level. With the expiry of the early year subsidy in 2007 and Pamir Energy’s increasing energy delivery (due to better winter performance and lower losses), subsidy funds were contributed only by the Swiss Economic Cooperation Organization. Under current projections, Swiss funding is to expire in 2011—implying financial self-sufficiency for Pamir Energy. This projection assumes further tariff increases, and substantial growth in demand in off-main grid regions due to 135


Poverty and Social Impact Assessment

better performance of Pamir Energy’s small power plants. However, the $27,015 projected year-end deficit for 2011 suggests that Pamir Energy may still need some operating subsidies from donors. If no additional subsidies are forthcoming, some 30-40 percent of customers may be unable to finance their current electricity consumption levels.

Electricity sector investment needs and funding. The energy crisis that took hold in the winter of 2007-2008 led the government to intensify efforts to finance major investments in its power sector, and to better use its own resources to improve national energy security and increase Tajikistan’s electricity export potential. Since then, investments have been attracted into the electricity sector from the governments of Russia, Iran, and China. The investments were channelled into the construction of the Sangtuda-1 (Russia) and Sangtuda-2 (Iran) hydropower plants, and of the south-north electricity transmission line (China). Figure 4.9 describes the committed and realized investments from internal and external sources. The largest projects already funded and functioning include:

Rehabilitation of the Centralnaya hydropower plant (30 megawatts), with a credit from the Asian Development Bank;

Development of a feasibility study for the modernization of the Kayrakkum hydropower plant (126 megawatts),

The Varzob hydropower plants cascade (25 megawatts), via a USTDA grant; and

The Nurek hydropower plant upgrade and rehabilitation (taking the installed capacity from 3000 to 3200 megawatts).

The Sangtuda-1 hydropower plant is located on the Vakhsh river in the Khatlon 136

region, 160 kilometers south of Dushanbe. It consists of four units with total capacity of 670 megawatts and producing 2.7 KWh of electricity per year. Sangtuda-1, one of Tajikistan’s three largest hydropower plants, along with Nurek and Baipaza (which are also located along the Vakhsh cascade), was officially commissioned in July 2009. Some $720 million were invested in its construction during the 2005-2009 period. The company that owns Sangtuda-1 is a Russian-Tajikistani joint venture: 75 percent of the shares initially belonged to Russian government and commercial partners,147 while the government of Tajikistan held 25 percent plus one of the shares. As of February 2010, the Russian stake increased to 84.03 percent, while Tajikistan’s share had dropped to 15.97 percent. However, since virtually all the electricity generated at Sangtuda-1 is sold to Barqi Tojik, this monopsonistic position affords the Tajikistani side considerable power in this relationship. The launching of Sangtuda-1 increased Tajikistan’s generation capacity by some 1215 percent, and the utilization of the country’s hydropower resources by 3 percent. According to some estimates, Sangtuda-1 could add 1 billion kWh to Tajikistan’s annual electricity export potential. However, Statistical Agency data indicate that Tajikistan’s electricity exports collapsed in 2010, with the dissolution of the Central Asian integrated transmission network. Likewise, despite the increases reported in 2009 and 2010, power generation in 2010 year remained some 6 percent below the 17.1 billion kWh peak recorded in 20052006.

147

In addition to the RAO UES electricity utility, these included Power Machines, the Chekhov Gidrostal Plant, ChirkeyGESstroy, Zarubezhvodstroy, Zagranenergostroymontazh, and Trust Gidromontazh.


Chapter 4 - Tajikistan’s Energy Sector Figure 4.9— Tajikistan: Committed and actual (disbursed) investments in large Barqi Tojik energy projects (million US$)

400 373.4 356.8 350

300

Actual amount Committed amount

250

200

180

150 121.5 109.3 91.2

100

50

21.9 27.9 15.5 13 19.3 8.5 12.3 11.3 10.1 4.7

36.9

37.7

39.2

53 42.5 36

59.7

20 0

6.1

0

Source: Barqi Tojik Investment Program 2010.

The South-North transmission line. Whereas most of Tajikistan’s electricity is generated from hydropower plants along the Vakhsh river cascade in the south, the country’s northern regions (especially Khujand and Penjikent) have not traditionally been connected to the rest of the national transmission grid. These regions instead relied on locally generated power (chiefly hydro) and imports from Uzbekistan (and sometimes Turkmenistan, trans-shipped through Uzbekistan). The vulnerability of these arrangements was underscored by the winter

crisis of 2007-2008, and subsequently by the dissolution of the regional integrated electricity grid and by growing tensions with Uzbekistan (linked in part to Tajikistan’s construction of the Roghun hydropower station—see below). The government of Tajikistan responded by accelerating the construction of new high-tension power lines to these northern regions from the Vakhsh river cascade. Thanks to a loan from China’s ExImBank, the first stage in this construction was completed 137


Poverty and Social Impact Assessment

in 2010. The 1600 megawatts of capacity added to Barqi Tojik’s transmission infrastructure helped to ease energy tensions this past winter. Longer term, the construction of additional north-south transmission infrastructure could boost Tajikistan’s electricity exports to Kyrgyzstan, as well as the possible transit of summer electricity from Kazakhstan and Kyrgyzstan to Afghanistan and Pakistan, within the framework of the Central Asia-South Asia Regional Electricity Market (CASAREM) project. The completion of this work—the total costs of which are reported to be in the range of $280-$340 million—is expected in June 2011.

Construction of the Sangtuda-2 hydropower plant (with 220 megawatts of installed capacity) commenced during the Soviet period in the 1980s, but halted in the beginning of the 1990s due to lack of funding. With financing provided by a joint venture with the government of Iran, construction recommenced in February 2006, on a buildoperate-transfer basis with Iran’s Sangob Company serving as the general contractor. Most of the financing for this $256 million project is coming from the government of Iran (which is providing a $180 million loan); Sangob is contributing another $36 million. Barqi Tojik is providing $21 million in financing, while the state budget is to provide the remaining $19 million. During the first 12.5 years of Sangtuda-2’s operation all profits are to accrue to the Iranian partners; after this time ownership would revert entirely to the government of Tajikistan. Whereas Sangtuda-2 was planned to become operational by the end of 2012, recent reports indicate that the construction is running ahead of schedule and could be finished in 2011. The anticipated 1 billion kWh of additional electricity generated from Sangtuda-2 could significantly reduce Tajikistan’s autumn-winter electricity shortages.

138

Because of its large size and financing needs, Roghun’s construction has evoked domestic and international controversy. The Roghun hydropower station. Like the Sangtuda stations, the Roghun hydropower station is an attempt by the government of Tajikistan to complete a Vakhsh cascade construction project begun during the Soviet period. However, because of its large size (3600 megawatts of installed capacity as planned) and financing needs (variously estimated at $2-3 billion by the government of Tajikistan148), and because it would be a multiyear storage facility—with greater potential to alter water flow in the Vakhsh and then AmuDarya rivers (the latter of which flows into Uzbekistan and Turkmenistan, as well as forming part of Tajikistan’s border with Afghanistan)—Roghun’s construction has evoked both domestic and international controversy. Following an estimated $800 million invested during the Soviet period, construction at Roghun recommenced in 2006. Tajikistan’s 2010 budget allocated 650 million Somoni ($148 million) for Roghun construction—a sum roughly equal to three quarters of Tajikistan’s 2010 social protection expenditures. This robust support from the state budget reflects Roghun’s national priority status, with which Tajikistan’s president has identified himself. Through mid2010 some $68 million (perhaps 2-3 percent of the total funds needed) had been disbursed for Roghun construction.149 Roghun’s construction is to be completed in two phases. Phase I, which involves construction of the dam to two thirds 148

According to the IMF’s December 2010 country report for Tajikistan, the full costs of Roghun’s construction have not yet been determined. 149 IMF (2010): “Tajikistan Country Report No. 10/374”, Washington DC.


Chapter 4 - Tajikistan’s Energy Sector

of its full height, includes the reconstruction of two existing tunnels, the building of a third tunnel, the creation of the regulating reservoir, and the installation of two generation units. The electricity output of this Phase I would be about 4,300 GWh, and it would also enable the generation of an additional 400 GWh at the Nurek hydropower station (currently Tajikistan’s largest). Phase II would extend the dam to its full planned height (335 meters) and install 2400 megawatts of additional generating capacity. After the completion of Phase II, Roghun would generate roughly 13,000 GWh, while also boosting generating capacity downstream at the Nurek hydropower station. Roghun’s construction faces two sets of challenges. The first are financial, reflecting the gap between the project’s estimated $2-3 billion price tag and the more modest capacities of Tajikistan’s state budget. To mobilize additional resources, the Ministry of Finance in early 2010 issued shares in the Roghun Open Joint Stock Company (the stateowned corporation discharging some of Roghun’s ownership functions), for purchase by legal entities and citizens of Tajikistan. A national marketing campaign helped sell some 1.7 million unregistered shares, yielding 820 million Somoni ($166 million) in revenues.150 However, this campaign met with significant criticism from the international community, in part because of concerns about quasi-obligatory share purchases. The World Food Programme’s April 2010 food security monitoring report151 estimated that 10-15 percent of expenditures of rural households (including those suffering from food insecurity) went to the purchase of Roghun shares; many households seemed to be spending more on these shares than on health or education. To this were added governance concerns: due in part to the lack of a unified database of registered shareholders, the first shareholder meeting did not take place for at

least a year after the equity sales commenced. The IMF found that the “campaign to finance this project through equity sales has curtailed disposable incomes and is estimated to weigh on the growth prospects…. As such, the phasing out of the equity campaign is welcome.”152 The second and related set of challenges is political. The gap between Tajikistan’s financial capacity and Roghun’s construction costs suggests that foreign investment may be the only possible way to finance the project. However, Roghun’s construction faces adamant opposition from the government of Uzbekistan, which claims that Roghun’s operation could have significant environmental and economic costs for downstream countries. Uzbekistan since 2009 had been conducting a high-profile diplomatic campaign against Roghun’s construction, in effect presenting prospective investors with a political “Tajikistan or Uzbekistan” question. The government of Uzbekistan also undertook measures to limit the trans-shipment of goods and services—including electricity imports from Turkmenistan—destined for Tajikistan across its borders. These political controversies, combined with the large construction costs, have thus far limited prospective investor interest in Roghun. Two new combined heat and power plants (in Dushanbe—270 megawatts; in the north—300 megawatts) are under construction. The plant in Dushanbe is being constructed under an agreement with China (2009) with a price tag of $400 million.153

The Kayrakkum hydropower station along the Syr-Darya river in northern Tajikistan was put into operation in 1957. While not being particularly large (total installed capacity is 126 megawatts), Kayrakkum has traditionally played an important role in the electricity balance of 152

150

Ibid. 151 WFP and UK Aid (2010): “Tajikistan Food Security Monitoring System”, Dushanbe.

IMF (2010a): “IMF completes Reviews Under Extended Credit Facility Arrangement with Tajikistan, and Approves Request for Augmentation and Waivers”. 153 http://www.highbeam.com/doc/1G1-194947609.html

139


Poverty and Social Impact Assessment

Tajikistan’s Sughd oblast. Management of the water in Kayrakkum’s reservoir sometimes evokes controversy with Uzbekistan, whose eastern regions rely on the Syr-Darya for drinking, irrigation, and other uses. Like many other parts of Tajikistan’s electricity infrastructure, Kayrakkum has undergone considerable depreciation since its construction; losses are currently around 310 million kWh annually. The US Government (USTDA) in 2005 provided grant funding for a feasibility study on Kayrakkum’s reconstruction; the total amount needed was estimated at $172 million. The EBRD and EIB are now working on this.

If the lion’s share of its hydropower potential could be tapped, Tajikistan could export significant amounts of electricity to Central and South Asian countries, as well as China. The Nurabad-1 hydroelectric power station. In 2009, the government concluded an investment package with China that included $650 million for the construction of Nurabad-1 350 megawatt hydroelectric power station on the Khingob River in eastern Tajikistan. It is being constructed by China Theban Electric Apparatus Stock Co LTD (CTEAS).

amounts of electricity to Central and South Asian countries, as well as to China. The creation of a Central Asian and South Asian regional electricity market (CASAREM) has therefore been a major objective of the government, as well as development banks, key bilateral donors and international financial institutions. This vision, much of which is realized within the framework of the Central Asian Regional Economic Cooperation (CAREC) programme, rests on two pillars:

Generation: The construction of large hydropower plants in Tajikistan (e.g., Roghun) and Kyrgyzstan (e.g., Kambarata1), in this way boosting development prospects in Central Asia’s low-income upstream countries by significantly expanding their electricity production and exports (supplemented by increased use of coal-fired power plants); and

Transmission: The construction and rehabilitation of transmission infrastructure in Tajikistan (and possibly also Uzbekistan and Turkmenistan), to allow for the transmission of increased electricity exports to Afghanistan and Pakistan, and potentially India.

Specific projects falling under this heading include:

CASA 1000: The first phase of this transmission project would wheel Kyrgyzstan and Tajikistan to South Asia. Cost estimates run from $500 million to $1 billion, of which $300 million would come from the participating countries and $700 million from the World Bank, the Asian Development Bank, the Islamic Development Bank, and private investors.

Afghanistan-Tajikistan transmission project: Under a twenty-year power purchase agreement signed in September 2008 between the two governments, Tajikistan is to export annually up to 500 gigawatt hours of electricity to Afghanistan (initially during the summer monthly only). These exports are to be

The Zarafshan hydropower plant (150 megawatts) on the Zarafshan river. An engineering, procurement, and construction contract has been concluded with the Sinohydro Corporation (China). Credit financing from China is expected. The 154 estimated cost is around $258 million. Tajikistan and regional electricity projects. If the lion’s share of its hydropower potential could be tapped, and the appropriate regional transmission infrastructure put in place, Tajikistan could export significant 154

Barqi Tojik Investment Program 2010.

140


Chapter 4 - Tajikistan’s Energy Sector

made possible by the completion of the Sangtuda hydropower stations, and by the “Afghanistan-Tajikistan 220kV transmission project”, which would upgrade and extend Barqi Tojik’s transmission infrastructure 274 kilometers south from the Vakhsh cascade to the border with Afghanistan. This extension of the transmission grid, which is being financed by the Asian Development Bank, was to have been completed in 2010. However, rail traffic disputes with Uzbekistan (linked to the Roghun controversy) delayed the necessary construction work; in the meanwhile, Uzbekistan completed the southern extension of its own national transmission grid to the border with Afghanistan. The realization of this regional energy market vision could handsomely benefit Tajikistan, as both a producer and transshipper of electricity exports (from Kyrgyzstan). However, its prospects face three non-commercial obstacles:

The incongruities of planning to significantly increase electricity exports while millions of Tajikistani citizens go without electricity in the winter; and

The government of Uzbekistan has yet to fully sign on to the pillars under-pinning this vision—particularly the generation pillar, which assumes the expansion of Tajikistan’s hydropower generation capacity at Roghun. As long as the viability of this vision remains in doubt, Tajikistan may face serious difficulties in attracting the billions of dollars in foreign investment needed to develop its electrical energy sector.

Annex 3.1 describes recent projects, those in progress, and projects planned in the near future, as described in Tajikistan’s 20102012 Poverty Reduction Strategy Paper. The total funding gap for the near future is estimated at some $2.2 billion. This table does not include the prospective hydropower plants planned by government in the medium to long run (e.g., Roghun).

Uncertainties associated with the military and political situation in Afghanistan and Pakistan;

141


Poverty and Social Impact Assessment

Oil and gas Gas. While Tajikistan’s proven gas reserves are estimated at 5.7 billion m3,155 the extractive sector is not well developed, so that the industry primarily revolves around the import, transmission, and distribution of gas from Uzbekistan. TajikTransGaz,156 a vertically integrated state-owned monopoly with 17 regional gas distribution enterprises, manages Tajikistan’s gas pipeline infrastructure, which is divided into two parts. In the south, gas is received from Uzbekistan from the Kelif-Dushanbe pipeline. In the north, the transmission line is part of the gas transmission system from the west to the east of Uzbekistan. (Regulatory practices to encourage third-party access to TajikTransGaz’s pipelines have not yet been developed). Table 4.11—Tajikistan’s gas balance Million m3

2004

2005

2006

2007

2008

2009

Production

36

29

20

17

16

20

Imports

622

631

637

647

513

217

Consumption

658

610

657

663

529

237

supplies are constrained by underdeveloped pipeline infrastructure (particularly in distribution). Figure 4.10—Tajikistan: Gas consumption by sectors (in million cubic meters)

700 600

81.5

500

140.2

households communal services

85.6

400 162.5

300 200

others

39.6 418.2

31

261.6

100

158.2

0 2007

2008

2009

Source: Tajikistan Statistical Yearbook, 2010

Some 95 percent of Tajikistan’s gas needs are met by imports from Uzbekistan, the terms of which are specified in annual bilateral agreements, which set the prices for imported gas, as well as stipulate the terms under which Uzbekistan’s gas is trans-shipped across northern Tajikistan, and Tajikistan’s access to Uzbekistan’s rail and gas transit systems.

Source: Tajikistan Statistical Yearbook, 2010.

Table 4.12— Tajikistan: Purchase price of natural gas from Uzbekistan (per 1000 m3)

As Table 4.11 shows, gas imports and consumption have dropped sharply since 2007, due to increased purchase prices for natural gas from Uzbekistan (see Table 4.18). (Tajikistan has no gas exports.) Households account for about half of gas consumption (see Figure 4.10). About 30 percent of households in Tajikistan have access to the gas supply network, with the greatest numbers of these (39 percent of the total) in Dushanbe.157 Prospects for increasing household gas

2006 $42

155

https://www.cia.gov/library/publications/the-worldfactbook/fields/2179.html 156 TajikNefteGaz, another state-owned monopoly, is responsible for gas extraction, most of which occurs in fields in Khatlon province in southern Tajikistan. 157 Energy Charter Secretariat (2010): “Tajikistan: Indepth review of the Investment Climate and Market Structure in the Energy Sector”.

142

2007 $55

2008 $145

2009 $240

2010 $254

Source: TajikTransGaz.

Efforts to reduce losses in the gas system, primarily by improving distribution systems and metering, have been a major priority for TajikTransGaz. As the data in Table 4.13 below indicate, these efforts seem to have met with some success—losses were reduced from 24 percent of total consumption in 2005 to 12 percent during the first half of 2010. However, these figures are still high by international standards.158 158

The international standards for gas losses vary in different countries but 90 per cent of the developed countries have achieved less than one per cent system losses.


Chapter 4 - Tajikistan’s Energy Sector

Table 4.13—Tajikistan: Gas transit losses

Type of loss

2004

2005

2006

2007

2008

2009

2010*

Transmission (trunk pipeline)

7%

14%

13%

11%

13%

n.a.

n.a.

Distribution (technical losses)

1%

1%

3%

4%

2%

n.a.

n.a.

Commercial losses

13%

9%

4%

2%

2%

n.a.

n.a.

Total losses

22%

24%

19%

18%

17%

14%

12%

* January – July data., Source: TajikTransGaz.

The tariff structure is based on the prices at which TajikTransGaz imports natural gas from UzTransGaz. These prices have more than quadrupled in the last six years as world and especially regional gas prices rose sharply. In 2010, the price was $254/1000m3, bringing domestic gas prices in Tajikistan up to European levels. TajikTransGaz bases its national tariffs on the import price: retail tariffs are then approved by the Antimonopoly Commission. Tariffs for users have therefore increased accordingly. Arrears and debts to Tajikistan’s gas distribution company run high, with the Dushanbe Heating and Power Plant, TALCO, and the Tajik Cement Company being key debtors. As a result, TajikTransGaz often falls into arrears to Uzbekistan’s UzTransGaz, which sometimes responds by cutting gas supplies. For example in December 2009, imports were cut by 50 percent because arrears to UzTransGaz had reached $9 million, while TajikTransGaz’s accounts receivable at the time were some nearly $19 million.159 Developing domestic gas production in Tajikistan, through more extensive exploration and attracting foreign investment, is among the government’s energy sector priorities. Russia’s Gazprom and Canada’s Tethys Petroleum are currently developing gas fields in Tajikistan, but they will not become productive for some time. News reports from December 2010 concerning Gazprom’s development of the Sarikamysh gas field are 159

Uzbekistan cuts gas supply to Tajikistan, Nigina Sharipova, 2009-01-06, www.centralasiaonline.com

particularly optimistic:160 seismic surveys have led Gazprom to estimate that the field holds some 60 billion m3 of gas—the equivalent of 100-150 years of annual gas consumption. If quickly brought into production, a field of this size could certainly increase the supply of gas in Tajikistan.

TajikTransGaz’s efforts seem to have met with some success— losses were reduced from 24 percent of total consumption in 2005 to 12 percent during the first half of 2010. However, these figures are still high by international standards. However, it is not clear that the owners of the gas—Gazprom, the government of Tajikistan, or both—would choose to sell the gas at a discount on the domestic market, relative to the prices that could be obtained if the gas were exported. This could limit the extent to which domestic gas prices would fall after the new gas would come on line. Finally, significant investments in the expansion of Tajikistan’s gas distribution infrastructure would be needed in order to bring the benefits of the country’s prospective gas bounty to

160

Major Natural Gas Find in Tajikistan Set to Change Regional Dynamic, Written by Editorial Dept, Wednesday, 22 December 2010 16:33, by Iskander Aliev, Source: RFE/RL.

143


Poverty and Social Impact Assessment

households outside of Dushanbe, Khujand, and a few other cities.

Information from the Ministry of Energy and Industry suggests that Tajikistan

Table 4.14—Tajikistan: Coal production (thousand tons)

Production

2003

2004

2005

2006

2007

2008

2009

47

93

99

105

181

199

176

Source: Statistical Yearbook of Tajikistan, 2010.

Oil. While Tajikistan’s crude oil reserves are estimated at 117 million tons, most are located deep underground; their exploitation is not seen as commercially viable. Tajikistan’s annual crude oil production levels have declined since 1992, when 1,311 barrels/day were produced. The 1992-1997 civil war, the country’s macroeconomic troubles, and a lack of investment in the oil infrastructure have contributed to this decline. In July 2001, Tajikistan brought its first small oil refinery online at Konibodom. The refinery has a capacity of 400 barrels per day, and produces gasoline, diesel fuel, kerosene, and fuel oil.161 However, as with gas, almost all of Tajikistan’s oil and refined products are imported. Whereas most gas imports come from Uzbekistan, most oil and refined product imports come from Russia.

Coal. Competition and private ownership play more important roles in Tajikistan’s coal sector than they do in oil and gas. Some 14 commercial enterprises, including joint ventures with foreign companies, are engaged in coal production and distribution. In principle, the state’s role is limited to enforcing environmental and safety regulations, promoting technological progress, and supporting the development of the associated road and rail transport systems.

has over 40 coalfields with confirmed reserves of 15.3 billion tons—an estimated 41 percent of Central Asia’s total coal reserves.162 Retail prices have been liberalized and are now close to international prices (see Figure 4.11). At the end of the Soviet period, Tajikistan’s annual coal consumption was some 1.5 million tons, of which over 1.0 million tons were supplied domestically. Coal production fell sharply after the collapse of the Soviet Union; and while output has increased significantly in the last decade, it remains a fraction (20 percent) of pre-civil war levels (see Table 4.14). Figure 4.11—Retail prices for hard coal in Tajikista, and the European Union (per ton)

$160 $140 $120 $100 $80 $60 $40 $20 $0 2007

average price Tajikistan

http://www.globalsecurity.org/military/world/centralasi a/tajik-energy.htm

144

2009

2010

market price EU

Source: MOEI for Tajikistan, BP Statistics for EU.

The government is considering projects to boost electricity output by collocating coalfired power plants with coal fields. (The Fon Yaghnob, Nazar Ailok, and Shurob mines are 162

161

2008

Press release, Third International Tajikistan Exhibition, "Mountainous Equipment, Enrichment and Extraction of Ores and Minerals," Dushanbe, September 18-20, 2008


Chapter 4 - Tajikistan’s Energy Sector

particularly important in this context.) In 2007, the Ministry of Energy and Industry signed an agreement to this effect with the Chinese Development Bank, China’s ExImBank, the National Development Bank of Kazakhstan, the Eurasian Development Bank, the World Bank, and other partners. Increased coal production and sales could also offer households (with boilers or furnaces) better access to a decentralized heat source. For these reasons, the government’s development programme for the coal sector called for annual production to rise to 445,000 tons in 2010, and to 815,000 tons by 2015. (By country, production in 2009 was only 176,000 tons.) Despite these ambitious plans, Tajikistan’s coal industry remains in poor condition, mostly due to the decapitalization of sectoral and associated transport infrastructure, inadequate financing and working capital, and managerial shortcomings. The decline in coal output in 2009—when Tajikistan was suffering from an energy shortage—is likewise a troubling sign. The attraction of Tajikistan’s coal fields to investors may be mostly based on the possibilities for strip mining, which substantially reduces risks and costs. However, although many deposits are suitable for strip mining, they are found in mountainous regions, where the requisite transport and other infrastructure are not in place, and which are often subject to extreme weather conditions. Significant outlays from the state budget would seem to be necessary, in order to promote the coal sector’s development (see Annex 3.2). Tajikistan’s 2010-2012 Poverty Reduction Strategy Paper estimates total investment needs for this sector at around $137 million; this financing has not yet been secured.

District heating Tajikistan’s heat generation and distribution infrastructure, which was developed during the Soviet period, is largely

The district heating infrastructure has undergone considerable decapitalization, as household tariffs have not kept pace with the market prices charged for gas, oil and other fuels. confined to Dushanbe (which has a combined heat and power plant and several large district heating systems); several secondary cities also have district heating systems based on hot water supplied from heat-only-boiler (HOB) plants. Heat supply is only available during November-March. The district heating system worked more or less satisfactorily until the 1990s, since Tajikistan received significant amounts of natural gas, fuel oil, and electricity from neighbouring Soviet republics (as well as domestically produced coal) at nominal prices. Since then, however, the district heating infrastructure has undergone considerable decapitalization, as household tariffs have not kept pace with the market prices charged for gas, oil, and other fuels. High levels of leakages/losses due to outdated pipeline network design and inadequate insulation, as well as frequent outages, are the result.163 The two main heat generators—the combined heat and power plants at Dushanbe164 and Yavan— are highly depreciated.165 In addition to these plants, there are 181 heat boilers in the country, but a majority of these are currently

163

Energy Charter Secretariat (2010): “Tajikistan: Indepth review of the Investment Climate and Market Structure in the Energy Sector”. 164 The Dushanbe combined heating and power plant is 100 percent owned by Barqi Tojik. 165 The Government has recently provided financing for the reconstruction of the Dushanbe combined heat and power plant; the TALCO aluminium company has likewise provided funds to reconstruct the Yavan combined heat and power plant.

145


Poverty and Social Impact Assessment

out of operation.166 A December 2010 report from the Ministry of Energy and Industry found that only Dushanbe had district heating—and only to a limited extent (in the city centre).167 The Dushanbe District Heating Enterprise produced only around 92 gigacalories of heat in 2008-2009, meeting some 20 percent of the city’s needs. No heat or hot water metering exists in residential, commercial or public sector buildings. Consumers are therefore unable to regulate the heat supply, resulting in too low or too high room temperatures and further heat losses from the open windows that cool overheated apartments. The piping layout and the lack of control by the residents over the heat consumed make consumption-based billing or cutting off non-paying consumers very difficult. Maintenance of equipment inside buildings (residential and office) is the responsibility of the housing and utilities administration: the poor financial situation of these structures contributes to the poor condition of the district heating infrastructure.

Figure 4.12)—is an additional cause of the deterioration of the district heating system. Barqi Tojik sells heat to wholesale consumers/re-sellers at the following rates (without VAT):

• • •

Figure 4.13 shows current and prospective retail heating tariffs for the residents. Figure 4.13—Tajikistan: Household heating tariffs (somoni per gCal)

4 3.11 3 2

Figure 4.12—Tajikistan: Communal service tariffs and wholesale gas price trends, (2007 = 100)

500

2.43 1.82

1.84

2009

2010

2.73

1 0

Communal service tariffs Wholesale gas prices

400

2011

2012

2013

Source: Forecast parameters of social and economic development of Tajikistan 2009-2013.

300 200 100 2007

2008

2009

2010

Source: State Statistical Agency, UNDP calculations.

The sharp increases in gas prices— reflecting the higher prices of imports from Uzbekistan, which were not fully offset by increases in communal service tariffs (see 166 Energy Charter Secretariat (2010): “Tajikistan: Indepth review of the Investment Climate and Market Structure in the Energy Sector”. 167 Ministry of Energy and Industry of the Republic of Tajikistan (2010): “Status of Regulation of District Heating and CHPs”, M.M. Sharipov – Head of the PTL sector, Power Department Kiev, 2-3 December 2010.

146

budgetary institutions: $6.8 (30 Somoni) per gigacalorie; households: $0.9 (4 Somoni) per gigacalorie; main consumers (industry, commercial structures, etc.): $26 (117 Somoni) per gigacalorie.

Household heating tariffs are crosssubsidized by Barqi Tojik. At present, the combined heating and power plants are not profitable; all expenses are covered by Barqi Tojik. Collection of payments and services are low: debts accumulated by consumers reached $1.5 million in 2010.168 With district heating companies increasingly unable to supply reliable heating services, many urban households have turned to electricity for heating. The electricity 168

Ministry of Energy and Industry of the Republic of Tajikistan (2010): “Status of Regulation of District Heating and CHPs”, M.M. Sharipov – Head of the PTL sector, Power Department, Kiev, 2-3 December 2010.


Chapter 4 - Tajikistan’s Energy Sector

system is not always able to cope with the additional winter demand, leading to frequent local power (and heating) outages. In sum, Tajikistan’s district heating system is near collapse. In response, the government intends to convert boilers from burning of gas and fuel oil to burning coal. Construction of the two new combined heating and power plants will help somewhat to improve the heat supply. The construction of one of the CHPs (200 megawatt, with a price tag of $30 million) has started already and is expected to be completed within 18 to 20 months. However, the bigger issue is developing and implementing a sustainable heating strategy which will seek to shift the spaceheating load away from electricity.169 The Government of Tajikistan has acknowledged the urgency for developing such a strategy (based on the use of gas, coal and biomass) but no documents seem to be available at the moment.

incomeand employment-generation opportunities for rural households—for many of which, migration has become the dominant coping mechanism. Chronic energy shortages have generated a number of unfortunate side effects. Due to the extensive use of wood for fuel (as well as increased land use for farming and grazing), Tajikistan’s mountainous regions have lost up to 70 percent of their wood covering since the late 1990s.171 Recent research172 indicates that Tajikistan’s forests have more or less disappeared, and are limited to very small relicts in remote and sparsely populated areas. In addition to the loss of biodiversity, this deforestation has significant environmental, economic, and human costs in terms of flooding, landslides, and mudslides. According to an OSCE report, some 50,000 hectares of arable are lost every year to floods, mudslides and erosion.173 The negative impact of smoke inhalation and respiratory disorders on health,174 and of greenhouse gas emissions on climate change, may also be important.

Decentralized Renewables The absence of reliable electricity (and other energy) supplies in rural areas has become a major development issue for Tajikistan. Barqi Tojik data indicate that, despite comprising nearly three quarters of the country’s population, rural households during 2008-2010 accounted for only 8-11 percent of Tajikistan’s total electricity consumption.170 It is estimated that over 1 million residents (primarily in rural areas) have little or no access to adequate electricity/energy supplies—particularly during the winter, when it is common to have spells of more than six weeks without any electricity. The absence of reliable electricity supplies means reduced access to health, education, and other public and social services in rural areas. It also limits 169

Energy Charter Secretariat (2010): “Tajikistan: Indepth review of the Investment Climate and Market Structure in the Energy Sector”. 170 “Intermediate Strategy for Renewable Energy Sources”, p. 8.

Due to extensive use of wood for fuel, Tajikistan’s mountainous regions have lost up to 70 percent of their wood covering since the 1990s. On the other hand, GTZ (2010) argues that, since for many rural households, the alternative to burning firewood is burning 171

This figure, which appears in a number of sources, is not reflected in the official statistics on forest cover— which show no significant changes in the last two decades. For more on this, see GTZ (2010): “Forest Sector Analysis of the Republic of Tajikistan”. 172 Ibid. 173 Central/South Asia: Deforestation Makes Mark On Region, Its People, Radio Free Europe/Radio Liberty citing OSCE environmental officer Saulius Smalys, August 29, 2006. 174 Bringing Power to the Poor in the Pamirs, World Bank, press release. Available here: http://lnweb90.worldbank.org/eca/eca.nsf/0/66201C6D C20F591785256C32006D471A?OpenDocument

147


Poverty and Social Impact Assessment

dung, the health effects of which can be even more serious, this argues for combining extensive reforestation efforts (which are needed in any case) with the establishment of sustainable firewood plantations to meet rural energy needs (and provide income generation opportunities).

rates dropped 40-50 percent. Schools, orphanages, support facilities for street children and homes for the elderly struggled to maintain minimum service levels.177 The extent to which this slippage has been reversed since 2008 is not clear.

Table 4.15—Tajikistan: Small hydropower plants (1991-2010)

Number of small scale HPPs

District Khatlon Sughd Gorno Badakhshan Districts of Republican Subordination Total

Capacity (megawatts)

20 86 40 107 253

905 3868 5619 5019 15,410

Funding sources State budget and grants 9 7 38 32 86

Local investment 11 79 2 75 167

Source: Ministry of Energy and Industry

175

the According to a WHO report, winter energy crisis of 2007-2008 had a negative effect on public health: access to basic health care declined significantly, as reduced heat and electricity supplies (and therefore running water) forced many hospitals and health centres to close or work restricted hours—in some cases discharging patients. Prolonged exposure to cold temperatures, the health and safety risks associated with unsafe use of alternative home heating equipment (e.g., gas, kerosene; stoves burning wood and charcoal), the lack of running water, isolation or over-crowding—all this adversely affected general health conditions status. Higher incidence of acute respiratory diseases, waterborne diseases, and preventable maternal and infant deaths resulted; pregnant women, children, the elderly, and the mentally ill were particularly at risk.176 A 2008 UNESCO report found that no heating systems were operating in 26 percent of Tajikistan’s schools that winter; attendance 175

WHO and the Ministry of Health of Tajikistan (2008): “Health Assessment for Tajikistan”. 176 Ibid.

148

Large investment projects in the electricity sector are unlikely to come on line very soon. For the small, isolated rural or mountainous settlements, the expansion of decentralized renewable energy technologies—chiefly for electricity generation, but also for heat, hot water, and biogas—might be the only option for reducing energy insecurity likely to make a difference in the short run. Moreover, the construction and maintenance of decentralized renewable energy facilities can themselves create incomeand employment generation opportunities—one study finds that the addition of one megawatt of installed capacity in a small hydropower plant can generate 40 “green jobs”.178 Such calculations underscore how the expansion of decentralized renewable energy technologies, rural development, and poverty reduction in Tajikistan are inextricably linked.

177

“Tajikistan country case study”, Country profile prepared for the Education for All Global Monitoring Report 2008 : Education for All by 2015: will we make it?; by Vladimir Briller, UNESCO, 2007 178 According to the Concept of the development of small scale hydropower (2009)


Chapter 4 - Tajikistan’s Energy Sector

The expansion of decentralized renewable energy technologies is therefore a priority for the government and many donors. Official data indicate that, since 1991, some 15 thousand megawatts of installed capacity have been added to the country’s electricity generation assets, via the construction or reconstruction of 253 small scale hydropower plants (see Table 4.15). In comparison to big hydro and hydro-carbons, decentralized renewable projects are relatively inexpensive, and can attract at least some of the financing needed for their construction and maintenance from donors, private entrepreneurs and the communities in which they are located.179 Some of the basic legislation and strategic programming frameworks needed to promote the development of decentralized renewables in Tajikistan have been put in place. The Complex Programme on Alternative Energy Sources such as small rivers, solar, wind, biomass, and geothermal energy for 2007–2015 (2007), is divided into three phases: •

Phase 1 (2007–2009): Compiling a cadastre of alternative energy sources; assessing the potential effectiveness of various technologies, taking into account Tajikistan’s geo-climatic conditions; and developing new renewable energy technologies;

Phase 2 (2010–2012): Introducing pilot programmes to test the effectiveness of renewable energy technologies; establishing an industrial base for production; training and capacity building; and

In December 2010, two important developments occurred that could accelerate the development of the appropriate regulatory environment for renewable energy sources. These include the approval by the Ministry of Energy and Industry of: •

a methodology for calculating tariffs for electricity generated by independent power producers (including those using small hydro and other renewable technologies) who are connected to Barqi Tojik’s grid (Decree No.131); and

a basic model contract for power purchase agreements between Barqi Tojik and independent power producers (Decree No.112).

As long as Barqi Tojik manages most of Tajikistan’s power generation assets and the country’s transmission grid, its incentives for providing rival power producers with access to the grid may be weak. Despite this progress, four aspects of the commercial and regulatory environment for decentralized renewable remain problematic: •

179

In principle, decentralized renewables (and energy efficiency) projects can also attract carbon finance, under the Kyoto Protocol’s clean development mechanism (CDM). However, there are apparently no CDM projects under implementation in Tajikistan, due to the lateness (2009) of the country’s ratification of the Kyoto Protocol, and continuing legal and regulatory lacunae.

Phase 3 (2013–2015): Production of equipment for alternative energy generation.

As long as Barqi Tojik manages most of Tajikistan’s power generation assets and the country’s transmission grid, its incentives for providing rival power producers with access to the grid may be weak. (The same applies to Pamir Energy in Gorno Badakhshan). According to the changes in the Energy Law that was made in February 2009, 149


Poverty and Social Impact Assessment

household energy insecurities is now widely recognized. Likewise, funding from donors and government sources for decentralized renewables is increasingly available. However, mechanisms to attract and transparently manage these resources remain under-developed. UNDP’s proposed National Trust Fund for Renewable Energy and Energy Efficiency could provide an important missing piece of the institutional development puzzle.

Barqi Tojik is now obligated to buy electricity from (mostly small-scale) independent power producers. However, since the actual tariffs must be agreed upon with Barqi Tojik (or Pamir Energy), developers of decentralized renewables projects continue to face regulatory uncertainty about their viability.180 •

180

As long as tariffs for electricity generated from other sources remain relatively low (e.g., compared to other countries), Barqi Tojik has little reason to purchase “expensive” electricity from decentralized renewable producers. Such purchases must ultimately be subsidized, either from Barqi Tojik’s retained earnings, from cross-subsidies implicitly paid by consumers of power generated from other sources, from the state budget, or by donors. In the meanwhile, according to information obtained from Barqi Tojik, out of some 340 small hydropower plants built in Tajikistan—and at present, only one is connected to the grid. Neither private- or public-sector agencies in Tajikistan possess the technical expertise needed to construct and maintain decentralized renewable energy installations. Communities that construct small power plants (often with donor assistance) are sometimes unable to repair or maintain them. Likewise, the regulatory and technical requirements—and the capacity to enforce them—needed to ensure technological standardization and connection to centralized grids is not fully in place. The potential of decentralized renewable energy technologies to address Tajikistan’s national and

For off-grid communities, matters are simpler: tariffs are defined by local governments, and can therefore be determined in advance.

150

Small hydropower. For settlements located near to small rivers and streams, the construction of small-scale hydropower plants can improve access to reliable year-round electricity services. The programme to develop small hydropower stations, featuring the construction of 71 facilities with installed capacity of approximately 80 megawatts, was adopted by the government in 2009. It distinguishes three types of small hydropower plants (see Table 4.16). Table 4.16—Tajikistan: Ministry of Energy and Industry’s small hydropower plant (HPP) classification scheme

Definition

Symbol

Power range

Micro

µHPP

100 kilowatts or less

Mini

mHHP

101 – 1,000 kilowatts

Small

sHHPs

1,001 – 30,000 kilowatts

Source: Ministry of Energy and Industry

Since 1991, some 15 thousand megawatts of installed capacity has been added to the country’s electricity generation assets via the construction or reconstruction of 253 small hydropower plants (see Table 4.15). Another 29 small hydropower plants are currently under construction, and will in turn add another 11.6 thousand megawatts of installed capacity when they are built. On the basis of research covering 530 large and small rivers with a total length of 14,316 kilometres, Tajikistani specialists have


Chapter 4 - Tajikistan’s Energy Sector

concluded181 that the exploitation of only 10 percent of the hydroelectricity potential of small rivers in middle and high mountainous zones would allow for power supply for up to 70 percent of the settlements and agricultural entities. In particular, they estimate that in the Rasht area alone more than 100 small hydropower plants could be built.182

small hydropower plants each with 100 kilowatts of installed capacity would provide each of these households with one kilovolt of electricity—sufficient for year-round access to indoor lighting. The cost of constructing a small hydropower facility with 100 kilowatts of

Table 4.17—Tajikistan: Costs and benefits of scaling up small hydropower plants for 100,000 vulnerable households

Capacity provided per household (kilowatts kW) Total installed power capacity needed (megawatts) Total number of small hydropower plants needed (average unit capacity = 100 kW) Total investment required Financial return to the local economy Total jobs created Income for local community generated by sale of excess electricity to the national grid Reductions in annual firewood consumption Reductions in greenhouse gas emissions (tons of CO2 equivalent) Total financing needed183

Some illustrative scenarios showing the possible impact of scaling up existing support for small hydro has been developed by UNDP-Tajikistan (see Table 4.17).184 If the number of inhabitants without access to reliable electricity supplies in Tajikistan is taken as 1 million, and if the households in which they live are assumed to have 10 members each, then the construction of 1000 181

PREGA (Promotion of Renewable Energy, Energy Efficiency and Greenhouse Gas Abatement), Tajikistan Country Report: Country Energy Situation and Prospects, 2006 (for the Asian Development Bank). Original source “Power resources of Tajik SSR”, Bowels of Earth, Leningrad, 1965.

182

Ibid. Total investments required plus surplus electricity buyback. 184 These calculations are taken from “Intermediate Strategy for Renewable Energy Sources”, pp. 49-50; UNDP-Tajikistan, 2010 (available at http://europeandcis.undp.org/uploads/public1/files/vuln erability/Senior%20Economist%20Web%20site/Final_ Draft_IntermediateStrategyTajikistan.pdf). 183

1 kW

2 kW

3 kW

100

200

300

1000 $100 mil. $200 mil. 4000

2000 $200 mil. $400 mil. 8000

3000 $300 mil. $600 mil. 12,000

$1.75 mil.

$3.5 mil.

n/a n/a

500,000 m3 900,000

$5.25 mil. 2.5 million m3 4.5 million

$101.75 mil.

$203.5 mil.

$305.25 mil.

installed capacity based on intermediate (best available local) technologies is estimated at around $100,000. This scenario posits an average construction cost of $1/kW/household. Satisfying the unmet basic electricity needs of 1 million vulnerable people therefore carries a price tag of $100 million. Moreover, each additional megawatt of installed capacity in a small hydropower plant is estimated to generate more than 40 direct and indirect jobs.185 Using best locally available technologies also increases the multiplier effect of this spending on the national (and local) economy. It has been estimated that, with the current capacity of Tajikistan’s small hydro-related industry, some 50 of the funds invested in the construction can be attributed to supporting 185

This estimate is based on Nepal’s experience, where it was estimated that more than 500 jobs were created for less than two megawatts of installed capacity. However, the number of jobs created seems likely to decrease with each next megawatt installed, particularly in light of specific Tajik conditions.

151


Poverty and Social Impact Assessment

national economy. This roughly corresponds to a multiplier of 2, so that $100 million invested in the construction of small hydropower plants would generate $200 million in additional output and incomes for Tajikistan.186 In conducting the scaling up exercise, we assumed that average on-grid time per annum is 3500 hours (40 percent of the total possible), with actual generation at 50 percent of capacity. The price Barqi Tojik would pay to purchase this electricity is assumed to be $0.01/kWh, while the guaranteed purchase price paid to the small hydropower producers is assumed to be $0.02/kWh. The difference between the two would be financed by the national trust fund. In calculating the savings in CO2 emissions, it was estimated that the average household uses 0.5m3 of firewood for cooking only and another 2.5 m3 of firewood for combined cooking and heating, annually. This biomass absorbs 1.8 tons of CO2 emissions per cubic meter.

Small hydro clearly represents a viable economic and technological option for many households in remote, isolated locations. If resources could be found to double or triple the electricity delivered (i.e., provide two or three kilowatts of installed power per household), the benefits of such investment— in terms of access to energy, job creation, reductions in deforestation and greenhouse gas emissions (from burning coal, firewood, dung), and health benefits (in terms of fewer respiratory ailments associated with burning

186

As per standard macroeconomic multiplier analysis, according to which the change in total income and output resulting from an initial change in expenditure is given by the expression ∆income = ∆expenditure/leakages. Here, ∆expenditure = $100 million and leakages = .5.

152

firewood, coal, likewise expand.

peat,

or

dung)—would

However, small hydro projects in Tajikistan can face two important drawbacks. •

The first is seasonal: the small streams on which many of these facilities are located are more likely (than larger rivers) to freeze in the winter. These facilities can therefore be rendered inoperable during the season when power and heat are in greatest demand, and when central grids are unable to compensate. By contrast, many communities that are connected to Barqi Tojik’s (or Pamir Energy’s) grids draw on them during the summer time (when power is relatively abundant), thereby reducing the demand for off-grid power and decreasing the commercial viability of small hydro (and other decentralized renewables) projects.

The second is hydrological: these small streams are also more likely (than larger rivers) to undergo significant reductions in water flow during periods of drought, thereby reducing their de facto generation capacity.

Exclusive reliance on small scale hydropower plants is unlikely to be sufficient to provide the electricity needed for rural households. Nor can it by itself allow for significant developments of rural industry. Still, small hydro clearly represents a viable economic and technological option for many households in remote, isolated locations. Its utilization can be scaled up in a matter of few years with minimal up-front costs, provided certain policy and hydrological conditions are met. The recent introduction of regulations requiring Barqi Tojik to purchase power from small scale generators should help to improve the situation. Regulatory issues aside, important questions about the sustainability of


Chapter 4 - Tajikistan’s Energy Sector

constructing and operating small scale hydropower plants—despite political will and donors’ efforts—remain. Ongoing efforts by the Government of Tajikistan supported by the international community (including those of UNDP) will be important to in order to create a sustainable model for the operation of small hydropower plants. Apart from the regulatory improvements, other areas which will be important to address include: •

Increasing the capacity/knowledge base of local companies in constructing and especially maintaining small hydropower stations; Improving the availability of spare parts for small hydropower stations; and

Attracting finance for these projects— including, potentially, via the trust fund for decentralized renewables and energy efficiency proposed by UNDP (Box 4.1). Solar power. The estimated potential for solar power is about 25 billion kWh/year in Tajikistan. It is also estimated that the utilization of available solar energy in Tajikistan could satisfy as much as 10-20 percent of national energy demand (note that the estimate was in the light of the tariffs prevailing in 2007).187 Local experts estimate that the climatic conditions of Tajikistan are favourable for using solar energy, especially in mountain territories, and in East Pamir, in particular, where hydropower potential is

limited.

According to experts, the wide introduction of biogas technologies using animal or agricultural and household wastes could reduce annual methane emissions by 5-8 thousand tons. The majority of this solar potential is not exploited; there is no industrial solar energy capacity in Tajikistan. There is some use of solar resource for water heating purposes; this could be developed further. Small-scale photovoltaic technologies (primarily for public building) could also be introduced, particularly in very remote areas with low population densities where grid reinforcements or new connections seem infeasible. Wind power. Wind power potential in Tajikistan remains largely uninvestigated. Local experts believe that wind energy can be commercially viable in certain regions, where the average annual wind speed is around 5-6 meters/second (such as Fedchenko and Anzob, territory around Lake Sarez in Gorno Badakhshan).188 Four wind power plants (see Table 4.25) were under construction by Barqi Tojik in 2010. Biomass. It is estimated that Tajikistan has the potential to produce around 2 billion kWh/year of electricity from biomass sources (note that the estimate was in the light of the tariffs prevailing in 2007). Currently around three quarters of the population use biomass in their housekeeping.189 In the countryside, where there is no access to natural gas, biogas technologies could be very promising. 188

187

Kabutov, K., “Tajikistan: Priority Directions and Status of Research in the Field of Renewable Energy Sources,” Published in Geliotekhnika, 2007, No. 4, pp. 91-96.

Ibid. This section relies heavily on Kabutov, K., “Tajikistan: Priority Directions and Status of Research in the Field of Renewable Energy Sources,” Published in Geliotekhnika, 2007, No. 4, pp. 91-96.

189

153


Poverty and Social Impact Assessment

According to experts, the wide introduction of biogas technologies using animal, agricultural, or household wastes could reduce annual methane emissions by 5-8 thousand tons. The most promising option of biomass utilisation is biogas generation by means of anaerobic fermentation of manure. A few experimental biogas generators are currently operating in Tajikistan. Opportunities to produce energy via thermochemical conversion of cotton residues may also be present. In 2008 some 253,000 hectares of cotton were planted. Each hectare generated around 100,000 cotton plants, the stalks of which are used for winter heating in rural areas. Geothermal. Tajikistan’s geothermal resources are small and poorly studied. Data about using thermal waters are generally absent, although the development of thermal water in vicinity of Khodja-Obi-Garm is anticipated. It is estimated that Tajikistan could produce 45 billion kWh annually from geothermal sources.190

Energy efficiency Energy intensity in Tajikistan is almost twice the world average, and three times higher than most developed countries, which means that Tajikistan needs three times more energy to produce one unit of GDP than highly developed countries.191 Energy efficiency potential in Tajikistan is assessed by the Ministry of Energy and Industry at 30 percent of current power consumption.192 Some estimates put potential savings much higher; recent UNDP research has found that 190

Kabutov, K., “Tajikistan: Priority Directions and Status of Research in the Field of Renewable Energy Sources,” Published in Geliotekhnika, 2007, No. 4, pp. 91-96

191

UNDP , “Energy Efficiency Master Plan for Tajikistan: Energy Efficiency for Economic Development and Poverty Reduction”. 192 Ministry of Energy and Industry, http://minenergoprom.tj/index.php?lng=en.

154

houses in rural areas lose 50-60 percent of the heat generated.193 The draft “Energy Efficiency Master Plan for Tajikistan” supported by UNDP estimates that adoption of basic energy efficiency measures could reduce primary energy consumption for electricity and heat production from thermal power and heating stations by at least 10 percent; and electricity transmission and distribution losses by 10 percent.194 National energy saving legislation, which was adopted in 2002, recognizes energy efficiency as a national priority and defines a number of priority areas. The implementation of this legislation has been somewhat lacking, however; activities have largely been limited to an ongoing campaign to replace incandescent with compact fluorescent lamps.

Tajikistan needs three times more energy to produce one unit of GDP than highly developed countries. The 2002 legislation could be strengthened via the adoption of energy efficiency regulations pertaining to building codes, minimal energy efficiency standards for appliances and equipment, energy efficiency labelling, energy audits, and energy statistics and energy balance.195 The design and implementation of a national energy efficiency plan, to set national and sectoral energy efficiency targets and outline the resources and policy instruments to realize them, could also be undertaken. The draft “Energy Efficiency Master Plan for Tajikistan” supported by UNDP calls for the application of different energy efficiency measures to rural and urban areas, 193

UNDP, “Analysis of the energy efficiency in rural Tajikistan”, November 2020”, Dushanbe. 194 UNDP: “Energy Efficiency Master Plan for Tajikistan: Energy Efficiency for Economic Development and Poverty Reduction”. 195 Ibid.


Chapter 4 - Tajikistan’s Energy Sector

in light of the differences between them. For rural areas, an integrated rural development approach is recommended, to provide additional incomeand employmentgeneration opportunities and reduce poverty while also saving energy. For urban areas, the introduction and enforcement of appropriate building codes, to regulate the level of energy consumption in buildings, is advocated with an estimate of the potential to reduce energy consumption by 20 percent. It is also estimated that public awareness campaigns could reduce energy consumption in public buildings by some 10 percent; a national programme to improve the efficiency of public lighting systems could reduce losses in this sector by up to 20 percent.

neighbours. A recent World Bank report197 finds that, without $3.3 trillion in new investments in primary energy development and power sector infrastructure over the next 20 years, the Europe and Central Asian region could be moving from being a large (at present) net energy exporter to becoming a net energy importer by 2030.198

In sum, energy efficiency is one of the most cost-effective ways of reducing imbalances between energy demand and supply. Efforts to promote energy efficiency should become an integral part of national energy planning.

Summary The Asian Development Bank projects that Tajikistan’s electricity demand will grow at an annual rate of 1.9 percent during the 2010-2030 period.196 Tajikistan’s energy sector has the potential to meet this demand— particularly in hydro (big and small) and coal; possibly also in gas. The potential returns on energy savings and energy efficiency investments also seem significant. However, potential does not necessarily translate into development. Like many developing countries, Tajikistan does not possess the domestic capital, technology, or managerial capacity needed to fully realize its energy opportunities. Moreover, it is in competition for this capital and technology with many other countries—including its

197 WB (2010b):“Lights out?: the outlook for energy in Eastern Europe and the former Soviet Union” 198

196

APEC/ADB Energy Outlook for Asia and Pacific, 2009.

http://web.worldbank.org/WBSITE/EXTERNAL/COU NTRIES/ECAEXT/0,,contentMDK:21722062~menuP K:258606~pagePK:146736~piPK:146830~theSitePK:2 58599,00.html

155


Poverty and Social Impact Assessment

199

Box 4.1—Tajikistan: A national trust fund for decentralized renewables and energy efficiency

UNDP has designed a national trust fund for decentralized renewables and energy efficiency. This trust fund would be a legal body with the following responsibilities:200 •

Collecting fees to finance the introduction of decentralized renewable energy technologies and energy efficiency measures;

Underwriting contracts with Barqi Tojik for the purchase of electricity generated with decentralized renewable technologies;

Supporting decentralized renewable and energy efficiency activities in rural areas, inter alia via: o o o

Campaigns to promote decentralized renewables and energy efficiency; Education and training programmes for professionals working in decentralized renewables; and Financial support for the preparation of decentralized renewable energy and energy efficiency projects, including investment studies (up to 40 percent of total costs).

Fund raising for decentralized renewables and energy efficiency projects in Tajikistan; initiating and supporting international cooperation and micro-financing for decentralized renewables and energy efficiency programming;

Establishing and maintaining a database of decentralized renewables and energy efficiency projects; and

Providing investment subsidies for other decentralized renewables and energy efficiency activities.

Tajikistan’s prospects in this competition depend on a number of factors, including the extent of national commitment to the energy sector, the legal and regulatory environment, and domestic tariffs/prices. Its successes in attracting foreign investment in national energy projects like Sangtuda-1 and the South-North energy transmission network, and its strong support for the Roghun hydropower plan, underscore the strength of the government’s commitment to developing Tajikistan’s energy sector. However, in other respects—particularly when it comes to attracting private capital, as well as Tajikistan’s reliance on the development of trans-national electricity transmission infrastructure—the picture is not so rosy. The relative paucity of private capital and technology inflows suggests that the burden of developing Tajikistan’s energy sector will continue to be born largely by the state budget. In light of the other development challenges 199

UNDP (2011): “Energy Efficiency Master Plan for Tajikistan: Energy Efficiency for Economic Development and Poverty Reduction”. 200 Ibid.

156

facing the country—in terms of income and non-income poverty, and access to improved water and sanitation (just to name a few)—the state budget’s ability to bear this burden—and the opportunity costs associated with trying to do so—seems likely to remain in question for the foreseeable future, especially given the country’s large budget deficits and public debt. The alternative—improving the energy sector’s enabling environment via more ambitious reforms of legal and regulatory conditions, combined with higher tariffs and elimination of cross subsidies, as well as greater attention to the social protection of those households most vulnerable in the face of higher tariffs—is worthy of serious consideration.


Chapter 4 - Tajikistan’s Energy Sector

The energy/poverty e nexus Tajikistan’s Statistical Agency has been conducting living standards surveys (LSS) periodically since 2002, most recently in 2007 and 2009. As the same 1500 households were surveyed in 2007 and 2009, a living standards panel data set is now present, and is available on the World Bank website.201 Figure 4.14— Tajikistan: Share of households reporting orting power cuts, winter 2007 (%) (

and villages than in Dushanbe. Unfortuna Unfortunately, the 2009 LSS data do not contain information on electricity usage in kWh kWh. However, the LSS 2007 data indicate that 26 percent of the population used 101-200 200 kWh per month. This would be enough to light a room for a couple of hours per day. (Another 16 1 percent used Figure 4.15—Tajikistan: Share of households rep reporting power cuts, summer 2007 (%)

93

93

90 77

75

74

68

49 30 11 2 Average

9

7

18 24

14

Dushanbe

Never Several times a month

1

1

8

Other town

1

33

Village

Every day Several times in a week

22

15 6

4

0

Average

Dushanbe

0

Never Several times a month

3 Other town

6

6

5

Village

Every day Several times in a week

UNDP calculations, based on LSS 2007 data

Energy poverty Household access to and use of electricity. The LSS data indicate that electricity connection rates in Tajikistan are high—98.4 98.4 percent in 2009, up from 96.1 percent in 2007. Unfortunately, the 2009 household questionnaire naire did not contain a question about power cuts, while there was one in 2007. Figure 4.14 and Figure 4.15 below describe the share of households reporting power cuts in 2007. The average duration for 2007 was 5.5 hours per day. As expected, cuts were mostly during the winter, and dramatically more common in other towns

100 kWh per day or less). The 2009 living standards survey database contains information on household electricity expenditures. Based on this, and on information about tariffs in 2009, we calculated usage in kWh (see Figures 4.16 and 4.17).

201

http://econ.worldbank.org/WBSITE/EXTERNAL/EXTDEC/E XTRESEARCH/EXTLSMS/0,,contentMDK:21878045~page PK:64168445~piPK:64168309~resourceurlname:LSMS_ISA _poster31.pdf~theSitePK:3358997~isCURL:Y,00.html#Docu mentation.

157


Poverty and Social Impact Assessment

Figure 4.16—Tajikistan: Share of households using electricity, urban vs. rural (%)

Figure 4.17—Tajikistan: Average monthly electricity consumption per household, urban vs. rural (kWh) (

99

98

98 97

390

436 412 357

364

327

96 95

average 2007

Urban

Rural 2009

average

Urban 2007

Rural 2009

UNDP calculations, based ased on LSS 2007and 2009 databases. database

These data indicate that, while the share of households (both rural and urban) using electricity increased during 2007 2007-2009, average household electricity consumption declined by some 8 percent during this time. (National ational data on apparent electricity consumption202 also show an 8 percent decline during 2007-2009.) Whereas a startling 14 percent increase was reported in Dushanbe, this growth was offset by declines in electricity consumption of 10 and 24 percent in rural and other urban areas, respectively. It seems ms that whereas households in Dushanbe responded to reductions in district heating supplies and sharply higher gas prices by using more electricity for heat and cooking, households in rural areas and especially other cities simply had to make do with less electricity. 2008 Barqi Tojik data also illustrate the low electricity consumption levels in rural areas (see Table 4.18 4.18): whereas 72 percent of the population lives in the rural areas, they consume less electricity than in the urban areas. On the other hand, living standards survey data suggest that electricity consumption increased for poor households while falling for the non-poor, poor, although the latter was still higher than the former. As Figure 4.19 shows,, electricity consumption by 202

Defined as generation minus net exports.

158

the poorest households (first quintile) increased slightly (2 percent) percent during 20072009. By contrast, electricity consumption in the wealthiest households (fifth quintile) dropped by some 47 percent. Factors other than income would seem to be constraining household electricity consumption. These data also indicate that electricity consumption by the poorest (first quintile) households in 2009 was larger than households in the second and third quintile, and only slightly below cconsumption by households in the fourth quintile. Household size does not seem to be correlated with reductions in electricity use. The data in Figures 4.20 and 4.21 indicate that households with three or more children during 2007-2009 2009 reported a slight in increase in electricity consumption, while households with at least six members reported a 4 percent decline in electricity consumption (half the national average). By contrast, electricity consumption in households with at least one elderly member dropped by 18 percent during 2007 2007-2009— more than double the national average.


Chapter 4 - Tajikistan’s Energy Sector

Table 4.18—Tajikistan: Tajikistan: Household electricity consumption during 2006-2008 2008 2006

2007

2008

Million kWh

Share

Million kWh

Share

million kWh

Share

Households

3,314

25%

3,044

22%

2,818

23%

- Urban

1,841

13%

1,786

13%

1,74 1,745

14%

- Rural

1,473

11%

1,258

9%

1,074

9%

Total consumption

13,652

13,967

12,515

Source: Barqi Tojik Sales Department

Figure 4.18—Tajikistan: Share of households using electricity, by quintile (%)

99 97

98 97

Figure 4.19—Tajikistan: Average monthly electricity consumption per household, by quintile (kWh) (

99

98

748

97 96

96

95

357 222

Poorest Second Third Fourth Top quintile quintile quintile quintile quintile 2007

2009

286

333

261

324 417 361

399

Poorest Second Third Fourth Top quintile quintile quintile quintile quintile 2007

2009

UNDP calculations, based ased on LSS 2007and 2009 databases. database

Gender—specified specified as femalefemale versus male-headed households—is is not correlated with differences in household access to electricity. Inter-household household differences in 2009 were not large either in terms of shares of households using electricity, or in terms of average erage daily consumption levels (see Figures 4.22 and 4.23).

159


Poverty and Social Impact Assessment

Figure 4.20—Tajikistan: Share of households using electricity, by household size (%)

99

98 96

average

99 97

97 96

Many member HHs 6+

Figure 4.21—Tajikistan: Average monthly electricity consumption perr household, by household size ((kWh)

406

364 365

333

96

With many children 3+

2007

387 371

390 357

HH with elderly

average

Many member HHs 6+

With many children 3+

2007

2009

HH with elderly

2009

UNDP calculations, based ased on LSS 2007and 2009 databases. database

Large differences between ethnic Tajik and Uzbek households, either in terms of access to electricity or consumption, are not apparent. For households of “other nationalities”, however, the picture is rather Figure 4.22—Tajikistan: Average age monthly electricity consumption, by gender ender of household head (kWh)

390 357

438 351

strange. While the share of these households using electricity rose from 79 to 98 percent during 2007-2009, 2009, their daily consumption dropped by a third (see Figure 4. 4.24 and 4.25).

Figure 4.23—Tajikistan: Share of households using electricity, by gender of household head (%)

98

379 359 96

average

Female headed 2007

Male headed 2009

UNDP calculations, based ased on LSS 2007and 2009 databases. database

160

average

99

98 96 96

Female headed 2007

Male headed 2009


Chapter 4 - Tajikistan’s Energy Sector

Household access to and use of gas. Household consumption of piped gas declined dramatically following the large price increase and interruptions in gas supply that began b in Figure 4.24—Tajikistan: Share of households using electricity by nationality (%)

96 98

97 99

98

95 97 79

2007. These declines were most notable in Dushanbe, and in villages (see Figures 4.26 and 4.27).

Figure 4.25—Tajikistan: Average monthly electricit electricity consumption per household by nationality ((kWh)

390

396

357

371

362

404 329 268

average

Tajik

Uzbek

2007

Others

average

Tajik

2009

Uzbek

2007

Others 2009

UNDP calculations, based ased on LSS 2007and 2009 databases. database

Figure 4.26—Tajikistan: Gas as pipeline users by location (%)

90

76

73

81

73

Figure 4.27—Tajikistan: Tajikistan: Gas pipeline users: poor vs. non poor (%)

46

45 25

21

21

13

Average Dushanbe

2007

71

19

11

Other towns

Villages

2009

Average

Poor

2007

Not poor

2009

UNDP calculations, based on LSS 2007and 2009 databases. database

In contrast to piped gas, LPG use increased dramatically across the count country. Small cylinders were the category mostly used both among the poor and non--poor (see Figures 4.28 and 4.29).

Household access to and use of different sources of heat. The living standard survey data indicate that a majority of the population relies on electricity and wood for heating (with the share of the latter increasing), followed by coal and dung (the share of both of which decreased in 2009 compared to 2007), as shown in Figure 4.30.

161


Poverty and Social Impact Assessment Figure 4.28—Tajikistan: Share of LPG users, poor vs. non-poor (%)

60

60

50

50

40

40

30

30

20

20

10

10

0 2007 2009 2007 2009 2007 2009 Average

Poor

Figure 4.29—Tajikistan: Share of LPG users by location (%)

0 2007 2009 2007 2009 2007 2009 2007 2009

Not poor

Average

Dushanbe

Other towns

% of hhs that use 10kg cylinder % of hhs that use 15kg cylinder % of hhs that use 20kg cylinder other

Villages

UNDP calculations, based on LSS 2007and 2009 databases. database

Figure 4.30— —Tajikistan: Household heating sources (%)

40

44 32 26 17 10

13

12 5

1. Wood

2. Electricity

3. Dung

Average 2007

4. Coal

3

5. Other

Average 2009

UNDP calculations based on LSS 2007, 2009 data.

Access ccess to and use of district heating. While the share of households who are connected to central heating systems actually increased (from 9 to 12 percent)) during 20072007 2009, they remain small in number (and are mainly in Dushanbe). When asked “if if they have central heating, whether they actually are able to use is as a source of heating”, over 90 percent of respondent households—both both poor and non-poor—answered answered in the negative (see Figures 4.31 and 4.32). Use of gas for heating. The picture for using gas for heating is different. The vast majority of households in Dushanbe and other 162

towns used gas (LPG) for heating in 2007 (compared compared to 45 percent in rural areas areas), and the pattern is the same across po poor and nonpoor. The situation wass rather different in 2009: only 45 percent used gas for heating in the small towns, 21 percent in Dushanbe, and a only 11 percent in the rural areas. Across poor and non--poor households, usage levels dropped from over 70 percent in 2007 to 20-25 25 percent in 2009. These declines seem to reflect the rising prices for LPG that took hold towards the end of 2009. Interestingly, LPG consumption trends were quite similar across poor and non non-poor households in both 2007 and 2009, reflecting its availability and desirability (compared to


Chapter 4 - Tajikistan’s Energy Sector

Figures 4.33 and 4.34).

piped gas) in 2007,, but decreasing affordability for both groups in 2009 (see Figure 4.31—Tajikistan: Share of households having and using district (central) heating, by location (%)

90.6

91.8

95.1

Figure 4.32—Tajikistan: Share of households having, using district (central) heating, poor vs. non poor (%).

100

92.3

90.6

91.8

100

91.7

90.1

90.5

75 62

60

38.8 23.5 8.5

12.2 3.7

3.7

0.6

0.5

2007 2009 2007 2009 2007 2009 2007 2009 Average

Dushanbe

Other towns

Villages

8.5

12.2

6

4.5

10.3

16.9

2007 2009 2007 2009 2007 2009 Average

Poor

Not poor

Share of households that reported central heating, % Share of households that reported availability of central heating but dwelling was never heated by this source

UNDP calculations based on LSS 2007and LSS 2009 database

Figure 4.33—Tajikistan: Share of households using LPG for heating, poor vs. non poor (%)

Figure 4.34—Tajikistan: Share of households using LPG for heating by location (%)

89.5 76

72.7

80.5

72.7

70.7

45.5 44.6 25

20.9

19.3

2007 2009 2007 2009 2007 2009 Average

Poor

Not poor

20.9

13.3

11.1

2007 2009 2007 2009 2007 2009 2007 2009 Average

Dushanbe

Other towns

Villages

UNDP calculations, based ased on LSS 2007and 2009 databases. database

Use of other fuels. Use of firewood increased in towns (to 36 percent of households, up by 5 percentage points) and in villages (at 57 percent, up 4 percentage points). The use of oil and petrol increased substantially in “other towns” (to 19 percent) percent at the expense of using dung and coal (the

latter due to increased prices, most likely). In the villages, the use of electricity for heating increased from 5 to 15 percent, at the expense of burning dung. The use of coal slightly decreased in villages as well. In Dushanbe, the use of electricity for heating ting rose from 88 to 97 percent (see Figure 4.35).

163


Poverty and Social Impact Assessment

Figure 4.35—Tajikistan: Tajikistan: Sources of heating, by location (%)

5. Other 4. Dung 3. Coal 2007

2009

2007

Dushanbe

2009

2007

Other towns

2009 Villages

2. Wood 1. Electricity

UNDP calculations, based on 2007, 2009 LSS data.

For poor households,, the use of wood increased from 40 to 51 percent with decreasing shares of coal (from 15 to 11 percent)) and dung (from 22 to 15 percent). The share of electricity remained roughly constant at 19 percent.. For the non-poor, non the use of electricity increased from 31 to 39 percent (see Figure 4.36).

expenditures, willingness to pay is best assessed through dedicated “willingness-to-pay” pay” surveys, with “contingent valuation” being the most common method used. •

Some 78 percent of rural households reported use of wood as a heating ting source in 2009, up from 68 percent in 2007. The sa same is true for the small towns, where 68 percent of households were heating with wood in 2009, up from 45 percent in 2007. Even in Dushanbe 15 percent of households reported heating with wood, up from 11 percent in 2007.

Estache et al. (2002)203 distinguish between the affordability of access and the affordability of consumption. This is an important distinction in the context of Tajikistan, where in the high mountainous areas there are households not connected to the electricity and especially gas grids.

Energy affordability Affordability concerns the ability of certain consumers or consumer groups to pay for minimum levels of service. It should be distinguished from a number of related notions: •

“Affordable” versus “low “low-cost”. Services may be low-cost, cost, but not necessarily affordable.

Ability to pay versus willingness to pay. While affordability can be studied via an analysis of income and

164

203

Estache, Q. Wodon and V. Foster (2002), “Accounting for poverty in infrastructure reform: Learning from Latin America's experience”, World Bank, Washington DC.


Chapter 4 - Tajikistan’s Energy Sector

Figure 4.36—Tajikistan: Tajikistan: Sources of heating, poor vs. non-poor (%)

5. Other 4. Dung 3. Coal 2007

2009

2007

Poor

2. Wood

2009

1. Electricity

Not poor

UNDP calculations, based on LSS 2007and 2009 databases database

Table 4.19—Benchmarks used ed in measuring affordability, share of total household income/expenditure (%)

Source

Electricity

World Bank (2002)204

10-15

WHO (2004)205

10

UN/ECE

206

UK government

Heating

Water 3-5

15 207

US government208

10

3

6

2.5

Source: “Can Can poor consumers pay for energy and water? An affordability analysis for transition countries”, Fankhauser and Tepic, EBRD (2005).

uch affordability measures warrant further Such comment, in a number of respects:

Using household expenditures rather than income tends to provide more accurate information, as income data rarely capture all sources of household income. This is particularly true for transition economies,, where informal activities often provide substantial shares of household income.

Utility expenditures can be defined either in terms of actual payments or as billed amounts. In transition economies, the differences can be substantial, since many utilities only collect a fraction of payments due.

Average expenditures on utility services relative to average household incomes can provide general information about affordability. However, assessing expenditure ratios for households belonging to different

204

World Bank (2002), Sourcebook for poverty reduction strategies, core techniques and cross cuttingissues, Washington DC 205

WHO (2004), “Energy sustainable development and health”, in Chapter 3 of “Access to electricity and heating”, WHO background papers, June , Geneva 206

Available online at http://www.unece.org/env/europe/reps.pdf

207

The UK government ment set 3 % as a burden threshold for the lowest income decile (see http://www.sustainabledevelopment.gov.uk/sustainable/quality04/maind/04j06. htm and http://www.scotland.gov.uk/library5/environment/sfps.p df). 208

The US Environmental Protection Agency (http://www.epa.gov).

165


Poverty and Social Impact Assessment

income/expenditure deciles/quintiles and for specific vulnerable groups (such as pensioners and recipients of social assistance) can provide more accurate and more useful information information. Changes in affordability should be assessed under different policy scenarios, taking into account the effects of tariff increases and income growth on the demand for utility services.

The he shares of household expenditures devoted to electricity for or the poorest quintile rose to 3.6 percent in 2009, from rom 1.7 percent in 2007. This is nott high by international standards.

However, many households use generators for lighting—expenditur expenditures for which may not be captured by these data. Also, heating expenditures can be a major burden for vulnerable households. The 2009 survey data on the shares of other energy expenses in household budgets in 2009 (see Figure 4.37) indicate that expenditures on wood and coal were a major burden on households, particularly those in the poorest quintiles. The share of wood in these expenses is highest for the poorest quintile (23 percent). ). Spending on coal was the next largest expense (around 14 percent). The burden of energy expenses was largest for the

Figure 4.43: : Share of electricity payments in total Figure 4.37—Tajikistan: Shares of household expenditures on various energy services/sources, 2009 (%) monthly household expenditure, average (%)

Country

3.6

Dushanbe

2.6

25 1.7 20

2.0

1.6

1.7 1.9

20

2.2 1.4

1.2

15

15 10 10 2007 2009 2007 2009 2007 2009 2007 2009 2007 2009

5

Poorest quintile

0

Second quintile

Average

Third quintile

Q1

Electricity

Q2 LPG

Fourth quintile Q3

Firewood

Top quintile Q4

Q5

Coal

Other

5 0 Average

Q1

Electricity

UNDP calculations, based on LSS 2007and 2009 databases database

Other towns

Q2 LPG

Q3 Firewood

Q4

Q5

Coal

Other

Q4

Q5

Coal

Other

Villages

35

25

30

20

25 20

15

15

10

10 5

5 0

0 Average

Q1

Electricity

Q2 LPG

Q3 Firewood

Q4

Q5

Coal

Other

UNDP calculations, based on LSS 2009 database. database

166

Average

Q1

Electricity

Q2 LPG

Q3 Firewood


Chapter 4 - Tajikistan’s Energy Sector

Figure 4.38—Tajikistan: Share of average verage annual energy consumption in total al household consumption, poor vs. non-poor 2009 (%)

12.8

Poor

8.9

Not poor

UNDP calculations, based on LSS 2009 database

poorest quintile in the towns: about 31 percent of their expenditures went to firewood purchases.

expenditures to wood. The poorest quintile’s use of electricity was slightly lightly larger in Dushanbe (4.5 percent of expenditures), and around the same (3.5 to 3.7 percent) in other towns and rural areas. The largest share of expenditures on coal wass in small towns at 18 percent. Approximately proximately the same share (3.2 percent) of expenditures wass devoted to LPG purchases across the country. The living standard survey data indicate that the average household in 2009 devoted some 10 percent of its expenditures to energy purchases. For poor households, this share rose to 13 percent. For rural households, this share was 11 percent; households in Dushanbe averaged only ly 5 percent. These figures indicate that shares of

Figure 4.39—Tajikistan: Average annual energy consumption in total household consumption for the poorest quintile by location, 2009 (%)

18

16

Figure 4.40—Tajikistan: Average annual energy consumption in total household consumption, by location, 2009 (%)

17 10

10

9

11

5

average

Dushanbe

Other towns

Rural

Poorest quintile

average

Dushanbe

Other towns

Rural

UNDP calculations, based on LSS 2009 database

The next in line wass the poorest quintile in rural areas, with around 22 percent of household expenditures devoted to wood. Households in the poorest quintiles quintile in Dushanbe devoted around 16 percent of their

spending on energy are within the international benchmarks on affordability. It should be reiterated however that these are averages across the year and spending on energy during the winter would be higher.

167


Poverty and Social Impact Assessment

Figure 4.41—Tajikistan: Tajikistan: Share of household expenditures on food, 2009 (%))

63.4 60.9 58.8

58.3

58.9

59.8

59

58.3

60.4 58.9

58.6

56

55.3

UNDP calculations, based on LSS 2009 database database.

Energy, food, and household indebtedness. As a result of the compound crisis of 2007-2008, vulnerable households in rural areas saw their stocks of food and seed destroyed. In urban areas the most vulnerable were forced to spend their income on alternative heating, leaving less for food and healthcare. The 2009 living standard survey sur data indicate that poor households devoted 63 percent of their expenditures to food (see Figure 4.41). When combined with data on energy expenditures, this implies that poor households spend nearly 80 percent (more, (more for selected groups, and especially in the winter time) of their income on food and energy alone—leaving little left over for other spending. This his is confirmed by the results of the regular surveys conducted by the World Food Programme and DFID (see Figure 4.42), 4. which show significant growth in the numbers

168

of vulnerable households contracting new debts during 2008-2010. Figure 4.42—Tajikistan: New debts incurred in past three months, 2008 2008-2010

58%

57% 51% 44%

47%

37%

Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 Oct 08 Jan 09 May 09 Jul 09 Oct 09 Feb 10

Source: “Tajikistan Tajikistan Food Security Monitoring System”; WFP and UK Aid, April 2010.


Chapter 4 - Tajikistan’s Energy Sector

Poverty alleviation and energy sector development The benefits of improving household electricity supplies. The large investment projects described above should help Tajikistan to reduce or eliminate energy shortages. Sangtuda-1 and -2 alone should add 890 megawatts of installed capacity in power generation. Tajikistan’s Centre for Climate Change209 estimates that the implementation of the government’s small hydropower programme should provide or improve electricity supplies to more than 240,000 rural households (around 1.5 million people).210

Willingness-to-pay studies, in which respondents are asked what they would be willing to pay to avoid a power outage;

209

Kayumov A. and Kabutov K. “Socio-Economic Assessment of the Production and Consumption of Renewable Energy Sources in the Republic of Tajikistan”, Centre of Climate Change, HydroMeteorological Agency, Dushanbe, Tajikistan.

210

This estimate, which was developed within the framework of the UNDP project on the “Development of electricity supply for rural communities of Tajikistan”, is based on an assumption that the mean level electricity demand per rural household is 5.7 kWh/day (cooking—1.6 kWh/day; lighting—0.5 kWh/day; heating—2.0 kWh/day; television—0.3 kWh/day; refrigerator—1.1; other—0.2) or 2000 kWh annually.

Production loss surveys, in firms are asked to estimate sales lost due to a real or hypothetical power outage; or

Captive generation costs, which analyze the costs of back-up electricity options (as a proxy for willingness to pay).

Tajikistan’s Centre for Climate Change estimates that the implementation of the government’s small hydropower programme should provide or improve electricity supplies to more than 240,000 rural households.

How would this affect income poverty? While there are no established indicators directly linking changes in energy/electricity supplies to poverty, the value of lost load (VoLL) shows the costs of power outages, both directly to consumers (e.g., added fuel costs to run back-up generators) as well as indirect costs, such as the opportunity costs of not having electricity to use. VoLL is typically estimated through: •

No credible VoLL estimates are available for Tajikistan; comparative studies suggest an estimated VoLL of 1.36 Euro/kWh for countries in similar circumstances.211 Country assessments can vary considerably, however. A common short-term response to energy shortfalls—especially by households and other small-scale users—is to arrange for emergency power capacity (e.g., via diesel generators). Unlike traditional power generation projects, such capacity can be brought on line in a matter of days. But because of diseconomies of scale, the costs of 211

“Regulation of continuity of supply in the electricity sector and cost of energy not supplied”, Ilaria Losa ERSE S.p.A. and Osvaldo Bertoldi - ENGINET S.r.L; available at http://www.iccgov.org/iew2009/speakersdocs/Losa-etal_paper_RegulationOfContinuityOfSupplyInTheElectr icitySector.pdf

169


Poverty and Social Impact Assessment

Figure 4.43—Tajikistan: Top ten constraints on business investments in Tajikistan

25% 23%

18%

6%

Electricity

Tax Rates

Access to Finance

Practices Informal Sector

5%

Inadequately Educated Workforce

5%

Corruption

5%

Licences & Permits

4%

Access to Land

3%

Customs & Trade Reg.

3%

Tax Admin -istration

Source: Enterprise perception survey, EBRD/WB

generating electricity in this way are typically $0.20–$0.30 $0.30 per kWh. For some countries, the price tag can be 4 percent of GDP.212 The Energy Sector Management Assistance Program (ESMAP)213 reports a willingness to pay of around $0.10– –$0.40 per kWh, for lighting and television alone.214 Based on this, a World Bank study estimates that, by adding public good benefits (such as street lighting, and “global benefits” of reduced carbon dioxide emissions, where applicable) the benefit for an average

212

World Bank (2008): “The Welfare Impact of Rural Electrification: A Reassessment of the Costs and Benefits”, IEG, Washington DC. 213 Established in 1983, the Energy Sector Management Assistance Program (ESMAP) is a global, multidonor technical assistance trust fund administered by the World Bank and cosponsored by 13 official bilateral donors. 214

ESMAP (2003): “Rural Electrification and Development in the Philippines: Measuring the Social and Economic Benefits.” ESMAP P Report 255/03, World Bank, Washington, DC.

170

household consuming 30–40 40 kWh a month is about ut $60 per month per household.215 Increased access and electricity supply reliability would also improve the business environment in Tajikistan— —especially for small and medium-sized sized enterprises. As shown in Figure 4.43, access to electricity was cited as the most important constraint on business investments in Tajikistan (by 25 percent of respondent firms) in a recent sample.216 According ccording to the Association of Power Engineering Specialists of Tajikistan, the construction and nd maintenance of small hydropower plants, and the output they generate, creates 40 new jobs per megawatt of installed capacity.217 215

World Bank (2008): “The Welfare Impact of Rural Electrification: A Reassessment of the Costs and Benefits”, IEG, Washington DC. 216

http://rru.worldbank.org/BESnapshots/Tajikistan/defaul t.aspx. 217

According to thee Concept of the development of small scale hydropower (2009).


Chapter 4 - Tajikistan’s Energy Sector

In addition, the collection of firewood, dung, and other combustibles that may be gathered (as well as purchased) imposes a costly time burden of up to 8 hours a week— time which could be used productively. On the basis of a 15-year (1988-2003) panel data set, a recent World Bank study found that the number of home businesses grew significantly faster in communities that became electrified than in either communities that did not, or in those that had been electrified earlier. Electrification also extends the working hours of home businesses, increasing net income from these activities.218 The elimination of the power cuts, and better access to electricity, can be expected to significantly reduce non-income poverty (in terms of access to quality health, education services, etc.). A recent World Bank study found that in urban areas, improving household electricity supplies can reduce both infant and under-five mortality rates; and that the effect is large, significant, and independent of income levels.219 Less combustion of traditional solid fuels such as crop residue, and dung can be expected to reduce the incidence and severity of respiratory ailments— especially for the elderly, women, and young children. A review of existing studies showed that exposure to indoor cooking using traditional methods increased the risk of premature death by a factor between two and five.220

concentration of the smallest particles per cubic meter (PM10). The extra risk of respiratory sickness from exposure to certain levels of PM10 is captured in the hazard ratio (the relative probability of the exposed versus unexposed being sick), which is 3.5. Lost adult work days average 3.0 per year, and the additional under-five mortality rate is 2.2 per 1,000. Substituting electric lighting for kerosene lamps therefore is estimated to have a quantifiable health benefit of $2.50 per household per day221. This is very important for Tajikistan, with its heavy reliance on firewood, dung, and kerosene/oil products.

The elimination of the power cuts, and better access to electricity, can be expected to significantly reduce non-income poverty. Improving energy sector governance. Movement toward cost recovery tariffs, and improving the governance and regulation of energy companies, are essential if Tajikistan is to attract the investments it needs into its energy sector. In many developing countries, price-gap comparisons understate the total size of the difference between actual and costrecovery tariff levels, particularly because of theft and poor collection rates. Such hidden costs, which are estimated to typically average

Improvements in indoor air quality can also come about through changes in lighting sources. The harmful particulates emitted by kerosene lamps can be measured by the 218

World Bank (2008): “The Welfare Impact of Rural Electrification: A Reassessment of the Costs and Benefits”, IEG, Washington DC.

219

Wang, Bolt and Hamilton (2003), ‘Estimating the potential lives saved from improved environmental infrastructure’, Environment Department, World Bank, Washington DC.

220

World Bank (2008): “The Welfare Impact of Rural Electrification: A Reassessment of the Costs and Benefits”, IEG, Washington DC.

221

Ibid.

171


Poverty and Social Impact Assessment

Figure 4.44—Tajikistan: Potential impact of increased electricity electricity tariffs on affordability (% of household expenditure by quintiles)

Dushanbe

Other towns

8 7 4

6

5 3

2

Average

5 3

5 4

3

3 2

Q1

Q2

Q3

Q4

3

3

2

Q5

Average

5

5

4

Q1

Q2

4 3

2

Q3

2

Q4

Tariff = $0.0186 (7.5 dirham)

Tariff = $0.0186 (7.5 dirham)

Tariff = $0.034 (13.1 dirham)

Tariff = $0.034 (13.1 dirham)

Rural

Q5

Country average

6 6 4 3

5

4

3

3

2 2

Average

2

Q1

Q2

Q3

2

Q4

4

4

4

2

3

2

1

Q5

Average

Q1

Q2

3

2

2

Q3

Q4

Tariff = $0.0186 (7.5 dirham)

Tariff = $0.0186 (7.5 dirham)

Tariff = $0.034 (13.1 dirham)

Tariff = $0.034 (13.1 dirham)

3 1

Q5

UNDP calculations, based on LSS 2009 database. database

4 percent of GDP, for Tajikistan have been estimated at above 10 percent of GDP.222 The government’s intention to encourage private investment in infrastructure, and in the energy sector in particular, is clear. Public-private private partnerships have provided an important vehicle for attracting investment into utilities and infrastructure in many transition economies. Evidence of improvements in the quality of service provision rovision as a result of private participation is

222

Ebinger, Jane, 2006, “Measuring Financial Performance in Infrastructure: An Application to Europe and Central Asia”, Policy Research Working Paper 3992, World Bank, Washington, DC.

172

abundant.223 In fact, Tajikistan has its own good example of the importance of good governance and public-private private partnerships in utilities, in Pamir Energy. Pamir Energy generates more electricity in Gorno Badakhshan dakhshan than Barqi Tojik had done; it increased collections, reduced losses, improved service delivery, and combined higher tariffs with donor-supported supported social protection. The better-than-expected expected revenue picture reflects public outreach (inter ( alia via working orking with local village organizations), a willingness to disconnect non-payers, non and the effective use of lifeline tariffs. The social 223

While in some cases the results are mixed, and in others unsatisfactory, not all of these projects were equally well conceived to start with. For more on this, see Ibid.


Chapter 4 - Tajikistan’s Energy Sector

protection arrangements made transparent for the first time the amount of subsidies that were being given, and to whom. Understanding where these subsidies went contributed to increased payment discipline. Rising energy tariffs and prices: potential impact on poverty. Tariff increases in conditions of widespread poverty and food insecurity can have a dramatic impact impac on vulnerable households, particularly if they are not mitigated by effective, adequate social protection.

Figure 4.45—Tajikistan: Potential impact of increased electricity tariffs on average household energy affordability

Share of energy in total household expenditures (poorest quintile)

In Figure 4.45 we show the simulated impact of increased electricity tariffs on the share of energy expenses in household budgets. (We assumed no changes in real household incomes or in the relative prices price of other energy sources). ). The poorest quintile in “other cities” would spend the m most, with around 20 percent of household expenditure going to energy. These are annual averages; the shares of energy expenditures in some household budgets will reach more than half of total monthly expenditures.

19%

13%

Average

In Figure 4.44 we show the simulated impact on household budgets of the increased electricity tariffs alone: taking as a benchmark the cost recovery ery level at 13.1 dirham ($0.034).224 The effect on poor households will be dramatic, with the poorest quintile devoting 6-8 percent of household expenditures to electricity alone. (We We assumed that household electricity icity consumption and overall household expenditures nditures remain unchanged.)

20%

19%

Dushanbe

Other towns

Rural

Poorest quintile

Share of energy expenditures in total household expenditures 12%

11%

13%

7%

Average

Dushanbe

Other towns

Rural

Average

Share of energy in total household expenditures (poor vs non-poor) non 11%

13% 9%

Average

poor

non poor

UNDP estimates.

224

Forecast parameters of social and economic development, Tajikistan, 2009.

173


Poverty and Social Impact Assessment

These figures have a rough character; more rigorous estimates would require the application of more complicated theoretical tools (e.g., computable generable equilibrium models) with demanding data requirements. In making such calculations, it is important that all relevant costs and benefits—including environmental and health externalities—are captured by the relevant modeling framework (see Figure 4.46 below).225 According to the government’s programme for tariff increases, the process of

raising tariffs toward cost-recovery levels is to continue at least through 2012. As the government skipped the increases that had been planned for July 2010 and January 2011, this process could extend into 2013 and beyond. Whether this pace is gradual enough is a very difficult question to answer, for both analytical and philosophical reasons. The example from Armenia indicates that sharper increases in tariffs are possible, if coupled with effective reform of the social assistance system (see Annex 1).

Figure 4.46—Social, economic, and environmental impact of energy subsidies (UNEP)

Source: “Analysis of the Scope of Energy Subsidies and Suggestions for the G-20 Initiative”, IEA, OPEC, OECD, World Bank Joint Report, prepared for submission to the G-20 Summit Meeting Toronto (Canada), 26-27 June 2010, adapted from United Nations Environment Programme (UNEP) 2004. The Use of Economic Instruments in Environmental Policy: Opportunities and Challenges. Division of Technology, Industry and Economics. Paris.

225

The IEA/OECD/OPEC/World Bank joint report contains some examples of such empirical studies.

174


Chapter 4 - Tajikistan’s Energy Sector

Social assistance to mitigate the impact of rising energy prices Energy sector reforms are usually based on the proposition that households should face prices that cover most (at least) of the costs of the energy they consume. If energy prices are not set at cost-recovery levels, the investments needed to modernize old and build new infrastructure will not be forthcoming. This is particularly the case in conditions of rising global primary energy product prices and, for Tajikistan, disrupted regional energy trade and supplies. With more than two thirds of their budgets going to foodstuffs and another 13 percent going for heat and electricity, low-income households are poorly placed to withstand further tariff increases. This raises questions about the magnitude, form, and effectiveness of energy subsidies—and social protection in general— for low-income households.

Lovei et al (2000)226 recommended detailed surveys of the consumers and analysis of the sector before applying alternatives in a given country. The World Bank’s 2006 regional infrastructure study227 summarizes the benefits and shortcomings of the main subsidy schemes (see Table 4.20). Key questions in selecting an appropriate subsidy scheme include:228

Coverage: What percentage of the poor is reached by this instrument?

226

Lovei L., Gurenko E. Haney M. O’Keefe P. and Shkaratan M. (2000): “Maintaining utility services for the poor, policies and practices in central and eastern Europe and the former Soviet Union”, World Bank, Washington DC.

Targeting: What percentage of the subsidy goes to the poor? Predictability: Can the poor be sure what they will receive the subsidy, and plan accordingly? Distortions: While all subsidies have distortionary elements, are these of unacceptable dimensions or character? Administrative costs: How expensive and practical is the scheme’s administration? Target consumption: How good is the scheme at ensuring that households enjoy minimal access to heat and power? Balanced approach: How good is the scheme at balancing the financial needs of power, heat, and fuel suppliers with consumers’ ability to pay? Cross-subsidization: Is subsidization taking place within a sub-sector (for example, power) or between subsectors (for example, between power and heat)?

Social protection in Tajikistan: the current system The utility payment assistance programme has been in place in Tajikistan since at least 2003. Under this programme, low-income households are granted compensation for their usage of electricity and natural gas.

227

World Bank (2006): “Infrastructure in Europe and Central Asia region: approaches to sustainable services”, DC. 228 World Bank (2010b): “Lights out?: the outlook for energy in Eastern Europe and the former Soviet Union”, DC.

175


Poverty and Social Impact Assessment

Table 4.20—Benefits and shortcomings of social assistance mechanisms Mechanism Notional burden limits approach

Benefits Benefits can be predicted with reasonable certainty. Administrative costs are relatively low. Conceptual simplicity, based on definition of realistic minimal energy needs.

Shortcomings Coverage and targeting of the poor is usually relatively low; there are heavy administrative burdens on the poor associated with its application. One of the most distortionary mechanisms of all utility subsidy mechanisms on the demand side. Costly for the budget. A network of offices is needed to administer the scheme The administrative costs to identify and target the poor may still be considerable.

Other earmarked cash transfers

The targeting ratio is relatively high; the net financial burden on utilities is low.

Coverage of the poor as achieved by earmarked cash transfer schemes is highly uncertain, and in most surveyed countries was low. It is administratively demanding. Might operate better if the amounts are transferred directly to the utility or paid to the consumers only upon production of receipts for full payment of utility bills.

Non-earmarked cash transfers

Coverage depends on the ability and willingness of the poor to meet the eligibility criteria. Least distortionary of utility subsidy mechanisms. There are no additional administrative requirements if a social assistance system is already in place. There is no financial burden to utilities or other (non-household) consumers.

Life-line tariffs

High coverage of the poor. Targeting ratio improves as the size of the initial block decreases. The benefits received are highly predictable, especially through a twoblock life-line tariff. The scheme is simple to administer.

Targeting ratio is often problematic. The percentage of cash transfers reaching the poor rarely exceeds the percentage of the poor in the population. Significant fiscal costs make this practical only in those countries where poverty levels are low (generally less than 10%), state budgets can accommodate the administrative and substantive costs, and informal incomes are low or insignificant so as not to distort the means testing necessary to design the income supplement. Payment discipline must be strictly enforced, since the recipients of cash transfers would have the freedom of choice as to how they use their income supplement. Since the poor tend to be under-represented among those with utility connections, many may not benefit. Reliable (tamperproof) metering or a reasonable proxy (such as apartment size for heating) are needed to estimate consumption. Disciplined meter readers/controllers are needed. There is a significant burden on the budget, on the finances of the utility, or on other (industrial) consumers (in the case of crosssubsidization).

Compiled from different sources. Main reference: Azerbaijan: Issues and Options Associated with Energy Sector Reform, World Bank (2005)

176


Chapter 4 - Tajikistan’s Energy Sector

Table 4.21—Tajikistan: State budget allocations for the electricity and gas assistance programme

Year

State budget allocation (million Somoni)

Allocation to electricity (million Somoni)

Share of electricity subsidies in programme

2007

25 ($7.3 million)

17 ($4.9 million)

68%

2008

28 ($8.2 million)

19 ($5.5 million)

68%

2009

36 ($8.8 million)

25 ($6.1 million)

69%

2010

36 ($8.2 million)

25 ($5.7 million)

69%

Source: ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009.

The amounts available in the past are shown in Table 4.21.229 Based on the official statistics, as of 1 January 2009 Tajikistan had 7,392,000 residents living in 1,218,500 households. Some 241,000 households (20 percent) were beneficiaries of these subsidies; their geographical distribution is presented in Table 4.22.230

Over time, the programme has undergone several modifications, in order to increase coverage and reduce leakage. In 2009, by presidential decree energy saving light bulbs were provided at no charge to 241,000 low income households, during the second half of the year. Approximately 2 million light bulbs at the cost of 18 million Somoni ($4.5 million) were distributed, absorbing more than half of the programme’s

Table 4.22—Tajikistan: Number of electricity and gas assistance programme recipient households by region

Region

Population

Households

Poor households

191,000

33,000

7,000

Sughd

2,171,000

382,000

76,000

Khatlon

2,657,000

397,000

77,000

695,000

139,000

28,000

Subordinated cities and regions

1,678,000

268,000

54,000

TOTAL

7,392,000

1,219,000

241,000

Gorno Badakhshan

City of Dushanbe

Source: ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan,

229

ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009. 230 Ibid.

177


Poverty and Social Impact Assessment Table 4.23—Tajikistan: Allowances under the social assistance program for electricity and gas

Type of service

Allowances

Usage indicators – electricity

April- September, Up to 100kWh per household

Natural gas service - available

October – March, 12 cubic meters per person

Usage indicators – electricity

April- September, Up to 150kWh per household

Natural gas service - unavailable

October – March, 30 cubic meters per person

Usage indicators - natural gas

April- September, 10 cubic meters per person

Natural gas meter - absent

October – March, 12 cubic meters per person

Usage indicators - natural gas

April- September, 20 cubic meters per person

Natural gas meter - present

October – March, 30 cubic meters per person

Source: ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009

budget. This assistance was provided in lieu of direct subsidy payments due. At present, the government pays a benefit of 9 dirhams ($0.02) per kWh, which is thought to be equivalent to the cost of a basic allocation for consumption of electricity and natural gas (see Table 4.23). The amount of compensation provided to each low-income household is based on the time of the year, actual (or estimated) electricity consumption, the availability of natural gas service to the household, and whether or not such service is metered. Households that are not connected to the electricity and gas grids do not receive these benefits, even if they are very poor. The compensation scheme involves the following steps:231

231

The description is taken from ADB (2009),“Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009: and World Bank (2010a): Tajikistan: Delivering Social Assistance to the Poorest households, Report No: 56593-TJ, Washington DC.

178

1) Identification of eligible households. Beneficiaries are identified by the district administration (hukumat), with the deputy heads of the district hukumat serving as the chairmen of the commissions that selects beneficiaries. These commissions consist of representatives from the jamoats (subdistricts), local offices of the Ministry of Labour and Social Protection, local financial departments, local statistics office, local electricity and gas offices, and others. The secretary of this commission, the only paid position (from the social assistance budget) updates the list of approved households. 2) Application. Low-income households apply for the programme, providing information on specially designed application forms while visiting the jamoat office. Applicants must submit copies of internal passports; verification of address; statements of earnings from salaries, pensions, and other sources (including remittances); and in-kind agricultural production (if any). The number of questions has expanded and become more detailed over time. Beneficiary households


Chapter 4 - Tajikistan’s Energy Sector

Table 4.24—Tajikistan: Income threshold to quality for compensation

Income threshold to qualify for assistance

Decision No.145, 03/04/07

Decision No.379, 03/08/08

Decision No. 306, 28/05/09

Official minimal monthly salary

50% of official minimal monthly salary

50% of the average monthly wage at place of residence

Source: Strengthening Corporate Management of Barqi Tojik. Prepared by Eric Harrison Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009.

must reapply every six months in order to remain eligible.

per household member) and comparing it to the monthly threshold (which is currently less than 35 Somoni— slightly under $8). The jamoat subcommissions also consider whether the household receives electricity and/or natural gas (as described above). In the case of competing households being in similar financial situation, priority is given to: families with many dependent children which have lost their bread-winner and have disabled children; families consisting of more than two people who are disabled; families headed by an unemployable disabled person; and single elderly and disabled people with unfavourable material situation when compared to other residents.234 Once qualified, the eligible household is entered into the programme for one year.235

3) Creation of lists of eligible households. In each jamoat, a subcommission prepares a list of candidates for approval by the committee at the district level. The members of the sub-commission are the rais-mahalas (village or neighbourhood headmen or women) and jamoat representatives, as well as representatives of local housing and utilities offices, if any. In practice, the jamoat and rais-mahalas do not search for eligible families. Rather, families come to the jamoat office to apply; sometimes they ask the rais-mahalas to bring them to the jamoat to apply. Once the list of beneficiaries in the jamoats is completed, the jamoats send the lists to the districts.232

5) Allocation and transfer of funds from the central budget to Amonat Bank. Once the state budget is approved, the Ministry of Finance transfers funds to the accounts for compensation payments in districts, which are managed by the district treasuries. These accounts are located at the local branches of Amonat Bank (Tajikistan’s

4) Assessment. The assessment process evaluates the applications by the local and regional commissions according to the predefined guidelines. The income threshold for needy families has been steadily lowered over time, as is shown in Table 4.24.233 The key step is establishing a household’s financial situation (its average monthly income 232

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC. 233 ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009.

234

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC. 235 ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009

179


Poverty and Social Impact Assessment

Table 4.25—Tajikistan: Budget expenditures on social assistance in 2009

Annual budget

Share of GDP

Social assistance plus social pensions

$22 million

0.45%

Social pensions

$12 million

0.25%

Social assistance

$10 million

0.2%

Electricity and gas subsidies

$5 million

0.1%

Subsidies for needy families with school children (conditional cash transfer)

$3 million

0.06%

Programmes

Main social assistance programmes

Sources: Ministry of Finance, State Agency for Social Insurance & Pensions, World Bank.

who do not use these funds to settle their electricity bills may be excluded from programme.

state-owned savings bank with an extensive branch network).236 6) Beneficiary households receive subsidies, either by collecting funds in cash from Amonat Bank (and then paying their utility bills in person), or via a transfer of funds to their Barqi Tojik accounts. Households can therefore choose between receiving the payment in cash or transferring it directly to the utility company (see Figure 4.47). Amonat Bank handles the payments to beneficiaries, paying them every two months. Beneficiaries normally pick up their benefits at branch offices, but in rural areas, some receive cash payments at their homes from bank cashiers. Beneficiaries have three months to collect their benefits, after which time any remaining money is sent back to the Treasury.237 The payments made get recorded in the lists held at the district level. These lists form the only existing record: there is no central registry.238 Beneficiaries 236

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC 237 In Dushanbe, about 7 percent of deposits were sent back. In Varzob District only about 1 percent was sent back. 238 World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC.

180

7) Reporting back. The districts report to the Ministry of Finance on the use of funds under the programme; they do not report back to the Ministry of Labour and Social Protection. Because the jamoats must return undisbursed funds to the Ministry of Finance, the Ministry has an incentive to monitor the districts’ management of these monies.

Weaknesses in the current system In the 2009 state budget, social assistance totaled $10 million, with another $12 million allocated for social pensions (which the State Agency for Social Insurance and Pensions pays to poor elderly people who have not contributed to the pension fund), as in Table 4.25. Jointly this constituted 0.5 percent of GDP, which is far below the average of 1.6 percent of GDP for the Europe and Central Asian region as a whole.239

239

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC.


Chapter 4 - Tajikistan’s Energy Sector

Figure 4.47—Tajikistan: Tajikistan: Current system of electricity ectricity and gas compensation

Barqi Tojik

Ministry of Finance

Utility Bill Information

Central Budget Funds

Utility Bill Payment Utility Bill

Lists of Households

Direct Transfers

Amonat Bank

Low-income income Households

Cash Payments

Lists of Households

Lists of Households

Regional Evaluation Commissions

Ministry of Labour and Social Protection

Semi-annual Reports

Legend Money flow Information and data flow Source: Strengthening Corporate Management of Barqi Tojik. Prepared by Eric Harrison Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009 2009.

The main social assistance programmes are the electricity and gas compensation programme, and a conditional cash transfer scheme for low-income income families with school-aged aged children. In 2009, about $5 $ million was budgeted for electricity and gas compensation.240 A recent World Bank study identifies a number of shortcomings ngs in Tajikistan’s social assistance system in general, and the electricity and gas subsidy programme in particular. These include:241 •

240 241

Small size. Social assistance in 2009 amounted to less than 3 percent of percapita monthly consumption of the poorest 20 percent (lowest quintile) of the population. This was the lowest share in the Europe and Central Asia region. Poor targeting. Only 23 percent of social assistance payments reached the

poorest quintile of the population in 2009. The second nd poorest quintil quintile received 27 percent,, while more than 30 percent of social assistance leaked to the top two quintiles (Figure 4.48). •

Poor coverage. Only about 20 percent of the poorest quintile of households received any social assistance in 2009. This was the lowest sh share in the Europe and Central Asia region.

Figure 4.48—Tajikistan: Distribution of social assistance oayments across quintiles of consumption expenditure in 2009 (%)

27.0 22.6

19.9

18.4 12.1

Q1

Q2

Q3

Q4

Q5

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, TJ, Washington DC.

Ibid. Ibid.

181


Poverty and Social Impact Assessment

As a result, Tajikistan’s social assistance system has almost no effect on poverty; the World Bank study estimates that it lowered the poverty rate by only 0.3 percentage points in 2009.

An ADB study finds that systems of bill collection and social assistance with electricity/gas are prone to fraud, due to heavy reliance on door-to-door visits. According to the World Bank’s analysis, the ineffectiveness of the electricity and gas subsidy programme reflect the following problems: •

242

The Ministry of Finance’s annual budget for this programme is based on the past year’s budget rather than on information on beneficiaries or the population in need. Instead, the Ministry adjusts the past year’s budget for changes in the price of electricity and gas and for overall inflation.

o While financial reporting (to the Ministry of Finance) imposes a degree of financial discipline on the districts, there is less accountability to the Ministry of Labour and Social Policy for non-financial aspects of the programme, such as the accurate identification of beneficiaries.243 o The absence of a central registry of beneficiaries, which limits options for analyzing the effectiveness and improving the current system. o The rais-mahalas’ weak incentives to seek out beneficiaries. o Potential difficulties for residents of remote areas (since they must visit the local jamoat office in person in order to become beneficiaries).

Ministry of Labour and Social Policy officials report some weaknesses in the work of jamoat commissions and raismahalas, including:

o The administrative burden on beneficiaries may be considerable— particularly since they need to resubmit applications every six months.

o lack of motivation in selecting beneficiary households, because they are not paid for this work, and because inspections are infrequent, resulting poor discipline.

o The assessment includes “income from various sources” as eligibility criteria: in Tajikistan as in all transition countries significant levels of economic activity goes on unrecorded.

o poor understanding of the rules of and inaccuracies in the assessments; nepotism and corruption in the system (an ADB study finds that systems of bill collection and social assistance with electricity/gas are prone to fraud, due to heavy reliance on door-to door visits242).

o Most importantly, households that are not connected to the electricity and gas systems cannot receive electricity and gas subsidies.

ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the

182

o potential conflict of interest by raismahalas, who are responsible both for collecting fees for electricity and gas use (as well as for communal service provision) as well as for disbursing funds under the electricity and gas subsidies programme.

Lifeline subsidies at Pamir Energy. A short comparison with the subsidy scheme employed by Pamir Energy is instructive. In winter time the threshold for the subsidy is Ministry of Energy and Industry, Republic of Tajikistan, September 2009 243 This list is based on World Bank (2010a), unless otherwise noted


Chapter 4 - Tajikistan’s Energy Sector

200 kWh. If a consumer has consumed, for example, 300kWh, s/he will pay: 200kWh x 0.25 US cents (the lifeline tariff) + (300-200) kWh x 2.75 US cents (normal tariff), i.e. $3.25. The threshold for the lifeline tariff in the summer is 50kWh.

means a targeting system for Tajikistan that would meet the criteria of:244 •

accuracy—the empirical ability to measure “means” and distinguish between the poor and the non-poor without distorting work disincentives;

simplicity and administrative feasibility, taking into account institutional capacity and economic conditions (such as informality); and

transparency in the weighting of eligibility criteria and consistency in their implementation across applicants.

Tajikistan’s social assistance reform programme High poverty levels combined with ineffective social assistance create a strong case for social policy reform. The poorly targeted nature of Tajikistan’s current system discourages donors from “topping off” these programmes with grant funding—as they have done, for example, in Kyrgyzstan. The example of Pamir Energy—which has successfully combined lifeline tariffs with grant funding to finance subsidized service delivery—shows that other approaches are possible in Tajikistan. The World Bank therefore recommends reform of Tajikistan’s social assistance system, by: •

consolidating the two programmes currently in operation (i.e., the electricity and gas subsidy programme and the conditional cash transfer for low-income families with children) into a single social assistance programme;

using proxy means testing to improve targeting and coverage; and

improving the management of social assistance (e.g., via the establishment of central electronic registries of applications, beneficiaries, and payments).

Of these three elements, the introduction of effective proxy means testing seems likely to be the most difficult. The World Bank argues that an appropriate proxy

The government in January 2011 started piloting this scheme, in Istaravhsan and Yavan rayon of Khatlon oblast; the pilot is to run for two years. Annual allocations will be 400 Somoni ($91), paid out quarterly through Amonat bank. The 13 criteria for the proxy means targeting were identified based on 2009 living standards survey database (see Table 4.26). In the pilot, the eligibility threshold is established at 130 points; households receiving more than 130 points will be entitled this benefit.

Income transfers versus lifeline tariffs One of the most contentious power sector reform debates concerns the desirability of tariff-based subsidies (such as lifeline tariffs) versus direct income transfers to lowincome households. Critics of tariff-based subsidies argue that these measures are expensive, poorly targeted (in that they can benefit both the poor and the non-poor), discourage energy conservation, and are impractical in situations of incomplete meterage. Defenders of tariff-based subsidies respond that, while theoretically attractive, income transfers in practice often fail to reach a large share of the poor because social policy

244

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC.

183


Poverty and Social Impact Assessment

Table 4.26—Tajikistan: Indicator composition of Tajikistan’s proxy means testing pilot

Characteristic Household size Gas oven Generator Electric radiator Refrigerator Satellite dish Car or truck Computer Household head’s employment sector Household Head’s Education Housing Roof Material Number of Children under 15 Oblast No of disabled of 1st category and disabled children in the household1 Total Number of Variables

Used for urban households *

* * * * * * * * * * * 11

Used for rural households * * * * * *

* * *

9

Source: Robert van Leeuwen (Team leader), EU/Mott MacDonald, “Presentation of the PMT system in Tajikistan”, November 2010.

institutions do not possess the capacity needed for effective targeting.245

In assessing subsidy schemes against various criteria (coverage of the poor, targeting effectiveness, benefit predictability, distortions and other side effects, and administrative costs and difficulties), Lovei et al found that instruments performing well on some criteria perform poorly on others.246 Not all subsidy mechanisms are applicable or perform equally well across all countries and utility services, and no single instrument outperforms all others.247

to protect the poor during and after tariff reform, the World Bank’s 2010 regional energy study recommends providing cash transfers to the poor as the preferred instrument, and lifeline tariffs as a second best if meters are installed.248

A 2010 IEA/OPEC/OECD/World Bank joint report argues that cash transfers have many advantages over universal subsidies and other transfers.249 The authors argue that social safety nets are much more efficient and equitable: out of 24 countries surveyed in the 2005-2008 period, 16 were transferring more than 50 percent of their social protection

A number of recent influential reports argue in favor of the income transfers. For example: 245

Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC. 246 Lovei, L., E. Gurenko, M. Hany, and P. O'Keefe. 2000. "Maintaining Utility Services for the Poor. World Bank Report 20874. Washington, D.C. 247 Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC

184

248

World Bank (2010b): “Lights out?: the outlook for energy in Eastern Europe and the former Soviet Union”, Washington DC

249

IEA, OPEC, OECD and World Bank Joint Report (2010): “Analysis of the Scope of Energy Subsidies and Suggestions for the G-20 Initiative”, prepared for submission to the G-20 Summit Meeting Toronto (Canada), 26-27 June 2010.


Chapter 4 - Tajikistan’s Energy Sector

funds to the poorest 25 percent of the population.250 However, Lampietti et al argue that income transfers tend to only be well targeted in countries with a small percentage of the population below the poverty line.251 When there are enough funds to finance the administration of social assistance and the informal sector is small, means testing is easy. By contrast, it is harder to produce welltargeted income transfers in countries where nearly half the population is poor, budget resources are insufficient, and means testing (or proxy means testing) is complicated by the presence of a large informal sector. The authors also argue that coverage of the poor varies inversely with the share of the subsidy that goes to the poor—the more households are eligible for assistance, the more likely it is that non-poor households will be eligible for assistance. That is: if a benefit system covers a large percentage of the population, the system is likely to be poorly targeted. Lampietti et al (2007) find that the case for lifeline tariffs is stronger in countries with high poverty rates, high inequality, high access of poor households to the subsidized network, and poor targeting of social transfers. Policy makers in such countries may find lifelines a more efficient way to deliver mitigating measures than direct income transfers channelled through questionable social protection systems. The report notes, however, that lifeline tariff blocks should be kept small (below 50-100 KWh), and that the government (or, by extension, donors) must

250

Nemtsov R. (2010): “Developing Effective Reform Strategies: Safety nets to protect the poor and vulnerable groups from the negative impacts of reform, Increasing the momentum of fossil-fuel subsidy reform”, presentation at the WTO/WB joint conference, 2010, Geneve, Switzerland.

251

Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC.

compensate utilities for the social transfers they provide.252 Komives et al253 compare lifeline tariff schemes that subsidize some consumption of all customers versus schemes in which only households consuming below a certain threshold are subsidized. While they find that cash transfers are generally better, they also find that differences in terms of their targeting effectiveness are not very large.

It is harder to produce welltargeted income transfers in countries where nearly half the population is poor, budget resources are insufficient, and means testing is complicated by the presence of a large informal sector. Tajikistan seems to tick off many of the boxes suggested by Lampietti et al for using lifeline tariffs. These include: high income poverty rates; relatively high connection rates (at least for electricity), compared to other developing economies; relatively high rate of metering; and serious administrative difficulties of implementing cash transfer programmes. Moreover, as described above, Pamir Energy has successfully used lifeline tariffs in Gorno Badakhshan (Tajikistan’s poorest region), financed by donor grants. Tajikistan had lifeline electricity tariffs until 2007, when they were abolished. If a decision were to be taken about reinstituting lifeline tariffs, their disadvantages should be kept in mind. These include:

252

Ibid.

253

Komives K., Foster V., Halpern J. and Wodon Q (2005).: “Water, Electricity, and the Poor: Who Benefits from Utility Subsidies”, World Bank, Washington DC

185


Poverty and Social Impact Assessment

Table 4.27—Tajikistan: Share of electricity expenditures in total household expenditures, by poverty level, under different tariff scenarios

All households Poor Very poor Not poor

Average monthly electricity consumption (kWh)

Baseline 2009 7.5 dirham per kWh

Scenario 1 2011 tariff at 9.9 dirham per kWh

Scenario 2 Cost recovery level 13.1 dirham per kWh

Scenario 3 Cost recovery level (13.1 dirham/kWh) with lifeline (100 kWh at 1.0313 dirham/kWh)

357 332 339 373

2.0% 2.8% 3.6% 1.7%

2.7% 3.9% 5.2% 2.2%

3.6% 5.3% 7.2% 2.9%

2.6% 3.7% 5.1% 2.1%

UNDP calculations, based on 2009 LSS data.

Cross-subsidies between households;

Leakages to non-poor households;

Increased room for corruption, in light of current bill collection practices (e.g., door-to-door visits);

A greater administrative burden on Barqi Tojik (particularly in the context of the company’s ongoing structural changes);

The month in which the survey was conducted (November 2009) is assumed to be an average month for electricity consumption for the year (i.e., no adjustments for winter/summer consumption patterns were made). The survey provides information on total household spending on electricity. We calculated average monthly spending for electricity and applied the tariff for the year 7.5 dirham/kWh ($0.018/kWh)254 to calculate average household electricity consumption for 2009. Thus our baseline scenario reflects the Using the 2009 average Somoni/$ exchange rate.

186

• • •

Weaker energy conservation incentives, and other distortionary effects.

Using the 2009 living standards survey data, we simulated the effects on poverty of the potential reintroduction (or rather expansion beyond Pamir Energy in Gorno Badakhshan) of a lifeline regime in Tajikistan, as follows:

254

tariff structure as of 2009 (see Table 4.27). From this baseline we consider three alternative scenarios:255 Scenario 1: 2011 tariff at 9.9 dirham/kWh ($0.025/kWh); Scenario 2: cost recovery tariff at 13.1 dirham/kWh ($0.03/kWh); and Scenario 3: cost recovery + lifeline tariffs. The lifeline threshold is up to 100 kWh at 1.03 dirham/kWh ($0.002/kWh); the tariff shifts to 13.1 dirham/kWh ($0.03/kWh) when consumption exceeds 100 kWh—the tariff regime employed by Pamir Energy.

We then calculate household electricity expenditures using the different tariffs under the three scenarios. (The reported averages are only for those households appearing in the LSS data as having non-zero electricity consumption.) The main assumptions under these calculations are: a) household electricity consumption remains unchanged; b) households consume the same amount of electricity under different tariffs; c) household incomes remain unchanged (i.e., they do not receive additional funds to cover the extra spending on electricity); and d) lifeline limits are applied to all households. 255

Using the 2010 average Somoni/$ exchange rate.


Chapter 4 - Tajikistan’s Energy Sector

Table 4.28—Tajikistan: Impact on poverty under different tariff increases

Share of poor households (not weighted)

Share of poor households (weighted)

Share of population (weighted)

Poverty level (baseline scenario)

38%

39%

46%

Poverty level under scenario1 Poverty level under scenario 2

40% 41%

41% 42%

48% 48%

Poverty level under scenario 3

40%

41%

47%

Extreme poverty level (baseline scenario)

13%

14%

20%

Extreme poverty level under scenario 1 Extreme poverty level under scenario 2

15% 15%

16% 17%

18% 20%

Extreme poverty level under scenario 3

14%

15%

19%

Poverty levels

Extreme poverty levels

UNDP calculations, based on 2009 LSS data. Table 4.29—Tajikistan: Subsidies under different scenarios

Subsidy value (baseline scenario)

Somoni 20,402,968

$256 4,945,934

Subsidy value under scenario 1 Subsidy value under scenario 3

11,658,839 13,946,359

2,826,248 3,380,772

Source: UNDP calculations

Next, we compared electricity expenditures under each scenario to those under the baseline scenario, and calculated the impact on poverty of moving from the baseline to the alternative scenarios. Table 4.28 shows the impact on poverty and extreme poverty levels, respectively. Thus, under scenario 3 the extreme poverty rate would fall by 5 percent (from 20.2 to 19.1 percent). Next, we calculated the total subsidy amount under each scenario (Table 4.29). If the introduction of a lifeline tariff were to be combined with raising tariffs to the cost recovery level, the government (or donors) would need to spend $3.4 million in order to reduce extreme poverty by 5 percent. The effectiveness of the proxy means tested income transfer now being piloted would also be constrained by the small scale of the fiscal resources behind it. The World Bank’s analysis finds that the extreme poverty 256

Using the 2009 average Somoni/$ exchange rate.

rate would fall by less than one percent if the social assistance budget for 2009 was consolidated and if benefits were perfectly targeted to the extreme poor. In other words, the lifeline regime would be more effective in terms of reducing poverty. Although our calculations are very rough, they are in line with the conclusions from Lampietti et al, described above. However, one of the main drawbacks of the lifeline regime is the leakage of subsidies to non-poor. Coady et al (2005)257 conducted a poverty and social impact assessment of Tajikistan’s electricity sector in 2005, and recommend cash transfers over the lifeline tariffs, primarily on these grounds. In our case, 40 percent of the $3.4 million subsidy would leak to households in the fourth and fifth quintiles. Additionally: 257

Coady D., Gassmann F. and Klytchnikova I (2005), “An Evaluation of the Welfare Impacts of Electricity Tariff reforms and alternative compensating mechanism in Tajikistan”, Maastricht University Maastricht Graduate School of Governance, Working paper, MGSoG/2006/WP2005/001, April 2005.

187


Poverty and Social Impact Assessment

The cash transfer scheme requires the authorities to know who the poor are, and understand their needs better, unlike the lifeline scheme where the poor are “anonymous”. This “knowledge” could be utilized for the purposes of channeling other assistance schemes to the identified poor.

Many poor households live in isolated mountainous areas where they do not have reliable access to the electricity and natural gas supply networks. The World Bank’s 2007 poverty assessment reported that poverty rates rise with the altitude of residence in the mountainous areas (e.g., in Gorno Badakhshan). These poor households are the least likely to have access to electricity, and thus to electricity and gas compensation through 258 Pamir consumption subsidies. Energy does operate a lifeline tariff regime, and while it is largely regarded as a success, numerous press reports indicate that in many remote and highaltitude areas, where households are not connected to Pamir Energy’s grid, the situation is desperate. In the sparsely populated Eastern Pamir region, which represents 26 percent of Tajikistan’s total area, there is virtually no electricity supply. Most residents there do not benefit from the lifeline regime, or from other state social assistance programmes.

258

With the current winter energy situation, in which electricity cuts reach 10-12 hours per day, reductions in electricity tariffs may not matter very much. Cash transfers would instead help finance purchases of alternate fuels (e.g., LPG, firewood).

World Bank (2010a): “Tajikistan: Delivering Social Assistance to the Poorest households”, Report No: 56593-TJ, Washington DC

188

While metering levels are high (96 percent),259 many meters are old and of uncertain quality, and are located within the apartments. These circumstances in many transition economies are associated with higher levels of corruption and bribery among meter-readers, since the billing and bill collection is done mostly door-to-door. Such problems would limit the effectiveness lifeline tariffs, should they be introduced now. With donor assistance (particularly from the World Bank, Asian Development Bank, and the Swiss Economic Cooperation Organization) new meters are being installed at building entrances, but this remetering process will take some time to complete.

In light of the above, the government’s endorsement of income transfers based on proxy means testing is understandable. Still, the small fiscal resources standing behind this system do not bode well for its impact on reducing poverty. Countries that have opted for proxy-means tested income transfers have had mixed results. (The authorities in Tajikistan would benefit from a review of these experiences; in Annex 1 we provide a case study of the Armenian experience.) These considerations, combined with Pamir Energy’s successes with lifeline tariffs, suggest that additional or transitional schemes employing lifeline tariffs could be employed. Is there room for transitional lifeline tariffs? The World Bank’s recent report on social policy reform in Tajikistan concludes that the government would have to budget an estimated 312 million Somoni ($70 million, or 1.4 percent of GDP) annually, to eliminate extreme poverty through payment of social assistance. This sum would be about seven times the amount currently budgeted for social assistance, and beyond Tajikistan’s fiscal 259

UNICEF and State Committee on Statistics Republic of Tajikistan (2009) Tajikistan Living Standards Measurement Survey 2007: Indicators at a Glance.


Chapter 4 - Tajikistan’s Energy Sector

capacity. Hence, the report recommends that the government increase the budget for social assistance, in order to finance delivery of a single proxy means tested income benefit. Having a single consolidated social assistance programme in a small country has important advantages, particularly in terms of simplification and reduced administrative costs. But introducing the pilot, testing its results, and then rolling out the programme across the country could take 4-6 years. During this time, poor households will face the burdens of tariffs rising toward costrecovery levels. Hence, there is a strong case for continuing/introducing tariff-based electricity subsidies, at least until the new social benefit scheme is rolled out nationally. Options here include: (a) continuing the present system; (b) introducing a lifeline regime; and (c) combining a lifeline regime with some categorical targeting. The benefits of transitional lifeline tariffs are described in a number of recent reports. For example: •

Lampietti et al conclude that, while poverty-targeted income transfers may be more efficient, they can take years to become operational. Ideally, tariff-based subsidies should not be phased out until targeting is significantly improved.260 Coady et al conclude that, while a comprehensive safety net system which explicitly targets poor households would be more effective than lifeline tariffs, a gradual approach combining tariff increases and the introduction of income benefits with lifeline tariffs is probably desirable. The authors also recommend a stronger differentiation between the lifeline and main tariff rates.

260 Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC

Komives et al261 explain how means testing can be used in combination with lifeline tariff schemes. The authors argue that, if well implemented, such a hybrid scheme generates solid improvements in subsidy performance.

Hence even in the medium term, Tajikistan could combine lifeline tariffs with an income transfer scheme based on proxy means testing. Or, it could subsidize only large households (or a subgroup, such as large households with many children), since the 2009 living standards survey database indicates a strong correlation household size and poverty (particularly energy poverty). The joint IEA/OPEC/OECD/World Bank study similarly argues that, since putting in place effective social safety nets can take time, governments may want to consider assisting the transition of the poorer sections of the community, depending on the specific circumstances and challenges. The suggested options include, inter alia: •

Temporarily maintaining universal subsidies on those fuels that are more important in poor household budgets;

Where poor households are connected to the grid and metering is in place, using lifeline tariffs;

Where poor households are not connected to the grid, providing subsidies to offset connection charges; and

Prioritizing broader structural expenditures that can benefit the poor, including road and rural-electrification schemes, but also expenditures on health, and education.

261

Komives K., Foster V., Halpern J. and Wodon Q (2005): “Water, Electricity, and the Poor: Who Benefits from Utility Subsidies”, World Bank, Washington DC.

189


Poverty and Social Impact Assessment

Table 4.30—Tajikistan: Tariffs at Pamir Energy and Barqi Tojik net of subsidy, for residential consumers

Average consumption per year Low (1650 kWh/year) Currency

Medium (4200 kWh/year)

High (6000 kWh/year)

Pamir Energy

Barqi Tojik

Pamir Energy

Barqi Tojik

Pamir Energy

Barqi Tojik

US$

1.62

3.40

6.81

8.65

10.94

12.36

Somoni

7.08

14.85

29.78

37.80

47.81

54.00

Source: Pamir Energy

Extending transitional lifeline electricity tariff schemes may be justified by Pamir Energy’s good experiences with their use, as well as the anticipated tariff increases during the coming years. A recent ADB study262 finds that many features of Tajikistan’s existing assistance programme could be combined with such a scheme, thereby facilitating its introduction. In particular: •

The list of households eligible for the electricity and gas subsidy programme could be used to determine initial eligibility for lifeline subsidies;

The mechanism set up with Amonat Bank to allocate central budget subsidies could be redeployed, to facilitate their transfers to Barqi Tojik (rather to households);

Barqi Tojik could provide utility bills, calculated using the lifeline tariffs, to qualified households with a copy to Amonat Bank; and

Eligible households could pay the balance of their utility bills to Barqi Tojik.

Table 4.30 compares the tariffs at Pamir Energy and Barqi Tajik net of subsidy.

262

ADB (2009): “Republic of Tajikistan: Strengthening Corporate Management of Barqi Tojik”, Harrison E., Corporate Solutions, under a contract with ADB for the Ministry of Energy and Industry, Republic of Tajikistan, September 2009.

190

Pamir Energy’s nominal tariffs for households are higher than those charged by Barqi Tojik. However, after subsidy, all Pamir Energy consumers actually pay less than Barqi Tojik consumers in the same category. Connection subsidies. In regions in which electricity and gas (or other) grids have not yet been fully extended, connection subsidies. These approaches generally take the form of a one-off subsidy per household, covering part of the costs of establishing the initial connection/access to the grid.263 Such subsidies have been shown to be quite effective, especially when combined with complementary non-price approaches to making utility services accessible and affordable to poor households.264 Heating allowances. Data from Tajikistan (as from many other transition economies) suggest that, on average, the poor spend almost twice as much of their household budgets on heating as do the non-poor. They also indicate that poor households’ demand for heat is less income- and price-elastic than nonpoor households’ demand. It should therefore be possible to design a heating subsidy that 263

IEA, OPEC, OECD and World Bank Joint Report (2010): “Analysis of the Scope of Energy Subsidies and Suggestions for the G-20 Initiative”, prepared for submission to the G-20 Summit Meeting Toronto (Canada), 26-27 June 2010.

264

Komives K., Foster V., Halpern J. and Wodon Q (2005).: “Water, Electricity, and the Poor: Who Benefits from Utility Subsidies”, World Bank, Washington DC.


Chapter 4 - Tajikistan’s Energy Sector

will benefit the poor more than the non-poor. However, as Lampietti et al note,265 several problems present themselves in attempting to design and implement heat subsidies in transition economies in general, and in Tajikistan in particular. For example: •

Household access to (non-electric) networked heat systems is very limited outside of Dushanbe.

Households have shown some ability to substitute other fuel sources (e.g., electricity, LPG, wood) for central heat, in response to price/tariff changes.

Most household heating Tajikistan is not metered.

use

in

These considerations argue against the introduction of lifeline tariffs for central heat. But providing cash grants may also be problematic, particularly in light of the negative externalities (e.g., deforestation, greenhouse gas emissions, fire and health risks) associated with off-grid heat sources (e.g., firewood, dung, coal). Subsidies or vouchers to encourage the use of clean fuels (e.g., LPG) could offer important opportunities in this respect. In the longer term, the expansion of centralized heat generation and distribution capacity—perhaps via electricity cogeneration in combined power and heating plants—could be an important part of the solution. Tajikistan is planning to build two combined heating plants in Dushanbe, and other coal-fired plants elsewhere. In addition to increasing the numbers of households with access to dedicated heat networks, these plants will increase the scope for employing lifeline heating tariffs.

Other possible energy subsidies. International evidence indicates that subsidies for LPG (and kerosene) are generally less regressive than electricity subsidies, as these fuels are more likely to be used by the poor for cooking, and for lighting in rural areas.266 In contrast to coal and fuel oil, LPG and piped gas have a relatively modest environmental footprint. These considerations argue for the reinstatement of LPG subsidies—possibly via donor financing. Likewise, as suggested in Lampietti et al,267 subsidies for energy efficient household appliances, insulation, and other energy-saving technologies could be considered—possibly via vouchers. The experience of Ukraine’s Housing and Municipal Services Allowance programme, which was in effect during 19952004, may be important in this context. Subsidies were paid under this programme to households if their total monthly housing expenses exceeded 15 percent of their average gross income during the preceding three months. This programme allowed the government to increase utility prices (and housing rents) substantially. One report found that, without these allowances, half of all Ukrainian households would have had to devote more than 50 percent of their income towards housing costs.268 Other, subsequent studies269 painted a less flattering picture; concerns were raised about the programme’s financial 266

Nemtsov, R. (2010): “Developing Effective Reform Strategies: Safety nets to protect the poor and vulnerable groups from the negative impacts of reform, Increasing the momentum of fossil-fuel subsidy reform”, presentation at the WTO/World Bank joint conference, 2010, Geneva.

267

Lampietti, J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC 268

265

Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC.

Mikelsons M, “Other Country Experience with a Consumer-Based Housing Subsidy”.

269

RAND Graduate School (2010): “Developing a Safety Net for Ukraine”, Rohozynsky O., PhD dissertation, 2010

191


Poverty and Social Impact Assessment

sustainability, targeting, and coverage. In response, Ukraine’s Ministry of Labour and Social Policy and the World Bank launched the Social Assistance System Modernization project in 2006. The World Bank provided $99 million for this project, aiming to enhance the effectiveness of the social assistance system.

192

The key principle of the new system is the “one stop shop� model, similar to the system used in Great Britain. While the Housing and Municipal Services Allowance programme was eventually replaced in Ukraine, it can potentially work well.


Chapter 4 - Tajikistan’s Energy Sector

Conclusions and recommendations Access to energy sources/services. While the electricity connection rate is high on average (98.4 percent in 2009), in remote and mountainous regions the connection rate is lower. Households across the country, and most notably in rural areas, experience power cuts reaching 10-15 hours a day, especially in the winter time, with negative impacts on the health status of the population, social services, education and environment. Households, especially those in the urban areas use electricity for cooking and heating, given the deterioration of the gas supply and district heating in the recent years. Households in the rural areas have shifted to burning wood and dung. In contrast to natural gas, use of LPG was high in 2007, but then declined by 2009 (from around 70 percent to 25 percent) due to high retail prices. The same is true for coal. This leaves limited alternatives for poor households; the use of wood is therefore as high as 68-69 percent in the villages and small towns. This leads to deforestation, with Tajikistan’s mountainous regions having lost up to 70 percent of wood cover since the late 1990s. Affordability. The share of expenses on electricity in the households’ expenditure budgets in 2007 and 2009 were 1.7 percent and 3.6 percent on average respectively, for the poorest quintile. These are not high by international standards. While average spending on energy sources and services was 10 percent and for the poorest quintile 13 percent (17.5 percent for the poorest quintile in the towns and 16.7 percent in villages), for selected groups of households this share can be much higher. Potential impact of restoring supply reliability. The implementation of the large investment projects in generation and transmission infrastructure would allow Tajikistan to eliminate energy shortages. Sangtuda 1 and 2 will add 890 MW in

capacity and 3.7 billion KWh in generation annually. The implementation of the state programme of small scale hydropower plants has a potential to radically improve access and use of electricity in the remote rural areas. Increased access and supply reliability will also improve the business enabling environment in the country, facilitating business development and attracting investment. The need to attract private capital. According to Asian Development Bank projections, Tajikistan’s electricity demand will grow at an annual rate of 1.9 percent, during 2010-2030.270 Tajikistan’s energy sector has the potential to meet this demand— particularly in hydro (big and small) and coal; possibly also in gas. The potential returns on energy savings and energy efficiency projects also seem significant. However, potential does not necessarily translate into development. Like many developing countries, Tajikistan does not possess the domestic capital, technology, or managerial capacity needed to fully realize its energy potential. Moreover, Tajikistan is competing for this capital and technology with many other countries—including its neighbours. A recent World Bank report271 finds that, without $3.3 trillion in new investments in primary energy development and power sector infrastructure over the next 20 years, the Europe and Central Asian region could be moving from being a large (at present) net energy exporter to a net energy importer by 2030.272

270

APEC/ADB Energy Outlook for Asia and Pacific, 2009.

271

WB (2010b):“Lights out?: the outlook for energy in Eastern Europe and the former Soviet Union” 272

http://web.worldbank.org/WBSITE/EXTERNAL/COU NTRIES/ECAEXT/0,,contentMDK:21722062~menuP

193


Poverty and Social Impact Assessment

The need to improve the investment environment. Tajikistan’s prospects in this competition depend on a number of factors, including the extent of national commitment to the energy sector, the legal and regulatory environment, and domestic tariffs/prices. Its successes in attracting foreign investment in national energy projects like Sangtuda-1 and the South-North electricity transmission line, and its strong support for the Roghun hydropower plan, underscore the strength of the government’s commitment to developing Tajikistan’s energy sector.

Like many developing countries, Tajikistan does not possess the domestic capital, technology, or managerial capacity needed to fully realize its energy sector potential. However, in other respects— particularly when it comes to attracting private capital, the picture is not so rosy. The relative paucity of private capital and technology inflows suggests that the burden of developing Tajikistan’s energy sector will continue to be born largely by the state budget. In light of the other development challenges facing the country, the state’s ability to bear this burden—and the opportunity costs associated with trying to do so—seem likely to remain in question for the foreseeable future. The alternative—improving the energy sector’s enabling environment via more ambitious legal and regulatory reform, combined with higher and rationalized tariffs (with elimination of cross-subsidies), and greater attention to the social protection of those households most vulnerable in the face of higher tariffs—need to be considered. The implementation of the government’s small hydropower programme would likewise benefit from measures to create a more K:258606~pagePK:146736~piPK:146830~theSitePK:2 58599,00.html

194

attractive enabling environment, particularly in terms of further progress in developing feed-in tariffs for electricity generated from independent power producers. Tajikistan has its own example of the impact of good governance on electricity utility performance: Pamir Energy, the first public-private partnership project implemented in Central Asia in the energy sector. Commercially oriented management at Pamir Energy was able to: generate more electricity; increase collections; reduce losses, and provide an improved service. Better-thanexpected revenues were achieved as a result of public relations efforts, a willingness to disconnect non-paying users, payment flexibility, and work through village organizations. The agreements on social protection made transparent for the first time the amounts of subsidy given and to whom they were given. This helped people understand where their payments go, which further contributed to payments discipline. The need for parallel reforms in social assistance. The current system of social assistance, whereby the government pays a benefit of 9 dirhams (0.09 somoni) per kWh to around 240,000 households for electricity consumption levels between 100-250 KWh (depending on whether the service is metered or not and the time of the year) is ineffective and poorly targeted. The government intends to consolidate all current social assistance programmes into a single transfer programme based on proxy means testing. The arguments for this form of social assistance for Tajikistan—despite its administrative costs— are well grounded. The scheme is currently being piloted in the Khatlon region; the pilot phase is intended to last through the end of 2012. Then comes the pilot’s evaluation, which will take another 6-12 months. If assessed positively, the new system will be rolled across the country; such an exercise will take another 1-2 years. Thus, reforming Tajikistan’s social assistance network will take at least 4-6 years.


Chapter 4 - Tajikistan’s Energy Sector

Potential parallel/transitory social assistance schemes for the energy-poor. During this time, Tajikistan has two options: keeping the current system in place (possibly with some modifications to improve coverage and targeting); or combining it with a supplementary scheme, ideally one that can be quickly introduced and can be based on the current administrative system. The option of reintroducing lifeline electricity tariffs, coupled with some categorical targeting (for example, for large households), and supported in part by donor financing (as per the Pamir Energy model) during this 4-6 year transitional period should therefore be considered. Such a system can be introduced quickly, and could be corrected as needed during the transition period. Irrespective of whether a transitional lifeline option should adopted, the existing system could be improved in the identification of qualified households, in line with the strongest contributors to poverty identified in the 2009 LSS data, and further elaborated in the pilot scheme for the introduction of the proxy means tested targeted cash transfers— most notably, household size. Stronger donor support for subsidized grid connections (for electricity and piped gas) and for installing/updating meters and for LPG are recommended. Assistance for sustainable firewood harvesting might also be considered. Implications for technical assistance. With the passage of the 2010 Law on Renewable Energy, regulations for the law’s effective implementation need to be developed. The same is true for the energy efficiency. UNDP is helping the Government with the development of a national Energy Efficiency Master Plan, as well as with the concept of Renewable Energy and Energy Efficiency Fund. UNDP’s assistance package could be expanded to include support for the development of a national heating strategy. As for the other funding agencies, special credit lines through local banks for funding small hydropower projects could contribute greatly to the implementation of the national

programme, as was the case in many other transition countries. For this to happen however, a better enabling environment should be created for the renewable energy sector. Tajikistan urgently needs to design and implement a national heating strategy. This could resolve uncertainties regarding the extent of the desired rehabilitation or expansion of the central heating system, as opposed to other alternatives, either of a centralized (e.g., electricity, piped gas) or decentralized (e.g., LPG) nature. The articulation of such a strategy could also help clarify the appropriate assistance programmes to mitigate the impact of higher heating costs on low-income households (e.g., via connection subsidies, lifeline tariffs, or income transfers). Need for better energy poverty related data. The importance of questions of access to and affordability of energy for poor households in Tajikistan underscores the need to improve the living standards surveys and the data they produce. There is likewise an urgent need to conduct willingness-to-pay survey research, to improve our understanding of household preferences, energy demand elasticities, and their possible implications for setting correct lifeline tariff levels. A multisector willingness-to-pay study, looking at energy, water and sanitation, and potentially other communal services, could be invaluable. Implications for future research. Some estimates of the impact on GDP, household incomes, and poverty rates of measures to improve energy supply reliability are made above. But these are rough approximations developed within the framework of partial equilibrium analysis; more comprehensive macroeconomic linkages are not captured. The complex relationships between energy infrastructure development and poverty reduction are best analyzed via the use of computable general equilibrium or macroeconomic models, such as PAMS (poverty analysis macroeconomic simulator). The design of such a model for Tajikistan 195


Poverty and Social Impact Assessment

would improve policy makers’ ex ante understanding of how specific policy changes or reforms (e.g., increased tariffs, the Roghun financing campaign) will affect national aggregates (e.g., personal consumption), household incomes and expenditures, and poverty levels.273 The use of such techniques requires extensive investments in data collection and model building. While these could be costly exercises, the costs of not making these investments could be greater. Such models can be constructed according to the principles of social costbenefit analysis, to evaluate the inter-temporal impacts of energy subsidy changes on social welfare, through identifying, measuring, and discounting future and external (as well as private) costs and benefits associated with different policy options or investment decisions. Such an approach could help align commercial investment decisions made within the energy sector with criteria for mitigating— and especially adapting to—climate change.274 For example, the commercial feasibility of large hydropower projects like Roghun may depend on the pace of glacier melt in the Amu-Darya basin. Should global warming and glacier melting accelerate, a temporary surge in water flow along the Vakhsh cascade could result—followed by significant subsequent reductions, if the glaciers’ size and “water tower” capacities were to be considerably reduced. As a multi-billion dollar water project can only be commercially viable if it is amortized over many decades, the correct inclusion of the possible environmental costs associated with climate change adaptation could be crucially important in Roghun’s future.

273

Pereira da Silva L.A., Essama-Nssah and Samaké I.: “Linking Aggregate Macro-Consistency Models to Household Surveys: A Poverty Analysis Macroeconomoic Simulator (PAMS)”

274

IEA, OPEC, OECD and World Bank Joint Report (2010): “Analysis of the Scope of Energy Subsidies and Suggestions for the G-20 Initiative”, prepared for submission to the G-20 Summit Meeting Toronto (Canada), 26-27 June 2010.

196


Chapter 4 - Tajikistan’s Energy Sector

Appendices Appendix 4.1—Armenia Case Study Armenia – electricity sector reforms Background: In the late 1980s and early 1990s, Armenia’s economy suffered a catastrophic earthquake, the breakup of the Soviet Union, protracted conflict, and the closure of borders with Azerbaijan and Turkey. Political and economic isolation—landlocked and entirely dependent on imported oil and gas—compounded the effects of rising energy prices. The cost of supplying electricity and central heating skyrocketed, while residential electricity prices remained very low. Unable to cover internal maintenance costs and crippled by the shutdown of the country’s nuclear power plant and weekly interruptions in natural gas supply, by 1992 electric utilities were on the verge of collapse. Residential consumers bore the brunt of the utility crisis. From 1992 to 1995, most of the population received only two to four hours of electricity per day, and central heating and natural gas supplies were virtually terminated. The economy also suffered as public infrastructure and the industrial sector were hit by shortages. Consumers stopped paying their utility bills, and in 1994, payment for electricity fell to only 10 percent of billings. With district heating also gone, residents of the capital Yerevan burned trees, telephone poles, and books to get through the winter, and deforestation for firewood took place on a devastating scale. Initial recovery: In 1995 the economy began to stabilize, and the government embarked on reforms to put the energy sector back on its feet. The focus was on restructuring and regulating the energy sector, raising tariffs, improving payment discipline, and making the electricity supply more reliable. The result was a dramatic improvement in the supply of electricity: by 1999 most households were again receiving service 24 hours a day, and outages were shorter and less frequent. Increasing cost recovery by utilities became a cornerstone of the government’s economic reform program. Until 1999, Armenia had a lifeline tariff structure. The first 100 kWh of electricity consumed cost households $0.029/kWh; the second bloc (100–250 kWh) cost $0.038/kWh; and the third bloc (above 250 kWh) cost $0.048/kWh. The first 100 kilowatt hours (kWh) of electricity consumed was highly subsidized for all households. Dramatic tariff adjustment: A pivotal moment in Armenia’s electricity sector reform was a tariff increase in January 1999, several years into the reform program and after the height of Armenia’s energy crisis. The increase was large—to $0.048/kWh—for all households. For poor households, this shift from a lifeline to a single rate system amounted to an average tariff increase of some 47%. These higher tariffs were coupled with the introduction of generalized social transfers targeted at low-income households. Restructuring the tariff had a disproportionately negative impact on the poor. Poor households cut electricity consumption by more than nonpoor households; the percentage of poor households with arrears was higher; and the average size of their arrears increased more. Poor households reduced electricity consumption by some 20% (from 152 kWh per month in 1998 to 121 kWh in 1999)—enough for a refrigerator and a few light bulbs. Consumption declined significantly more among rural households (26%) than urban (13%)—probably because rural households had greater access to substitutes. Despite this 20% reduction in consumption by poor households, their average bills increased by 13 percent (from dram 2,680 in 1998 to dram 3,020 in 1999). The tariff increase was 50% greater than what was originally conceived when the mitigating transfers were formulated, so their impact on poor households was underestimated—underscoring the need for careful modelling and mitigating the reform’s consequences. Parallel social assistance reform: The government took two actions to minimize the impact on the poor of the 1999 tariff increase. First, a newly designed family benefit, targeted at the 28% of the households living below the poverty line, was introduced in 1999. Second, an additional 9% of households not eligible for the family benefit, but expected to have difficulty paying their electricity bills, received a smaller sum to assist with electricity payments. While the reduction in electricity consumption and increases in total electricity expenditures experienced by these households were similar to poor households overall, the households receiving cash transfers increased average monthly payments to the utility by 4%. It is not clear whether the cash transfers offset the adverse impact of the tariff increase. However, these transfers may have prevented greater declines in consumption and larger arrears among poor households. Lessons: Opponents of the pre-1999 lifeline tariff system correctly argued that it benefited 100% of consumers, when only 33% were classified as poor. But with the new tariff structure, only 55% of the poor actually received the income transfers, leaving almost half of them uncompensated for the 47% tariff increase. With limited access to low-cost substitutes, increases in tariffs and collection rates meant greatest hardship for the urban poor, who devoted 16% of their monthly cash expenditures to electricity. Moreover, the revenues accruing to electricity suppliers only increased by about 6% from sampled households, thanks to falling consumption and growing arrears. This suggests that the benefits of the reform programme did not materialize as quickly or easily as intended, and that tariff increases must be accompanied by moves to encourage greater payment. Further developments: As electricity consumption dropped, consumption of natural gas and wood for heating has increased. This highlighted the need for a heating strategy. Since 2002, Armenia has had a national heating strategy. Use of gas for heating has increased, while use of wood has declined. The economy has continued to grow and poverty has declined significantly since 1999. Armenia has continued to reform the electricity sector, including the sale of its loss-making distribution network to foreign strategic investors. By 2004, collections had reached almost 100%. Long term impact: The Armenian reform experience raises important questions about the poverty and social impact of energy sector reform, particularly over a longer period. How can cost recovery be improved while maintaining a balance with social protection? If affordability is an issue, how can tariffs and collections both be increased at the same time? And what difference, if any, does it make if a private operator— with stronger motivation to improve revenue collection—enters earlier in the process? Source: Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC.

197


Poverty and Social Impact Assessment

Armenia’s Family Poverty Benefit Programme Armenia’s social protection system has become better targeted since 1999, and efforts to improve it further continue. Currently four types of benefits are in place in Armenia: family benefit, lump-sum childbirth allowance, childcare allowance, and benefits to the families of deceased national heroes of Armenia. Much importance is attached to improving targeting of these programmes. Eligibility for the family benefit is determined by the “family vulnerability score”, which reflects household incomes, the number of household members unable to work, domicile size and conditions, pensioner status, and the like. Each indicator has its numerical value, and the family vulnerability score is determined by the product of such values. The higher the score, the more vulnerable the family is judged to be, and the greater its eligibility for benefits under this programme.

Targeting: By 2006, coverage under the family benefit programme among the bottom consumption decile improved to 61% by 2006. Approximately 45% of programme benefits resources went to the poorest 10% of Armenia’s households. Lessons: Challenges in further improving the family benefits programme include improving the:  Motivation/compensation for the social workers who determine family vulnerability scores. The system, especially in its early days, was prone to fraud and corruption.  Inclusion/exclusion indicators in the family vulnerability scores. For example households who receive a refrigerator or a computer as gift can be included in the programme.  The drastic differences in living standards in Yerevan versus other regions need to be better reflected in the income transfers to households in other regions.  The current mechanisms for defining benefit size direct assistance to families with children, and to families living in high mountainous and border regions. Other categories of vulnerable households may suffer from this emphasis. Source: “Social Workers Perception of the Family Vulnerability Assessment System in Armenia, USAID Armenia, 2008.

198


Chapter 4 - Tajikistan’s Energy Sector

Armenia’s connection subsidies for heating In line with the national heating strategy (developed in 2002), the use of gas for heating has increased, while use of wood and electricity for heating declined. This was made possible in part by a $7 million programme ($530,000 of which came from the government, and $470,000 of which came from participating households; the rest came from donors) funding connection subsidies for poor households (so defined under the family benefits programme). This programme allowed some 10,000 households to install gas heaters and in some cases boilers.

Lessons learned: 1) Institutional barriers were significant in terms of limiting the installation of gas boilers in apartment houses. Underdeveloped housing associations and difficulties in ensuring the realization of collective payments limited the application of this programme to such buildings. 2) Early and active involvement of central and local governments is crucial for taking such programmes to scale. Early involvement in implementation—particularly in terms of informing potential beneficiaries, collecting applications, encouraging household cofinancing, and the subsequent recognition of their contributions—was particularly important. 3) Adequate public outreach is critical. A number of tools were employed; direct mailing worked best. 4) When household gas tariffs accelerated sharply toward the end of the past decade, household

interest in gas heat declined and demand for electric heat rose. This underscored the need for effective long-term forecasting of energy sector pricing and other commercial trends.

Sources: •

Lampietti J.A., Banerjee S.G., Branczik A. (2007): “People and Power: Electricity Sector Reforms and the Poor in Europe and Central Asia”; World Bank, Washington DC;

IEA, OPEC, OECD and World Bank Joint Report (2010): “Analysis of the Scope of Energy Subsidies and Suggestions for the G-20 Initiative”, prepared for submission to the G-20 Summit Meeting Toronto (Canada), 26-27 June 2010;

GPOBA (2008): Armenia OBA scheme for provision of gas and heating, SDN week, Washington DC.

199


Poverty and Social Impact Assessment

Appendix 4.2—Major development projects, partners in Tajikistan’s energy sector Development Partner

Assistance

Project Name

Duration

Amount ($ million)

ADB

Loan

2000-2009

34.0

ADB

Loan

2002-2005

5.0

ADB

Loan

2006-2010

22.0

ADB

Grant

Power Rehabilitation Project – rehabilitation of power transmission and distribution facilities in the war-damaged areas of Khatlon and Dushanbe regions Emergency Baipaza Landslide Stabilization Project – restoration of water flows in the Vakhsh River to pre-earthquake flows Regional Power Transmission Interconnection Project – construction of a 220 kV double-circuit transmission line to link the power grids in Afghanistan and Tajikistan Nurek 500 kV Switchyard Reconstruction Project

2006-2014

55.0

ADB

TA

Improving Barki Tojik’s Billing and Collection System

2000-2005

0.5

ADB

TA

Introducing International Accounting Standards at Barki Tojik

2000-2003

0.5

ADB

TA

Strengthening Corporate Management of Barki Tojik

2006-2010

1.5

ADB

TA

Regional Power Rehabilitation

2009-2010

0.5

ADB

TA

Development of an Energy Conservation Program

2002-2005

1.1

ADB

TA

Power Sector Development

1999-2003

0.9

ADB

TA

2003-2006

0.4

2006-2008

55.0

China Eximbank

Loan

Improving the Accounting and the Financial Management System of the Subsidiaries of Barki Tojik Construction of the transmission line 220 kV “Loiazor-Khation”

China Eximbank

Loan

Construction of the transmission line 500 kV South-North

2006-2009

318.0

Government of India IsDB

Loan

Rehabilitation of Varzob HPS-1

2008-2012

13.0

Loan

2000-2009

15.0

IsDB

Loan

2006-2010

10.0

IsDB

Loan

Power Rehabilitation Project – rehabilitation of power transmission and distribution facilities in the war-damaged areas in Khatlon and Dushanbe regions Regional Power Transmission Interconnection Project – construction of a 220 kV double-circuit transmission line to link the power grids in Afghanistan and Tajikistan Completion of the Sangtuda-2 hydropower plant, 220 MW

2006-2012

180.0

IsDB

Loan

Construction of Small Hydro Power Plants in Rural Area

2004

9.0

KFAED

Loan

Rehabilitation of Dushanbe City Distribution Network Project

2004-2009

13.0

Nurek Switchyard 220 kV Rehabilitation Project

2008-2011

38.0

Regional Power Transmission Interconnection Project – construction of a 220 kV double-circuit transmission line to link the power grids in Afghanistan and Tajikistan Completion of Sangtuda-1 hydropower plant, 670 MW

2006-2010

9.0

2005-2009

800.0

2000-2009

8.0

2007-2011

8.0

2003-2012

5.0

KfW

Loan/Grant

OFID

Loan

RAO-UES

Loan

SECO

Grant

SECO

Grant

SECO

Grant

Power Rehabilitation Project – rehabilitation of power transmission and distribution facilities in the war-damaged areas of Khatlon and Dushanbe regions Energy Losses Reduction Project – Reduction of the commercial losses in the electricity and gas systems Pamir Private Power Project

World Bank

Loan

Pamir Private Power Project

2002-2010

10.0

World Bank

Grant

Energy Loss Reduction Project

2005-2012

18.0

World Bank

Grant

Energy Emergency Project

2008-2010

6.5

ADB = Asian Development Bank, HPS = hydroelectric power station, IsDB = Islamic Development Bank, KFAED = Kuwait Fund for Arabic Economic Development, kV = kilovolt, MW = megawatt, OFID – OPEC Fund for International Development, RAO-UES = Unified Energy System, SECO = State Secretariat for Economic Affairs (Switzerland), TA = Technical Assistance Source: Asian Development Bank: “Country Partnership Strategy: Tajikistan, 2010–2014”, Manila, Philippines

200


Chapter 4 - Tajikistan’s Energy Sector

Appendix 4.3—Energy sector investment projects: Budgets and funding gaps Investment projects in electricity (US$)

Measure

Total funds needed

Budget

Approved USD Foreign Donors

Gap

In Country Projects 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Implementation of the project “”Development of Regional Electricity Market” Conducting of an analysis and implementation of measures to reduce energy loss Determination and implementation of the schedule of consumer supply – heat, electricity and gas, primarily in winter season Gas exploration, establishment of a condensed gas station network (Gazprom contract) Investment Program for the construction of ETI-220 kW Loladzor-Kulob, the substation of Khatlon 220kW Implementation of the project “Dushanbe -2 Heating and electricity centre (270 mWt) FDI Investment project Sangtuda 2 1st phase of Roghun HPP Feasibility study and independent assessment Roghun Concession contract for Electric networks in GBAO Ecological expertise for major projects (underground) Investment restoration of Nurek HPP Restoration and rebuilding of Nurek New Investment in Nurek Construction of 220 kWt ETL Khujand-Aini Investment in Kayrakkum HPP Investment in Varzob HPP New Investment in Varzob HPP Investment in Vakhsh HPP New Investment in Vakhsh Restoration of the energy sector in 1,2 phases Project on ETL of Sangtuda Dushanbe 500kW 228km with substation at Sangtuda Construction of 500kWt ETL Sangtuda 1-Major HPPAfghanistan Construction of HV line 220 kWt Loladzor- obi-Mazor and substation Power Loss reduction project Project on construction of 220kW ETL of Kayrakkum-Asht (86 km) Subtotal – in country

25500000

25500000

48000000

48000000

500000

500000

8000000

8000000

58200000

58200000

400000000 256000000 280000000 700000 26500000 405000000 87500000 321000000 21600000 125000000 250000 40000000 240000000 57000000 68400000

40000000

400000000 216000000 280000000 700000 26500000 405000000

4000000

17000000

5900000

62500000

321000000 21600000 125000000 250000 23000000 240000000 57000000 0

161000000

161000000

32000000

32000000

58100000 30000000

2900000

55200000

52800000

1703900000

54000000 300000 450000000 504300000

14000000 500000

40000000

14500000

40000000

2500000 450000000 452500000

3270850000

67300000

1743900000

2213850000

16300000 2766550000

0 30000000 16300000 1761350000

Regional projects 27 28 29

Project for regional intersystem electricity transfer line Project on electricity trade between Central and South Asia Construction of ETL within CASA_1000 (750 km) Subtotal – Regional projects Total

Source: adapted from PSRP Tajikistan 2010-2012

201


Poverty and Social Impact Assessment

Investment requirements for key coal mines (in US$ million) Budget financing

External financing

Names of enterprises

2007-2008

2008-2010

2010-2015

2007-2008

2008-2010

2010-2015

Section construction “East” deposit FonAygnob

19.0

-

-

44.2

-

-

Plan construction by production of coke

-

4.44

-

-

10.36

-

Section construction and mine “Ziddi”

-

19.05

-

-

44.45

-

Section construction and mine “NazarAylok”

-

-

20.9

-

-

48.78

Source: Ministry of Energy and Industry of Tajikistan

Examples of small scale hydropower plants funded with IFI support Funder

SHPP

Asian Development Bank (ADB)

Dushokhzamin plant in the Nurabad district Kalandak plant in the Rasht district

The Islamic Development Bank funded the construction of eight small power plants in rural Tajikistan. The loan of US$ 9.3 million provided by the IDB will support the construction of another five small hydropower stations in the rural Tajikistan.

2750 KW Marzich station (Ayni district in Sughd) 667 KW Sangikor station in the Rasht district (northeast) 600 KW Fathobod station in Tojikobod (northeast of Tajikistan) 850 KW Pitavkul station in Jirgatol (northeast of Tajikistan) 100 KW Shahboloi station in Nourobod (eastern Tajikistan)

The Government allocated US$ 2.4 million to utility company “Barqi Tojik” for construction of following small hydropower plants:

360 KW Khorma station in Baljuvon (Khatlon) 500 KW Toj station in Shahrinav (central Tajikistan) 700 KW Shirkent station in Tursunzoda (central Tajikistan)

UNDP

30KW Doshtmandi, Baljuvan district 40 KW Yol, Shuroobod district 10 KW Hissorak Shuroobod district 10 KW Safedob Shuroobod district….. 100 KW Nurofar, Vahdat district

Based on Tajikistan: In-depth review of the Investment Climate and Market Structure in the Energy Sector, Energy Charter Secretariat, 2010

202


Chapter 5: Water and Communal Services in Kyrgyzstan

Author: Gabriel Regallet Editor: Ben Slay

203


Poverty and Social Impact Assessment

Executive summary posting a 119 percent increase (i.e., more than doubling). On the other hand, the most recent (2009) official household survey data indicate that households in the three poorest deciles devote only 0.35 percent of their expenditures to water and sanitation, and another 0.1 percent to solid waste. This is well below the suggested 4-5 percent international affordability benchmark. However, as part of the national policy dialogue on water issues, the government has pledged to significantly increase tariffs during the next two decades. Should tariffs continue to increase at the rates recorded during 2007-2010, questions about the affordability of these services for poor households will become more important.

This chapter assesses current trends in Kyrgyzstan’s communal services sector— water, sanitation, and solid waste management services—with a particular focus on: •

poor and vulnerable households whose welfare could be affected by tariff increases and sector reform;

customers’ perceptions of, satisfaction with, access to, delivery and prices of communal services, and relevant social protection measures for poor households; and

price- and non-price mechanisms affecting vulnerable households’ access to communal services, and of the policies and reforms affecting them. •

Key findings Regarding access to, and quality of, communal services in Kyrgyzstan: •

204

Coverage of the rural population with improved water supply has increased since 2004, to reach 50-60 percent of the rural population, due to the construction of new and reconstruction of existing infrastructure (with funding international financial institution and other donors, as well as the central government under the Taza Suu programme). However, less than 25 percent have access to sewerage. For Kyrgyzstan’s 25 largest cities, drinking water coverage varies from 60-90 percent, depending on the city. It is less than 40 percent for improved sanitation and solid waste collection. Official consumer price data show that household water tariffs rose 87 percent during 2007-2010, with Bishkek

There is little evidence to suggest that Kyrgyzstan’s social protection system has a significant impact on vulnerable households’ access to communal services. Nor are there special benefits intended to promote such access. Regarding the financial and institutional viability of the sector:

275

The financial viability of many communal service providers is at risk, due to tariffs set below cost recovery levels, coupled with high (35-45 percent) non-revenue water levels275 and inadequate institutional capacity among service providers and the local government bodies that supervise them. Prospects for meeting the investment needs for infrastructure

Water that is produced but “lost” before it reaches the customer. These losses reflect leaks (physical losses), theft or metering inaccuracies (apparent losses), and unbilled authorized consumption (e.g., water used in public fountains or for firefighting, or water provided for free to certain users).


Chapter 5 – Water, Sanitation, and Communal Services in Kyrgyzstan

rehabilitation, renewal and expansion from internal sources (e.g., by raising tariffs) are clouded by legal, regulatory, and political uncertainties, and by outdated commercial and managerial practices—particularly regarding billing and contracting. The national policy dialogue group assessed the financing gap associated with maintaning existing water infrastructure at $37.5 million in 2009. •

Decentralization initiatives in Kyrgyzstan have transfered significant amounts of water and communal service infrastructure (along with responsibility for its maintenance and service delivery) to local governments. However, there is no national coordinating and regulatory body for communal service provision as a whole; nor are there comprehensive, integrated national water supply, sanitation and waste management policies. There is instead a lack of coordination between water sector policy and policies for related sectors (e.g., housing). Likewise, development policies for communal services sectors are not well integrated into the national annual and medium-term budgeting processes. Regarding customers’ satisfaction with, and willingness to pay for, communal services:

Survey results about willingness to pay more for better quality communal services are contradictory. Whereas research conducted by the Asian Development Bank and USAID indicate a strong willingness to pay for improved services, UNDP research suggests that a majority of poor customers (in both rural and urban households) are unwilling to pay more for communal services even if their quality improves. Likewise, some survey research indicates that much of

the population—including poor households—seems generally satisfied with the quality of communal services. Policy recommendations: •

Affordability analyses for communal services, with an emphasis on low income households, should be mandatory for tariff revision and development planning for communal service providers. Such analyses should underpin efforts to protect consumer rights and promote appropriate public participation in communal service delivery.

Given the low shares of household (including poor household) expenditures devoted to communal services, it is not clear that social protection measures to improve household access to these services are necessary. Emphasis should instead be placed on mobilizing resources—from donors, as well as from central and local governments—to reverse the decapitalization of water and communal service infrastructure, and to extend access to households that are not connected to this infrastructure.

A national policy framework for water and communal services, linked to health and environmental protection, and integrated water resources management, should be established. It should be supported by the creation of a national body for water, sanitation, and solid waste policy and investment planning, and for coordination with other relevant sectors—including the private sector and non-governmental organizations.

The legal responsibilities and division of labour between local governments and service providers should be clarified, as should the disposition of the debts inherited by local communal 205


Poverty and Social Impact Assessment

service providers from central government institutions; contracting and other legal instruments governing the use of municipal infrastructure and assets by service providers, as well as oversight and reporting relationships; and rights regarding ownership and management of local water infrastructure. •

International agencies can support these recommendations by:

o Helping to accelerate the transfer of expertise and mentoring of Kyrgyzstan’s experts on water sector

206

management, on affordability assessments, and the like;

o Supporting the implementation of the recommendations made by the national policy dialogue; and o Supporting research and analysis to develop disaggregated household survey data and qualitative survey results to capture the various dimensions of poverty, vulnerability and social exclusion, pertaining to water and communal services.


Chapter 5 – Water, Sanitation, and Communal Services in Kyrgyzstan

Sectoral overview Basic characteristics About one third of Kyrgyzstan’s 5.4 million citizens in live in urban areas; two thirds live in rural areas. The water, sanitation and solid waste management systems on which they rely face a series of technical, socio-economic, environmental, geographic, financial, and institutional challenges. Rugged, mountainous topography and many remote locations raise the costs of extending service infrastructure; while seismic activity and harsh climatic conditions drive up maintenance costs. While acces to improved water sources276 in rural areas has improved since 2004, almost half of the population remains without access to adequate drinking systems, and an even higher share is without access to adequate sanitation facilities. While access to drinking water, sanitation, and solid waste management services in Kyrgyzstan’s 25 cities is better than in rural areas, it also varies widely across (and within) cities. Most of Kyrgyzstan’s urban and rural water supply and sewage systems were constructed 40-50 years ago. At the time of their construction, they generally met the basic operating parameters of adequate water supply 276

As in many countries, there are discrepancies in Kyrgyzstan regarding access to drinking water and sanitation services in the data the national government, as opposed to estimates generated by the joint UNICEFWHO monitoring assessment and by other donors. These discrepancies sometimes reflect different definitions of what constitutes an “improved system”, as well as a lack of updated baseline data for rural and urban systems. Most figures refer to existing systems that are not necessarily fully functional. In addition, local governments and service providers do not always have good information on the numbers of houses and apartments in each city, or the residents therein. Numbers of people living in informal settlements around big cities are even more difficult to assess. These uncertainties affect estimates of service recipients.

systems: water delivered to the population in sufficient quantity (24-hour service, at adequate pressure) and quality (meeting basic health standards). However, because water was considered a right instead of a business, water supply and sewage systems were constructed without due regard for their financial sustainability. Over time, these systems became increasingly expensive to operate and maintain, because vodokanals277 lacked the necessary funds for adequate repair and maintenance. As a result, much of this infrastructure has undergone significant decapitalization— particularly in areas that have not been covered by the government’s Taza Suu programme (which is co-financed by the Asian Development Bank, World Bank, and other international agencies). At present, many of the 1,074 centralized piped water supply systems are not operating effectively. Official data indicate that the volume of water collected and distributed declined by some 8 percent during the 2007-2010 period—12 percent in per-capita terms. For households that do have access to piped water supplies, service levels, in terms of reliability, predictability and water quality is often poor. Kyrgyzstan’s Sanitary and Epidemiological Surveillance service reports that, on average, about 2 percent of chemical and 10 percent of microbiological water quality tests failed to reach national norms in 2007.278 277

The term “vodokanal” refers to a service provider responsible for potable water and/or sewage supplies. The KKP (republican communal enterprise) acronym refers to providers of solid waste collection and disposal services and possibly, but not necessarily, liquid waste collection and disposal services and/or city improvement services. KKPs include public utilities such as Tazaliks and, to a much lesser degree, private operators. 278 Cited in OECD, National Policy Dialogue on Financing Strategy for Urban and Rural Water Supply

207


Poverty and Social Impact Assessment

These challenges of inadequate access and poor service quality cannot be addressed within the financial parameters facing many service providers.279 Even the Bishkek vodokanal does not have sufficient funds to fully maintain the network. In fact, the financial viability of many service providers is put at risk by tariffs that are set below cost recovery levels, coupled with technical leakages and inadequate institutional capacity. Tariffs for water and communal services do not fully cover basic operational costs (e.g., salaries, energy, chemical usage), equipment upgrades and replacement (via a depreciation fund), not to mention the costs of infrastructure expansion.

infrastructure, and communal providers remain uncertain.

Prospects for covering the costs of infrastructure rehabilitation, renewal, and expansion from internal sources (e.g., by raising tariffs) are clouded by legal, regulatory, and political uncertainties, and by outdated commercial and managerial practices—particularly regarding billing and contracting. Since most household water consumption is not metered, non-revenue water levels cannot be measured accurately; nor can water bills be directly linked to actual consumption levels. The absence of water meters also precludes the use of lifeline tariffs (i.e., lower tariffs for small amounts of consumption, to keep basic services affordable for low-income households). The legal and institutional relationships between local governments (who are formally responsible for communal service provision),

Water service coverage. According to Bishkek and Osh vodokanal representatives interviewed for this research, 90 and 75 percent of these cities’ populations were provided with safe drinking water in 2010, respectively. (Areas of these cities not covered by the water infrastructure are new settlements, some of which have a quasiinformal character). A 2006 estimate by the Urban Institute280 placed that water coverage in Kyrgyzstan’s other 23 cities at between 60 and 90 percent, on the basis of census data, registered customers (as specified in residential customer service agreements), and the estimates of vodokanal staff. According to the Urban Institute, most vodokanal directors believe that customers underreport the numbers of users in their household, because invoices are usually based on a per capita usage norm (typically 170 liters/capita/day).

and Sanitation in the Kyrgyz Republic, ENV/EPOC/EAP/WATER (2009). 279 Opinions on the financial status, and quality of management, in the water and communal services sector differ widely. Representatives from the Antimonopoly Agency, the Kyrgyzhilkommunsoyuz communal service supplier, and the Community Development and Investment Agency (ARIS) interviewed for this research all agreed that management quality remains problemaric. But while ARIS specialists pointed to improvements over the last five years, Kyrgyzhilkommunsoyuz representatives argued that municipal service providers were nearly bankrupt. Antimonopoly agency specialists agreed that the final situtation was most dire in small cities.

Water consumption. Only 3-4 percent of Kyrgyzstan’s households had water meters in 2010 (less than 8 percent in Bishkek),281 which reduces the accuracy of water consumption estimates. Water consumption norms in Kyrgyzstan vary by location and

208

service

These issues are made more pressing by the fact that Kyrgyzstan is a low income country, in which official household survey data indicate that almost a third of the population lives below the national poverty line. Many citizens of the Kyrgyz Republic believe that they have a right to receive water at nominal cost, and view water supply and sewage as a government problem beyond their responsibility.

Drinking water in urban areas

280

Urban Institute, “Development of Performance Improvement Action Plan for Delivery of Local Infrastructure Services”, 2006. 281 According to information provided by Chinara Makeshevna, Antimonopoly Agency, 10 December 2010.


Chapter 5 – Water, Sanitation, and Communal Services in Kyrgyzstan

number of conveniences/cars/animals (i.e., indoor toilets, indoor bath, cows, etc.); they range from 30 liters/capita/day for a street connection to 60 liters/capita/day for house connection, with a maximum of 250 liters/capita/day for apartment blocks in Bishkek, and 210 liters/capita/day for apartment blocks in other cities.282 According to 2010 OECD data, real water consumption for household/domestic needs in Kyrgyzstan is 71.1 liters/capita/day; expert assessments of the average water consumption volume for rural and urban populations range from 50 to 125 liters/capita/day.283

Focus group discussions held in Osh indicate that major rainfalls result in service interruptions of 3-5 days due to mud accumulation in the Papan reservoir. Results of focus group discussions in Bishkek conducted for this research indicate that the installation of water meters allows households with less than four members to reduce their water bill by 25-50 percent. Some focus group participants who paid for the installation of water meters in Bishkek mentioned they recouped their investment (about 2000 som, or $44, per meter) within 912 months. However, information from vodokanal staff indicates that water consumption by households with more than four members typically exceeds the relevant norms, in part because households respond to the tariff structure (which links water consumption payments to household size) by under-reporting household size. Water availability. While vodokanals in Bishkek, Osh, and some other cities provide 282

These norms are for combined cold and hot water consumption. 283 OECD, Private Sector Participation in WSS in EECCA region: Status Paper, ENV/EPOC/EAP (2010). p.74.

24-hour service, others do not. These included (in 2010) Kok-Jongak (where water was provided for four hours a day), Kyzyl-Kia (7 hours), Kochkor-Ata (8 hours), Nookat (20 hours), Sulukta (12 hours), Uzgen (20 hours), and Isfana and Batken, where service was only provided intermittently. Even 24-hour systems have frequent service interruptions. For instance, participants in focus group discussions held in Osh report that major rainfalls result in service interruptions of 3-5 days due to mud accumulation in the Papan reservoir. Billing and non-revenue water. In developing countries, non-revenue water levels are often above 40 percent; in developed countries, they average about 16 percent.284 However, data are not always available, and what is available is not always reliable, as utilities have incentives to underreport these losses in order to conceal their own inefficiency.285 Assessments of nonrevenue water levels in Kyrgyzstan differ. According to a 2006 Urban Institute study and an ARIS 2010 update, estimated leakages in distribution networks are in the 35-45 percent range, as is typical in many developing countries. However, since most systems are not metered and the amount of water pumped into the system and distributed to consumers is not closely monitored, this estimate has an approximate character.286

284

N. Tynan and Kingdom, W, “A Scorecard for Water Utilities in Developing Countries,” online discussion, April 2002. 285 Bill Kingdom, Roland Liemberger, Philippe Marin, “The Challenge of Reducing Non-Revenue Water (NRW) in Developing Countries”, Water Supply and Sanitation Sector board Discussion Paper 8, Dec. 2006. 286 During the Soviet period utilities were required to keep non-revenue water levels below 18 percent. In other former Soviet republics, utilities continue to report such low levels, even when their actual losses are higher. Source: Frauendorfer, R. and R. Liemberger, The issues and challenges of reducing non-revenue water, Asian Development Bank, 2010.

209


Poverty and Social Impact Assessment

Table 5.1—Kyrgyzstan: Revenue/cost ratios for drinking water supply in smaller cities (January 2010)

City Balykchi Batken Cholpon Ata Isfana Jalalabad Kant Kara-Balta Water Co Kara-Balta Electric Power and Water Supply Industrial Association Kara-Kul (2006) Kara-Suu (2004) Kochkor-Ata

Revenue/cost ratio 1.05 0.5 1.02 0.32 1.07 1.50 0.57 1.00 0.33 0.60 1.03

City Kok-Jongak Kyzyl-Kia Mailuu-Suu (2004) Naryn Nookat Shopokov (2008) Sulukta Talas Tokmok (2004) Uzgen (2008) Unweighted average

Revenue/cost ratio 1.30 1.08 1.32 1.07 1.02 1.00 0.8 1.175 1.06 1.10 0.95

Source: ARIS 2010.

Interviews conducted in Bishkek and Osh indicate that illegal connections are not a significant problem in Kyrgyzstan’s cities. This suggests that non-revenue water levels reflect the absence of metering, as well as leaks in water lines that are not detected (most vodokanals do not have leak detection equipment or portable flow meters). Reducing non-revenue water levels would therefore require significant investments in metering (and not only of end users), as well as more accurate consumer service agreements and improved invoicing (via computerized accounting and billing systems). Vodokanals would also need to examine unbilled authorized consumption and consider reducing the number of entities entitled to receive water without paying for it. Interruptions in water supply, which are often caused by excessive leakage, tend to be a particular hardship on the urban poor, who often have to buy water from vendors. Reducing physical losses would also make more water available and enable utilities to improve poor communities’ access to water. Water quality and health. In many cities in Kyrgyzstan, water is piped from wells or springs of reasonable quality. But some cities (Kyzyl-Kia, Mailuu-Suu, Nookat, Shopokov, Sulukta) use solely river water, or use river water as a backup during the dry season. Many pipes are made of asbestos, which can pose health risks (according to the 210

World Health Organization).287 Some 20 percent of Kyrgyzstan’s water supply systems do not comply with sanitary norms; many lack sanitary protection zones, water treatment plants, and disinfection systems (SES, 2008). Water treatment, where it is practiced, is typically limited to chlorination with occasional water quality testing by the State Sanitary and Epidemiological Surveillance service. Many systems deliver untreated water much of the time, even in systems relying on river water as their primary source. Additionally, in water systems that deliver water intermittently, the potential for contamination is significant since contaminants can more easily leach into unpressurized pipes. Since there are few asbuilt drawings of the water system that show the location of other underground utilities (e.g., heating, sewage) and different entities have constructed water infrastructure ove