Social protection, poverty exit and household’s behaviour: a multidisciplinary perspective by Luigi Peter Ragno

Social protection, poverty exit and household’s behaviour: a multidisciplinary perspective Paper prepared for the Conference:

Luigi Peter Ragno Social Protection Consultant Doctor of Philosophy‐ Candidate‐ School of Environment and Development Brooks World Poverty Institute (BWPI) and Institute for Development Policy and Management (IDPM) The University of Manchester Oxford Road, Manchester. M13 9PL. UK Luigi.Ragno@postgrad.manchester.ac.uk www.bwpi.manchester.ac.uk

“Scaling‐Up Social Protection In Bangladesh: Building Effective Social Ladders And Safety Nets” Dhaka, Bangladesh 9 & 10 of October 2011

DO NOT QUOTE WITHOUT PREVIOUS PERMISSION BY THE AUTHOR Abstract: Social Protection strategies are being depicted as ladders, trampolines, ropes and springboards. Their role of preventing people from falling into poverty is now being expanded and directly linked to growth and poverty exit. Findings from several evaluations and studies suggest that social protection instruments can lift restrictions on the productive capacity and create new prospects for households living in poverty in terms of, for instance, participating to savings scheme, securing of credit, improving consumption, as well as improving households’ resource allocations towards higher return investments in the short, medium and long term (e.g. children’s education). In turn, this can contribute to micro level growth and poverty exit by generating growth outcomes such as accumulation and protection of assets, supply of labour, and several types of effects on the local economy. This paper suggests that the contribution of social protection interventions in lifting barriers to the production capacities of households is not sufficient in generating growth outcomes and poverty exit. The choices that households make in relation to grasping the new prospects that social protection contributes in opening are not certain; rather, these are dependent upon the household’s individual risk taking behaviour which is influenced by several factors, endogenous and exogenous to the households itself. Behavioural research can provide policy makers with insights on the household’s risk taking behaviour: in turn, this may be useful in designing, for instance, context or group specific interventions and defining complementary measures and incentives to maximize the impact of social protection policies on poverty reduction.

Contents 1

Introduction .................................................................................................................................... 2

Poverty Entry and Social Protection ............................................................................................... 4

Poverty Exit and Social Protection .................................................................................................. 5

Households Risk Taking Behaviour and Poverty Exit ...................................................................... 8

Policy Formulation and Household behaviour .............................................................................. 12

Conclusion ..................................................................................................................................... 13

Bibliography .................................................................................................................................. 15 1

Introduction

Social Protection strategies are being depicted as ladders (Kabeer, 2002, Kabeer, 2007), trampolines (Conway and Norton, 2002) ropes and springboards (World‐Bank, 2001) by several academics and researchers in leading and influential research organizations such as the Overseas Development Institute (ODI) and World Bank (WB), setting the ground for a new role for social protection in development policy. Their role of reducing poverty by ‘preventing entries’, by preventing people from falling into poverty (Lustig, 2001) is now being expanded to promoting household‐level growth and exits from poverty.

Definitions of social protection World Bank (WB) and Asian Development Bank (ADB) (WB) ‘public interventions (i) to assist individuals, households, and communities better manage risk, and (ii) to provide support to the critically poor’ (Holzmann and Jørgensen, 2000: 9).

(ADB) ‘the set of policies and programs designed to reduce poverty and vulnerability by promoting efficient labour markets, diminishing people’s exposure to risks, and enhancing their capacity to protect themselves against hazards and the interruption/loss of income’ (Baulch et al., 2008: 7); .. ‘The set of policies and programs that enable vulnerable groups to prevent, reduce, and/or cope with risks, and that: are targeted at the vulnerable groups; involve cash or in‐kind transfers; and are not activities that are usually associated with other sectors such as rural development, basic infrastructure, health, and education’ (Baulch et al., 2008: 13)

Overseas Development Institute (Conway), Institute of Development Studies (IDS) and International Labour Organization (ILO) (ODI) ‘Public actions taken in response to levels of vulnerability, risk and deprivation which are deemed socially unacceptable within a given polity or society’ (Conway et al., 2000: 2) (IDS) ‘All public and private initiatives that provide income or consumption transfers to the poor, protect the vulnerable against livelihood risks, and enhance the social status and rights of the marginalised; with the overall objective of reducing the economic and social vulnerability of poor, vulnerable and marginalised groups’ (Sabates‐Wheeler and Devereux, 2007).

(ILO) ‘As the set of public measures that a society provides for its members to protect them against economic and social distress that would be caused by the absence or a substantial reduction of income from work as a result of various contingencies (sickness, maternity, employment injury, unemployment, invalidity, old age, and death of the breadwinner); the provision of health care; and, the provision of benefits for families with children’ (ILO, 2000: 29).

This shift has been accompanied by a series of ‘conceptual frameworks, analytical tools, empirical evidence, national policy processes, heavyweight agencies and big names’ (Devereux and Sabates‐Wheeler, 2007: 1) which have contributed in changing perception on the role of social protection and its impact on growth and poverty exit. The main objective of this paper is to bring in the discourse on poverty exit and social protection, a multidisciplinary perspective that can contribute in the formulation of appropriate policies and strategies. This paper first presents an overview of social protection and poverty reduction, focusing on its ‘preventing entries’ role. The analysis then shifts on the existing literature on social protection and ‘poverty exit’ and its foundations in the economic theory. This is followed by introducing a multidisciplinary perspective that builds on insights from behavioural economics, cognitive psychology and cultural anthropology and reinterprets the poverty exit process. The final section of the paper looks at 2

how social protection policy formulation can incorporates these new insights and approaches to promote poverty exit. Social Protection is a common expression used by donor agencies, academics (and students), governments and NGOs which means many things to different people. There is in fact no consensus on what social protection is and several authors and organizations have adopted different conceptualizations. It is widely accepted however that social protection policies in developing countries have a strong focus on poverty and poverty reduction and are important strategies in addressing households’ vulnerability and favour growth1 (Arjona et al., 2002, Arjona et al., 2003, Barrientos and DeJong, 2006, Barrientos and Hulme, 2005, Barrientos et al., 2005, Devereux, 2002, United Nations. Economic and Social Commission for Asia and the Pacific., 2002, Voipio, 2007, Holmes et al., 2007, John Farrington, 2004, John Farrington, 2007, Frota, 2008, Barrientos and Hulme, 2008). In the definition of Social Protection presented in the box, there are some clear basic common principles: social protection policies are initiatives of governments or private actors (e.g. insurance providers and NGOs), which attempt to reduce poverty addressing households’ vulnerability and exposure to risks. These apparently similar definitions, which highlight the role of vulnerability and risk, do differ substantially when the ‘how’ and the processes of delivery are looked at in detail. The WB and ADB’s definition of social protection highlights that policies should aim at (WB) assist individuals, households, and communities better manage risk and (ADB) enhancing their capacity to protect themselves, emphasizing the central role of the individuals (or households) in managing risk and reduce vulnerability. In this conceptualization of social protection, public actions have the task of providing measures for the individuals to manage risk. This notion drastically differs from the ODI, IDS and ILO’s definitions where public actions are meant to directly address people’s risk and vulnerability: all public and private initiatives that provide income or consumption transfers to the poor, protect the vulnerable against livelihood risks (IDS).... and ....public measures that a society provides for its members to protect them against economic and social distress (ILO). Measures which fall under the umbrella of social protection can be grouped under 5 headings (Baulch et al., 2008: 14‐15 –adapted‐): Labour Market: Direct employment generation (microenterprise development and public works); Labour exchanges and other employment services; Skills development and training (only if targeted to vulnerable groups); labour legislation (including minimum age, wage levels, health and safety) Social Insurance: Programs to cover the risks associated with unemployment, sickness, maternity, disability, industrial injury, and old age (contributory schemes) 1

In developed countries, the emphasis of social protection is on income maintenance and on protecting living standards for all BARRIENTOS, A. 2011b. Social protection and poverty. International Journal of Social Welfare, 20, 240‐249..

Social Assistance (social transfers) and Welfare; Welfare and social services targeted at the disabled, poor elderly, the indigent, those affected by disasters, and other vulnerable groups; Cash/in‐kind transfers (food stamps, health cost exemptions, or subsidies); Temporary subsidies for utilities, housing, etc. Micro and Area‐Based Schemes: Micro insurance/microfinance schemes; Agricultural insurance; Disaster preparedness and management and Social funds Child Protection: Child rights and advocacy/awareness; Early child development activities; Educational assistance (school feeding, scholarships, fee waivers); Health assistance (health cost reduced fees/ subsidized medicines for vulnerable groups; Street‐children/child worker/orphan initiatives; Family allowances. In developing countries, social protection has a developmental role and is meant to contribute to poverty reduction in two ways: by preventing people from falling into poverty (Poverty Entry) and by supporting people and households already living in poverty in their ‘graduation’, lifting out (Poverty Exit).

Poverty Entry and Social Protection

A household, if exposed to a risk (e.g. potential loss of the shelter due to an incoming cyclone) or if experiencing a shock (the actual loss of the shelter) is obliged to respond and adopt ex ante or ex post strategies to deal with the potential and actual losses. Some of these risk management strategies might be detrimental to the household’s well being and might push the household under a minimum acceptable benchmark level of welfare (poverty). The responses vary from households to households and are mediated by its endogenous characteristics and other exogenous factors, such as number of active family members, asset ownership, and social networks. Examples of households’ responses to risks and shocks are very well documented and can lead to selling assets, taking children out of school, reduce the quantity and quality of food intake and child labour. Morduch (1994 :221) defines this type of poverty as ‘stochastic poverty’, a situation where households might fall into poverty because of the realization of a risk and the lack of appropriate risk management options. Social protection and stochastic poverty are clearly interlinked. Social protection measures do not deal with the risky event (Risk) per se but with the risk responses, the ex ante and ex post strategies individuals and households adopt to smooth consumption and avoid irreversible or temporary falls below an acceptable benchmark: e.g. pensions, or weather related insurances (risk mitigation), or public works and cash for work programs in the aftermath of a natural disaster (risk coping options). The lack of appropriate risk management options ‘may be intrinsically detrimental to the poor, and, one could also consider a measure of lack of access to consumption‐smoothing mechanisms just as deprivations in health and nutrition may be considered as part of an expanded poverty concept’ (Morduch, 1994: 224). In the Figure below, I try to visualize this process. 4

Let’s assume that a household starts off with a predefined level of welfare, above an acceptable minimum benchmark. The literature on stochastic poverty suggests that the realization of a risk (shock), such as the death of the income earner, can lead the households in a downward trajectory (X) below a benchmark (such as a consumption poverty line). The mobility occurred because the household did not have appropriate risk management instruments, such as a life insurance to, for instance, smooth consumption. From the new welfare level, the household can further fall below the benchmark, as shown in the trajectory D, maintain the same level of welfare trajectory C or improve, trajectory A and B. It is well accepted that social protection measures do have a role in avoiding a trajectory X, or, in case the household finds itself already below a benchmark, avoid trajectory D. The direct contribution of social protection is therefore reducing poverty by ‘preventing entries’ (CPRC, 2008), by diminishing the number of household falling into stochastic poverty. Instead, what role social protection instruments play in poverty exits2 trajectory A or B, in supporting the household ‐coming from trajectory X or households which have been living below a benchmark level for a long period of time as represented in trajectory Y‐ is very controversial and conceptualized differently in the literature.

Poverty Exit and Social Protection

Devereux (2002), while presenting findings from three Southern Africa countries, suggests that, even if often unintended, social protection measures promote exit from poverty by triggering ‘asset creation (by the project) and investment behaviour (by project beneficiaries or participants)’ (Devereux, 2002: 662). Social protection, by addressing and managing uninsured risks can support upward trajectories (A and B) by promoting more entrepreneurial decisions (Binswanger, 1980, Binswanger and Rosenzweig, 1989), ‘encouraging moderate risk taking behaviour by smoothing income stream against adverse entrepreneurial outcomes’ (Devereux, 2002: 665; see also Morduch, 1999, Morduch, 1995); by taking more risks and subsequently having higher returns (Holzmann and Jørgensen, 2000). Evidence on how social protection measures may promote poverty exits, exists but is far from being systematically discussed in the literature. Barrientos (2008; 2011) has attempted to address this by developing a basic framework on how social protection, and in particular social transfers, may contribute to micro level growth and promote exits in households living in poverty.

Lifting out of poverty, graduation from poverty, promotion, among others, are often used as a synonymous of poverty exits.

Findings from evaluations and studies in several countries, suggest that social protection instruments can contribute to positive growth outcomes in relation to a) accumulation and protection of assets, to b) supply of labour, and to c) several types of local economy effects (Barrientos and Scott, 2008, Barrientos, 2011a). Asset Accumulation and Protection: Gertler (Gertler et al., 2006) has provided evidence that beneficiaries of the Oportunidades program in Mexico, a part form increasing consumption directly from the transfers, have been able to generate additional income by increasing their investments and participation in farm assets, agricultural activities and micro‐enterprises. Accumulation of human capital (e.g. education and health) have also been documented in programs not directly targeting human capital accumulation, such as social pensions schemes in Brazil and South Africa. In addition, evidence from Nicaragua’s Red de Proteccion Social (Maluccio, 2005) and Ethiopia’s Productive Safety Net Program (Sabates‐Wheeler and Devereux, 2010), suggest that beneficiaries were better capable to protect their assets. In Nicaragua, beneficiaries were able to protect their productive asset and human capital better than the non beneficiaries during a sudden drop of coffee prices. Similarly in Ethiopia, households targeted by the program were more likely to successfully protect their income generating activities in the 2008 food crisis. Supply of Labour: The review of evidence (Barrientos, 2011a, Barrientos and Scott, 2008) of the effects of social protection on the supply of labour suggest that labour supply of children and elderly decreases while some evidence suggest that this is compensated by reallocation of labour supply among the working age members of the households. Samson (2009), in his review on the linkages between employment and social transfers, argues against the belief the social transfer functions as a labour disincentive. He discusses several studies which suggest that social transfers may have positive impacts in labour supply by providing additional resources for funding job search, by increasing labour market participation and by contributing to build human capital, which is essential for long term employment impacts (Samson, 2009). Local Economy Effects: Local economy impacts are instead well documented. For instance, cash for work programs, while transferring cash to the workers, build community infrastructures, like roads or embankments which have a direct impact on markets, by facilitating the movements of goods, transportation services and so forth. Social transfers particularly can also have a direct impact of increasing the demand for goods and services such as food, productive assets, transportation and labour. Finally, evidence exists that even the non beneficiary households of areas where a program is being implemented, benefit by an increased number of loans and gifts from the beneficiary households as well as by ‘demonstration effects’ in production techniques, schooling and health services utilization (Barrientos and Scott, 2008, Barrientos, 2011a, Barrientos and Sabatés‐Wheeler, 2010, Barrientos and Sabates‐Wheeler, 2009). But how can social protection interventions contribute in triggering these productive outcomes? Social protection instruments provide additional and predictable liquidity as well as insurance features to households; this contributes in lifting restrictions on the productive capacity that households living in poverty face. By lifting these barriers and unlocking the productive capacities of the households, social protection opens up new prospects and opportunities: some examples of new prospects are the participation in 6

saving scheme, securing of credit, improving consumption, as well as improving households resource allocations towards higher return investments in the short, medium and long term (e.g. children’s education) (Barrientos and Scott, 2008, Barrientos, 2011a). However, it is important to highlight that the lifting of, for instance, liquidity barriers per se does not guarantee the participation of households to saving scheme, to securing of credit, and more in general, to higher return investments. The household in fact still needs to choose and decide upon the seizing of these new prospects that social protection has contributed in opening. Therefore the questions here are: will the beneficiary households grasp all the opportunities brought by the access to social protection instruments? Will households respond by seizing the new prospects and choosing higher return and riskier decisions such as starting saving and taking loans? Traditional economic theory suggests that rational actors make choices to maximize their expected utility: Holzmann (2000:7), referencing Ravallion’s 1997 study on famines and Morduch’s (1995; 1999) extensive work on insurances and consumption smoothing, suggests that the threat of destitution, income variability and closeness to a ‘survival line’, influence the linear relation between choice and maximization of expected utility, and causes the rational actor to act ‘irrationally’, outside the economic axiom. The Expected Utility (EU) is the standard hypothesis in the neoclassical economic theory3, and suggests that in the context of an economic decision, when agents are confronted with several potential outcomes (decision under risk and uncertainty), the agent chooses the outcome which maximizes its utility. The choice selection is the result of a weighting process of the mathematical probabilities of the possible outcomes which maximize the agent’s expected utility. Considering that the shape of the EU function can be concave, convex and linear, agents do not choose to maximize their expected utility at the same extent as in some cases the shape shows an increasing, constant or decreasing marginal utility. These changes are explained in the EU function by the degree of risk aversion which actors have: actors can be risk averse (concave shape with decreasing marginal utility) if they prefer certain to riskier but with higher utility returns outcomes; risk seeking (or prone with a convex shape and increasing marginal utility) if they tend to choose for riskier outcomes and risk neutral, linear shape of the function with constant marginal utility. The degree of risk aversion strictly interrelated to the threat of destitution, income variability and closeness to a ‘survival line’: This threat of destitution and non‐survival renders the poor very risk adverse and as a result makes them very reluctant to engage in higher risk/higher return activities (Holzmann and Jørgensen, 2000: 9). 3

Several other Nonexpected‐Utility hypothesis and theories have been developed (see CAMERER, C., LOEWENSTEIN, G. & RABIN, M. 2004. Advances in behavioral economics, New York Princeton, N.J., Russell Sage Foundation ; Princeton University Press.) however the EU hypothesis remains central to the field of economics.

This is supported by several studies mostly based on the analysis of field data from south India – ICRISAT‐ (Binswanger, 1980, Binswanger and Rosenzweig, 1989, Morduch, 1995) which have concluded that households whose consumption levels are most vulnerable to income shocks devote a greater share of land to safer, traditional varieties of rice and castor than to riskier, high‐yielding varieties (Morduch, 1995: 110); and more in general, Morduch suggests that risk averse households show reluctance in adopting new technologies and taking advantage of new economic opportunities (Morduch, 1995: 104). In line with this discourse, traditional economics suggests the access to social protection instruments allows individuals and households to reduce income variability, to smooth income and consumption (Morduch, 1995), to reduce the threat of destitution and non survival, and as a consequence, influences positively the degree of risk aversion. It increases the agent’s attitude and behaviour to take riskier choices. It promotes the re‐establishment of the positive linear relation between choices and marginal wealth utility maximization, encouraging households to take advantage of the opportunities created by the access to social protection instruments, such as securing credit and investing in higher return strategies. This, in turn, will contribute in achieving productive outcomes, micro level growth and exiting poverty. While the literature on poverty traps provides important insights on non‐linearities in the productive outcomes of social protection (Barrientos, 2011a, Barrientos and Scott, 2008; Carter and Barrett, 2007), the next section introduces a multidisciplinary perspective that challenges the traditional economics’ assumption presented here. This perspective, which builds on insights from behavioural economics, cognitive psychology and cultural anthropology, reinterprets the poverty exit process of grasping new productive opportunities brought in by social protection, as strictly dependent on a context‐bounded household’s risk taking behaviour.

Households Risk Taking Behaviour and Poverty Exit Risk judgements can never be neutral nor individualistic but rather are always shaped through shared understandings and anxieties about phenomena (Tulloch and Lupton, 2005: 7).

The decision makers and risk takers are rational actors as well as part of a social, political and psychological construct which defines their vision and perception of events as well as decisions and actions to be taken. In standard economics, there is a clear tendency to ‘tackle risk using a rigorous conceptualization of rational actions’ placing ‘stringent requirements on people’s capacity to process information and estimate probabilities (Taylor‐Gooby and Zinn, 2006: 8), abstracted from social and cultural settings and influences.

Risk taking, as a component of a household decision making, can be fully understood only within a framework where the perception and acceptability of risk and its linkages with context specific endogenous and exogenous factors play a role. Risk taking behaviour is the response to the acceptability of a perceived risk: this implies that decisions are taken and action made based upon a process that defines the tolerability of risk as a Behavioural and Experimental Economics consequence of being perceived as such (Yates, Behavioural Economics 1994).

Behavioural economics is a field in the economics discipline based on the notion that ‘increasing the realism of the psychological underpinnings of economic analysis will improve economics on its own terms generating theoretical insights, making better predictions of field phenomena, and suggesting better policy’. (Camerer et al., 2004: 7)

Behavioural and Experimental Economics: two separate fields in economics

Behavioural economics is defined not on the bases of the research methods adopted, but on the incorporation of psychological concepts in economics. Experimental economics instead is defined by the use of laboratory and no laboratory field experiments as research methods and attempt to test existing economic models rather than producing new theories. Differently from behavioural economics which adopts which widely adopting research methods from the field of psychology (including experiments), Experimental economists have developed new rigours methods suitable for analyzing economic phenomena. (Camerer et al., 2004:8)

The basic assumption of the neoclassical economic theory, that self‐interest and rationality govern people’s choices, has been criticized by what are now known as behavioural and experimental economics. Though behavioural and experimental economics are considered to be two separate sub‐fields in the economics disciplines, the innovations, insights, and use of observed empirical evidence brought in, have challenged fundamental hypothesis and assumptions in economics. It is widely accepted that both behavioural and psychological elements were present in earlier work, even in Adam Smith’s less known ‘Theory of Moral Sentiments’; but the sophistication and systematization brought in the second half of the twentieth has allowed swiping conceptual changes.

As well as traditional economists, the goal of behavioural economists is the prediction of phenomena and economic decisions to design better and more appropriate policies. However they highlight the need of having more realistic assumptions which reflects the psychological underpinning of economic models. One of the key concepts in the field of behavioural economics in relation to decision making under uncertainty, is the reinterpretation of risk as a not purely mathematically quantifiable probability, but as a construct of individual human behaviour. To the mathematically quantifiable probability of risk, behavioural economists replace the probability judgement (Camerer et al., 2004: 9), a mental process that includes perception, emotions, experience and memory of previous events. In the probability judgement, individuals ‘edit’ (Camerer et al., 2004: 127) the outcomes/prospects which are re‐coded as loss or gains in relation to a reference point. As well as the prospects, the available choices go through a process of classification and are filtered by what Kahneman names

the ‘framing effect’ and ‘context effect’ (Camerer et al., 2004: 12). These processes define people’s preferences, behaviour and choice’s selection. Behavioural economists suggest that the way a particular choice is framed and presented, influences the preference that is after revealed. A common example made by Tversky and Kahneman is the Asian disease (Tversky and Kahneman, 1981). In this example, a group of people were asked to choose between two options 1 and 2 in the first frame and 3 and 4 in the second frame. In the first frame of the problem, when people are confronted by two choices of saving 200 lives for sure (1) or 33% chances of saving all and 66% of saving none (2), most people select option 1 over 2. In the second frame, people are requested to choose between (3) 400 people dying for sure or 66% chances of saving none and 33% of saving all (4). Considering that choice 1 and 3 are the same (same effect), the majority of people will choose 4 over 3. The context of available choices also shapes the preferences and behaviour: in opposition with the independence‐cancellation axiom of the EU hypothesis, choices are selected based on the alternative choices available. Therefore, preferences are ill‐defined, highly malleable and dependent on the context in which they are elicited. Nevertheless, when required to make an economic decisions—to choose a brand of toothpaste, a car, a job, or how to invest— people do make some kind of decision. Behavioural economists refer to the process by which people make choices with ill‐defined preferences as ‘constructing preferences’ (Slovic, 1995; Camerer et al., 2004: 15, Payne et al., 1992). In line with behavioural economists’ discourse, Yates represents the risk taking cycle as six stages process (Yates, 1994: 328‐329), in which risk perception is at its genesis. Important contributions to this topic where given by Slovic, Taylor‐Gooby, Yates, Chicken and others (Chicken and Posner, 1998, Dake, 1991, Douglas, 1986, Douglas, 1992, Furedi, 1997, Lupton, 1999, Slovic, 2000, Taylor‐Gooby and Zinn, 2006, Tulloch and Lupton, 2005, Yates, 1994), mostly for their multidisciplinary approach and their attempt to identify bridges and common grounds among disciplines in understanding risk and behaviours. As Taylor (Taylor‐Gooby and Zinn, 2006: 55) suggests Recent work demonstrates the complexity of the risky choices that people make in their everyday lives. Issues of social identity, and group membership are bound up with risk taking and risk avoidance. Some people actively seek out some risks which others just as strenuously avoid. It becomes clear that risk can not be understood simply in terms or rational judgement. Emotional factors supply the drive necessary to make choices... The current debate has attempted to identify important factors, defined by Dake (Dake, 1991) as ‘orienting mechanisms’, that influence risk taking behaviours, in particular related to risk perception, its acceptability and ultimately the response to it. Some of these factors have been reported here, however these are not conclusive and stand‐alone ‘slots’. These factors are not explanatory variables; in other words the factor themselves are the construct of sets of beliefs and other interrelated sub‐factors. Gender: As Slovic reports (Slovic, 2000: 396), several studies have suggested that the gender of the individual assessing risk, influences the outcome, as for instance, males usually downsize risk in 10

comparison to women. Gender has to be understood not as simply biological (physically more vulnerable) differences but as the social construct which influences human development. For instance, where women are less exposed to education and labour markets, might not have the required knowledge to properly assess risk. Worldviews (Dake, 1991): This concept can be understood as the ‘dispositions’, similar to Tulloch’s cultural disposition (Tulloch and Lupton, 2005: 6), the general attitudes in the ‘world and social organizations’ (Slovic, 2000: XXXiii and 405), which influences individuals in their perception and response to risk. Example of Dake’s dispositions are egalitarianism, fatalism, hierarchy and so forth (Slovic, 2000: 402). Emotions/experiences: All perceptions, such as visual images, are coloured with emotions: ‘we do not see a house; we see a handsome house, or an ugly house’ (Slovic, 2000: 403). Emotions and moods (Yates, 1994: 203‐207) result from a situation an individual has experienced in the past and create a bias in the risk assessment process. For instance, having been targeted by a thief in the past, will influence the individual to perceive a higher risk of being robbed again, even if the number of theft have remained constant. In this case, the individual which has experienced the theft will perceive risk differently than other individuals which have not experienced, and respond (risk taking) differently and more conservatively. Group risk perception: When a risk response is taken within a group, such as a household, the individual perception is mediated to produce collectively a single action, which might entail sub‐ actions (Yates, 1994: 164). Collective models of intra‐households decision and allocation (Haddad et al., 1997, Haddad et al., 1994, Rogers et al., 1990) tend to suggest that individuals, with different perceptions and preferences are bounded by social and economical contracts which drive them to take collective decisions. Within the group, several factors influence the role of the individual in the final decision: ‘chronological age’, as defined by Taylor (Taylor‐Gooby and Zinn, 2006: 180) to incorporate both the biological age and the experience and wisdom of an old person, is an important variable which can strongly influence the final choice, mostly in terms of risk aversion. Inequality: While individuals and groups still face the same risks (poverty, exclusion etc), the way risk is perceived has changed (Taylor‐Gooby and Zinn, 2006: 243). Spatial inequalities, in terms of where the individual live and the access to services, also shape the perception of risk (Tulloch and Lupton, 2005). Knowledge and information: Knowledge is the ‘sum of what we know, both theoretically and in practice’ (Chicken and Posner, 1998: 9); its distribution, as continuous process of information flow is uneven and asymmetric. Knowledge is continuously filtered by media4 (Taylor‐Gooby and Zinn, 2006: 59) and can, in line with Foucault and Gramsi’s discourse (Coben, 1998, Mayo and Lange, 2001, Burchell et al., 1991), act a as tool to control people and their decisions.

See also the Social Amplification Framework in SLOVIC, P. 2000. The perception of risk, London, Earthscan Pub.

When applying these insights to the poverty exit process, the causal link among social protection instruments, the lifting of restriction in the productive capacities of households, and the agent’s attitude and behaviour to take riskier choices and to take advantage of the new prospects brought in, is challenged. This multidisciplinary perspective disputes the notion that social protection, by addressing and managing uninsured risks, guarantees more entrepreneurial decisions and encourages moderate risk taking behaviour. Instead, this perspective suggests that the choices that households take in relation to grasping these new prospects that social protection has contributed in opening (e.g. participation of households to saving scheme, to securing of credit, and more in general to higher return investments), are dependent upon each household’s risk taking behaviour, which in turn, is influenced by a complex interaction of factors endogenous and exogenous to the households itself, rather than being the certain effect of reduced income variability and consumption smoothing. In this view, the contribution of social protection intervention in lifting barriers to the production capacities of the households is not sufficient in generating productive outcomes and poverty exit. Understanding and defining households risk taking behaviour and the factors that influence it, becomes an important aspect of designing social protection interventions with ‘ladder’ outcomes. If policy makers fail to do so, social protection strategies might not contribute in supporting households in exiting poverty.

Policy Formulation and Household behaviour

The key issue now is how social protection policy formulation can incorporate behavioural insights of the targeted households to favour poverty exit. In other words, what are the essential elements and incentives that might influence households to grasp the opportunities brought in by social protection measures. ‘Behavioural research’ approaches and tools might contribute to integrate risk taking behavioural concerns of the targeted households within the policy formulation framework. However, it is important to say that this section only introduces basic concepts of behavioural research and does not attempt to identify or develop specific approaches to be used in policy formulation. Additional applied research needs to be done to develop tools that can be applied to the formulation of social protection policies. Behaviour as defined by Rosenberg (Rosenberg and Daly, 1993: 66), is the output of a complex dynamic system that is shaped and nurtured by past experiences, current condition and heredity. Like an elegant goulash that is the end product of a complex recipe, behaviour is determined by the blending and processing of many ingredients. Behavioural research is an umbrella concept (Rosenthal and Rosnow, 1991: 6), that refers to several disciplines’ field of inquiries. The behaviour of early primitive humans, humans as political animals, economic animals, social animals, talking animals, humans as logicians‐ these are heuristic expressions which serve to indicate or point out some particular aspects or relationship concerning human nature. Rosenthal also adds that such relationships and aspects are the concerns of psychologists, communication and educational researchers, sociologists, economists (mostly 12

behavioural and experimentalists), psycholinguistic, behavioural biologists and even some statisticians (:6). In behavioural science and behavioural research, the empirical research strategies can be broadly classified into two main categories: experimental and non experimental approaches. Experimental methods (Rosenthal and Rosnow, 1991: 71) involve direct manipulation and control of variables. The researcher manipulates the first variable of interest and observes the response (Cozby, 2009: 82). Experimental research has been applied to several disciplines from psychology to economics, defining a paradigm shift and the birth of sub fields of experimental and behavioural economics. Experimental economics for instance attempts to examine basic system which triggers and define economic choices and to identify which features of naturally occurring economic contexts are likely to impact economic decisions and outcomes (Weber and Camerer, 2006: 188). The field of experimental research relies on ‘experiments’ to test models and behaviours of actors: experiments in economics have been broadly classified in laboratory and field experiments (Harrison and List, 2004: 1013). In the experimental approach to behavioural research, the laboratory is seen as a ‘wind tunnel’ (Royal Swedish Academy, 2002: 9) where researchers have the opportunity to investigate human behaviour creating a simplified setting which mirrors the environments of economic interaction. In a non‐experimental approach, behaviours and relationship are studied by observing the actors in their natural environment. This can be done by analyzing public records and other secondary information, by direct observation, by asking individuals to describe and recall their behaviours as well as through more advanced psychological tests (Cozby, 2009: 81). Non experimental methods use observation, in a broad sense, to describe behaviours rather than identifying causes‐effect relationship. By describing behaviours and relationships, non experimental methods can however ‘relate’ behaviours to variables and factors as well as propose hypothesis about correlations as relational principles (Rosenthal and Rosnow, 1991: 10). Behavioural research can offer policy makers with insights on risk taking behaviour of households with access to social protection interventions. These insights can be important in designing, for instance context or group specific interventions and defining complementary measures and incentives to maximize the impact of social protection policies on poverty reduction.

Conclusion

In developing countries, social protection has a clear developmental role and can contribute to reduce poverty in two ways: by preventing people from falling into poverty (Poverty Entry) and by supporting people and households already living in poverty in their ‘graduation’, lifting out (Poverty Exit). 13

While the direct contribution of social protection in preventing poverty entries is well accepted, a growing body of evidence exists on its impact in facilitating exiting from poverty. Findings from several evaluations and studies suggest that social protection instruments can lift restrictions on the productive capacity that households living in poverty face in, for instance, participating to saving scheme, securing of credit, improving consumption, as well as improving households resource allocations towards higher return investments in the short, medium and long term (e.g. children’s education). This, in turn, can contribute to micro level growth and poverty exit by facilitating the accumulation and protection of assets, supply of labour, and several types of local economy effects. However, this paper moves away from the standpoint of standard economics, which suggests that the access to social protection instruments, by reducing income variability, the threat of destitution and non survival and by smoothing consumption, increases the agent’s attitude and behaviour to take riskier choices and, in turn, contributing in achieving productive outcomes, micro level growth and exiting poverty. Instead, this paper presents a multidisciplinary perspective which proposes that the behaviour and choices of the households in grasping the new opportunities brought in by social protection are not certain and do not fit the traditional hypothesis of utility maximization. As an alternative, the paper presents insights from other disciplines (behavioural economics, cultural anthropology and cognitive psychology) which indicate that households’ risk taking behaviour is dependent upon a complex interaction of factors, endogenous and exogenous to the households itself. In this view, the contribution of social protection intervention in lifting barriers to the production capacities of households living in poverty is not sufficient in generating productive outcomes and poverty exit. Approaches and techniques from the field of behavioural research may be useful in understanding the risk taking behaviour of households targeted by social protection instruments. Behavioural research can provide policy makers with insights on the risk taking behaviour which, in turn, may be useful in designing, for instance, context or group specific interventions and defining complementary measures and incentives to maximize the impact of social protection policies on poverty reduction. 14

Bibliography

ARJONA, R., LADAIQUE, M. & PEARSON, M. 2002. SOCIAL PROTECTION AND GROWTH. OECD Economic Studies, 2002, 7‐45. ARJONA, R., LADAIQUE, M. & PEARSON, M. 2003. Growth, Inequality and Social Protection. Canadian Public Policy, 29, S119. BARRIENTOS, A. 2011a. (Forthcoming) Social Transfers and Growth: What do we know? What do we need to find out? World Development. BARRIENTOS, A. 2011b. Social protection and poverty. International Journal of Social Welfare, 20, 240‐249. BARRIENTOS, A. & DEJONG, J. 2006. Reducing child poverty with cash transfers: A sure thing? Development Policy Review, 24, 537‐552. BARRIENTOS, A. & HULME, D. 2005. Chronic Poverty and Social Protection: Introduction. European Journal of Development Research, 17, 1‐7. BARRIENTOS, A. & HULME, D. 2008. Social protection for the poor and poorest : concepts, policies and politics, Basingstoke, Palgrave Macmillan; University of Manchester. Brooks World Poverty Institute; Chronic Poverty Research Centre. BARRIENTOS, A., HULME, D. & SHEPHERD, A. 2005. Can Social Protection Tackle Chronic Poverty? European Journal of Development Research, 17, 8‐23. BARRIENTOS, A. & SABATES‐WHEELER, R. 2009. Do transfers generate local economy effects? BWPI Working Paper 106. BARRIENTOS, A. & SABATÉS‐WHEELER, R. 2010. Strategic complementarities and social transfers: how do PROGRESA payments impact nonbeneficiaries? Applied Economics, 43, 3175‐3185. BARRIENTOS, A. & SCOTT, J. 2008. Social Transfers and Growth: A Review. Brooks World Poverty Institute, University of Manchester Working Paper 52. BAULCH, B., WEBER, A. & WOOD, J. 2008. Social protection index for committed poverty reduction. Volume 2: Asia, [Manila], Asian Development Bank. BINSWANGER, H. P. 1980. Attitudes toward Risk: Experimental Measurement In Rural India. American Agricultural Economics Association. BINSWANGER, H. P. & ROSENZWEIG, M. R. 1989. Wealth, Weather Risk And The Composition And Profitability Of Agricultural Investments. ECONOMIC DEVELOPMENT CENTER, Department of Economics, Minneapolis, Department of Agricultural and Applied Economics, St. Paul, UNIVERSITY OF MINNESOTA. BURCHELL, G., GORDON, C. & MILLER, P. M. 1991. The Foucault effect : studies in governmentality: with two lectures by and an interview with Michel Foucault, London, Harvester Wheatsheaf. CAMERER, C., LOEWENSTEIN, G. & RABIN, M. 2004. Advances in behavioral economics, New York Princeton, N.J., Russell Sage Foundation ; Princeton University Press. CARTER, M., R. & BARRETT, C. 2007. Asset Thresholds and Social Protection: A Reply to Dercon. IDS Bulletin, 38, 43‐44. CHICKEN, J. C. & POSNER, T. 1998. The philosophy of risk, London, Thomas Telford. COBEN, D. 1998. Radical heroes : Gramsci, Freire, and the politics of adult education, New York, Garland Pub. CONWAY, T. & NORTON, A. 2002. Nets, Ropes, Ladders and Trampolines: The Place of Social Protection within Current Debates on Poverty Reduction. Development Policy Review, 20, 533‐540. CONWAY, T., NORTON, A. & DE HAAN, A. 2000. Social Protection: New Directions of Donor Agencies. Social Development Department. London, UK: Department for International Development. COZBY, P. C. 2009. Methods in behavioral research, Boston, McGraw‐Hill Higher Education. CPRC 2008. Chronic Poverty Report 2008‐2009: Escaping Poverty Traps, Chronic Poverty Research Centre.

DAKE, K. 1991. Orienting dispositions in the perception of risk: An analysis of contemporary worldviews and cultural biases. Journal of Cross‐Cultural Psychology, 22, 61‐82. DEVEREUX, S. 2002. Can Social Safety Nets Reduce Chronic Poverty? Development Policy Review, 20, 657‐675. DEVEREUX, S. & SABATES‐WHEELER, R. 2007. Editorial Introduction: Debating Social Protection. IDS Bulletin, 38, 1‐7. DOUGLAS, M. 1986. Risk acceptability according to the social sciences, London, Routledge & Kegan Paul. DOUGLAS, M. 1992. Risk and blame : essays in cultural theory, London, Routledge. FROTA, L. 2008. Securing decent work and living conditions in low‐income urban settlements by linking social protection and local development: A review of case studies. Habitat International, 32, 203‐222. FUREDI, F. 1997. Culture of fear : risk‐taking and the morality of low expectation, London, Cassell. GERTLER, P., MARTINEZ, S. & RUBIO‐CODINA, M. 2006. Investing Cash Transfers to Raise Long‐Term Living Standards. World Bank Policy Research Working Paper 3994, August 2006. HADDAD, L. J., HODDINOTT, J., ALDERMAN, H. & INTERNATIONAL FOOD POLICY RESEARCH, I. 1997. Intrahousehold resource allocation in developing countries : models, methods, and policy, Baltimore; London, Johns Hopkins University Press. HADDAD, L. J., HODDINOTT, J., ALDERMAN, H., WORLD BANK. POLICY RESEARCH DEPT, P. & HUMAN RESOURCES, D. 1994. Intrahousehold resource allocation : an overview, Washington, D.C., World Bank Policy Research Department Poverty and Human Resources Division. HARRISON, G. W. & LIST, J. A. 2004. Field Experiments. Journal of Economic Literature, 42, 1009‐ 1055. HOLMES, R., FARRINGTON, J. & SLATER, R. 2007. Social Protection and Growth: The Case of Agriculture. IDS Bulletin, 38, 95‐100. HOLZMANN, R. & JØRGENSEN, S. 2000. Social Risk Management: A new conceptual framework for Social Protection, and beyond. Social Protection Discussion Paper No. 0006. The World Bank. ILO 2000. World labour report 2000 : income security and social protection in a changing world, London, International Labour Office. JOHN FARRINGTON, R. S. A. R. H. 2004. Social protection and pro‐poor agricultural growth: what scope for synergies? ODI Natural Resource perspectives, Number 91, . JOHN FARRINGTON, R. S. A. R. H. 2007. Linking social protection and the productive sectors. ODI Briefing Paper, 28. KABEER, N. 2002. Safety Nets and Opportunity Ladders: Addressing Vulnerability and Enhancing Productivity in South Asia. Development Policy Review, 20, 589‐614. KABEER, N. 2007. A 'Vision' Thing? Debate and Difference within the OECD‐DAC Poverty Network Approaches to Social Protection. IDS Bulletin, 38, 51‐53. LUPTON, D. 1999. Risk and sociocultural theory : new directions and perspectives, Cambridge, Cambridge University Press. LUSTIG, N. 2001. Shielding the poor : social protection in the developing world, Washington, D.C., Brookings Institution Press. MALUCCIO, J. A. 2005. Coping with the “Coffee Crisis” in Central America: The Role of the Nicaraguan Red de Protección Social. International Food Policy Research Institute, FCND Discussion Paper 188 MAYO, P. & LANGE, E. 2001. Gramsci, Freire and adult education: possibilities for transformative action. International review of education, 47, 395‐396. MORDUCH, J. 1994. Poverty and vulnerability. The American Economic Review ( Nashville ), 84, 221 (5 pages). MORDUCH, J. 1995. Income Smoothing and Consumption Smoothing. The Journal of Economic Perspectives, 9, 103‐114.

MORDUCH, J. 1999. Between the State and the Market: Can Informal Insurance Patch the Safety Net? . The World Bank Research Observer, 14. PAYNE, J. W., BETTMAN, J. R. & JOHNSON, E. J. 1992. Behavioral decision research: A constructive processing perspective. Annual Review of Psychology, 43. ROGERS, B. L., SCHLOSSMAN, N. P. & WORKSHOP ON METHODS OF MEASURING INTRA‐HOUSEHOLD RESOURCE, A. 1990. Intrahousehold resource allocation : issues and methods for development policy and planning, Tokyo, United Nations University. ROSENBERG, K. M. & DALY, H. B. 1993. Foundations of behavioral research : a basic question approach, Fort Worth, Tex. ; London, Harcourt Brace Jovanovich College Publishers. ROSENTHAL, R. & ROSNOW, R. L. 1991. Essentials of behavioral research : methods and data analysis, New York ; London, McGraw‐Hill. ROYAL SWEDISH ACADEMY 2002. Foundations of Behavioral and Experimental Economics: Daniel Kahneman and Vernon Smith. Advanced information on the Prize in Economic Sciences. SABATES‐WHEELER, R. & DEVEREUX, S. 2007. Social Protection for Transformation. IDS Bulletin, 38, 23‐28. SABATES‐WHEELER, R. & DEVEREUX, S. 2010. Cash transfers and high food prices: Explaining outcomes on Ethiopia's Productive Safety Net Programme. Food Policy, 35, 274‐285. SAMSON, M. 2009. Social Cash Transfers and Employment A note on empirical linkages in developing countries. In: OECD (ed.) Promoting Pro‐Poor Growth: Employment. Paris: Organisation For Economic Co‐Operation And Development. SLOVIC, P. 1995. The construction of preferences. American Psychologist, 50. SLOVIC, P. 2000. The perception of risk, London, Earthscan Pub. TAYLOR‐GOOBY, P. & ZINN, J. 2006. Risk in social science, Oxford ; New York, Oxford University Press. TULLOCH, J. & LUPTON, D. 2005. Risk and everyday life, London, Sage. TVERSKY, A. & KAHNEMAN, D. 1981. THE FRAMING OF DECISIONS AND THE PSYCHOLOGY OF CHOICE. SCIENCE, 211, 453‐458. UNITED NATIONS. ECONOMIC AND SOCIAL COMMISSION FOR ASIA AND THE PACIFIC. 2002. Reducing poverty and promoting social protection, New York, United Nations. VOIPIO, T. 2007. Social Protection for Poverty Reduction: The OECD/DAC/POVNET View. IDS Bulletin, 38, 45‐50. WEBER, R. A. & CAMERER, C. F. 2006. "Behavioral experiments" in economics. Experimental Economics, 9, 187‐192. WORLD‐BANK 2001. Social protection sector strategy : from safety net to springboard, Washington, DC, World Bank. YATES, J. F. 1994. Risk‐taking behaviour, Wiley Series in Human Performance & Cognition.