Box II.1. Statistical Capacity Indicator

Across these four broad categories, the Middle East and North Africa (MENA) region faces longstanding issues that accentuate the difficulties in forecasting. These include lack of information due to data opacity, conflict, and exposure to commodity price volatility. The latter two can be characterized together as “structural volatility.” The Statistical Capacity Indicator (SCI)—which measures the timely production of credible statistics, a proxy for the degree of information forecasters may have in their arsenal (see box II.1)—shows that in aggregate, the developing economies in the MENA region have the lowest score. Moreover, the score has been falling over time. Public debt reporting standards, as well as measures used to characterize labor markets, are not up to international standards (Arezki and others 2020). With limited or imprecise data, forecasters have fewer inputs to load into their models, which reduces the accuracy of their forecasts.

Several economies in the region are periodically in conflict. The MENA region accounted for 68 percent of global battlerelated deaths from 2013 to 2017 (World Bank 2020). Yemen has the world’s largest humanitarian crisis, with about 80 percent of its population in need. The cumulative GDP losses in Syria through 2017 are estimated to be four times Syria's GDP in 2010. When conflict occurs, information systems are often discontinued or destroyed and data become scarce. Forecasting under such conditions can be trying.

Finally, the direct or indirect dependence of MENA economies on commodities makes them vulnerable to strong fluctuations in commodity prices. Many MENA economies are major and highly concentrated oil exporters and significant food importers. The ensuing macroeconomic volatility is associated with larger forecast errors as models face difficulty in accommodating wild swings.

Forecasters and their errors have been much studied in the economics profession. In the next section, recent global data on growth forecast errors are analyzed to uncover patterns of forecast errors in MENA and beyond, relating them to the interplay of data opacity, growth volatility, conflict, and exposure to commodity shocks.

Box II.1. Statistical Capacity Indicator

The World Bank’s Statistical Capacity Index (SCI) is a measure of data transparency (see box II.2). The overall SCI score is based on a diagnostic framework to assess the capacity of national statistical systems over time. The framework has three dimensions: source data; methodology; and the periodicity and timeliness of socioeconomic indicators. A composite score for each dimension and an overall score combining all three dimensions are derived for each country on a scale of 0–100. A score of 100 indicates that the country meets all criteria. Each dimension is evaluated on criteria based on metadata information obtained from the World Bank, International Monetary Fund, United Nations, the UN Educational, Scientific and Cultural Organization (UNESCO), and the World Health Organization (WHO). The SCI is collected by the World Bank’s Data Group. The overall SCI score is the average of the three sub-indicators calculated for each dimension (see table BII.1). The source data dimension reflects whether a country conducts data collection activity in line with internationally recommended frequency (periodicity), and whether data from administrative systems are available and reliable for statistical estimation purposes. This dimension covers the micro-data aspect of data transparency that is essential because microdata are at the foundation of a country’s data system. Specifically, the criteria used are the frequency of population and agricultural censuses and of poverty- and health-related surveys, and completeness of vital registration system coverage. A country can achieve a perfect score if it has conducted at least one population census in the past

continued on next page