2017/73 k nKonw A A weldegdeg e ol n oNtoet e s eSrei r e ise s f ofro r p r&a c t hteh e nEenregryg y Etx itcrea c t i v e s G l o b a l P r a c t i c e The bottom line Forecasting Electricity Demand: An Aid for Practitioners Forecasts of electricity demand in the developing world are central to any discussion Why are electricity-demand forecasts important? This Live Wire provides practitioners with a quick reference to historical trends for electricity demand in the developing world and about investment plans for Accurate forecasts of electricity demand inform to new econometric forecasts at the country level. By providing a power generation. Yet demand investment decisions about power generation succinct overview, it allows the patterns and interrelationships of forecasts are not always well founded and may be subject to and supporting network infrastructure economic growth (and other relevant drivers) to come into focus. The note also reviews demand-forecasting methodologies and discusses optimism bias. This Live Wire Of major interest to energy policymakers, power utilities, and their applicability in different contexts. provides practitioners with private investors alike, forecasts are also essential for development The table at the end of the note provides historic data on growth ready evidence on growth in professionals. Inaccurate forecasts, whether they over- or under in electricity demand as well as customized 5- and 10-year projec- demand for electricity, together predict demand, can have dire social and economic consequences. tions for 106 developing countries. with information on historic Underestimating demand results in supply shortages and forced trends and predictions for 106 power outages, with serious consequences for productivity and eco- developing countries. nomic growth. Overestimating demand can lead to overinvestment in How has electricity demand grown historically? generation capacity, possible financial distress, and, ultimately, higher On average, global growth in demand for electricity electricity prices. has been slowing since 1970 Jevgenijs Steinbuks Despite their central role in economic analyses of the power sector, demand forecasts are sometimes opaque. Demand forecasts Electricity demand is influenced by aggregate demand, changes is an economist in the Development Economics are critical in the design of least-cost generation plans for the power in energy intensity, and shifting input prices. Aggregate demand is Research Group at the sector, as well as in investment appraisals of individual power-gen- affected, among other factors, by economic and demographic growth, World Bank. eration projects. Nevertheless, demand forecasting often receives and access to electric power. Energy intensity changes with industri- Joeri de Wit is an less technical attention than other components of the power sector alization or deindustrialization—and with any consequent shift in the economist in the World analysis; the basis for forecasting can also sometimes be obscure. composition of industries. It also changes when technological progress Bank’s Energy and Given the paucity of data and the methodological challenges, affects energy efficiency. Shifting prices for inputs used to produce Extractives Global Practice. demand forecasts are sometimes derived from simple heuristics: electricity, or for intermediate products that are substitutes for (or Arthur Kochnakyan for example, using GDP-based demand-growth forecasts as proxies complements to) electricity, also affect electricity demand. is a senior economist for the growth in demand for electricity (sometimes with an adjust- GDP or other measures of economic output are often the in the same practice. strongest correlates of electricity demand. Since GDP forecasts are ment factor). Moreover, demand forecasting may be subject to optimism bias, which can be driven by the political economy or the available for a range of developing countries,1 they are often used Vivien Foster is a engineering culture surrounding large investment projects. Demand to forecast electricity demand. Figure 1 shows global demand and lead economist in the forecasting can be particularly challenging in developing countries, real GDP growth rates from 1970 to 2012, based on an unweighted same practice. where data is often elusive, political influences are often brought to average across countries. Two points are noteworthy. First, in bear, and historic electricity demand itself is more volatile owing to 1. See, for example, IMF World Economic Outlook Database, http://www.imf.org/external/pubs/ft/weo/2015/02/weodata/ instability, whether macroeconomic or political. weorept.aspx. 2 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Figure 1. Electricity demand and real GDP growth rates over time in OECD countries, 1970–2012 a. OECD countries b. Non-OECD countries 16 16 14 14 Electricity demand Electricity demand “Electricity demand has 12 Real GDP 12 Real GDP Growth rate (percent) Growth rate (percent) 10 10 grown faster than GDP for 8 8 GDP growth rates up to 5 6 6 percent and slower than 4 4 GDP for GDP growth rates 2 2 0 0 above 5 percent” -2 -2 -4 -4 -6 -6 1970 1980 1990 2000 2010 1970 1980 1990 2000 2010 Source: World Bank WDI database and IEA Extended Energy Balances Database. Figure 2. Real GDP growth rates against electricity-demand both the industrialized countries of the Organisation for Economic growth rates, 1990–2012 Co-operation and Development (OECD) and in non-OECD countries, long-term growth in demand for electricity is trending downward. 15 Second, for the non-OECD countries, there has been a marked convergence between electricity demand growth rates and GDP Percentiles Electricity demand growth rate (percent) 75th growth rates over this time period; whereas throughout the 1970s 10 50th and until the mid-1980s rates of growth of demand for electricity far 25th exceeded real GDP growth rates, the two came closer together since 5 the mid-1990s. Nevertheless, individual countries show significant variation in 0 rates of growth in demand for electricity around their respective GDP growth rates. Figure 2 plots real GDP growth rates, rounded to -5 the nearest percentage point, against the electricity-demand growth rates for the period since 1990. The relationship between electrici- 45º ty-demand growth and GDP growth does not lie on a 45-degree line, -1 -10 -5 0 5 10 15 meaning that the two—though correlated—are not typically equal Real GDP growth rate (percent) in value. In fact, electricity demand has grown faster than GDP for GDP growth rates up to 5.0 percent and slower than GDP for GDP growth rates above 5.0 percent. There is considerable variation across countries, however, with half the countries falling within four percentage points of this mean (table 1). 3 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Table 1. Real GDP growth rates against electricity-demand growth Though much of the variation in historic growth in energy rates: Linear fit (in percentages) demand can be attributed to differences in real GDP growth rates, other factors play a role. Figure 3 plots the average electricity Real GDP Electricity demand growth rates output and real GDP growth rates since 1990 across country growth rate Mean 25th perc. 50th perc. 75th perc. regions, income categories, electric system capacity, access rate, “Growth in electricity -1.0 1.1 -1.4 0.7 3.4 and oil-export orientation. Growth in electricity demand has been demand has been 0.0 1.7 -0.9 1.4 4.2 systematically higher in Africa, Asia, and the Middle East, and among 1.0 2.3 -0.4 2.0 4.9 countries having lower income, smaller power systems, and lower systematically higher in access rates—as well as for oil exporting countries. On average, 2.0 3.0 0.1 2.7 5.6 Africa, Asia, and the countries in groups with high electricity-demand growth also had 3.0 3.6 0.6 3.3 6.3 Middle East, and among high real GDP growth. Where this pattern breaks down, other factors 4.0 4.3 1.0 4.0 7.0 countries having lower are at play. For example, low-income countries experienced high 5.0 4.9 1.5 4.6 7.7 electricity-demand growth while showing relatively low growth in income, smaller power 6.0 5.6 2.0 5.3 8.4 real GDP . Economic transformation—characterized by intensifying systems, and lower access 7.0 6.2 2.5 5.9 9.1 production and greater access rates, among other factors—changes rates—as well as for oil 8.0 6.8 3.0 6.6 9.8 growth rates for electricity demand beyond growth arising from exporting countries.” 9.0 7.5 3.5 7.2 10.5 output shifts alone. Figure 3. Average electricity demand growth rate across different country groups, 1990–2012 (percent) Real GDP growth rate 7.5 8 6.8 7.0 6.5 6.5 growth rate (percent) 7 6.4 Electricity demand 5.8 5.6 6 4.9 4.7 4.1 4.7 5 4.3 3.7 3.9 3.1 2.5 3.3 4 2.5 3 2 1 0.6 0 World Europe and Central Asia OECD Latin America and Caribbean Middle East and North Africa Sub-Saharan Africa South Asia East Asia and Pacific High Upper-middle Lower-middle Low > 100 GW [10 GW,100 GW) [1 GW,10 GW) [0 GW,1 GW) <50% > 50% No Yes Region Income group System capacity Access rate Oil exporter? 4 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Table 2. Accuracy of best performing time-series forecasting models vis-à-vis ad hoc forecasting methods (benchmark models) 5-year forecast horizon 10-year forecast horizon Accuracy Statistical Accuracy Statistical improvement significance (percent improvement significance Benchmark heuristics (percent) cases (percent) (percent cases) “When good data are Constant historical electricity growth rate 77 85 74 68 available, practitioners Electricity growth rate proportional to GDP growth 184 88 124 73 can employ econometric methods to deliver accurate forecasts of demand.” How should electricity demand be forecast? As an alternative to these ad hoc approaches, we present econo- metric forecasts based on a “horse race” benchmarking of popular Both methodology and data availability make time-series econometric forecasting methods. First, a number of forecasting electricity demand a challenge for experts electricity-demand forecasts are computed based on a time series of There are various forecasting approaches, models, and methods: electricity production in 106 developing countries from 1960 to 2012. econometric analysis, general equilibrium economic models, and The econometric analysis includes the most popular time-series engineering or mathematical methods—many of them also data model families. Altogether, 33 model specifications are estimated for intensive. They capture economic and population growth, variations stationary electricity production data series; 36 model specifications in input prices, and prevailing weather conditions and patterns. are estimated for nonstationary series. These econometric specifi- Moreover, while the ultimate goal is to forecast electricity demand, in cations comprise both univariate time-series models, where historic most countries only electricity production data are available for fore- electricity production is the sole determinant of future demand, and casting purposes. As electricity is a nonstorable and poorly tradable multivariate time-series models, which include other predictors, such commodity, output is a reasonable proxy for total final consumption. as countries’ GDP and demographics.2 The accuracy of the resulting But using electricity output data will lead to downward-biased electricity production forecasts is then evaluated for each country. forecasts in the presence of market distortions and correspondingly Specifically, econometric models are tailored to the historic series, high suppressed demand. leaving out the final five and ten years for model validation. Sample Ideally, demand data should be broken down across customer econometric forecasts are then compared against the actual values categories, such as residential and nonresidential demand, which of electricity output in the validation sample; the best-performing may behave quite differently. A significant time series of 10–20 years models, which yield the lowest forecast error, are selected. is usually needed, ideally with quarterly resolution to capture sea- Time-series econometric forecasts are shown to systematically sonal patterns. When good data like these are available, practitioners outperform heuristic methods of demand forecasting; they also can employ econometric methods to deliver accurate forecasts (see perform quite well relative to more refined microeconometric box 1 on page 6 for a recent example from Armenia). models. Table 2 compares the accuracy of best-performing When data are of lesser quality or do not exist, practitioners may forecasting models against two benchmark heuristics as follows: resort to simple heuristics—for example, assuming that electricity (i) electricity demand grows at the same constant rate as in the demand will continue to grow at a predetermined rate (most past, or (ii) electricity demand grows proportionally to GDP growth. frequently, the historical average) or that it is somehow proportional The comparison between the time-series econometrics and the to projected GDP growth. 2. Further details on models and methods used are available in Steinbuks (forthcoming 2017). 5 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Figure 4. Time-series econometric forecast errors by country categories 12 5-year horizon (percent) 10.7 9.4 Forecast error over 10 9.0 8.9 8 6.0 6.0 5.7 6.0 5.9 6.3 6 5.0 5.3 4.9 5.2 4.6 4.3 “Time-series econometric 4 2.7 2.4 1.7 forecasting methods 2 0 yield accurate forecast World Europe and Central Asia Latin America and Caribbean Middle East and North Africa Sub-Saharan Africa South Asia East Asia and Pacific High Upper-middle Lower-middle Low > 100 GW [10 GW,100 GW) [1 GW,10 GW) [0 GW,1 GW) <50% > 50% No Yes predictions for the majority of developing countries.” Region Income group System capacity Access rate Oil exporter? heuristics is based on two measures: the “accuracy improvement” forecast predictions for the majority of developing countries, with a or mean percent difference between model average within sample mean average error of 6 percent over the five-year forecast horizon. forecast errors, and the “statistical significance,” or the percentage of This implies that, for many countries, time-series econometric cases for which the difference in model forecast errors between the methods perform well relative to more data-intensive econometric two approaches is statistically significant. Table 2 demonstrates that methods (as illustrated in box 1). The quality of electricity-production forecasts based on time-series econometric methods significantly forecasts diminishes, however, for the countries of Sub-Saharan outperform projections based on either of the ad hoc methods Africa, low-income countries, and countries with low electricity-ac- used by practitioners, although the advantage diminishes slightly as cess rates and small electricity-generation systems. The forecasting the forecasting period is extended from 5 to 10 years. The massive accuracy of time-series methods is also greatly diminished for accuracy improvement over the benchmark model that assumes countries that have recently undertaken major investments (e.g., that electricity demand is proportional to GDP may at first seem Ethiopia, Cameroon, and Myanmar) or disinvestments (e.g., Lithuania) counterintuitive, given the high popularity of this approach among in electricity-generation assets; countries that have volatile electricity energy practitioners. But, as shown above, GDP growth is not always production or rely heavily on electricity imports (e.g., Albania, Benin, strongly correlated with electricity demand. And when it is, multi- and Botswana); and countries affected by war (Iraq, Libya, Syria) or variate econometric time-series models estimate a more accurate disaster (Haiti). For these countries, the application of more rigorous GDP-energy multiplier when compared with a simple 1:1 ratio. forecasting methods is advised. Time-series econometric models produce accurate forecasts, For a country-by-country table showing both historic demand although certain types of countries are systematically more difficult growth trends (2000–15) and projected demand growth rates for to forecast. Figure 4 plots the average electricity-output forecast 2015–20, see table 3 at the end of this note. The table also provides errors based on the best time-series model across country regions, information on the preferred time-series forecasting model selection income categories, electric system capacity, and oil-export orienta- for each country, as well as the mean forecast error. Practitioners tion. Time-series econometric forecasting methods yield accurate may find this helpful both as a quick check at the country level on 6 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Box 1. Forecasting electricity demand in Armenia Armenia recently underwent an intensive debate about investment in electricity generation—particularly whether an expansion scenario based on an oversized nuclear power plant should be preferred over one based on gas-fired generation. A robust forecast was needed to avoid economically unjustified investments. Forecasting electricity demand in Armenia is greatly facilitated by high-quality historical data, including quarterly series for aggregate income (GDP) and end-use electricity prices. As the electricity sector is a relatively minor part of the country’s GDP and the changes in prices are primarily supply-driven owing to changes in costs (regulated tariffs), one can plausibly assume the exogeneity of explanatory variables. The least squares regression approach was therefore employed to forecast electricity production. Quarterly dummy variables were included to capture the seasonal changes in electricity demand. Separate models were estimated for both residential and nonresidential categories of consumers. The preferred models were selected based on which performed best out-of-sample. Each model was then back-tested using historical quarterly data for 2003–10. The models were evaluated based on the Root Mean Square Error (RMSE) for the forecast years. The model with the lowest RMSE was selected and then refitted for all available quarters (2003 to 2010). The first table below shows the outcome of the estimated residential model. Overall, the best model explained 91.6 percent of the total variation in residential demand for the period 2003 to 2010. GDP and seasonal dummy variables were found to be statistically significant. The resulting income elasticity is 0.31. Estimated residential model: ln DRESt = Ln b0+ b1Ln Yt+ b2Q2+ b3Q3+ b3Q4+et Coefficients Estimate t Stat b0 Constant 2.168 2.798 b1 GDP 0.310 5.139 b2 Q2 -0.433 -13.683 b3 Q3 -0.391 -12.348 b4 Q4 -0.102 -3.200 b0 Constant 2.168 2.798 The second table shows the outcome of the estimated nonresidential model. Overall the model explains 91.7 percent of the total variation in nonresidential electricity demand for the period 2003 to 2010. All included variables are found to be statistically significant. The estimated elasticity for income is 0.38; for price, -0.38. Estimated residential model: ln DNONt = ln b0+ b1Log Yt+ b2Ln PNONt+ b3Q2 +et Coefficients Estimate t Stat b0 Constant 2.617 2.556 b1 GDP 0.379 7.014 b2 Price -0.375 -2.447 b3 Q2 -0.101 -7.334 With accurate and reliable electricity demand forecasts in hand, the government of Armenia has gone through several iterations of an investment plan for least-cost electricity generation. It decided to reconsider its earlier approach to investing in excessive generation capacity. 7 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s forecasts of electricity demand, or as a first approximation when no econometric methods, which are useful reference points when more other information is available. Those wishing to develop their own elaborate forecasts are either unavailable or too costly to obtain. econometric models may benefit from indications showing which These forecasts are demonstrated to be an order of magnitude more specifications proved most successful for each country. Graphic accurate than the crude heuristics commonly used, which assume representations of each country’s historic electricity-demand trend that future electricity production grows at a constant historic rate or and forecast will also be available in Steinbuks (forthcoming 2017) . is proportional to GDP growth. On average, they are able to predict electricity demand with a forecasting error of only 6 percent. They How is this useful? are also more applicable in data-scarce environments than are more traditional microeconometric approaches, which rely on detailed, This note provides both off-the-shelf historical quarterly times-series of electricity demand broken down by trends for electricity demand growth and customer category, combined with parallel information on economic forward-looking 10-year demand forecasts for growth, demographics, and pricing. Careful judgment and common sense should be exercised in 106 developing countries applying the forecasts described here to specific country or project Based on a review of historic evidence and a systematic economet- settings. Trends shift over time, and forecasts are likely to be less ric effort, this note has shown that while electricity demand growth meaningful in power sectors undergoing major structural shifts. is strongly related to GDP , naïve heuristics based on GDP alone can be quite misleading. Other factors, such as overall level of economic References development, the size of the electric system, and industrial structure Steinbuks, J. Forthcoming 2017. “Assessing the Accuracy of Electricity also play a role in shaping future electricity demand. Demand Forecasts in Developing Countries,” World Bank Policy This note provides both off-the-shelf historical trends for Research Paper, Washington, DC. electricity-demand growth and forward-looking 10-year demand forecasts for 106 developing countries, which are available in This Live Wire was peer reviewed by Debabrata Chattopadhyay and table 3 at the end of this note. The latter are based on time-series Elvira Morella. 8 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Table 3. Electricity demand: Historical and forecast growth rates to 2020 2000–15 historic growth rates, in percent 2015–20 forecast growth rates, in percent Average forecast Country 2000–05 2005–10 2010–12 5 percent CI Mean 95 percent CI (percent error) Sub-Saharan Africa 6.6 4.7 5.9 n/a n/a n/a n/a Angola 18.6 19.1 3.5 -2.1 3.2 7.0 9.4 Benin 5.5 8.0 1.2 -7.5 0.0 2.3 29.4 Botswana -1.7 -7.7 -2.0 -20.0 -10.0 -4.9 34.5 Cameroon 3.0 9.5 2.1 1.3 1.5 1.7 0.7 Congo, Dem. Rep. 4.7 1.3 0.1 -2.1 0.1 1.5 1.6 Congo, Rep. 9.3 16.3 14.2 -20.0 4.3 8.7 38.2 Côte d’Ivoire 3.7 1.0 2.1 2.0 1.9 1.8 2.7 Eritrea 7.4 1.6 5.2 4.3 4.1 3.8 2.2 Ethiopia 14.0 15.0 22.3 23.2 24.4 25.4 6.0 Gabon 3.8 5.3 2.8 1.6 2.1 2.6 4.1 Ghana -1.2 10.0 3.3 1.6 1.4 1.2 18.9 Kenya 8.6 5.0 4.8 4.1 4.4 4.6 2.7 Mauritius 5.6 3.7 2.2 1.4 2.2 2.8 0.7 Mozambique 7.4 5.1 1.1 -5.1 -1.1 1.1 5.8 Namibia 3.6 -4.3 1.5 0.3 0.4 0.5 6.7 Nigeria 12.0 2.2 2.3 3.3 3.0 2.8 8.1 Senegal 11.7 4.2 3.9 3.6 3.5 3.4 2.2 South Africa 3.3 1.2 0.1 -1.0 0.2 1.1 2.4 Sudan 9.8 19.2 22.0 26.6 26.3 26.0 4.1 Tanzania 8.8 9.2 0.5 -20.0 0.5 71.1 1.9 Togo 1.6 -1.1 -4.2 n/a -2.8 5.7 18.8 Zambia 2.9 5.3 5.3 6.6 8.5 9.7 2.4 Zimbabwe 6.8 -1.6 0.7 -9.6 -3.3 0.1 4.4 East Asia and Pacific 7.8 6.0 7.3 n/a n/a n/a n/a Brunei 5.7 3.2 1.6 1.4 2.4 3.2 1.4 Cambodia 23.0 0.6 -0.5 -11.3 -1.6 1.2 16.6 China 16.9 13.5 17.1 12.1 12.4 12.7 3.7 Indonesia 7.4 6.6 7.8 6.5 6.9 7.3 1.7 Korea, Dem. People’s Rep. 3.6 -1.1 0.8 n/a 4.0 212.8 5.6 Malaysia 3.9 10.2 5.8 -2.7 -1.4 -0.2 3.2 (continued) 9 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Table 3. Continued 2000–15 historic growth rates, in percent 2015–20 forecast growth rates, in percent Average forecast Country 2000–05 2005–10 2010–12 5 percent CI Mean 95 percent CI (percent error) Mongolia 3.2 5.2 6.4 3.9 5.8 7.4 4.6 Myanmar 3.5 5.1 21.4 n/a 3.4 203.3 11.1 Philippines 5.0 4.0 2.7 2.3 2.7 3.0 2.4 Singapore 4.1 3.7 2.2 1.6 2.2 2.7 1.8 Thailand 7.5 4.1 1.3 2.1 2.0 2.0 1.7 Vietnam 20.4 15.4 13.9 12.8 12.9 13.0 1.7 Europe and Central Asia 2.5 1.3 3.4 n/a n/a n/a n/a Albania 3.0 7.9 -4.5 -4.1 0.7 2.0 17.4 Armenia 1.2 0.6 7.9 n/a 6.8 8.6 5.4 Azerbaijan 4.5 -3.6 5.0 -20.0 -1.3 -0.7 6.4 Belarus 3.7 2.5 -2.3 -0.5 -1.1 -1.5 5.5 Bosnia and Herzegovina 4.2 7.2 -1.6 1.6 3.3 4.4 7.6 Bulgaria 1.6 0.9 2.8 n/a 1.8 n/a 2.9 Croatia 3.1 2.7 0.8 1.0 1.6 2.0 10.1 Cyprus 6.0 4.3 -1.5 -1.5 0.0 1.2 4.8 Georgia -0.4 7.9 9.2 13.6 12.6 11.8 4.3 Hungary 0.3 0.9 -0.7 -1.4 0.2 1.5 8.1 Kazakhstan 6.4 4.4 4.3 -3.5 3.3 7.6 2.4 Kyrgyzstan -0.1 -3.7 0.8 -0.7 0.1 0.7 9.4 Latvia 3.7 7.0 -6.1 n/a -0.2 n/a 6.7 Lithuania 5.9 -13.1 8.1 37.3 15.1 14.6 28.2 Macedonia 0.4 0.9 0.5 1.4 1.2 1.1 4.7 Malta 3.4 -1.1 0.8 -1.9 -0.2 1.1 3.7 Moldova 1.4 0.4 0.4 0.4 0.4 0.4 2.3 Poland 1.7 0.2 1.2 -1.2 0.5 1.8 2.5 Romania 2.9 0.4 1.7 n/a 1.0 n/a 3.6 Russia 1.7 1.8 1.3 1.4 1.9 2.4 1.6 Serbia 1.4 0.5 0.4 0.9 0.5 0.3 1.5 Tajikistan 4.0 -0.8 -0.4 -0.3 -0.3 -0.2 4.0 Turkey 5.9 6.1 7.0 -20.0 2.8 53.4 5.1 Turkmenistan 6.0 6.0 2.6 1.4 1.4 1.3 1.3 (continued) 10 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Table 3. Continued 2000–15 historic growth rates, in percent 2015–20 forecast growth rates, in percent Average forecast Country 2000–05 2005–10 2010–12 5 percent CI Mean 95 percent CI (percent error) Ukraine 1.7 0.3 0.2 -2.8 -2.5 -2.2 4.6 Uzbekistan 1.0 1.0 2.9 3.6 3.4 3.3 1.0 Latin America and Caribbean 4.1 3.7 4.9 n/a n/a n/a n/a Argentina 3.7 3.7 4.2 3.4 3.8 4.1 3.2 Bolivia 5.2 8.4 6.4 6.3 7.1 7.8 2.4 Brazil 3.1 5.6 4.1 3.8 4.0 4.2 1.3 Chile 6.2 3.0 3.9 -1.0 1.6 3.8 3.1 Colombia 3.3 3.6 2.4 2.2 2.1 2.0 0.8 Costa Rica 3.9 3.2 2.3 n/a 2.0 n/a 1.2 Cuba 0.4 2.7 0.0 -0.8 0.5 1.4 2.5 Dominican Republic 9.7 4.2 2.6 2.3 2.1 1.9 2.2 Ecuador 4.0 10.7 3.7 -0.6 2.9 6.0 4.3 El Salvador 8.6 4.8 -0.6 -2.0 -0.2 1.2 2.6 Guatemala 6.6 2.1 2.2 1.8 1.8 1.7 2.3 Haiti 0.3 1.1 13.6 3.3 2.7 2.3 38.5 Honduras 10.7 4.2 3.5 2.2 2.1 2.1 2.0 Jamaica 2.5 -8.4 1.7 2.3 2.0 1.8 20.1 Mexico 3.9 2.2 3.6 2.1 2.0 2.0 2.3 Nicaragua 6.0 4.0 3.6 3.1 2.9 2.8 3.8 Panama 3.8 5.5 5.0 3.0 2.9 2.8 4.2 Paraguay -0.9 1.1 3.5 -0.4 2.2 2.8 2.1 Peru 5.6 8.2 9.1 7.0 8.8 10.4 1.8 Trinidad and Tobago 5.9 4.0 5.2 n/a 6.5 n/a 1.5 Uruguay 0.2 8.6 3.7 2.3 3.1 3.7 6.2 Venezuela 4.7 2.4 2.3 1.7 2.2 2.7 1.1 Middle East and North Africa 5.7 6.3 4.9 n/a n/a n/a n/a Algeria 6.7 7.0 6.3 5.1 5.8 6.3 7.6 Bahrain 8.0 4.2 2.3 0.6 2.2 3.5 1.1 Egypt, Arab Rep. 7.8 7.0 6.3 4.7 4.7 4.6 1.1 Iran, Islamic Rep. 9.3 6.2 2.9 2.2 2.2 2.3 0.9 Iraq -0.9 13.0 3.9 n/a 2.1 n/a 26.4 (continued) 11 F o r e c a s t i n g El e c t r i c i ty D e m a n d : A n A i d f o r P r a c t i t i o n e r s Table 3. Continued 2000–15 historic growth rates, in percent 2015–20 forecast growth rates, in percent Average forecast Country 2000–05 2005–10 2010–12 5 percent CI Mean 95 percent CI (percent error) Israel 2.8 4.1 2.1 2.1 2.0 1.9 2.0 Jordan 6.2 10.6 0.7 -20.0 1.3 20.7 2.5 Kuwait 7.1 6.1 3.2 3.4 3.3 3.3 1.4 Lebanon 5.5 5.3 0.7 1.6 2.5 3.1 4.6 Libya 9.3 8.9 -6.9 n/a 17.6 86.7 6.6 Morocco 10.0 4.5 6.0 2.2 3.8 5.1 5.5 Oman 7.8 11.3 13.6 13.6 14.3 15.1 5.1 Qatar 11.5 19.1 13.5 14.1 13.9 13.7 2.1 Saudi Arabia 7.9 7.3 6.3 5.1 6.0 6.8 5.7 Syria 7.7 6.6 -14.6 -20.0 -20.0 -20.0 9.7 Tunisia 3.9 6.4 3.1 3.2 3.7 4.1 1.5 United Arab Emirates 10.4 12.2 1.8 2.1 2.1 2.0 4.7 Yemen 7.9 12.5 -1.1 2.5 2.3 2.1 7.6 South Asia 7.6 5.3 5.5 n/a n/a n/a n/a Bangladesh 13.5 11.6 10.0 10.3 10.4 10.5 0.8 India 5.1 7.4 9.2 9.7 9.2 8.6 1.9 Nepal 10.5 5.3 7.3 6.1 6.0 5.9 1.7 Pakistan 7.5 0.2 0.2 -1.6 -0.1 1.0 1.4 Sri Lanka 6.6 3.2 3.8 2.2 2.1 2.0 2.9 Note: CI = confidence interval; n/a = not applicable. Get Connected to Live Wire Live Wires are designed for easy reading on the screen and for downloading The Live Wire series of online knowledge notes is an initiative of the World Bank Group’s Energy and self-printing in color or “Live Wire is designed and Extractives Global Practice, reflecting the emphasis on knowledge management and solu- black and white. tions-oriented knowledge that is emerging from the ongoing change process within the Bank for practitioners inside Group. For World Bank employees: and outside the Bank. Professional printing can Each Live Wire delivers, in 3–6 attractive, highly readable pages, knowledge that is immediately It is a resource to relevant to front-line practitioners. also be undertaken on a customized basis for share with clients and specific events or occasions Live Wires take a variety of forms: counterparts.” • Topic briefs offer technical knowledge on key issues in energy and extractives by contacting GSDPM Customer Service Center at • Case studies highlight lessons from experiences in implementation (202) 458-7479, or sending a • Global trends provide analytical overviews of key energy and extractives data written request to cgsdpm@ • Bank views portray the Bank Group’s activities in the energy and extractives sectors worldbank.org. • Private eyes present a private sector perspective on topical issues in the field Each Live Wire will be peer-reviewed by seasoned practitioners in the Bank. Once a year, the Energy and Extractives Global Practice takes stock of all notes that appeared, reviewing their quality and identifying priority areas to be covered in the following year’s pipeline. Please visit our Live Wire web page for updates: http://www.worldbank.org/energy/livewire e Pa c i f i c 2014/28 ainable energy for all in easT asia and Th 1 Tracking Progress Toward Providing susT TIVES GLOBAL PRACTICE A KNOWLEDGE NOTE SERIES FOR THE ENERGY & EXTRAC THE BOTTOM LINE Tracking Progress Toward Providing Sustainable Energy where does the region stand on the quest for sustainable for All in East Asia and the Pacific 2014/29 and cenTral asia energy for all? in 2010, eaP easTern euroPe sT ainable en ergy for all in databases—technical measures. This note is based on that frame- g su v i d i n had an electrification rate of Why is this important? ess Toward Pro work (World Bank 2014). SE4ALL will publish an updated version of 1 Tracking Progr 95 percent, and 52 percent of the population had access Tracking regional trends is critical to monitoring the GTF in 2015. to nonsolid fuel for cooking. the progress of the Sustainable Energy for All The primary indicators and data sources that the GTF uses to track progress toward the three SE4ALL goals are summarized below. consumption of renewable (SE4ALL) initiative C T I V E S G L O B A L P R A C T I C E ENERGY & EXTRA • Energy access. Access to modern energy services is measured T E S E R I E S F O R T H EIn declaring 2012 the “International Year of Sustainable Energy for energy decreased overall A KNO W L E D G E N Oand 2010, though by the percentage of the population with an electricity between 1990 All,” the UN General Assembly established three objectives to be connection and the percentage of the population with access Energy modern forms grew rapidly. d Providing Sustainable accomplished by 2030: to ensure universal access to modern energy energy intensity levels are high to nonsolid fuels.2 These data are collected using household Tracking Progress Towar services,1 to double the 2010 share of renewable energy in the global surveys and reported in the World Bank’s Global Electrification but declining rapidly. overall THE BOTTOM LINE energy mix, and to double the global rate of improvement in energy e and Central Asia trends are positive, but bold Database and the World Health Organization’s Household Energy for All in Eastern Europ efficiency relative to the period 1990–2010 (SE4ALL 2012). stand policy measures will be required where does the region setting Database. The SE4ALL objectives are global, with individual countries on that frame- on the quest for sustainable to sustain progress. is based share of renewable energy in the their own national targets databases— technical in a measures. way that is Thisconsistent with the overall of • Renewable energy. The note version energy for all? The region SE4ALL will publish an updated their ability energy mix is measured by the percentage of total final energy to Why is this important ? spirit of the work initiative. (World Bank Because2014). countries differ greatly in has near-universal access consumption that is derived from renewable energy resources. of trends is critical to monitoring to pursue thetheGTF in 2015. three objectives, some will make more rapid progress GTF uses to Data used to calculate this indicator are obtained from energy electricity, and 93 percent Tracking regional othersindicators primary will excel and data sources that elsewhere, depending on their the while the population has access le Energy for All in one areaThe goals are summarized below. balances published by the International Energy Agency and the the progress of the Sustainab respective track starting progress pointstowardand the three SE4ALL comparative advantages as well as on services is measured to nonsolid fuel for cooking. access. Accessthat they modern to are able to energy marshal. United Nations. despite relatively abundant (SE4ALL) initiative the resources and support Energy with an electricity connection Elisa Portale is an l Year of Sustainable Energy for To sustain percentage of by the momentum forthe the population achievement of the SE4ALL 2• Energy efficiency. The rate of improvement of energy efficiency hydropower, the share In declaring 2012 the “Internationa energy economist in with access to nonsolid fuels. three global objectives objectives, andathe means of charting percentage of the population global progress to 2030 is needed. is approximated by the compound annual growth rate (CAGR) of renewables in energy All,” the UN General Assembly established the Energy Sector surveys and reported access to modern universalAssistance The World TheseBank and data are the collected International using household Energy Agency led a consor- of energy intensity, where energy intensity is the ratio of total consumption has remained to be accomplished by 2030: to ensure Management Database and the World of theenergy intium of 15 renewable international in the World Bank’s Global agencies toElectrification establish the SE4ALL Global primary energy consumption to gross domestic product (GDP) energy the 2010 share of Program (ESMAP) relatively low. very high energy services, to double Database. measured in purchasing power parity (PPP) terms. Data used to 1 t ’s Household provides Energy a system for regular World Bank’s Energy the global rate of improvemen and Extractives Tracking Framework Health (GTF), which Organization in the energy intensity levels have come and to double the global energy mix, Global Practice. (SE4ALL 2012). based on energy. of renewable The sharepractical, rigorous—yet energy given available calculate energy intensity are obtained from energy balances to the period 1990–2010 global reporting, Renewable down rapidly. The big questions in energy efficiency relative setting by the percentage of total final energy consumption published by the International Energy Agency and the United evolve Joeri withde Wit is an countries individual mix is measured Data used to are how renewables will The SE4ALL objectives are global, economist in with the overall from renewable energy when every resources. person on the planet has access Nations. picks up a way energy that is consistent 1 The universal derived that isaccess goal will be achieved balances published when energy demand in from energy their own national targets through electricity, clean cooking fuels, clean heating fuels, rates the Bank’s Energy and countries differ greatly in their ability calculate this indicator are obtained to modern energy services provided productive use and community services. The term “modern solutions” cookingNations. again and whether recent spirit of the initiative. Because Extractives Global rapid progress and energy for Energy Agency and the United liquefied petroleum gas), 2 Solid fuels are defined to include both traditional biomass (wood, charcoal, agricultural will make more by the refers to solutions International that involve electricity or gaseous fuels (including is pellets and briquettes), and of decline in energy intensity some t of those of efficiency energy and forest residues, dung, and so on), processed biomass (such as to pursue the three objectives, Practice. depending on their or solid/liquid fuels paired with Energy efficiency. The rate stoves exhibiting of overall improvemen emissions rates at or near other solid fuels (such as coal and lignite). will excel elsewhere, rate (CAGR) of energy will continue. in one area while others liquefied petroleum gas (www.sustainableenergyforall.org). annual growth as well as on approximated by the compound and comparative advantages is the ratio of total primary energy respective starting points marshal. where energy intensity that they are able to intensity, measured in purchas- the resources and support domestic product (GDP) for the achievement of the SE4ALL consumption to gross calculate energy intensity Elisa Portale is an To sustain momentum terms. Data used to charting global progress to 2030 is needed. ing power parity (PPP) the International energy economist in objectives, a means of balances published by the Energy Sector International Energy Agency led a consor- are obtained from energy The World Bank and the SE4ALL Global Energy Agency and the United Nations. Management Assistance agencies to establish the the GTF to provide a regional and tium of 15 international for regular This note uses data from Program (ESMAP) of the which provides a system for Eastern Tracking Framework (GTF), the three pillars of SE4ALL World Bank’s Energy and Extractives on rigorous—yet practical, given available country perspective on Global Practice. global reporting, based has access Joeri de Wit is an will be achieved when every person on the planet The universal access goal heating fuels, clean cooking fuels, clean energy economist in 1 agricultural provided through electricity, biomass (wood, charcoal, to modern energy services The term “modern cooking solutions” to include both traditional and briquettes), and Solid fuels are defined the Bank’s Energy and use and community services. biomass (such as pellets 2 and energy for productive petroleum gas), and so on), processed fuels (including liquefied and forest residues, dung, involve electricity or gaseous at or near those of Extractives Global refers to solutions that overall emissions rates other solid fuels (such as coal and lignite). with stoves exhibiting Practice. or solid/liquid fuels paired (www.sustainableenergyforall.org). liquefied petroleum gas