68400 ACCESS TO ELECTRICITY IN AFGHANISTAN: A REVIEW OF RECENT DATA AND RECOMMENDATIONS TO IMPROVE UTILITY OPERATIONS April 2007 Introduction In contrast, the estimate based on the most current It is generally believed that Afghanistan has one of the Breshna data on residential consumers suggests an access lowest levels of access to electricity in the world; the figure of level of about only 11% on the national level. Of the urban 6% access on the national level is often cited. While it is households that participated in the survey, about 64% reported certainly true that overall access to electricity is low in this access to electricity through Breshna. overwhelmingly rural and poor country, a review of recent Access to electricity in Kabul alone is presented in Table household survey data and updated national utility data 2. In addition to the NRVA 2005 data and the Breshna utility suggests that the access to electricity is somewhat higher than data, the table shows data from a survey commissioned by the has generally been supposed, particularly in urban World Bank and carried out by the Central Statistics Office Afghanistan. and Kabul University in August 2006 called the Kabul This note presents the new data and examines the Household Energy and Water Survey (KHEWS). The data implications of higher levels of access for policy makers and from the two household surveys that were carried out in Kabul the managers of the power system in Afghanistan, particularly suggest that the utility data on consumers underestimate the in Kabul. Recommendations to improve the quality of the actual level of access by a degree even greater than was utility’s knowledge of the power system in Kabul follow the observed with the national-level data. According to the results analysis of the survey data. of the NRVA 2005 survey, about 59% of households reported access to electricity through Breshna. In August 2006, about Recent Household Surveys and Updated Utility Data 68% of the households that participated in the KHEWS Table 1 presents data on access to electricity in reported access to electricity through Breshna, in contrast to Afghanistan from the National Risk and Vulnerability the estimate of 34% that was derived from the Breshna utility Assessment (NRVA), a household survey, and from Da data on residential consumers in Kabul. Afghanistan Breshna Moassese (Breshna), the national utility. The NRVA, a large nationwide survey carried out in 2005, showed access to electricity through the public grid (Breshna) of about 16%. Table 1: Access to Electricity in Afghanistan: Comparison of Survey and Utility Data Public Access Access from any n= (Breshna) Source* % % NRVA 2005 National data (including kuchis) 30,020 15.6 24.8 of which, Urban 5,755 64.2 73.0 of which, Rural 22,601 4.3 14.1 Breshna utility data (2007) National level** 435,433 11.2% — * For NRVA, “any source� includes public supply, government generator, personal generator (engine), personal generator (micro-hydro), community generator (engine), community generator (micro-hydro), solar. ** Extrapolation based on Breshna data on residential customers and population data from the Central Statistics Office. If kuchi are excluded from this extrapolation (on assumption that kuchi would not consume power from public supply), the access rate becomes 11.9%. Table 2: Access to Electricity in Kabul: Comparison of Survey and Utility Data Public Access From Any n= (Breshna) Source* % % NRVA 2005 2,275 58.7 73.6 KHEWS 2006 2,000 68.2 88.6 Breshna Kabul utility data** 138,711 34.0 — * For NRVA, “any source� includes public supply, government generator, personal generator (engine), personal generator (micro-hydro), community generator (engine), community generator (micro-hydro), solar. For KHEWS, “any source� includes Breshna (own house), Breshna (through neighbor’s connection), personal generator (own), generator (neighbor’s), generator (locally privately owned, not neighbor’s), and community generator. ** Extrapolation based on Breshna Kabul data on residential customers and population data from the Central Statistics Office. Interpreting the Discrepancy between Utility and Survey have yielded some results in urban Afghanistan, and rural Data Afghanistan has seen a modest increase in access primarily As has been demonstrated by the data in Tables 1 and 2, thanks to the National Solidarity Program. there is a discrepancy between the level of access to electricity reported by people in household surveys, and the level of access to electricity that is derived from the Breshna data. A few factors might account for the discrepancy between the Breshna data and the data gathered by the two household surveys. First, the estimate of 6% nationwide access was probably inaccurate even when it was calculated some years ago. This estimate was an extrapolation of poor quality data on registered utility customers that were available after the fall of the Taliban on average household size and the overall population. The household surveys, on the other hand, were designed to yield results that are representative of the general Household Energy Survey being completed population, using established sampling and surveying methodologies. Third, the private efforts of individuals and domestic entrepreneurs have also contributed to the increased access to electricity and other energy services. Finally, some part of the discrepancy between the Breshna data and the survey data may be accounted for by unregistered connections to the public supply and/or poor record-keeping by the utility. In this regard, the case of Kabul is of particular note: the percentage of households that reported access to Breshna in the World Bank KHEWS survey is twice the percentage of households with access according to the data from Breshna. The analysis of the KHEWS data also considered the possibility that so-called “minor consumers� might account for Second, in the years of reconstruction, efforts by the the discrepancy between the estimates of access to electricity Government of Afghanistan and Breshna to increase access more accurately reflect the actual level of access to electricity in Afghanistan and particularly in Kabul. At the very least, it would seem reasonable to conclude that the estimate based on Breshna data represents the minimum boundary of the estimate of access to electricity, while the survey data represent the maximum boundary, with the actual level somewhere in between and likely closer to the survey results. Recommendations It is impossible to manage a power system efficiently and derived from the survey as opposed to the utility data. “Minor� effectively without accurate information on the number of consumers (in Dari, “fari� meaning “branch�) are those who consumers in the system and their demand characteristics legally receive electricity through “main� consumers (in Dari, (quantity and time of power consumption). The reasons for “asli� meaning “trunk�) who have the account with Breshna. this are obvious: in the short run, in addition to making the Because Breshna does not keep records on the number of already weak financial condition of Breshna worse (by not “minor� consumers, theoretically a large number of “minor� being part of the company’s formal billing and payment consumers could explain the discrepancy that has been system), a high percentage of unregistered consumers causes described. However, the data from the KHEWS show that demand to significantly exceed supply, resulting in voltage “minor� consumers account for only about 1.5% of the fluctuations, outages, equipment failure, and generally poor households that participated in the survey. Thus, this is not an quality of electricity supply for all consumers. In the medium- important factor in explaining the discrepancy. to-longer term, a poor knowledge of its customer base will The analysis also considered the possibility that people make it impossible for Breshna to forecast its electricity participating in the survey did not completely understand what demand (including willingness and ability of different was meant by “public supply� and “Breshna� as opposed to customer groups to pay for electricity services) and to generate other sources of electricity. However, it was clear from or contract sufficient power supply to meet this demand. discussions with the interviewers and the researchers that Breshna is well known in Afghanistan (indeed, it is essentially a household word) and that people understand well the difference between the relatively low-cost public supply of electricity and other sources of electricity that are much more expensive to the individual household. Finally, the analysis considered whether people would have any incentive to overstate their access to electricity through Breshna. However, there is no credible reason for people to do this, indeed, the opposite incentive would seem to prevail (i.e., people underreporting access to Breshna in order to keep unregistered connections from coming to the attention The primary recommendation coming from this review of of the authorities). recent data on access to electricity is for the Ministry of In conclusion, considering all the facets of the analysis as Energy and Water and Breshna to carry out a series of presented above and the fact that the two surveys gave results measures to improve the quality of the data on the power that are very similar, it seems likely that the survey results distribution system in Kabul. Kabul represents fully half of the national power system in Afghanistan according to data on fluctuating voltage levels (which damage machines and power billed; it could be that Kabul is, in fact, even more appliances), power surges, and blackouts are common. There significant in the national power system than the currently are still extreme supply-side constraints for existing available data suggest. Without accurate data on the number of consumers (power from the public grid is available only a few consumers on the system (as well as more detailed data on hours daily in most places), to say nothing of the supply that consumers’ locations on the system and relative levels of would be required to fuel economic growth; and industrial demand, and so forth), there is a high risk that investment and load is virtually nonexistent. The high cost of private operational decisions could be misguided or suboptimal. generation is prohibitive for most of the population, allowing for only very limited generation when the public supply is not available. The government and its international partners need to keep a strong focus on addressing these fundamental problems while improving the operations of the national utility, Da Afghanistan Breshna Moassese (Breshna), in particular, dramatically improving the quality of data collection in the national power system. Specific measures to consider are: A comprehensive census of the Kabul power consumers to accurately re-assess the system and create an accurate and comprehensive customer database; Regularizing and/or disconnecting unregistered consumers; Computerizing record keeping and providing training for workers in using the computerized system. At present, most data are recorded by hand and on paper, which leads to a high level of errors and giving a high level of discretion to bill collectors and other workers in the system who interface directly with consumers; Outsourcing meter reading and customer management (with incentives to reduce commercial losses) and installing an automated meter reading system for the larger-volume customers. The World Bank Closing Thoughts In Washington: Michael Haney Notwithstanding the higher-than-supposed level of access Phone: 1-202-473-3680 to electricity in Afghanistan, particularly in Kabul, it should be E-mail: mhaney@worldbank.org stressed that the electricity sector continues to severely Website: constrain economic and human development in Afghanistan. www.worldbank.org.af The power supplied is generally of poor quality, meaning