44512 noTE no. 36 ­ May 2008GRIDLINES Sharing knowledge, experiences, and innovations in public-private partnerships in infrastructure Does the private sector deliver on its promises? Evidence from a global study in water and electricity Katharina Gassner, Alexander Popov, and Nataliya Pushak I s private operation better than public when Despite the obvious difficulties, understanding the it comes to utilities? A recent global study tradeoffs between public and private management funded by the World Bank and PPIAF exam- is critical for policy makers and their advisors. ines the effect of private sector participation The private sector has long been advocated as in electricity distribution and water and sanita- a solution to the service delivery gap faced by tion services. Using a data set of more than developing countries. But the wide range of results 1,200 utilities in 71 developing and transition observed case by case has led to strong feelings economies, the study finds that privately oper- both for and against private involvement in util- ated utilities convincingly outperform state- ity services, and any resolution of the debate has run ones in operational performance and labor often seemed far away. productivity. Whether privately managed utilities outperform A global study for robust results those run by the state is an old question. What makes it difficult to answer is that utilities such as To address the question as rigorously as possible, water and electricity distribution companies do not and distill universally applicable results from the operate in competitive markets, where a change multitude of evidence, the World Bank and PPIAF from public to private management is expected to funded a global study examining the impact of lead to cost savings and efficiency gains driven by private sector participation (PSP) in water and the profit motive. Indeed, studies looking at priva- electricity distribution (Gassner, Popov, and tized firms operating in competitive markets have Pushak 2008). The research analyzes a sample of reported increases in labor productivity, output, 301 utilities with PSP and 926 state-owned enter- service quality, and investment.1 prises (SOEs), in 71 developing and transition economies, over more than a decade of operation The empirical results in electricity distribution (table 1). The data set compiled is unique in its and water and sanitation services are far less clear coverage, and its size and composition make it cut. These services have features traditionally possible to address for the first time methodologi- used to justify public involvement rather than a cal problems that have plagued empirical research competitive market. They are natural monopolies and hampered conclusive results. (when the service is provided through networks), generate externalities, and are (particularly in the Studies on natural monopoly industries have tradi- case of water services) considered a human right tionally suffered from small sample size and taken and an important element of social and develop- ment policies.2 The question whether privately managed utilities outperform publicly run ones is especially difficult to answer in developing coun- Katharina Gassner is a senior economist in the World tries, where the effect of weak or inappropriate PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY Bank's Sustainable Development Network Vice Presidency, legal and institutional environments must also be Finance, Economics and Urban Department. Nataliya taken into account. Pushak is a consultant in the same department. Helping to eliminate poverty and achieve sustainable development through public-private partnerships in infrastructure PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY TABle 1 The sample: utilities by region and sector electricity Water and sanitation Both sectors Region PSP SOe Total PSP SOe Total PSP SOe Total East Asia and Pacific 1 2 3 10 87 97 11 89 100 Europe and Central Asia 35 21 56 29 365 394 64 386 450 Latin America and the Caribbean 111 44 155 94 330 424 205 374 579 Middle East and North Africa 1 2 3 4 29 33 5 31 36 South Asia 3 3 6 0 0 0 3 3 6 Sub-Saharan Africa 9 18 27 4 25 29 13 43 56 Total 160 90 250 141 836 977 301 926 1,227 Source: Authors' calculations. Note: PSP = utilities with private sector participation. SOE = state-owned enterprises. the form of case studies, which cannot produce Not all private participation is generalizations. The global study, by contrast, uses the same a database covering all the electricity distribution and water and sanitation companies that experi- Much past research has concentrated on "pure enced PSP between the beginning of the 1990s privatization"--permanent private control over and 2002. Moreover, the study compares these PSP firms with a set of sufficiently similar state- business assets and associated rights. But because owned utilities to establish meaningful--"like with of natural monopoly features and social and politi- like"--comparisons. Finally, given the long period cal considerations, full divestiture of assets is rare in electricity and especially so in water and sani- The private covered, the study is able to address the question of the counterfactual in a dynamic way, showing tation. The study therefore examines the broad sector clearly how the performance of firms with PSP changed range of legal arrangements for involving the private sector--management and lease contracts, delivers on over time and comparing that change with the performance of firms that remained state run. concessions, and partial as well as full divesti- operational tures--using the transfer of operating rights to determine whether a utility is privately operated. performance To achieve robust results and isolate as much as possible the impact of introducing PSP from other external and internal changes that may also affect The results are differentiated by type of contract. firms' performance, the study uses a dual esti- The strength of the PSP impact is expected to vary mation strategy. This dual approach controls for by contract type, and the predominant type differs the fact that a utility is unlikely to be randomly by sector: divestitures (full and partial) account selected for PSP and the possibility that the analy- for most PSP cases in electricity distribution, sis might compare PSP cases with fundamentally and concessions for most in water and sanitation dissimilar SOEs and thus produce biased results. (figure 1). The study produces two sets of results: the first using a larger but potentially biased data set includ- Performance does improve . . . ing all available SOEs as comparators; the second using a smaller set of SOEs carefully selected for The results of the study show that the private sector their comparability. There is a tradeoff between the delivers on expectations of higher labor produc- two approaches: the larger data set allows a richer tivity and operational efficiency, convincingly differentiation of results by PSP type and period, outperforming a set of comparable companies that while the smaller one provides a more rigorous remained state owned and operated. Comparing estimation but at the cost of fewer observations average annual values for performance indicators and results. To ensure robust findings, the study from the pre- and post-PSP periods shows that reports only results confirmed by both models. PSP is associated with: Does the private sector deliver on its promises? ˇ A 12 percent increase in residential connections for water utilities ˇ A 54 percent increase in residential connections FIgURe 1 per worker for water utilities, and a 29 percent Divestitures dominate in electricity, concessions in water and sanitation increase for electricity distribution companies ˇ A 19 percent increase in residential coverage for 2% sanitation services 5% 5% ˇ A 32 percent increase in electricity sold per 9% But lack of worker, and an 18 percent increase in water sold 24% per worker investment 43% ˇ A 45 percent increase in bill collection rates in raises electricity 47% 66% ˇ An 11 percent reduction in distribution losses concerns for electricity and a 41 percent increase in the about the number of hours of daily water service. electricity water and sanitation sustainability All these changes, occurring over a period of five full divestiture partial divestiture of the gains years or more, are over and above those recorded concession lease or management for the state-owned companies. Source: Authors' calculations. . . . but staff reductions also occur The clear improvement in operational performance and coverage. The study finds mixed evidence on is encouraging for proponents of PSP. But the this issue and so cannot conclude that investment results also confirm one reason that introducing always increases with PSP (despite the evidence of the private sector so often provokes fierce political increases in water connections). resistance and public hostility: the labor productiv- ity gains are linked to a reduction in staff numbers The investment question is best examined at a in both water and electricity (no separate results disaggregated level. For electricity divestitures, as are available for sanitation), with the strongest economic theory predicts, investment per worker effects for divestitures. Following the introduction increases with PSP. For lease and management of PSP, average employment falls by 24 percent contracts, particularly relevant for water and sani- in electricity and by 22 percent in water. In other tation, there is generally no investment obligation words, on average state-operated utilities use more for the private party--and the results suggest employees than privately run ones to produce the that the public asset holding company does not same level of output. increase investment even if PSP brings operational improvements. For concession contracts there is Policy makers need to weigh the tradeoff between no conclusive evidence that investment increases. an increase in output and service quality and a reduction in staff. But it's worth bearing in mind Investment data are notoriously difficult to that while the staff reductions are significant at measure, and the results need to be interpreted the level of the utility, they occur over a number with care. Nonetheless, the evidence points to of years and are small relative to the national a lack of investment--public or private--in the labor force. Only in a few exceptional cases did maintenance and expansion of utility networks as the reductions in a utility represent more than a general rule, even where PSP leads to an increase 2 percent of national unemployment.3 Moreover, in operational efficiency. That raises concerns the study considers all staff reductions--whether about the long-term sustainability of the opera- layoffs or natural attrition--to be the same. tional improvements achieved. What about investment? And what about prices? Proponents of PSP long hoped--and political lead- A final key result relates to tariffs: except for elec- ers sometime rashly promised--that greater private tricity concessions, the study finds no evidence involvement in utility services would lead to vastly of a systematic change in residential prices as a greater investment and thus to greater capacity result of PSP. Yet in developing countries, where below-cost pricing of essential utility services is of them in detail. Still, the global study produces well documented,4 tariff increases for all but the clear findings that the private sector delivers on poorest households are often recommended as part operational performance and labor efficiency. of reform, to give a utility enough resources to address shortfalls in service. But the clear practical implications for labor means that governments need to address the employment The lack of any substantial difference in tariffs question seriously. Even though the observed staff between utilities with PSP and SOEs could have reductions improve utilities' productivity and are two explanations: tariffs changed in equal measure small relative to national unemployment, measures in both categories, or they did not change signifi- to mitigate the effects should be put into place cantly in either of them. The second explanation early on. The question is one of trading off the seems more likely in countries where affordability social costs of reform against the social costs of is a real concern for much of the population. The inaction. result may point to the economic and political diffi- culties of aligning tariffs with the costs of service The two other key findings relate to investment provision. Its implications for revenue streams call and tariff trends. Though the available data need into question the sustainability of private involve- to be further refined and analyzed, the study ment unless there are explicit subsidy payments. points to a worrying lack of investment in utili- The result might also explain the lack of public or ties by the public or private sector. And it finds no private investment. indication of tariffs moving closer to cost-recovery levels. These two findings are probably linked, and the subject deserves further attention from both Where do the efficiency gains go? researchers and reformers. If the efficiency gains associated with the entry Notes of a private operator do not translate into higher 1. See, for example, Megginson, Nash, and van Randenborgh investment or lower prices, where do they go? One (1994); Frydman and others (1999); La Porta and López-de-Silanes (1999); and Brown, Earle, and Telegdy (2006). possible explanation is that services are initially so 2. See Galiani and others (2005) for a discussion of what sets water underpriced that even significant efficiency gains supply apart from other goods and services. do not produce a financial equilibrium or justify 3. For a detailed discussion of the relative size of staff reductions, price reductions. Instead, the efficiency gains trans- see also McKenzie and Mookherjee (2003). late into better operational performance, such as 4. For evidence in Latin America, see Foster and Yepes (2006). reductions in distribution losses, and the govern- References ment spends less subsidizing its utilities. Brown, J., J. Earle, and A. Telegdy. 2006. "The Productivity Effects of Privatization: Longitudinal Estimates from Hungary, Romania, Another explanation may be that the private Russia and Ukraine." Journal of Political Economy 114 (1): 61­99. operator reaps all the gains through profits. Given Foster, V., and T. Yepes. 2006. "Is Cost Recovery a Feasible Objective the young regulatory environments in developing for Water and Electricity? The Latin American Experience." Policy countries, which often lack sufficient capacity for Research Working Paper 3943, World Bank, Washington, DC. supervising service contracts, this is a possibility Frydman, R., C. Grey, M. Hessel, and A. Rapaczynski. 1999. "When Does Privatization Work? The Impact of Private Ownership on that needs to be considered. Corporate Performance in the Transition Economies." Quarterly Journal of Economics 114 (4): 1153­91. Gassner, K., A. Popov, and N. Pushak. 2008. "An Assessment GRIDLINES Conclusion of Private Sector Participation in Electricity and Water and Sanitation Services in Developing and Transition Countries." Gridlines share emerging knowledge La Porta, R., and F. López-de-Silanes. 1999. "The Benefits of on public-private partnership and give an For each electricity or water utility Privatization: Evidence from Mexico." Quarterly Journal of Economics overview of a wide selection of projects from that shifts from public to private 11 (99): 1193­242. various regions of the world. Past notes can be operation, the potential for McKenzie, D., and D. Mookherjee. 2003. "The Distributive Impact found at www.ppiaf.org/gridlines. Gridlines are a improving performance depends of Privatization in Latin America: Evidence from Four Countries." publication of PPIAF (Public-Private Infrastructure PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY on a host of variables, observ- Economia 3 (2): 161­233. Advisory Facility), a multidonor technical assistance Megginson, W., R. Nash, and M. van Randenborgh. 1994. "The facility. Through technical assistance and knowledge able and unobservable. No Financial and Operating Performance of Newly Privatized Firms: An dissemination PPIAF supports the efforts of policy study can deal with every one Empirical Analysis." Journal of Finance 49 (2): 403­52. makers, nongovernmental organizations, research institutions, and others in designing and implementing strategies to tap the full potential of private involvement in c/o The World Bank, 1818 H St., N.W., Washington, DC 20433, USA infrastructure. The views are those of the authors and do PhONe (+1) 202 458 5588 FAX (+1) 202 522 7466 not necessarily reflect the views or the policy of PPIAF, PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY geNeRAl eMAIl ppiaf@ppiaf.org WeB www.ppiaf.org the World Bank, or any other affiliated organization.