WPS5704 Policy Research Working Paper 5704 The Politics of Power The Political Economy of Rent-Seeking in Electric Utilities in the Philippines Zahid Hasnain Yasuhiko Matsuda The World Bank East Asia Region Poverty Reduction & Economic Management Unit June 2011 Policy Research Working Paper 5704 Abstract This paper takes advantage of unique intra-country greater incentive to use the utility for rent-seeking than a variation in the Philippines power sector to examine national politician as they capture only a portion of the under what conditions politicians have an incentive to political gains from utility performance improvements “capture” an electric utility and use it for the purposes as some of the benefits of improved service will go to of rent-seeking. The authors hypothesize that the level other electoral jurisdictions within the utility’s catchment of capture is determined by the incentives of, and the area. The authors posit that three variables impact the interactions between, local and national politicians, magnitude of these incentives of local and national where the concepts of “local” and “national” are context politicians: (i) the local economic context, specifically specific. A local politician is defined as one whose the scale of rents that can be extracted from an electricity electoral jurisdiction lies within the utility’s catchment cooperative (ii) the degree of competitiveness of local area; by contrast, a national politician is defined as one politics; and (iii) the political salience of an electricity whose electoral jurisdiction includes two or more utility cooperative’s catchment area for national politicians. The catchment areas. These jurisdictional differences imply authors illustrate this framework through case studies different motivations for local and national politicians: of specific power utilities, and suggest some policy because of “spillover” effects, local politicians have a implications. This paper is a product of the Poverty Reduction & Economic Management Unit, East Asia Region. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at zhasnain@worldbank.org and ymatsuda@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team THE POLITICS OF POWER: The political economy of rent-seeking in electric utilities in the Philippines Zahid Hasnain* Yasuhiko Matsuda† JEL Codes: H41, H54, D72 Acknowledgements: This paper has benefitted from comments from Phil Keefer, Alan Townsend, Victor Dato, Beatriz Arizu, Takeshi Kawanakta and Motoky Hayakawa * The World Bank † The World Bank PHILIPPINES: THE POLITICS OF POWER 1. Introduction It is commonly known that public utilities in developing countries are particularly susceptible to political capture and that the use of these utilities for the purposes of rent-seeking is a major cause of their poor performance (Shirley 2002, Lal 2006). What is less well known however, is what drives this political capture in specific contexts — under what conditions politicians are more likely to interfere in the day-to-day operations of utilities and when will they have less incentive to do so. This paper takes advantage of unique intra-country variation in the Philippines power sector to show that rents do indeed drive capture, but the effect is contingent on the nature of local political competition. Electric cooperatives (ECs) provide power to roughly half of the households in the Philippines and are the main, if not exclusive, distribution utilities for consumers who live outside of the major metropolitan areas. While the overall financial health of the sector has improved steadily over the past decade, there is great diversity among the 119 cooperatives that exist, and many cooperatives continue to have high technical and financial losses, and require government subsidies. An earlier World Bank study of the sector identified a variety of geographical, regulatory, and corporate governance factors behind this variation in performance, but speculated that undue political interference in management as the core issue, and the minimization of political interference and the maximization of professional management as the heart of the reform agenda for the sector (World Bank 2000). This paper takes these observations as a point of departure and explores what explains the variations in the level of politicization of ECs across the Philippines. If it can be accepted that ceteris paribus, a public entity such as an electricity cooperative will perform better under a professional manager with relevant technical expertise and experience than otherwise, then an alternative way to pose the same question would be: why are some ECs used for serving the interests of politicians and a small number of their supporters through, for example, regular interference in day-to-day management in matters involving procurement, personnel recruitment and management, billing, and electrification while others are left alone to be run professionally by the ECs‘ managers? This research question draws on the earlier finding that the level of politicization has a major impact on EC performance, and therefore a framework that explains the variation in levels of politicization will also explain a large part of the variation in EC performance and therefore electricity service provision in the Philippines. The paper‘s main contribution is the development of an analytical framework, what we refer to as the ‗political economy of nested capture‘, to explain this variation. To put simply, we hypothesize that the level of politicization of an EC is determined by the incentives of, and the interactions between, local and national politicians, where the concepts of local and ―national‖ are context specific. A local politician is defined as one whose electoral jurisdiction lies within an EC‘s catchment area or, to state in another way, an EC catchment area contains two or more such electoral jurisdictions. By contrast, a national politician is defined as one whose electoral jurisdiction includes two or more EC catchment areas. In the Philippines context, given that a typical EC will include multiple municipalities and congressional districts, mayors and national congressmen can be considered local politicians. By contrast provincial governors usually, and in the extreme case the president, are national politicians. These jurisdictional differences imply different motivations for local and national politicians with respect to the EC. Each politician faces a choice: to use the EC as a source of rents to finance 2 their electoral campaigns and to build alliances with locally influential individuals who can help them in elections, or to help the EC be run professionally to improve electricity services. While using the EC as a source of rents can help finance elections and other political needs (e.g., maintenance of a private militia in extreme cases) it also imposes a cost to the locality in terms of production losses due to poor power supply. Presumably improved power supply would help the politician get elected; however, a local politician can capture only a portion of the political gains from EC performance improvements as some of the benefits of improved service will go to other electoral jurisdictions within the EC‘s catchment area. By contrast, a national politician can capture all of the political benefits of improved services but the EC is likely to be only a small source of rents given the size of their electoral jurisdictions. The implication therefore, is that local politicians have a greater incentive to use the EC as a source of rents and a lesser incentive to improve service delivery than national politicians. The latter, on the other hand, may have greater incentive to keep the EC performing well, but given the typically limited scope of EC operations in their electoral jurisdiction (i.e., nation- or province-wide), the EC is expected to occupy only limited space on their political calculation, dampening their incentives to actively intervene in its management and operations. We posit that three variables impact the magnitude of these incentives of local and national politicians: (i) the local economic context, specifically the scale of rents that can be extracted from an EC (ii) the degree of competitiveness of local politics (where local is defined as above); and (iii) the political salience of an EC‘s catchment area for national politicians. On the first point, a large EC in a largely agrarian economy will be a big source of rent but would impose fewer costs in terms of production losses than a smaller EC in an industrial area where the costs of disrupted power will be higher. The more competitive the local election, the more the need for local politicians to use these rents to target key voters to swing an election or otherwise obtain an advantage over the political rival, and conversely the more likely that these targeted interventions would be successful (as relatively fewer votes need to be ‗bought‘). The greater the political salience of the EC catchment area, the greater the incentive of a national politician to intervene and to prevent the EC‘s poor performance from becoming a national political issue. The level of politicization of an EC, and therefore its performance, will then be a function of these three variables. This finding that rent-seeking is actually greater in settings that are more electorally competitive is surprising but consistent with that of a rich academic literature on analytical political economy that has explored the limitations of elections for holding politicians accountable, and the incentives of politicians to provide targeted benefits to particular constituents (―pork-barrel‖ or ―clientelist‖ policies) as opposed to public goods that benefit a large number of voters. In applying these insights to a particular sector in a particular country, the paper attempts to provide additional insights into politicians‘ incentives controlling for many of the features, such as institutional structures (presidential vs. parliamentary systems, coalition vs. majority governments), electoral systems (majoritarian vs. proportional), and societal features (ethnically divided vs. homogenous) that the literature has emphasized as key explanatory variables. It is also hoped that the analytical framework will prove useful for the analysis of rent-seeking in other sectors, particularly in the infrastructure sectors with similarly high political visibility such as water supply and sanitation, and roads and transportation. The paper is structured as follows: the next section provides background on the power sector in the Philippines, giving an overview of the legal and institutional framework governing ECs, and of sector performance. Section 3 summarizes the relevant theoretical and Philippines-specific academic political economy literature. Section 4, the theoretical core of the paper, presents the political economy of nested capture framework. Section 5 puts some flesh to this framework by 3 through case studies of select ECs — these cases are illustrative and are not meant to be a ―test‖ of the theoretical argument. The final section presents some implications of the analysis for improving the sector. 2. Sector Overview1 Electric cooperatives as the institutional mechanism for rural electrification in the Philippines had their genesis in the 1960s with the launch of pilot projects assisted by USAID and modeled on such initiatives in the US. With the passage of the National Electrification Act in 1969, ECs were designated the country‘s primary electricity distribution system, and the sector expanded rapidly in the 1970s with cooperatives covering all areas of the country with the exception of metropolitan Manila and Cebu, and some of the more remote islands. Presently (2010), the 119 ECs in the Philippines account for approximately 57 percent of total connections and 24 percent of total energy consumption. The majority of ECs are connected to the grid and purchase electricity from the national transmission company. Access has increased rapidly, and 97 percent of barangays and 78% of households now have electricity connections. The legal framework for rural electrification in the Philippines is set out in the Electric Power Industry Reform Act (EPIRA) 2001, the Presidential Decree No. 269, and the Presidential Decree No. 1645 (which makes amendments to Decree 269). EPIRA stipulates that ECs can be either non-stock, non-profit cooperatives and be governed under PD 269; stock cooperatives and be governed under the Cooperative Code; or stock corporations and be governed under the Corporation Code. The vast majority of ECs —112 out of 119 as of 2008 — are non-stock and non-profit, owned by member consumers, and under the domain of PD 269. These ECs are under the domain of the National Electrification Administration (NEA) and while NEA is not a regulator in the strict sense of the word, it is tasked with providing financial and technical support to the ECs, and monitoring their performance. These non-stock ECs will be the focus of this paper. ECs are owned by their members/consumers on payment of a nominal fee, with the members electing a board of directors who are responsible for the overall policy setting of the organization. The mode of elections is specified in guidelines issued by NEA. The board, which numbers between 5 and 15 depending on the size of the EC, is elected on the basis of one-member one vote, with a once consecutive renewable term of three years, and with a third of the board up for election each year. The geographical area of the cooperative is demarcated into districts which usually encapsulate one or more municipalities, with each director representing a district. Aside from citizenship and membership of the cooperative, the only major restriction on eligibility for standing for the board elections is that the candidate cannot be holding any elective government office except that of barangay captain. The board, which among themselves selects a president, vice president, secretary, and treasurer, is de jure responsible for managing the overall business of the EC, and can adopt the necessary by- laws for this purpose. It is meant to be a policy-making body, with actual day-to-day management the responsibility of a general manager. The board is required to hold an annual general members meeting, with a quorum being 5% of members or a 100 whichever is less, and any additional members meetings as deemed necessary. For example, major policy decisions, such as the consolidation of two or more cooperatives, or a merger, need a two-thirds approval of the members of each of the concerned cooperatives. Similarly, the conversion of the EC into a 1 The description of the sector is based on data available circa 2008-09. 4 stock cooperative or a stock corporation requires the two-thirds ratification of members in a general members meeting. While the management of the EC is ostensibly accountable to the board, NEA, with its overall mandate to promote electrification under PD269 and in particular under the so-called ‗martial powers‘ introduced under PD1645, also exercises considerable technical, financial, and regulatory authority. Its powers include the technical and financial oversight of cooperatives, including the option of taking over management in extreme cases, administration of subsidies from the congressional appropriation for line expansion, guarantees to the cooperatives for power purchase from the electricity spot market, and administration of performance improvement plans. NEA is responsible for the overall performance monitoring of cooperatives and has a performance ranking system that includes measures on amortization, reductions in system losses, efforts at electrification, payments of generation and transmission charges, collection efficiency, and corporate governance. For ECs that score low, NEA has the power to appoint an acting general manager and/or a project supervisor, a decision which cannot be challenged by the board of directors, an authority that NEA has exercised a number of times. NEA‘s de jure administrative powers are aimed at curtailing the board of directors‘ influence on the day-to-day management of the EC. While the general manager of the EC does report to the board, his or her selection, appointment, and termination are governed by regulations issued by NEA that provide significant limitations on the board. These regulations are explicitly aimed at limiting political influence in personnel management by protecting the general manager from these pressures and ensuring transparent and merit-based recruitment. NEA is not an independent regulator but an agency under the Department of Energyand is governed by a Board of Administrators, made up of a chairman and four members, one of whom is the Administrator, with the chairman and members appointed by the president of the Philippines to serve a term of six years. These institutional arrangements make NEA potentially susceptible to political influence. For example, the president could always compel the NEA board to take specific actions he/she desires. The exact channels by which this influence is exercised are not entirely clear, but it appears that the interests of congressmen and the governor do play an important role in its decision on whether to intervene in a particular cooperative, as some of the case studies below show.2 In some cases it appears that NEA chose not to protect general managers from overly intrusive boards because of the board‘s political connections, and in other cases incompetent and malleable general managers were brought in from outside. Tariff policy is under the purview of the Electricity Regulatory Commission (ERC) — to be precise, the ECs submit their tariff petitions first to NEA for approval before sending them to the ERC. The tariff-setting methodology allows the ECs to recover their cash requirements for power purchase costs, non-power costs (operations and maintenance, administration, and other general expenses), loan amortization and interest payments, reinvestment (5% of total revenues), and system losses up to a cap. This cap has been steadily reduced over the years and is currently at 14%. As these are non-profit organizations, the tariff methodology does not allow for any retention of savings beyond the cap, and any additional loss reduction is required to be passed on to consumers in the form of lower tariffs. In addition to tariff setting, the ERC approval is also required for planned capital expenditures in excess of P500,000. 2 One common tactic used by the members of congress in the Philippines to exert influence over a particular government agency is to use the annual budget authorization process to ―extort‖ concessions. An imaginable scenario is one where a particular member of congress withholds approval of the proposed budget until the NEA leadership agrees to provide assistance to the EC in his constituency. 5 The diversity in performance of the ECs is the defining feature of the sector. Overall system losses of cooperatives have declined significantly over the past twenty years, from a national average of 25 percent in 1987 to 13.5 percent in 2008 (Figure 1, left panel) although the pace of improvement has slowed down of late. The variation around this national average is, however, considerable (Figure 1, right panel), with systems losses varying from a low of less than 7 percent for cooperatives such as Cebu III, Misamis Oriental I, and Bohol I and a high of over 23 percent for Laguna, Camarines Sur I, and Camarines Sur III. Eighteen cooperatives have losses under 10 percent while twelve have losses exceeding 18 percent. Other performance indicators, such as operating margin, repayment to NEA, and collection efficiency, paint a similar picture of great divergence around a generally improving overall national average. Figure 1: System losses over time and across cooperatives System Losses (percent) System losses across cooperatives (2008) 30 29% 25 26% National average, 13.5% 23% 20 20% 15 17% 10 14% 5 11% 8% 0 1987 1990 1993 1996 1999 2002 2005 2008 5% Ilocos Norte Isabela II Tarlac II Laguna Romblon Catanduanes VRESCO Camotes No Samar Misamis Or I So Cotabato I Source: NEA In terms of institutional indicators, overstaffing in cooperatives, a major feature of high administrative costs, has been reduced, with the consumer-to-employee ratio declining from 211 in 1998 to 350 in 2008, but remain high as ideally these ratios should be of the order of 400:1 for a dispersed rural island population and 600:1 for a compact urban population. As Figure 2, left panel shows, this ratio varies from 745 for Davao Sur to a low of 112 for Camiguin, with eight cooperatives with ratios of 200 or lower. Collection efficiency similarly varies greatly around the national average of 88%, a number that has remained stagnant over the past twenty years. That overstaffing has an effect on systems losses is suggested in figure 2, bottom panel, which shows a negative relationship between these losses and the consumer-to-employee ratio. There are geographic and technical factors, in addition to politicization, behind this variation. The political imperative of electrification, which will be discussed in more detail below, meant that the cooperatives network was expanded to connect remote and scattered households that were commercially nonviable. Approximately 90 percent of the 7.8 million customers of the ECs are households, which accounts for the relatively low share in energy consumption, and also the relatively greater difficulty, compared to Meralco, a private utility that serves Metro Manila, for example, in improving collection efficiency. As per the Bank‘s previous sector policy note, roughly a third of the ECs are in non-viable franchise areas, either due to geographical factors such as susceptibility to natural calamities like typhoons or location in small islands or mountainous regions, or have a small, predominantly residential consumer base. The World Bank‘s 2000 study identified political factors as the predominant reason for this diversity in performance in the economically more viable ECs. A number of the high performing ECs are characterized by professional management that has taken major initiatives to reduce 6 systems losses, made efforts to improve customer service while strictly penalizing customers who steal electricity and don‘t pay their bills. They have also made progress in streamlining their organizations by removing redundant staff, and introduced transparency in recruitment. By contrast, many of the poorly performing ECs are characterized by high management turnover, excessive involvement of the Board of Directors and of municipal, provincial, and national political actors in the day-to-day functioning of the EC, serious procurement irregularities, patronage-based employment, and high non-technical systems losses due to theft, often due to the collusion of employees and customers. To put it another way, the ECs‘ performances appear to correlate with the extent to which professional managers are granted sufficient autonomy to run them on technical, managerial grounds. The general managers‘ autonomy in turn is primarily a function of the extent of day-to- day interference by the board members. The boards‘ ability to dictate the ECs‘ management in turn depends on the level of autonomy they themselves enjoy from other local or national politicians. In short, given the highly localized nature of the ECs, the variable patterns of politicization that surrounds them unfold within a nested structure of state capture that is particular to each locality. We will elaborate on this structure below. Figure 2: Key institutional indicators of electricity cooperatives (2008) Consumers per employee (2008) Collection Efficiency 100% 800 700 National average, 350 90% 600 80% 500 70% 400 300 60% 200 50% 100 40% 0 Ilocos Norte Isabela II Tarlac II Laguna Romblon Catanduanes VRESCO Camotes No Samar Misamis Or I So Cotabato I System losses and overstaffing 30% 25% System loss 20% 15% 10% 5% 0% 0 100 200 300 400 500 600 700 800 Consumer/employee Source: NEA 3. The Philippines Political Context: A Review of the Relevant Literature State capture is a common theme of the Philippine political economy. According to a widely accepted perspective, the state operates at the mercy of economic elites (e.g., landed oligarchs) who wield their wealth and influence to steer public policies in their favor (Hutchcroft 1998, 7 Coronel 2004). On the other hand, a revisionist perspective has shown that many elites have acquired wealth and power after they captured access to public resources, for example, through winning elected posts and/or allying with influential national politicians (Sidel 1998). Either as a consequence or a cause of elite dominance, state capture emerges as a main culprit of poor governance and development constraint in any narrative on the Philippine political economy. State capture is pervasive in the Philippine politics because public resources offer attractive sources of rents for power-holders and power-seekers and because there are not sufficient checks on their opportunistic behavior. There is also a broad agreement that in the Philippine politics is intensely local, and this has to do with the dominant mode of political organization in the country, apart from the geography and history which for years has reinforced the natural spatial and cultural fragmentation of the Philippine society. One of the key institutional characteristics of the Philippine democracy is the absence of a well-institutionalized political party with nation-wide presence. Instead of political parties, a family (sometimes including extended kinship, friendship, and other informal personal ties) has become the primary unit of organization for political activities. Since no single family can establish a nation-wide bailiwick, emergence of political clans is inexorably linked to specific localities and their political economies (McCoy 1994). Anecdotes abound of political families in a variety of localities establishing their dominance through their control of key economic activities and access to political power. When some of these families entrench themselves and manage to remain dominant in a particular area spanning generations, they become ―political dynasties‖, a term recognized (and derided) even in the Constitution.3 How these families or clans establish political and economic dominance in a particular locality largely determines the nature of governance and a prospect and a pattern of socioeconomic development in this area. In a highly patronage-based political system such as the Philippines‘, control of wealth (or better yet a good part of the local economy) is an essential means to retain power and influence in a locality. With wealth, local political clans are able to buy votes directly to win elections, organize a local political machine to maintain captive voters, often with armed ―enforcers‖ to intimidate opposition and general voters into submission, and simply accumulate more wealth for personal enrichment and political power-building. Capturing public apparatuses and using their political power and influence to consolidate their hold on local economic activities are among the most common means for wealth accumulation among Filipino political elites. As important as the local dimensions of politics in the Philippines, local politics does not exist and unfold in isolation from the national politics (and vice versa). local politicians‘ varying positions within, and relations with, politicians at higher tiers of the political system are among the variables that determine the nature of local political economy. On the one hand, the president (and presidential aspirants) relies heavily on local politicians to mobilize votes for them because of the absence of party machinery. On the other hand, many elected local politicians rely on the national governments‘ fiscal largesse to supplement their generally scarce fiscal resources.4 local politicians can also benefit from preferential regulatory treatments (e.g., on matters related to land use) and other political favors if they are in good terms with the president who also enjoys vast 3 Article II on Declaration of Principles and State Policies Principles, Section 26 states ―The State shall guarantee equal access to opportunities for public service and prohibit political dynasties as may be defined by law.‖ 4 The Philippine Constitution grants the president with a broad range of budgetary powers and discretion which he/she can, and does, use for the purpose of political alliance making with both members of the House of Representatives who represent particular geographic areas (districts) and local chief executives governing provinces, cities and municipalities. 8 appointive powers to control the national bureaucracy. In addition, local politicians can tap fiscal and regulatory protection of the national government through Congress, often having one of their own family members elected in the national legislature as part of their strategy to secure flow of national funds to their districts and thus solidify their control over local public affairs. Hence gaining and maintaining access to the vast resources of the national government is an essential tool of political survival for most local politicians in the Philippines. Despite the pervasiveness of the political dynasties and their ―elective affinity‖ with state capture and the central importance of local politics, the existing literature does not provide a satisfactory explanation of the variations in the extent and the nature of state capture across different local jurisdictions. It is these variations, however, that are critical for understanding different degrees and patterns of rent-seeking observed at the local level, such as the case of the 119 electricity cooperatives, and the literature is largely silent on this critical question: What are the key characteristics of local political economies that account for the observed variations in the extent of state capture and rent-seeking at the local level? As an exception, de Dios (2007) offers an insightful interpretation of the interactions between the economic endowments of a particular locality and the likely pattern of its political and economic development. In summarizing his argument, de Dios (2007, 154) presents the following schematic depiction of variation in local political economies: 1. Hobbesian situations where contending political clans vie for power with inconclusive or impermanent results will be associated with the worst development outcomes. The absence of order discourages both investment and productive activity owing to the insecurity of property rights; even contending clans themselves can ensure their own economic interests only to the extent of their capabilities for violence. 2. The emergence of a politically dominant clan or strongman is a step forward in terms of attaining order, regardless of whether this takes the form of clientism or warlordism. Where political hegemony is combined with overweening economic dominance, however, the problem of credible commitment affects potential external investors, who fear possible expropriation or harassment; such areas will manifest significant investments directly undertaken by or in close association with the dominant clan itself. 3. Where stable political conditions are guaranteed under a hegemonic political clan but the latter does not play an overweening local economic role, greater room is left for external investors to assume a larger role in the local economy. A minimal signal of commitment to an investment-friendly regime is needed, however, and this notably takes the form of investment in social infrastructure and maintenance of peace and order. Such conditions are attainable even under secular bosses and fiscal brokers. 4. Without such a commitment and effort, the likely result is a regime of petty bosses thriving on parasitic and illegal activities while the local economy stagnates (as seen in some small towns and provinces in Southern Tagalog, for example). 5. The stability of investment-friendly policies is enhanced most when political power becomes less based on the charisma or personal reputation of the local leaders and instead relies more on a civic constituency with a pronounced stake in the provision of local public goods. The few suggestive examples indicate that the class composition of this constituency is less important than the fact that it should be sizeable and influential (though not necessarily the majority population) and conscious of its interests beyond the persona of individual politicians. Depending on the locality‘s social structure, this constituency may emerge from a relatively affluent commercial class (Cebu City), middle-class residents and small businesses (Marikina City) or the marginalized sectors themselves, participating in governance through people‘s organizations (Naga City). It is also at this point that personal political exchange is close to being transcended and the political relationship approaches the prescriptive. 9 Although de Dios‘ analysis focuses on the effects of the nature of the local political economy on investor incentives and the consequent impacts on the area‘s economic development, similar logic applies, at least partially, to the relations between the nature of the local political economy and the local political elite‘s incentive to provide public services, such as universal access to quality electricity distribution. On the one hand, to the extent that the maintenance of power in a particular locality depends on catering to broad segments of the constituencies, the politicians will have a stronger incentive to ensure the EC operates efficiently (e.g., with limited service interruption) and economically (e.g., without having to raise tariff too high). The politician, however, will balance this incentive to cater to the interest of the general population with a counter-incentive to channel benefits to particular constituents or accumulate rents. Appealing to particular constituents at the cost of the general public can be politically functional if the imperfections inherent in the electoral system limit the general voters‘ ability to punish recalcitrant politicians and if those particular constituents are in a position to offer the politician disproportionate benefits in terms of votes, financing or other desirable goods. The politician needs rents for a variety of purposes including financing of his own campaign and related political activities, dispensing political favors to allies and supporters and enriching himself and his families for pure personal gain. Finally, this stylized politician does not exist in a vacuum. He operates within his own network of political allies and enemies, including in the particularity of the center-local relations discussed above. His degree of freedom in either protecting the EC‘s managerial autonomy or raiding it for rent-seeking purposes is conditioned by the disposition of other political actors who may also approach the same EC with their own motives. The next section elaborates on our framework for accounting for variations in patterns of capture across the case study ECs. 4. The Political Economy of Nested Capture: A Framework This section presents a framework for explaining why some ECs are used by political actors for serving their private interests while others are not and are reasonably effective in providing this vital public good. The framework is based on the interests of two sets of actors: (i) local politicians, which are defined as those politicians whose electoral jurisdiction is within an EC‘s catchment area; and (ii) national politicians, defined as those whose electoral jurisdiction is larger than the EC‘s catchment area or, in other words, more than one EC catchment area is within their electoral jurisdiction. Defined in this way, municipal mayors and congressmen will usually be local politicians, while the provincial governors and, in the extreme case, the president will be national politicians. 5 Our argument is that the variations in the levels of politicization of the EC are related to the varying nature of political incentives and nested political relations between the EC, local politicians, and national politicians. Local and national politicians will impact the EC through actors within the EC, namely the board of directors and the general manager. In particular since the board of directors are elected, their political incentives also need to be understood. Therefore, the incentives of the three sets of actors are elaborated below. The Incentives of the Board of Directors Given the institutional structure of ECs as elected bodies, we hypothesize that the ‗average‘ board of directors will be politicized. One would expect board elections to be similar to local government elections in the Philippines, albeit on a smaller scale, with local power brokers playing an important role, and with active canvassing by mayors and congressmen on behalf of 5 The few exceptions include Albay and Palawan which have only one EC in the entire province. 10 their candidates.6 In particular, one would expect the economics of elections would incentivize boards to use their authority for rent-seeking. While board elections are smaller in scale than local government elections, and usually less competitive (voter turnout figures of 20% or less were commonly quoted), anecdotal evidence suggests that candidates spend on average between five hundred thousand and one million pesos in these campaigns. These include the standard expenditures on, for example, transporting cooperative members (i.e., consumers) to the voting booths, as well as vote buying, with a commonly quoted figure of 200 pesos per vote. The fact that voter turnout in board elections is significantly less than that in municipal elections implies that these elections are more susceptible to manipulation as there are fewer votes to buy to determine an outcome. Given the importance of key individuals to deliver votes, the informal contract between a board member and these individuals would involve the delivery of votes in exchange for the rents that the EC can provide. These rents can include lucrative suppliers contracts to politically influential individuals with associated kick-back schemes; recruitment of staff, particularly low-skilled casual employees, sometimes ahead of board elections; and intervention on behalf of consumers to forgive penalties for non-payment of bills, or to prevent disconnection. Some board members may use the EC board as a ―stepping stone‖ for other elected posts (e.g., municipal council) for which the entry barrier for electoral contests is higher (and elections more expensive). These members are expected to use their positions on the board to both build positive reputation among the local constitudents and by building strengthening their campaign ―war chest.‖ Given that an ordinary voter/consumer is not in a position to see direct contribution of an individual board member on the EC‘s performance, the reputation building by a board member is more likely to take the form of addressing specific needs and demands of individual constituents such as a request for deferring bill payments or for employment. From the cases examined for this paper, the frequency of EC board members aspiring for higher elected posts appears to be low. Given the political process by which board members are elected, we expect each EC to have a baseline value of ―a certain degree of politicization.‖ This stems from the electoral incentives of the board of directors combined with weak bottom-up accountability due to imperfections in EC board elections. We a priori expect this to be the norm across all cooperatives. Furthermore, the reality of local elections in the Philippines where local clans, as opposed to nationally-organized political parties, operate as the key organizational unit for political activities means that a candidate for an EC board is unlikely to succeed electorally without support of one of the local clans. Typically, a single board member represents one of the several local governments in the catchment area. Hence obtaining endorsement of the mayor of that local government, or an alternative local political force, is crucial for his/her electoral success. This makes most board members effectively dependent on mayors or other influential political groups in the locality. The Incentives of Local Politicians Local politicians have the choice of either intervening in the functioning of the EC to gain access to EC rents to finance their own political campaigns and to aid in political alliance building (as alliances are built and maintained on the basis of exchanges of favors) or to address EC-wide performance issues that appeal to voters. The former would include brokering employment 6 For an analysis of flaws in the Philippines‘ national elections, see ―CEPPS Philippines Election Observation Program, Strengthening Electoral Process, IFES Final Report.‖ International Foundation for Electoral Systems, August 2004. 11 opportunities for those who seek patronage jobs and rent-seeking in procurement and contracting. The latter includes the following: (i) to keep the tariff low if politicians fear that constituents‘ complaints about the tariff level may somehow affect their political fortune (and this will require that the EC operate at some level of efficiency since the tariff level is capped by the national regulator); (ii) to extend electricity lines to areas still without access; (iii) to address service problems that their constituents may bring to their attention; or (iv) to intervene forcefully to seek more permanent solutions if the performance failure is severe and general across the jurisdiction. We posit that the motivations for rent-seeking will be higher than the concerns about the EC-wide performance issues as the latter approximate the incentive to provide public goods which are typically scarce in a political context dominated by patronage and clientelism. Given the typical EC‘s board structure where a mayor has influence over just one or only a few of several (or as many as a dozen) districts, and given that the politician‘s electoral jurisdiction lies within the EC catchment area, few local politicians have either the incentives or authorities to address EC-wide performance issues. However, since interfering in the EC‘s management for the purposes of rent-seeking involves costs in terms of lower production due to poor electricity supply, local political actors will choose to intervene only when the ―returns‖ are judged to be ―worth the effort.‖ We argue that the returns are determined by two factors: (i) the rent-seeking potential of the EC — the bigger the EC in the local economic context the bigger the available rents and the greater the incentive of local politicians to intervene to capture it; and (ii) the competitiveness of local politics — the more competitive the local elections, and the more intense the political contest between rival clans, the higher the marginal value of the EC as a source of additional rents to swing an election. Relative potential of an EC as a source of rent Our hypothesis is that local politicians will not interfere in ECs unless these provide sufficiently large rents that warrant the transaction costs involved in such interference. The local politicians‘ incentive to interfere is likely to be higher for those ECs which provide a relatively larger source of rent compared to other rent-seeking opportunities in the locality. The larger the EC relative to the size of the local economy (including the size of the fiscal activities of the local governments), other things being equal, the more the rent-seeking opportunities, the greater the outside interference, and the greater the degree of politicization above the baseline value. Cooperatives in economically more developed environments are likely to be relatively small players in terms of the rent seeking opportunities given alternatives as compared to relatively larger cooperatives in more agrarian settings.7 Similarly, in localities where significant illegal economic activities exist, politicians in search of rents may find it more lucrative to control these informal economies than capture the EC. Apart from small local economies, controlling ECs in calamity-prone areas is particularly attractive to politicians as i) calamities create ample opportunities for political visibility, ii) the ECs‘ access to national relief funds provides resources for rents, and iii) rehabilitation efforts can be skewed towards the politicians‘ constituencies. The capturing of the EC by local politicians is done through proxies, either the GM or Board members. How the capturing takes place in turn is a function of the nature of political competition across municipal jurisdictions given that no municipal mayor has the ability to select 7 Typical methods of rent-seeking in the Philippines include preferential access to public works projects through companies owned or controlled by ―local‖ politicians, and these opportunities are expected to be greater in localities with relatively larger economies. 12 (informally) all the candidates for EC Board members who represent different geographic districts. Competition, visibility, and the types of banditry A second factor impacting the incentives of local politicians is the intensity of political competition in the locality. The higher the intensity of political competition, the greater the incentives of politicians to seek additional rents from capturing the EC (as well as any other source of rent) to gain a marginal edge over their political rivals. This intensity of competition is intrinsically linked to the nature of the clan divisions in the area, given the primacy of the clan in Philippine politics. In areas where there are rival and equally powerful clans, then political competition and the battle for economic rents are likely to be intense and hence the higher the temptation to involve the EC (as well as many other sources of rents) in these battles. When these battles play out in the EC, they may translate into severe factionalism or frequent changes in the composition of the board and the constant changes in GMs, hence undermining the EC‘s operational performance. In areas where there is a dominant clan, in contrast, the incentives for interventions in the EC‘s internal affairs are, other things being equal, likely to be less because, by definition, the dominant clans in those areas are more electorally secure and also are likely to hold dominant economic positions within the localities. In these cases, the degrees of rent-seeking through the EC may not be much beyond the baseline value. On the other hand, an EC with a relatively large source of rents in a politically stable area is likely to lead to its monopoly capture by the dominant local clan. In these cases, the dominant clan is expected to act like a ―stationary bandit‖ whose interest is in continuing with some level of rent-seeking while preserving the EC as a long-term source of rent (Olson 1993). The Incentives of National Politicians National politicians in our framework refer to provincial governors, the president and possibly national senators. We argue that since national politicians‘ electoral jurisdictions are larger than the EC catchment area they have less of an incentive to interfere in EC operations for the purposes of rent-seeking. By the same token, their incentive to interfere in EC operations at all is likely to be limited under normal circumstances. However, once they choose to intervene, they are more likely to be motivated to address EC performance issues or to come in aid of their local allies. The EC is a relatively small source of rents for the national politician; however, EC-wide performance failures can become a political liability for him. A typical EC catchment area can contain something like a third of the provincial vote pool and there is a risk that at some point the blame for the EC‘s poor performance is placed on the governor. The likelihood that a higher-tier politician gets involved in an EC matter in turn depends on the political salience of the EC or the EC catchment area for his/her broader political calculation beyond the dynamics of the local politics. The higher the salience, the greater the incentive to intervene. The national politician will most likely work through NEA to intervene in the EC. This NEA intervention could be low-key and routine, as in NEA providing the regular checks to ensure that the EC complies with rules, knowing that it has the implicit backing of the national politician. Or it could be more dramatic if EC performance failures become severe and NEA intervenes and impose corrective measures through its considerable de jure powers. The framework implies that NEA‘s de fact authority — the extent to which it exercises its ‗martial powers‘ — is an outcome that is determined by the political interests of national politicians. Whether or not NEA‘s interventions will be successful depend on the relations and the balance of 13 power between national politicians on the one hand and the local politicians who are behind an EC‘s politicization on the other. The Political Economy of Nested Capture To summarize, three variables determine the incentives of local and national politicians and therefore the level of politicization of an EC beyond the ―baseline level‖ expected of any average EC: the potential of an EC as a source of rent; the competitiveness of local politics, and the political salience of the EC catchment area. How these three variables interact is depicted in Figure 3 below. On the left half of the figure, if the EC is a low source of rent then it is unlikely to attract the interests of local politicians and the extent of politicization will be determined entirely by the EC board. The degree of competitiveness of local politics will not be important as local politicians would not be interested in the EC‘s affairs, and therefore this variable has been excluded from the left half of Figure 3. Given the economics of elections in the Philippines, we would expect some degree of interference by the board, which, if unchecked by national politicians, will result in ―baseline politicization‖ of the EC. This outcome is depicted in the extreme left branch of the figure. On the other hand, if the political salience of the EC catchment area is high and therefore the EC performance matters for the political fortune of the national politician (likely to be either the provincial governor or the national president), then a national politician will be motivated to encourage NEA to intervene in the EC to improve EC performance. These efforts are likely to be successful since there are usually no local politicians who have a stake in the EC and who are powerful enough to resist this intervention. We would therefore expect smaller EC‘s in areas with high political salience in general to have relatively low levels of politicization. The right half of the figure depicts the situation where the EC has high rent potential. In this case local politicians will interfere, and the precise nature of this intervention will be conditioned by the intensity of local political competition. In cases where local political competition is intense, local politicians will have considerable incentive to use the EC as a source of rent. If the EC also happens to be in a catchment area that has low political salience for national politicians, then we will observe a high degree of EC politicization as there will be few external checks on local politicians. These are likely to be the worst performing ECs. In cases where the political salience is high, NEA may intervene, but this intervention is likely to be resisted by local politicians resulting in considerable conflict in the EC. Whether or not these interventions are successful will depend on the balance of power between local and national politicians; NEA‘s interventions are more likely to be successful in EC‘s that are located in areas with less intense political competition and where the stakes are lower for local politicians. This framework implies that the worst performing ECs will be the ones with (i) high rent potential, (ii) in areas with highly competitive local politics, and (iii) in areas with low national salience; and the best performing ones will be those with (i) low rent potential, and (ii) in areas with high national salience. The other combinations should result in intermediate cases with medium levels of politicization. Two limitations of this framework need to be highlighted. First, the framework says nothing about when an intervention by NEA with the backing of a national politician in EC‘s with high political salience will take place. For example, the performance of the EC may have to deteriorate considerably before the intervention occurs and improvements begin. 14 Second, there is one important actor that has not been explicitly included in the framework, namely the general manager (GM). Our position is that it is only when mayors or other local politicians refrain from interfering in the EC‘s day-to-day management for rent-seeking or patronage purposes can the GM have sufficient space and autonomy to manage the EC on purely technical-managerial criteria. However, a good GM can have an independent impact on the probability of politicization that is not captured in our framework. As some of the case studies highlight if, due to accidents of history, a good manager is able to build a reputation then politicians are less likely to interfere in the EC for fear of a backlash. The risk of such negative popular reactions is higher in those ECs which have performed relatively well. In contrast, the potential political cost of interfering in an EC that is already poorly performing is not particularly high because the consumers are already ―used to‖ such a situation and may not have the information to attribute a marginal worsening of the EC‘s performance to the local politicians‘ additional interference. It follows that a part of the political economy of EC may be characterized by virtuous and vicious circles where well-performing ECs discourage political interference while poor performing ECs invite more of it, until the performance problem reaches a crisis proportion. However, to the extent that whether or not an EC gets a good manager is random, this factor is not included in the theoretical framework. 15 Figure 3: The political economy of nested capture Potential of EC as a source of rent Low High Competitiveness of local politics Political salience of catchment area for Low High ―national‖ politician Political salience of Political salience of catchment area for catchment area for Low ―national‖ politician ―national‖ politician High Baseline Low High Low High Politicization: Board dominates Intermediate cases: High politicization: Medium politicization “local” politicians dominate Low politicization: External protection allows EC to be run Indeterminate: professionally Intermediate cases: Confrontation between Medium politicization “local” politicians and NEA 16 5. The Case Studies The objective of the case studies is to illustrate the theoretical framework by examining some of the best performing and worst performing ECs in the Philippines. These cases were deliberately selected on the variables of interest; therefore, by definition, they do not serve as an empirical test of the framework. Rather, the aim of the analysis is to bring out the interests of the actors involved and to provide some flesh to the preceding abstract theoretical discussion. The selected cases are four poorly performing ECs, namely Albay Electric Cooperative (ALECO), Pampanga Electric Cooperative (PELCO) II, Central Negros Electric Cooperative (CENECO), and Sorsogon Electric Cooperative (SORECO) II; and two well-performing ECs, namely Palawan Electric Cooperative (PALECO), and Cebu Electric Cooperative (CEBECO) III. In order to exclude geographic factors, all of these ECs are in viable franchise areas. Table 1 provides some basic data on these six cases. The poor performing ECs are in general characterized by high systems losses, high financial losses, dependence on subsidies from the national budget, and overstaffing; the good performing ones have low systems losses, high collection efficiency, and generally good financial performance. The discussion below summarizes the cases along the variables of interest, namely the extent to which the level of politicization is driven by the hypothesized impact of rent-seeking potential, the competitiveness of local politics, and the political salience of the EC catchment area. Table 1: Basic data on the Cases (2008) CENECO ALECO PELCO II SORECO II PALECO CEBECO III Consumers 138,544 162,924 129,492 74,050 74,125 60,890 MWH sold 505,079 203,873 241,477 48,167 112,266 133,660 Systems loss 11.36% 24.8% 18.1% 20.82% 11.3% 5.2% Consumer/employee 381 325 187 182 372 383 Operating revenue 2,455,496 1,320,320 1,379,865 386,416 924,623 516,440 (000s) Net margin (loss) (104,866) (242,037) (236,543) (29,133) (30,365) 2,296 (000s) CENECO CENECO in the province of Negros Occidental is the second largest EC in the Philippines in terms of power sold. While its systems losses at 11.4% are better than the national average, it can be categorized as a poor performer given its consistently high financial losses despite a favorable and compact franchise area that includes the city of Bacolod with several large industries, and with commercial and industrial customers accounting for half of power sold (which makes it easier for the EC to collect tariff payments). CENECO is characterized by a number of governance failures. A high proportion of the systems losses are non-technical, again due to pilferage and collusion, and reflective of the patronage- based hiring of employees. Numerous instances of procurement irregularities were cited by the interviewees, including the award of a contract for the supply of a transformer to the 3rd lowest bidder (resulting in overpayment of P30million) and against the recommendations of the procurement committee. When the lowest bidder filed a complaint, the Board simply decided to procure another transformer, even though it was not needed. This then triggered NEA‘s intervention to remove a member of the Board and the general manager. 17 NEA‘s interventions however, have not been successful as evidenced by a high turnover of general managers, with 6 in the past 10 years, and an additional two NEA appointed project supervisors. Most recently, a NEA appointed general manager was forced out by the Board due to his investigation of procurement anomalies involving the over-invoicing of gasoline purchases. While we were not privy to all the facts, it appeared that NEA was unwilling or unable to protect this reformist general manager against the opposition of the Board, who ousted him in a split 4-3 vote, and a faction of the local politicians who apparently stood behind the Board. In this episode, opposition by the EC union to some of the efficiency-enhancement measures introduced by the ousted general manager might also have been a contributing factor. local politicians have been actively engaged in CENECO‘s affairs, which is unsurprising given the EC‘s large size and high rent seeking potential. The politics of CENECO are dominated by the politics of Bacolod City. Of CENECO‘s nine districts, five are located within Bacolod City and the remaining four distributed among three other cities each having one board member as its representative and two municipalities which share one member on the board.10 Although the ―normal‖ politics of the Board is definitely an element of the governance failures in CENECO, a more important factor is the nature of the politics in and around Bocolod City, including the fallout from the intense rivalry between the congressman and the city mayor. The two belong to opposing clans whose rivalry has a long history, and is symptomatic of the generally intense political competition in the province. The congressman was in his last term and was allegedly eyeing to run for the city mayor‘s office in the 2010 election whereas the mayor was still in his second term and is eligible for one more re-election.11 The Board is deeply divided between supporters of the congressman and supporters of the mayor. The most visible aspect of this conflict has been the recent cutoff of power of the city hall for non-payment of bills and the retaliatory closure of the CENECO offices by the city for non-payment of taxes. To summarize therefore, this paper‘s theoretical framework would predict that CENECO would be one of the worst performing ECs given its large size and therefore high rent-seeking potential, the intensity of the local political competition in Bacolod city, and the lack of any particular significance of the EC‘s catchment area for national politicians, which may explain NEA‘s repeated failures in trying to impose management changes. ALECO ALECO is one of the largest and worst performing ECs in the Philippines, with systems losses that increased from 21.2% to 24.8%, and collection efficiency that decreased from 94% to 78%, between 2003 and 2007. It is one of the three ECs in Philippines that was under close monitoring by NEA at the time the research was conducted (the other two being PELCO II and PELCO III). Its sustained financial losses resulted in indebtedness to the National Power Corporation (NPC) of P1.5 billion, and to NEA of P0.5 billion, repeated threats of power disconnection by NPC, and finally a nationally brokered management takeover by the NPC in 2008. It is therefore, in contrast to CENECO, an example of poor performing EC in which NEA eventually successfully intervened. 10 This contrasts with the situation in CASURECO II, for example, where Naga City accounts for 30-40% of the consumption in the EC‘s catchment area and yet have only two of the 12-member board seats. 11 As in many other localities, the politics of Bacolod appears to be intensely personal as well. The mayor also told us of personal rivalry between him and the congressman dating back to the 1980s when he first ran for an elected office and defeated the congressman. 18 ALECO shares many of CENECO‘s governance problems. First, its electricity tariff is too low to cover costs even if systems losses are reduced and collection efficiency improved, as the ALECO management never applied to the ERC for an increase in the tariff between 2003 and 2008, in large part due to the Board‘s fear that a tariff increase would have negative electoral ramifications. Second, its collection efficiency is low because of sustained inaction against non- paying and politically protected consumers, most notably the Barangay Power Associations (BAPAs) which are major delinquents but successfully lobby municipal mayors to intervene whenever ALECO issues them disconnection notices. Third, the EC‘s high systems losses have a big non-technical component, primarily pilferage (while these are not measured separately, officials stated that they account for roughly Figure 4: Employment trends and board elections half of the losses), which in turn results in ALECO from collusion between consumers and Number of Employees ALECO linemen. Fourth, patronage-based 650 employment has been a constant feature, in particular of contract staff whose six month Election year 600 terms are repeatedly renewed or who are regularized. While ALECO‘s consumer-to- 550 employee ratio of 325 is in line with the national average, this does not take into 500 account staff on contract — who are not considered employees — who number 450 about 100. It is also interesting to note that 400 there was a surge in employees prior to the 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 board elections of 2007 and 2001 (Figure 4), Source: NEA again indicative of this phenomenon of patronage. Finally, and most significantly, are the alleged cases of procurement fraud that have been reported in the local press, such as over-invoicing for transformers,12 and the delivery of equipment even before the bidding process was completed, resulting in equipment that did not conform to the technical specifications. The EC has been characterized by considerable involvement of the governor and congressman in EC management. ALECO scores very high on the variable ‗rent-seeking potential‘ as it is the only EC supplying the province of Albay, a regional center of the Bicol region with a population of 1.19 million. ALECO‘s catchment area covers 15 municipalities and 3 cities. Given ALECO‘s province-wide reach, this is a case where the governor, by our definition, is also considered a local politician. The fact that Albay is also in a typhoon prone area and therefore a regular recipient of reconstruction and rehabilitation funding from the national government has also increased its rent-seeking potential and generated considerable political interest in its affairs by the governor, congressmen, and, to a lesser extent, mayors. Under the Bicol Calamity Assistance for Rehabilitation (CARE) funds the province received P770 million (out of a national allocation of P9 billion), of which P220 million were for the rehabilitation and/or replacement of dilapidated electric facilities to be administered by NEA and the provincial government, a considerable avenue for patronage and rent-seeking. The governor, and to a lesser extent, the congressmen have also regularly intervened on behalf of ALECO every time the NPC threatened to cut power on account of ALECO‘s debts. ALECO, unlike CENECO, also scores highly on the variable, ‗salience of EC catchment area for national politicians‘. As a regional center, Albay is a politically important province, as it has access to the means of transportation and communication crucial in election campaigns and in 12 The Bicol Chronicle, September 11, 2004 19 vote counting, and is the locus of major government facilities and line agencies, and therefore the levers of resource distribution, which may affect electoral outcomes. The province is also home to a highly influential advisor to former president Arroyo, and his interest in the EC was a major factor behind NEA‘s eventual intervention that resulted in a management contract with NPC. PELCO II Like ALECO, PELCO II has been a sustained poor performer and is under close monitoring by NEA. It is a large EC, and although its customer base is almost entirely residential, it is located in a compact, viable franchise area. It has consistently had systems losses in the twenty percent range, has seen declining collection efficiency, a decreasing consumer-to-employee ratio, and has been incurring huge losses requiring subsidies from the national budget. The governance problems in PELCO II are similar to those of CENECO and ALECO: significant pilferage, collusion between consumers and employees, patronage driven overstaffing, high management turnover (4 GMs in the past 5 years), and considerable procurement irregularities as noted for example in NEA audit reports. The Board has been highly interventionist in the day-to- day management of the EC, authorizing payments to the NPC for power purchase, payments to suppliers, and managing the staff payroll. One interviewee stated that for the Board ―the bottom line is to make the EC a milking cow‖, and the Board‘s culpability in the above-mentioned governance problems was widely cited. The Board is also factionalized between two rival groupings, and there was a case of some of the board directors overstaying their tenures. Inter- clan rivalries have also strained the relationship between the Board and municipal politicians, as evidenced by the retaliatory actions taken between the EC and the Municipal Government of Mabalacat in 2004 involving PELCO II cutting the power supply of the municipal offices on non- payment of bills, and the municipality responding with shutting down the EC offices for non- payment of taxes. PELCO II, being a large EC, has high rent-seeking potential. Local politics is also highly competitive. Therefore, it is not surprising that municipal mayors are involved to a considerable degree in the EC‘s affairs. However, congressmen are surprisingly not involved, which may be explainable by the somewhat unique status of Pampanga as the home province of the former President of the Philippines, and therefore the considerable salience of the EC‘s catchment area to this national politician. The former President‘s interests were to showcase Pampanga as a prosperous province, which implied improving the services of the 4 ECs. Given her national constituency, it would seem reasonable to suppose that this service delivery objective, in terms of cultivating votes, would dominate any rent-seeking motivations that the ECs could support, which are by definition small in the national context, and indeed her behavior suggests as much. The President has been quite proactive in the province, with her interests apparently operationalized in the province by NEA, and with the strong support of the then presidential advisor on rural electrification who also played an important role in sustaining good performance CEBECO III (to be discussed below). As mentioned, in response to the long-standing problems in PELCO II NEA has intervened and brought in a project supervisor and a general manager to improve the performance of the EC. Both these individuals have roots in the former communist underground resistance of the Marcos era, and were chosen precisely because it was assumed that they would be able to withstand the considerable political pressures that were likely to fall on them as they attempted to reform the EC. 20 In sum, PELCO II appears to be a case of an EC that was poor performing due to its high rent- seeking potential and intensity of local competition; however, when Arroyo became president, NEA received the necessary backing from a national politician to intervene in the EC. While NEA‘s interventions are strong, they do leave the Board with some options for retaliation, as exemplified in the removal of the previous NEA appointed general manager by the Board for alleged mismanagement and corruption. It is also too early to tell whether the current reforms are sustainable. SORECO II SORECO II is one of the 7 ECs in the Philippines that are registered with the Cooperatives Development Authority (CDA), and the larger of the two ECs supplying power to Sorsogon province. Given its status as a stock cooperative, it does not fall under NEA‘s jurisdiction, thus potentially depriving national politicians, if they so desire, a channel of influencing the EC. It is a mid-sized cooperative with roughly 56,000 consumers, 64 percent of whom are residential with only 2 percent industrial customers; its rent-seeking potential, therefore, is low. SORECO II is therefore, a good test of the hypothesis of baseline politicization that we would expect in ECs given the economics of elections in the Philippines. The bulk of SORECO II‘s history since its formation in 1977 has been characterized by an intense power struggle between some of the key stakeholders, namely the board of directors, the management, the employee union, and NEA. The Board‘s decision to register with the CDA — in 1996 — was motivated in large part by the desire to control a powerful and long-serving General Manager who had generated a core following from a subset of the employees through the distribution of favors. The registration with the CDA however, sparked management action and union counteraction that led to the creation of two boards of directors and two sets of managers in 1998 resulted in organizational paralysis. Over the past decade however, the board has been dominant in the EC‘s affairs. De jure, in CDA-registered ECs there are few checks and balances on the authority of the Board from NEA. NEA can no longer invoke P.D. 269 and P.D. 1645 and hire or terminate general managers, or suspend the Board, officers, or employees of the EC. NEA however retains financial powers requiring that any application for loans by SORECO II require prior NEA approval. The history of SORECO II in the past decade has been highlighted by the strained relations between the Board and management, with a high turnover of general managers (five in past seven years), and numerous cases of illegal dismissals of managers either pending in the court or resulting in compensatory payments to former general managers. Anecdotal evidence also reveals that the Board has become involved in almost all facets of managerial decision- making within the EC, in contravention to its role as only a policy-making body. These including very micro operational decisions such as the purchase of tee-shirts for linemen, on a decision on whether or not the EC needs to buy a new car, the transfer of employees from one office to another, and, most significantly, the crowding of the pre-qualification, bidding and awards committee (PBAC) with the members of the Board. The Board has also used its powers to increase its per diem allowances five-fold over the past two years. Its small size and limited rent-seeking potential would suggest limited involvement of local politicians in the day-to-day affairs of the EC, and this indeed is the case, in sharp contrast to ALECO, PELCO II, CENECO, and CASURECO II. local politicians do have an interest in extending electrification to all the barangays, and under this pressure SORECO II has extended its lines to many non-viable areas. They have also used Board elections as a means to test their political machine or to demonstrate their political strength by way of preview for the more 21 important municipal elections. However, this is short of the detailed interventions seen in the other cases. CEBECO III CEBECO III has consistently been the best performing EC in the Philippines, and the recipient of numerous awards and recognition from NEA. It is a medium-sized cooperative, with residential, commercial, and industrial/large load consumers making up 50%, 25%, and 11% of sales respectively. At 5.2% for 2008, it has the lowest systems losses in the country, has maintained 100% collection efficiency, and has been consistently returning a profit. The other two ECs in the province, CEBECO I and CEBECO II, are also good performers with single digit systems losses, and there is a high level of coordination and cooperation among all three ECs. Indeed, from a management and corporate governance perspective, as will be elaborated below, the three can be viewed as a single entity. CEBECO III‘s excellent technical and financial performance is matched by an equally impressive corporate culture. There is a strong sense of service delivery in the EC with rapid response teams that are mobilized in the event of power breakdowns or other customer complaints, high staff morale and a strict work ethic. Indiscipline and corruption are clearly not tolerated, and customers who do not pay their bills have their connection promptly cut off. There is also apparent harmony between the management and the board, and no hint of the factional rivalries or the governance problems that are so evident in the poor performing ECs. The key ingredient to the CEBECOs‘ success is the dominant role played by a politically powerful patron who has been intent to make these ECs a role model for the rest of the country. A former presidential adviser on rural electrification, and who hails from Cebu, effectively runs the three ECs and the corporate culture of these organizations is his creation. It would not be an exaggeration to state that the staff of the ECs have a religious devotion to this presidential adviser , who has a high level of respect in Cebu as a priest. All key decisions require his blessing, including commercial negotiations, large procurements, and the hiring of top-level management. Discussions with the business community and the mayor of Toledo city suggest that it is the combination of this presidential adviser‘s social and political influence plus his demonstrated ability as a good manager that has made the CEBECOs largely immune from negative external political interference. This is not to say that such pressures are not exerted from time to time. In the early 1990s, the factional rivalries between the mayor and vice mayor of Toledo city had an impact on the CEBECO III board, and politicians do try to field candidates for the Board elections. However, the fact that the EC is high performing combined with the presidential adviser‘s considerable political influence serve as an effective deterrent against politicians. For a politician to be exposed as meddling with the affairs of these well-performing ECs would be politically costly. Instead it appears that other avenues of rent-seeking were being pursued, in particular around the environmental and labor regulations around the mining operations in the area and control of the water districts. This relative lack of involvement is also helped by the fact that the CEBECOs offer limited avenues for rent-seeking as compared to the alternatives, such as from the environmental and labor regulations around the mining operations in the area and control of the water districts. Cebu Province also has a fairly diversified economy with the Chinese Filipinos, who are known for not getting directly involved in politics, dominating a lion's share (―70%‖) of the local economy. The broader political culture of the province also appears to be less disruptive, as the intensity of competition between rival clans is less pronounced. 22 Many interviewees believe that the good management of the CEBECOs has set off a virtuous circle with politicians deterred from interfering and disrupting management, thus risking the wrath of the media and voters, and this lack of interference in turn allowing professional managers to prosper. There was confidence that the CEBECOs would continue to perform well even after the presidential adviser‘s departure due both to these reputational effects and the relative political unimportance of the EC within the province. PALECO Palawan Electric Cooperative (PALECO) is a medium-sized cooperative that is the sole distribution utility for the island of Palawan. Its customer base is primarily residential and commercial, and although the overall franchise area is viable, its 11% systems losses, and consumer-to-employee ratio of 372, indicate reasonably strong management, given the challenges posed in supplying power to the more remote municipalities scattered over the area of the island. The EC had, until very recently, been recovering its costs (as stated, the tariff methodology does not allow for significant profits), and steadily improving its consumer-to- employee ratio. In August 2006 PALECO registered with the CDA and is now a stock cooperative. On governance, PALECO also scores generally positively. The current general manager has been in his position since 1994, and is one of the longest-serving managers in the country. Under his leadership the EC has taken strong measures to reduce non-technical losses by disciplining staff caught colluding with customers and has gone after delinquent customers by cutting off their connections for non-payment of bills. No significant procurement irregularities were brought to our attention in interviews with the local chamber of commerce. In general the picture that emerged was of a general manager that dominated the proceedings of the EC, and a Board that was subservient to his authority, a surprising finding given the de jure authority that the Board has in CDA-registered cooperatives, as exemplified by the case of SORECO II. While some of PALECO‘s Board members come from political families, interviewees stated that the Board was less politicized than in other ECs. Indeed after the conversion to the stock cooperative, the Board has continued to abide by the NEA guidelines on the selection of the general manager to ensure merit and transparency in the process. The Board has also maintained the general manager‘s administrative authority over line departments, and does not overrule the decisions of the bids and awards committee on procurement decisions, again in sharp contrast to SORECO II. While a priori one would expect the rent-seeking potential of PALECO to be quite high given that it is the sole distribution utility for the island, it appeared that politicians‘ attention was more focused on the money to be made through illegal fishing. One respondent stated that this could provide as much as P10-15 million of campaign funds, far larger than what PALECO could net. A number of interviewees also stated that politics in Palawan were less intense than in Bicol, for example, one reason perhaps being that Palawan is outside the typhoon belt and therefore is not a recipient of the calamity relief funds. Nonetheless, the lack of interference in the EC is surprising given its large size in the local context. 23 Summing up Figure 5: sums up how the cases conform to the theoretical framework set out in Section 4. The sharpest distinctions suggested by our framework should be in the cases along branches (2) and (5) of the figure. Our framework suggests, as backed up by the case study, that CENECO would be highly politicized as it has high rent-seeking potential, is located in an area with intense political competition, but lacks national salience to enable a successful NEA intervention. PELCO II, in the pre-Arroyo period, would also fit this profile. This is indeed the case. By contrast, CEBECO III would be expected to be the least politicized as the EC is a relatively small player in the local economic context — the licensing regime around mining in Toledo appears to be the focal point of political attention rather than the potential rents from the mid-sized EC — and has high political salience for a powerful national politician.13 ALECO and PELCO II (in the Arroyo period) are example of branch (6). ALECO is a large EC in a typhoon-probe area with high rent-seeking potential; and the fact that it is also the only EC in a politically important province also gives it high salience for national politicians. ALECO is also interesting as it reveals that the interest of politicians in the EC only became prominent after the three original cooperatives were merged, thereby increasing its rent-seeking potential. Indeed part of the motivation for the merger was precisely to increase the political usage of the EC. Therefore, our framework would suggest considerable intervention by local politicians but also a high likelihood of successful NEA intervention, predictions that are in general supported by the case study. Given the current NEA intervention, we can code its level of politicization as ‗medium‘. SORECO II is a case of baseline politicization as predicted by the framework. It has elicited little interest from congressmen and mayors, despite the absence of NEA, because of its relatively small size, and therefore lower rent-seeking potential. Instead, the board has been unchecked in its pursuit of rent-seeking. PALECO is the one EC that does not conform particularly well to the framework as given its large size one would expect relatively high rent-seeking potential which — coupled with its lack of national political salience and lack of protection from NEA — given its status as a stock cooperative — would suggest higher politicization than is the actual case. However, PALECO is a good example of how a strong general manager can build a reputation and deter politicians from interfering in the EC for fear of a backlash, a feature that was discussed in Section 4 but not included in the framework. 13 At least for the period when the field work was conducted, the mayor of Toledo was a relative of the provincial governor and as such, she may have needed less rent-seeking on her own as she could probably count on political and financial backing of the governor, who herself had ―inherited‖ the post from her own father. 24 Figure 5: The predicted location of the cases Potential of EC as a source of rent Low High Competitiveness of local politics Political salience of catchment area for Low High ―national‖ politician Political salience of Political salience of catchment area for catchment area for Low ―national‖ politician ―national‖ politician High Baseline Low High Low High politicization: SORECO II (1) Medium politicization: High politicization: PALECO CENECO; PELCO II (pre-Arroyo) (6) Low politicization: (5) CEBECO III (2) Indeterminate: PELCO II; ALECO 25 6. Implications Political Economy Profiling of ECs The political economy framework presented here can be used as a means for profiling ECs and for assessing the realism of technical reforms proposed for a particular EC. It can be used to identify in which ECs the level of politicization is expected to be above or below the baseline value based on the three variables of interest. In theory, all ECs can be ―scored‖ along these three variables, though, realistically, this scoring will require a fair amount of subjective assessment. This scoring can give an indication of where reforms are likely to work and where they are likely to fail, thus providing a much-needed complement to technical analyses. Reforms unlikely to work The risk profiling aside, the analysis does point to certain reforms and institutional arrangements that are unlikely to work in most circumstances. These are as follows: Conversion to stock cooperatives Conversion of ECs to stock cooperatives is viewed by many reformers as a necessary interim step towards conversion to stock corporations. However, this paper underlines the risks with this strategy. In stock cooperatives registered with the CDA, only consumers/owners with an equity share are allowed to vote in Board elections or in the annual general members meetings. Importantly, each shareholder is allowed only one vote, irrespective of the size of their equity. For PALECO for example, this implied that the voting population has been reduced from 18,000 to approximately 1200 after the conversion. The positive aspect of this restriction is that the member voters who have contributed money are more likely to participate in elections and in annual meetings; the negative aspect is that it is less costly for politicians to capture elections as there are fewer voters to buy off and their identities are easily known. In balance, one would not expect the conversion to stock cooperatives to change the fundamental institutional structure of ECs as elected bodies. Unlike non-stock cooperatives, stock cooperatives are not subjected to the checks and balances provided, at least in theory, by NEA, and therefore do not provide national politicians with a vehicle to intervene positively in the EC‘s affairs. Therefore, we would, ceteris paribus, expect higher levels of politicization in these ECs. The example of SORECO II clearly provides evidence to this proposition. More stringent qualifications for board members In discussions, general managers, businessmen, and mayors were unanimous in recommending that more stringent qualification criteria and exclusion clauses be introduced in order to reduce Board politicization. These included preventing board members from contesting other elected offices for a certain period, requirements for educational qualifications and professional background, and reservation of certain proportion of Board seats for representatives of the business community and consumer advocacy organizations. Given the primacy of clan politics in the Philippines, we are skeptical that any introduction of qualification criteria or exclusion clauses would fundamentally alter the dynamics of electoral politics, and hence Board politicization. The experience of the three-term limit rule is evidence to this, with mayors and congressmen bringing in family members to stand in on their behalf during 26 the mandatory waiting period. Similarly, minority representation for consumer groups in the Board is unlikely to have much impact, and we can foresee that such a representation could get easily marginalized from the major decisions taken by the Board. Mergers of ECs NEA has been advocating the merger and consolidation of ECs to achieve economies of scale and to attract private sector interest. Indeed many ECs are small and economically unviable, and consolidation is necessary on technical grounds. However, the political economy framework alerts us to the risks entailed in increasing the size and therefore the rent-seeking potential of a cooperative. ALECO is a case in point which after the merger became much more attractive to politicians. ALECO also demonstrates the moral hazard problem with large ECs that may be deemed too big to fail. Greater NEA independence Greater regulatory independence for NEA is a standard technical prescription. While NEA is not a regulator (the ERC is) in the strict definition of the term, it does have considerable supervisory/regulatory powers and therefore should have the same governance structure as a regulator. In theory, effective regulation requires that the regulatory regime be established by law, with the role of the regulators, the mode of their appointment, service conditions, powers, immunities, and responsibilities defined in law and implemented in a transparent manner. The regulatory budget should be independent, and funding should be secure. All regulatory procedures should be well defined and widely publicized, and the scope of government intervention in regulatory processes should be clearly specified by law. Regulatory decisions should be thoroughly explained and made after a transparent and participative process that is open to all stakeholders and decisions should be subject to appeal before an appellate forum. Our analytical framework however, suggests that given the political economy of nested capture, NEA‘s interventions depend critically on the support of national politicians for their success, and giving NEA independence in this political context would potentially deny it this necessary support. An independent regulator can only be effective if the overall political economy context provides a generally supportive environment; where this environment is not supportive, as in the Philippines, regulatory independence is unlikely to make a difference and is probably not sustainable. Reforms likely to work Institutional innovations in the less politicized ECs Given that ECs are a suboptimal institutional design, conversion to stock corporations and eventual privatization is the right strategy for the sector in a long run. One step in this direction has been the attempted contracting out of cooperative management to the private sector. These so-called Investment Management Contracts (IMCs) were conceived for cooperatives that could be turned around with management expertise and relatively low investment requirement from the private sector, and not those that required a substantial capital investment. The program however, got derailed due to a variety of reasons, in particular political resistance from vested interests. For example, one attempted IMC, in PELCO III, was terminated because of violent protests by the employees of the cooperative who feared downsizing, and by political interference from the concerned mayors who were not paying their electricity bills. Another in CASURECO IV failed because the EC Board was highly factionalized with members representing the different political 27 clans in the province and the private party, Salcon Philippines, was unable to secure the support of rival factions. Our recommendation is that the choice of ECs in which IMCs are attempted should be informed by the political economy framework in addition to the technical criteria used to date. Attempted IMCs in highly politicized ECs are likely to fail. However, IMCs can and should be attempted in relatively less politicized ECs — ones that have relatively low rent-seeking potential, are in an area that has relatively less intense local political competition and has high national political salience. This would mean, somewhat ironically, that the IMC model is not recommended to situations where an intervention is most badly needed. It is highly probable that the same political economy factors that have created the performance problems in the first place would undermine the attempted IMC interventions in those worst-performing ECs – here the usual dictum of the political economy of reform that a crisis creates incentives to reform is unlikely to hold. Therefore, a management intervention may be more suitable for those ECs whose performance is at the ―mid-range‖ of the performance variation, although as stated above, specific underlying conditions should be evaluated before a decision to apply an IMC is taken. Now there is a risk that the very act of introducing an IMC will increase the rent-seeking potential and salience of the EC as politicians can use the contract negotiations as a means of extracting rents. Nevertheless, with the appropriate safeguards, IMCs in such select ECs can have important demonstration effects for the eventual institutional transformation of the sector. Increasing consumer voice Although some civil society actors who advocate ―popular democracy‖ and work with cooperatives in other sectors tend to see ECs as a variant of cooperatives which should be run on the basis of the ―voices of the popular sector‖ (consumers, in this case), evidence of effective consumer activism is limited. This is understandable given the collective action problem expected in the nature of the electricity distribution service. The business sector, by virtue of being organized, may be a partial answer to this problem. Polls have shown that the unreliable and insufficient supply of power at reasonable cost is a major perceived obstacle to improving the investment climate in the Philippines. This sentiment was echoed by some of the business sector representatives we interviewed for this study. But here again, there are certain built-in obstacles which may be difficult to overcome, at least in some localities. First, in a large number of small ECs, the weight of the business sector (commercial and industrial users) among the EC clientele is limited, and some of the large businesses (e.g., a large mining company in Toledo City, in the catchment area of the ―best-managed‖ CEBECO III) supply their own electricity or rely on their own back-up generators (e.g., a large sugar mill operator in Bacolod). Second, given the nature of the Philippine political economy often characterized with local factionalism cutting across not only political-partisan but also economic-business lines, mobilization of the business sector could simply amplify the political divide that may be undermining the EC‘s governance and management. Nevertheless, increasing consumer voice through the creation of advisory and supervisory fora with participation of sub-groups of consumers with the greatest stakes in improved services, notably the business community, can provide some of the much-needed checks and balances and improve the governance of ECs. Palawan‘s power commission discussed earlier is a promising initiative that also has the support of the Philippine Chamber of Commerce and Industry who are viewing it as a pilot which can be replicated in other ECs. Our finding is that such an approach is likely to be promising in some ECs but not in most. 28 References Coronel, Sheila, et al. 2004. The Rulemakers: How the Wealthy and Well-born Dominate Congress. Philippine Center for Investigative Journalism. De Dios, Emanuel S. 2007. ―Local Politics and Local Economy,‖ in Arsenio M. Baliscan and Hul Hill (eds.) Dynamics of Regional Development: The Philippines in East Asia. Edgard Elgar Publishing. Hutchcroft, Paul D. 1998. Booty Capitalism: The Politics of Banking in the Philippines. Cornell: Cornell University Press. Lal, Samir, 2006. Can Good Economics Ever Be Good Politics? Case Study of India’s Power Sector. World Bank Working Paper No. 83 McCoy, Afred M., (ed.) An Anarchy of Families: State and Families in the Philippines. Madison: University of Wisconsin Press. Olson, Mancur 1993. ―Dictatorship, Democracy, and Development.‖ American Political Science Review 87(3): 567-576. Shirley, Mary M. (ed.). 2002. Thirsting for Efficiency: The Economics and Politics of Urban Water System Reform. Oxford. Elsevier Science/World Bank Sidel, John. 1999. Capital, Coercion and Crime: Bossism in the Philippines. Palo Alto: Stanford University Press. World Bank. 2000. Philippines: Rural Power Sector Policy Note. 29