36586 W O R L D B A N K W O R K I N G P A P E R N O . 8 3 Can Good Economics Ever Be Good Politics? Case Study of India's Power Sector Sumir Lal THE WORLD BANK W O R L D B A N K W O R K I N G P A P E R N O . 8 3 Can Good Economics Ever Be Good Politics? Case Study of the Power Sector in India Sumir Lal THE WORLD BANK Washington, D.C. Copyright © 2006 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First Printing: June 2006 printed on recycled paper 1 2 3 4 5 09 08 07 06 World Bank Working Papers are published to communicate the results of the Bank's work to the devel- opment community with the least possible delay.The manuscript of this paper therefore has not been pre- pared in accordance with the procedures appropriate to formally-edited texts. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the International Bank for Reconstruction and Development/The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the govern- ments they represent. TheWorld Bank does not guarantee the accuracy of the data included in this work.The boundaries,col- ors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank of the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted.Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission promptly to reproduce portions of the work. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, Tel: 978-750-8400, Fax: 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, Fax: 202-522-2422, email: pubrights@worldbank.org. ISBN-10: 0-8213-6680-7 ISBN-13: 978-0-8213-6680-6 eISBN: 0-8213-6681-5 ISSN: 1726-5878 DOI: 10.1596/978-0-8213-6680-6 Sumir Lal is Senior External Affairs Officer in the New Delhi office of the World Bank. Library of Congress Cataloging-in-Publication Data has been requested. Contents Foreword v Introduction 1 Power Reform in India: Issues, Perceptions, and Players 2 Politics in India: Form and Substance 14 The Politics of Reform: Style and Strategy 17 Prospects for the Power Sector 22 Conclusions and Lessons Going Forward 24 References 25 iii Foreword R eform programs all over the world often come to grief because of the lack of sustained political will.A reform strategy might be technically and economically sound and the intended beneficiaries many, but implementation often stalls because of lack of sustained political commitment to reform. In recent years, the power sector in several developing countries has suffered from this frustrating gap between strong, pro-reform rhetoric at the political level and weak, hesitant implementation of the reform measures on the ground. This paper takes a fresh look at the problem of the "rhetoric-implementation gap" by taking the lack of political will as its starting point, and analyzing its causes in the current context of India. Though focused specifically on the recent experience of power sector reform in India,its observations and conclusions will be of interest to those advocating sim- ilar reform programs,in democratic environments,in developing countries across the world. Working from the assumption that people and institutions are not impartial but rather respond to political and economic incentives, this paper examines the incentives, informal relationships,andintereststhatgovernthebehaviorof keystakeholdersindifferingtimesand circumstances, and searches for the openings and opportunities that reformers must pursue if they are to obtain support for their initiatives. The paper concludes that political will for reform can be created if reformers first dis- play flexibility that will allow governing politicians to negotiate options and mediate con- flict among the multiple constituencies they serve. The reformers'technical, regulatory and implementation formulae can work only if the right incentives for those who must imple- ment the reform are first developed in this manner. Or, stated differently, the ultimate suc- cess of any reform program requires that there be an explicit melding of the"political"with the"technical."If these two tracks are not joined together,the proposed reforms are not likely to succeed. Praful Patel Vice President, South Asia Region April 2006 v Can Good Economics Ever Be Good Politics? Case Study of the India Power Sector Introduction1 Reform programs all over the world often falter for reasons that mystify those devising and supporting the reform, because governing politicians stall in implementing the measures fully, though they may have publicly agreed with their formulation--and despite the reform being technically and economically sound, with many intended beneficiaries. In recent years, the power sector in several developing countries has suffered from this frus- trating gap between strong, pro-reform rhetoric at the political level and weak, hesitant implementation of the reform measures on the ground. Focusing on the recent experience of power sector reform in India, this paper looks afresh at the problem of the "rhetoric- implementation gap" by taking the lack of political will as its starting point, and identify- ing the ingredients that comprise it in the current context of India. This paper explains how the lack of political will to follow through on announced reforms often reflects rational political behavior. Using this more realistic framework, it examinestheincentives,informalrelationshipsandintereststhatgovernthebehaviorofpeo- ple and institutions in different times and circumstances, and searches for the openings and opportunities that reformers must pursue if they are to obtain support for their initiatives. 1. This paper was first written in 2003 as an informal discussion note for colleagues in the World Bank. A shorter version was subsequently published under this same title in the Economic and Political Weekly, Mumbai, India (Vol XL, No 7, 12 February 2005). This is a marginally updated version. The author wishes to thank Penelope Brooke, Ian Alexander, Bhavna Bhatia, Bernard Tenenbaum, Salman Zaheer, and Ashish Khanna, who commented on different versions of this paper, and to numerous oth- ers for their support and encouragement. The findings, interpretations, and conclusions in this paper are the author's alone and should not be attributed to the World Bank, its affiliated organizations, members of its board, or the countries they represent. 1 2 World Bank Working Paper Material for this paper was collected through interviews with policymakers, academics, consumerandfarmactivists,politicalstrategists,andpoliticalscientists.Informationwasalso derived from select reading of academic articles and operational documents related to power reform in India, and from the considerable theoretical literature on the political economy of reform. This paper does not approach the subject through any orthodox theoretical frame- work and confines itself to a practical analysis based on the particularities of India. In its attempt to present a national overview, this paper tends towards generalized observations. The author acknowledges that, given India's diversities, for every comment made here, an exception is likely. Roadmap This paper has four sections: I The first section identifies the issues, perceptions, and players prevalent in the power sector in India and analyses causes for the gap between reform policy and its implementation; I The second section describes the key structural features of the Indian state and aspects of India's political culture that influence reform implementation in general; I The third section identifies the preferred political style and strategy for reform implementation in India; and I Thefourthsectionextractsthelessonsapplicabletothepowersectorandassesses the prospects for further reform. Power Reform in India: Issues, Perceptions, and Players The situation of the power sector in India may be briefly described as follows: the public sec- tor controls about 90 percent of generation, almost all of transmission, and, except for a few pockets, most of distribution. Federal utilities account for a proportion of generation and transmission, but most of the sector is in the hands of state electricity boards (SEBs), which are vertically integrated utilities owned by the respective state governments. Most problems plaguing the sector are at the state level.2 Politically captured, the sector suffers from huge deficits (cumulatively estimated at 1.5 percent of gross domestic product and contributing greatly to fiscal crisis in India's states), unsustainable and regressive subsi- dies (particularly benefiting large farmers), and large-scale theft (more than a third of gener- 2. India, a federal democracy, has adapted the Westminster model with an elected parliament and a government led by a prime minister at the federal level and a separately elected assembly and government led by a chief minister in each state. District and municipal bodies form a still-evolving third, local tier of elected government. There is thus a large political class with diverse interests. The Constitution defines the subjects under the jurisdiction of each tier. Electricity is on the "Concurrent List," which contains sub- jects on which both the federal government ("the center") and the states have jurisdiction. Central law prevails in concurrent subjects, giving the center a lead policy-setting role. The center-state dynamic in the power sector is discussed in the subsection on federalism. Can Good Economics Ever Be Good Politics? 3 ation). The utilities are inefficient and overstaffed, with corruption and political interference at all levels. Electricity supply is erratic and of poor quality. To top it all, nearly three-fourths of poor households, especially in rural areas, have no access to electricity. Reform started in 1991 with the liberalization of generation, but many independent power producers (IPPs) subsequently withdrew. The focus of power sector reform has since shifted to distribution, the area where most of the dysfunction lies. Since 1993, the World Bank has played a role in helping place power reform on India's policy agenda. Its formula of unbundling SEBs into separate entities for generation, transmission, and distribution, corporatizing these, and looking at privatization of distribution as an option, while improving governance and the regulatory environment, has been attempted in six states.3 Power sector reform has thus been firmly established at or near the top of the public policy agenda in India for more than a decade. Despite opposition and controversy, it has never been jettisoned, though models, prescriptions, and content might have changed. Most significantly, by now almost all states--governed, among them, by all the major political parties--are pursuing "reform" programs with the federal government holding out incen- tives for them to stay the course. The key features that characterize power sector reform as played out in India so far are: I Recognitionbypartiesinpoweratthecenterandinvariousstatesthatsuchreform is needed; I Consumer and civil society dissatisfaction with the present situation in the power sector; and I A gap between this apparent appreciation of the need for reform and actual imple- mentationofreformmeasures,particularlythesetrelatingtoprivatization,anti-theft measures, and tariff rationalization, with the agriculture sector proving a formidable roadblock. Explaining this gap is, in essence, the theme of this paper. The Rhetoric-Implementation Gap Politicalendorsementforpowersectorreformhascomeregularlyandincreasinglyassertively from the prime minister, finance minister, and state chief ministers belonging to all par- ties at significant forums ranging from parliament to the National Development Council (a venue where the prime minister and all chief ministers discuss development priorities in a non-partisan manner). Until recently, newly elected chief ministers who won elections on populist slogans have also taken up the reform mantra on assuming office, though this 3. The six are Andhra Pradesh, Haryana, Karnataka, Uttar Pradesh, Orissa, and Rajasthan. A sev- enth, Delhi, has gone this route without World Bank assistance. For an analysis of the technical, eco- nomic, and regulatory dimensions of the July 2002 privatization of the Delhi distribution system, see Agarwal, Alexander, and Tenenbaum (2003). After the reform-enabling Electricity Act was passed in 2003, a number of other states like Maharashtra and Gujarat have also taken the route of unbundling and corporatization. 4 World Bank Working Paper trend has apparently halted after the latest round of national and state elections in May 2004.4 The usual explanation for ruling politicians adopting reform policies is pragmatism: the fiscal distress that most states are in compels them to give priority to power reform in par- ticular, as this sector is where most of the state's financial hemorrhaging occurs. This expla- nation is correct to the extent that it is a starting point for changing the politician's behavior (from populist to reformist), but it does not tell the whole story. The key to understanding the gap between the Indian politicians' reformist rhetoric at the policy level and their failure to follow through lies in recognizing that they speak simultaneously to two audiences. These are positioned back-to-back and rarely see or hear each other.5 The first audience, which has grown increasingly important in the last decade, is the pol- icy and financial elite, with both a domestic metropolitan component and an international donor and investor component. In order to appease this audience6 and attract its endorse- mentandfunds,thepoliticianassumestheroleofthereformerandadoptsthepolicyrhetoric of the day. That is, as the key to accessing development loans and investments that politicians can crow about, "reform" is the expedient position for any party once it is in power. Because political parties in India are unconstrained by any fixed economic ideology, all ruling parties adopt the reform rhetoric. (By the same token, because reform also requires some pain, opposing it comes easily to all parties out of power.) The politician's second audience is his or her political constituency, which he/she must protect and nurture. The actions the politician pursues (or does not pursue) in respect to this second audience are usually dismissed as "populist," but they encompass both legitimate attempts to address grassroots concerns, and distorted attempts to placate swing-voter cate- gories such as big farmers.7 Reform stalls because of the gap between the two audiences; that is,whenaspectsofthepolicyacceptedbythe"reformist"politicianconflictwithactualgrass- roots priorities either in content or sequence, or when they hurt a crucial voting bloc. Power reformhasrunupagainstbothlegitimateanddistorted"populism":highertariffsareunpop- ular with all consumer groups and thus politically infeasible if not accompanied or preceded by improved utility performance; and a combination of the structural problems in Indian agriculture and political expediency inhibits politicians from taking tough measures, even against the "pampered" class of big, subsidized farmers. A question could arise as to why even politicians who have broken from the traditional mould of caste and clan patrons to position themselves as reformers should also move slowly on certain critical reforms, especially in the power sector. (Reforming states led by new age 4. J. Jayalalitha of the regional party, All India Anna Dravida Munnetra Kazhagam (AIADMK), in Tamil Nadu (elected in 2001) and Amarinder Singh of Congress in Punjab (elected in 2002) were the lat- est examples of chief ministers turning reformers on assuming office, but after May 2004, power reform has, temporarily in the author's view, become politically unfashionable inducing both these chief minis- ters to reverse their initial reform measures. 5. Expression used by political scientist Yogendra Yadav, in an interview with the author, New Delhi, December 2002. 6. For governing politicians in states, this audience includes the federal power ministry bureaucracy which both influences the channeling of vast reform-linked funds to the states from central financing bod- ies and attempts to ensure that states are adhering to reforms envisaged in the Electricity Act 2003. The federal Power Finance Corporation and Rural Electrification Corporation now provide the bulk of funds for distribution reforms and rural electrification, and increasingly for generation investments. 7. The term "grassroots" in this paper denotes stakeholders in their primary role as electricity con- sumers and voters. Can Good Economics Ever Be Good Politics? 5 chief ministers of this type have typically made commendable progress on aspects of power reform such as billing collection and theft control, but have stalled on aspects that directly affect farmers like metering, supply regulation, and tariffs.) The key distinction here is that the new breed of politician is not committed to reform per se, but to the political aspiration ofbeingseenasamodernizer.Reformisanecessarytooltoachievethis,butitispursuedonly within the limits of what is politically feasible. Typical of this group of politicians is the adoption of a corporate persona and pursuit of high-profile projects, such as hi-tech cities, e-governance showpieces, and state-of-the-art airports that establish their modern credentials. Meanwhile, actual structural reform of var- ious sectors proceeds only at a meandering, politically manageable pace.8 This highlights a crucial nuance: the word "reform" carries a different meaning and connotation for the tech- nocrat who devised it and the politician who implements it. For the former, it is a particular, economicallysoundformula;forthelatter,itisaspectsandmodificationsoftheformulathat are marketable to political constituents. In their dialogue, the technocrat and politician thus often speak past each other. The audience gap could arguably be bridged (1) through strategic communication campaigns that would demonstrate to political constituents the link between reform and public benefit and (2) by executing the necessary improvements in power supply to make rational tariffs more acceptable. Most reforming states do have some form of communi- cation strategy in place and several can justifiably claim to have improved both quality and service relative to the pre-reform period. The challenge is that absolute improvement--an uninterrupted and stable supply--will still take some years due to the utilities' investment constraints and implementation capacity. It is in calling for patience that the reforming politician runs into the most vital factor undermining reform's acceptability: the lack of credibility of both the reform program and theelectricityutilities.Further,attemptstohastenperformanceimprovementareobstructed by powerful forces within stakeholder groups, such as corruption among staff and inertia among bureaucrats. Inshort,mostpoliticiansadoptreformpolicyandrhetoric,whetherinthepowerorother sectors, out of compulsion, and to position themselves as modern and forward-looking. After being elected, they balance implementation against pushback from five distinct factors: legit- imate grassroots concerns about the impact and sequencing of the reform; a credibility gap that leaves audiences unconvinced of the necessity of the reform; the need to placate certain voting blocs; obstruction by interest groups with a stake in the present system; and an inert bureaucracy. We discuss these factors in turn and then examine political will in light of that discussion. Grassroots Perceptions and Credibility The civil society broadly agrees on the necessity of power reform,9 but maintains signifi- cant resistance to certain aspects of it, so criticism of the reform program continues in the 8. An interesting take on this aspect of reform can be found in Pani, Narendar, "Icons and Reform Politics: The Case of S. M. Krishna." Forthcoming monograph (2002). 9. This conclusion is based on feedback from interviews spanning an ideological and class spectrum. This section attempts to enumerate issues and perceptions that carry political weight without either endorsing or rejecting them. 6 World Bank Working Paper public discourse. Broadly speaking, reforms that find wide, in-principle endorsement are: insulating state power utilities from political interference by keeping them at arm's length from government, establishing an independent regulator which actively seeks to enhance public participation in the regulatory process, and introducing transparent accounting mechanisms. On the other hand, key areas of dispute are removal of subsidies and cross- subsidies, unbundling, and privatization. Immediately apparent are (1) the consensus that the sector must be better governed and made more efficient and (2) the dispute over the appropriate way to do so. At the same time, experience shows that state governments have not been particularly enthusiastic about tack- ling even the undisputed, governance-related issues. We can thus distinguish those aspects of the reform that are being blocked by vested interests or simple inertia and those that are being resisted due to legitimate concerns or alternate viewpoints. With respect to the latter, a key, and not entirely misplaced, perception of the present reform program is that it is essentially a "bankruptcy workout" with no social objective. Ele- ments of the reform program, such as better targeting of subsidies to protect the truly poor or vulnerable, have not found much receptivity either because of the political clout of the potential losers of this subsidy or because of the absence of credible alternate mechanisms of subsidy delivery. Despite the strong rationale that a financially sound power sector would help the gov- ernment free up funds for health, education, other social sectors, and even for energy subsi- diestothetrulyneedy,whileincreasingthesector'sowncapacitytoexpandaccessofthepoor to electricity, the reform program finds few supporters on the ground.10 Quite simply, when there is no immediate, visible, or commensurate benefit to compensate for a lost subsidy, it makes little sense for a citizen, rich or poor, to give it up. As one advocate of good governance observed: "In a corrupt and decrepit system in which most players in the state structure seem to be prospering, it does not carry conviction if a villager is told that he has an obligation to reduce the fiscal deficit or the losses in the energy sector" (Narayan 2000). In other words, given the overall weak accountability of public spending and poor quality of service delivery in India, especially in sectors like electricity, health, and education, this particular reform promise is not believable. Within the policymaking and academic communities, an articulate section emphasizes the "public benefits" agenda. The main concerns of this group are that "the dictates of eco- nomic efficiency and profitability will leave little space for the articulation and promotion of the public interest in the electricity sector," and that it is not clear how and when the social benefits mentioned in the context of reform will actually occur.11 Further, electricity is essen- tial to the supply of basic goods themselves (water, health, education, and so forth), so "it may still be appropriate for the state to invest in this infrastructure rather than leaving it to (the) market" (Reddy 2001). These arguments are not confined to academic corridors. Many consumers are uncon- vinced that the reform is designed to help them, and few among those who are convinced 10. For example, in a recent pre-election (and post-reform) survey in Delhi conducted by the Centre for the Study of Developing Societies (CSDS) for Hindustan Times, slum dwellers, that is, the urban poor, were the only class of consumers who felt the power situation had worsened since 1998 ("More power to the people: Supply better, say voters, but private players get no credit" Hindustan Times, October 25 2003). 11. Points argued forcefully in Dubash and Rajan (2001), a relatively objective account of the reform process, including the role of the World Bank. Can Good Economics Ever Be Good Politics? 7 believe the promises that reform will eventually improve the supply and services from the utilities.12 In response to these perceptions, we could search for instrumentalist solutions: how better to demonstrate the link between the reform program and benefits for the poor; howtoensurethatcompensationforlostsubsidy(improvedqualityofsupply)occurssimul- taneously with tariff rationalization; how to direct the unavoidable pain of reform to the rel- atively better-off consumer; and how to assure critics that the public monopoly will not be replaced by a private, perhaps less accountable, monopoly. Even if such solutions are devised, they will not address the fundamental issue: credibil- ity. In virtually every state, the utilities are publicly viewed as corrupt and mismanaged, and their financial plight as of their own making. In influential circles (such as sections of the political leadership, bureaucracy, and media) both rural and urban, privatization is not seen as a panacea, especially after the corporate accounting scandals in the United States and India's own IPP experiences.13 In the states where the World Bank supports the reform pro- gram,itsimageasanoutsider"imposing"aparticularmodelalsocannotbediscounted.Peo- ple believe that reform is being forced from above and that it will require them to pay for the utilities' inefficiencies and corruption. That is, the reform program itself as well as the indi- vidual power utilities face a credibility problem. If these perceptions are not recognized and addressed,itwillbedifficultforanypoliticianorgovernmenttomakeconsumersbelievethat the reform effort is indeed intended to benefit the wider public. Without that belief, the rul- ing politician is unable to create pro-reform constituencies. The World Bank (1995) itself describes reform as being possible only when it is polit- ically desirable, politically feasible, and credible. This has been emphasized by others; a recent article (McMillan, Rodrik, and Horn Welch 2002) has commented: The . . . responses that will make reform successful are likely to be forthcoming only when there is sufficient credibility attached to the change in the policy regime. That . . . requires creative thinking on credibility enhancing mechanisms as an integral part of reform. The ingredients of credibility include achieving full government ownership of the reform, managing expectations, building in compensatory mechanisms with believable assurances of carrying them through, and committing to stability of the new policy. These, in turn, depend on the government's reputation with its constituents, the prevalence of political checks and balances, and the binding of the new policy to wide ownership and statutory commitments. In the context of India's power sector, we could add two specific requirements: the need for not just government ownership, but public ownership of the reform program; and the need to enhance the corporate credibility of each state power utility by giving primacy to immediate improvements in its technical and managerial performance (where such 12. Observations in this and the following paragraph are based on interviews, the findings contained in IDC (1999) and Revi (2002), which cited several surveys establishing post-reform consumer dissatisfaction. 13. The CSDS survey for Hindustan Times cited above found that a clear majority (54 percent) did not think privatization had benefited Delhi. Indicative of the perception challenge before reformers was that, among different classes, only the rich and those living in group housing societies felt that privatization had been beneficial. This found dramatic expression in the summer of 2005, when middle-class residents of Delhi came out on the streets to protest against alleged faulty metering and billing by the private compa- nies, an episode that has raised severe and lasting questions in India about the credibility of the concept of privatization not only in electricity but across sectors concerned with delivery of public services. 8 World Bank Working Paper scope still exists) as well as customer services and relations, so that assurances of future quality of supply can become more believable. This points to the clear need, as well, for careful sequencing of reforms to make them more palatable to the consumer. In the power sector, policymakers have grappled with a "chicken-or-egg" syndrome with regard to quality of supply and tariffs. From the point of viewofpoliticalfeasibilityandcredibility,itisobviouswhichshouldcomefirst.Asmentioned earlier, progress has been made in reform implementation in some states, yet an apparent disbelief remains in the public mind.14 Publicity of what has been achieved may not always workandcouldevenboomeranginsomecasesasitwouldbedisbelievedbecauseofthecred- ibility problem. Perhaps it is regulators who should more proactively take the lead in nego- tiating what the utilities will deliver to consumers in return for a tariff hike.15 The reform-enabling Electricity Act of 2003 has taken this forward by mandating regulators to issue regulations on standards of performance, including financial penalties to be payable to consumers in case of non-compliance. A related question addresses time horizon. Politicians are deservedly criticized for fail- ing to think beyond the next election, but more pertinently, citizens everywhere are con- cernedwithshort-termconsequences.Nohouseholderorentrepreneurfindsreliefinhearing that after some immediate pain, things will improve in the medium term.16 This universal psychology is reinforced and powerfully articulated in India because of the frequency of its elections forcing even big-picture politicians to think short term. The usual time horizon for an Indian federal or state government to implement "tough" reform is three years after being elected, but the last two years of the term are spent mollifying the electorate in anticipation ofthenextelection.Designersofreformprogramsmightfinditbeneficialtoadjusttheirown often rigid time horizon to this practical reality. Swing Voters and Rich Farmers The issue of "rich farmers" deeply perplexes those advocating power reform in India. This groupofelectricpump-usingfarmers,clearlyaminority,hasbeenproventobethemainben- eficiaryofelectricitysubsidiesthatarefiscallydraining,yetnopoliticiandaresdisturbit.17This 14. The author witnessed a poignant example of this disbelief at a gram sabha (legally mandated assembly of all voting adults of a village) at Village Kotda Athamana in Kutch district of Gujarat on November 4, 2003. The villagers discussed various development issues and problems in a cheerful and constructive manner but turned angry and derisive when the local district development officer announced a new scheme of the chief minister that promised 24-hour electric supply to any village whose residents agreed to install meters and pay full tariffs. Since you can't deliver electricity for even a few hours without interruption, was the villagers' unambiguous message, first let us see you deliver 24-hour supply, and then we'll think about paying. 15. This may be an over-optimistic expectation given the experience of regulation so far in India, but not entirely unrealistic. For a discussion on the weaknesses of Indian regulation and a proposal to strengthen it, see Dakovic, Tenenbaum, and Woolf (2003). 16. An insightful analysis of this mindset can be found in Kanbur (2000). 17. In Andhra Pradesh, Karnataka, and Tamil Nadu, about 20 percent of farmers depend on power for irrigation; in Haryana and Maharashtra, about 25 percent, in Rajasthan 12 percent, in Uttar Pradesh 4 percent, and in Punjab 67 percent. Pump-owning farmers belong to all socio-economic classes, but medium (2­4 hectares) and big (more than 4 hectares) farmers predominate and have gained dispropor- tionately from subsidies. Can Good Economics Ever Be Good Politics? 9 group appears impervious to the non-sustainability of the present regime and continues to demand protection. At the same time, surveys have repeatedly shown that individual farmers are willing to pay realistic prices for electricity if quality of supply is ensured. However, in state after state, power reform has lurched to a halt the moment it runs up against the agri- culture sector, whether the context is subsidies or installing meters to better monitor supply. The reason is simple. Farmers, especially big and intermediate farmers, are by far the most important political constituency that the power sector impacts in India. Reformists may well point out that it is these "rich" farmers who are being subsidized at the expense of the poor, and that reform will ultimately benefit the numerically superior poor masses; but the fact is, the mistargeting of subsidies was not a mistake, and politically, the big farmers, though small in number, are more important than the poor masses (who in any case are splintered). The big farmers are usually the patriarchs of their clans and communities, and function as political intermediaries who deliver blocs of votes to their favored political party. The pump-owning class is also the most articulate rural class. In an era of fragile coalitions and volatile vote swings, the big farmer's control over bloc votes is a potent weapon.18 This group commands tremendous "swing power," and it is very risky for political leaders to alienate it. The tendency to tinker with power tariffs to buy popularity with the rich farming class began in the 1970s. Thereafter, such tinkering became a nation-wide contagion, and ruling parties have found that subsidizing agricultural inputs, and thus gaining big-farmer support, to be easier than developing long-term agrarian investment and growth strategies. The 1967­77 decade had seen, for the first time, the decline of India's historically dominant Con- gress party and, simultaneously, the economic and political rise of middle-class and middle- castefarmcommunities.Increasedpoliticalcompetitiontowoothesesectionsledtotheoffer of populist sops like power subsidies, and this had a demonstration effect across states and political parties. Other political constituencies, such as the urban middle-class, endorsed these policies because of the national desire to attain self-sufficiency in food production. Over the next two decades, rural meters stopped being monitored in state after state, as politicians, upwardly mobile farmers, and SEBs across the country found a common inter- est in the subsidy regime--politicians for the votes it delivered, farmers for the free power (and therefore water) it afforded, and SEBs because it provided an opaque accounting head under which they could hide their transmission and distribution losses.19 The governing class is now trapped: fiscal necessity demands that subsidies be reduced, but undoing them could prove politically harmful. What then of the surveys that show farm- ers as willing to pay for quality supply? This has to be juxtaposed against the field experience showing that in political and other forums, farm leaders actively oppose subsidy withdrawal, privatization, and even metering, despite appreciating the logic of the measure being pro- posed.Sothesituation,ineffect,isthatfarmersarewillingtopayinprinciplebutnotinprac- tice.Theycanthusbeviewedaseitheralatentpro-reformconstituencythatismerelyhesitant for some reason or as hopelessly selfish and obdurate. Either way, this contradiction needs to be objectively analyzed. 18. According to political scientist Yogendra Yadav (2002), India has the highest "volatility index" (extent of vote swing between elections) among all democracies. 19. This is the author's interpretive summary of the history of the politics of agriculture subsidies pro- vided in Dubash and Rajan cited above. 10 World Bank Working Paper One presumable reason for farmer resistance is that "quality supply" is an unconvincing promise that no one will believe until it is actually delivered, as we saw in the Gujarat gram sabha note earlier. This leads back to the question of credibility and governance of utilities, and cracking the chicken-egg syndrome concerning the supply-tariff sequence. A second, very possible, explanation is that a small section of the most landed, articulate, and politically influential farm leaders is simply blocking reform due to its own stake in the present system, andpreventingmediumandsmallerfarmersfromcooperating.Thissituationmayvaryfrom state to state depending on the local social-political structure, which could explain why farm leaders have managed to block any agriculture tariff hikes in Andhra Pradesh and Karnataka, in contrast to Punjab, where farmers have accepted a modest increase. If indeed there is a disjunction at any location between farm elites and the farming com- munity in general, it would provide an opening to mobilize farmers in favor of reform. The challenge in such cases is to recognize the conditions and devise appropriate methods to bypass the elite and mobilize the communities. The key will lie in getting the local farm com- munities to "own" the problem (rather than forcing a solution on them), and empowering them to negotiate their own solutions with electricity utilities, such as, where conditions per- mit, forming cooperatives to control local distribution.20 However, the most powerful reason for non-cooperation is that all farmers have a real fear of paying higher electricity prices. So the question that really needs to be probed is, why shoulda"rich"farmerfearpayingslightlymoreforelectricitywhenheunderstandsthatulti- mately it is he who will benefit from a reformed and self-sustaining utility? The key point to consider here is that reduction of power subsidy is not the only threat the Indian farmer faces. In the general reform environment of the last decade, subsidies for other farm inputs--fertilizer and seeds, for example--are also under the scanner as are the numerous other protections the government has provided in the last several decades, such as assured prices and purchase of produce. The answer will therefore have to be found by placing the pump-using farmer at the cen- ter of an analytic work examining the costs and prices of all these inputs and outputs; their interlinkages(especiallythatofelectricityandwater);thefarmer'svulnerabilities;aswellashis access and incentive framework with respect to markets, credit, cropping patterns, terms of trade, and the broader agriculture policy and structural environment. This would help situate electricity in its context as a farm input, leading to better understanding of its place in the farmer's economy, and thus open routes to more creative strategies to resolve this problem. The opening may be found in an area that has nothing to do with electricity: risk insur- ance and rural credit reform, possibly. What is apparent from the experience so far is that electricity subsidies for farmers are difficult to rationalize without reference to farm issues in general. A multi-sector (agriculture-water-power-risk-credit), consultative approach thus is necessary, and politically a "package deal" will have to be struck with the farmer.21 Such a holistic approach might also help jumpstart reform in related sectors. 20. The experiment by the non-governmental organization Prayas in Maharashtra and the long- running Hukeri Rural Electric Cooperative Society in Belgaum district of Karnataka may be worthwhile case studies. 21. This is not a new or revolutionary insight. The Andhra Pradesh Power Sector Restructuring Pro- gram Socio-Economic Assessment Report on Baseline Survey (TARU Leading Edge 2001) forcefully stresses "the need and urgency for a comprehensive strategy that addresses such issues." Can Good Economics Ever Be Good Politics? 11 In parallel, an urgent need for targeted communication with farmers is clear. Much of the resistance may simply arise from ignorance, fear of the impact of measures like metering, and alienation as a result of not being consulted on the manner and timing of the reform. However, communication will be credible only if the reform is seen as addressing whichever farmer concerns are genuine. Corruption and Vested Interests The number of players in the electricity sector is large. Primary stakeholders include utility staff (managers of various levels); engineers and field-level technical and commercial per- sonnel; entrenched contractors providing services such as equipment supply, maintenance, billing and collections, food services; and various categories of consumers: farmers (big and small), domestic consumers (middle-class and poor, further divided into urban and rural), industry (big and small), and the commercial sector. Intermediaries include policymakers, the political class (at the federal, state, and local levels), trade unions, civil society, media, and donors. In addition, post-reform will include electricity regulators. An important set of sec- ondary stakeholders comprises the key players of other sectors--such as agriculture, indus- try, and transport--whose policies and behavior influence the power sector. This paper has already discussed some critical consumer categories that resist reform on grounds of legiti- mate concerns, low credibility, and political power. We now look at the important players in the system--utility staff, the bureaucracy, and the political class--elements of which have a vested interest in the current regime. As has been observed, despite the wide agreement on the need to improve the gover- nance of the sector, most state governments have approached this task in a cosmetic, half- hearted fashion. For example, they have been reluctant to grant autonomy to their power utilities, contriving to retain management control or influence through top-level appoint- ments, even where they have unbundled and corporatized. Indeed, analysis would likely reveal a positive correlation between timing of changes in state political leadership and that of topmost utility management. Similarly, most state governments have met the federal legal requirement to establish regulatory commissions, but have given them very limited jurisdiction. Even in states where regulators have been given a wider mandate, questions arise about their independence, process of appointment, and powers of enforcement. Karnataka, for example, did not imple- mentthecommission'starifforderin2002;Rajasthandidnotallowtheutilitiestofilefortar- iff increase; and Andhra Pradesh has not sought any increase for agricultural consumers in the last three tariff filings. Clearly, some parties in the system are obstructing effective reform. The most obvious culprit is the corrupt utility employee, who has done the most harm to the public image of utilities, and thus greatly undermined the overall believability of the reform effort. An assis- tant engineer in a district town, it is said, can rake in Rs200,000 (about US$4,500) per month through corrupt means (as against an actual salary of Rs10,000­25,000), while the national theft of electricity is estimated to cost Rs20,000 (roughly US$4 billion) per year.22 22. The estimate for an engineer's income is from Narayan (2000); the national estimate of theft was declared by Prime Minister Atal Behari Vajpayee in his Independence Day speech (2002). The exchange rate in June 2004 was approximately US$ 1 = Indian Rs 46. "Crore" is an Indian unit of measure equiva- lent to 10 million. 12 World Bank Working Paper The internal structure and lax procedural practices of the state-run utilities, which allow for "informaldiscretionarypower"inthefield,ithasbeenargued,aretherealcausefortheircul- ture of inefficiency. That is, political interference and rent-seeking are a result--rather than the cause--of the utilities' mismanagement.23 This is certainly true at an operational level where field-level discretion and lack of accountability allow local politicians, utility staff, and some consumers to collude in order to influencedecisionsrelatedto:timinganddestinationofsupply;locationoftransformersand substations; award of contracts; carrying out of metering, billing, and collection; exemptions from disconnections; and appointments, transfers, and postings of field staff. Such opportu- nities are particularly lucrative for field staff, as is evident from a 2002 survey by the India chapter of Transparency International. This nationwide study of petty corruption in India ranked the power sector third most corrupt after the police and health in terms of public per- ception and second most (after health) in terms of social impact. Key findings for the power sector included: I One in two members of the public who interact with the electricity utility experi- ences corruption. In two-thirds of these cases, money is directly demanded. I Main areas of corruption are "improper" supply, excess billing, new connections, and meter readings. I Themostcorruptamongthestaffarelinemen,"officers,"meterreaders,andbilling clerks. I Total"outflow"ofpettycorruptioninthepowersectorisestimatedatRs5,764(a little over US$1 billion) per year--Rs669 per capita. If the last estimates are correct, it should be easy to calculate the per capita receipts per employee as well. That figure alone would explain the internal resistance to governance reform, but there is more: the higher corruption involving political kickbacks. There is no exactfiguretoquotehere,butthesectorisacknowledgedtobearichsourceofelectionfunds, withpoliticiansatthelocallevelexactingrentsinexchangeforsomeofthefavorslistedabove, andpartiesfillingtheircoffersthroughthecontractsandprocurementroute(Rao2002).This is another area where technical solutions for the power sector run up against larger realities. It is a well-documented and understood fact in India that the country's election-funding mechanisms are the root cause of its widespread institutional corruption and unscrupulous politics.24 Candidates and parties incur astronomical, undeclared expenses and then seek "returns on investment" when in office. In many ways, the prerequisite for successful power and governance reform in India is election-funding reform. However, funds are not all that the politician extracts from this sector. A far more valu- able currency is being exchanged: favors in return for votes. As is evident from the earlier dis- cussion on farmers, the power sector has an important bearing on a party's political support. Distorted tariffs and cross-subsidies breed conflicting interest groups, a situation that auto- matically creates political constituencies. A further political space is opened up in the present era, when uncertainty precedes the promised benefits of reform, a space that the opposition 23. From Ruet (2001), an inference one could draw is that improved internal management is a pre- condition for controlling political interference. 24. Numerous official, legislative, civil society, and academic studies testify to this assertion. Can Good Economics Ever Be Good Politics? 13 party in every reforming state has visibly sought to occupy. Power sector policy thus forms a core component of a politician's political strategy, concretizing the difficulty in "depoliticiz- ing" it. Besides politicians and corrupt staff, are other vested interest groups such as suppliers and contractors. Merely feeding off the system, they are not the primary cause of the rot, and so are passed over in this paper. Utility Staff and the Bureaucracy Corrupt elements apart, utility reforms are bound to generate concern and resistance among utility staff, causing them to emerge as the strongest internal lobby against reform. Surveys reveal that senior managers resist reform for fear of losing control, and middle management and field staff for fear of retrenchment. Some also, quaintly, fear "increased workload" (IDC 1999). These responses convey an instinctive, anti-reform mindset, which should not catch anyone by surprise given that for vast swaths of the Indian lower and middle classes, a gov- ernment job is primarily a passport to security, and that the concepts of productivity and merit are relatively weak in the public sector work culture. Public sector staff also tend to be highly unionized and politically aware, and thus very effective in safeguarding their interests. In order to gain staff support for reform (or, at the very least, minimize staff resistance), therefore, the utilities' top management, backed by the reforming government and its polit- ical leadership, needs to put in place proactive human resource and communication strate- gies from the start in order to address employees' natural fears and insecurities. As in the case of farmers and other consumer groups, utility staff constitutes a vital stakeholder group that needs to be drawn into the reform process and made to "own" the problem. Tactics need to be devised to both engage union leaders and bypass them. Despite occasional headline- grabbingstrikesandotherformsofindustrialaction,unionleadersinIndiaareusuallypolit- ically flexible and open to negotiation. Moreover, the lower bureaucracy, as a class, is docile and unlikely to question authority if it is not insecure. Staff resistance, in other words, while undoubtedly difficult, is not completely intractable either; it can be addressed by a resolute and creative management. Yet, visionary leadership is hardly the hallmark of the senior bureaucrats leading the reform.25 A safety-first culture and uncertainties in the politician-bureaucrat relationship mean that bureaucrats tend to over-read political messages and play to politicians' fears and ignorance. This works both ways: officers might defer more eagerly than is warranted to a politician attempting to block a measure or, on the other hand (and perhaps more often), dampen the enthusiasm of a politician seeking reform by emphasizing only the difficulties of implementation rather than possible solutions. Perhaps India's politicians could have been goaded into implementing more reforms than they have so far had their top bureaucrats not been so ingratiating or, alternatively, so discouraging.26 An added complication, which 25. Utility chiefs, for the most part, continue to be drawn from the elite bureaucratic cadre, the Indian Administrative Service (IAS), which also runs the state and federal administrations overseeing and coor- dinating reform. 26. These comments are admittedly sweeping and do injustice to the crop of bold and constructive officers present in every state cadre and to the pro-reform lobby in the bureaucracy. The attempt here is to describe a predominant and influential administrative style. 14 World Bank Working Paper neutralizes even the more effective officers, is the frequency of transfers, often for political reasons, in the states. This practice precludes bureaucrats from developing sector expertise or a long-term interest in a reform program. Political Will We probably still do not have a clear answer as to why a committed reformer should not be able to cut through the resistance with the political backing of the beneficiaries of power reform: Why is the political will to clean the system missing? The foregoing analysis suggests some tentative answers. One, reform has no visible political allies. Potential candidates like the "poor" neither understand the link to themselves nor are they part of the discourse, barring a few experimental locations. Furthermore, industry, which underwrites the cross- subsidy to agriculture and households, has possibly found it more expedient to steal elec- tricity or secede from the system, and so has no spontaneous interest in the reform process. Two, the problem may not be one of political will alone but, as described above, of bureau- cratic will as well: government officials and utility managers likely exaggerate the "political resistance" they meet in the field, using the politician or trade union leader as a scapegoat to mask their own inertia. A final issue relates to incentives. A politician, ultimately, will push for a change only for political expediency. This is discussed at a theoretical level below, but the point here is that Indian politicians, especially at the state level, so far have had little political incentive to take on the interests who rule the power sector. Further analysis to examine why presumed nat- ural allies of the reform have not mobilized, to gain better understanding of the bureaucrat- politician relationship, and to identify incentives for reform-minded politicians could perhaps point us to a feasible political strategy for implementing power sector reform. Politics in India: Form and Substance The preceding section examined specific hindrances within the power sector that have impeded implementation of reform. Even so, references to broader issues such as the politi- cal economy of agriculture, political time horizons, election-induced corruption, and the cul- ture of the bureaucracy were inescapable. This paper now shifts to the big picture. In order to understand the political factors affecting the power sector, it is necessary to assess how the architecture of the Indian state and the country's social structure combine to generate a par- ticularstyleofpolitics,whichinturninfluencesthemannerofimplementationofreformpro- grams. This paper also departs from the pessimism of the previous section to search for clues to the implementation of whatever reform has occurred so far in India in various sectors. Democracy and Populism Pro-reform elites in India tend to blame democracy for much of the delay and inconsistency with which the country has gone about implementing its reform agenda, and superficially they are correct. This group's main arguments are that election-driven politicians tend to be risk-averse, populist, and prone to interest-group capture, and possess short time horizons. Thisdescriptioncouldperhapsapplytopoliticiansinanydemocracy,buttheparticularmal- Can Good Economics Ever Be Good Politics?15 ady in India appears to be that neither politicians nor the constituents who drive them are able ever to sacrifice short-term, partisan gain for more durable, wider public benefit. While populism is often an attempt to address genuine grassroots concerns, in practice it manifests itself through the political instruments of patronage and clientelism: politicians extract favors from the system to benefit particular groups of political clients (usually the more powerful and better organized elements of society), who in turn reward the politician for doing just that (and expect no more from him). The combination of dysfunctions in India's power sector--poor service, corruption, unsustainable subsidies, and so forth--are one outcome of this culture. Yet, as argued earlier in the subsection on corruption, the pop- ulist politician is perhaps more a result than the cause of the rot. This explanation is of necessity simplistic, but it sheds light on two intriguing aspects of Indian politics. For one, the servicing of favored clients by political patrons helps us under- standwhythefrustrationevidenteverywherewithpoorgovernanceandlackofdevelopment does not translate into cumulative political pressure to bring about enduring change; and for another, it provides an insight into the lack of ideology and clearly identified positions on economic issues among political parties. (An example of which, alluded to above, is the ease withwhichpartieschangetheirpositiononelectricityreformdependingonwhethertheyare in or out of power.) However, it would be wrong to conclude that Indian politics are entirely bereft of vision or ideology. The country as a whole is consciously shifting from one development paradigm (centrally planned, public sector-driven socialism) to another (market-driven growth with government as facilitator). Because of the complexity of its politics, the management of this process cannot follow a script. It is improvised from day to day, with politicians using their skills and networks to push a measure or pull back from one, depending on their calculations and compulsions of the moment. What can be said in mitigation is that democracy trains politicians to be opportunistic, flexible, capable of negotiating new alliances, sensitive to changes in the concerns of their electorates, and publicly accountable; to mediate and resolve conflict; and strike deals and compromises. Democracy also provides early warning of opposition and enables ruling politicianstoadjusttacticsaccordingly.Thus,ratherthanrailatIndia'sdemocracy,thispaper takes the view that, though the experience of reform implementation has been frustrating in India, it is these political skills that have actually helped India implement whatever reform has occurred so far.27 Federalism India's federal structure has the huge drawback, from the economic purist's view, of keeping India's political class in almost perpetual election (and populist) mode, as every year one or more major states face elections. However, on balance, federalism has been the single most decisive factor that has sustained reform in the last decade. Its advantages have been many. Most crucially, as liberalization, deregulation, and other reform across multiple sec- tors impacted different states in different ways both positively and negatively, no political party has been able to outright reject the broad reform agenda, and no common ground 27. This argument is incisively developed in Jenkins (1999). 16 World Bank Working Paper has risen to rally opposition. Second, liberalization has provoked increased competitive- ness and demonstration effects among states (power reform is a strong example), provid- ing momentum to the national reform effort. Third, the division of the country into federal units has correspondingly divided anti-reform interest groups. Finally, reform introduced in a staggered fashion across the states has confined political discontent at any one time to a few sectors in a few states. In the case of power, which is a concurrent subject (where the center and the states have jurisdiction, see footnote 2), the center plays a key role in framing policy, such as through the landmark, reform-enabling Electricity Act of June 2003, funding the sector (through instru- ments like the reform-linked incentive fund for states--the Accelerated Power Development and Reform Program28--and agencies like the Rural Electrification Corporation and Power Finance Corporation), and providing services through its own generation and transmission entities. In the past, the center contributed to the profligacy that is destroying the sector by allowingstatestocontinuetodrawontheseserviceswhiledefaultingmassivelyonpayments. Inrecenttimes,however,thecenterhashelpedcreateanenvironmentforpowerreformwith the prime minister and other federal ministers and forums giving it a prominent place in the nationalpoliticaldiscourse.Thecenterhasachievedthisbetterenvironmentbyusingitscon- vening role to get states to agree to certain basic principles; by issuing guidelines; by working through the policy and funding mechanisms just listed; and, of late, by imposing stronger payment discipline on SEBs purchasing from its entities. Because parties ruling at the center and states have similar interests to protect, and there wouldbemuchpoliticalgiveandtakeamongthem,thecentralreformframeworkisnotpar- ticularly stringent, nor has the center pushed the states to implement agreed measures. So this appears to be another case where political compromise has created a gap between policy and practice. The Electricity Act is a case in point. Thisact,passedaftermonthsincoldstorageandbehind-the-scenesnegotiatinginapar- liamentary standing committee: consolidates existing laws on generation, transmission and distribution; supersedes conflicting state legislation (a matter of some contention); intro- duces legislation to govern new concepts like trading and open access; liberalizes and deli- censes generation (except hydro); expands the definition and activities of captive plants; and allows competition in distribution. In other words, the act is a significant milestone, virtu- ally legislating the reform agenda being attempted since the mid-1990s and putting the SEBs under increased pressure to perform, else they risk losing their more lucrative customers, such as big industrial units, to competition. Yet, typically, the act does not go the full way. Much has been left vague and will depend on the rules of implementation to be issued by the power ministry, its interpretation by the variousstateregulators,andthecaselawthatwilldevelop.Thatis,thereisstillplentyofroom fornegotiationandcompromise,fordelayandobfuscation,andforvariousprovisionsofthe act to be either diluted or strengthened, depending on which lobby wins on a particular day. At the same time, amendments are pending that would further tie the states' hands in terms of timing of key policy steps, such as implementing open access arrangements and phasing out of cross-subsidies. 28. Under the Accelerated Power Development and Reform Program, states agree to achieve time- bound reform milestones in exchange for access to federal soft-funding facilities. In addition, there is an incentive component of APDRP that provides cash grants against reduction of financial losses of utilities. Can Good Economics Ever Be Good Politics? 17 However, if we cast aside the expectations of quick results for a moment, we can discern thatthecenter'srealcontributionhasbeenthatofagendasetting,andherepoliticalskillsplay a role. Since ruling politicians, cutting across parties, understand the need for power reform, but face huge resistance to implementing it in their states, they have contrived to keep the reformist pressure on at a national level. The center thus speaks the correct language on reform, but acts softly. This gives the ruling state politician a public alibi to pursue reform, yetleewaytoimplementitatacomfortablepace.ThisistypicalofreforminIndia:itisagreed to in principle; the policy or legislation that follows is a compromise; and implementation lags significantly. Nevertheless there is net, albeit incremental, progress. The Politics of Reform: Style and Strategy India's reform record, especially the vast unfinished agenda in various sectors and the mean- dering,half-heartedpacewithwhichthefederalandstategovernmentspursuereformimple- mentation, is usually described in despairing terms. The in-country reform lobby and the country's external well-wishers draw repeated and unflattering comparisons with China in particular. Among pro-reform circles, most discussions on why reform does not occur swiftly, smoothly, and decisively in India focus on listing the usual suspects: the limitations of democracy, politicians' shortsightedness and corruption, and lack of political will. These argumentstendtobecircularandshedlittlelightonhowtocrackthesyndrome.Armednow with some insights into the motivations and compulsions of India's politicians, we can attempt to isolate their strategy for managing reform. To do this, it is necessary to set aside frustration for a moment and look back at India in 1991. Relative to that starting point, significant reform has occurred in a number of sectors. Reform in trade and industrial policy has been far-reaching; external liberalization has pro- gressed considerably, and so has financial sector reform; at the state level, deregulation and sales tax reform have moved in several states and so, indeed, have some aspects of power sec- tor reform; and, in real terms, subsidies in several sectors, including some related to the holy cow of agriculture, have been curbed. If one were to delve further into each sector and take any reasonable cut of time in the last decade, progress would be discernible, albeit slow, between the start and end points. In addition, it is worth remembering that Indian reform is government led, that is, initiated by ruling politicians. Furthermore, it has occurred in cir- cumstances in which reform beneficiaries are ignorant or unbelieving of their status and in aneraoffragile,fractiouscoalitionsandfrequentelections.Fromthisperspective,thereform record is not entirely discouraging.29 True, a breakdown of this record would show that reform has proceeded in relatively "easier"sectorswhileithasstalledin"difficult"ones,andwithinsectors,"simple"reformhas moved ahead of "complex" reform. Yet, this argument is again both self-fulfilling and circu- lar.First,thissituationisnaturalandnotuniquetoIndia.Second,onemustconsiderthepos- sibilitythatthereformmightnotbesocomplexorthesectorsodifficultortheimplementing government so reluctant if political feasibility--credibility, sequencing, public ownership, relevance to grassroots priorities, compensatory formulas, and so forth--was factored into its design at the very start. 29. For a recent, upbeat account of India's reform record since 1991, see Kelkar and Shah (2003). 18 World Bank Working Paper In any case, by now we have something of a storyline for reform in India. At a broad "vision" level, the country is attempting to undergo a paradigm shift. In practice, political leaderspreferexpediency,acceptingreformatapolicylevelbutimplementingitinselectsec- tors at a low-risk pace. Within sectors, policies are put into practice and begin to have a direct impact on stakeholders, frequently meeting, as in the case of power, with strong resistance. The analysis below draws on theoretical concepts in order to understand the generic politi- caldynamicsthatoperateinareformcontextandexaminestheirmanifestationinIndia,par- ticularly in the power sector.30 Politicians' Incentives to Reform It is axiomatic that politicians will be willing to reform if they have an incentive to do so. In simple terms, the reform must fulfill at least one of the following conditions: a) enhance their politicalsupport,b)notmeetwithoverwhelmingopposition,andc)provideopportunitiesto increase the resources they control through increased economic benefits (for their support- ers), patronage opportunities, or kickbacks. Applying this to the Indian experience, it is clear that politicians have not had an ideological bias for or against reform. They have neither opposeditwholeheartedlynoradvocateditcoherently.Insectors(orwithinsectors),theyhave pursued reforms where they have had incentives to do so; otherwise they have not. Drawing inferences from the case of power reform in India, it becomes apparent that reform is less likely in areas where its costs are concentrated on a small number of powerful actors while the benefits are dispersed among a wide number of prospective beneficiaries (who may not even be aware of their beneficiary status). Conversely, reform is more likely in sectors where the costs are incurred by poorly organized groups, including taxpayers, while thebenefitsaregainedbythewell-organized.(ThishappenedduringtheIPPrushoftheearly 1990s, in which ruling politicians and the international power industry saw much profit, while consumers and those affected by the location of the projects were hardly consulted.) Extending the logic, patronage-based parties will have difficulty implementing reforms that curb patronage. Conversely, they will retain the support of groups whose interests are hurt if they are able to generate other incentives for their supporters to remain loyal--hence the importance of cross-sector solutions to crack the farm subsidy conundrum. Finally, policy reform with indirect or delayed consequences is easier to implement than one whose consequences are swift and direct. Itisnowobviouswhy,forinstance,laborreform,privatization,andthereductionofsub- sidies have generally proved to be such stumbling blocks for India's reformers and, indeed, why some elements of power reform have proceeded while others have proved difficult. Power reform, in short, is by definition politically tough--and reformers must go in expect- ing a rough ride. Interest Groups Muchoftheacademicliteratureoneconomicreformreinforcestheintuitiveassumptionthat "interestgroups"--especiallyrent-seekinginterests,politicianswithshorttimehorizons,and competing electorates--constitute an immutable impediment against reform. India's expe- 30. Main references for this section are: Echeverri-Gent (2000), Rodrik (1996), and Jenkins (1999). Can Good Economics Ever Be Good Politics? 19 rience with power reform would substantiate this view. However, before settling on a firm conclusion, the political behaviors of economic interest groups are worth examining. Broadly,thegroupsformtwocategories:thosewhowieldinfluencethroughthevoteand those who wield it mainly through access and money. The first set includes segments like big farmers and government employees who, even after being morphed and subdivided into their caste or religious identities (the basis of political mobilization in India), would consti- tute significant voter blocs for whom reform impact would be a key (but not the only) deter- minant of electoral choice. The second set includes industrial and commercial interests who bargain for favorable policy outcomes through networking and contribution to political cof- fers. (A third category includes activist interest groups such as the environmental, consumer, civil rights, and women's rights movements, whose influence is based on lobbying the bureaucracy, judiciary, and media. This category is not material here.) Viewed through these lenses, it can be argued that interest groups are a dynamic entity that can be negotiated with and manipulated; can adjust to changing circumstances; and are prone to splintering, erosion, or compromise. Even those hostile to reform "favor limited battles rather than total war" (Jenkins 1999). Further, the natural opportunism of politicians enables them to create new alliances in response to changing circumstances and to find new avenues to garner political resources. These are important factors in the Indian context, where reform has been largely top- down and government led: it is worth reiterating that ruling politicians and policymakers are the ones who have initiated and pushed reform in India. They have improvised as they have gone along, pausing and reviewing at every stage. Incremental, instead of "big bang," reform hasenabledthemtoerodeorchipawayatopposinginterestgroups,creatingdivisionswithin their ranks and preventing effective anti-reform mobilization.31 Anti-reform interests have adjusted their own tactics--or lost numbers from their ranks--at each turn. They then begin to expect the next dose and are relatively more docile when it comes. One has only to look at sectors such as telecom and external liberalization, where reform has made considerable advances, to see this process in play. Telecom,inparticular,isoftencitedasashowpieceforreform,thoughitisapparentthat reforminthissectorisfarfromcomplete,withstronggovernmentinfluence,insufficientreg- ulation, continued scope for political interference, and an uneven playing field. The point is, however, that compared to the starting point of the mid-1990s, government control in tele- com has lessened, a regulator is in place, day-to-day political interference has been reduced, services have improved, and most crucially, pro-reform constituencies (consumers and cor- porateinterestswhohavebenefitedfromcompetition)nowexist.Itisalsoevidentthat,build- ing on this accumulated reform of a decade or so, further reform is on the way. That is, telecom reform has been a continuous (though not always steady) process. Ofcourse,incontent,telecomisnotanidealexampletociteinapaperonpowerreform, because raising capital--not bankruptcy--was the concern in this sector. Neither was the sector as corrupt as the power sector, nor was it burdened with politically significant sub- sidies. Moreover, while the unmistakable public message of telecom reform is price cuts for consumers, the opposite is the case in the power sector. Nevertheless, the analogy being 31. The privatization of the Delhi power distribution system in 2002 can possibly be categorized as "big bang" reform. Yet, as an almost wholly urban and relatively homogenous city-state, Delhi's circum- stances were unique, and it can be argued that this is a case of the exception proving the rule. 20 World Bank Working Paper drawn here is one of process: the political management of telecom reform is a classic exam- ple of how India implements reform in general.32 It has been argued that such "successful" reforms are more feasible in federally admin- istered sectors such as telecom, as they do not impinge directly on powerful voting interests suchasfarmers,forexample,whomstateadministrationsconfrontincontextssuchaspower reform. It is true that states must confront anti-reform interests more frequently than does the center because of the nature of subjects under state jurisdiction, but the divide between "difficult" and "easy" reform is not a center-state one. The center, for instance, has not dared to proceed overtly on labor reform and is having trouble even beginning to rationalize the income tax regime, while several states have made advances in, for example, sales tax reform and deregulation. Reform is politically difficult in all sectors, federal and state. A better difficult/easy cate- gorization can be made as follows: where the gulf between compensation and pain (in con- tent, perception, and time) for politically and financially powerful interest groups is narrow, reform proceeds in stealthy, incremental doses. Where it is very wide (or appears to be so), as in power and income tax, reform apparently grinds to a halt. Winners and Losers This observation leads to another theoretical assumption about reform politics. A number ofmodelsassumethatreforminitiativescreateclear-cutcategoriesofwinnersandlosers,and that the difficulty arises when benefits for the winners are not quick or visible. This is intu- itive but does not capture the complexity of practical politics. In reality, such a neat and instant divide is rare. The most critical political factor when a reform is implemented is uncertainty--in the minds of the so-called winners that: a) they actually are winners and b) if so, will one day actually receive the benefits. Similarly, so-called losers are: a) uncertain that theywilleverbecompensatedandb)hopefulthattheymightbeabletoblockthereform. In real time, therefore, the categories overlap as, at a micro-level, players are not cer- tain to which category they belong. This clearly is the situation with power reform at this juncture. As described earlier, presumed winners such as the rural poor and industrial con- sumers are either unaware or uncertain of the benefits coming their way, while potential losers such as big farmers and utility staff are uncertain of being adequately compensated. Both believe they should block the process. Further, as a reform is implemented, traffic continues between the two groups as indi- viduals adjust behaviors. The politician managing the reform has to try to tilt, while con- stantly judging, the balance in this dynamic situation. In India so far, the chosen strategy, as indicated in the telecom example (or, for that matter, in the ambiguity left in the Electricity Act), has been incremental reform. This helps create an initial constituency of early "win- ners" who sense real benefit in further reform, while lulling "losers" into believing they have reaped a good compromise. These initial winners develop into an interest group with a stake in reform and give the implementing politician confidence to push further. The "losers" begin to cede ground in slices; over time, they lose members to the other side. This is exactly what is happening in telecom. It is slow and imperfect, but it is generally acknowledged to 32. An informative comparison of telecom and power reform in India can be found in Mukherji (2004). Can Good Economics Ever Be Good Politics? 21 have worked. If power reform is to proceed, therefore, a momentum needs to be imparted to the loser-winner traffic. For this to happen, the factors creating uncertainty will need to be addressed, as described previously.33 The Strategy Revealed The above discussion suggests an implementation strategy in India of gradual, incremental reform, even in the so-called simpler sectors. Without going into judgments on the merits or demerits of gradualism, it is evident that the governing class in India has found comfort in this style of reform. Gradualism, of course, is not unique to India, and on occasion has been found to be the most acceptable method in even relatively homogenous societies like New Zealand and the Netherlands, where, for example, legislation to enable postal reform came more than a decade after the reform was initiated.34 One generous appraisal concludes that the process of gradualism has allowed India to plan and sequence its reforms coherently, ensured democratic buy-in, prevented policy reversals, and made all major political parties associate themselves with the reform agenda through the election cycle, which has seen them all govern at the center and in various states atonepointoranother.However,theauthorspointoutthatthereareseveralsectors,includ- ingpower,inwhichIndiacouldandshouldhavemovedfaster(WilliamsonandZagha2002). Under the umbrella of gradualism, we can now isolate the specific tricks and tactics that governingelitesemployinIndiatoimplementreform.Asourearlierdiscussionshaveshown, incremental (and stealthy) reform enables the politician to use his skills and networks to: 1) keep his own incentives in the process alive by enabling him to devise rules and strata- gems thatwillaugmenthisownresourcesandrewardhissupportersorcollaborators;2)gain time to divide, or lull opponents of the reform, exploiting the fluidity of interest groups; 3) resolve, avoid, defer, or shift conflict through compromise, obfuscation, deal-making, blame-shifting, or stalling; and 4) identify and harness incipient supporters of the reform who will help push subsequent reform.35 In short, since much of reform is politically unmarketable, either because it runs up against powerful interests, does not coincide with the politician's time cycles, or fails to address legitimate concerns, the implementation game is all about stealth, ambiguity, and following the path of least resistance. It is not a very pretty formula and does not conform to sanitized theories of transparent, consultative, democratic functioning or good governance. It can be faulted for its outcome of slow, compromised reform. On the other hand, it has ensured the implementation of whatever reform has occurred so far. The Electricity Act is a perfect example: it comes after a decade of slow, back-and-forth reform, but makes whatever 33. A possible starting point is a study to identify who exactly bears the cost of power subsidies in terms of higher taxes, poor social services, etc. Until this is fully understood, the benefits of removing sub- sidies cannot be convincingly demonstrated. Potentially, "losers" who are being asked to give up subsi- dies might actually prove to be "winners" if it is shown that they are unknowingly paying for the subsidy in other forms. 34. Quoted by E. A. S. Sarma in his address "Postal Reforms in the Indian Context," delivered at the Seminar on India Post, New Delhi, November 12, 2002. 35. Jenkinsprovidesseveralexamplesofsuchtactics.Twoparticularlyilluminatingcasestudiesarethose of Rajasthan sales tax reform and Sharad Pawar's handling of the sugar lobby in Maharashtra (Jenkins 1999, pp. 155­63). 22 World Bank Working Paper hasbeenachievedsofarvirtuallyirreversible.Nevertheless,itleavesenoughleewayforoppo- nents to feel able to negotiate and (ultimately) buy into the process. Lookingahead,twoquestionsarise:Canthisstrategyof"stealthygradualism"helpIndia completely achieve its paradigm shift? Is it possible to initiate an alternative, more efficient process of reform? The present style clearly has its limits, for not only does it keep India per- manently behind the curve and deeply impede its prospects for growth, it also perpetuates the culture of poor governance. A quicker and more efficient style is unlikely to emerge overnight on a national scale, though alternative approaches are plausible within sectors or locations. However, even at a local level, any "reform of reform implementation," to coin a phrase, would be critically dependent on ensuring democratic ownership of the reform by negotiating its content with the beneficiary populations and mobilizing grassroots political reform champions. Prospects for the Power Sector WenowhaveacompletepictureoftheIndianreformprocess:agenuineparadigmshiftbeing attempted at a vision level; political expediency in practice; strong resistance for a variety of reasons in various sectors; and an outcome of stealthy gradualism. With respect to the power sector, optimists will see a glass half full and will counsel patience, maintaining that India's reformers will, in their gradual style, eventually outmaneuver those resisting reform. Pes- simists will see the half-empty glass, and pronounce that whatever happens will be too little, too late. The latter could well be right, for it can be argued that the chip-and-erode style of reformhasreacheditslogicalendinthepowersector.Thereformagendahasreachedapoint where there is no space for behind-the-scenes maneuvers; stark decisions have to be taken that will directly affect large blocs of powerful voters and vested interests. If politicians con- tinue to hesitate, the fiscal situation will become unsustainable, and a critical infrastructure sector will remain dysfunctional. What, then, are the prospects for further power reform? This paper has focused mainly on identifying the impediments to power reform and on describing India's compromised reform implementation strategy. The positive factors that the country's power reform lobby must strategically seize are: I Theunambiguousandovertplaceonthenationalpoliticalagendaoftheissuesof unsustainable power subsidies and power theft: though this appears as empty rhetoric, it represents the significant and essential first step of the political class as a whole recognizing and facing up to the problem. It is accompanied by frank debate on these issues at grassroots forums. I Providing free power to farmers remains a politically tempting ploy and the tariff question will remain a tricky one until broader agricultural issues are addressed.36 36. In May 2004, the newly elected Congress chief minister of Andhra Pradesh Y. S. Rajasekhara Reddy reversed his predecessor's tentative tariff reform and reintroduced free power to farmers in his drought-hit state, while the chief ministers of Tamil Nadu and Punjab, as mentioned earlier, reversed their own incip- ient reform policy to do the same. This was followed by the rival political formations in Maharashtra promising free power to farmers ahead of elections in that state. Earlier, (then) chief minister Digvijay Singh Can Good Economics Ever Be Good Politics? 23 However,thereisenoughawarenessnowto utilizesuchpolitical giveaways as oppor- tunities to trade off other aspects of reform such as: insisting that the burden of such subsidies be taken off the utilities and that they be made transparent by placing them on the state budget (thereby also forcing greater fiscal discipline on the state government); introducing measures to improve the financial management of util- ities; encouraging competition as envisaged by the Electricity Act; and (in a deal with the beneficiaries of free supply) installing meters and implementing anti-theft measures.37 Indeed, political strategists privately believe that if only they could cred- ibly assure quality power supply even at a price, this would carry greater weight with voters than these rash sops of free but erratic power.38 I ThefactthatnearlyallstateshavesignedontotheAcceleratedPowerDevelopment and Reform Program: this is admittedly a weak reform framework, but the trend is significant in a context of incremental reform. I The emergence of a committed and articulate pro-reform constituency in the bureaucracy and national media and the gradual percolation of the benefits of reform into the public domain through events like hearings of the (albeit weak) state regulatory commissions.39 I Thebroadacceptancesincethe2004electionsthattheinfrastructuresectorsarethe show-stopper in India's growth and poverty alleviation efforts, and the prominent place of this subject on the public agenda. I Asenseofexpectationamongthepowerfulanti-reformlobbies--farmersandutil- ity staff--that further reform is inevitable. That is, these groups are increasingly amenable to negotiation and compromise. I Finally,thepassageoftheElectricityActwhich,forallitsflaws,isawatershed. These factors might appear superficial and weak in comparison to those lined up in opposi- tiontoreform.Nevertheless,theyrepresentthebuildingblocksforcreativestrategy:theyindi- cate that those accused of obstructing reform are actually themselves seeking solutions. Of the various impediments to power reform described previously, at least two--accommodating of Madhya Pradesh (politically positioned as progressive and reformist), in the run up to elections in his state, on September 25, 2003, wrote off the previous three years of electricity dues for 1.2 million consumers (for farmers using pumps of up to 5 horsepower capacity and poor people with a single-point connection). Though Singh was forced to suspend his plan on a directive from India's autonomous election commis- sion which objected to the announcement as a violation of the election code of conduct, and was also defeated, his successor found enough political merit in the idea to implement it on assuming office. 37. As if to prove this point, Andhra Pradesh chief minister Reddy has been quoted by the Press Trust of India (The Indian Express, Delhi, August 23, 2004) as saying that free power will be given only to those farmers who have authorized connections. 38. Interview with a key political decisionmaker, who asked to remain anonymous, of the then ruling federal coalition, November 2002. More significantly, at the India Today State of the States Conclave held at New Delhi on August 6, 2004, at least three chief ministers belonging to different political parties-- Vasundhara Raje (BJP) of Rajasthan, Uma Bharati (BJP) of Madhya Pradesh (since resigned) and Virbhadra Singh (Congress) of Himachal Pradesh--categorically declared that what farmers and others wanted was "quality power, not free power." 39. An indication of changing public perception is available in the CSDS-Hindustan Times (2003) pre- election survey in Delhi. Despite the differing perceptions among classes, overall, 46 percent of respon- dents felt the power situation had improved since 1998, as opposed to 31 percent who disagreed. 24 World Bank Working Paper legitimate grassroots concerns and building credibility for the process--are within the reform lobby'scapacitiestoaddress.Thatis,theirresolutiondependslessontheusualsuspectofpolit- ical will and more on "reformers' will" to explore and negotiate options. Another major impediment--political protection of the farm constituency--potentially has solutions in the multi-sectorapproach,ininnovativemodelsthatempowercommunities,andintiminginter- ventions in accordance with the political cycle. Momentum gained through such approaches would reinforce the positive points just enumerated and eventually result in the isolation and gradual breaking down of the only real opponentsofthereform,thecorruptvestedinterests.Thisisnotanattempttopaintanoverly optimistic picture, but a pointer to the possible breaks in the impasse that appears to have gripped the power sector in India. Conclusions and Lessons Going Forward The key message that emerges from this long answer to the question we started with--Why is there a gap between the politician's reformist rhetoric and what he or she actually imple- ments on the ground?--is that a technocratic solution alone does not make a reform pro- gram. As we have seen, despite the clear public dissatisfaction with the present state of the Indian power sector and the sound logic of the reform formula itself, implementation has proved challenging and politically difficult. In hindsight, the weakness of the Indian power reform program has been that while it has focused appropriately on sorting out distortions in the relationship between the owner- government and power utilities through the unbundling and regulation model, it has failed to carry credible assurances that this will improve the equation between the reformed utili- ties and their consumers. In addition, it has not factored in the mutual relationships between consumers, its claimed beneficiaries, and the politicians whose behavior it is ultimately seek- ing to change. To set this in the framework of the World Bank's World Development Report 2004, which looks at service delivery as a three-way relationship between policymakers, service providers, and clients, the Indian power reform program has concentrated on redefining the policy- maker-provider (that is, government and utilities) compact, but has not addressed the provider-client (utility-consumer) and client-policymaker (consumer-politician) dynamic. In short, it has tinkered only with the technical side of the service delivery triangle. Clearly then, a politically feasible reform program needs to address the other two sides of the trian- gle as well. That is, the managers and devisers of the reform must recognize that each of the three relationships described above has an impact on the other two, and no one can be amended in isolation. Critical in this is recognizing the central role of the politician and legitimacy of concerns of the constituencies he or she represents, both poor and non-poor. At the same time, the emphasis must shift so that major stakeholders, including entrenched interest groups, are made to first own the problem, after which solutions can be sought and negotiated. In sum then, in order to build coalitions of support and mobilize beneficiaries, a reform program should have as its primary ingredients a nuanced understanding of the local polit- ical economy, a flexible and creative technical approach, and an integrated underpinning of strategic, multi-dimensional communication. It is possible that a reform program con- Can Good Economics Ever Be Good Politics?25 structed on these lines would help change the incentive structure for ruling politicians more fluently than India's present strategy of stealthy gradualism that must otherwise necessarily be the only way. References Agarwal, Manish, Ian Alexander, and Bernard Tenenbaum. 2003 (pre-publication version). "The Delhi Electricity Discom Privatizations: Some Observations and Recommenda- tions for Future Privatizations in India and Elsewhere." World Bank Energy and Min- ing Sector Board Discussion Paper Number 8. Available at www.worldbank.org/energy. Dakovic, Tonci, Bernard Tenenbaum, and Fiona Woolf. 2003. "Regulation by Contract: A New Way to Privatize Electricity Distribution?" World Bank Energy and Mining Board Discussion Paper Number 7. Dubash, N. K., and S. C. Rajan. 2001. "Power Politics: Process of Power Sector Reform in India." Economic and Political Weekly, September 1, 2001, pp. 3367­3390. Echeverri-Gent, John E. 2000. "Politicians' Incentives to Reform: Globalization, Partisan Competition, and the Paradox of India's Economic Reform." Draft available via http://www.people.virginia.edu. IDC (Institute for Development and Communication). 1999. "Power Reforms in Haryana: Communication Need Assessment Survey." Chandigarh. Irwin, Timothy, and Chiaki Yamamoto. 2003. "Improving the Governance of State-Owned Power Utilities: Some Options and Evidence Drawn from International Experience." Background Paper prepared for the World Bank's South Asia Energy and Infrastructure Unit. Jenkins, Rob. 1999. Democratic Politics and Economic Reform in India. Cambridge, UK: Cambridge University Press. Kanbur, Ravi. 2000. "Economic Policy, Distribution and Poverty: The Nature of Dis- agreements." Invited presentation to the Swedish Parliamentary Commission on Global Development on September 22nd. Kelkar, Vijay, and Ajay Shah. 2003. "Why is This a Very Happy Diwali? Top Answer is Reforms." The Indian Express (New Delhi), October 22, 2003, p. 1. McMillan, Margaret, Dani Rodrik, and Karen Horn Welch. 2002. "When Economic Reform Goes Wrong: Cashews in Mozambique." Working Paper Number 9117. Cam- bridge, Mass.: National Bureau of Economic Research, Inc. Mukherji, Rahul. 2004. "Managing competition: Politics and the building of independent regulatory institutions." Revised version of paper for the SAIS South Asia Studies Sem- inar: India's Next Generation Economic Reforms--Sectors, Actors and Prospects, Washington, D.C., May 17, 2000. Narayan, Jayaprakash. 2000. "Energy Sector Reform and Governance." Loksatta Times 1(2):3­5. "Overview: Charging Ahead with Reforms." 2003. Indian Infrastructure 6(1):18. Rao, S. L. 2002. "The Political Economy of Power." Economic and Political Weekly, August 17, 2002, pp. 3433­3444. Reddy, A. K. N. 2001. "Indian Power Sector Reform for Sustainable Development: The Public Benefits Imperative." Energy for Sustainable Development 5(2):74­81. 26 World Bank Working Paper Revi, Aromar. 2002. "India's Power Sector Reforms: Managing the External Environment." Presentation at the CRISIL Power Sector Seminar, New Delhi, September 20th. Rodrik, Dani. 1996. "Understanding Economic Policy Reform." Journal of Economic Litera- ture 34:9­41. Ruet, Joel. 2001. "Winners and Losers of the SEB Reforms: An Organizational Analysis." New Delhi: Centre de Sciences Humaines. TARU Leading Edge. 2001. Report on Baseline Survey (The Andhra Pradesh Power Sector Restructuring Program Socio-Economic Assessment). Hyderabad/New Delhi: Pre- pared for UK Department for International Development (DfID). Team HT-CSDS (Centre for the Study of Developing Societies). 2003. "More Power to the People: Supply Better, Say Voters, but Private Players Get No Credit." Hindustan Times 25(October):2. Transparency International India and ORG-Marg. 2002. "Corruption in India: An Empiri- calStudy."Availableathttp://www.ti-bangladesh.org/ti-india/documents/corrInd.htm. Vajpayee, Atal Behari. 2002. Independence Day speech. Williamson, John, and N. Roberto Zagha. 2002. "From the Hindu Rate of Growth to the Hindu Rate of Reform." Working Article Number 144, Center for Research on Eco- nomic Development and Policy Reform, Stanford University. World Bank. 1995. Bureaucrats in Business: The Economics and Politics of Government Own- ership. New York: Oxford University Press. ------. 2003. World Development Report 2004: Making Services Work for Poor People. Washington, D.C.: Oxford University Press. Yadav, Yogendra. 2002. Interview with the author. December. Eco-Audit Environmental Benefits Statement The World Bank is committed to preserving Endangered Forests and natural resources. We print World Bank Working Papers and Country Studies on 100 percent postconsumer recy- cled paper, processed chlorine free. The World Bank has formally agreed to follow the rec- ommended standards for paper usage set by Green Press Initiative--a nonprofit program supporting publishers in using fiber that is not sourced from Endangered Forests. For more information, visit www.greenpressinitiative.org. In 2005, the printing of these books on recycled paper saved the following: Trees* Solid Waste Water Net Greenhouse Gases Electricity 463 21,693 196,764 42,614 79,130 * Pounds Gallons Pounds KWH 40" in height and 6-8" in diameter Can Good Economics Ever Be Good Politics? is part of the World Bank Working Paper series. These papers are published to commu- nicate the results of the Bank's ongoing research and to stimulate public discussion. This publication is the third in a series of Working Papers sponsored by the Development Communication Division (DevComm) of the World Bank's External Affairs Vice-Presidency. This series is designed to share innovations and lessons learned in the applica- tion of strategic communication in development projects. Together with other donors, NGOs, and private sector partners, DevComm seeks to mainstream the discipline of development communication in development practice. In recent years, the power sector in several developing countries has suffered from a frustrating gap between strong, pro-reform rhetoric at the political level, and weak, hesitant implementation of the reform measures on the ground. Focusing on the recent experi- ence of power sector reform in India, this paper looks afresh at the problem of the "rhetoric-implementation gap" by taking the lack of political will as its starting point, and identifying the ingredients that comprise it in the current context of India. Assuming that peo- ple and institutions are not impartial but instead respond to politi- cal and economic incentives, it explains how the lack of political will often reflects rational political behavior. Using this more realistic framework, it examines the incentives, informal relationships, and interests that govern the behavior of people and institutions, and searches for the openings and opportunities that reformers must pursue. World Bank Working Papers are available individually or by subscription, both in print and online. DevComm ISBN 0-8213-6680-7 THE WORLD BANK 1818 H Street, NW Washington, DC 20433 USA Telephone: 202 473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org