WPS5855 Policy Research Working Paper 5855 Enrichment with Growth Michael Klein The World Bank Sustainable Development Network Finance, Economics and Urban Department October 2011 Policy Research Working Paper 5855 Abstract This essay first sets out the “business model� problems wealth diverted may be smaller. The most effective pro- entailed by corruption and their effects as well as growth polices that help enrich an elite resemble fairly implications for economic growth. Key issues are the orthodox economic policy prescriptions. Eventually need for secrecy and co-operation with partners in crime. the abolition of corruption holds the greatest promise Dealing with these leads to behavior which is ostensibly to enhance growth and with it the wealth of elites. The bizarre and undermines economic efficiency, but is expectation of such growth may explain why more and in fact a rational way of managing corrupt practices. more political elites pursue “sound� economic policy and However, different economic policies can be pursued may embrace anti-corruption efforts, while securing legal that are compatible with corruption. Some are more ways of enrichment for themselves. Country examples pro-growth than others. Pro-growth corrupt policies illustrate policy approaches with different combinations hold the promise of enriching the corrupt elite more of enrichment and growth properties. in absolute terms even though the share of national This paper is a product of the Finance, Economics and Urban Department, Sustainable Development Network. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at mklein@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Enrichment with Growth Michael Klein 1 JEL classification: D73, O17 Key words: Corruption, economic growth 1 Professor, Frankfurt School of Finance and Management, Frankfurt, Germany, and Senior Visiting Professor, Johns Hopkins School of Advanced International Studies, Washington DC. Former Vice President, Finance and Private Sector Development, The World Bank. Table of Contents page Introduction 3 Extracting Bribes and Effects on Economic Performance 3 Finding partners 5 Managing corrupt relationships 7 Monitoring and accounting 8 A country case: Nigeria 9 Improving the Corrupt Business Model 11 Ownership of corporations 12 Sharing in portfolios of firms 13 Corruption and the end of history 15 Concluding Remarks 17 2 Introduction2 In 1974 the World Bank published a prominent study: “Redistribution with Growth�. It set out thinking about achieving the Holy Grail of development economics – how to obtain overall growth and a better life for all at the same time. The developmental discourse faces two issues. Why would powerful people heed the call of technocrats for policy reform when it does not explicitly aim at assuring that they come out well? Not only that, but standard developmental discourse has recently emphasized that corruption is a scourge that prevents countries from growing. If this type of enrichment of the powerful may actually be at odds with technocratic policy advice, why follow the advice? Many people actually running bits of the world want power and wealth. The spectacular growth performance of many countries across the globe has shown that traditional ways of ruling and robbing a country may not be very effective at reaching those goals. How then can enrichment best be pursued by powerful elites? One might call the goal of the greedy ruler “Enrichment with Growth�. In the following I set out arguments that a Consigliere to the “rich and shameless� might want to think about. First, I discuss key mechanisms that characterize “traditional� corruption. Second, I set out an approach to enrichment that nowadays seems to be followed by many autocrats – ownership of professionally managed corporations – and its limitations. Third, arguments are provided for a potentially more promising approach to enrichment, one that looks suspiciously like some forms of industrial policy. Fourth, the most effective form of enrichment is described – running an economy like a portfolio manager with decent corporate governance. Finally, the paper concludes with perspectives on the debate about enrichment of powerful people, growth and legitimacy. Extracting Rents and Effects on Economic Performance Corruption is supposed to make the corrupt better off. In doing so it may undermine the welfare of others, not just to the degree of direct stealing, but by destroying the very basis for economic activity. It is, for example, utterly clear that systems of what Mancur Olson called “roving bandits� fail (Olson, 2000). Competing rent-extracting “warlords� destroy the basis for wealth and end up with little wealth themselves compared to what can be obtained in well-functioning countries (see Shleifer and Vishny (1993) for a basic model). The warlords themselves are typically not terribly rich by the standards of today’s world and they are somewhat limited in enjoying the fruit of their stealing. They might be interested in 2 I am indebted to Alexander Libman and Ignacio Mas for helpful comments 3 developing the economy, if they can benefit in the bargain. Even in less dysfunctional systems corrupt elites see what is apparently possible in today’s world. Countries like China and India are still plagued by corruption; yet they grow extremely fast. The ideal for the corrupt would be to support general economic growth in a way that allows them to make more in absolute terms. The basic arithmetic is simple. Assume you want to make money off extracting wealth from a country. In a country with the income level of the Democratic Republic of Congo you need to extract 100 percent of GDP to obtain the same amount of money that extracting just 1 percent in a country like Korea gives you (assuming equal populations). In Tunisia one would need to collect 5 percent and in Nigeria 15 percent of GDP for the purpose. All this simply reflects the level of income per person in the countries. Economic analysis has at times suggested that corruption drags down growth (see Mauro, 1995 for an initial empirical study). Yet, the progress of analysis has left us with the finding that it is hard to tell how corruption affects growth (Svensson, 2005). In particular, it is by no means clear that all forms of corruption are inimical to growth. There is thus hope for corrupt autocrats to find ways to enrich themselves even more by modernizing at the same time. The following discussion explores basic features of policies that may make this possible. First, consider how the mechanisms corruptly to extract rents often undermine economic growth. Shleifer and Vishny (1993) argue that one feature of corruption that leads to pervasive ill-effects is the need for secrecy. Corruptors would thus tend to favor projects where corruption can be easily hidden, but which may not be efficient. Second, corruptors would want to enhance the ability to extract rents and to keep others out of the corruption racket. Hence they would act against new entry and thus undermine innovation in an economy. In the following I explore this line of thought further. In doing so I use a number of anecdotes derived from my work in Nigeria in the 1980s and early 1990s3. Let’s start by considering the “business model� of the corrupt. Those who wish to enrich themselves need to solve several problems. They need to find partners - partners to obtain favors from and partners to collaborate with to obtain such favors. They then need to monitor whether the partners stick to their commitments and enforce the commitments. They need to do this largely without the help of formal institutions such as contracts or courts. 3 Intriguingly Nigeria has improved in recent years. Growth accelerated, particularly that of the non-oil economy, which virtually stagnated during the oil booms of the seventies and early eighties. Aggregate perception indicators suggest that among countries with some of the worst reputation for corruption Nigeria has made exceptional progress (The Globalist, March 10, 2011) 4 Finding partners Somehow the corrupt partners must find each other. To some degree the formal power structures in a government or corporation simplify the match-making. It is usually well-known which formal positions go hand in hand with the ability to dispense favors and punish defection. Still, it is not always clear whether all people in position of power are corrupt. Also, match-making cannot always be done in straight-forward ways for at least residual risk of prosecution, although this varies, of course, with the degree of control corrupt parties have over formal institutions such as the police or judges. Here it can help to create and maintain a general atmosphere of corruption. Paradoxically, even a somewhat vocal press can help make the point that corruption is widespread. As long as the allegations are not too specific, sometimes outlandish and not verifiable in a court of law this may help prepare the terrain by biasing the people interested in doing business in the country in question to those willing to be corrupt. To some degree, therefore, public accusations of corruption may in fact facilitate the establishment of corrupt business relations. Nigeria’s rambunctious somewhat uncontrollable press may just have played this role. Accusations of corruption abound, yet little is done and many accusations may just be fabrications – a strategy of disinformation. Where everybody expects corruption, many will choose to play the game. Thus expectations are validated. That leaves the issue of identifying specific individuals with whom to enter into corrupt relations. Some form of ostentatious wealth relative to the rest of society, like an expensive house or car, may help signal that a particular individual might be corrupt. At the same time the level of ostentatiousness needs to be calibrated such that it might just plausibly be due to legitimate sources of income such as true business success or pre-existing family wealth4. Such signals pave the way to conversations that can establish willingness “to deal�. As in the case of gangsters who seek partners (Gambetta, 2009), it is not straightforward to establish relations of trust. There need to be places where potential partners can meet for legitimate reasons and where they can gauge each other’s dispositions. The assessment of whether to trust someone to be willing to commit an illegal act tends to be a drawn-out process, as potential partners observe a host of subtle clues about each other, before broaching the true deal directly. A typical setting for meeting and assessing potential partners would be at the private residence of important intermediaries, where streams of visitors of all types from official agencies, politics or businesses come and go throughout the day. The setting is relaxed; no obvious agenda is visible 4 There is a corruption joke that goes as follows: The minister of public works in country x visits counterparty in country y. He gets invited to the sumptuous private home with pools, gardens, tennis courts etc. Asks: how can you afford this on your official salary? Answer by minister from y: “see that road over there?� – minister from x says: “yes!� - minister from y says: “15%!� The year after minister of y gets invited to minister in country x. Visits private home of counterpart – even more grandiose and full of fancy cars. Asks: How can you afford this? Answer: “see that road over there?� – “no!� – “see, 100%!� The joke is funny, because stealing 100% would mean an action beyond plausible deniability and without a “developmental� smokescreen. 5 other than a friendly visit. Chance conversations are possible – an ideal place to meet and subtly explore. Before a deal can be done between the principals in a corrupt transaction, they might use intermediaries to prepare the terrain. Intermediaries are matchmakers. They know corrupt people. They may have long-lasting relationships with multiple parties on both sides and are thus able to build trust. In some contexts their contacts and relationships may even allow them to be more stable partners than changing top policymakers or changing CEOs. (Think “Klaus Schwab� and “World Economic Forum� for a legitimate variant of matchmaking.) The use of intermediaries also allows the principal some plausible deniability until the final deal is closed. They might get their income from fees that can look legitimate – the “consulting fee� - without actually participating directly in grand corruption. In fact, intermediaries may command extra trust, if they stick to their normal “fee schedule� and clearly stay away from asking for a special cut in deals. Deal clinching meetings follow. In one case, the CEO of an engineering company sought to meet the managing director of a large public enterprise. After long periods of petitioning for a meeting and waiting, he finally was ushered into the office of the managing director, who sat in his chair behind the desk and faced the wall. For four hours he remained in silence while the radio played. Then he turned. A deal was closed in the ensuing meeting5. At first sight such behavior is bizarre. Yet, “waiting games� may serve two functions. They select people who are interested to wait and thus seem sufficiently desperate for a deal. Silence also serves to place pressure on the waiting party to show his hand, possibly by offering a “good deal�, thereby revealing willingness to enter into corrupt relations. At the same time offering to make corrupt payments provides a form of compromising evidence that exposes the briber6. The complex process of match-making explains some prima facie “weird� behavior. It also means that it takes time. The only study I am aware of about the actual transaction costs (excluding technical or legal studies) of making deals for large private infrastructure projects shows these costs can be substantial. Accordingly, normal transaction costs tend to amount to some 3 percent of project cost, whereas in complicated environments these costs can go often to 10 percent or more (Klein, So and Shin 1996). The extra costs are essentially due to time wasted by highly paid lawyers or contractors waiting for meetings, not extra costs of writing complex contracts. It is normal to argue that the time spent so inefficiently demonstrates the low capacity of some governments or their disorganization and lack of political direction. In fact, it may be that the same observable phenomena reflect rational strategies to explore the potential for corrupt transactions. 5 I have no evidence that corrupt payments played a role. Yet, the behavior would just be bizarre, if it was not at least an attempt to assess the readiness to be corrupt. 6 The typical strategy to elicit bribes in the Hotel I used to stay in in Lagos would be to let arriving visitors wait without giving them the room they booked. 6 Managing corrupt relationships The costs of establishing corrupt relationships militate against expanding them too much. Moreover, keeping the number of parties to a corrupt relationship limited helps managing it. Fewer people may be more easily monitored and they present fewer chances for whistleblowing. In addition, it is helpful, if it is always the same few people so that repeated interaction can limit the incentives to renege on agreements. Importantly, the group of participants has an incentive to obtain and maintain incriminating information on the others to be able to retaliate should one renege. Providing such information to each other voluntarily is actually helpful for cementing cohesion. Loyalty is important so that whistleblowing and reneging is unlikely. Hence the use of family members in corruption rackets or members of the same ethnic group. If those participating in corruption rackets are not very competent at doing anything but being corrupt, loyalty can be further enhanced. Those who have no promising alternative employment have lessened incentives to be disloyal. Corruption becomes detrimental when multiple parties ratchet up bribe demands for complementary services they provide such as protection from the police together with a license for construction or when multiple “police forces� all ask for protection money without being able to ensure protection from the competing police. In this case, rent-seeking will erode gains from corruption. Hence it is rational to maintain de facto “fee schedules� of who needs to be paid and with how much. It is also necessary to keep new entrants out and to arbitrate disputes should they occur. Allocation of funds and arbitration may be easiest, where sources of corruption (“spigots� – for example, rents from oil, gas and mining operations) can clearly be allocated to as few parties as possible. Physical control over the source of corruption provides the best assurance that agreements can be enforced. We might thus expect contracts to go to areas controlled by the principals in a corruption racket. The process of reaching agreement, allocating spigots and arbitrating disputes may lead to uneconomic delays. The various partners in a corruption racket presumably do not trust each other a whole lot and might require lengthy discussion and review to come to a shared view. Grand corruption requires large amounts of money to be diverted. To keep the number of parties involved limited both for reasons of enrichment by the principals and to keep the chances of whistleblowing down, it may be useful to go for large projects when feasible. It may be a bit like the choice of beverages in places where alcohol is officially prohibited. One then tends to find a disproportionate share of drinks with very high alcohol content – more bang for the buck per unit of smuggled merchandise. Also, by choosing large, complex contracts to let, the use of non- standard procurement methods becomes prima facie reasonable and cost comparisons are rendered difficult (Shleifer and Vishny, 1993). 7 What types of activities lend themselves to grand corruption? It is sometimes argued that it is monopoly businesses where lots of rents can be extracted. However, it is not clear that this is the main mechanism. Extracting rents means monitoring profit and loss statements and charging profit-maximizing prices. The very information contained in comprehensive accounts that allow this to be done, makes it more likely that corrupt practices will be found out. High profits or high prices are very visible signs that attract attention of critics. For example, it is standard to argue in textbooks on infrastructure regulation that rate-of-return regulation is less likely to be considered illegitimate, because it seems to be better at rent extraction for the consumer than price cap regulation, which may lead to observable high ex-post profit, even when all-in costs and prices may be lower. In reality, effective prices are typically too low in corrupt countries – below full cost and often even below operating costs. Grand corruption often seems to go together with stealing from government funding of large projects and substandard service rather than high consumer prices and corporate profits. Low prices conveniently provide a “developmental narrative� that surrounds projects. Such narratives are a part of corrupt strategies. The narrative remains popular even though many studies suggest that the distributional effects of low prices tend actually to favor better-off people. Low prices also mean that funding has to come from government sources. If an infrastructure venture were dependent in full on payment by customers and funded by private debt markets, reasonably solid accounts would need to be maintained, which would increase the likelihood of finding out wrong-doing7. Corrupt payments may thus more easily be taken out of public funding, not privately paid revenues. Monitoring and accounting The discussion suggests that basic accounting is an enemy of corruption. It also highlights that a measure like profit exerts relatively tight management discipline, because, to be derived, it requires consistent and comprehensive accounting. Nigeria provides apposite anecdotes. In 1978 a scandal broke for the national oil company, NNPC, facing allegations that it had diverted something in the order of 2.8 billion dollars. The allegations were never proven, not least thanks to the inexistence of accounts for the company. A public commission that dealt with the matter recommended in 1980 among other things that NNPC keep proper accounts. Consultants were hired. Yet, accounts kept being late by more than a year – unsuitable for management. In the mid-1980s a fire broke out in the accounting department and set the effort back further. The breaking out of fires, not in the kitchen, but in the accounting department had a good tradition in the country. West Africa’s then tallest tower, the NITEL building in Lagos, stood charred for many years from a fire that had broken out in its accounting department. A lack of accounts prevents serious management and the lack of control is acutely felt by management. A colleague and I once developed a financial model of the Nigerian National 7 In addition, as is often argued taxpayers may be less organized and able to resist than other social groups. 8 Petroleum Corporation (NNPC) based on public data on oil production, global oil prices, standard cost estimates and cost patterns from outdated accounts of the company. The then Managing Director of NNPC was so interested that he spent four hours discussing the financial model. In no other case was a person at this level willing to spend that much time discussing substantive issues without even a hint of a deal on the table. While comprehensive accounts are problematic for corrupt practices, some form of monitoring is needed on whether bribers paid and whether bribes received their agreed share. When IMF staffers once probed the fiscal accounts of Nigeria they could not reconcile numbers. When pressing for explanations, they were - allegedly - in the end able to see an accounting of major expenditures maintained at the very center of government in an exercise book. While this is no way to keep professional accounts, it might be a rational method for keeping track of corrupt payments. Such informal accounting need not be as comprehensive as full accounts, but just needs to track critical payments. It need just be used for monitoring of corrupt agreements not for managing a company or government finances. A Country Case: Nigeria What might be the overall outcome of a system that embodies the features discussed above for a whole country? Consider Nigeria. One can easily document a few striking patterns. First, policy-making for large projects has been extremely stable – no external shock, no shift in relative prices used to dislodge or reshape projects, once they were agreed. Until recently, almost all large projects were already contained as proposals in a consultant report by Arthur D. Little in 1964. They progressed inexorably, albeit excruciatingly slowly at times. The poster child for the syndrome is the multi-billion dollar Ajaokuta steel plant: conceived in 1964, started in 1979 with outdated Soviet technology, completed in 2004, shut down again and re-opened in 2010. Official commission reports occasionally recommended that several projects be stopped, including Ajoakuta – with no effect8. The slow progress may reflect periodic shortage of funds, shift in parts of the economics and also hold-ups by important players in the system. The Bonny LNG project provides a case in point. In 1964 a plane with buyers willing to sign off-take agreements for the gas was on the tarmac in Amsterdam ready to fly to Lagos. Yet, the then petroleum commissioner from the North of the country stopped the process and ordered a review, during which time North Sea oil was discovered. This killed the economics of the project for many years. It was not until 1999 that the plant became operational. The development of large infrastructure projects followed similar patterns. For example, hydro-power plants continued on course, even though gas would have been cheaper (Klein, 1987). 8 For example, the Onosode Commission of 1984 9 Second, the key players in the system have remained in important positions for very long periods of time. To this day candidates for top political positions are old-timers, such as Muhammadu Buhari, author of the 1983 military coup, running for President in 20119. George Obiozor of the Nigerian Institute of International Affairs once called Nigerian politics “a game of rocking chairs, a game in which unlike in the game of musical chairs, there is no movement�.10 This is at odds with a country often depicted as rocked by coups and erratic political change. Yet, it may be that elites at the top rotate, while the underlying power structure remains little changed. Third, large, complex projects were distributed across the major parts of the country, for example, one refinery in the East at Port Harcourt, one in the West at Warri and finally one in the North at Kaduna, some 800 km from the nearest port. Roughly this corresponds to the territories of the three main ethnic groups of the country, the Yoruba, Hausa and Igbo. Early on in 1962 the approach was to provide different regions of the country with equity shares in refineries in economically efficient locations. However, this proved not to be a stable solution politically. The results were bizarre projects. The Kaduna refinery was designed to produce specialty products that others did not produce, including asphalt for road construction. The refinery was built. Then the country had to import heavy oil to obtain the desired product slate via roughly 800 km of oil pipelines. When asphalt was finally produced the government ran out of money following the decline in oil prices in the mid-1980s. The response was to try exporting asphalt by rail back to the port ostensibly to obtain valuable foreign exchange. It was like buying a house knocking it down and selling the parts. Fourth, as the system is focused on extracting rents from funding of complex construction projects and undermines management capabilities and incentives, it may easily lead to provision of poor quality services. Whereas proper accounting allows one comprehensively to “follow the money�, the causes of substandard quality are harder to pin down. For example, officials in Nigeria almost reflexively blame suppliers for sub-standard quality of equipment or construction regardless of the real reason. They might be right, they might not be. One cannot easily find out. Funds for maintenance are lacking. Operations by local staff in government firms have often been deficient. The country is full of private fallback solutions for deficient infrastructure services like standby generation or private water vendors. Finally, the various arrangements that make corruption manageable provide a rational explanation for often observed uneconomic decision-making, weak “capacity� and the like. Reviews of Nigerian policies by the country itself have routinely come out with views that “priorities are not properly defined, planning seems to be synonymous with aggregation of projects, and coordination by government departments is poor�. This particular assessment was made for NNPC by a managing director. Hence, top managers understand the issues. At the same time Nigerian staff performs effectively in different settings, for example, in the private oil 9 For a series of examples see Klein (1987) 10 Newswatch October 6, 1983 p.25 10 companies in the country. Key managers of NNPC were in fact drawn from the private oil companies. Yet, capable and rational people have a hard time operating effectively in a system that militates against sound management information, that tries to limit people in the know, compartmentalizes information and may promote staff into key position who are loyal and not so competent11. As a result, possibly of a corruption driven syndrome, project choices have been uneconomic, massive delays abound, operations have been inefficient and service quality has been sub- standard. Critical inputs for the economy as a whole were either not provided at all or in sub- standard quality. Funds syphoned off have been of significant amounts and presumably been spent abroad to an important extent. Improving the Corrupt Business Model Running a traditional corruption system is possible but it undermines economic growth. Bad projects are chosen. Quality of service is reduced. Sound management is undermined. To up the game, new forms of enrichment are needed. Many powerful people dream of making their country richer, more “competitive�. They send their children to be educated at Harvard. They seek medical treatment in Singapore. They know that, if there was higher productivity growth in their economy, there could be so much more wealth going around, including for themselves. The basic challenges in establishing an effective system of corruption remain. Partnerships have to be formed. Agreements have to be monitored and enforced, all this without the explicit help of official institutions and with at least plausible deniability. The challenges do not become easier if one seeks a more efficient system. Better project choices are needed as is better management. It means monitoring will be easier and secrecy harder to maintain. Yet, corrupt elites the world over are ready to embrace the challenge. So what would the strategy of powerful people look like, if they wanted to combine enrichment with better growth? Somehow they need to be directly interested in generating growth. This requires waiting until efficient projects produce profit derived from productivity growth. It means preventing others from breaking ranks and from stealing upfront in ways that undermine the achievement of productivity growth. Andrew Wedeman (1997) has called this the search for “dividend-collecting� systems of corruption. 11 Intriguingly, the counterparts of one of the authors in the procurement process for a large gas pipeline seemed genuinely proud at the end of the somewhat tortuous process for having been part of at least a relatively clean process. 11 Ownership of Corporations One thing about economic growth seems clear. One needs to have dynamic firms with professional management. Compared to the previous discussion of a corrupt business model, the corrupt need to surround themselves not just with people who are loyal, but also competent and thus with outside options. Professional managers may jump ship when they see dubious behavior or they have the potential to apply their competence to feather their own nest to the detriment of the corrupt “principal�. This places a special burden on supervision of the professional by parties loyal to the principal, for example by family members of the principal. The number of suitable family members or otherwise close associates limits the manageable control span and thus the size of firms. Supervision might also fail, because the supervisor might not be able to understand enough of what managers are actually doing. Hence it becomes particularly important to obtain and keep incriminating evidence on them. Infractions like tax evasion, evasion of regulatory requirements or other personal dirt come in helpful to build an incriminating dossier. The challenges of hiding corrupt activities become more “market-oriented�. At one level they become easier. Once a corrupt principal has gained ownership of a firm, he has a legal right to derive benefits from it. Much effort thus goes into acquiring or disposing of assets by manipulating valuation. For example, the acquisition of ownership rights and the purchase or sale of other assets is preferably conducted through negotiation rather than some form of effective auction. This provides one explanation for the popularity among politicians of negotiated deals for privatization or, for example, large infrastructure contracts such as Independent Power Providers. Disguising beneficial ownership can help obscure the matter further. Other such time-honored ways of bad corporate governance can also be used for stealing such as using related party transactions for “tunneling�. Trying ones hand at owning corporations has become very popular for important political leaders in many countries. It has been tried with varying success by leaders across the world. Control of business empires is often exercised by family members, who might provide a degree of loyalty. Elsewhere the ruling party may play that role. Famous examples are Suharto or the King of Morocco. Businesses in Indonesia with connections to Suharto suffered systematic share price losses in response to news of Suharto’s health (Fisman, 2001). In the transition to market economies, leaders in countries of the former Soviet Union or in China or currently in Africa promote the move to more market-oriented systems while obtaining ownership rights in new fully or partly private companies. A bit of emphasis on firms and professional management has provided some promising results – compare Nigeria and Indonesia. Yet, the recent troubles in Tunisia exemplify some of the limits of the approach. Important enterprises in the country were controlled by the family of the deposed ruler, Ben Ali. Tunisia did well economically relative to other countries in the region that do not benefit from oil. However, growth remained modest. While the country had 12 improved its policies, the heavy hand of government remained palpable for all who wanted to do business, from street vendors to larger players attempting to enter markets. The essential limitation of the “corporate model of corruption� is its attitude to entry and exit. The ruling elite, once transformed into owners, likes keeping competition out (Shleifer and Vishny, 1993). For growth to be more dramatic and to be sustained, the forces of creative destruction need to be allowed to play. It is then not clear that firms owned by powerful people will survive and with them the rents they were meant to provide. Sharing in Portfolios of Firms It seems that it is not enough to invest in a firm and try the professional way. If one also needs to pursue competition, the Holy Grail for a greedy ruler would be to “invest� in a portfolio of companies in a dynamic economy. Still corrupt people need to obfuscate information and maintain discipline in the system. Assuming that a solution exists, what might some features look like ex post? It seems plausible to expect that a country that found a successful “portfolio approach� to corruption does not possess significant natural resources. There would then not have been the easy option to steal resource rents. Instead, the only way of generating rents would have been via productivity growth. The core corrupt system itself might consist of a limited number of powerful players with restricted entry into the system. At the same time there needs to be a market that selects good performers from bad ones. The measurement of success in that market needs to be compartmentalized so as to make it hard to trace money flows. Imagine you create some large conglomerates and limit competition at home. Then tell them that favors depend on success in export markets. Such success tends to reflect productivity growth thus creating gains that can be distributed among the corrupt. Monitoring export success can be done independent of keeping good accounts. Export numbers are partial indicators of financial performance, but de facto often correlated with success. They can also be cross-checked relatively easily by looking at import numbers abroad. To maintain discipline the interplay between taxation, regulation and firms provides means for hostage taking. First, higher taxes or complex regulations may make it more tempting to avoid them and thus provide fodder for incriminating dossiers. At the same time compliant firms may easily be able to avoid trouble with the law and thus not be unduly hindered as long as they “behave�. Second, in return for taxes and favors authorities may provide firms with benefits - subsidies of some form. Alternatively, one could just extract the net difference between taxes and favors on the one hand and subsidies on the other. However, asking for higher taxes with the 13 option of paying some back in the form of subsidies is a way of taking “hostages� to enforce corrupt agreements. If one also controls the banks and allows numbered bank accounts one has a way of moving money in ways that are hard to trace. These various features are what one used to observe ex post in Korea. The government aggressively promoted the Chaebols, Korean conglomerates, under Park Chung-Hee in the 1960s. Business associations helped manage state-business collaboration. Initially, the goal was a “self-reliant� import-substituting economy under state control. Control was exercise through the banking system via preferential terms for credit. Even when banks were formally private, managers were appointed by the government. The use of anonymous accounts in banks was allowed. Accounting in Chaebols allowed detailed money flows to be obscured. A critical later innovation by the Park regime was that favors were provided to the Chaebol, only if they performed in export markets and paid unofficial political contributions (You, 2005). This set up a contest that provided incentives for the Chaebol to raise productivity (Stiglitz, 1996). “Cultivating good relationships with the President became a golden road for business expansion� (You, 2005). The good relationships were punctuated by occasional prosecution of business owners or politicians for tax evasion or corruption, with cases typically dropped later, possibly because the desired disciplinary effect had been achieved. In one case the owner of Samhak, one of the largest conglomerates in the late 1960s, supported the opposition candidate Kim Dae-Jung in the 1971 election. He was convicted of tax evasion and the conglomerate forced into bankruptcy. The outcome was spectacular economic growth and probably rising amounts of corruption in absolute terms till the early 1990s (based on data on actual corrupt payments culled from newspaper reports and trials – see You, 2005). One might thus potentially interpret Korean industrial policy not just as a clever technocratic response to market failure, but also as a device to allow corruption to function in a system that generates productivity growth. Shleifer and Vishny (1993) pointed out that the Korea case was a puzzle. Maybe the design of “industrial policy� holds the key to the answer. Yet, even the “Korean Model� has limitations. During the Asian crises of the late 1990s problems of corporate governance were prominently displayed. The structure of corporate arrangements placed minority shareholders at risk. Problematic financial practices such as excessive leverage came to haunt corporations. The large corporations were hard to understand and difficult to control and restructure. Close ties to politics may have exacerbated excessive risk-taking in the hope of political support in case of trouble. Limited transparency has been a legacy of the past. In addition, one of the macro-economic outcomes of exposing corporations to competition in export markets, while protecting them at home, is a dualistic structure of the economy – world class corporations honed in global competition live side by side with low- productivity companies catering to a protected domestic market, just like in Japan. 14 How to do better then? What would be needed is an approach that provides for participation of the elite in a well-diversified portfolio of corporations that are exposed to competition and governed in a professional manner. To prevent professional managers abusing their control over corporations, effective corporate governance would need to protect shareholders. All the while the ruling elite would ensure that there be a policy environment that enables business to function well. The destructive features of traditional corruption would need to be eliminated. The remuneration of the elite would need to come from profits of the portfolio of companies. To be able to maintain control over the system, a level of state ownership of critical players in the system, such as critical holding companies, would be useful. State ownership provides extra discretion to elites to assign positions and rights to income from firms and to discipline those breaking rank. These requirements are met, for example, by Singapore. The economy is organized around large state companies that act as holding companies for a range of businesses. Temasek holdings is renowned for its professional competence and invests in a global portfolio – the ultimate portfolio for sustained enrichment. Former government ministers, top civil servants and their families benefit, holding well-paid positions in these companies. Government ministers are among the highest paid in the world. The prime minister earned US$ 2.9 million last year and awarded himself an 8 months bonus. Singapore has one of the best business environments in the world. Corruption has been tackled effectively. Corporate governance standards are high. At the same time income distribution in Singapore is among the more unequal in the world. Corruption and the End of History In the literature on corruption there is a debate about whether corruption (or lack thereof) drives economic growth or whether the level of economic development (or lack thereof) drives corruption. The arguments of this paper suggest that corruption can drag down economic performance significantly. The expectation of economic growth can, however, change the form that corruption takes. The corrupt have incentives to adopt growth-oriented policies. Indeed growth seems compatible with some forms of corruption and that sort of growth need not lead to lower absolute enrichment of elites. In the process legal forms of enrichment supplant illegal ones. Yet, the distribution of positions of power remains a murky area. Switching enrichment regimes requires policy shifts. Policies like privatization or public-private partnerships can be part of the transitions from one regime to another. So can industrial policy. In normal technocratic discourse, the respective proponents of privatization on the one hand and industrial policy on the other tend to be considered to lie on opposing sides of the ideological spectrum. From the perspective of elites in search of riches both recipes may come in handy. Making the switch to more “productive enrichment� regimes can be hard. For example, political scientists have argued that land owners in the Philippines have prevented reforms that occurred 15 in countries with effective land reforms like Korea and Taiwan, China (You, 2005). Failure to switch regimes provides a possible reason why some countries get trapped at particular income levels, not just “poverty traps�, but also middle income traps for countries like Tunisia. This is a variant of the arguments used by Mancur Olson who argued that initially successful systems may ossify, when interest groups prevent further desirable regime change including through “state capture� by influential businesses. Ultimately, the debate is about “using positions of power to extract rent�. Sometimes this involves illegal abuse of a public position for private gain, commonly called corruption. Sometimes this involves legal arrangements that happen to favor an elite12. Wealth creation is enhanced when people are rewarded on merit and not by virtue of a position of power they may happen to hold. Yet, the interest in maintaining position may undermine wealth creation. The ultimate requirement for wealth creation is allocating positions of power based on merit. Some rulers like Deng Xiao Ping and Lee Kuan Yew saw the need to step back and promote at least a modicum of merit-based succession. In the end, less personal ways of arranging for succession may turn out to be more effective. Democratic elections are a way to do so. Just like competition in markets, elections allow for challenge from new promising ideas and for an end to failed approaches. The process itself helps create legitimacy. Things can still go wrong, but at least one no longer needs to rely on the wisdom of powerful rulers. Just recently, in several elections during 2011 Singaporeans sent an unexpectedly strong message of distrust to the ruling party. Stellar economic performance has not been enough it seems. From the perspective of an advisor to the powerful, Francis Fukuyama’s thesis of the “End of History� remains relevant. As long as different jurisdictions race each other in search for “competitiveness� or productivity growth, democracy may find some unexpected allies. Democracy not only provides citizens with greater freedom, but may ultimately also be the most promising complement to efficient economic management. 12 I encountered an amusing, albeit damaging, example in Greece a few years before the current financial crisis. The government of the day tried to reform business registration for small and medium firms. In most countries this type of reform is popular and relatively easy to implement – not in Greece. The complex Greek registration system required a series of steps with fees being paid for several. The fees went to fund the pension fund of lawyers in Greece. Hence change was impossible. In substance the difference between unofficial and illegal “fee schedules� for bribes may be small. 16 Concluding Remarks First, this review of types of corruption suggests that a variety of weird forms of behavior observed in struggling economies may in fact be perfectly rational ways to behave for the elites seeking enrichment. By the same token an apparent “lack of capacity� may be a part of rational strategy for the corrupt. Second, corruption can be detrimental for economic welfare. Yet, different types of enrichment systems exist. Some allow elites to do very well by lifting all boats. Third, it is not just that different systems of corruption exist. The argument of this essay suggests that the very experience of rapid, historically unprecedented growth is motivating elites to seek new ways of enrichment that can actually lead to a better life for all. Call it economic revolution from above. Fourth, ultimately the highest degree of wealth may be reached when people obtain their position in society on the basis of merit. Choosing elites well is part of the issue. The search for enrichment is one part of the forces driving towards elections of rulers. Fifth, once upon a time, when elites had accepted rights to the spoils of the land, enrichment was not “abuse� in a legal sense. The emergence of less personal systems of rule subject to the rule of law “forced� rapacious elites to resort to corruption, the abuse of position for personal gain. The forms of enrichment that seem most effective today again allow the elites to obtain their wealth legally, even though the acquisition, for example, of property rights may have occurred illegitimately. Proudhon’s dictum that “property is theft� might apply in such cases. 17 Bibliography Fisman, Raymond (2001) “Estimating the Value of Political Connections� in: American Economic Review, 91:4 Gambetta, Diego (2009) Codes of the Underworld, Princeton University Press Klein, Michael (1987) Natural Gas Utilization in Nigeria, World Bank, Washington D.C. mimeo Klein, Michael, Jae So and Ben Shin (1996) “Transaction Costs in Private Infrastructure Projects: Are they too high?� in: Public Policy for the Private Sector, World Bank, Washington D.C. Lambsdorff, Johann (2005) “Consequences and Causes of Corruption – What do We Know from a Cross-section of Countries?� Diskussionsbeitrag Nr. V-34-05 Volkswirtschaftliche Reihe ISSN 1435-3520, Universitaet Passau Olson, Mancur (2000) Power and Prosperity: Outgrowing Communist and Capitalist Dictatorships, Oxford University Press Shleifer, Andrei and Robert W. Vishny (1993) “Corruption� in: The Quarterly Journal of Economics, August Stiglitz, Joseph E. (1996) “Some Lessons from the East Asian Miracle� in: The World Bank Research Observer, August Svensson, Jakob (2005) “Eight Questions about Corruption� in: Journal of Economic Perspectives Vol. 19 No. 3 Summer Wedeman, Andrew (1997) “Looters, Rent-scrappers, and Dividend Collectors: Corruption and Growth in Zaire, South Korea and the Philippines� in: The Journal of Developing Areas Vol. 31 You, Jong-Sung (2005) Embedded Autonomy or Crony Capitalism: Explaining Corruption in South Korea Relative to Taiwan and the Philippines, Focusing on the Role of Land Reform and Industrial Policy, Paper presented at the Annual Meeting of the American Political Science Association, Washington D.C. September 1-4 18