The Political Economy of Corruption: A Philippine Illustration, Study notes of Political Economy

Download The Political Economy of Corruption: A Philippine Illustration and more Study notes Political Economy in PDF only on Docsity! The Political Economy of Corruption: A Philippine Illustrationa James Roumasset b Department of Economics, University of Hawai’i at M
noa & University of Hawai’i Economic Research Organization Working Paper No. 08-05 October 27, 2008 Abstract This essay explores the nature, causes, and consequences of corruption as it pertains to entire regimes. Grand corruption is modeled as a type of unproductive rent-seeking at the highest levels of government. The economic costs of corruption are assumed to increase in the decentralization (and relaxation) of its governance, increase convexly in the percentage extracted, and decreasing in the opportunities for productive rent-seeking. Combining these assumptions with the benefits of corruption yields the results that optimal corruption revenues are increasing in greed of the regime and in economic opportunities but that the economic costs of corruption may be highest in the least avaricious regime. The theory is illustrated with a stylized account of corruption in three Philippine administrations, from 1973-1998. Policy implications are discussed, including the role of the economist in making corruption less attractive. Keywords: Corruption, Philippines, kleptocracy JEL codes: H11, K42, O5 a This paper was originally drafted for the Western Economics Association International meetings in Seattle, July 1997, and was issued as University of Hawaii at Manoa Working Paper 97-10 under the title, The Political Economy of Corruption.” Thanks to Lee Endress and Roger Blair for helpful discussions, Majah-Leah Ravago for research and editorial assistance, and to an anonymous referee for constructive comments. All remaining errors are my responsibility. b Professor at the Department of Economics, University of Hawai’i, M
noa, and Environmental Director for University of Hawai’i Economic Research Organization. Saunders 542, 2424 Maile Way, Honolulu, HI 96822 USA. E-mail: jimr@hawaii.edu 2 THE POLITICAL ECONOMY OF CORRUPTION: A PHILIPPINE ILLUSTRATION James Roumasset 1. Introduction Corruption, according to Rose-Ackerman (1996, p. 365), “occurs when officials use their positions of public trust for private gain.” It is “an extralegal institution used by individuals or groups to gain influence over the bureaucracy” (Leff, 1964, p. 8). That is, corruption involves transactions, typically between private parties and public officials, designed to manipulate the machinery of government. It may be of the permission-seeking type (quotas, licenses, permits, passports, and visas), the enforcement avoiding type (tax evasion, illegal pollution) or the competition-harassing type. Corruption is closely associated with bribery that has been recognized since the 15 century B.C. as “a gift that perverts judgment” (Noonan, 1984, p. 12). Most of the literature on corruption and bribery is implicitly applied to lower level public officers (e.g. Mocan, 2008).1 This essay explores the nature, causes, and consequences of corruption as it pertains to entire regimes – what Rose-Ackerman (1996, 1997) calls “grand corruption” or kleptocracy, including high level manipulation of policies and projects. We shall see that grand corruption is similar to rent-seeking at the highest levels of government and may be usefully regarded as part of the same “third-best” theory of government (Dixit, 1996), also known as political economy. The purpose of the present paper is to extend the theory of corruption and imbed it in a more general political economy of public policy. To dramatize the theoretical points, I provide a stylized account of corruption in three Philippine administrations, from 1973-1998.2 Inasmuch as hard evidence about corruption at the highest levels of government is generally unavailable, 1 “Because an increase in perceived corruption in a country is thought to be associated with a slower rate of economic growth (Mauro 1995), a sizable literature has emerged recently to examine factors that impact the level of corruption across countries. For example, Ades and DiTella (1999) found that corruption is higher in countries where domestic firms are sheltered from foreign competition. Graeff and Mehlkop (2003) documented the relationship between a country’s economic freedom and its level of corruption. Brunetti and Weder (2003) found that a higher freedom of the press is associated with less corruption. Van Rijckeghem and Weder (2001) showed that the higher the ratio of government wages to manufacturing wages, the lower is corruption in a country” (Mocan, 2007). 2 Marcos’s first two constitutional terms, 1965-73 are excluded from this account. 5 Thus, unlike a kleptocracy modeled along the lines of Brennan and Buchanan’s (1977) Leviathan, Marcos was not free to directly pocket the proceeds from the 18% production tax. Further entrepreneurship was needed to develop schemes that transferred wealth of the quasi- public entity, UNICOM, into private hands. One clever scheme involved using UCPB funds to purchase and plant an African hybrid variety on a plantation given to the head of UNICOM as “compensation” for a relatively small amount of land that he lost under the country’s land reform program. Moreover coconut farmers were mandated to replant their coconut farms to these same hybrids on the grounds that their higher yields would improve both farmers’ yields and the national economy. In summary, under the guise of what appeared to be legitimate public programs, a single Export Ban UCPB (18%) Land Reform Mandatory Hybrid Purchase UNICOM Domestic Monopsony COPEC Dwarf-Hybrid Monopoly Lower Prices to Farmers Higher Prices to Foreign Consumers Higher Cost to Farmers Figure 1: “Legitimate” Corruption of Philippine Coconut Policies 6 private individual was given a monopsony on coconuts, a monopoly on coconut oil, and a monopoly on the one variety of coconut trees that farmers were required to buy. Moreover the oil and hybrid monopolist was not required to buy the industries but was but was largely gifted them thru the allocation of public funds. 2.2 Secrecy and Stupidity Lucrative as it was, however, the scheme was “too clever by half.” The coconut oil monopoly, popularly known in the Philippines as “COPEC,” tried to imitate OPEC and stockpiled coconut oil in order to raise the world price. The plot backfired when the oil went rancid and, when it was eventually sold, spoiled the reputation of Philippine oil. As the result of this 18% tax and the nuisance value of the replanting requirement, the Philippines irreversibly lost its position as the world’s leading exporter of coconut oil. Moreover, the government was not in a position to exploit its oil monopoly on the domestic market. Cooking oil, like rice and gasoline is regarded as a basic commodity among urban consumers and the political repercussions of a rise in price can be severe. Finally, the African dwarf hybrid program was a bust. Farmers were unwilling to replant, even under the threat of law and a highly subsidized price, and the program became unenforceable. These errors suggest a corollary to the proposition that bribery is more distortionary than taxation. The necessity of secrecy leads government officials to discourage more transparent activities in order to stimulate the more “bribable” sectors (S-V, 1993), to say nothing of differences in revenue disposition. While the coconut scheme described above was not entirely secret, neither was it the product of thorough public debate, which likely would have partially overcome the failure to foresee adverse consequences. That is, secrecy not only begets inefficiency in intersectoral resource allocation, it begets collective stupidity. Another famous scheme involved the construction of a nuclear reactor project on the Bataan peninsula. The Westinghouse power plant, “originally estimated at $500 million for two 7 reactors, ended up costing $2.8 billion for a single reactor.”7 Again the transfer to private hands was achieved through relatively sophisticated means, relative to Mrs. Marcos’s more blatant alleged tactic of simply demanding a share of government contracts. A number of specialized or dummy corporations were set up to obtain contracts for the construction of the power plant, to insure the plant (for $688 million), and to handle other aspects of the project. Both Marcos and his crony Herminio Desini had substantial interests in these companies. Desini was also apparently paid $80 million for consulting services largely for his role in helping Westinghouse win the contract.8 The high level of corruption in the Marcos administration accords with Persson and Tabellini’s (2005) thesis that parliamentary governments tend to have a higher level of corruption than presidential ones. Marcos was able to establish himself as both President and Prime-Minister and to effectively limit checks and balances on his power (cf. footnote 6). 3. Post-Marcos Evolution of Corruption After the legitimacy of the Marcos administration had waned and the government of self- proclaimed “housewife” and widow of Benigno Aquino came into power, corruption returned to the decentralized, Spanish-mestizo-dominated form that had prevailed before Marcos declared martial law on Sept 21, 1972. The loss of central control of corruption has two deleterious effects. First, the number of expropriators and the level of effort expended in government plundering increases, much as in the Tragedy of the Commons.9 First the degree of “theft” from the highest level of government increases. This in turn feeds the further spread of corruption thru a kind of Gresham’s Law (corrupt officials outcompete honest officials for jobs and bribe-giving producers drive out their competitors: “observance of law does not survive in a competitive market” (S-V, 1993 p. 604). 7 Adams, 2008. 8 Ibid. 9 Hardin, 1968. 10 Either authority is centralized in an honest “principal” government that designs incentive compatible mechanisms to motivate corruptible bureaucrat-agents (Rose-Ackerman, 1978), authority is centralized in a kleptocratic (Rose-Ackerman, 1996), Leviathan (Brennan and Buchanan,1980), monarchy (S-V), or authority is not centralized (and corruption is “S-V inefficient”). As Shleifer and Vishny themselves confess, this approach does not illuminate “the far deeper question” (S-V, 1993 p. 608) concerning the ability of high government officials to centralize corruption. The case of the Philippines suggests that legitimacy is an important determinant of central authority. Power does not, as Chairman Mao suggested, emanate entirely “from the barrel of a gun,” though legitimacy and the ability to monopolize violence are clearly joint inputs in the production of central authority. To the extent that such authority is “gun-produced,” leaving it as exogenous in economic models is understandable. On the other hand, high-level and low-level corruption are co-evolutionary. High-level or “grand” corruption may be distinguished from rent-seeking in shaping public policies on the basis of whether it is inside or outside of the law. Thus, Imelda Marcos’ extortion may be said to be corruption and the coconut scam said to be mere rent-seeking. This distinction may turn out to be superficial and elusive in practice, however. What surreptitious threats were used, for example, to deter opposition to new laws governing coconut policy? It may be recalled, for example, that Senator Palaez was shot for revealing consequences of these policies for coconut farmers. Similarly in the U.S., there is a fine line between policies than have been influenced by legal and illegal campaign contributions. Politicians’need for campaign gifts in effect offers “protection for sale,” and special interest groups are willing buyers (Grossman and Helpman, 1994). In any case, the economic causes of corruption and rent-seeking are very much the same, and the theory of political economy can be applied to grand corruption. Indeed, it has been suggested above that Imelda’s larceny was very much dependent on President Marcos’ ability to 11 maintain popular support and the appearance of legitimacy about his government.11 Since petty corruption and grand corruption are also usefully seen as part of the same theory, this suggests a unified theory of rent-seeking which encompasses corruption as well. Figure 3 illustrates the application of neoclassical political economy12 to the problem of how cooperation becomes more or less centralized. The downward sloping excess burden costs represent the S-V (1993) theory discussed above. For the same corruption revenue, the excess burden costs fall with the degree of centralization. The upward sloping curve, CCM and CCA represent the costs of centralization under Marcos and Aquino respectively. Marcos’ entrepreneurial skill and credible commitment to punishing corruption-seeking independents is represented as a lower cost of organizing a 11 Peter Kann, publisher and CEO of the Wall Street Journal, once referred to Marcos’ martial law government as “smiling martial law.” Marcos himself was extremely persuasive in arguing that martial law was itself consistent with the emergency power granted to the President under the Philippine constitution. 12Neoclassical political economy is also known as the “economics of the third-best” ( Dixit, 1996; Roumasset, 1989). Figure 3: Equilibrium Centralization of Corruption $ Degree of Centralization ACA CCA CCM CA CM Excess burden costs of corruption ACM 12 sustainable central system of corruption. Agency costs, ACA and ACM,13 represent the corresponding vertical summation of the two curves for the two administrations. The optimal degrees of centralization are where the agency cost curves are minimized, at CA and C M. Figure 4, illustrates the plausible consequences of centralization and corruption opportunities for the three regimes discussed in this paper. The marginal costs of corruption, MCA and MCM, are derived by combining figures 1 and 2. As corruption revenue is increased, the excess burden costs shifts increasingly upward leading to the rising MC curves in figure 4. The position of MCA above MCM reflects the greater excess burden per dollar of corruption rents under the more decentralized Aquino administration relative to that of Marcos. The marginal cost of corruption under Ramos is shown as even lower than that of Marcos due to the opportunities before the Asian financial crisis, to take a smaller share from the greatly expanded $ C o rru p tio n ($ ) C A C M ,C R F ig u r e 4 : C o m p a r a t iv e C o r r u p t io n M B R M B A M B M M C A M C M M C R private foreign investment. In addition, the marginal benefits are drawn to reflect the characterization that the Marcos’s and their cronies were the most avaricious in their quest for corruption revenues. The figure illustrates the possibility that a less avaricious regime (Aquino) can be more costly to the economy relative to more centralized regimes (Marcos and Ramos) 13 Jensen and Meckling (1976) pioneered the use of agency theory in their theory of corporate governance, especially regarding corporate finance. 15 to pay bribes, thereby violating both efficiency and equity conditions of desirable taxation. The problem that remains to be solved is under what conditions privatizing tax collection can be effective. What is needed to prevent this from merely degenerating into the above result concerning bonuses or commissions is the possibility of developing guidelines for the tax collector’s auditing procedure and the possibility of auditing the auditor. This of course presumes that some part of the central government is incorruptible, but it would still represent an advance over existing literature. A related and also unsolved problem concerns how a kleptocratic center enforces centralization. Anecdotal accounts suggest that simple pyramid sharing-schemes are commonplace, with lower officials receiving higher percentage shares but higher officials receiving larger bribe revenues due to the nature of the pyramid. 5. Conclusions and Policy Implications Grand corruption is modeled as a type of unproductive rent-seeking at the highest levels of government. The economic costs of corruption are assumed to increase in the decentralization (and relaxation) of its governance, increase convexly in the percentage extracted, and decreasing in the opportunities for productive rent-seeking. Combining these assumptions with the benefits of corruption yields the results that optimal corruption revenues are increasing in greed of the regime and in economic opportunities but that the economic costs of corruption may be highest in the least avaricious regime. The theory is illustrated with a stylized account of corruption in three Philippine administrations, from 1973-1998. Most of the corruption literature presumes a social engineering perspective. From this vantage point, one can recommend the following sequencing of reforms analogous to the “sequencing of liberalization” literature. First, corruption should be centralized under the highest level of government. This will dramatically lower the costs of corruption and set the stage for the second step. Next, the per unit price of corruption should be reduced. This will promote mechanisms of corruption that are more conducive to growth without even necessarily reducing 16 the “take” of high officials. A third step involves liberalizing those policies whose enforcement increases corruption. The fourth step involves auditing systems that render government enforcement incentive compatible. Since there is no such thing as a social engineer, however, one may ask about more pragmatic approaches. It is widely recognized that a free press and a demand for investigative journalism is one of the most effective weapons against bribery and rent-seeking generally. The government can be more proactive by improving auditing and reporting requirements so as to render the nature and consequences of government programs, policies, and projects more transparent to the public. Mandating benefit-cost analysis and reports such as environmental impact statements can similarly increase transparency. Indeed the Australian government at one time actually created a Transparency Agency to monitor government programs and to document who gained and lost from programs and how much. And Transparency International14, based in Berlin, ranks countries according to corruption as perceived by private companies. In the same vein, the Philippine government has over the years15 recognized that there is a need to fight graft and corruption and thus established agencies that are mandated to rebuke erring presidential appointees and government officials. In President Cory Aquino’s time, two other special agencies were created in reaction to Marcos's abuses: the Commission on Human Rights and the Commission on Good Government (PCGG)16. The latter is tasked to repossess ill-gotten wealth allegedly stolen by the Marcoses and cronies. Treisman (2000) finds evidence that the process of economic development may lessen corruption thereby promoting a virtuous circle inasmuch as corruption impedes growth. Education offers another way out of the stagnating sort of corruption inasmuch as education promotes civil vigilance, in part by leading to a broader social understanding of predatory practices and policies. The more people understand that win-lose strategies tend to generate lose- 14 http://www.transparency.org/ 15 Starting from Pres. Elpidio Quirino’s Integrity Board of President in 1950 to Presidential Anti-graft commission of Pres. Gloria Macapagal-Arroyo. < http://www.pagc.gov.ph/pagc_predecessor.htm > 16 See e.g. <http://www.pcgg.gov.ph/ > and Dolan, 1991. 17 lose results in situations where agents repeatedly encounter one another, they will be more inclined to employ win-win strategies. By characterizing and sometimes quantifying not only aggregate benefits and costs of projects and policies but their distribution, economists can make an important contribution to transparency. I have emphasized the nature of the political regime as a cause of corruption. Other causes have and will be explored in the future as statistical patterns regarding the correlates of high and low corruption emerge. One example that may apply to the Philippines regards the role of religion. Treisman (2000) suggests that corruption tends to be higher in predominantly Catholic countries due to less tolerance of challenges to authority, less weight on individual responsibility (relative to human fragility protected by a forgiving church), less separation between church and state (leading to less autonomous monitoring of the state), and more weight on family over individual, leading to greater tolerance of “amoral familism” and nepotism (Treisman, 2000, p428). These forces may become part of a nation’s social capital and may help explain why the Philippines is often characterized as a Latin American country in Asia.