Corruption and Economic Development Stephen J. H.

Dearden
Manchester Metropolitan University, Department of Economics Email: s.dearden@mmu.ac.uk DSA European Development olicy Study !roup Discussion aper "o. #$, %cto&er '(((

This paper discusses the relevence of economic theory to the analysis of corruption and reviews the econometric evidence for its causes and consequences. Finally it outlines the anti-corruption measures which have been taken by international organisations. It is in many ways surprising how little attention has been paid in the academic literature to the problem of corruption in developing countries. While anecdotes may flow freely at conferences it rarely appears as an issue in formal debate. Yet with the end of the cold war and a more critical and focused view of foreign aid, the institutional, political, social, and cultural obstacles to sustained growth are now being addressed. Institution building, the rule of law and good governance are phrases that are appearing with increasing frequency in the international debate amongst policy makers. While neo-classical market economics has become the universal model and structural adjustment the means, the essential requirement of an institutional framework is now coming to be recognised. he transition of the economies of the former !oviet "nion tomarket economies has illustrated the problems most graphically. While the #$ech %epublic and &oland were able to build upon their prewar e'perience and institutional memory, %ussia has struggled to create the conditions necessary for a successful market economy. he failure to establish an independent judiciary and a framework of company law to enforce contracts, define property rights and impose accountability( to create an administration capable of honest and efficient administration and to undertake equitable and efficient ta' collection, have undermined the economic recovery of the nation. he creation of effective legal structures and civil institutions requires a political will and the international community regards the commitment to democracy as the necessary and in many ways sufficient condition. hus corruption can be seen as a manifestation of institutional failure which in turn reflects political failure, but such political failure in turn can only be e'plained by an analysis of the underlying forces at work in a society. ) tradition of authoritarian regimes is seen as lacking the checks and balances necessary to avoid corruption. here is no tradition of the separation of powers, of an independent and critical press, or of an independent judiciary. *ven where there is the panoply of representative democratic government, where political control is based upon clientalism, political patronage or corporatism, corruption is likely to take root. ) social structure characterised by a high degree of racial, tribal, religious or geographical divisions may be particularly vulnerable to political clientalism, where the ruling political elite buys its support from the groups that will sustain its power. he position of such groups will be reinforced by the process of corruption that gives

them continued privileged access to societies resources and thereby sustains their importance to the elite that wishes to maintain control. +owever the conflict over the distribution of national resources in such a clientalist political system may in the long run become acute and lead to political and social disintegration, particularly in states whose legitimacy is not beyond question. ,urther the marginalisation of the formal legal system is the inevitable by-product of the emergence of such a political mechanism which offers privileged access, thereby undermining its legitimacy beyond the activities of the privileged elite groups. #artier--resson. has suggested five economic conditions which appear to encourage the flourishing of corruption within a society. he e'istence of an e'ploitable natural resource /eg. oil0 providing the opportunity for !tate authorities, both administrative and political, to obtain payments. !econdly, the general scarcity of public assets relative to demand, accompanied by policies of fi'ed official prices, which creates opportunities for informal rationing through bribery. hirdly, low wages in the public sector are also likely to be associated with e'tensive low-level corrupt payments. ,ourthly, high levels of !tate intervention1planning /ie. protectionism, !tate-owned enterprises, price controls, e'change controls, import licenses, etc.0 which has characterised many developing countries. ,inally, economies in transition are likely to have particular problems as they undertake privatisation and establish the relevant legal framework of company and contract law, etc. he relevance of these various factors is evaluated in the empirical review in the following section, however first I wish to consider in more detail the likely economic effects of e'tensive corruption in an economy. )nalysis 2iscussion about the economic costs of corruption has focused upon a number of different aspects. 3ne approach has emphasised the cost of the diversion of resources from productive activity to avoidance of 4predation4. In the absence of an effective rule-of-law resources must be diverted to protect property rights. ) number of authors have developed theoretical models to e'amine the economic consequences of such situations where protection against predation is incomplete /5jungqvist 6 !argent .778, )cemoglu .778, 9urphy et al .77.0. hey suggested the possible e'istence of multiple equilibria, some involving low production where diversion /ie. theft0 offers a high pay-off. he government has two roles in the prevention of predation - control is more effective if provided collectively and the government itself, in that it determines law creation and enforcement, can be a major source of diversion. %ent-seeking is the major e'pression of government diversion /:rueger .7;<0 and the principal component of what we understand by corruption. )n alternative approach has focused upon the theoretical impact of bureaucratic inefficiency upon investment and growth. In endogenous growth models such inefficiency may directly reduce investment or may lead to the misallocation of resources between capital projects. !imilarly in the neoclassical growth model, such misallocation of investment may reduce the steady-state level of income and the private marginal product of capital. ) broader economic perspective on the welfare costs arising from corruption can be obtained by focusing upon the distortions that it brings to the efficient functioning of

markets. echnical, allocative and distributional efficiency will all be compromised by the various forms of price distortion in the factor and goods markets arising from corruption( we have situations of &areto inferiority. +owever it is the central role of price distortions in the welfare assessment that has lead some authors to defend the role of corruption in developing countries. In economies already characterised by substantial price distortions - wage regulation, unprofitable state-owned enterprises, state agricultural marketing boards, food subsidies, import controls, e'change controls and unsustainable e'change rates, amongst others - the rule of second best may apply. he price distortions of corruption may merely be offsetting the other state created distortions of the economy. hus the underpayment of public sector workers may merely be offset by the opportunities offered to such workers to supplement their pay by bribery, providing incentives for the efficient discharge of their responsibilities and raising their pay to market clearing levels. !imilarly the problems of rationing import licenses or allocating public contracts can be achieved through the shadow pricing of the bribes paid to obtain them. It must be remembered that welfare analysis is not concerned with issues of the distribution of income but only with the effectiveness of the pricing signals that reflect the scarcity of resources. -ut despite these theoretical arguments the prices that arise from these informal transactions are unlikely to be those of an efficient competitive market. #arried out secretly, without the crucial information flows of efficient markets, and with restricted participants, the correct market clearing prices cannot be established and economic welfare ma'imised. #orruption has usually been defined as 4the abuse of public office for private gain4. his definition has been widened recently, as attention has turned to corruption within the private sector, to cover 4giving or receiving undue advantage in the course of business activities leading to acts in breach of a person=s duties 4/ ransparency .7770. +owever it can be argued that the role of the public sector is more significant, since it is here that the opportunities for corruption are most e'tensive and the public administration creates the opportunities for corruption in the private sector by creating the necessary institutional and market conditions. In both cases corruption can be viewed as an e'ample of the &rinciple-)gent problem, where the core difficulty lies in the mechanism to monitor the actions of those to whom authority is delegated but where the information is possessed asymmetrically by the agent. In creating a framework to categorise the forms of corruption we can begin by e'amining its impact upon the private firm or individual. #orruption can both increase or decrease the production costs faced by the firm or the product costs faced by the individual. -ribes paid to e'pedite government regulatory activities, to obtain import licenses or to win government contracts, will all raise costs. -y contrast bribes paid to circumvent government regulations - eg. health and safety, environmental requirements, ta'ation - may reduce the net cost to the company, even if it imposes social costs. In the e'treme case of payments made to gain access to a valuable state resource - eg. mining concessions, privatised enterprise - the e'istence of the enterprise itself may depend upon such payments. ) government policy of protectionism, with Import !ubstituting Industrialisation, usually creates conditions of monopoly1oligopoly and therefore substantial economic rent for e'ploitation.

,rom the perspective of the public sector costs will be both financial and economic. #orruption resulting in ta' evasion will clearly reduce the governments ta' revenues, but corruption may also add to the governments costs through the allocation of contracts to higher priced contractors. !uch increases in prices will be essential to meet the costs of the bribes that are paid. -ut the economic costs may be even more substantial as projects may be undertaken merely to generate bribes rather than as a reflection of economic development priorities. !uch major capital investment projects, which were never economically viable, may remain a drain on the scarce resources of developing countries for many years. ,rom the broadest perspective corruption distorts prices throughout the economy, as the costs of bribes are passed onto the final consumer and it creates delays in economic transactions and additional uncertainty, which may be a crucial element in undermining investment and encouraging capital flight. *mpirical !tudies ) large number of studies have attempted to isolate statistically both the causes and consequences of corruption( these are comprehensively reviewed in 5ambsdorff /.7770. "nfortunately they suffer from recurring and perhaps insurmountable problems, which while not confined to this area of study, are nonetheless particularly apparent when attempting to analyse the significance of corruption. he first difficulty lies in attempting to establish a meaningful and robust measure of the degree of corruption present in a society. 2ifferent studies have employed a wide variety of pro'ies to reflect the e'tent of corruption, while broader corruption indices have been created by a number of institutions. 3ne of the most comprehensive currently available is that created by ransparency International which has been published since .778. +owever even this does not claim to be an objective measure of the e'tent of corruption but of the 4perception of corruption4. It draws upon several surveys of e'pert and general public opinion from around the world >. ?ualitative assessments, with arbitrary scales, also make the interpretation of the estimated coefficient values on the corruption variable problematic. )lternatively one could look to the risk assessments of the international credit risk agencies, but the e'tent of corruption need not be the dominant factor in such assessments. he second difficulty lies in the causal relationships that these studies claim to establish. Is corruption the primary cause of any statistical relationship or also a consequence /ie. the endogeneity problem0@ ,or e'ample, does an association between low public sector salaries and the level of corruption reflect the undermining of the ta' base to fund public e'penditure /corruption causing low pay0 or does it reflect the need for corrupt pay supplements to provide adequate salaries for public sector employees /low pay causing corruption0. )lternately corruption and low pay may be co-determined by factors that have not been identified in the statistical analysis ie. the political and economic interests that maintain a low ta'1public e'penditure regime. he final problem arises from the robustness of the other variables that are included in the studies, often leading to econometric weaknesses.

With these general difficulties in mind we can turn to a brief review of the various factors that have been suggested as the determinates of the level of corruption and then to the efforts to identify what the economic consequences may be. )ttempts have been made to establish, amongst others, a statistical relationship between corruption and the si$e of the public sector, relative public sector pay levels, or the openness of the economy. -ut beginning with Aohnson et al /.77B0, they considered the hypothesise that in economies where ta'ation and regulation of business allows discretion to officials, corruption will flourish, and firms will respond by leaving the official economy and moving into the informal sector. hey e'amined the relationship between the share of the unofficial economy in C2& and the quality of regulation, the si$e of the ta' burden and the level of corruption. In three separate regressions they utilised data for between D> and <7 countries in the 3*#2, 5atin )merica and the former !oviet block for the mid .77Es. "sing four opinion surveys they found a positive correlation between high levels of regulation and the si$e of the informal economy. In their two surveys of ta'ation it was the perceived ta' burden, rather than the level of the highest marginal ta' rates, which encouraged the growth of the unofficial economy. ,inally, data from si' surveys confirmed the e'pected relationship between the e'tent of perceived corruption and the si$e of the informal sector. +owever a number of difficulties arise with this study. ,irstly, measuring the si$e of the informal sector raises serious problems. !econdly, the study also found a clear negative association between the si$e of the informal sector and the level of C2&. -ut the reduction in the si$e of the unofficial economy as national income increases can be e'plained in a number of ways. )s the economy grows the resources available to the public sector may allow more effective public administration. )lternatively the relationship may be interpreted as indicating the crucial role of administrative reform in enhancing the rate of growth. ,inally, both corruption and the e'tent of the informal economy may reflect underlying institutional weaknesses, which in turn may be determined by social and cultural influences. %ijckeghem and Weber /.77;0 e'amined the relationship between corruption and public sector pay. In a sample of >B developing countries they found the higher relative civil servant pay the lower the level of corruption. +owever other similar studies have failed to confirm the relationship. 5eite and Weidemann /.7770 established a robust statistical relationship between a countries e'ports of oil and minerals, relative to CF&, and the level of corruption for a sample of ;> countries in .7;E. he hypothesis here being that abundant natural resources, earning substantial hard currencies, provides the opportunity for rentseeking. )des and 2i ella /.77;0 e'amined the impact upon corruption of interventionist industrial policies, believing that they provide the opportunities for rent-seeking amongst public officials and politicians. "tilising World #ompetitiveness %eport indices for the openess of public procurement procedures and equality of fiscal treatment between enterprises for D> countries between .7B7 and .77>, they found a negative relationship with their corruption inde'. his negative correlation was confirmed by the use of alternative quantitative measures of industrial interventionism

- ie. the si$e of public subsidies to public and private enterprises. hey also attempted to address the causality question of whether interventionist industrial policy created opportunities for corruption, as hypothesised, or whether corrupt politicians deliberately pursued interventionism in order to create opportunities for rent e'traction. "sing instrumental variables to address this simultaneity bias confirmed their interpretation of the central role of interventionism in creating the conditions for corruption. urning specifically to the institutional dimension, studies have attempted to e'amine the correlation between corruption and the degree of press freedom, e'tent of democratic institutions, degree of political decentralisation /ie. federalism0 or the quality of the judiciary. 3ne of the few robust studies, the .77; World 2evelopment %eport, found that its measure of the predictability of the judiciary was a significant variable in its sample of 87 countries. While 9auro /.7780 found a high correlation between -usiness International=s survey assessments of nine institutional and political aspects of 8B !tates. his included three dimensions of institutional efficiency corruption, bureaucracy, judicial integrity - and si' measures of political stability. ,inally we can consider studies of the underlying cultural determinants. 5a &orta et al /.77;0 in a study of DD countries using the World Galue !urvey, suggested that a higher incidence of 4trust4 in a society reduced the level of corruption( while the dominance of a hierarchical religions, such as %oman #atholicism, *astern 3rthodo' or Islam, was positively associated with higher incidences of corruption. +owever in a subsequent study of ..< countries the relationships weakened when C2& per capita was introduced into the equations. +usted /.7770 introduced the concept of 4power distance4 - the e'tent to which the less powerful members of society e'pect and accept unequal influence - and found it positively associated with corruption, as were the e'tent of dominance of materialism and the perceived threat of uncertainty. ,inally 2ollar et al /.7770 established that the percentage of women in the labour force and in &arliament had a significantly negative effect upon the level of corruption in their study of HH countries. +owever any measure of female participation in a developing countries labour force must raise serious concerns and the interpretation of the findings remains unclear. Indeed all of these studies of cultural factors raise serious problems of objective measurement. In terms of the evaluation of the economic consequences of corruption Cupta et al /.77B0 e'amine the impact of corruption upon the distribution of income in D; countries. hey discovered a positive association between corruption and income inequality /measured by the Cini coefficient0 and a negative impact of the growth of corruption upon the income growth of the bottom >EI of the income distribution. hey also found that corruption may have additional indirect effects via its adverse impact upon educational opportunity and land distribution. )s in some other studies the authors attempted to address the question of the direction of causation by the use of instrumental variables. +owever the reliability of income data, especially in developing countries and in !tates where corruption is high, raises serious problems, as can the use of the Cini coefficient as an accurate reflection of changes in the distribution of income. ) similar approach employing instrumental variables has been taken by +all and Aones /.7770 to address the causation relationship between productivity and

corruption indices. "tilising data for .>; countries, compiled by &olitical %isk !ervices for the period .7BH-78, they established a strong relationship between the level of output per worker and their measure of 4social infrastructure4. his inde' included assessments of law and order, bureaucratic quality, risk of e'propriation and repudiation of contracts and corruption. he primary role of social infrastructure was tested by employing the geographical and linguistic characteristics of the countries as instrumental variables. 3ther studies have focused upon the influence of corruption on the level of C2& per capita /:aufman et al /.77700 and upon the rate of growth of C2&. :eefer 6 :nack /.7780, &oirson /.77B0 and 5eite 6 Weidmann /.7770 all confirmed the significance of corruption levels for growth rates. 9auro /.7780 identified reduced investment as the mechanism whereby growth was reduced by the influence of corruption. "tilising data from -usiness International for the period .7BE-B8 for 8B countries 9auro employed nine country assessment categories, grouped into indices of bureaucratic efficiency, political stability and overall institutional efficiency, so as to reduce measurement error. he bureaucratic inde' consisted of assessments of the efficiency of the bureaucracy, judiciary and of the incidence of corruption. -oth the bureaucratic inde' and corruption alone were found to be negatively correlated with the rate of investment, even while controlling for other significant variables. o test for the endogeneity of these variables 9auro employed a measure of national linguistic homogeneity /*thnolinguistic ,ractionalisation0, confirming the causal relationship. )s investment is an important determinant of growth 9auro proceeded to e'amine the relationship between corruption and growth. -oth the corruption inde' and the bureaucratic efficiency inde' were significantly related to per capita C2& growth over the period .7HE-B8, although the results were less robust than in the case of the relationship to investment. -ut more importantly once investment was introduced into the growth equations the bureaucracy and corruption indices became insignificant. )des and 2i ella /.77;0 also e'amined the impact of corruption upon investment and found that it reduced the positive impact of interventionist policies upon investment by between .HI and ;>I. ) number of other studies, using different inde'es of corruption, have confirmed these results /eg. :eefer 6 :nack .7780. his suggests that the primary economic impact of corruption may be as a deterrent to investment rather than in reducing the productivity of capital once an investment is made. +owever problems of heteroscedasticity /the probability distribution of the disturbance term is not constant0 have often compromised many of the estimates. -ut other authors regard the predictability of the corruption as a more important influence upon investment than the overall level. he World 2evelopment %eport /.77;0 identified the importance of this factor. #ampos et al /.7770, utilising the same World -ank survey data, carried out a cross-section study of H7 countries for the years .77E-7<. #ontrolling for C2& per capita and secondary school enrolments - as a measure of human capital investment - they found that the perceived unpredictability of corruption was a more robust and significant determinant of the rate of investment than the e'tent of corruption. hus he argued countries with e'tensive, but predictable corruption, would deter less investment than those with similarly high levels of corruption where payments could not guarantee the desired outcome. +owever his measures of the predictability of corruption and its e'tent were highly correlated with each other and the overall e'planatory power of his equations very weak. #ampos

also recognises that more stable institutionalised corrupt regimes might well present even greater obstacles to reform in the longer term than those that are unpredictable. +e suggests just such a situation may be facing some of the countries of *ast )sia. he conclusion that a category of stable but corrupt regimes can e'ist must also be reconciled with evidence such as that of 9auro /.7780 who found a high correlation between his measures of political instability and bureaucratic inefficiency, including corruption. #orruption may also have an impact upon the flow of foreign resources to a country whether as foreign investment or as aid. "nfortunately the evidence is rather mi'ed. Whereas Wei /.77;0 detected a significant negative effect in a study of the capital flows between .< source and <8 recipient countries between .77E and .77., Wheeler and 9ody /.77>0 and )lesina and Weder /.7770 found that corruption was an insignificant determinant of foreign direct invesment. )lesina and Weder also found that corruption appeared to have no influence upon the flows of aid from 3*#2 donors. +owever focusing upon bilateral aid flows they suggest that the !candinavian donors did appear to be sensitive to the corruption levels of recipient countries, while the flow of "! aid appeared to be positively associated with their corruption measure. )lthough attempts at the econometric evaluation of the origins and consequences of corruption present considerable methodological problems, some tentative conclusions can be drawn. here is confirmation that corruption is encouraged in economies with substantial e'port earnings from a natural resource or where the !tate is interventionist, as suggested by #artier--resson. #orruption is also strongly associated with inefficient public administration and an ineffective judicial system. Whether these institutions have been compromised to provide the means for corruption or are themselves a manifestation, like corruption itself, of underlying political and social conditions, remains to be considered. he economic consequences of e'tensive corruption include low labour productivity, reduced investment and consequently lower growth. he distribution of income is also likely to be more unequal. he International Institutions 3*#2 !o far the most significant international agreement addressing the problem of corruption has been that of the 3*#2 4#onvention on #ombating -ribery of ,oreign &ublic 3fficials in International -usiness ransactions4. his was adopted in Fovember .77; and came into force in ,ebruary .777. !o far .; countries, including seven within the *", have ratified the #onvention which requires the signatories to make it a criminal offence to offer bribes in order to obtain a business advantage. It also encourages !tates to end the ability of firms to treat bribe payments as business costs that are ta' deductible. he success of these measures is being monitored by an 3*#2 Working Croup on -ribery in International -usiness ransactions. he 3*#2, the *" and C; are also collaborating in the fight against money laundering. he Inter-governmental ,inancial )ction ask ,orce /,) ,0 is composed of representatives from all of the major international financial centres and produces recommendations for anti-money laundering legislation and procedures. ,) ,

reviews compliance with these recommendations and is assisting in the establishment of similar agreements in other regions of the world. 3ne such has already been established in the #aribbean. !imilarly in .77B the "F had published a report on 9oney 5aundering, followed in .777 by another initiative, the 43ffshore Initiative4 of the "F Clobal &rogramme against 9oney 5aundering. his identified a 4white list4 of well regulated 3ffshore ,inancial #entres that were cooperating with the international organisations. hose not included were to be encouraged to adopt anti-money laundering legislation and to increase their cooperation. -y contrast in .77B the 3*#2 had published a %eport on "nfair a' #ompetition that had identified a 4black list4 of ta' havens that facilitate unfair international ta' competition. Interestingly !wit$erland and 5u'embourg declined to sign up to the 3*#2 report and are therefore not liable to listing. &ressure on the appro'imately fifty 3ffshore ta' havens has been mounting. Within the *"s jurisdiction the ta' policies of the Isle of 9an, Aersey and Cuernsey have been found to be harmful and could be required to be amended by >EED. 9any of the 4black listed4 3,#s face the threat of economic sanctions by the 3*#2. his will raise serious problems for a large number of island economies, such as the 9arshall Islands, !eychelles, -ermuda and the -ritish 3verseas erritories of the #aribbean, for whom offshore financial activity is an important source of revenues. In response #)%I#39 is taking the actions of the 3*#2 in regard to ta' havens and the ,) , on money laundering to the W 3. he #ouncil of *urope he #ouncil of *urope has also adopted two conventions covering #riminal and #ivil 5aw. hese cover #entral and *astern *urope as well as the *". he #riminal 5aw #onvention e'tends the definition of corruption to include the activities of the private sector, while the #ivil 5aw #onvention provides for compensation for individuals suffering a loss as a result of corrupt practises by an official. -oth #onventions provide protection for 4whistleblowers4. !o far .> 9ember !tates of the *" have ratified the #riminal 5aw #onvention and only 8 the #ivil 5aw #onvention. he *# will be able to accede to the #onventions in its own right once the #onventions come into force. he *uropean #ommunity Within the *" two anti-corruption #onventions have been drafted. he Auly .778 #onvention is concerned with the protection of the *#s financial interests and the 9ay .77; #onvention with corruption amongst *# officials or officials of the 9ember !tates. +owever neither has been ratified by a large number of its 9ember !tates. &ublic &rocurement within the *" has been subject to review /#39/7B0.<D0, with the possibility of the introduction of anti-corruption pledges by all parties to contracts, as has the role of the !tatutory )uditor. In terms of the *"=s e'ternal relations the #ommon !ervice %ele', which is responsible for e'ecuting most of the *"= development aid programme, has adopted 4ethical clauses4 in its 9anual of &rocedural %ules for its procurement procedures. he possibility of 4blacklisting4 companies or individuals who infringe these rules

has, for the first time, been introduced. "nfortunately this 9anual only covers the contracting phase and complementary rules covering the implementation phase have yet to be adopted. he #ommission has however taken general powers to suspend or cancel aid if partner governments fail to take satisfactory action to combat corruption when it is discovered. he abolition of the ta' deductibility of bribes, which had been a frequent practice within *" !tates, will be required with the adoption of thee 3*#2 #onvention. )ssociated with this practice has been the issue of the treatment of the costs of bribery as part of *'port #redit Insurance. -ribery has been estimated to add .EI to >EI to the costs of any contract and these e'port guarantees are usually ultimately government funded. )s the *" has competence in regard to e'ternal trade responsibility clearly falls upon the #ommission to instigate action in this area. he heightened priority being given by the *" to the problem of corruption in developing countries and the general approach that is being taken, is best illustrated by e'amining the new &artnership )greement /#otonou0 between the *" and the )#& states that succeeded the 5omJ #onventions in >EEE. In this )greement 4the respect for human rights, democratic principles and the rule of law4 are regarded as 4essential elements of the partnership4 while a commitment to 4good governance4 is a further 4fundamental element4. In )rticle B, where the process of 4political dialogue4 is outlined, specific reference is made to the 4regular assessment of the developments concerning the respect for human rights, democratic principles, the rule of law and good governance.4 "nder )rticle 7 the )greement recognises that 4democracy based upon the rule of law and transparent and accountable governance are an integral part of sustainable development4. !pecifically 4the structure of government and the prerogative of the different powers shall be founded on the rule of law, which shall entail in particular effective and accessible means of legal redress, an independent legal system guaranteeing equality before the law and an e'ecutive fully subject to the law.4 Cood governance is defined as 4the transparent and accountable management of human, natural, economic and financial resources for the purpose of equitable and sustainable development. It entails clear decision making procedures at the level of public authorities, transparent and accountable institutions, the primacy of law in the management and distribution of resources and capacity building for elaborating and implementing measures aiming in particular at preventing and combating corruption.4/)rticle 7.D0 he introduction of good governance into the )greement had faced considerable opposition from the )#& !tates. )s )damou !alao of Figer commented 4the main contention concerns its inclusion as an essential element in the future agreement ... none of the )#& countries want to see good governance become an essential element, which if violated would trigger the non-e'ecution clause and potentially lead to sanctions4 / he #ourier Fo..;; 3ct-Fov. .7770. he )#&s were willing to strengthen the political dialogue, affirming their 4adherence to the principles of democracy, the rule of law, human rights and good governance. Fonetheless these principles should not be used by the *" to suspend development assistance unilaterally or render it conditional4D. he result of the negotiations was a compromise solution distinguishing between 4essential4 /ie. human rights, democracy, rule of law0 and 4fundamental4 /ie. corruption0 elements of the )greement, with two separate procedures created to address cases where there has been violation of these principles.

In the case of corruption there is a consultation procedure and a strong reserve power for the *# - 4in cases of special urgency appropriate measures may be taken without prior consultation4. he )greement is qualified, - only 4serious cases of corruption, including acts of bribery leading to such corruption, as defined in )rticle 7;, constitute a violation of the )greement4. -ut 4if the consultations do not lead to a solution acceptable to both &arties or if consultation is refused, the &arties shall take the appropriate measures. In all cases, it is above all incumbent on the &arty where the serious cases of corruption have occurred to take the measures necessary to remedy the situation immediately. he measures taken by either &arty must be proportional to the seriousness of the situation. In the selection of these measures. priority must be given to those which least disrupt the application of this agreement. It is understood that suspension would be a measure of last resort4 /)rticle 7;.D0. he )greement also qualifies the regular assessment of of the )#&s progress in the 4promotion of human rights, processes of democratisation, consolidation of the rule of law and good governance4 in that it is to 4take account of each country=s economic, social, cultural and historical conte't.4/)rticle 7.<0. #onclusion )lthough the international aid organisations can make a significant contribution to the erosion of the endemic corruption that is compromising the economic development of many countries, ultimately it is the level of commitment to reform within the countries themselves that is crucial. ,or progress to be made political commitment to 4good governance4 within the developing countries is crucial. his will ensure that the elements necessary for what ransparency International has termed a 4Fational Integrity !ystem4 can be put in place. hese elements include an independent judiciary, political pluralism with an accompanying effective opposition, appropriate anti-corruption legislation, effective monitoring and auditing systems /including paralegal independent inspectorates and administrative commissioners0, structures to ensure integrity in the awarding of public contracts, the encouragement of the monitoring role of civil society through the establishment of an independent press, and the 4declaration of interests4 by both politicians and civil servants. he international community can encourage these developments by a programme of technical assistance and targeted aid. ,or institution building to be successful it is essential that there be a political commitment in the developing countries. he international donors must be willing to reward such political commitment to reform and penalise those !tates whose governments fail to meet these international standards. -ut it must be recognised that significant difficulties may arise in the attempt to objectively and consistently assess the progress of the recipient states. !ome institutions, such as the World -ank, believe that there is a high correlation between simple quantitative indicates and progress in general reform. ,or the *" the #otonou )greement=s requirement for the regular assessment of the )#&s progress in achieving reform will quickly bring this issue to the fore. It will be essential to assuage the )#&s fear that such assessments may be arbitrary or politically motivated. he *" at least has the advantage of more diluted political interests and a greater degree of overt dialogue than the situation in many individual 9ember !tates. he #otonou )greement also provides considerable fle'ibility in the operation of its 4corruption clause4 while retaining sanctions in the case of serious infringments.

he difficulties that such fundamental transformations present for many developing !tates must also be recognised and an appropriate long-run perspective adopted. ,or e'ample, the pressure upon public sector budgets arising from structural adjustment programmes can e'acerbate the problem of low pay in the civil service, encouraging corruption. he emphasise upon 4partnership4 in much of the debate is a clear recognition of the nature of the challenge that reform offers. -ut in those cases where the e'isting political elite is unresponsive then donors have already identified the crucial role that will have to be played by other actors in the developing countries 4civil society4. his strategy of by-passing unsatisfactory governments has yet to be tested. he political difficulties of fostering other 4actors4 in a society may be considerable, perhaps compromising their relationship with their own domestic governments. It also fails to address the problem of those civil societies where such 4actors4 do not e'ist. -ut the developed countries also have a clear responsibility to create an international environment that discourages corrupt practices. he recent moves by the 3*#2 and the *" to make the offering of bribes by international companies illegal is an indication of significant progress and a clear demonstration of a commitment that the international community can now require the developing world to reciprocate. *ndnotes .. ) general non-technical review of the issues involved in the debate about corruption can be found in a series of articles in he )#&-*" #ourier, Fo .;;, 3ctober .777. It is also available at httpK11www.europa.eu.int1comm1development1publicat1courier1inde'L.;;Len.h tm >. 2ocumentation for the ransparency International #orruption &erception Inde' can be found at httpK11www.uni-goettingen.de1Muwvw D. D. &ara.B< K )#& Croup Fegotiating 9andate, -russels, DE !eptemeber .77B. )vailable at httpK11www.oneworld.org1acpsec1gb1lome1future1negmanLe.htm -ibliography )cemoglu 2. .778. %eward !tructures and the )llocation of alent K *uropean *conomic %eview Gol. D7, pp..;-DD )des ). 6 2i ella %. .77;. Fational #hampions and #orruption, !ome "npleasant Interventionist )rithmetic K he *conomic Aournal Gol. .E; pp. .E>D-.E<> )lesina ). 6 Weber -. .777. 2o #orrupt Covernments %eceive 5ess ,oreign )id@ K Fational -ureau of *conomic %esearch Working &aper ;.EB #ampos A, 5ien 2. 6 &radhan !. .777. he Impact of #orruption on Investment( &redictability 9atters K World 2evelopment Gol. >; pp. .E87-H;. 2ollar 2. et al .777. )re Women %eally the 4,airer !e'4@ #orruption and Women in Covernment K &olicy %esearch %eport on Cender and 2evelopment, Working &aper Fo<, he World -ank, Washington.

Cupta !., 2avoodi +. 6 )lonso- erme %. .77B. 2oes #orruption )ffect Income Inequality and &overty@ K International 9onetary ,und Working &aper 7B1;H +all %. 6 Aones #. .777. Why 2o !ome #ountries &roduce !o 9uch 9ore 3utput &er Worker than 3thersK ?uarterly Aournal of *conomics Gol ..< pp. BD-..H. +usted .777. Wealth, #ulture and #orruption K Aournal of International -usiness !tudies Gol. DE />0 pp DD7-HE Aohnson !. et al. .77B. %egulatory 2iscretion and the "nofficial *conomy K he )merican *conomic %eview Gol. BB, pp. DB;-7> :aufmann 2. et al .777. Covernance 9atters K World -ank &olicy %esearch Working &aper >.7H. :eefer &. 6 :nack !. .77H. Institutions and *conomic &erformance, #ross-!ection ests "sing )lternative Institutional 9easures K *conomics and &olitics, Gol .> pp. >E;->; :rueger ). .7;<. he &olitical *conomy of the %ent-!eeking !ociety K )merican *conomic %eview Gol. H< pp. >7.-DED 5ambsdorff A. C. .777. #orruption in *mpirical %esearch - ) %eview K ransparency International Working &aper. 5a &orta et al .77;. rust in 5arge 3rganisations K he )merican *conomic %eview, Gol. .D; />0, pp DDD-B. 5a &orta et al .777. he ?uality of Covernment K he Aournal of 5aw, *conomics and 3rganisation, Gol. .8 /.0 pp >>>-;7 5eite #. 6 Weidmann A. .777. 2oes 9other Fature #orrupt @ Fatural %esources, #orruption and *conomic Crowth K International 9onetary ,und Working &aper 771B8 5jungqvist 5. 6 !argent . .778. he !wedish "nemployment *'perience K *uropean *conomic %eview Gol. D7 pp. .E<D-.E;E 9auro &. .778. #orruption and Crowth K ?uarterly Aournal of *conomics Gol. ..E pp. HB.-;.> 9auro &. .77;. he *ffects of #orruption on Crowth, Investment and Covernment *'penditure( ) #ross #ountry )nalysisK in #orruption and the Clobal *conomy, Institute for International *conomics, Washington. 9urphy :., !hleifer ). 6 Gishny W. .77.. he )llocation of alent, Implications for Crowth K ?uarterly Aournal of *conomics Gol. .EH pp. 8ED-8DE &oirson +. .77B. *conomic !ecurity, &rivate Investment and Crowth in 2eveloping #ountries K International 9onetary ,und Working &aper 7B1<

%ijckeghem #. 6 Weber -. .77;. #orruption and the %ate of emptation. 2o 5ow Wages in the #ivil !ervice #ause #orruption@ K International 9onetary ,und Working &aper 7;1;D. ransparency International .777. ,ighting #orruptionK What %emains to be 2one at *" 5evel K Working &aper. "nited Fations 2evelopment &rogramme .77;. #orruption and Integrity Improvement Initiatives in 2eveloping #ountries. Wei !. .77;. +ow a'ing is #orruption on International Investors K Fational -ureau of *conomic %esearch Working &aper HEDE Wheeler 2. 6 9ody ). .77>. International Investment 5ocation 2ecisions, he #ase of "! ,irms K Aournal of International *conomics Gol. DD pp. 8;-;H