The World Bank A u g u s t PREMnotes 1 9 9 9 n u m b e r 2 7 Economic Policy Lessons from large adjustment loans A Quality Assurance Group review of World Bank loans made in response to the recent global financial crisis shows the importance of borrower ownership, country knowledge, and selectivity In 1998 the Bank approved a number of The five loans considered by the QAG extremely large adjustment loans that panel were prepared or finalized under formed its core response to the global finan- extraordinary time pressure--pressure gen- cial crisis that began in Asia, hit Russia, and erated by the need for official financial sup- The key lessons threatened Latin America. Five of these port as private capital fled emerging markets loans--involving total commitments of after July 1997. Despite these unusual global are simple and $6,830 million--were recently subjected to conditions, the lessons from the loans are Quality at Entry assessments by a Quality broadly applicable to the Bank's adjustment incontestable-- Assurance Group (QAG) panel (table 1). operations. This note presents the lessons from the and important assessments that are likely to be most useful Build on borrower ownership to country directors and task teams prepar- An adjustment operation's policy objectives enough to bear ing new adjustment operations. In their nar- are much more likely to be achieved when row formulation the key lessons are simple there is substantial borrower ownership. repetition and incontestable--and important enough Ownership means not only a commitment tobearrepetition.Thefiveadjustmentloans to the policy conditions among key poli- show that applying these basic lessons is not cymakers but also broad support for the new always straightforward, however, and some- policy direction among politicians. Even timesinvolvesmakingtradeoffsamongBank when the Bank knows that borrower own- objectives. Considering the application of ership is weak or uncertain, it may decide thesebasiclessonsinthecontextofeachloan that the risk is worth taking. In that case the yields a richer set of corollary lessons. Bank must not only acknowledge this in Table 1 The adjustment loans reviewed by the Quality Assurance Group panel Amount Name of Board Number of (millions of Country operation date tranches U.S. dollars) Argentina Special Structural Adjustment November 1998 Three front-loaded 2,525 Argentina Repurchase Facility Support November 1998 One 505 Korea, Rep. of Structural Adjustment Loan II October 1998 Two equal 2,000 Malaysia Economic Recovery and Social Sector June 1998 One 300 Russia Structural Adjustment Loan III August 1998 Three back-loaded 1,500 from the development economics vice presidency and poverty reduction and economic management network its documents but also pay special attention reignited growth, generating widespread during appraisal to government plans for support for government policies and mak- dealing with resistance. When a plausible ing it easy for the government to gain sup- plan is missing, the Bank should structure port for new reforms. By contrast, Russia's its support in a way that narrows differences reforms have not resulted in growth, so gov- of view within the country, builds con- ernment reformers have faced widespread stituencies to overcome resistance, or both. skepticism. The Bank encouraged efforts by Among the five operations reviewed, the Russian reformers to engage Parliament degreeofborrowerownershipforthereforms in a dialogue on key issues, but this process on which the loan was based varied enor- was not well advanced when the crisis com- Building broad mously. In Argentina the government team pelled a decision to move forward with the had essentially designed the policy reforms Structural Adjustment Loan III. Partly ownership is andwascommittedtotheirimplementation. because of these circumstances, the Russian Moreover,therewasbroadconsensusamong loan had three back-loaded tranches, and time-consuming leading political parties on the overall pol- the Bank pushed for a "permanent institu- icy direction. In the Republic of Korea, tionalization of a participatory reform but essential althoughthereformsrepresentedaprofound process." As it has turned out, the second change of policy direction, there was a con- and third tranches have not been released viction among the political elite that the because political turbulence has inhibited reformswereneededtorecoverfromthecri- progress on reforms. sis.Inaddition,thenewlyelectedgovernment of President Kim Dae Jung appeared to have Generating broad political ownership robustpublicsupportforthereformprogram. Second, ownership of reforms by the core In Malaysia, by contrast, support for the economic policy team is not enough. A reforms occasionally seemed uneven. And broader political base--including sufficient in Russia the economic policy team at the support in the legislature--is required. center was deeply committed to the reforms When such broader ownership appears but faced strong opposition within Parlia- absent, the Bank needs to ask the economic ment and among the general public. These policy team to articulate a credible politi- diverse experiences generated interesting cal strategy for generating sufficient own- corollary lessons. ership to implement and sustain the reforms. Establishing support for new policies In Russia the inner circle of policymak- First, building borrower ownership takes ers had impressive intellectual capacity and time. When time is short, a loan should be deep convictions. But their political support, used to encourage processes that build own- especially in Parliament, was weak. The QAG ership. Policy ownership builds up as the panel suggested that the Bank should have result of favorable outcomes of related poli- "required from the authorities a clearer and cies and as the result of internal political dia- articulated strategy of how they were plan- logue. When financial crisis compels the ning to deal with . . . implementation issues." Bank to provide an adjustment loan quickly, Including political strategy in the coun- it is unlikely that sufficient time will be avail- try dialogue is difficult. When the core Bank able for the Bank to stimulate internal polit- team resides in the country, strong personal ical dialogue and to help build ownership relationships may facilitate such discussions. for new policy directions. Second-best But when the Bank is starting a new rela- options include deferring policy conditions tionship with a government whose trust must for which ownership is weak to later tranches be won, as in Malaysia, this process is espe- and putting in place participatory or other cially difficult. In that case, when the depth political processes that help build ownership. of borrower ownership is uncertain and the For example, Argentina's economic pro- government does not articulate a medium- gram had stopped hyperinflation and term strategy for building political support, PREMnote 27 a single-tranche operation--based on Bank team's ability to share good practices actions already taken--makes sense. from other countries with its Korean coun- terparts helped establish its credibility and Involving an incoming administration utility. For example, learning about Mex- Finally, engaging the key players in an incom- ico's Pacto Solidaridad helped the Koreans ing government before its inauguration can designtheirowntripartiteagreementamong help build ownership for reforms. This government, businesses, and workers. approach worked in Korea, for example. By contrast, even though Malaysia's gov- The economic crisis hit in November 1997, ernment was preoccupied with dealing with just before a national election. The Bank the financial crisis, a key mission did not initiated dialogue with the incoming eco- include experts on bank reform--experts Combined, the nomic team of newly elected President Kim who could have added much value to what Dae Jung, a move that helped assess and nur- had been accomplished on an earlier 1998 Bank's country ture deeper ownership of structural reforms. mission by the Bank and the International Monetary Fund. knowledge and Combine country knowledge and global expertise Maintaining a minimum knowledge base global expertise Deep knowledge of a country is essential to Finally, the Bank should invest in a mini- quality operations. The Bank's value added mum knowledge base for countries where can generate ismuchgreater,however,whenitcontributes it is not active but where it might be called knowledgeaboutwhathasworkedelsewhere. in quickly. Even though the Bank had been quality operations The quality of the operations in Argentina inactive in Korea and Malaysia for some time, waspartlyduetotheconsiderableknowledge it was able to call on staff who knew a great among Bank staff of the country and its insti- deal about the politics and economics of tutions--knowledge that created a profes- both countries. But the Bank might not sional consensus within the Bank and with always be so lucky. For inactive countries that the borrower on the right policies. In Korea might require assistance quickly, the Bank and Malaysia the Bank had been inactive should at least develop an informed opin- for a number of years before the Asian cri- ion--drawing on third-party work--of the sis. Still, in preparing the adjustment oper- country's key problems. ations, the Bank relied on staff who knew a lot about these countries. Tilt toward selectivity InmostcasestheBankneedstotiltawayfrom Forging local partnerships broad coverage and toward selectivity in the Even when Bank staff members have good policy reforms supported by a single opera- knowledge of a country, their understand- tion. The five operations assessed involved ing can be enhanced by working with domes- conditionalities on a fairly comprehensive tic research institutions and think tanks. This set of reforms. This approach reflected an approach was especially helpful in Korea, understandable tendency by Bank staff to where many of the operation's analytical push as many elements of reform as possi- underpinnings were provided by local insti- ble through the window of opportunity cre- tutions engaged in policy research. A simi- ated by a crisis. But given the need for lar approach was used in Argentina. borrower ownership, analytical clarity, and institutional capacity, this tendency should Drawing on experience elsewhere be resisted. Excessively broad conditional- In addition, the Bank's credibility with gov- ity may reduce the probability that real ernment counterparts and value added in progress will be made on key reforms. policy design are significantly enhanced when it assembles staff with expertise about Setting priorities and sequencing reforms what has worked in other countries in areas During the design phase the Bank and the that are of interest to the government. The government should pay more attention to August 1999 setting priorities and sequencing reforms. Conclusion Doing so may help narrow conditionality. Stated in their simplest form, these lessons Setting priorities is difficult--especially in may seem obvious and trite. Yet the QAG's transition economies such as Russia, where assessments show that they are sometimes creating an environment that attracts domes- not given sufficient weight in the design tic and foreign investment requires that of adjustment operations. Furthermore, reforms proceed along an ambitiously broad their application in diverse circumstances front. Still, realism about the government's requires careful consideration and good implementation capacity--both technical judgment. The more specific corollary and political--requires that priorities be set. lessons may be helpful in that regard. Excessively broad Sustaining focus on key reforms The Quality at Entry assessments on which this In a multi-tranche operation, selectivity can- note is based were performed by a panel of expe- conditionality not be achieved by excluding from second- rienced Bank staff and an outside economist: trancheconditionssustainedperformanceof Joe Wood, who recently retired as Vice President may reduce the conditions from the first tranche. Adequate for the South Asia Region; John Page, then Chief performanceoffirst-trancheconditionsshould Economist for the Middle East and North Africa probability of be a condition for the second tranche, and Region and now Director, Economic Policy, PREM hence this performance requires consistent Network; and Vittorio Corbo, a Chilean econo- real progress on monitoring over the life of the operation. mist who has worked in and with the Bank for With Korea's loan, for example, despite decades. The panel reviewed the operations' doc- key reforms the central role of corporate restructuring umentation and interviewed many of those to the operation's success and the inclusion involved, including government officials. of related conditions in the first tranche, This note was written by Daniel Morrow dur- monitorable conditions focusing on ade- ing his tenure as a Fellow in Economic Policy, quate progress in that area were absent from PREM Network, with assistance from Victo- the second tranche. This approach may give ria Elliot (Principal Economist, Economic Pol- the appearance of selectivity, but it runs the icy, PREM Network). If you are interested in risk of losing focus on reforms that take time similar topics, consider joining the Country to implement. (In Korea corporate restruc- Strategy and Performance Thematic Group. turing work nevertheless remained promi- Contact Victoria Elliot, x80535, or click on The- nent in the country dialogue.) matic Groups on PREMnet. This note series is intended to summarize good practice and key policy findings on PREM-related topics. PREMnotes are distributed widely to Bank staff and are also available on the PREM website (http://prem). If you are interested in writing a PREMnote, email your idea to Sarah Nedolast. For additional copies of this PREM- note please contact the PREM Advisory Service at x87736. Prepared for World Bank staff