NUMBER 143 * ED Precis 119 Operations Evaluation Department April 19 Liberalizing Trade and Industry in Madagascar In the late 1980s, the government nant role in the economy and tightly investment program and the con- of Madagascar attempted a series of controlling private enterprise. Af- dition of the financial sector. As a structural reforms with helpfrom the ter per capita GDP declined and result, real income per head grew World Bank. Two IDA adjustment the external debt burden grew, the in 1988-90 for the first time since credits had very different outcomes. government began reforms aimed the early 1970s, more industrial Trade liberalization and budgetary at financial stabilization supported capacity was used, and nontradi- reforms succeeded because of strong by the International Monetary tional exports surged. government commitment and intensive Fund, followed by Bank-assisted Bank efforts, while reform of public structural adjustment. PSAC recorded some gains- enterprises and of the civil service notably reform of the budget pro- suffered from the government's lack The Industry and Trade Policy cess and the first private participa- of determination and reduced commit- Adjustment Credit (ITPAC, ap- tion in banking. But failure to re- ment by the Bank. proved in 1987) sought primarily solve several basic issues led to to liberalize international and do- delays in the second tranche and An OED audit of the projects' con- mestic trade, and to take the first cancellation of the third. Project cluded that, had the Bank paid closer steps in reforming public enter- design was flawed. Reforms were attention to early warning signals that prises, the banking sector, and rushed, under pressure from the the government was not fully commit- public investment. The Public Sec- Bank, even though the borrower ted to some of the reforms, the results tor Adjustment Credit (PSAC, ap- lacked commitment and wanted might have been different. Privatization proved in 1988) aimed at improving more time to consolidate the re- can be used as a tool to prevent reforms the allocation of public resources forms. Supervision and monitoring from backsliding. The Bank and the by strengthening the formulation of targets were poor. borrower must agree to a set of general and monitoring of public spending principles for privatization in advance, and investment, reforming the Lack of clarity on public enterprise to avoid getting bogged down in nego- civil service, rationalizing public reform. Conditions and targets tiating details later. Privatization enterprises (through liquidation, were poorly defined, so it was dif- must be carried out transparently; divestiture, or restructuring), and ficult to ensure that the govern- well-defined performance indicators restructuring the state-owned can clarify conditions for all parties; banking system. and social safety net programs linked Performance audit report, "Mada- to reforms must be well publicized and Results gascar: Industry and Trade Policy carried out early in the reform process. Adjustment Credit," by Alice Early gains. Most of the ITPAC Galenson, Report No. 14774, June Madagascar is one of the poorest reforms were successful. The gov- 30, 1995. Available to Bank ex- countries in the world. The govern- ernment reformed the exchange ecutive directors and stafffrom ment that came to power in 1972, rate regime, reduced and rational- the Internal Documents Unit and the one that followed it in 1975, ized import tariffs, lifted quantita- andfrom regional information pursued a strategy of import sub- tive restrictions, removed most service centers. Precis written by stitution, giving public enterprises controls on prices and profits, and Stefano Petrucci. and marketing monopolies a domi- took steps to improve the public ment was actually privatizing im- was poorly conceived, and did not proach that allows loan conditions to portant enterprises and not just offer a coherent strategy to cushion be met on an individual, case-by-case paper firms. The government ap- the effects of adjustment on the basis. Well defined performance indi- parently viewed the agreement to poor. Few substantial measures cators are essentialfor monitoring "rationalize the public enterprise were carried out until five years conditionality. Instead of agreeing sector" as an agreement to divest after the start of the project they to the principles of privatization itself of non-performing public en- were meant to accompany. and applying them, the govern- terprises. It was never convinced of ment and the Bank got bogged the need to privatize the seemingly There is little evidence, however, down debating whether various successful enterprises. At the time that political unrest was linked to actions constituted "satisfactory of the audit in 1995, more than 100 the social costs of adjustment, and progress," and supervision commercial and financial enter- none of the parties taking part in amounted to no more than count- prises remained under state owner- the ensuing elections advocated an ing the enterprises on the auction ship, burdening the economy. end to reform efforts. In fact, the block. Conditions for trade re- adjustment process appears to have forms, on the other hand, were Unfinished business. Despite the had a positive impact on most of based on explicit steps to be taken. banking sector reforms, by 1990-91 Madagascar's people. Nevertheless, the portfolios of two state-owned more timely action on a number of * Privatization is a powerful tool banks were again in serious trouble. fronts could have helped protect for preventing the reversal of public Civil service reform never got off certain groups, and could have led enterprise reform. Cleaning up the ground. Economic liberaliza- to a more successful implementa- Madagascar's bank portfolios tion led to a demand for political tion of the reform program, giving could not prevent excesses from liberalization. During a period of the government greater credibility recurring, but had all the banks political upheaval (1991-93), some in its efforts to reassure the public. been privatized, they would have policies adopted earlier were re- been far less likely to permit the versed and adjustment was dis- Lessons credit expansion that set off a li- rupted. Project funding was ulti- quidity crisis in 1991. mately canceled in 1993, after a * Trade liberalization and budgetary new government failed to resolve reform succeeded because of close gov- * Divestiture of public enterprises some fundamental issues. In par- ernment participation and intensive should be transparent so that all con- ticular, it suspended public enter- Bank efforts, whereas reform of public cerned parties are convinced they are prise reforms, adopted a new civil enterprises and of the civil service suf- being treatedfairly and given an equal service statute with several coun- feredfrom reduced commitmentfrom chance to compete. Creation of a fund terproductive measures, and failed both sides. A strong active partner- to facilitate broader Malagasy par- to control public spending. ship with the government early on ticipation in privatization would would have helped reinforce com- have alleviated fears of excessive Social aspects of adjustment mitment to reform. Closer attention foreign investment or of unbalanc- by the Bank to early warning sig- ing ethnic relationships. Measures to alleviate the social nals of low government determina- impact of adjustment and provide tion may allow the Bank to take * Sustaining reforms requires contin- technical assistance were included actions either to increase the likeli- ued popular support. It is important in a companion Economic Manage- hood of success or to proceed to develop a cohesive safety net, ment and Social Action Support more cautiously. implemented in a timely manner, Project, approved in 1988. The and to publicize it well, especially project was extremely demanding * A general agreement on principles when there is public hostility on both local and Bank resources, of reform is more effective than an ap- toward adjustment. OED Pr&cis is produced by the Operations Evaluation Department of the World Bank to help disseminate recent evaluation findings to development professionals within and outside the World Bank. The views here are those of the Operations Evaluation staff and should not be attributed to the World Bank or its affiliated organizations. This and other OED publications can be found on the Internet, at http:// www.worldbank.org/html/oed. Please address comments and inquiries to the managing editor, OED, tel: 1-202/458-4497, fax: 1-202/522- 3200, e-mail: eline@worldbank.org April 1997