Alison Decker is a second-year student in the international development program and a senior editor of SAIS Perspectives.

International financial institutions (IFIs) bear unique risk in their operations. They lend to often-fragile governments, which means that they cannot use the same risk mitigation strategies employed in private loans, like collateral and shorter loan maturities. Instead, IFIs work with borrowing countries to foster policy environments as conducive to loan implementation as possible. One of the primary ways is through loan conditions, or “conditionality.”

When IFIs first started embedding conditions into their loans, they were viewed as a “substitute for collateral” or a method of assuring the borrowing country would be able to solve its structural imbalances and, ultimately, repay the loan.[i] As projects increased in complexity—tackling institutions instead of infrastructure, for example—the number of conditions attached grew in tandem, and their scope broadened.[ii] Originally, conditions focused on quantitative criteria that signaled a country’s efforts to reform its balance of payments issues. Over the second half of the twentieth century, conditions began to incorporate criteria for financial sector reforms, military spending, anti-corruption efforts, and environmental safeguards, to name a few.[iii] These conditions, built into loan packages, served multiple purposes. For one, they operated as a filter to “screen countries which do not intend to increase adjustment effort.”[iv] If a country wouldn’t agree to IFI conditions, they likely wouldn’t commit to meaningful macroeconomic reform. In addition, conditions allowed IFIs to base lending decisions and disbursements on verifiable outcomes, rather than promises of reform, and they transferred some of the risk back to the borrowers. However, the explosion of conditions is often critiqued, particularly when examining loans through the 2008 Accra Agenda and 2011 Paris Declaration on Aid Effectiveness country ownership lens.[v]

During the 1990s, conditionality’s popularity sank and non-compliance rose: between 1993 and 1997, just over 25 percent of 141 IMF arrangements were actually in compliance.[vi] There are some potential structural justifications for non-compliance. IFI clients pose an adverse selection issue: often they are countries in crisis, which makes adhering to the multiple, stringent conditions difficult. Loan conditions are signed in what some may consider “conditions of duress.” [vii] Furthermore, stipulated conditions are often inherently complex and difficult to quantify.  Even in cases where governments enact recommended reforms, desired outcomes may not follow. Paul Collier noted that IFIs overestimate their own power, particularly in Africa, and “ignored domestic politics” to loan detriment.[viii] Conditionality can even have outright negative effects, undermining domestic process by “supplanting public policymaking” and stepping on the toes of countries’ sovereignty.[ix] There are also few clear links between loan success and the conditions imposed on that loan. Indeed, economists have noted that between 1982-1995 when condition use peaked, aid was associated with increases in corruption and reduced quality of governance.[x] In response to these criticisms, IFIs have reformed their conditionality guidelines, following IMF recommendations that conditions be kept to a minimum and aligned with country-led development strategies. In theory, this adherence means borrowing countries have more input over their loan conditions and a more meaningful seat at the table. With these overarching themes in mind, examining conditionality in two countries with radically different forms of governance and macroeconomic constraints, Nigeria and Botswana, illustrates what truly influences loan conditions and their effects: the country context.

Nigeria, with a reputation of corruption and fiscal leakage, offers a typical scenario in which loan conditions would be employed to create the necessary policy environment for effective implementation. Though Nigeria has had strong growth rates (6-8% annually),[xi] its reliance on oil production as a driver of its economy—coupled with a tendency for oil revenues to go directly into the pockets of the state—leaves the country vulnerable to external shocks. In a 2018 loan designed to foster fiscal transparency and prudent fiscal federalism, the World Bank thus employed conditions which were varied and encompassed both broad issues and specific performance criteria, including 40% average female participation in the budget process, citizen engagement in the budget process, and specific numbers of states using biometric registration in payroll to reduce fraud.[xii] While the conditions were numerous, there were few macroeconomic suggestions and even the broadly scoped conditions were highly relevant to the project. This was not always the case, as demonstrated via a comparison of a 1998 loan to support small and medium enterprises which was not successful—there was no significant change in outcomes for small and medium enterprises. Final reports on that loan blamed macroeconomic deterioration and lack of Government commitment for the failure. Conditions included use of market-based interest rates, which the Government did not maintain. There was additional non-compliance: the Bank purports that the Government did not promote line of credit options sufficiently, nor take ownership of the pilot financing components.[xiii] The 1998 and 2018 loans both used conditions, but differently. In its 1998 loan, conditions were numerous and focused on setting countrywide economic policy. In contrast, the 2018 conditions focused on specific performance indicators and for the most part were limited to the relevant sector in which the loan operated. Examining the use of conditions in Nigeria shows that indeed, IFIs have changed their conditionality strategy over time. It is also noteworthy that in this difficult policy environment, the conditions were often not met—and projects fizzled.

In contrast, Botswana has a reputation for a sound policy environment. It has solid macroeconomic policy, efficiently manages its natural resources, and maintains strong governance indicators. The country’s economic growth has been markedly pro-poor and the largest decline in poverty occurred in rural areas, reducing income inequality.[xiv] Botswana seems to be in little risk of default and this is reflected in how IFIs deal with the country. In current loans, World Bank Group conditions on Botswana are minimal and read as nudging the government to create its own strategy. This is demonstrated in the 2017 Water Security and Efficiency Project, which supports investments in water supply infrastructure. Conditions were grounded in the expectation that the government does not need stringent supervision.[xv] The government demonstrated prior commitment to the necessary reforms, initiating a Water Sector Reform Program before the project even began. In addition, the government agreed to during-loan reforms, including devising a Water Resource Management Water Efficiency Strategy and Water Policy to provide water access, and ensuring financial sustainability by monitoring water tariffs.[xvi] In these conditions, the scope is limited to the water sector and they act as suggestions for development of government policy. Though these conditions are not particularly onerous, this type of conditionality strategy has not always been the case, as indicated in a 1984 financial services sector loan. For a project aimed at financing non-mining small-scale enterprises and strengthening development finance institutional capacity, the Bank employed multiple conditions to specifically alter governmental macroeconomic policy. For this loan, the government had also demonstrated commitment to reform: it created the Botswana Enterprises Development Unit to promote indigenous small-scale enterprises and the Ministry of Commerce and Industry carried out multiple private investment feasibility studies. As conditions of the loan, the government agreed to several reforms: to maintain a consolidated debt, review lending rates of the small-scale finance enterprises, limit their operations growth rates, raise interest rates on microcredit operations in the agricultural sector by 2 percent, and maintain a specific debt-to-equity ratio during the loan’s operations.[xvii] While the loan had some successful outcomes, it did not fully meet the development objectives. These conditions are far more intrusive into government operations and expand past the scope of the financial sector. The shift in condition strategy from the 1984 loan to the 2017 loan shows the way IFI thinking on the issue has evolved, even for a country considered to be a stable policy environment.

As might be expected from the literature on loan conditionality, the total number of conditions declined across IFI loans in recent years and the conditions themselves are tighter in scope and aligned with country development strategies. Current conditions are less cumbersome, nudge governments to devise their own strategy, and more efficiently support specific projects rather than push for macro reform.

Through examination of loans in two different environments, a few lessons are clear. One is that conditionality has evolved since its first application, as displayed by World Bank loans in both Nigeria and Botswana. In earlier loans, conditions regulated macroeconomic stability in both a stable and difficult policy environment, potentially infringing on governmental sovereignty and creating cumbersome requirements. Now, even in more challenging policy environments, conditions focus more on specific performance measures in the relevant sector. Ultimately, however, the country context is still critical. In Nigeria, conditions were numerous, structural, and quantitative, aimed at creating institutions that would underpin and support the loan. In contrast, the conditions written into loans for Botswana were simple, read more as suggestions for Botswana to deepen already existing policies, and focused on the sector relevant to the loan. All IFIs made note of Botswana’s strong policy environment as one of the critical factors in determining conditions and disbursing loans quickly, efficiently, and in risky sectors—even in extreme circumstances like the severe fiscal losses after the financial crisis. The conditions gave Botswana agency: to develop their own strategies to address development problems that the government had already identified. Conditions acted more as a “push” towards specific reforms that would assist in a path forward than Nigeria’s conditions, which functioned as ways to protect the integrity of the loan from fiscal leakage. When designing condition strategy, these lessons are important to keep in mind: conditionality use as a risk mitigation strategy is more necessary when operating in a difficult policy environment, as linking disbursement funds to reforms critical for success can ensure funding is not used ineffectively. In stable environments, conditions can function less as a mitigation strategy and more as a tool to provoke country-led development.

      [i] Kapur, Devesh. "Conditionality and Its Alternatives." In Buira, Ariel. The IMF and the World Bank at Sixty. London: Anthem Press, 2005.

      [ii] Kapur, Devesh. "Conditionality and Its Alternatives." In Buira, Ariel. The IMF and the World Bank at Sixty. London: Anthem Press, 2005.

[iii] Goldstein, M. “IMF Structural Programs.” In M. Feldstein (Ed.) Economic and financial crises in emerging market economies, 363-437. Chicago: University of Chicago, 2000.

[iv] Marchesi, Silvia, and Jonathan P. Thomas. "IMF Conditionality as a Screening Device." The Economic Journal 109, no. 454 (1999): 111-25. doi:10.1111/1468-0297.00420.

[v] OECD.  The Paris Declaration on Aid Effectiveness and the Accra Agenda for Action. London: Organization for Economic Cooperation and Development, 2005.

[vi] Kapur, Devesh. "Conditionality and Its Alternatives." In Buira, Ariel. The IMF and the World Bank at Sixty. London: Anthem Press, 2005.

[vii] Kapur, Devesh. "Conditionality and Its Alternatives." In Buira, Ariel. The IMF and the World Bank at Sixty. London: Anthem Press, 2005.

[viii] Santiso, Carlos. “Good Governance and Aid Effectiveness: The World Bank and Conditionality.” The Georgetown Public Policy Review, 2001.

[ix] Santiso, Carlos. “Good Governance and Aid Effectiveness: The World Bank and Conditionality.” The Georgetown Public Policy Review, 2001.

[x] Knack, Stephen. "Does Foreign Aid Promote Democracy?" SSRN Electronic Journal, 2001. doi:10.2139/ssrn.260047

[xi] International Monetary Fund. “Nigeria: 2018 Article IV Consultation.” Washington DC: International Monetary Fund, 2018.

[xii] World Bank Group. “Program Appraisal Document: States Fiscal Transparency, Accountability, and Sustainability Program for Results.” Washington, DC: World Bank Group, 2018.

[xiii] World Bank Group. “Implementation Completion Report: Federal Republic of Nigeria. Private Small and Medium Enterprise Development Project.” Washington, DC: World Bank Group, 1995.

[xiv]World Bank Group. “Country Partnership Framework for Republic of Botswana.” Washington, DC: World Bank Group, 2015.

[xv] World Bank Group. “Project Appraisal Document: Botswana Emergency Water Security and Efficiency Project.” Washington, DC: World Bank Group, 2017.

[xvi] World Bank Group. “Project Appraisal Document: Botswana Emergency Water Security and Efficiency Project.” Washington, DC: World Bank Group, 2017.

[xvii] World Bank Group. “Staff Appraisal Report: Second Development Finance Companies Project.” Washington, DC: World Bank Group, 1984.

Photo credit and license: Creative Commons Attribution 2.0 Generic license.