Research & Policy Briefs From the World Bank Malaysia Hub No. 11, Nov 2017 Are Public Credit Guarantees Worth the Hype? Facundo Abraham Sergio L. Schmukler Public credit guarantees have become a popular instrument to try to expand lending to financially constrained firms. In many instances, these schemes have proven useful to increase access to finance. However, public interventions to extend guarantees need to be designed, evaluated, and monitored properly. Otherwise, they could prove unnecessary or bring about important costs, such as increased moral hazard or implicit subsidies. Credit guarantee schemes have existed for a very long Credit guarantee schemes can emerge for different time, with the oldest schemes dating back to the nine- reasons (Honohan 2010). Guarantors can have an teenth century. However, it was not until the 1990s informational advantage over lenders. For example, that these instruments gained notoriety and expanded credit guarantee schemes established by small worldwide. Since then, brand new schemes have business associations can screen member firms better appeared in many countries and existing schemes have than financial institutions. Moreover, guarantors might been reformed, expanding their scale and outreach. By be able to diversify risks better than financial institu- 2015, they were present in virtually every country in tions. Whereas financial institutions can be geographi- the world (Pombo, Molina, and Ramírez Sobrino cally concentrated or focused on specific types of 2015). borrowers, guarantors can spread risk by guaranteeing loans in different financial institutions with different The public sector has been a key force behind the lending profiles. In some cases, these schemes can also propagation of credit guarantee schemes. Govern- develop in response to regulatory arbitrage. For ments started to increase the use of guarantees as a example, guarantees can make otherwise insufficiently way of channeling credit toward specific sectors, unsecured loans comply with regulatory requirements. geographical regions, and firms (typically, small and medium enterprises, or SMEs) that tend to be finan- Although these arguments imply that credit guaran- cially constrained. Furthermore, public schemes have tee schemes would be able to emerge and develop greatly expanded since the 2007−08 global financial privately with no state intervention, in many cases crisis as a means to boost private lending countercycli- governments participate in these schemes, often cally. directly. Public credit guarantee schemes typically take two forms. On the one hand, the state can set up and This policy brief attempts to answer two important manage its own guarantee scheme. On the other hand, questions regarding credit guarantees. First, is there a the government can partner with the private sector need for state intervention for these guarantees to and establish a public-private guarantee scheme. In emerge and persist over time? Second, to what extent this case, the state can retain either a majority or a have public credit guarantees had a positive impact in minority stake in the scheme. terms of expanding finance to financially constrained firms and enhancing performance of those firms that In response to the state’s involvement in credit have received guaranteed lending? guarantee schemes, the World Bank has issued a set of principles for the design of public guarantees that are How Do Public Credit Guarantees Work? efficient and financially sustainable (World Bank 2015). For example, management needs to be independent When a loan is guaranteed, a third party, known as the from political interference and be selected according to guarantor, promises to pay back to the creditor a part clearly defined criteria. Furthermore, private participa- or the total amount of the loan if the borrower tion is encouraged to promote governance and respon- defaults. In exchange for providing the guarantee, the sibility. Coverage ratios (representing the fraction of guarantor collects a fee from the creditor. Because the the loan value that is guaranteed) need to leave creditor faces lower risk when a loan is guaranteed, it enough risk to the lenders to motivate them to prop- can offer better lending conditions and require lower erly assess and monitor borrowers. In addition, estab- collateral to guaranteed borrowers. Banks using guar- lishing costless and speedy claim procedures can antees can reduce their loan loss provisions, which increase credibility and encourage lenders to partici- increases profits and capital levels. pate. Transparency and oversight are also key features Affiliation: Development Research Group, the World Bank. E-mail addresses: fabraham@worldbank.org, sschmukler@worldbank.org. Objective and disclaimer: Research & Policy Briefs synthesize existing research and data to shed light on a useful and interesting question for policy debate. Research & Policy Briefs carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions are entirely those of the authors. They do not necessarily represent the views of the World Bank Group, its Executive Directors, or the governments they represent. Acknowledgement: We received very useful guidance and comments from José De Luna Martínez and Norman Loayza. Global Knowledge & Research Hub in Malaysia Are Public Credit Guarantees Worth the Hype? Table 1. Characteristics of Public Credit Guarantee Schemes Outstanding SMEs Outreach Default rate Regions Age Employees guarantees served (%) (%) (% of GDP) Global median 21 99 0.11 1,383 1.6 2.5 Africa 26 26 0.01 77 0.3 17.1 Asia 27 371 0.10 17,293 2.7 1.2 Europe 22 93 0.29 1,139 0.9 2.9 Middle East and North Africa 12 40 0.12 829 2.2 3.8 Western Hemisphere 19 64 0.05 6,531 3.4 2.0 Source: Calice 2016. Note: All values are median values for 2014. Outreach is defined as the percentage of served SMEs to total SMEs in the country. The default rate is defined as the percentage of nonperforming guarantees to total outstanding guarantees. Data include 60 public credit guarantees in 54 countries. SMEs = small and medium enterprises. Western Hemisphere includes North America, and Latin American and the Caribbean. of a successful public scheme. With respect to pricing, Public credit guarantees are relatively new, having the level of the fees would depend on the scheme’s been in place for around 20 years in general. The purpose. If the goal of the scheme is to address a median scheme around the world has outstanding market failure such as asymmetric information, it might guarantees equivalent to 0.11 percent of GDP and have to temporarily offer subsidized fees to attract fewer than 100 employees. In addition, it serves less lenders while borrowers learn about them. If the than 2 percent of total SMEs in a country. However, the scheme is not addressing a market failure, higher fees size, outreach, and performance of these schemes would be justified. But in the latter case, the question widely vary across regions (figure 1). would be why private sector participants are not providing the guarantees themselves. In general, when the public sector manages the guarantee schemes, the private sector does not partici- Public Credit Guarantees Around the World pate (Calice 2016). The types of guarantee schemes led by the state differ across countries. In some countries, More than 30 percent of credit guarantee schemes public schemes are administered in a centralized around the world have some form of state ownership, manner, whereas in others, these schemes consist of according to a survey by Beck, Klapper, and Mendoza local and regional funds overseen by a central institu- (2010). Public credit guarantee schemes are particu- tion. In some cases, the public sector can be directly larly important in developing countries, where they are involved in granting guarantees, whereas in other cases the main type of guarantee scheme. In contrast, in the state provides strategic direction but has little developed countries private schemes tend to prevail. control on how the scheme is run (see box 1 for Box 1. Types of Public Credit Guarantee Schemes. sets the eligibility criteria for firms applying for a guaran- tee and provides the funding. However, the BBB does Countries across the world have adopted different not decide on guarantees applications, which is done models of public credit guarantees. In many countries, a directly by financial institutions. single state agency provides the guarantees (examples include Chile, Estonia, Indonesia, the Republic of Korea, Other countries have opted for public-private guar- Thailand, and the United States). antee schemes with different degrees of government participation. For instance, in France, credit guarantees In other cases, public schemes can operate in a more are offered through an organization owned 90 percent decentralized manner. For example, in Japan, there are by the state and 10 percent by banking groups. In Spain, 51 state-run credit guarantee corporations under the guarantees are provided by private schemes. The state umbrella of the Japan Federation of Credit Guarantee intervenes through the Compañía Española de Reafian- Corporations (JFG). zamiento (CERSA), which is a public institution that In some countries, the state is not directly involved in grants counter-guarantees to private schemes, provides granting guarantees. In the United Kingdom, the British tax reductions for their operations, and sets the cover- Business Bank (BBB), a state-owned development bank, age ratios of guarantees. 2 Research & Policy Brief No.11 examples of different designs of public credit guaran- increased their sales and survival rates (Oh et al. 2009). tee schemes around the world). Furthermore, public Similarly, the French loan guarantee program led to schemes differ in terms of their coverage ratio, pricing, higher growth of participating firms relative to nonpar- eligibility criteria, and debt recovery arrangements, ticipating ones (Lelarge, Sraer, and Thesmar 2010). In among other features. contrast, credit guarantees do not seem to have increased the performance of firms in Italy (D’Ignazio The Impact of Public Credit Guarantees and Menon 2013). Furthermore, there is evidence that in Japan the performance of firms that received guar- The performance of public credit guarantee schemes anteed loans not only did not increase but even dete- can be measured by whether they generate “financial riorated. additionality” and “economic additionality.” Financial additionality refers to whether these schemes increase In addition, some public schemes are not financially credit and enhance lending conditions to targeted sustainable, requiring constant capital injections from firms. Economic additionality refers to whether firms the government, potentially creating a fiscal burden for that receive guaranteed loans improve their perfor- the public sector and losses for taxpayers. For example, mance (in terms of growth, investment, employment, this could occur because the schemes are being used among other indicators). to pursue political goals (such as to create employment or prevent unprofitable businesses from failing). In The literature has found some evidence of financial some cases, these schemes are also being used to additionality. For example, 67 percent of loans guaran- cover the loan portfolio of state-owned banks. teed by the Canada Small Business Financing Program have been granted to SMEs that otherwise would not Policy Discussion have obtained credit (Seens and Song 2015). Similarly, public credit guarantee schemes have increased lend- Policy makers across the globe have resorted to public ing to SMEs in Italy (de Blasio et al. 2014) and Japan credit guarantee schemes to try to alleviate financing (Uesugi, Sakai, and Yamashiro 2010). Evidence of finan- constraints. However, it is not clear from either a theo- cial additionality has also been found in developing retical or an empirical standpoint that state interven- countries. The National Guarantees Fund in Colombia tion in these schemes is always required. has been successful in increasing loans to SMEs (Castillo Bonilla and Girón 2014). Likewise, Chile’s Supporters of public schemes argue that they are State-Owned Guarantee Fund for Small Entrepreneurs helpful to reduce problems of information asymme- (FOGAPE) has expanded the volume of lending to tries. However, this argument implies that the state has microenterprises and small firms (Cowan, Drexler, and an informational or enforcement advantage over Yañez 2015). private agents, which might not necessarily be the case. A second argument in favor of public schemes is that However, public credit guarantee schemes can also they can be used to subsidize the initial costs of learning have negative effects. As in Italy and Japan, the credit- about new groups of borrowers. But this argument worthiness of firms that participated in the Malaysian would only support a temporary use of public schemes: scheme declined and their default rates increased once financial institutions acquire the necessary knowl- (Boocock and Shariff 2005). In Germany (as well as in edge, these schemes would need to be ended. A third other countries, like the Netherlands), these schemes argument is that the state can spread risks more have been associated with higher risk-taking by banks, broadly than markets because of its ability to deal an indicator of increased moral hazard (Gropp, better with collective action frictions (situations in Gruendl, and Guettler 2014). Furthermore, there is which no individual agent has incentives to act to solve evidence that in some countries guaranteed loans a problem, even though that would benefit everyone). were extended to financially unconstrained firms, Thus, public guarantees could promote lending in the generating deadweight losses by providing scarce presence of high risk or high aversion to risk. financing to firms that did not need it, resulting in a loss to the economy (Zia 2008, among others). Even if the arguments in favor of guarantees are valid, the positive effects need to offset the increased Evidence on economic additionality is also mixed. moral hazard (the pervasive incentives for banks to Guaranteed loans have been found to increase lend to riskier borrowers) that accompanies publicly employment in the United States (Craig, Jackson, and led schemes. Because risk is partially transferred to the Thompson 2008) and Central and Eastern Europe government, banks might have less incentive to screen (Asdrubali and Signore 2015). In the Republic of Korea, and monitor borrowers. Furthermore, it is not straight- firms participating in public credit guarantee schemes forward that public schemes are better than other 3 Are Public Credit Guarantees Worth the Hype? available tools to encourage lending to new markets legal framework. Other work could investigate when (such as direct lending requirements for financial public schemes are warranted. institutions). The answer to this question requires conducting a cost-benefit analysis of public credit guar- Future analyses should try to overcome the existing antee schemes and comparing them with alternative methodological challenges in measuring the impact of state interventions. public credit guarantee schemes (Gozzi and Schmukler 2015; Ioannidou et al. forthcoming). Comparing firms The empirical evidence has not been helpful in that receive guarantees with firms that do not might be settling this debate. In some cases, the use of public difficult because they could be systematically different, schemes has been beneficial, whereas in other cases, or other reasons might be affecting firms’ perfor- they have imposed costs with their net effect being mance. Even when firms with and without guarantees unclear. The varying degree of success of public are otherwise comparable, comprehensive data on a schemes depends in part on whether they are effec- large and representative sample of both groups of tively addressing a market failure. Differences in the firms might not be available. Timing is also important design of public schemes could also explain heteroge- because guarantee programs are usually put in place in neities across countries. the presence of negative shocks. Thus, comparing the Further work is needed to accurately evaluate the effects of this policy with a counterfactual scenario that impact and long-term sustainability of public credit does not involve a downturn might lead to biased guarantee schemes. This work could focus on the results. An extra difficulty is accurately measuring critical factors that can promote sound performance of whether credit guarantee schemes are leading to addi- credit guarantee schemes (such as governance, tional lending or, instead, creditors are switching from pricing, and risk management practices). Additional unguaranteed to guaranteed credit, so that on net work could also try to understand why private schemes there is no additional lending. Another obstacle is might not develop on their own and the role of differ- determining how much time is needed before results ent factors in their development, such as the need for can be observed. Lastly, even when firms receiving reliable data on SMEs, the cost to collect debt after credit guarantees obtain more credit and grow faster, defaults, the existence of secondary markets for they could be displacing other firms, with no aggregate nonperforming loans, and the importance of a sound effect on growth. References Asdrubali, P., and S. Signore. 2015. “The Economic Impact of EU Guarantees Gropp, R., C. Gruendl, and A. Guettler. 2014. “The Impact of Public on Credit to SMEs: Evidence from CESEE Countries.” European Guarantees on Bank Risk-Taking: Evidence from a Natural Experiment.” Economy Discussion Paper 2, European Commission. Review of Finance 18 (2): 457−88. Beck, T., L. Klapper, and J. C. Mendoza. 2010. “The Typology of Partial Credit Honohan, P. 2010. “Partial Credit Guarantees: Principles and Practice.” Guarantee Funds around the World.” Journal of Financial Stability 6 (1): Journal of Financial Stability 6 (1): 1−9. 10−25. Ioannidou, V., J. Liberti, T. Mosk, and J. Sturgess. Forthcoming. “Intended and Boocock, G., and M. N. M. Shariff. 2005. “Measuring the Effectiveness of Unintended Consequences of Government Credit Guarantee Credit Guarantee Schemes: Evidence from Malaysia.” International Programmes.” In Finance and Investment: The European Case, edited Small Business Journal 23 (4): 427−54. by C. Mayer, S. Micossi, M. Onado, M. Pagano, and A. Polo. CEPR and Oxford University Press. Calice, P. 2016. “Assessing Implementation of the Principles for Public Credit Guarantees for SMEs: A Global Survey.” Policy Research Working Paper Lelarge, C., D. Sraer, and D. Thesmar. 2010. “Entrepreneurship and Credit 7753, World Bank, Washington, DC. Constraints: Evidence from a French Loan Guarantee Program.” In International Differences in Entrepreneurship, edited by J. Lerner. Castillo Bonilla, J. A., and L. E. Girón. 2014. “Cuantificación de la Importancia University of Chicago Press. del Fondo Nacional de Garantías en la Movilización de Créditos a las Pymes.” Estudios Gerenciales 30: 18−24. Oh, I., J.-D. Lee, A. Heshmati, and G.-G. Choi. 2009. “Evaluation of Credit Guarantee Policy Using Propensity Score Matching.” Small Business Cowan, K., A. Drexler, and A. Yañez. 2015. “The Effect of Credit Guarantees Economics 33 (3): 335−51. on Credit Availability and Delinquency Rates.” Journal of Banking and Finance 59 (C): 98−110. Pombo, P., H. Molina, and J. N. Ramírez Sobrino. 2015. The Guarantee Systems: Keys for the Implementation. Spanish Association of Craig, B. R., W. E. Jackson, and J. B. Thompson. 2008. “On Government Accounting and Business Administration. Intervention in the Small-Firm Credit Market and Its Effect on Economic Performance.” In Entrepreneurship in Emerging Domestic Seens, D., and M. Song. 2015. “Requantifying the Rate of Incrementality for Markets, edited by G. Yago, J. R. Barth, and B. Zeidman. Springer. the Canada Small Business Financing Program.” SME Research, Industry Canada. de Blasio, G., S. De Mitri, A. D’Ignazio, P. Finaldi Russo, and L. Stoppani. 2014. “Public Guarantees to SME Borrowing. An RDD Evaluation.” Uesugi, I., K. Sakai, and G. M. Yamashiro. 2010. “The Effectiveness of the Dipartimento di Scienze Economiche e Aziendali Marco Fanno, Public Credit Guarantees in the Japanese Loan Market.” Journal of the Università degli Studi di Padova. Japanese and International Economies 24 (4): 457−80. D’Ignazio, A., and C. Menon. 2013. “The Causal Effect of Credit Guarantees World Bank. 2015. Principles for Public Credit Guarantee Schemes for SMEs. for SMEs: Evidence from Italy.” Working Paper No. 900, Bank of Italy. Washington, DC: World Bank. Gozzi, J. C., and S. L. Schmukler. 2015. “Public Credit Guarantees and Access Zia, B. H. 2008. “Export Incentives, Financial Constraints, and the to Finance.” European Economy: Banks, Regulation, and the Real (Mis)Allocation of Credit: Micro-Level Evidence from Subsidized Export Sector 2: 101−17. Loans.” Journal of Financial Economics 87 (2): 498−527. 4 Global Knowledge & Research Hub in Malaysia