The Big Business of Small Enterprises

The Big Business of Small Enterprises An IEG Evaluation of WBG Experience with Targeted Support for SMEs 2006-12 Andrew H. W. Stone IEG, Private Sector JOINT MNSFP-MENA Chief Economist Seminar January 30, 2014

WBG uses systemic and targeted reforms to assist SMEs Systemic reforms are critical to SME priorities, as evidenced in enterprise surveys: Reliable power supply Good governance (reduced corruption) Reasonable taxes Political Stability Fair competition (and restraints on informal competition) Adequate worker skills Access to Finance Development finance institutions also target help to SMEs through: Financing through loans, investments or guarantees Advice to governments and financial institutions Business Development Services and Training Supply/value chain interventions, network and cluster development Many interventions work better in parallel or sequence with others. Evaluating their individual development impact can prove challenging 2

Targeted SME support is a big business for the WBG Averaging $3 billion/year Out of U.S. 963 Projects $18 billion WBG TSME Support by Number of Projects and Total Commitments, Expenditures, and Gross Exposure, FY06-12 A2f= Access to Finance; BDS=Business Development Services, SIP = Small Investment Program, TA-AS=technical assistance and advisory services. (Source: IEG Portfolio Review)

These projects span the globe, as did the evaluation. Case SME Support Study Coverage in IFC Investment, of WBG TSME MIGA Support Guarantee, and World Bank FY06-12 Investment by Portfolios number of projects FY06-12 by number of projects

Two kinds of reasoning for SME Support SMEs make special contributions to developing economies to growth, employment, productivity, and investment; they merit special support. (But literature doesn t provide conclusive evidence of a bigger contribution than large firms to growth and employment in developing country contexts.) SMEs face special challenges. Addressing these challenges levels the playing field contributing to the resolution of systemic economic constraints, and hence to better functioning of markets and institutions. This would allow SMEs to realize their full potential for generating jobs and growth in developing economies. Wellrooted in a number of country and cross-country diagnostics showing size-based differentials in how firms experience the investment climate and business services such as finance.

What does WB enterprise survey data tell us? How firms are constrained depends not only on size, but on the interaction of size with country conditions, especially income level. Any holistic view of SMEs needs to focus a great deal of attention on systemic (nontargeted) reforms.

Theories of Change: From Intentions to Impacts A credible theory of change for TSME interventions must focus on leaving a sustainable supply of the service by establishing well-functioning markets and institutions and not simply providing a temporary supply of benefits to a small group of firms during a project s lifespan. A Targeted interventions need to be strategic, producing broader benefits for institutions and markets. B

Except in small economies, the service gaps are too large for direct support to fill, so impact must be systemic. Nicaragua: Up to 850 MSMEs receive a matching grant, averaging $6,000; c. $5-6 million c. 366,000 MSMEs in Nicaragua 23.2% identify finance as major constraint,.232*366k*6k= $509 million

Problems in WBG approach to TSME What is the definition? Existing definitions do not align and are often contextually inappropriate. For some countries, IFC/MIGA include almost the universe of formal firms The WB often uses local standards. Does it not have a view? At the least, some criteria for defining please! Who should be targeted? Only a minority of projects identify which firms are eligible for benefits for example, 82% of IFC board documents did not define SME and 65% of WB project appraisal documents failed to do so. Provisions often silent on eligibility. Why are we targeting? For IFC advisory, only 27% identify a market failure they are trying to address. While 75% of World Bank TSME lending projects identify a market failure, under half specify what it is. What works? Little guidance on efficacy of the most common forms of TSME support either for direct beneficiaries or for markets and economies, much less appropriate sequencing and complementarities.

Diverse WBG Portfolio with a Heavy Financial Sector focus

For IFC, SME support is seen as a strategic objective based on SMEs job creation potential 17 percent total projects and 16 percent of commitments are TSME. Most IFC investment projects do not define SMEs and do not connect the intervention to correcting a market failure. At the frontier? TSME Investment Portfolio by Product Line (% of commitment value) SME Definition in Board report No. of projects % of projects Not defined 204 82% Greater than IFC criteria 18 7% IFC criteria 14 6% Less than IFC criteria 9 4% Average loan or loan proxy (<$2 million) 5 2% Total 250 100% Source: IEG portfolio review.

Source: IEG portfolio review Best near the frontier, where projects provide demonstration effect or catalyze market improvements IFC is most effective when it works strategically through its targeted investments, often in conjunction with more systemic approaches, to expand the supply of SME oriented financial (or other) services, increase competition and improve market functioning. Distribution of IFC Investment Portfolio by Country Income Level, Region and Industry Group(% of commitment value)

Performance Targeted SME projects less successful than overall portfolio and FM portfolio. However, they improved their performance over time. IFC lacks sufficient monitoring and evaluation information about its TSME projects to truly understand their development impact. Advisory services performed better than rest of portfolio, except in poor countries. Investments accompanied by advisory better impact. Source: IEG portfolio review

MIGA SIP High relevance, questions on effectiveness, efficiency, work quality 57% of MIGA SME projects were underwritten under SIP, but account for less than 8% of gross SME exposure and 2% overall. SIP projects: highly relevant to MIGA s operational priorities of supporting investments in IDA countries, in conflictafflicted or fragile environments, and South-South investments. However, few SIP projects have shown positive development outcomes. Higher risk due to the location of most projects, inherently riskier nature of smaller firms. SIP s streamlined processing of guarantees has not produced anticipated produced efficiency gains, reduced processing time.

Regular MIGA Guarantees: Efficient, but do they reach SMEs? Can channel substantial political risk coverage to benefit SMEs, but currently no mechanism for ensuring funds will go to SMEs. Concentrated on a few clients driven by home country regulations. Underperformed comparable financial sector guarantees in business performance, economic sustainability, and contribution to PSD. No evidence that long-term tenor passed on to endborrowers. The lack of systematic tracking at the intermediary and borrower levels makes it difficult to determine whether expected results were achieved.

WB Lending TSME projects: about 7% of projects and 2% of commitments. Lines of credit, matching grants, and business development services dominate the portfolio. Despite substantial engagement in low income and FCS countries, relatively low level of commitments in IDA countries and high level of commitments in upper middle income countries. Reaching the frontier and building markets where they are weakest?

WB Lending

Challenges in World Bank Lending Decentralized: TSME support is broader than suggested by formal strategic focus, likely more driven by country and regional strategy and demand. Additionality not always clear: Some projects rated as successful for impact on beneficiaries, provide little evidence of impact on systemic obstacles. Work quality strengths: linkage to prior analytic work, high rate of successful development outcomes, high rate of realism in self evaluations. Work quality weaknesses: overly complex designs, overly optimistic timeframes and risk assessments, and the frequent need for delays, restructuring and partial cancellation. Table 4.10. Problems Identified in IEG Evaluations - World Bank TSME Lending Projects Source: IEG. Problem Descriptions Number % of Projects of Projects Inadequate M&E framework, poor data 27 51% quality PIU experience with the WB 26 49% Inadequate technical design 24 45% Implementation disrupted by a crisis 21 40% Inadequate supervision 14 26% Inadequate risk assessment 13 25% Overly complex design 12 23% Inadequate political or institutional 9 17% analysis Inadequate baseline data or unrealistic 9 17% targets Inadequate prior analytic work 9 17% Project Restructured 9 17% Inadequate partner financing or 8 15% coordination At Least One Driver Identified 48 91% Number of Closed Projects Reviewed 53 100%

World Bank TSME AAA A tiny fraction of portfolio Both relevant and important to SME challenges. Delivered mainly to governments. Self-ratings only: high and rising level of success for TA. technical assistance. TA the context of lines of credit appears effective at strengthening institutional performance and therefore in producing positive outcomes. ESW appears effective to inform lending and build capacity but had limited traction in influencing government policy or the development community.

Lessons on sequencing, complementarity and collaboration from 6 mission case studies. In general, a solid base of analytic work both helped to tailor the design of interventions and to build World Bank Group credibility on reforms. Often policy, legal and regulatory frameworks needs to be in place for targeted investments to succeed. Nonetheless, perfection of the policy, legal and regulatory was not required as a precondition to targeted interventions where the state had reasonable credibility with investors and the direction of reform is understood. Each institution should play to its strengths. Where there are shared objectives, coordination and communication strengthen performance. Counterpart/client capacity is key. Problems when too little coordination: 1. Missed opportunities for leverage 2. Issues that fall between the cracks 3. Duplication of efforts 4. Contradiction of efforts

Recommendations (1): WBG Management Should: CLARIFY ITS APPROACH TO TARGETED SUPPORT TO SMEs. Clarify objectives and analytic justification for TSME support, how it relates to systemic reform, where it is appropriate, what main forms it will take and how it will be monitored and evaluated. Root TSME support in an understanding of how interventions sustainably remove SME constraints. ENHANCE RELEVANCE AND ADDITIONALITY. Refine focus to shift benefits from better-served firms and markets to frontier states (e.g. with underdeveloped financial systems), frontier regions and underserved segments. INSTITUTE A TAILORED RESEARCH AGENDA, to understand: A policy- and contextually-relevant definition of SME; contributions of SMEs to jobs and growth; how the design of interventions should relate to country conditions; a project-relevant definition of the frontier ; the correct sequencing and combinations of systemic and targeted interventions; and the actual performance and impact of key types, combinations and sequences of interventions.

Recommendations (2): WBG Management Should: STRENGTHEN GUIDANCE AND QUALITY CONTROL Project documents for TSME projects should define and justify the beneficiary group, provide specific targeting mechanisms, and include impact indicators in its results and M & E frameworks. REFORM MIGA s SMALL INVESTMENT PROGRAM Radically rethink its approach to providing guarantees for investments in SMEs through the SIP program, considering either a merger with its regular program or a fundamental redesign to improve performance.

Pocket Slides

Does support to SMEs (vs. other size classes of firms) produce more jobs and growth? Smaller firms create more jobs, but they also destroy more jobs. Ayyagari, Demirguc-Kunt, and Maksimovic (2011), using cross-sectional survey World Bank enterprise survey data, show there is more job creation in smaller and younger firms. Cross-sectional survey data do not control for survivor bias and composition effects and distinguishing between net and gross job creation. Studies that use panel data, allowing for firms to exit over time, raise questions about a special role in job creation for smaller firms. Haltiwanger, Jarmin, and Miranda 2010; Biggs and Shah 1999, Page and Sonderbom 2012: IEG literature review questions whether evidence shows SMEs making a disproportionate contribution to growth, poverty reduction, or employment. As economies grow, the share of SMEs tends to increase, but no evidence yet that having more SMEs other things equal causes more growth. Substantial evidence that systemic improvements in the business environment and the financial sector can promote growth by improving market dynamics and leveling the playing field, especially for SMEs. But there is little rigorous evidence to support the positive (or negative) impact of targeted programs.

Access to finance is only sometimes a leading SME priority Although SMEs often identify access to finance as their biggest obstacle, the ranking is not robust to parallel measures: The likelihood that a firm will have a loan increases with both firm size and country income level. When the interaction of these two factors is controlled for, the statistical difference in the rate of access for firms with 100-299 firms and firms with over 300 firms disappears.

Review of 10 Lines of Credit, 8 closed and evaluated Ave. 10 components. 60% had to be extended (an average of.9 years), half were at least partially cancelled. Eighty percent had a management risk flag. (Near TSME norm). Work quality challenges: Consideration of alternatives inconsistent and incomplete. M&E system indicators focused on outputs, not outcomes; only 2/8 on subborrowers. No counterfactuals, examination of long-term impact on intermediaries. Follow-ups focused on consumer satisfaction, subject to response bias, based on recall data. Project documents reflected inconsistent observance of OP/BP 8.30: Several did not discuss why IFC was not playing the leading role. Some did not reflect the mandatory Bank/IFC coordination. Some were not justified in terms of important sector and policy reform objectives. Some appeared to offer financial intermediaries better than prevailing market terms.