Report No. 22803-AR Argenti na Small and Medium-Sized Enterprises in Argentina A Potential Engine for Economic Growth and Employment August 2002 Finance, Private Sector and Infrastructure Sector Management Unit Country Management Unit 7 Latin America and the Caribbean Region Document of the World Bank CURRENCY EQUIVALENTS Currency Unit: Argentine Peso (AR$) AR$ 1.0 = US$0.2739726 (August 1, 2002) WEIGHTS AND MEASURES Metric System FISCAL YEAR July I -June 30 ACRONYMS ABPPRA Association of Banks AFIP Administraci6n Federal de Ingresos Publicos BCRA Banco Central de la Republica Argentina BDS Business Development System BICE Banco de Inversi6n y Comercio Exterior BNA Banco de la Naci6n Argentina CDE Centros de Desarrollo Empresarial CEDES Centro de Estudios de Estado y Sociedad CEMA Universidad del CEMA (Argentina) CEPAL Centro de Estudios para America Latina CFI Consejo Federal de Inversiones CONICET National Council for Science and Technology (Argentina) CUlT Clave de Identificaci6n Tributaria DPPJ Direcci6n Personas Juridicas de la Provincia de Buenos Aires. DGI Direcci6n General de Impuestos DOP Direcci6n de Obras Particulares EMS Environmental Management Systems EPH Encuesta Permanente de Hogares FDI Foreign Direct Investment FIEL Fundaci6n de Investigaciones Econ6micas Latinoamericanas FONCYT Fondo para la Investigaci6n Cientifica y Tecnol6gica FONTAR Fondo Tecnol6gico Argentino FUNDES Fundaci6n para el Desarrollo Sostenible en Anidrica Latina IADB-FUNDES Inter-American Development Bank-Fundaci6n para el DesarroLlo Sostenible en Amdrica Latina IAMC Industrial Asset Management Council ICT Information and Communication Technology IDEB Instituto para el Desarrollo Empresarial Bonaerense IERAL Instituto de Estudios sobre la Realidad Argentina y Latinoamericana IMPE Instituto de Producci6n y Empleo INDEC Instituto Nacional de Estadfstica y Censos de la RepubLica INTI Instituto Nacional de Tecnologfa Industrial IPO Initial Public Offer LAC Latin America and Caribbean Region MIF Multilateral Investment Fund MSME Micro, Small and Medium-Sized Enterprises OECD Organization for Economic Cooperation and Development PIU Project Implementation Unit RGS Sociedades de Garantfa Reciproca SECyT Secretarfa de Ciencia y Tecnologia (Argentina) SEPyME Secretaria de Pequeiias y Medianas Empresas SMEs Small and Medium-Sized Enterprises TFP Total Factor Productivity UIA Uni6n Industrial Argentina UIA-IDB Uni6n Industrial Argentina/Inter-American Development Bank IBRD ' Vice President David de Ferranti Chief Economist Guillermo Perry Country Director Axel van Trotsenburg Sector Leader Juan Gaviria Task Manager J. Luis Guasch Small and Medium-Sized Enterprises in Argentina: A Potential Engine for Economic Growth and Employment* Table of Contents Executive Summary .............. 1. Introduction .x 2. SMEs are critically important but have not realized their potential .xii 3. SMEs are very heterogeneous agents .xiii 4. Business Environment Ridden with Institutional Rigidities .xv 5. Significant Problems with Access to Finance .xix 6. Weak Demand for Knowledge .xxi 7. Current Government Policies: Multiplicity of Programs, Uncertain Impact. xxi 8. The Path Ahead: Towards Incentive-Driven SME Policies ......................................... xxiii 1. Introduction...1 1.1 Objectives .1 1.2 Data and Methodology .2 1.3 Main Findings .3 1.4 Organization of the Report .4 2. SMEs as Potential Engines to Economic Growth and Employment: Anatomy of the Sector . 5 2.1 The Role of SMEs in Argentine Economy .5 2.2 SMEs in the City of Buenos Aires .6 2.3 SMEs are Different from Large Firms: Size Matters .8 2.4 Productivity in the Manufacturing Sector .9 3. SME Response to Economic Liberalization: Determinants of Success and Growing Heterogeneity of Firms ...12 3.1 Uneven supply response to reforms ............................................................... 12 3.1.1 Manufacturing SMEs: Inter-industry Differential Responses ......................................................... 13 3.1.2 Intra-industry Heterogeneity of Manufacturing SMEs ............................................................... 15 3.2 Evidence of Successful Exporting SMEs................................................................s15 3.2.1 Successful SME exporters.................................................................16 3.2.2 Exporting as an incremental learning process ............................................................... 17 3.2.3 Export Project Cycle: From Identification of Export Opportunity to Sustainable Exports . 17 Identification of exports oppopunities .17 Acquisition of sustainable export capabilities .18 3.3 Dynamics of Industrial SMEs................................................................D.18 3.3.1 W hat Detem ines Sales Growth? .................................................................20 This report was written by Jose Luis Guasch (Task Manager), Yevgeny N. Kuznetsov, and Susana M. Sanchez (LCSFR). Contributions by Osvaldo Giordano, Jorge Colina, Gerry McDermott and FIEL are gratefully acknowledged, as well as FUNDES for their gracious courtesy of sharing their work and data for this report. Cara Zappala and Jocelyn Troncoso provided research assistance. Peer reviewers: Paul Ballard, Claudio Frischtak, Margaret Miller, Juan Llisterri, Peter Mousley and Javier Gonzalez Fraga Danny Leipziger, Paul Levy and-William Maloney provided useful comments. iv 3.3.2 Determinants of Labor Productivity in the City of Buenos Aires ................................................... 22 3.4 The Service Sector SMEs: Preliminary Evidence ............................................ 23 3.5 Implications for SME policy ............................................ 23 4. Institutional Rigidities: Environment for Business Activities ................................................. 26 4.1 Business Registration System .26 4.1.1 Commercial Registration .27 4.1.2 Operational Permiits and Licenses at the Municipal Level .28 4.1.3 Registration for Tax Purposes .30 The Simplified System to Pay National Taxes - The Monotributo ......................................... 30 4.1.4 Labor registration .31 4.2 Periodic Costs of Business Formalities .32 4.3 Labor Market Rigidities .32 4.4 Logistics and Infrastructure Issues .34 4.5 The Consequences of a Complex Business Environment .35 5. Financial Markets and SMEs ............................ 38 5.1 Participation in Credit Markets .38 5.2 Capital Structure .......... 39 5.3 Bank Lending to Small and Medium Enterprises .40 5.3.1 What are the Determinants of SME Lending by Banks? .41 5.3.2 What Explains the High Interest Rates of SME Loans? .45 5.3.3 Relationship Lending .46 5.3.4 Borrowing Concentration .47 5.3.5 Financing Exports .48 5.4 Leasing and Factoring .48 5.5 Risk Capital Investment - Private Equity Investments .48 5.6 Reciprocal Guarantee Societies .50 Regulation of Reciprocal Guarantee Societies .50 5.7 Under-developed Financial Infrastructure .51 5.7.1 Information Problems .51 5.7.2 Regulation Problems .52 5.7.3 Legal Framework for'Secured Transactions .53 6. Knowledge Constraints ......................................................................... 54 6.1 Low-level equilibrium trap as a result of the knowledge constraint ............................................... 56 6.2 Knowledge constraints common to all SMEs ......................................................................... 56 6.3 Differentiation of knowledge constraints according to capabilities of firms .................................. 57 6.4 Agencies and programs to alleviate knowledge constraints: lessons from evaluation .................... 58 7. Toward the Reform of SME promotion: current policies and lessons from best practice ........................ 62 7.1 Why Foster SME Development? .62 7.2 Current government policies and programs .63 7.2.1 Government Efforts on Business Deregulation .63 Simplification of Labor Regulations .64 Simplification of Tax Regulations .64 7.2.2 Financial Markets .65 7.2.3 Business Development Services .67 8. The Path Ahead: Policy Recommendations .74 8.1 General Strategy .74 8.2 Reform of SME promotion system: Crisis as a window of opportunity .77 8.2.1 The growth pay-off to dramatic reform of SME promotion system are both significant and immediate 77 8.3 Eliminating Government Failure: Business Deregulation .82 v Managing the Process: Deregulation for growth. Short, medium, and long-term horizons ............................. 82 8.4 How to Deepen Financial Markets? ......................................................... 83 8.5 SME Assistance Programs: Institutional Design and Directions for Reform ................ ................. 84 8.5.1 Consolidation of existing programs ......................................................... 84 8.5.2 Piloting of new demand-driven decentralized SME programs ....................................................... 84 8.5.3 Design and nation-wide discussion of a new system of business development services to SMEs. 84 8.6 Longer-term options to reform the SME promotion system ................ ......................... 85 8.6.1 Transition at the Federal level ......................................... 85 8.6.2 Institutional design of the new decentralized system ......................................... 86 The Matching Grant Principle .......................................... 87 The Business-Climate Ranking System ......................................... 87 8.7 Conclusion ......................................... 91 9. References .......92 vi List of Tables Table 2.1: Argentina's SMEs and International Comparison ......................................................6 Table 2.2: Characteristics of Argentine Enterprises in the City of Buenos Aires, 1999 ...........................8 Table 2.3: Comparing SMEs and Large Firms in Argentina ......................................................9 Table 2.4: Average Productivity of Labor by Firm Size (Total Index = 100) ......................................... 11 Table 2.5: Average Productivity of Capital by Firm Size (Total Index = 100) ....................................... 11 Table 2.6: Working Capital (As a Percentage of Sales) ............................... I11 Table 2.7 Capital-Labor Ratio According to Size (Total Index = 100) ................................................ 11 Table 3.1: Rates of Changes in SME Production and Employment ........................................ : 14 Table 3.2: Economic Liberalization: SME Winner and Losers ....................................... 14 Table 3.3: SME Production space: Changes in Output Share and in Foreign Trade Ratios in Sectors with 'Active', 'Passive' and 'Adaptive' Responses, 1984-93 .15 Table 3.4: SME Production Spacea: Changes in Employment in Sectors with 'Active', 'Passive' and 'Adaptive' Responses, 1984-93 ...................................... 15 Table 3.5: Structure of Industrial Exports in 1999 ...................................... 17 Table 3.6: Kendall's Correlation Coefficient ...................................... 20 Table 3.7: Industrial SMEs in 1996 ...................................... 21 Table 3.8: Industrial SMEs ...................................... 22 Table 3.9: Industrial SMEs - Determinants of Success 1996-1998 ...................................... 23 Table 3.10 Regression Results: Determinants of Labor Productivity Among Manufacturing SMEs in the City of Buenos Aires, 1999 ...................................... 24 Table 3.11: Taxonomy of SMEs in Argentina: 'Success', 'Muddling Through', 'The Case for Exit' ..... 26 Table 4.1: Business Registration in Selected Latin American Countries ............................................... 29 Table 4.2: Costs of Complying with Business Formalities by Business Size . 31 Table 4.3: Structure of Labor Taxes .............................................. 35 Table 4.4: Logistic Cost as % of product value per country .............................................. 36 Table 4.5: Direct Import Costs .............................................. 36 Table 4.6: List of Steps (Tramites) to Create and Register a Business .............................................. 38 Table 5.1: Number of Financial Intermediaries .............................................. 42 Table 5.2: Commercial Debt Balances with Banks by Firm Size: 1998-2000 ........................................ 43 Table 5.3: Determinants of SME Lending among Argentine Banks .............................................. 46 Table 5.4: Non-performning loans by Type of Firms (%) .............................................. 47 Table 5.5: Borrowing Concentration among Debtors in 1998 ............................................................ 49 Table 5.6: Sociedades de Garantfa Recfprocas (September 1999) .......................................................... 52 Table 8.1: Summary of Market-Oriented SME Interventions ............................................................ 68 List of Figures Figure 2.1: Underemployment by Firm Size (Percentage) ............................................................ 10 Figure 2.2: Unit Labor Cost (in constant pesos 1992=100)* ............................................................ 10 Figure 2.3: Median Labor Productivity by Firm Size (Total=100) ........................................................... 10 Figure 4.1: Average Explicit Costs of Business Registration in Argentina .............................................. 30 Figure 4.2: Labor Registration: Firm's Tramites ............................................................ 33 Figure 4.3: Periodic Cost of Business Formalities ............................................................ 34 Figure 4.4: Share of Informal Workers in Buenos Aires ............................................................ 37 Figure 4.5: Creating a Business is a Complex Process ............................................................ 39 Figure 5.1: SMEs' Access to Finance ............................................................ 40 Figure 5.2: Access to Bank Loans among Industrial SMEs in 1998 ........................................................ 41 Figure 5.3: Financial Structure of SMEs ............................................................ 41 Figure 5.4: Estimated Share of Lending to SMEs and Growth Rates of SME Lending by Bank Size for Foreign and Domestic banks in Argentina ............................................................ 44 Figure 5.5: Evaluation of Credit Services by Exporting SMEs ............................................................ 49 Figure 6.1: Surveyed Argentine SMEs Knowledge on E-Commerce Subjects ........................................ 56 Figure 6.2: SMEs that Perform Electronic Transactions ............................................................ 56 Figure 6.3: Reasons Why Surveyed Argentinean SMEs do not Perform Electronic Transactions ........... 56 vii Figure 6.4: Obstacles to the Development of E-Commerce in Argentina ...........................................:.56 Figure 7.1: BDS Programs for SMEs/GDP .64 List of Boxes Box 3.1: - Working Hypothesis - Fundes' Study .18 Box 5.1: Main shortcomings of the legal framework for risk capital investment in Argentina. 50 Box 6.1: Legal Aspects to be Considered for the Development of Electronic Trade .57 Box 7.1 Credit Guarantees .63 Box 8.1: Performance Oriented SME Organizations: Factors of Success .73 Box 8.2: Govermment as Facilitator of Private Learning: SMEs Adopt ISO-1400 Standards Through Innovative Supplier Development Program .74 Box 8.3: From top-down to bottom-up SME promotion .76 Box 8.4: Factors of Success of Sub-National Private Sector Development Agencies .77 Box 8.5: Argentina: Co-Investment and Business-Climate Ranking as Criteria for Funding Provincial Public Works .80 Box 8.6: Contests for federal funds between sub-national govermnents: Lessons from the World Bank projects .82 viii Executive Summary 1. Introduction In Argentina, convertibility law and economic liberalization in the early 1990s brought about dramatic changes in economic performance. For a while, the structural reforms proved reasonably successful in attracting FDI and generating growth through the expansion of large domestic conglomerates and multinationals. In the end, however, the economic path induced by the previous policy regime proved unsustainable and finally the Argentine economy precipitated into a deep and prolonged crisis. But even before the advent of the current difficulties, this growth was very unevenly distributed between regions, sectors, and economic agents. The productivity gap across the macroeconomic cycle between small and medium-sized enterprises and large firms remained significant signaling unrealized opportunity of SMEs as a potential source of dynamism, flexibility and growth diffusion as well as employment. To adjust effectively to increasing globalization and a series of external shocks (the crisis in East Asia, the default by Russia, the devaluation of the Brazilian real), SMEs were confronted with'the task of developing business strategies to secure their niches in the new arena. The, viability of these strategies, however, were obstructed by various constraints in the legal and economic framework, difficulties in acquiring information and technology, and insufficient access to finance. Ten years after the beginning of economic reforms, and in the midst of one of the worst economic and financial crisis of Argentina's history, SMEs, with small exceptions, still need to take off and realize their potential. The purpose of this report is to assess the constraints affecting Argentine SMEs in terms of creation, development, and performance, and identify appropriate and justifiable policies and institutions that can help improve their performance, and contribute to economic growth and job creation. In addition to the serious problems that SMEs face because of current macroeconomic and financial conditions, Argentine SMEs face constraints from many sources. Although this report cannot examine all of them, it focuses and analyzes the three key groups of constraints: (a) inadequacies of the business environment; (b). insufficient knowledge, information and technology; and (c) access to finance and high cost of borrowing. The central policy questions of this report are how to address these constraints; how to correct or ameliorate the effects of market and government failures; how to eliminate unnecessary regulations; what kind of support is appropriate and how best to design SMEs assistance programs. Overall, the report proposes policy interventions in the business environment and in the financial sector and proposes the facilitation of efficient institutions of public and/or private service delivery that as a public good benefit all firms, but particularly SMEs. This report shows that such institutions are better created at the state and regional level, then scaled to the national level. This second-generation institutional agenda is at the center of a sustainable and market-driven incentive framework for SME development. This report presents seven main messages: a) The SME sector is critical to the Argentine economy both from a growth/efficiency and equity standpoint yet on average, it has failed to realize its potential. SMEs account for 69-78 percent of employment and 61 percent of production. Another reason why the SME sector is critical to the Argentine economy is that it can constitute the basis for the reconstruction of a network of local dynamic firms. In fact, the evidence in many developing economies suggests that the lack of a national entrepreneurial basis can be the source of macroeconomic and even institutional instability, and also that its presence is crucial for fully taking advantage of the benefits derived from FDI and other features of the current process of globalization. In other words, localization of decision making matters. Small and medium-sized enterprises have not realized their potential, partially because of their own deficiencies but also because of government failures and standard market failures affecting economic activity. Hence the current predicament of SMEs is also an opportunity. Once onerous regulatory burden is relieved, one can expect SMEs to become x a source of growth and employment. The current crisis, and the need to jump-start growth is an important window of opportunity to reduce regulatory burden and start the reform of SME policies and programs. b) Smal and medium-sized enterprises are highly heterogeneous and are not just smaller-sized replicas of large firms. SMEs have special organizational and technological characteristics, which would have been irrelevant in the world of perfect markets and perfect govemment. It is market and/or govemment failure, which makes SMEs distinct: they have much lower adaptive capacity to deal with market and govemment failures, and they are the most cost impacted by burdening regulations, lack of markets deficient business environment. Of course, not all of those differences demand or justify public sector interventions but there seems to be a broad consensus on the positive impact that a well designed and implemented policy can have on SMEs performance. c) Argentina's business environment suffers from a high degree of institutional rigidities, which partly explains why roughly one quarter of economic activity is performed outside laws and regulations and only half of the labor force is registered with the social security. Registration and creation of a legal entity is a time consuming process for SMEs, with official costs of compliance ranging in 2001 from $1,755 to $6,500 depending on business size. Overall, the regulatory burden of the business registration system is very high, falling more heavily on small firms and giving large firms a comparative advantage. d) There is a need for policy action to deepen financial markets for SMEs. Small and medium enterprises received, on average, about 9.8 percent of bank credit during 1998-2000, but they contribute to about 61 percent of Argentina's value added. Other countries present similar findings, such as Peru, where SMEs receive 15.1 percent of bank loans, while generating 42 percent of the country's GDP. Nonetheless, until the recent collapse of the financial system access to finance per se did not seem to be a big problem. In fact, about 58 percent of SMEs in the City of Buenos Aires had access to finance in 1998-1999. The problem lied rather in the conditions under which SMEs received credit. Overall most SMEs had access to very expensive short-term loans in the form of overdrafts from current accounts, which may not be the most effective form to finance working capital or investment needs. At present, given the current conditions of the banking system, access probably became the most pressing SME's problem regarding financing matters. e) Knowledge constraints are significant. In Argentina, most SMEs do not use training and technical assistance because services available are directed towards large firms and are too expensive for SMEs. Overall, about 38 percent of SMEs purchased technical assistance services and 62 percent received training courses during 1997-1998, leaving 30 percent without such services. Although the data to date is scarce, small and medium enterprises appear on average to suffer from isolation. The special nature of knowledge of markets, technologies, and supplies, plus the fact that such knowledge would still require substantial effort on the part of the firm to assimilate it, put greater weight on demand factors for information rather than on supply factors. This ability to incorporate lessons and new ideas is a key factor to remain competitive in the new economic environment. f) In Argentina, there is a striking multiplicity of SME programs, yet with uncertain impact. The type of public intervention is important. It is not so much a matter of additionality of resources to the sector (as the amount spent for the sector in Argentina is ample) as it is the issue of the focus targeting, and particularly the incentive structure and the delivery of those services. g) Argentina needs to overhaul its SME policies and programs to make them more incentive- and demand-driven. Not all SMEs are created equal, and they respond differently. Assistance xi programs for them to be effective have to be discriminatory and responsive to performance. Those programs have to induce self-selection, be incentive driven, adapted to the capabilities of SMEs, and the assistance gradual, matched to the fi-ms' performance. 2. SMEs are Critically Important but have not ReaL;zed their Potential Small and medium-sized enterprises play a much more important role in economic growth than had been acknowledged previously. They are the single most important source of employment in many countries, while in Argentina they account for about 69-78 percent of employment. For some sectors, their share of employment is greater than their. share of production, indicating lower levels of productivity. In terms of dynamics, both employment and number of establishments in manufacturing SMEs have decreased quite significantly in the manufacturing sector, but increased in the service sector in inter-census period 1984- 94. h) The SME sector is characterized, on average, by a lack of dynamism and low productivity, (mainly in the service sector), and low rates of entry and growth. Although superficially some of the SME subsectors-i.e., manufacturing- look healthy, a deeper examination reveals that SMEs face major problems, including difficulties to start-up and to grow. Most SMEs in Argentina are still struggling to adapt to competition in the global marketplace ushered by trade liberalization and the reduced state role in the economy. Despite their continued economic importance, a limited number of new SMEs enter the marketplace and many more struggle to survive or die of attrition. Though the data is scarce, recent estimates by the Central Bank suggest that mortality rates reached 18 percent, about twice the birth rate during the 1990s. In many cases, SMEs focus on survival in the short-term rather than on repositioning for domestic, regional, and international competition. Hence, they are in a precarious position to integrate their activities into the global economy. The current situation of the Argentine's economy has certainly worsened their already weak position. SMEs are Differentfrom Large Firms: Size Matters Argentine SMEs, like those in other countries, have special features of their own: their economic rationale is of a particular kind, their behavior is specific, and they are impacted quite differently by regulations and risk factors. Generally speaking, these firms are family businesses and their administration falls directly on the owners (often untrained). Thus, their management often relies on untrained individuals, and the owner's outlook (generational cultural factor) is especially important in deciding actions that determine a firm's evolutionary process. In fact, SMEs not only differ in regard to how effective they are at strategic planning, but may also even pursue goals that differ from profit maximization, such as market survival, maintaining their jobs and sources of income, or keeping up the family business tradition. For all these reasons, SMEs might need to be treated as economic agents, which are not merely "smaller-sized large ones." There are significant differences relating to the size of the firm. Of particular relevance in the comparison between SME and large firms, are factors that significantly impact their unit costs, interest rates and transaction and logistics costs, all significantly higher for SME. While differences are to be expected and should not be, in principle, a cause for intervention, components of those differences that are induced by market and government failures are. xii. Productivity in the Manufacturing Sector Productivity and usage of inputs differ significantly according to firm size. Again this is not surprising given differences on modes of production, scale effects and the myriad of costs with differential impact according to size. In any event, it does clearly impact the prospects for growth and development of SMEs. According to Fiel (1996), among manufacturing firms, labor productivity for larger firms was 2 to 4 times higher than that of smaller firms both whenfirm size was measured by number of workers and gross production value. In 1999, the manufacturing sector of the City of Buenos Aires presented the same pattern. In terms of productivity of capital, Fiel (1996) presents the same dramatic differences by size of firm. Also, SMEs require a much higher proportion of working capital as a percentage of sales than larger firms, impacting unit costs and productivity. In summary, the case has been made that SMEs are characteristically and behaviorally different from large firms and are impacted differently as well (Table 1). That in itself need not be a reason for intervention. A closer look at the environment under which they operate, and the causes behind those differences, are essential in evaluating the case for intervention, at least on efficiency grounds. To unleash their potential, unit costs have to come down, and productivity and quality standards have to increase. Firms on their own can and should take measures to improve their predicament, but the government also needs to address and correct inefficiencies and market and government failures that enhance those differences across firms of different sizes. The questions remain who should intervene and what is it that needs to be done in order to achieve those results. Table 1: ComparingFirms by Size in Argentina Concept Small Medium Large Organization Individual Owner S.A. S.A. Avg. age 28 years 38 years 57 years % that sub-contract 4.5% 9.5% 52% Operate with bank Loans 76% 86% 98%- Avg. annual interest rates on >25% >23% 12% bank loans Informality (% of workers) 70% 32% 10% Transaction Costs (% of sales) 15% 7% 1% Logistic Costs (% of sales) 35% 26% . 15% R&D investment 0.3% 1.5% 3.8% Technological support Customers and Providers Custoners and Providers Providers and other sources Initial development based on: Previous experience Experience and access to new Access to new technology I _________________I________ ___________________________ technology Source: FIEL (1996, 2001) and Bank'Staff estimations. 3. SMEs are very Heterogeneous Agents As expected there was a highly uneven supply responses to economic liberalization of SMEs, highlighting how diverging firm-level trajectories affect inter-industry and intra-industry heterogeneity. The growing heterogeneity of SMEs emerges as the single most salient feature of the aftermath of economic liberalization. Small and medium-sized enterprises, like all other economic agents, differ from each other. This chapter identifies three types of SMEs: 'successful', 'muddling through' and 'failing' firms. This class of heterogeneity merits special consideration. Although inadequacies of the business environment, knowledge, information and technology constraints, and low access to finance affect all types of SMEs, each type of SMEs have different capabilities to face those growth constraints. Firm surveys, case studies and econometric analysis of panel data all pinpoint existence of three classes of SMEs: successful firms, 'muddling through' SMEs and 'clear failure'. Although these three ideal types are best to be perceived as the extremes and the middle point of a continuum of widely heterogeneous xiii economic agents, yet for the policy purposes it is useful to summarize the main features of these three types of SMEs (Table 2). The heterogeneity of SMEs and their responses to economic liberalization implies heterogeneity of Business Development System (BDS) and financing needs and constraints. Successful SMEs tend to have quite advanced technical, managerial and marketing capabilities whereas the 'persistentfailure' SMEs have none of these capabilities. The 'muddling through' firms fall somewhere in between. It does not mean that successful SMEs do not face BDS constraints and do not have BDS needs. On the contrary, as the surveys of successful exporters of FUNDES clearly demonstrated, successful finns display the highest effective demand for BDS services. What it does mean, however, that BDS services required by successful SMEs will be quite different from the services for 'muddling through' and 'surviving firms'. Successful finms require high intensity specialized services and are willing to pay full market costs for these services. Muddling through firms may require 'light touch' inexpensive BDS and may require limited-time subsidies on a matching fund basis to show the value of the BDS to the firm and to assure subsequent market demand for the services. 'Loser firms' in contrast may require very basic benchmarking design and assistance to the owner/manager to exit or switch activities of the firms to more promising and profitable lines of business. To cope with such heterogeneous features and needs it is crucial to stress the importance of local level involvement in the design and implementation of SME policies. Table 2: Taxonomy of SMEs In Argentina: Implications for Policy Muddling througb - surviving Clearly Unsuccessful, on their Successful SMEs SMEs Way to Liquidation The major chalenge Expansion and growth: Re-founding the firm on new Exit. Phasing out of production facing the firm specialized marketing and foundations on the face of activities technical support, access to dramaticaUy new opportunities and finance challenges opened by economic liberalization Capacity for auto- Quite high, in particular vis-&- Limited to very basic comparison vs. Virtually non-existent. There is a diagnostics and vis the domestic and Neighborhood firms and most direct need to convince the family owners benchmarking international competitors competitors that exit is in his/her best interests Technological capabilities Formal depamnent or informal Informal technology/ engineering - technology development teams groups may exist Marketing and managerial Usually there are formal Quite weak and tend to develop on ad- Rudimentary capabilities marketing departments (taking hoc basis up to 5% of personnel) Alliances with other firms 80% of successful exporters If alliances exist, they are mostly No alliances/networking. Survival involved in some kind of informal orientation and very short planning aliance. Transition from horizon precludes participation in infonnal agreements and inter-firms networks alliances to formal agreements Subcontracting Staring to emerge. The Occasional Almost non-existent arrangements owner/manager is actively seeking subcontracting arrangement with large firms Focus of public policies Facilitation of high-intensity Faciltation of 'light touch' relatively Assistance in exploring marketing and technical support inexpensive marketing and technical employment opportunities for based on almost full cost- support services. owners and employees to ease the recovery of services. exit and closure of the business Development of private-public service Development of private service providers (e.g. BDS centers of UIA) providers and thickening of the market for business development services xiv 4. Business Environment Ridden with Institutional Rigidities The environment for business activities refers to norms, rules, regulations, laws, and procedures that shape the behavior and operations of businesses in an economy, particularly those dictated by government policies. As a set of rules, the business environment can constrain or encourage business activities, with significant impact on productivity and unit costs. Business Registration System Business registration is an important element for start-ups in any economy, and the relative ease or difficulty in registration is an important component of vitality of the private sector. It is the first point of contact between an entrepreneur and the government, and as such it sets the tone for their entire future relationship. In Argentina, this relationship does not start off in the most positive way since SMEs typically face a process that can take up to several months and require several hundred dollars in fees. Compared to other Latin American countries, Argentina ranks among the worst performers next to Guatemala and Venezuela. It takes about 130-150 working days to comply with all registration requirements. As in other countries, business registration procedures can be grouped into four types: (a) commercial registration of businesses; (b) acquisition of operational permits and licenses at the municipal level; (c) compliance with tax regulations; and (d) labor and social security registration. To comply with each component of the registration process, businesses must perform a number of formalities (trdmites) at different levels of government, visit various agencies, provide documentation and fill out forms, and pay fees and commnissions. Moreover, for each formality, businesses have to wait a certain number of days for approval. Some registration requirements occur once usually when becoming formal, while other requirements entail periodic costs of formality (e.g., tax payments and workers' social security contributions) (Table 3). Overall, in 2001 the cost of commercial registration in Argentina ranged between $800 and $1,900, depending on the business size. Although small firms would pay a lower absolute value, relative to their monthly company value added, the fixed costs of commercial registration could reach 18 percent of the firm's monthly value added, with wide differences by business size (Figures 1 and 2). Figure 1: Periodic Cost of Business Formalities ~~~~~~~1 11 -25 2 4 41 0 Ag Source:, Authiors' calculations xv Figure 2: Average Explicit Costs of Business Registration in Argentina 25~~~M 120 181 *L~~-L Rgbkdbn klon Wr (jeimb & IbWane) Table 3: Costs of Complying with Business Formalities by Business Size (2000) US$ BusinessSize (rY oemplovees) 1 2 - 5 6-10 11- 25 26 - 40 41 - 50 51 + Avg. Average Monthly Value Added per Business a/ 980 2,W 11,873 23,746 59,409 82,83 340,565 6,125 1. Registration Costs at the National Level Cost of documents 270 270 270 270 270 270 270 270 Professional or consulting fees (includes cost of an 400 400 400 600 800 800 1,500 700 accounting study) Application for "urgent procedure 130 130 130 130 130 130 130 130 Total cl 800 800 800 1,000 1,200 1,200 1,900 1.100 % of avg. Monthly value added 82 31 7 4 2 1 1 18 2. Registration Costs at the Municipal Level b/ Cost of documents 200 200 200 200 200 200 200 200 Tax advances 200 250 300 350 400 400 400 329 Professional or consulting fees 300 300 600 800 1,000 1,200 1,500 814 Total 700 750 1,100 1,350 1,600 1,800 2,100 1,343 % of avg. Monthly value added 71 29 9 6 3 2 1 22 3. Costs of Tax Registration Cost of documents d/ 70 70 70 70 70 70 70 70 Cost of tax advances e 55 75 75 75 100 100 100 83 Professional or consulting fees 50 80 150 200 300 400 1,000 311 Total 175 225 295 345 470 570 1,170 464 % of avg. Monthly value added 18 9 2 1 1 1 0 8 4. Costs of Registering Workers Total f 50 80 150 200 300 400 1,000 311 % of avg. Monthly value added 5 3 1 1 1 0 0 5 5. Total Registration Cost (1+2+3+4) . - 1.725 1,855 2,345 2,895 3,570 3,970 6.170 3,219 -%bf avg. Monthly value added .. . 176 72 20 .12. 6 5 2 53 6. Cosis of Fonnaliies to Pay Taxes and Laor 50 - 100 499 . 807 1.485 1,988 5.108 165 Obligations per Month - _ - - - - - % ofavg. Monthly value added 51 39 4.2 3.4 2.5 2.4 15 2.7 Source: a/ Authors' calculations based on information fom the 1994 Economic Census, INDEC, Argentna b/ Authors' calculations based on information presented in FUNDES-IDB (2000), which are representative for the municipalities of the city of La Plata c/ Includes: costs of publication in the official gazette, costs of certifying documents, and costs of hiring professional accounting studies. Excludes: initial payment of one percent of capital for the Stamps Tax, annual levy as percentage of capital required by the Province of Buenos Aires, stamp tax to register accounting books. d/ Assumes a cost of $70 to obtain certified copies of the business' by-laws and rental contract or real estate ownership deed e/ Includes the cost of making tax advances the for the Gross Income Tax f/ Includes an estmate of the prices charged by accounting firms for this type of professional consulting and the costs of depositing the amounts deducted and in paying contributions. xvi The Judiciary and Supervision Slow and high cost legal procedures, and lack of arbitration mechanisms to solve commercial disputes between firms impose a heavier burden on SMEs than on large fimns. In many cases SMEs are left without the option to use existing regulatory and supervisory institutions. Deficient regulation of noncompetitive markets, like public utilities, impair the performance and competitiveness of the economy and, in particular, of SMEs. SMEs face noncompetitive markets even in the markets of some tradable inputs (i.e. steel). As is often the case, the creation of an institution like the "Secretarfa de Defensa de la Competencia" (Competition Agency) did not improve the situation. In the case of many tradable inputs, trade liberalization was undermined through special regimes for those sectors. Labor Market Rigidities In Argentina, workers have historically enjoyed strong job rights, including the right to a generous severance payments and to ultractivity clauses (collective bargaining agreement remains in place until a new one is signed). The level of non-wage labor cost is significantly high and its imperfect link to direct benefits induces a high tax wedge, one of the highest in LAC. Moreover Argentina has a very highly centralized collective bargaining agreement system. All types of firms, small, large, and independent of their geographical location and predicament are subject to the same agreements negotiated by the heads of the union (cupulas). Clearly, that adversely impacts SMEs not providing them with the flexibility they need to address their particular and individual needs and predicament. Some attempts to provide SMEs with additional flexibility have been tried but very lirmited practical success. In addition to a cumbersome and costly labor registration system, labor regulations also introduce other distortions, such as expensive costs of dismissal and hiring restrictions. Similarly, firms must contribute to union-sponsored health programs with limited benefits to show for. Finally, firms have limited flexibility in the type of labor contracts used. In short, labor regulations in Argentina remain rigid, resulting in substantial costs for all firms, particularly SMEs. As an example, according to a recent study on the effect of labor regulation on firms' employment decisions in Argentina, tightening of labor regulations reduces the number of jobs while increasing hours worked of those workers that kept their jobs, which is the opposite effect sought by regulators through the 1990s. Argentina has made efforts to reform the sector securing marginal improvements but with mixed results. The Labor Reform Law of 2000 aims at improving labor market conditions. Additional regulations and actions are required to achieve a larger impact. Logistics and Infrastructure Issues Cost related to logistics is a major factor impacting the way of doing business in Argentina. Logistics costs are a significant obstacle to the development of SMEs, reducing productivity, increasing unit costs and with the detrimental effect on the competitiveness of Argentine firmns. Among the major LAC countries, Argentina remains the second (after Peru) with the largest logistic costs in the Region. The Consequences of a Complex Business Environment Among other factors, shortcomings of the regulatory framework for business formalization have important implications for private sector competitiveness and efficiency, particularly among SMEs. These include: * Business Informality. Not surprisingly, given the onerous compliance costs associated with business formalities, a significant share of SMEs' activities is performed outside the laws and regulations. The regulatory framework is biased towards smaller firms. First, the costs of registering a business, such as those documented here are much higher, relative to the fimn's value added for smaller firms. Second, the monthly costs of complying with tax and labor regulations amount to a staggering 3 to 7 percent of the firm' s value added for fnrns with less xvii than 25 workers. Thus, labor and tax regulations are biased against smaller firms. The amount that they must spend on bureaucratic red tape is almost equivalent to what they must pay in income tax and the value added tax through the monotributo, and other local taxes. Thus, growth and development of SMEs are hindered. It is not only the individual firm that loses, but society as well, since the benefits, such as fiscal impact, social coverage, access to formal financial markets and access to the contract system, originate from compliance with laws and regulations. Business formality can be viewed as a necessary input to production, which firms decide how much to purchase along a continuum Rise in Labor Informalitv. * Rise in Labor Informality. The high bureaucratic cost of labor regulations may explain, among other factors, increasing labor informality. Argentina, for being a middle-income country, has one of the lowest compliance index on social security registration, not exceeding 50 percent. During 1980-1998 the share of informal salaried workers increased for all firm sizes, especially among smaller firms. This obviously has serious implications for the costs and coverage of social programs, especially health insurance, pensions, and income support during bouts of unemployment. * Reduction of Business Innovation. Although governments have good reasons to regulate economic activities, the Argentine cumbersome business regulation system imposes a constraint to business innovations, especially among businesses in the manufacturing and commerce sectors. For example, a cafe shop (negocio de gastronomia), which is an activity typically dominated by small businesses, will require municipal authorization to offer espresso coffee services, as well as approval of the location of the coffee machine and auxiliary units. * Disincentives for Geographical Expansions. To sell products or services and to buy supplies from businesses in other provinces, a business must register with the Convenio Multilateral (Multilateral Agreement). The multilateral agreement is a mechanism for transferring the corresponding part of Gross Income Tax to other provinces when a company generates income from a sale or makes payment for a purchase of goods in a province other than the one in which it is located. * Uncertainties about Compliance and Enforcement. An excessive amount of red tape creates losses for society and is highly susceptible to the discretion of government authorities and corruption while leaving plenty of room for errors and creating judicial insecurity. For businesses with operations in different municipalities, the three-tiered system of business formalities adds additional uncertainties during municipal inspections because definition criteria of economic activity vary widely across municipalities. Labor registration inspections are very ineffective due to lack of good information. * Inefficient Use of Government and Firm Resources. Argentina's cumbersome labor registration system imposes substantial costs on businesses. According to recent estimations, businesses used to spend about $700 million per year, representing 1.5 percent of overall salary costs (Giordano et al., 1999). For government agencies, the labor registration system entails duplication of activities and collection of information, and thus expenses. * Reduced Productivity and Increased Unit Costs. The increase transaction and operating costs due to that unfriendly business environment, with a large number of both fixed and variable costs-rule driven- adversely impact productivity and unit costs to all firms but particularly to SMEs * Reduced Number of New Economic Activities. The increased cost hinders the formation of new fiums and reduces investment expenditures of existing firms with the obvious adverse impact on employment and income generation. xviii 5. Significant Problems with Access to Finance Comparing SME's contribution to output with distribution of bank loans, we find severe imbalances and a need for policy action to deepen financial markets for SMEs. Small and medium enterprises receive, on average, about 9.8 percent of bank credit during 1998-2000, but they contribute to about 61 percent of Argentina's value added and to 69-78 percent of employment. Other countries present similar findings, such as Peru where SMEs receive 15.1 percent of bank loans, while generating 42 percent of Peru's GDP. The need for both short and long-term policy action in the fmancial markets has become particularly urgent in Argentina nowadays because the current collapse of the banking system is harming economic activity as a whole but is especially damaging for SMEs, particularly in terms of access to finance. Participation in Credit Markets Contrary to the widespread perception that SMEs lack access to finance, a recent survey showed that until the triggering of the current financial crisis a substantial share of SMEs borrowed money. Firms paying interest -our measure of access to finance- hovered around 58 percent in 1998 and 1999 among SMEs in the city of Buenos Aires. The manufacturing sector presented the highest rate of access to finance while the service sector presented the lowest rate. As expected, access to fmance was positively related to firm size. Firms with more than 50 employees not only were more likely to have access to finance, but also their access improved between 1998 and 1999. Smaller SMEs, with 6 to 50 employees, had lower rates of access and their access worsened in recent years. Bank Lending to Small and Medium Enterprises What are the Determinants of SME Lending by Banks? SME lending was never a main line of business for Argentina's banks. According to a recent Central Bank study, overall, SME lending accounted for around 20.9 percent of banks' loan portfolio in 2000, a decline from 23.6 percent in 1998, partly explained by the economic recession. Although the volume of lending to this sector declined 4.5 percent during 1998-2000, it did so at a lower rate than loans to large firms (5.8 percent). Lastly, loans to the non-financial public sector grew about 125 percent over the same period. Because a decline in the share of small business lending could arise from small business loans increasing or decreasing at a slightly slower rate than all types of lending, Clarke and others also analyzed the impact of foreign bank entry in the real growth rate of small business lending. They found that small business lending by domestically owned banks appeared to be growing faster than lending by foreign-owned banks in Argentina. Moreover, the authors found that bank size was positively correlated with growth of small business lending with a large effect among foreign banks. This indicates that small business lending was growing faster for large foreign banks than for small foreign banks. Lastly, one of the most interesting findings of the authors was that Argentine public banks did not come to the rescue of this sector and did not appear to surpass private banks in the extent to which they lent to small businesses. Thus, arguments saying that eliminating government resources invested in public banks or privatizing public banks would hurt small business lending do not have strong empirical evidence. What Explains the High Interest Rates of SME Loans? Argentine SMEs cited the high level of interest rates of bank loans as an important constraint. Several factors may explain the high levels of interest rates paid by SMEs: Credit Quality. Differences in credit quality may explain why SMEs paid higher interest rates on bank loans. According to the BCRA study, SME loans presented substantial higher levels of non-performing loans than large firm loans. Type of Products. When SMEs borrow from banks it is via credit cards and overdraft current accounts that carry interest rates much higher than rates of other types of loans, ceteris paribus (Sepyme's web xix page). In 1999, about 42 percent of SMEs with access to the banking system were using overdrafts in their current accounts (decubiertos). Moreover, the UIA's survey of 2000 found that about 50 percent of industrial SMEs were using overdrafts, paying on average an interest rate of 2.7 percent per month or 38 percent annually when prices were stable or even showed a declining trend. Administrative Costs. The transaction cost of lending to SMEs is high in relation to the size of loans, making such lending in many cases unprofitable. The lack of documentation available makes it difficult to obtain a clear picture on the financial situation and prospective success of such businesses. Most banks were using lending technologies based on standardized credit policies and standard information. Informational distance contributes to increase the costs of generating borrower-specific information. Collateral. Collateral may reduce the cost of intermediation. Banks may find it easier to assess collateral value than to evaluate project creditworthiness and in case of default, expected losses may be lower. Various studies have found that SMEs' loans presented a higher fraction secured by collateral. This seems to indicate that the legal framework for secured transactions present serious shortcomings. Using a database of business debtors as of October 2000, a team of researchers from CEMA and the BCRA assessed the determinants of the interest rate paid by large firms on their bank overdraft account, which averaged 3.1 percent monthly. After controlling for various explanatory variables, the econometric results show that firms that paid lower interest rates were larger, in terms of assets, had better ratio of sales over assets, and had a lower leverage ratio. Moreover, firms with poor credit quality and with loans with lower guarantees paid higher interest rates. Under-developed Financial Infrastructure Argentina has an under-developed infrastructure to support financial transactions that limits the variety of credit products available for SMEs. * As shown in Chapter 5, inadequate laws and regulatory regimes have had the unintended consequence of detracting risk capital investment, especially for SMEs. These shortcomings increase transaction costs for evaluating, structuring, managing, and exiting equity investment operations. Because investment size and firm size are related, these higher transaction costs have a disproportionately negative effect upon investments in SMEs. * Information problems are particularly acute for SMEs. Many of the smallest firms do not have audited financial statements that can be used by banks. Moreover, they maintain parallel accounting books to circumvent government regulations, rendering audited financial statements useless to assess creditworthiness. For some banks, knowledge about a client is more important for lending decisions than cash flow analysis from financial statements. Hence, lack of detailed micro data on small businesses is likely a major reason why banks have not developed large portfolios, as SMEs often cannot credibly convey their risk quality to banks without offering collateral. Although the Argentine Government has taken several steps in reforming the environment for secured transactions, partly improving some areas, many problems remain, suggesting that the Government may consider a full reform of its system, instead of piecemeal reforms. Registries are inadequate for registering security interests and for searching existing encumbrances. Each geographical area has its own pledge registry, requiring re-registration of goods that move between jurisdictions or against debtors that have multiple business locations. Registries operate manually and data can only be transmitted in person because fax or mail is not allowed. The execution of guarantees of movable goods is lengthy and cumbersome and the associated court fees are high, reaching between 4 to 6 percent of the total claim depending on the province. xx 6. Weak Demand for Knowledge SMEs face significant sources of knowledge constraints, examining the areas of marketing, managerial skills and technical capacities. There is an important difference between finance and business environment on one side and knowledge on the other as factors of SME growth. Special nature of knowledge - marketing, managerial and technical capabilities of SMEs - the fact the knowledge could be readily available but would still require substantial effort on the part of the firm to assimilate it, puts much greater weight on demand factors rather than supply factors. Whereas any SME would demand good business environment and credit which do not require sophisticated capabilities to consume and assimilate (any one can spend money or benefit from simpler regulations), advice from consultants is only as good as the firm's ability to assimilate and act on it. Consequently, weak SMEs do not even understand the nature of their problems. They tend to list finance constraint as the major problem whereas in most cases it is conspicuous lack of marketing, managerial, and technical capabilities. For example, a recent survey of SMEs finds that only 5.1 percent of the firms consider that ties with other organizations, such as firms and SME institutions, are important for their growth. One of the few systematic studies of non-financial support programs for SMEs in Argentina estimated that approximately US$222 million dollars was being spent annually by national, provincial, and municipal governments and non-government organizations for such programs. This figure was equal to about 0.08 percent of the GDP and 0.26 percent of total public sector spending of Argentina and translated into about US$2,800 per SME that was using some support program. However, if the support programs would expand their coverage the amount for each SME could drop to US$280 in the unlikely case that all SMEs are covered. Moreover, the prospect of a sharp decrease in the public resources that will be available in the future, given the current macroeconomic and fiscal conditions, adds extra concem on the continuity of those programs. A small number of SMEs use non-financial programs. Despite the growth in the 1990s of technical assistance and training programs in Argentina, the existing evidence reveals that SMEs have low access to or low motivation for such services. For example, the FUNDES- UNGS study shows that only 16 percent of SMEs have access to training and consulting services. It also shows that a staggering 78 percent of firms surveyed used no instrument of SME support (e.g., training, technical assistance, credit, export promotion) during 1996-1998, even when the Argentine economy was rebounding in 1996-97. Low-level Equilibrium Trap as a Result of the Knowledge Constraint Conspicuously low demand for knowledge sets the stage for a low level equilibrium trap - a vicious circle of mutually reinforcing lack of basic understanding of where the problems are, erroneous actions following the wrong diagnosis and resulting poor performance. Poor designed SME programs can easily reinforce this vicious circle and be strongly counter-productive if, for instance, they pump credit resources to firms which main limitation is on managerial and marketing side. Weak knowledge capabilities imply that the credit would be poorly utilized which would put both the SME in question and the bank in jeopardy. 7. Current Government Policies: Multiplicity of Programs, Uncertain Impact Argentina has a myriad of specially designed programs to offer training and technical assistance as well as financial support to small and medium-sized enterprises. According to the Secretary of Small and Medium-Sized Enterprises (Secretarfa de Pequefias y Medianas Empresas, SEPyME, created in 1997 under the Presidency but moved in 2000 to the Ministry of Economy, and again in 2002 to the Ministry of Production) Argentina has approximately 300 programs and lines of credit to support SMEs. xxi Business Deregulation During the past decade Argentina has experienced a radical change in the organization of its economic institutions. From a vastly over-regulated and state led model, which has driven the country to 25 years of economic decadence, the 1990s witnessed a structural reform emphasizing modernization through deregulation and market oriented policies. In spite of the deep ongoing crisis, the Argentine government still seems to keep interest in fostering private sector competitiveness by reducing the costs of business regulations. Recent efforts were focused on deepening the deregulation process initiated in the 1990s. The main activities in this regard are: competitive plans, one-stop-windows, and simplification of tax and labor regulations. Business deregulation is a complex process that requires and ongoing rather than a "one-shot" set of "massive deregulation" measures. Modern societies are involved in a permanent regulation - deregulation process. Financial Markets Throughout the world, governments have intervened extensively in financial markets. Argentina has not been immune to this ipproach. The Argentine government has a long history of using directed credit and subsidized lending to increase SMEs' access to credit services. By designing credit lines with subsidized interest rates, the Argentine Government tended to assume that banks face liquidity shortages and that SMEs cannot support high funding costs. Overall, in Argentina, as well as in other countries, these interventions have not been effective in lowering the total cost of credit for borrowers and have not resolved problems of asymmetric information, partly explaining their low penetration. It is important to point out that public credit guarantee programs, in and of themselves, do little to encourage banks to learn about SMEs and build relationships with them, creating another layer of transactions. Nevertheless, it is also worth to notice that private credit guarantee institutions (the so-called Sociedades de Garantia Recfproca) are relatively new in Argentina (and their legal framework was modified during 2000), and that during a four year recession it is difficult to evaluate its impact. Business Development Services In the 1990s, the number of business development programs for SMEs grew substantially. These programs offer services, such as training and technical assistance, human resource management, technologies, and export capacities. Overall, these government programs are characterized by the lack of strategic vision, coverage, coherence, and follow-up, raising questions about the elements of success and the cost-effectiveness of those programs. The existing evidence reveals that SMEs have low access to or low motivation for such services. For example, the Fundes study shows that only 38 percent of SMEs received consulting services in 1997- 1998. Public institutions (including municipalities) provided only 16 percent of all consulting services. Also, about 62 percent of SMEs received training services. More importandy, the Fundes study also shows that 30 percent of firms surveyed used no instrument of BDS support (e.g., training, technical assistance, export promotion), which adds extra concern about their growth potential. xxii 8. The Path Ahead: Towards Incentive-Driven SME Policies Policy recommendations are urgently needed to reform current government efforts to enable the business environment, increase the availability of financial services while reducing the cost of credit, and promote a market for business development services. Analytical considerations suggest the following reasons for government interventions: * Addressing the market failures that create cost disadvantages for SMEs, restrict their access to markets, and inhibit the development of markets for a diverse range of financial and non-financial services appropriate for small firns. This reason is stressed by the current collapse of the banking system; * Improving transactional efficiency in financial, product, and input markets relevant to SMEs, by facilitating access to information and developing mechanisms to manage risk; * Reconsidering public policies and regulations that discriminate against small fins or produce fixed costs that create a competitive disadvantage for them; and * Investing in public goods that open market access and build enterprise competitiveness- including infrastructure (information, communications, power, water, and transport) as well as education, technology development. Elements of an Effective SMEs Assistance Program A large number of SME interventions have failed, both in Argentina and worldwide. Few have succeeded. An efficient and efficacious SME program is even more rare. This crucial observation brings us to central policy issues: the identification of factors of success of efficient and efficacious SME initiatives and their adoption in the design a successful SME programs and related institutions in Argentina. Leaving momentarily aside the particular short-term requirements posed to SME policies by the ongoing economic crisis, evaluation studies in Latin America and Argentina allow discerning the following factors of success of good SME programs and institutions: * Development of a standard set of metrics to measure performance of SMEs and SME programs. For example SME growth and productivity rates, repayment rates, start-up survival rates, export growth, employment levels, linkages levels or index etc. This is essential to monitor effectiveness and adjust assistance and design. * Critical importance of entrepreneurial management. Successful organizations supporting micro, small and medium-sized enterprises (MSME) are often initiated by social entrepreneurs, individuals with unusual problem-solving and management skills and motivation for MSME development. In the examples of Foundation Chile and ProChile, success of MSME agency is predicated on such entrepreneurial manager at the top of the organization. Successful MSME organizations tend to evolve from a reliance on top key individual to a robust organization with efficient corporate governance. * Cost recovery. Successful MSME organization should target for growing cost recovery. For example, if most MSMEs learn mostly from each other and that by demonstrating that MSMEs can improve their business using business development services, some MSME organizations have been able to secure a stable demand for new business development services. Matching grant subsidies to accelerate the demonstration effect can be justified until such demand develops. xxiii * Extensive use offollow-up techniques (benchmarking). Successful programs helps-through benchmarking indicators- diagnose where firms are, what do they need to do to improve and alternatives for those unlikely to survive. * Facilitate and reward linkages within the supply chain. Successful program should inform and facilitate opportunities to link with relevant supply chains and provide coordination assistance * Leverage their effectiveness with use of ICT. Develop SME internet portals, internet immersion institutes, improve access to the internet; provide access to ICT instruments, train firms accordingly and inform about the benefits. A number of these initiatives should be developing at the state level and coordinated at the federal level, with the public sector as a model user, for example through e- government. * Participation of clients in the design of the programs. Clients not only need to pay for the services; they need to participate in the design and evaluation of programs. To assure that they do, SME programs shall never be run by the governments (whether federal or sub-national), rather by an autonomous private management contractor working in cooperation with the government but independently from it. * Incentive driven components and private sector participation. Successful programs are incentive driven and managed by the private sector. For example one of the most widely used are limited time matching grants to first movers. First movers that develop new ways of doing things, such as breaking into non-traditional export markets, are critical agents for economic development. Because SMEs learn most productively from each other, the success of a first mover provides powerful examples to follow and thus has positive spillover effects. Although some public subsidy to first-movers is justifiable because of the spillover effect, the main question for governments is how to deliver such subsidy. To avoid capture and resulting misallocation of resources, government shall not be involved directly in delivering such subsidy. Rather, a firm-level management governance scheme has proved to be a promising solution in various countries. This scheme included the creation of a temporary public fund managed by a private management service provider. To offset certain fixed start-up costs of breaking into export markets the government cover costs of knowledge (consultants' costs, travel to trade fairs), usually on a 50/50 basis. Once eligibility criteria are met, funds are provided on a "first come, first served" basis. The fund is temporary with a typical duration of 3 to 4 years because once exporting ceased to be innovative activity, thereby not leaving any learning spillovers, there is no reason for the public fund to exist. The private management service provider administering the fund operates under a time-bound contract with performance benchmarks and acts as global network information broker (by providing relevant information to firms), markets the scheme to the firms, provides up-front, free support in preparing firm's programs and acts as an impartial administrator. The relevant knowledge (about export markets and on the upgrading of the firm) is highly specific and can be provided only by independent service providers. Nevertheless, experiences like PREX (financed by WB) or PRE (financed by IDB) programs show shortcomings of this approach. The private sub-contractor often does not have a full understanding of local SMEs problems. Additionally, when the program finishes, the global network information broker role goes away together with the private operator. * Facilitating collective learning of SMEs. The government (or central authority in general) orchestrates decentralized learning by monitoring and assuring information sharing between local experiments. Central authority does not direct--it facilitates information sharing, encouraging improvisation, and experimentation. Information exchange through continuous networked interaction) is the underlying principle of the decentralized learning in many areas of SME performance. Box 7.3 presents an example. xxiv * Performance orientation. The successful programs are loaded and driven by incentives and induce self selection among SME firms, helping those that help themselves, illustrating how they can and should help themselves and suggesting exit to those that are not capable for internal or external factors to improve themselves. * Gradualism and Assistance Linked to Performance. Not all SMEs are created equal, as we have seen. Successful programs link assistance with performance and gradual assistance as SMEs improve themselves. A touch of realism assisting SMEs to identify their possibilities and potential is essential, followed with assistance on exit for those not viable. This approach contrasts sharply with traditional SME promotion strategies, which rely heavily on the direct and subsidized provision of financial and non-financial services to SMEs. It places much greater emphasis on creating an enabling environment for SME competitiveness, and on developing markets for SME-relevant services rather than substituting for them. It attempts to broaden the coverage and impact of government programs by using the private sector to deliver services, and utilize scarce public resources for facilitating market transactions and investing in public goods. Policy Chalenge The challenge is to initiate institutional development at the local and national level, which encourages transition from inward-looking firms, narrow search routines, and information-poor markets to learning-oriented firms and mature, information-rich markets. Securing an appropriate incentive framework involves amelioration of both government failure (deregulation and economic liberalization) and market failure. A strategy to improve the incentive framework for SMEs should be based on three blocks. The first block is the so-called first generation reform of elimination of government failure - deregulation and economic liberalization and better regulatory environment for businesses and investors. The second block should address the issues relating to SMEs' access to finance and to improve the now under- developed financial structure, including the new constraints posed by the current disruption in the functioning of the banking system. The third block -- second-generation agenda -- encompasses a set of initiatives of horizontal interventions to ameliorate market failure and accelerate positive changes in business' behavior. The latter includes schemes such as matching grants, business incubators, private-public seed funds, and facilitation of private-to-private collaboration through business associations and information centers. xxv Table 4. Summary of Market-Oriented SME Interventions Policy agenda U - Building Institutional Capacity and Measuring the Delvery of SME Policy agenda - Benchmarks and Poli agenda I -Deregulation Investment in Public Goods Performance Indicators Business Environrnent Business Registration * Evaluate whether it is worthwhile to expand the pilot Monitor progress on simplification of business registration in aUl provinces by * Streamline business registration procedures by reducing ject initiated in May 2001 by SEPyME surveying firms on the investment climate. Items to monitor are: licensing and registration requirements and administrative Take advantage of information and communication * Business registration process - cost (for all areas, cost includes in time fnologies to facilitate interaction between government and and fonmal and informal types of payments and contributions, including Eliminate labor registration requirements beyond the MEs bribes), procedures required, delays ocal security one Institute systematic monitoring of SME performance and * Business licensing - number and type required, cost, time and payments Agree on cost-benefit analysis to eliminate the of policies and interventions required gulations that accrue little or no net benefit for the economy a Improve access to information about markets, standards, a Acquisition of business premises - fonnalities, costs, constraints, and Coordinate business registration between national and and technologies) delays ocal authorities * Evaluate the experience of "competitive plans" to see * Import, export and customs regulations - formalities, costs, and delays Disseminate information about registration whether deregulation agenda can work in that context * Product, input,.and equipment certification - types, incidences, equirements and procedures. Create process of innovation formalties, costs, and delays where firms' feedback is incorporated to simplify business Tax administration - requirements, constraints, costs, number of taxes registration and forms * Business inspections - types (and agency responsible), costs, number, Labor regulation reform: and process followed Review labor taxes to identify which taxes do not bring * Quality of infrastructure and services benefit to the government, firms, and the labor force * Labor relations * Aow SMEs firms to sign decentralized-at the firm Conflict resolution and legal environment evel, collective bargaining agreements Number of collective agreements signed at the SME firm level Reform the severance payment regime to reduce its * Amount reduced of tax wedge ts and.make it more transparent and less prone to litigation Consider substituting the severance payment regime for vidually capitalized accounts Tax reform: Simplify the tax regime and compliance procedures Improve comiercial ransactions law xxvi - | Policy agenda 1 - Building Institutional Capacity and Measuring the Delivery of SME Policy agenda - Benchmarks and Policy agenda I -Deregulation Investment in Public Goods Performance Indicators Financial Markets Improve legal environment for secured transactions. * Creation of a registry for lien on movable property allowed Monitor number of security interest registered in the registry for liens on With WB assistance the GOA drafted a secured transaction nder a new secured transaction law ovable property law within the SSAL that could be submitted to Congress Promote innovation in loan products, lending Reduction of direct lending through public financial institutions Eliminate shortcomings of the legal framework for risk methodologies, delivery mechanisms, and risk assessment Reduction in volume of subsidized credit lines pital investment to promote equity investments in SMEs. nethodologies Share of SME lending and growth rate of SME loans Promote and allow information sharing of a'l positive Develop a credit-scoring tool for the SME market in Quality of SME loan portfolio d negative credit information histories from all types gentina using information available in all commercial banks sactions to eliminate fragmentation of credit histories. Build institutional capacity in private banks by Improve efficiency of database of social security ssemiating best-practice on SME lending by commercial detors banks Improve legislation and monitoring of reciprocal Reduce public sector involvenent in SME lending arantee societies Prepare a phase-out program of subsidized credit lines and nterest rates subsidies. Redesign interest rate subsidies programs o avoid perverse incentives and guarantee additionality of funds Rethink govemment participation in guarantee funds by owing greater involvement of private sector Business Development Services (BDS) I Target subsidies for market development to specific * Facilitate innovation in delivery mechanisms Cost recovery indicators for publicly-provided or subsidized services narket failures a Develop performance and impact indicators Coverage (number of SME clients), in particular outside major urban * Structure assistance on a gradual mode as SMEs * Improve training and technical assistance to private BDS espond to incentive structure providers Decentralization (extent of participation of municipal and provincial Facilitate poor perfonner exit mechanisms a Llrnit long-term subsidies for BDS to public goods (e.g., overments in BDS programs financing and design) information, labor and management training) Reliance on private provision of services (rather than in-house service Improve management and cost control in public BDS livery) institutions' Characteristics of program management (extent to which remuneration * Accelerate access, usage of ICT for SMEs f management or BDS services is tied to the performance of the programs) a Condition budgetary allocations to the achievement of impact a Reduce duplication across agencies in services provided * Decentralize service provision to provincial level a Introduce pilot initiatives combining deregulation and provision of business development services at a provincial level xxvii 1. Introduction In Argentina, convertibility law and economic liberalization brought about dramatic changes in economic performance. At least before the current macroeconomic problems, the structural reforms proved reasonably successful in attracting foreign direct investment and generating growth of large domestic conglomerates and multinationals. But even before the advent of the current difficulties, this growth was very unevenly distributed between regions, sectors, and economic agents. The productivity gap between small and medium-sized enterprises and large firms remains large signaling large unrealized opportunity of SMEs as a potential engine of growth and employment. What are small and medium-sized enterprises? The most frequently used variables to define SMEs are number of employees, sales volume, financial assets, and productive assets. According to the Economy Ministry the legal definition of SMEs in Argentina requires calculating an index number that summarizes number of employees, annual sales and productive assets in reference to pre-determined maximum values.' Due to the complexity of the official formula, and for consistency with recent studies that did not use the recent redefinition of SME, this report considers as SMEs independent firms employing fewer than 200 employees. Small-enterprises have generally less than 50 employees, while micro-enterprises have at most 10, or in some cases less than five employees. 1.1 Objectives The purpose of this report is to assess the constraints affecting Argentine SMEs -creation, development, and performance- and identify appropriate and justifiable policies and institutions that can help improve their performance, and contribute to economic growth and job creation. Argentine SMEs face constraints from many sources. Although this report cannot examine all of them, it focuses and analyzes the three key groups of constraints that emerge as a consensus among SMEs surveyed in a recent study:2 (a) inadequacies of the business environment; (b) access to finance and high cost of borrowing; and (c) knowledge, information and technology constraints. While clearly SME performance, and that of any other type of firm is highly impacted by the macroeconomic environment, and that has been quite an adverse factor in Argentina in the last few years, this report tries to stay away from the impact of such environment on SMEs. Most of the focus is in the impact of those three types of generic constraints and through the analysis of SMEs heterogeneous performance in the imputing of the determinants of success and proper assistance programs. 'The firm-specific index number equals P, = 310 o' * 10- 10 iA , where PO = number of employees, V OR VAR APR VA=annual sales and AP = productive assets. The sub-index R refers to the maximum value for the corresponding variable. Microenterprises have an index below 0.3, small enterprises between 0.3 and 1.5, and medium-sized enterprises between 1.5 and 10. 2 Obviously, growth constraints arise from tax and regulatory burdens, high input prices, financing problems, lack of competitiveness, low productivity, labor market rigidities, and an inadequate business environment (with some of them being much more severe than in other countries, i.e. labor market, regulations, registry, taxes). A recent study carried out by Fundes, shows the lack of consensus about the ranking of the main problems facing SMEs attributed to the heterogeneity of the sector, and the plethora of important problems, but nevertheless, most business problems can be grouped into three areas: (a) access to finance; (b) business environment and barriers to formalization and transaction costs; and (c) knowledge constraints. Obviously, an over-valued exchange rate has adversely impact the competitiveness of Argentine firms. 1 The central policy question of this report are how to address those constraints; how to correct or ameliorate the effects of market and government failures; and more important how to eliminate unnecessary regulations; what kind of support is appropriate and how best to design SMEs assistance programs. Overall, the report policy interventions in the business environment and in the financial sector and proposes the facilitation of efficient institutions of public and/or private service delivery that as a public good benefit all firms, but particularly SMEs. This report shows that such institutions are better created at the state and regional level, then scaled to the national level. This second-generation institutional agenda is at the center of a sustainable and market-driven incentive framework for SME development. 1.2 Data and Methodology The analysis of this report relies a combination of quantitative analysis and literature review, which together permits to consolidate the current knowledge about the three growth-constraints addressed in this report. The report uses a number of databases: * Argentine bank level database. The central bank of Argentina provided bank level information on balance sheets and income statements for all the commercial banks and the distribution of bank lending by size for the period 1998-2000. Total SME lending is approximated by adding to components: (a) loans between US$200,000 and US$1,000,000 from any individual bank; and (b) all commercial loans under $200,000 that are directed towards SMEs.4 This database is used to evaluate the impact of bank ownership and foreign entry on SME lending. * Firm survey of the City of Buenos Aires (INDEC- 1999). To estimate the geographic domestic product the statistical agency of the city of Buenos Aires carries out a comprehensive survey of firms operating in commerce, service, and industry sectors. This survey includes a limited number of variables on the composition of revenues, expenses, and employment as well as other information about the firm. In the case of industrial firms, the questionnaire contains a wider range of information about investment behavior and exports. * Survey of SME exporters of FUNDES. To impute the determinant of success for exporting SMEs, FUNDES undertook an extensive survey of exporting manufacturing SMEs. This is s very rich survey at the firm level. * Firm survey of FIEL. Based on sample form the Observatory Of Industrial SMEs of the Industrial Union of Argentina. Most of those firms are relatively successful, so it is not a random sample. The purpose of the study was to impute the dynamic patterns of industrial SMEs. The paucity of information about SMEs imposes four main limitations to the analysis. First, the overall characterization of SMEs relies on the 1993 economic census data since most recent data do not exist for Argentina. A more recent picture of SMEs is circumscribed to the city of Buenos Aires. Second, lacking panel data on SMEs it is difficult to characterize firm dynamics and evaluate thoroughly the determinants of success and failure. Third, although the data do not allow an empirical analysis of the relative impact of the three main constraints examined in this report, it provides an impressionistic view of the magnitude of the problems. Finally, the data do not permit a complete evaluation of government programs. The report takes advantage of various on-going research related to SME performance. The Inter- American Development Bank organized a regional research on the determinants and consequences of 4 To promote lending to small businesses, the Argentine Central Bank allows banks to treat commercial loans less than $200,000 as consumption loans for purposes of information and provisioning requirements, simplifying the approval process for banks. 2 financial constraints of firms that includes two studies about Argentina: (a) an assessment of relationship lending and the determinants of interest rates among a sample of medium and large firms carried out by a team of researchers from CEMA and the Central Bank of Argentina; and (b) the determinants of industrial SME receiving loans carried out by CEDES. 1.3 Main Findings This report presents seven main messages: i) The SME sector is a critical sector of the Argentine economy both from growth/efficiency and equity standpoint yet it failed to realize its potential. SMEs account for 78 percent of employment and 61 percent of production. Small and medium-sized enterprises have not realized their potential, partially because of their own deficiencies but also because of government failures and the standard market failures affecting economic activity. Hence the current predicament of SMEs is also an opportunity. Once onerous regulatory burden is relieved, one can expect SMEs to become an engine of economic growth and employment. The current crisis, and the need to jump-start growth is an important window of opportunity to reduce regulatory burden and start the reform of SME policies and programs. ii) Small and medium-sized enterprises are highly heterogeneous and are-not just smaller- sized replicas of large firms. SMEs have special organizational and technological characteristics, which would have been irrelevant in the world of perfect markets and perfect government. It is market and/or government failure, which makes SMEs distinct: they have much lower adaptive capacity to deal with market and government failures, and they are the most cost impacted by burdening regulations, lack of markets deficient business environment; not all of those differences, demand or justify public sector interventions. iii) Argentina's business environment suffers from a high degree of institutional rigidities, which partly explains why roughly one quarter of economic activity is performed outside laws and regulations and only half of the labor force is registered with the social security. Registration and creation of a legal entity is a time consuming process for SMEs, with official costs of compliance ranging from $1,755 to $6,500 depending on business size. Overall, the regulatory burden of the business registration system is very high, falling more heavily on small firms and giving large firms a comparative advantage. iv) There is a need for policy action to deepen financial markets for SMEs. Small and medium enterprises receive, on average, about 9.8 percent of bank credit during 1998-2000, but they contribute to about 61 percent of Argentina's value added. Other countries present similar findings, such as Peru, where SMEs receive 15.1 percent of bank's loans, while generating 42 percent of the country's GDP. Nonetheless, access to finance per se does not seem to be a big problem, about 58 percent of SMEs in the City of Buenos Aires had'access to finance in 1999 and 1998. The problem lies rather in the conditions under which SMEs receive credit. Overall most SMEs have access to very expensive short-term loans in the form of overdrafts from current accounts, which may not be the most effective form to finance working capital or investment needs. v) Knowledge constraints are significant. In Argentina, most SMEs do not use training and technical assistance because services available are directed towards large firms and are too expensive for SMEs. Overall, about 38 percent of SMEs purchased technical assistance services and 62 percent received training courses during 1997-1998, leaving 30 percent without such services. Although the data to date is scarce, small and medium enterprises 3 appear on average to suffer from isolation. The special nature of knowledge of markets, technologies, and supplies, plus the fact that such knowledge would still require substantial effort on the part of the firm to assimilate it, put greater weight on demand factors for information rather than on supply factors. This ability to incorporate lessons and new ideas is a key factor to remain competitive in the new economic environment. vi) In Argentina, there is a striking multiplicity of SME programs, yet with uncertain impact. The type of public intervention is important. It is not a so much a matter of additionality of resources to the sector (as the amount spent for the sector in Argentina is ample) as the issue of the focus targeting, and particularly the incentive structure and the delivery of those services. vii) Argentina needs to overhaul its SME policies and programs to make them more incentive- and demand-driven. Not all SMEs are created equal, and they respond differently. Assistance programs for them to be effective have to be discriminatory and responsive to performance. Those programs have to induce self-selection, be incentive driven, and adapted to the capabilities of SMEs, and the assistance gradual, matched to the frms' response. 1.4 Organization of the Report This report is organized in eight chapters. Chapter 2 states that small and medium-sized enterprises play a much more important role in economic growth than had been acknowledged previously. In Argentina, SMEs are the single most important source of employment. Chapter 3 presents the highly uneven supply responses to economic liberalization of SMEs, highlighting how diverging firm-level trajectories affects inter-industry and intra-industry heterogeneity. The following chapters analyze business constraints arising three sources. Chapter 4 describes the main features of the business environment in Argentina. It describes business registration procedures, finding that the regulatory burden of the business registration system is very high, falling more heavily on small firms, and giving large firms a comparative advantage. The subjects of lack of access to finance and bank lending to SMEs are considered in some detail Chapter 5. Chapter 6 moves to explains the sources of knowledge constraints, examining the areas of marketing, managerial skills and technical capacities. Chapter 7 analyzes current government policies to address the three main constraints analyzed in previous chapters. Lastly, Chapter 8 outlines policy recommendations to revise current government efforts to enable the business environment, increase the availability of financial services while reducing the cost of credit, and promote a market for business development services. 4 2. SMEs as Potential Engines to Economic Growth and Employment: Anatomy of the Sector Any assistance program with (the limited) government resources to assist exclusively SME, requires a motivation orjustification. The usual arguments ought to be based on efficiency grounds-market or government failures types-or on equity and poverty grounds. Often both types of arguments are used. To make that argument for SMEs, it is useful to analyze their similarities and their differences with the other firms. This chapter presents a set of characteristics by furm size and a bit of an anatomy of the Argentinean SME sector, and argues that indeed there are significant differences among firms of different sizes and a number of those differences might be linked to market imperfections and transaction costs. The economic importance of small and medium-sized enterprises (SMEs) in Argentina is also reviewed here. 2.1 The Role of SMEs in Argentine Economy Small and medium-sized enterprises play an important role in economic growth. They are the single most important source of employment in many countries. As shown in Table 2. 1, Argentine SMEs account for about 69-79 percent of employment. For some sectors, their share of employment is greater than their share of production, indicating lower levels of productivity. In terms of dynamics, both employment and number of establishments in Table 2.1: Argentina's SMEs and International manufacturing SMEs has decreased Comparison... NuCber mNetpannualiAvg.sNo. % of % of quite significantly in the Country (000's) Nrt annual Wo;kers eo n %oof of| manufacturing sector, but increased USA 6 2,1% 11 54% 31% in the service sector in inter-census EU-12 18,500 2,0% 7 72% 15% period 1984 -94. Overall, the MERCOSUR 5,700 2,9% 10 620% 12% numbers remain overwhelming. Chile 40 4,0% 1 0 80%/ 1 9%/ Argentina a/ 1,000 2,2% 10 69 - 79% 10% Given those numbers and the Uruguay 170 3 7% 67% 18% standard elasticities, productivity Brazil 4,050 3,0% I1 5 90/o 9Y. increases in SMEs would have a TOTAL 3020 2,2% 8 164% 1 90% Source: IAMC (I 999, p. vii) using data from Eurostat (UE); SBA (USA); CORFO major effect on economic activity (Chile); IMM (Uruguay) and IGBE (Brazil). and employment. That is the a/ net rate obtained by extrapolation between the 1994 and 1998 estimates. challenge. The geographical distribution of SMEs reflects the relative concentration of the population and economic activity in Argentina. The province of Buenos Aires, the Federal Capital, and the province of Santa Fe hold about two-thirds of all SMEs, while presenting a slightly higher share ofjobs created by SMEs. An interesting finding of the 1994 economic census is that the firm size distribution varies by geographical regions, with microenterprises being more active in less populated areas than SMEs and large firms. These differences in the geographical concentration of firms of all sizes reflect wide disparities in terms of business opportunities and linkages between large firms and SMEs and between SMEs and microenteprises. Overall, there are 29 SMEs per large firm in the country, compared to about 303 microenterprises per large firm. Another finding is that the workforce of SMEs also varies by geographical regions. Misiones and the Federal Capital SMEs have the largest contribution in terms of number of jobs. The SME sector is characterized, on average, by a lack of dynamism and low productivity, (mainly in the service sector), and low rates of entry and growth. Although superficially some of the SME sub-sectors, such as manufacturing, look healthy, a deeper examination reveals that SMEs face major problems, including difficulties to start-up and to grow. Most SMEs in Argentina are still struggling to adapt to competition in the global marketplace ushered by trade liberalization and the reduced state role in the economy. Despite their continued economic importance, a limited number of new SMEs enter the 5 marketplace and many more struggle to survive or die of attrition. Though the data is scarce, recent estimates by the Central Bank suggest that mortality rates reached 18 percent, about twice the birth rate during the 1990s. In many cases, SMEs focus on survival in the short-term rather than on repositioning for domestic, regional, and international competition. Hence, they are in a precarious position to integrate their activities into the global economy. Most SMEs are relatively older than micro and large firms, especially in the manufacturing sector. According to the 1999 INDEC survey of firms in the City of Buenos Aires, 42 percent of SMEs have been in business for at least 15 years, compared to 39 percent of large firms and to 30 percent of microenterprises (less than 5 workers). Furthermore, manufacturing SMEs are much older than SMEs in other economic sectors. Both the 1999 INDEC and the Uni6n Industrial Argentina (UIA) surveys illustrate this, with more than 40 percent of them operating for more than 30 years. There are indeed sectoral differences on SME performance. SMEs operate in a wide range of economic sectors, but their sectoral distribution presents substantial differences with respect to large and micro firms. For example, most SMEs operate in the service sector (44 percent), compared to microenterprises that work mostly in the commerce sub-sector (58 percent). Although SMEs in Argentina sell goods and services at the national and international levels, they have a limited scope of activities. SMEs concentrate their activities in local markets, purchasing and sales activities within the same locality and even within the same province, indicating a weak capacity to expand their geographical area of operation. Although SMEs are providers of goods and services to other businesses, many of which are large enterprises, the number of business clients is rather small. 2.2 SMEs in the City of Buenos Aires Using a 1999 survey of enterprises in the City of Buenos Aires, Table 2.2 presents indicators by firm size on the main characteristics of entrepreneurial activities. Businesses are divided into three subsectors. Service, which includes hotels and restaurants, transport, warehouse, real estate services, rental services, and business related services. The commerce sector includes retail and wholesale shops and the industry sector includes manufacturing enterprises. The main highlights are: * The dominance of the service sector in the City of Buenos Aires is consistent with the increasing percentage of GDP contributed by the service sector and with previous economic census data. In Argentina, the participation of the service sector has increased in the last two decades, reaching 67 percent of the GDP in 1999, up from 48 percent in 1989. According to the 1994 economic census, SMEs employed 62.6 percent of workers in the service sector, generated 61.7 percent of the aggregate value, and 63.2 percent of the gross productive value. * Although the service sector makes the largest contribution to both firm number and employment, it is clear that the commerce sector holds the greatest share of total income for firm with more than 5 workers. * The distribution of firm size varies by sector. Microenterprises (less than 10 workers) comprising most of the service and commerce sectors, and small firms (11-50 workers) accounting for the lion share in industry.5 * Industrial firms are substantially older than firms in the service and commerce sectors. About 60 percent of industrial firms have been in operations for at least 20 years. The youngest firms seem to be concentrated in the service sector. 5 Differences in economies of scale between operations in the industry sectors and other sectors may explain this finding. 6 * Service enterprises are the least likely to be stock corporations; this starkly contrasts with the industry sector, which has both the oldest firmns and the greatest proportion of limited companies among the three sectors. * Few industrial firms sell their goods in foreign markets. On average, about one in five industrial firms will export, but larger firms are more likely to export. * Labor regulations impose substantial costs for smaller firms, explaining their lower level of salaried workers. The reported non-wage expenses as a fraction of gross wages ranged between 57 percent and 36 percent for firms with less than 50 workers in the service and commerce sectors. These high rates may explain why the share of informal workers has soared for smaller firms as is shown below. Table 2.2 Characteristics of Argentine Enterprises in the City of Buenos Aires, 1999 Number of employees T_otal < 5 6 - 10 11 - 50 51-100 101-200 201-500 over 500 Service Market Share (lo in column category) Firms 63 90 62 50 42 60 81 88 Employment 62 90 63 51 44 61 84 88 ncome 34 77 48 33 19 26 39 48 Ceneral Characteristics /o firms 100 37 21 34 4 2 2 0 /o workers 100 5 7 32 11 10 23 13 /o income 100 14 9 52 9 5 7 3 /o age of firm with more than 20 years 20 15 15 26 33 22 19 29 /o limited companies 33 16 27 49 82 27 73 47 /osalariedworker 84 41 71 84 80 87 91 100 Gross wages/ total expenses (%) 21 6 13 18 28 35 40 37 Non-wage expenses / gross wages (%) 45 125 54 47 57 35 25 28 Commerce Market Share (%l in column category) irms 24 9 31 31 27 11 7 12 Employment 20 9 30 28 25 11 6 12 ncome 53 22 48 58 61 42 41 52 General Characteristics h firms 100 26 21 43 6 2 1 0 / workers 100 1 9 54 19 6 5 5 /oincome 100 3 6 59 19 5 5 2 age of firm with more than 20 years 29 6 24 34 33 29 39 60 A limited companies 62 49 42 71 93 87 83 100 A salaried worker 88 69 74 88 89 94 94 100 ross wages/ total expenses (%lo) 6 4 4 6 5 6 8 15 Non-wage expenses / gross wages (%/6) 29 32 40 28 36 27 30 11 ndustry tarke Share (/a in column category) Firms 13 1 7 20 32 29 12 n.a. Employment 18 2 7 20 31 28 10 n.a. ncome 12 0 4 10 20 33 20 n.a. General Characteristics % firms 100 2 12 66 14 5 I n.a. A workers 100 0 3 45 27 16 9 n.a. D income 100 0 2 42 27 18 10 n.a. %age of firm with more than 20 years 58 46 52 58 63 71 38 n.a. A limited companies 69 69 51 69 77 91 74 n.a. % salaried worker 94 79 86 93 94 97 99 n.a. ross wages/ total expenses (%/o) 21 6 13 18 28 35 40 n.a. on-wage expenses / gross wages (%) 26 65 27 30 19 17 13 n.a. Investment % firms that invested 57 44 21 55 85 92 100 n.a. nvestment outlays/total income (%) 3 11 2. 3 3 3 9 n.a. Export D firms that exported 19 - 7 18 28 44 56 n.a. exports/sales (%) 2 2 2 4 3 7 n.a. Source: Authors' calculation based on the 1999 survey of firms in the City of Buenos Aires carried out by INDEC. 7 2.3 SMEs are Different from Large Firms: Size Matters Argentine SMEs, like those in other countries, have special features of their own: their economic rationale is of a particular kind, their behavior is specific, and they are impacted quite differently by regulations and risk factors. Generally speaking, these firms are family businesses and their administration falls directly on the owners (often untrained) Thus, their management often relies on untrained individuals, and the owner's outlook (generational cultural factor) is especially important in deciding actions that determine a firm's evolutionary process. In fact, SMEs not only differ in regard to how effective they are at strategic planning, but may also even pursue goals which differ from profit maximization, such as market survival, maintaining their jobs and sources of income, or keeping up the family business tradition. For all these reasons, SMEs might need to be treated as economic agents, which are not merely "smaller-sized large ones." Table 2.3 and Table 2.4 illustrate some differences according to size. Of particular relevance in the comparison between SME and large firms, are factors that significantly impact their unit costs, interest rates and transaction and logistic cost, all significantly higher for SME. While differences are to be expected and should not be, in principle, a cause for intervention, components of those differences that are induced by market and government failures are. Table 2.3: Comparing Firms by Size in Argentina Concept Small Medium Large Organization Individual Owner S.A S.A. Avg. age 28 years 38 years 57 years % that sub-contract 4.5% 9.5% 52% Operate with bank Loans 76% 86% 98%- Avg. annual interest rates on 25% 23% 12% bank loans Informality (%of workers) 70% 32% 10% Transaction Costs (0/oof sales) 15% 7% 1% Logistic Costs (0/%of sales) 35% 26% 15% R&D investment 0.3% 1.5% 3.8% Technological support Customers and Providers Customers and Providers Providers and other sources Initial development based on: Previous experience Experience and access to new Access to new technology technology Source: FIEL (1996, 2001) and Bank Staff estimates. Labor market features also have a differential effect on SMEs. First, the heavily regulated structure of Argentina's labor market, again impacts adversely all types of firms but more so SMEs, mostly on the nature of costs spread over various scale levels as shown in Table 2.3, and also due to the lumpiness of some of those costs. For example non-wage costs reached 125 percent of gross wages for microenterprises in the service sector. Unemployment rates, another feature of labor markets, are substantially higher among small firms-the poverty impact angle. Small firms report four times larger unemployment rates than large firms, while medium firms report two times larger unemployment rates than large firms (Guasch, 2000). Additionally, as Figure 2.1 shows the underemployment rate of small and medium firms is over 50 percent while that of large firms is about 9 percent. Figure 2.2 displays unit costs by sector, showing their evolution through the 1990s. The service sector-mostly SMEs and non- tradables- shows its dismal performance, relative to other sectors. 8 Figure 2.1: Underemployment by Firm Size Figure 2.2: Unit Labor Cost (in constant pesos (Percentage) 1992=100)* 6D ---- ------ ------115 _ - _ 10~~~~~~~~~~~~~~~~~~~~~~~~~1 1~~~7o _ _ _ _ _ _ _ _ 7 E 0 - 105 7 40 - 1 S61 ' ) 2D unfmm E1 / +Cvus . 1 10 e 5 - . d s I h1 (1)EldbdbyMhtfi&&g Source: Guasch (2000). 2.4 Productivity in the Manufacturing Sector Productivity and usage of inputs differ significantly according to firm size. Again this is not surprising given differences on modes of production, scale effects and the myriad of costs with differential impact according to size. In any event, it does clearly impact the prospects for growth and development of SMEs. According to Fiel (1996), among manufacturing Firm Size (Totala100) firms, labor productivity for larger firms is 2 to 4 times higher than that of smaller firms both when firm size is measured by number of workers and - > . ,., gross production value (Table 2.5). In 1999, the .2. ,..,,manufacturing sector of the City of Buenos Aires o s_', _ ,_ napresents the same pattern (Figure 2.3). In terms of __i 9 1 !... I.X0 productivity of capital, Fiel (1996) presents the : - -.r;- _ r * 3 t- L same dramatic differences by size of firm (Table 2.6). Also Table 2.7 shows the working capital requirements according to size, illustrating the differential impact and costs. SMEs require a much higher proportion of working capital as a percentage of sales than larger firms, impacting unit costs and productivity. 9 Table 2.:' Average Productivity of Labor by Firm Size (Total Index = 100) Firm Size by Firm Size by Gross Employment PmeL PmeLH PMeLHC PmeL PmeLH PMeLHC Production Value Total 100.0 100.0 100.0 100.0 100.0 100.0 Total 1-5 65.2 59.9 . 64.3 45.5 37.3 40.8 Size 1 6-10 62.4 43.0 48.1 58.8 56.4 58.3 Size 2 11-50 68.1 81.6 85.4 79.8 104.1 110.7 Size 3 51-200 182.8 157.3 157.2 | 139.0 147.5 128.6 Size 4 201-500 119.0 128.6 112.2 101.6 193.8 186.6 Size 5 501+ 82.7 129.9 131.2 1630.0 878.3 912.1 Size6 PmeL = Average Productivity of Labor (defined as valued added per worker) PmeLH =Average Productivity of Labor (per hour worked) PMeLHC = Average productivity of work hour standardized by human capital Source: FIEL (1996, Table 12, p.152). Table 2.5: Average Productivity of Capital by Firm Size (Total Index = 100) Firm Size by Gross Production Firm Size by Employment Average Average Value Total 100.0 100.0 Total 1-5 36.2 31.8 Size I 6-10 53.2 79.6 Size 2 11-50 . 99.0 129.6 Size 3 51-200 148.6 199.9 Size 4 201-500 131.9 120.7 Size 5 501+ 77.1 71.3 Size 6 Source: FIEL (1996, Table 13, p. 152). Table 2.6: Working Capital (As a Percentage of Sales) Firm Size Standard Standard Firm Size Gross Employment Average Deviation Average Deviation Production Value Total 44.6 . 44.7 44.6 44.7 Total 1-5 45.1 27.4 54.4 36.6 Size I 6-10 52.7 32.5 47.0 36.8 Size2 11-50 50.2 41.5 45.2 64.5 Size 3 51-200 43.6 66.7 36.6 29.9 Size 4 201-500 30.9 . 27.1 28.0 23.4 Size 5 501+ 26.0 25.0 10.0 11.2 Size 6 Source: FIEL (1996, Table 16, p. 155) Regarding the capital labor ratio, as Table 2.7 shows, it increases with size of the firm, illustrating informality and the bias to use labor for capital in SMEs, again dissuading and delaying investments and the use of more effective technologies for product or process. Table 2.7: Capital-Labor Ratio According to In summary, the case has been made that SMEs are Size (Total Index = 100) characteristically and behaviorally different from Firm Size large firms and are impacted differently as well. by Gross Firm Size by Production That in itself need not be a reason for intervention. Employment Avg Avg. Value A closer look at the environment under which they Total 139.7 89.5° TotS operate, and the causes behind those differences, 6-10 69.4 79.3 Size 2 needs to be taken to evaluate the case for 11-50 75.1 100.7 Size 3 intervention, at least on efficiency grounds. To 51-200 111.3 75.1 Size 4 unleash their potential, unit costs have to come 201;500 26 5 363.2 Size 5 down, and productivity and quality standards have Source: FIEL (1996, Table 14, p. 153). to increase, the firms on their own can and should take measures to improve their predicament, but the government also needs to address and correct inefficiencies and market and government failures that 10 enhance those differences across firms of different sizes. The questions remain who should intervene and what is it that needs to be done in order to achieve those results. 11 3. SME Response to Economic Liberalization: Determinants of Success and Growing Heterogeneity of Firms What has been the SMEs' response to the major changes originated during the economic liberalization of the 1 990s? The growing heterogeneity of SMEs emerges as the single most salient feature of the aftermath of economic liberalization. Small and medium-sized enterprises, like all other economic agents, differ from each other. This chapter identifies three types of SMEs: 'successful', 'muddling through' and 'failing' firms. This class of heterogeneity merit special consideration. Although inadequacies of the business environment, knowledge, information and technology constraints, and low access to finance affect all types of SMEs, each type of SMEs have different capabilities to face those growth constraints. In this sense, this chapter is a bridge to subsequent chapters on business environment, finance, and knowledge constraints. Moreover, the differential impact of these constraints by type of firm also has implications for public policies to foster SMEs. In this dimension, the guiding principle of "matching policies to SMEs capabilities" proposed in this chapter will be further addressed in the policy chapter of this report. This chapter is organized as follows. First, it examines the uneven response of SMEs to the economic reforms in Argentina. The next sections explore this uneven response from various angles by relying on a variety of techniques. Section 3.2, on the basis of basic statistical data, provides a taxonomy of inter- industry responses. Section 3.3, on the basis of detailed case studies of manufacturing SMEs, provides a taxonomy of inter-industry SME responses by identifying successful SMEs, muddling through firms, and unsuccessful firms. Section 3.4 summarizes an extensive survey work of successful exporters recently completed by Fundes-Argentina. On the bases of case studies carried out by Fiel, Section 3.5 outlines factors of success of SMEs in the service sector. Using econometric analysis of a panel survey of industrial SMEs, Section 3.6 statistically identifies determinants of good SME performance. Lastly, Section 3.7 summarizes all the other sections and briefly examines policy implications of the SME heterogeneity. 3.1 Uneven supply response to reforms The response of firms to structural reforms has been highly uneven in Argentina. This unevenness has been, not surprisingly, both across sectors and within sectors. Predictably, privatized enterprises in telecommunications, utilities and financial services and some manufacturing sub-sectors demonstrated the largest supply response. Certain relatively labor-iiitensive commodities (such as apparel, electronic assembly, automobiles) demonstrated export growth. Some of these exports, however, were assisted by sector-specific measures and trade policies, as in case of automobiles. A clear picture of the SMEs' problems emerges from all the studies in this area. In the case of the manufacturing, Peres and Stumpo (2000) summarized the performance of SMEs after the economic reforms of the 1 990s in Latin America. In the case of Argentina, the authors find that SMEs increased their share of industrial production. At the sectoral level, SMEs adjusted their contributions to certain sub- sectors (Tables 3.1 and 3.2). Table 3.1 shows the response to structural reforms of industrial SMEs in Argentina ranks the third behind Chile and Mexico, yet ahead of Brazil. These numbers look relatively good but they should be taken with caution. And the performance of SMEs is very different in the service sector where stagnation has been more the norm than the exception. The caution is warranted for a number of reasons. Growth of SME production may mean that SMEs do not graduate into large firms, as they tend to do in knowledge-intensive sectors in developed economies. The evidence corroborate this hypothesis. The IAMC survey for industrial SMEs reveals that about 55 percent of SMEs have more 26 years. The UIA survey shows that 31 percent and 40 percent of industrial SMEs have, respectively, 16-30 years and more than 30 years. These data strongly suggest that the 12 Argentine SME market lacks turbulence. Argentine SMEs fail to grow or graduate from the SME class and fail to die, implying the existence of strong barriers to development and start-ups. Second, reduction in employment (and the resulting increase in labor productivity) may not be result of efficient labor saving technologies but a defensive response to cut labor in the presence of massive institutional rigidities in labor market. Chapter 4 on business environment will elaborate more on this. Overall the picture presented by this chapter is of peculiar dynamics - of limited adjustment, of defensive and survival dynamics rather than outright stagnation. This eases the SME predicament in the short-term (as any survival skills would) but exacerbates SME competitiveness in the long-run (as learning, managerial creativity and energy goes into survival modes rather than mastering export-oriented technical and marketing capabilities). Table 3.1: Rates of Changes In SME Production (Annual rates in percentages) Countries Production Argentina 1984-94 4.0 1995-2000 0.6 Brazil 1985-97 1.0 Chile 1981-90 -1.4 Chile 1990-96 9.1 Mexico 1988-93 9.5 Source: Peres and Stumpo (2000, Table 5, p. 1648) based on industrial SMEs, and information produced by CEPAL and INDEC. Table 3.2: Economic Liberalization: SME Winner and Losers Countries Winner sectors Loser sectors Argentina Non-alcoholic beverages, Textiles, garments, sawmills, Medicines, paints, iron and steel, pottery machinery of general use Brazil Footwear, furniture, and printing. Food, garments, chemical products Plastic products Chile (1981-90) Chemical products Food, sawmills, medicines, metal products Chile ( 1990-96) Sawmills. Medicines. Building Textiles, chemical products, materials. refining of copper, iron I Metal products Source: Peres and Stumpo (2000, p. 1649) 3.1.1 Manufacturing SMEs: Inter-industry Differential Responses6 Analyzing the inter-census period of 1984-1994, Yoguel (1999) identifies three groups of sectoral responses to economic liberalization within SME production space. The first group shows an active response to economic liberalization. As shown in Table 3.3 and Table 3.4, the 26 sub-sectors in this group increased their share in the economy and were generating half the SME output while experiencing a 20.7 percent fall in employment. This active response group still fared better than the overall SME sector 6 This section is based on Yoguel (1999). 13 (28.6 percent fall in employment). But this very significant decrease in employment in the group of the best firm is a very good illustration of SME predicament in Argentina: The second group of sub-sectors showed a passive response to economic liberalization. About 44 sub- sectors belong to this group that both lost output and increased the ratio of their imports to their output such as various branches of the metal products and machinery industries. In the passive response sub- sectors, the import ratio increased dramatically from 11 percent in 1984 to 40 percent in 1993. As a result, the ratio of these sectors in the SME production space decreased from 55 percent in 1984 to 39 percent of it in 1993. This is a passive adjustment because it resulted mainly in cutting inputs while employment in this group plummeted by a staggering 42 percent. This group faces a challenge of transition from passive response characterized by employment and cost cutting to a more active and adaptive adjustment related to technology adoption, staff training and other learning-intensive activities. One of the objectives of public policy should be about facilitation and acceleration of such a transition. Table 3.3: SME Production Space: Changes in Output Share and in Foreign Trade Ratios in Sectors with 'Active', 'Passive' and 'Adaptive' Responses, 1984-93 Share of total gross Value of output Export ratio Import ratio 1984 1993 1984 1993 1984 1993 . (%) (%) 0/) ( (%) ( %/ ) (%)0 Type of sectors Active response to trade openness (26 38.8 51.1 7.3 9.1 4.0 10.9 sectors) Passive response to trade 55.4 39.2 6.0 12.6 10.9 39.6 Openness (M sectors) . Adaptive response to trade openness 5.8 9.7 11.9 5.8 25.1 19.3 (12sectors) Total for above 82 sectors 100.0 100.0 . SME production space refers to the 82 specified sectors in which the share of output produced by SMEs (in the base year 1993 was between 30 percent and 100 percent Source: Yoguel, 1999, based on the 1984 and 1993 National Economic Censuses, foreign trade statistical series, and Bonvecchi, Kosacoff and Yoguel (1998) Table 3.4: SME Production Space': Changes in Employment In Sectors with 'Active', 'Passive' and 'Adaptive' Responses, 1984-93 Variation in Employment as a share of total employed in number all three tyes of sectors employed 1984 1993 1984-93 (%) (%) (%) Type of sectors Active response to trade openness (26 sectors) 46.7 51.9 -20.7 Passive response to trade openness (44 sectors) 47.0 38.6 -42.0 Adaptive response to trade openness (12 sectors) 6.3 9.5 +7.6 Total for above 82 sectors 100.0 100.0 -28.6 ' SME production space refers to the 82 specified sectors in which the share of output produced by SMEs (in the base year 1993 was between 30 percent and 100 percent Source: Yoguel, 1999, based on the 1984 and 1993 National Economic Censuses, foreign trade statistical series, and Bonvecchi, Kosacoff and Yoguel (1998) The third group of sub-sectors presents an adaptive response to trade liberalization. The 12 sub-sectors in this group are metal products and machinery industries, which in the 1980s were already exposed to competition from imports and characterized by very high import ratio. These empirical results for 14 Argentine SMEs demonstrate the impossibility of so-called vertical industrial policy based on picking the winning sectors. Ad hoc, it is virtually impossible to determine which kind of response -active, passive, or adaptive- a specific sector of the economy would demonstrate. Machine-building sectors, for instance, demonstrated all three types of responses in Argentina. Heterogeneity of the dynamic response calls for a humble role of the'public sector: acceleration of active responses by non-distortionary means focused primarily on the improvement of business environment. 3.1.2 Intra-industry Heterogeneity of Manufacturing SMEs' SME heterogeneity exists not only among economic sub-sectors but also within each individual sub- sector. As reported in a set of case studies in the manufacturing sub-sector, Gatto, Moori-Koening, and Yoguel (1993) identified three groups of SMEs: (i) a small group of successful SMEs; (ii) a large group of muddling through SMEs; and (iii) uncompetitive SMEs with little chance, of survival in the market and which in the long term should go out of business. Successful Manufacturing SMEs: inter-firm networks and cooperation agreements. Although the group of successful manufacturing SMEs is small, accounting for less than 5 percent of total output, these firms appear to acquire marketing and technological capabilities to increase sales and exports. One distinct characteristics of this group is inter-firm linkages. To take advantage of available preferential tariffs and the existence of bilateral sectoral accords, for example, some firms in this group have signed cooperation agreements with Brazilian and other foreign firms. These agreements have implied a two-way flow of goods, which has affected what gets produced by the firms in both countries (Yoguel, 1996). Typically, the firms in this sector se are medium-sized firms that export a significant part of their total sales (over 25 percent) to demanding markets (USA, EU and, to a lesser extent, Brazil). Muddling through firms: defensive strategies. This group comprises'the largest portion - bout two- thirds- of all manufacturing SMEs that have been facing the 'challenge to re-launch their businesses on new foundations' (Yoguel, 1993). Defensive strategies are largely about aggressive cost reduction. Technological and marketing capabilities of these firms remain outdated. 'Losers' with a little chance of survival. Firms in this group, which represent one third of the total, lack the most basic marketing, administrative and technological capabilities. Yet as in many family-run and owned firms, decision to close will depend not necessarily on rational economic principles but rather concerns of family businesses such as tradition and subsidies from. extended%family. In terms of SME policy, this group may require 'light touch' inexpensive assistance focused mainly on conveying on a range of alternatives available to these SMEs, organizing retraining to facilitate gradual downsizing of these firms on their way to closure. 3.2 Evidence of Successful Exporting SMEs8 During the 1990s, Argentina witnessed a quite significant increase of exports that was accompanied by a persistent deficit in the balance of payments and a concentration of exports in few large firms. In a recent analysis of exporting SMEs, Fundes (2001) shows increased share of industrial exports in the overall exports and increase in the number of exporting firms. As an indication of increasing duality of the industrial structure, the number of large-scale exporters grew by 66 percent while the number of SME exporters increased only by 26 percent. Although this extreme'concentration of exports in a few small firms can be partly explained by some SMEs graduating into large firms, this dynamic effect is apparently secondary. Exporting SMEs account for 37 percent of exporters but contribute 15 percent of aggregate exports, with 3 percent of firms generating 80 percent of industriil exports (Table 3.5). 7 This section is based on Gatto, Moori-Koening, and Yoguel (1993) and Yoguel (1999). 8 This section summarizes Fundes' study on successful exporting SMEs published in 2001. 15 Table 3.5: Structure of Industrial Exports in 1999 Exporting Firms Manufacturing Avg. Export Exports / Firm Size __________ Exports per Firm Sales Number % mulls$n % (US$ 'OOOs) % arge firms 668 6.4 13,919 85.( 20,837 18. SMEs 3,83 37.8 2,37 14.5 620 4. Successful 78 7.8 1,01 6. 1,296 n. Non-successful 3,048 30.1 1,358 8.3 446 n.a, Microenterprises 5,634 55.6 79 0.5 14 n.a Otal 10,134 100.0 16,375 100.( 1,616 15.7 Source: Fundes (2001, Table 8, p.30) a/ Successful SMEs includes firms that have been exporting for various years, which exports have been growing faster, and that export to a variety of markets. 3.2.1 Successful SME exporters Given the increasing export duality, it all the more relevant to understand determinants of success of successful SME exporters, though which consistently increased exports and the range of exported products. The group of successful SMEs studied by FUNDES is composed of approximately 800 firms whose represent about 20 percent of exporting SMEs and about 8 percent of exporting industrial firms. These are the SME firms 'who made it'. The important questions are how and what lessons can we leam. To answer these questions, Fundes selected about 80 firms (about 10 percent of total) for detailed case studies. These are mainly medium-sized firms with 45 employees on average and annual sales of 42 million. Exports, on average, represent 30 percent of the output. Although these firms grew strongly in the 1990s they were not, of course, immune to the recent crisis. Box 3.1 summarizes the working hypothesis of the Fundes' study. -Box 3.1 - Working Hypothesis - Fundes' Study HI. These firms export a higher proportion of their sales than the industry average H2. These firms have built their strength in the 1980s, which was consolidated in the 1990s. Although not continuously, their export experience cover more than 10 years. H3. These firms have improve the effectiveness of their distribution channels. H4. Following trade liberalization, they have increased their import content of their inputs and raw materials. H5. Even the smallest firms have formalized within their organizational structure an area for international trade. H6. They have broad commercial knowledge about: distribution channels, export potential of their products, main competitors, and clients' preferences. H7. They engage in frequent communication with external clients. H8. During the 1990s, they have made substantial investments, boxes fixed investments and in human capital H9. They rely on e-mail to contact foreign clients. Most of them have web-page, but they do not make electronic transactions. H10. They have broad innovative capacity and possess formal/informal research and development departments. They care about their quality assurance process and the importance of training. H 1l. The distinction between producing for domestic and international markets is blurred. H12. These firms use very few public support rograms. Source: Fundes (2001). 16 3.2.2 Exporting as an incremental learning process The overall picture behind the success is gradual painstaking learning process spanning many years. About 53 percent of the surveyed firms started exporting in the last ten years and introduced, over a time, a package of change which include investments both in fixed capital and 'soft' capital (marketing, managerial and technical capabilities). More specifically, these firms emphasize the following areas: * Quality control. More than half the firms comply with national and international quality norms. About 25 percent of firms have ISO-9000 certification. Significantly, since we are talking about 'best of the best' SMEs, this number is quite low by international standards suggesting huge room for improvement and untapped potential. * Training and R&D. These activities are carried out by two third of the firms. In a lower proportion of firms, there are informal groups for product and process. Some kind of research and development facilities exists in 30 percent of firms. * Export marketing capabilities. More than 70 percent use pamphlets to promote their products in foreign markets (in 75 percent of cases in English and in 40 percent in Portuguese). Eighty percent of the firms have visited clients abroad (an average of 20 per year), or have visited their clients in Argentina. Fair and exhibitions are used extensively; each surveyed firm visits annually three such events. Significantly, in most cases marketing capabilities are developed without the help of public organizations and with little support from private sector chambers. This is an important insight corroborating the need for dramatic overhaul, not marginal tinkering, of the existing SME promotion system (see Chapter 8 for an outline of the reform we propose). * Use of Internet. In the last few years the measures on foreign trade complimented with the use of information tools in digital formats and the use of the Internet and electronic mail with commercial purposes and as new modes to improve communications and relations with clients: more than half of the firms use information on foreign trade in CD ROM or Internet; about 55 percent has developed a web-page to publish its products and supply institutional information, and they routinely update their contents; and most are connected with foreign clients via electronic mail. The use of the Internet seems to be a-more useful tool to facilitate communications, obtain information about what happens in foreign markets, and promote products, than to detect business opportunities and the possibility of realizing businesses in foreign markets. * Linkages with domestic and foreign firms. There continuous buyer-supplier relationships, especially with firms with a long track record of activity, not less than 15 years have. 3.2.3 Export Project Cycle: From Identification of Export Opportunity to Sustainable Exports Export project cycle is a helpful practical concept to organize the discussion on how the firms achieved export success. It starts with identification of export opportunities and proceeds through development of marketing, technical, and managerial skills to transform opportunities into products and services for clients. Finally once, opportunities, capabilities, clients, and supplies are developing it takes consistent effort to maintain, expand, and diversify exports. Identification of exports opportunities Most of the surveyed firms started exporting through their own trials and errors. The identification of the first business opportunity, foreign trips (70 percent of the cases) were the mostly essential followed by participation in fairs and business missions (47 percent), and by initiatives of potential clients (29 percent). This is a crucial conclusion because it shows that the role of public sector is very limited at this initial stage of export project cycle. Public databases on exports, which export promotions agencies 17 develop, could useful as background information, but they alone cannot substitute intensive entrepreneurial search for new clients and new export markets. Acquisition of sustainable export, capabilities.. Acquisition of export capabilities that result in consistent export revenues is a learning process characterized by considerable lags. The consolidation of exporting capabilities meant was a process significantly more complex than its initiation. The time transpired between the identification of the opportunity and the execution of the first transaction was lower during the first year, while the consolidation of the foreign business demanded a period not lower than the two years. Only 37 percent of the firms reported one year or less as a duration of such learning, 36 percent reported that it took between one and five years to move to repeated export transaction, while in 27 percent of SMEs it took more than five years. Complexity of actions to acquire sustainable export capabilities explains these significant lags. The finding argues for patience and for limited and "light touch" but continuous support from the private sector, such as mentoring and coaching, rather than one-off involvements. Maintaining export success. Human responses focused on marketing and other trade activities representing 5 percent of labor in the surveyed firms) appear to be critical. In 95 percent of the firms, the persons directly involved in these activities constitute a stable group. These employees maintain important relationship with clients, suppliers, and other business people from the country. Knowledge and information in export project cycle. Basic information for initiation and consolidation of exports emerge mainly from direct contacts with clients and visits to fairs Embassies and state representation abroad have very little impact. Each firm uses an average of two channels of export sales: about more than 50 percent of exports are channeled through direct sales, 20 percent through firms' representatives, 15 percent through direct presence and 10 percent through traders. The rest corresponds to agreements with companies abroad, and in' a tiny number of cases (lower than I percent) to operations of electronic business. To assess whether all of the success factors mentioned above are inter-related, Fundes (2001) estimated a Kendall's correlation between a set of variables indicating that export dynamics are related to aggressive strategies for productivity increases and sales growth. These aggressive strategies encompass productivity increasing activities, such as quality control and staff training, as well as the management and marketing strategies, such as foreign sales departments, active promotion, and broad market knowledge. 3.3 Dynamics of Industrial SMEs9 One of the few sources of information about SMEs in Argentina is the survey of the Observatory of industrial SMEs of the Industrial Union of Argentina. From this survey, a panel of about 218 industrial SMEs can be constructed to evaluate dynamic patterns of industrial SMEs. Table 3.6 shows average levels of selected variables for these firms according to two measures of firm size: number of employees and sales volume in 1996. There is a positive correlation between sales volume and number of employees although not in the number of sales per employee that tends to increase in the larger firms. The larger firms in our sample pay a much higher average salary (above all based on the number of sales). Finally, the smallest firms seem to have more difficulty obtaining financing from banks and they seem to compensate by using their own capital and also depending on their suppliers (by delaying payment for inputs). 9 This section is based on a background paper prepared by Fiel for this report. 18 Table 3.6: Industrial SMEs In 1996 Average of Selected Variables by Number of Workers and Annual Sales Variables Number of Employees | 0-15 16-30 31-50 51-100 101-200 Number of observations 22 54 45 70 27 Annual sales ($) 782,555 1,640,970 3,251,574 5,009,689 8,599,217 Annual sales per employee ($) 72,991 68,263 82,321 68,379 69,289 Age of firm (year) 28 25 27 32 28 Annual wage ($) a/ 14,330 15,242 17,469 14,901 16,278 Cost of capital (%/6) 13.1 14.5 13.7 12.5 11.1 Sources of financing (% total) Own resources 41.9 37.5 37.8 39.8 43.6 Banks 36.2 45.2 43.2 41 39.4 Suppliers 21.9 17.3 19 19.2 17.1 Variables Annual Sales (S 'OOs) 0-1,000 1,0012,000 2,001-3,000 3,001-6,000 6,001-20,000 Number of observations 38 48 39 50 43 Annual sales ($) 19 40 49 67 96 Annual sales per employee (5) 42,675 50,674 65,238 76,181 122,010 Age of firm (year) 35 25 23 32 28 Annual wage ($) a/ 11,465 11,887 15,361 16,729 22,005 Cost of capital (%/6) 12.6 14.1 13.1 14.1 11.3 Sources of financing (% total) Own resources 46.2 36.6 45 32.2 43.6 Banks 35.2 43.7 36.5 47.1 39.3 Suppliers 18.6 19.8 18.5 20.6 17.1 Source: FIEL (2001) based on the survey of industrial SMEs of the Uni6n Industrial Argentina. a/ Include labor cost Fast-growing industrial SMEs are those that increased their sales in the last two years above the sample's average. Table 3.7 presents the averages of the selected variables for the two types of firms: "fast- growing" or "slow-growing." Although these two types of firms have similar sales volume, they have striking differences in age, membership to economic groups and exporting behavior. Fast-growing industrial SMEs seem to train their employees slightly more, which may explain their slightly lower average annual salary. They also invest more in machinery and equipment and have a lower percentage of labor costs compared to total production costs. Overall, about 44 percent of fast-growing industrial SMEs have been able to export their products, and financed their activities with bank loans. While facing more competition from foreign products, these firms have been able to remain in business and to keep their workforce. Slow-growing industrial SMEs finance their activities from internal sources of funds, from partners and business associates, or by delaying payment to suppliers. However, both types of firms face similar cost of capital, of about 12.5 percent. 19 Table 3.7: Industrial SMEs Average of Selected Variables by Growth of Sales in 1996-1998 Growth of Sales Variables Above Average Below Average Fast-growing Slow growing Number of observations 75 143 Annual sales in 1996 ($) 3,833,304 3,828,732 Number of workers in 1996 56 55 Growth of employment 1996-1998 16.7 -3.5 Age of firms (years) 25 34 Member of economic group a/ 22.7 17.5 Trained employees 62.7 44.1 Exports / Sales in 1996 (%) 14.2 7.7 Wage costs / Total costs in 1996 (%/6) 19.5 24.3 Annual wage in 1996 14,381 16,282 Exporting firm a/ 43.8 31.1 Foreign competitor a/ 37.3 24.5 Investment in machinery / Sales (%) 7.1 4.0 Cost of capital (%/6) 12.6 12.5 Sales to main client / Total sales (%) 21.4 17.9 Sales to three main clients / Total sales (%) 38.9 29.7 Sources of financing (% total) Own resources 38.6 41.2 Banks 44.5 39.6 Suppliers 16.9 19.3 Source: FIEL (2001) based on the survey of industrial SMEs of the Uni6n Industrial Argentina. a/ Percentage of firms with positive responses 33.1 What Determines Sales Growth? To assess the determinants of successful and unsuccessful industrial SMEs, two types of regressions were estimated. The first regression evaluates the variables influencing the growth of sales between 1996-1998 using ordinary least squares (OLS), while the second regression assesses whether growth rates are above the sample's average, for the same period, using a Probit model. Three sets of regressions were estimated using a panel of 218 industrial SMEs that answered the Structural Survey of the SMEI Observatory of the Industrial Union of Argentina between 1996 and 1998. As presented in Table 3.8, the first group of regressions (Model I) uses the number of employees as an indicator of firm size. Although this variable is not statistically significant, firms that faced competition from imports and that paid lower salaries in 1996 experienced higher growth rates in their sales. In the case of the Probit model, younger firms present higher growth rate than older firms, indicating that older firms may have reached optimum size of operations. The cost of capital, which approximates the firm's expected rate of return, increases the probability of success. It seems that fast growing firms are more likely to take on expensive debt. Also, the cost of capital captures the firm's access to external sources of finance. Lastly, firms that paid lower salaries in 1996 were more showed higher growth rates. 20 Table 3.8: Industrial SMEs - Determinants of Success 1996-1998 Model I Model |1 Model HI Growth Success a/ Growth Success a/ Growth Success a/ of Sales of Sales of Sales Type of regression OLS Probit OLS . Probit OLS Probit Marg. Marg. Marg. Explanatory Variables Coef. Coef. Effect Coef. Coef. Effect Coef. Coef. Effect Constant 4.1781 3.6312 4.9547 3.4424 4.9906 2.4445 (4.355) (2.193) (4.194) (1.683) (4.484) (1.259) Number of Employees 0.0022 -0.0014 -0.0005 (1.257) (-0.470) Log (annual sales in -0.0682 0.0079 0.0028 -0.0709 0.0032 0.0011 1996) (-0.932) (0.064) (-.968) (0.026) Age of the firm (years) -0.003 -0.0148 -0.0066 -0.0027 -0.0185 -0.0066 -0.0027 -0.0185 -0.0066 (-1.018) (-2.928) (0.936) (-2.975) (-0.927) (-2.964) Economic sector 2 0.1311 -0.1232 -0.043 0.1615 -0.1343 -0.0468 0.1734 -0.1565 -0.0542 (0.513) (-0.275) (0.635) (-0.300) (0.686) (-0.350) Economic sector 3 0.3465 0.5259 0.1991 0.3139 0.5433 0.2059 0.3157 0.5228 0.1979 (1.499) (1.332) (1.357) (1.374) (1.367) (1.322) Economic sector 4 0.3765 0.4973 0.1788 0.3596 0.5 0.1798 0.3698 0.4786 0.1721 (1.859) (1.414) (1.759) (1.405) (1.821) (1.348) Economic sector 5 0.286 0.2342 0.0862 0.2912 0.2308 0.085 0.2988 0.1949 0.0714 (1.311) (0.609) (1.332) (0.598) (1.376) (0.506) Log (cost of capital) -0.0361 0.6443 0.2305 -0.0079 0.6344 0.2269 (-0.172) (1.776) (-0.038) (1.750) Log (bank financing) 0.0139 0.1158 0.0414 (0.371) (1.758) Exports / Sales in 1996 0.d495 0.0957 0.0346 0.1145 0.0707 0.0255 0.1242 0.0769 0.0278 (%) (0.335) (0.393) (0.771) (0.289) (0.835) (0.314) Member of economic 0.1375 0.4234 0.1585 0.2518 0.3684 0.1374 0.2544 0.1688 0.0617 group (0.803) (1.445) (1.522) (1.309) (1.679) (0.655) Trained employees 0.1341 0.2158 0.077 0.165 0.2076 0.0741 0.1574 0.2064 0.0737 (1.100) (1.060) (1.338) (1.013) (1.278) (1.005) Log (share of sales to 0.0784 0.1092 0.0391 0.0853 0.1065 0.0381 0.0866 0.1002 0.0358 main client) (1.568) (1.278) (1.700) (1.248) (1.732) (1.180) Foreign competitor a/ 0.379 0.2603 0.0953 0.4133 0.2437 0.0892 0.4073 0.2404 0.0879 (2.743) (1.157) (3.013) (1.095) (2.971) (1.080) Log (wages in 1996) -0.5012 -0.3169 -0.1133 -0.4662 -0.3156 -0.1129 -0.4629 -0.316 -0.113 (-5.399) (-1.978) (-4.569) (-1.828) (-4.535) (-1.833) Number of observations 218 218 218 218 218 218 R2 0.2371 0.1301. 0.2341 0.1294 0.235 0.1296 Correct predictions (% of 74.3 72.9 73.4 observations)* a/ Success is defined as having growth sales above the average t-statistics in parenthesis Source: FIEL (2001) based on the survey of industrial SMEs of the Union Industrial Argentina. 21 The second group of regressions (Model II) uses the firm's volume of sales as proxy of firm size, instead of the number of employees. Regression results are very similar to those of Model I. The main change is that the proportion of sales to the first client has a positive effect on the rate of growth of sales. One percent increase in the participation of the first client, sales will increase by 8.5 percent. The last of group of regressions (Model III) add to Model II the ratio of bank financing to total financing but drops the cost of capital. The results are very similar to the previous models. Overall, it seems that fast growing firms are younger, have access to bank financing, are member of economic conglomerates, face competition from imported goods, and pay lower wages. These firms have been able to adapt to external forces by reducing their operating costs. More importantly, linkages to economic conglomerates have given these firms an important source of information to find new markets, leverage resources, and thus, they are better equipped to cope with competition. 3.3.2 Determinants of Labor Productivity in the City of Buenos Aires To continue trying to impute the determinants of success among SMEs, this section examines the determinants of labor productivity, defined as reported profits over total employees using the firm data census from the city of Buenos Aires. It is argued that labor productivity is influenced by the quantity and quality of inputs, efforts to improve technological levels, regulatory burden, and firm age. Table 3.9 shows ordinary least square and quantile regression analysis for a set of manufacturing firms in the City of Buenos Aires. To avoid endogeneity problems, all the explanatory variables are lagged one year. A larger amount of capital intensity, proxied by inventory levels per worker, increase labor productivity as confirmed in the empirical results. The results suggest that firms with higher levels of capital intensity enjoy higher levels of labor productivity. Firms with higher proportion of full-time workers, which are larger, face higher regulatory burden, thus the reducing impact on labor productivity. Another interesting result is that older firms enjoy lower levels of labor productivity, suggesting that younger firms are more dynamic and prompt to face market challenges and that older firms have reached their steady state level of operations. Table 3.9 Regression Results: Determinants of Labor Productivity Among Manufacturing SMEs in the City of Buenos Aires, 1999 Quantile Regression Explanatory Variables (lagged 1 year) OLS 50% Log(inventories/total employees) 0.063 * 0.085 * (1.95) (2.46) Invested in computer/software 0.435 ** 0.213 (1.99) -0.92 Non-wage expenses / salary expenses -5.581 ** -4.771 ** (2.45) (2.01) Log(salary costs/total employees) 0.739 *** 0.725 *** (5.24) (4.80) Finn age: > II years - Dummy -0.234 -0.246 (1.29) (1.27) Constant 1.524 1.554 (1.21) (1.15) Number of Observations 294 294 R-squared 0.15 Absolute value of t-statistics in parentheses significant at 10%; **significant at 5%; *** significant at 1% Note: Labor productivity is defined as annual profits over total employees 22 3.4 The Service Sector SMEs: Preliminary Evidence The following preliminary evidence was gleaned from interviews carried out by FIEL, in 2001 with representatives of the business chambers and associations in selected sectors (tourism, restaurants, cleaning companies, personnel agencies, insurance agencies). These findings are still being corroborated through interviews with the firms themselves: * The SMEs with the best performance over the last several years are those contracted out by large firms to provide services. In these cases, the large firms have played an important role in successful SME management by providing technological support and training and by pressure to cut costs. * The labor-intensive firms are another group that has also been successful. This is because of their ability to apply the organization of human resource management to certain niche markets that develop as previous markets decrease in profitability. This has meant that certain firms have reconverted and evolved from cleaning companies to security companies, by utilizing managerial capacity in handling personnel. In several of the selected sectors (insurance, transport and logistics), new operators have appeared and absorbed a large portion of the demand. Many of these new operators are the same size and use the same organizational structure as large firms, and they have become leaders in this sector. The 1999 survey of enterprises in the City of Buenos Aires also shows interesting features of the service sector in Argentina. Among SMEs, these firms accounted for 63 percent of firms and 62 percent of employment (Table 2.2). Because firms in the service sector are much younger, it seems that entry and exit rates are higher than in other sectors. Most service sector SMEs (58 percent) has been in business for less than 10 years, compared to 23 percent of manufacturing SMEs. Despite presenting the lowest share of salaried workers as percentage of total workers, service sector SMEs have the highest labor regulatory burden; non-wage expenses account for about 10 percent of total expenses, much higher than among commerce and manufacturing SMEs. Furthermore, service sector SMEs spend on taxes a higher share than other firms. 3.5 Implications for SME policy Overall, the major broad factor of success is the entrepreneurial effort of SMEs bolstered through business-to-business collaboration and partnerships. The Fundes study documented that successful exporting SMEs have access to inter-firm networks that are an important source of marketing and technical information. More importantly, the public sector does not play an active role in fostering learning capabilities and access to markets. The empirical evidence suggests that the overall business and incentive environment can hamper the development of effective marketing and technological support systems of inter-firm networks for SMEs. However, the results also caution against complacency and against the presumption that a liberalized private marketplace will be sufficient to secure SME export development. Thus, the government faces the challenge of how to build a capability-compatible incentive framework that would open learning opportunities for SMEs while being market driven. Such a framework will be introduced in the section on policy recommendations. Based on firm surveys, case studies and econometric analysis, this chapter identifies three classes of SMEs: successful firms, 'muddling through' SMEs, and 'clear failures'. Obviously, these three ideal types are best to be perceived as the extremes and the middle point of a continuum of widely heterogeneous economic agents, yet for the evidence discussed above, and for the purposes of policy, it is useful to summarize the main features of these three types of SMEs (Table 3. 10). 23 Table 3.10: Taxonomy of SMEs in Argentina: 'Success', 'Muddling Through', 'The Case for Exit' Muddling through - surviving Clearly unsuccessful, on their Successful SMEs SMEs way to liquidation The major challenge Expansion and growth: Re-founding the firm on new Exit. Phasing out of production facing the firm specialized marketing and foundations on the face of activities technical support, access to dramatically new opportunities finance and challenges opened by economic liberalization Capacity for auto- Quite high, in particular vis- Limited to very basic comparison Virtually non-existent. There is diagnostics and a-vis the domestic and vs. neighborhood firms and most a need to convince the family benchmarking intemational competitors direct competitors owners that exit is in his/her best interests Technological Formal department or Informal technology/ engineering -- capabilities informal technology groups may exist development teams Marketing and Usually there formal Quite weak and tend to develop Rudimentary managerial capabilities marketing departments on ad-hoc basis (taking up to 5% of personnel) Alliances with other 80% of successful exporters If alliances exist, they are mostly No alliances/networking. firms involved in some kind of informal Survival orientation and very alliance. Transition from short planning horizon informal agreements and precludes participation in inter- alliances to formal firms networks agreements Subcontracting Staring to emerge. The Occasional Almost non-existent arrangements owner/manager is actively seeking subcontracting arrangement with large firms Focus of public Facilitation of high-intensity Facilitation of 'light touch' Assistance in exploring policies marketing and technical relatively inexpensive marketing employment opportunities for support based on almost full and technical support services. owners and employees to ease cost-recovery of services. the exit and closure of the business Development of private-public Development of private service providers (e.g. BDS service providers and centers of UIA) thickening of the market for business development services Source: own elaboration on the basis of the studies mentioned in the text. What are the implications of firm heterogeneity for public policies? The heterogeneity of SMEs and their responses to economic liberalization implies heterogeneity of business development needs and constraints. As Table 3.10 indicates, successful SMEs tend to have quite advanced technical, managerial and marketing capabilities whereas the 'persistent failure' SMEs (the last column of Table 3.10) have none of these capabilities. The 'muddling through' firms fall somewhere in between. It does not mean that successful SMEs do not face BDS constraints and do not have BDS needs. On the contrary, as the surveys of successful exporters of Fundes clearly demonstrated, successful firms display the highest effective demand for BDS services. What it does mean, however, that BDS services required by successful SMEs 24 will be quite different from the services for 'muddling through' and 'surviving firms'. Successful firms require high intensity specialized services and are willing to pay full market costs for these services. Muddling through firms may require 'light touch' inexpensive BDS and may require limited-time subsidies on a matching fund basis to show the value of the BDS to the firm and to assure subsequent market demand for the services. 'Looser firms' in contrast may require very basic benchmarking designs that allow the owner/manager to exit or switch activities of the firms to more promising and profitable lines of business. 25 4. Institutional Rigidities: Environment for Business Activities The environment for business activities refers to norms, rules, regulations, laws, and procedures that shape the behavior and operations of businesses in an economy, particularly those dictated by government policies. Moreover, the term includes public and quasi-public goods, such as information and logistic infrastructure that affect businesses' interactions with markets, suppliers, and clients. Thus, as a set of rules, the business environment can constrain or encourage business activities, with significant impact on productivity and unit costs. The structural changes of the 1 990s and the associated quest for productivity increases have contributed to a rise in the demand for improving the environment for business activities in Argentina. A recent survey of 100 Argentine firms found that taxes and regulations are the main obstacles facing SMEs, imposing substantial costs (Batra and Stone, 2000). In contrast to other Latin American countries, Argentine businesses stand out for pointing out cumbersome taxes and regulations, ranked second only after Brazil (Lora, Cortds, and Herrera, 2001). The profitability and competitiveness of the SME sector are affected by the burden of complex business regulations through changes in their associated cost structure. Eliminating unnecessary or inadequate regulations and reducing excessive costs of regulation compliance could have important consequences in the economic activity. As presented in this chapter, a substantial amount of time and economic resources are invested and wasted in complying with existing rules and regulations (formality), some of them unnecessary, and that could otherwise propel economic activity in Argentina. This chapter presents the reality-conceming business and labor force registration procedures, and tax compliance procedures in Argentina and implications for SME development. It finds that Argentina suffers from a high degree of institutional rigidities, which partly explains why roughly one quarter of economic activity is performed outside laws and regulations and only 50 percent of labor is registered with social security.' Registration and creation of a legal entity is a time consuming process for SMEs, with official costs of compliance ranging from $1,755 to $6,500 depending on business size. Overall, the regulatory burden of the business registration system is very high, falling more heavily on small firms and giving large firms a comparative advantage. This chapter is organized in four sections. By examining the inventory of procedures of setting up a business at the federal and local levels, Section 4.1 estimates the costs of access to business formnalization in Argentina. Moreover, Section 4.2 addresses the main problems related to unfriendly and costly labor regulations, and Section 3.3 discusses excessive logistic costs arising from inadequate infrastructure. Finally, Section 4.4 summarizes the consequences, in terms of informality, lower firm productivity, and job creation. 4.1 Business Registration System11 Business registration is an important element for start-ups in any economy, and the relative ease or difficulty in registration is an important component of vitality of the private sector. It is the first point of contact between an entrepreneur and the government, and as such it sets the tone for their entire future relationship. In Argentina, this relationship does not start off in the most positive way since SMEs typically face a process that can take up to several months and require several hundred dollars in fees. 10 See Ahumada et al. (2000) for estimations of the evolution of the informal economy during 1980-1999 in Argentina. " This section is based on a background paper prepared by Giordano and Colina (2001) for this report. 26 Compared to other Latin American countries, Argentina ranks among the worst performers next to Guatemala and Venezuela (Table 4.1). It takes about 130-150 working days to comply with all registration requirements. Table 4.1 Business Registration in As in other countries, business registration procedures Selected Latin American Countries can be grouped into four types: (a) commercial registration of businesses; (b) acquisition of operational Country Number of Working Days permits and licenses at the municipal level; (c) Required compliance with tax regulations; and (d) labor and social Chile 12 - 12 security registration. To comply with each component of Bolivia 15 - 30 the registration process, businesses must perform a Brazil 31 - 60 number of formalities (tramites) at different levels of Colombia 40 - 70 government, visit various agencies, provide Ecuador 60 - 75 documentation and fill out forns, and pay fees and Uruguay 75 - 90 commissions. Moreover, for each formality, businesses Uruguo a/ 75 - 902 have to wait a certain number of days for approval. Some Mexico a/ 42 - 142 registration requirements occur once usually when Argentinab/ 130 - 150 becoming formal, while other requirements entail Guatemala 179 - 179 periodic costs of fornality (e.g., tax payments and Venezuela 170 - 310 workers' social security contributions). Source: Jansson and Sedaca (1999, p. 10, as reported by Tokam (1992)) To comply with business formalities, businesses have to a/ Consejo Coordinador Empresarial (2000) travel through a complex bureaucratic maze at the three b/ FUNDES-BID (2000) Note: Includes commercial, tax, health and levels of government. Thus, with 24 provinces and more security, location and labor registration. than 1,300 municipalities, each with its own regulations to authorize business operations, understanding the business registration system is a challenging endeavor. This section estimates the explicit costs of regulatory compliance including the official costs and the additional costs of preparing formns and documentation required for each formality.'2 For example, these include tax advances, legal fees of notaries, and fees of public accountants and/or special agents (gestores). In many cases, firms need to hire gestores to become aware of procedures and to expedite processes.13 Because some implicit costs are non-quantifiable, such as time spent processing business formalities and forgone output due to evasion activities and sub-optimal investment decisions, these are not estimated. 4.1.1 Commercial Registration The first step in the registration process is commercial registration. In Argentina, commercial registration takes place at the Inspecci6n General de Justicia. To make the registration effective, the entrepreneurs must formalize their signatures with a national public notary. Moreover, depending on the province, the entrepreneurs have to pay a stamp tax (impuesto a los sellos)14 of about one percent of the firm's capital. Next, the entrepreneurs must publish the registration of their company in the Boletin Oficial de la Republica Argentina (Official Gazette of the Argentine Republic), paying about $200, and present of 12 These costs include payments to gather the documents and/or to legalize documents for official recognition. 13 Due to excessive red tape, Argentina has developed a market for consultant services or gestores. The quality of these services varies significantly and thus, there is not guarantee about the reliability of the consulting. Usually, SMEs do not have economic resources to hire consultancy firms with prestige and experience. Moreover, the gestor is often the central figure in fraudulent practices, the most common of these being an agreement among civil servants and gestores to the effect that only applications made through intermediaries will be successful. 14 This is a provincial tax, which is not applicable to the City of Buenos Aires, and thus depends on the jurisdiction. 27 copy of the publication at the registry office. If entrepreneurs register their business in the City of Buenos Aires and in the province of Buenos Aires, they must deposit about 25 percent of the firm's capital in a public banking institution. This requirement adds to the transaction costs of registering a business because not all entrepreneurs maintain banking relations with public banks. The entrepreneur could pay an additional $130 - $135 to have their registration processed through the "urgent" track, which takes about 60 to 90 days.'5 Finally, accounting books must be registered. Figure 4.1 Average Explicit Costs of Business Overall, the cost of commercial registration in Registration in Argentina Argentina ranges between $800 and $1,900, depending on the business size (Table 4.2). Although small firms would pay a lower 20 ------0 absolute value, relative to their monthly company value added, the fixed costs of commercial registration could reach 18 percent 8 10 _ _ _ 7.z of the firm's monthly value added, with wide 3 _ lzii;l 8.1 difference by business size (Table 3.2). LL _ _ _ _ _4.1.2 Operational Permits and Licenses at N L~ tfU Ln T-R PrgUon ' Rqbt.UDol.. the Municipal Level Souna:. Auflbon' CI.liIo,.. * U.n) Registration at the municipality level depends not only among broad categories of economic activity, such as commerce, industry, or transport, but also among variants of a single activity. For example, in some municipalities, businesses must request various authorization to sell packed and unpacked food, as well as food for consumption on the business premises. To assess the costs in complying with business formalities at the municipality level, the following simplifications were made: (a) selection of a set of basic requirements that are applied to most activities in all municipalities and (b) selection of the municipalities of the City of La Plata (capital of the province of Buenos Aires). As a result, the estimates presented in this chapter are suggestive of the overall explicit costs of complying with business formalities.'6 Although the overall costs would vary by municipalities reflecting variations in economic development, the impact of the costs across firm size may be similar. The most important permits and licenses required by municipalities relate to location and construction, health and safety, occupational qualifications, and industrial activities. The principal operational licenses and approvals required in the city of La Plata are the following: * Zoning and construction. The entrepreneur must go to the appropriate office of the municipality to request a verification of the floor plan of the establishment and then request a zoning permit to use the business premises. The costs involved include preparing various certified copies of floor plans and rental contract or ownership deed, and payment of stamps. * Fire security permit. The entrepreneur must then go to the fire department to arrange for a technical inspection, pay another fee, and provide three copies of the floor plans that were verified initially. '5 Previously, the City of Buenos Aires had two procedural tracks: normal and "urgent." However, the City unified both tracks to avoid overlaps. Thus, instead of making the system more efficient, the government has institutionalized the extra cost involved in registration. 16This information on the formalities required by the municipality of La Plata has been extracted from FUNDES (2000). 28 Table 4.2: Costs of Complying with Business Formalities by Business Size US$ B siness Size (number of employees) 1 2 - 5 6 - 10 11 - 25 26 - 40 41 - 50 51 + Avg. Average Monthly Value Added per Business a/ 980 2,565 11,873 23,746 59,409 82,853 340,565 6,125 1. Registration Costs at the National Level Cost of documents 270 270 270 270 270 270 270 270 Professional or consulting fees (includes cost of an 400 400 400 600 800 800 1,500 700 accounting study) Application for "urgent" procedure 130 130 130 130 130 130 130 130 Total c/ 800 800 800 1,000 1,200 1 ,200 1,900 1,100 % of avg. monthly value added 82 31 7 4 2 ,1 X 18 2. Registration Costs at the Municipal Level b/ Cost of documents 200 200 200 200 200 200 200 200 Tax advances 200 250 300 350 400 400 400 329 Professional or consulting fees 300 300 600 800 1,000 1,200 1,500 814 Total 700 750 I,100 1,350 1,600 1,800 2,100 1,343 % of avg. monthly value added 71 29 9 6 3 2 1 22 3. Costs of Tax Registration Cost of documents d/ 70 70 70 70 70 70 70 70 Cost of tax advances e/ 55 75 75 75 100 100 100 83 Professional or consulting fees 50 80 150 200 300 400 1,000 311 Total 175 225 295 345 470 570 1,170 464 % of avg. monthly value added 18 9 2 1 1 1 0 8 4. Costs of Registering Workers Total f/ 50 80 150 200 300 400 I 311 % of avg. monthly value added 5 3 1 - 1 0 0 5 5. Total Registration Cost (1+2+3+4) 1,725 1,855 2,345 2,895 3,570 3,970 6,170 3,219 %ofav. monthly value added 176 72 20 12 6 5 2 53 6. Costs of Formalities to Pay Taxes and Labor 50 100 499 807 1,485 1,988 5,108 165 Obligations per Month I I I % of avg. monthly value added 5.1 3.9 4.2 3.4 2.5 2.4 1 .5 2.7 Source: al Authors' calculations based on information from the 1994 Economic Census, INDEC, Argentina b/ Authors' calculations based on information presented in FUNDES-IDB (2000), which are representative for the municipalities of the city of La Plata c/ Includes: costs of publication in the official gazette, costs of certifying documents, and costs of hiring professional accounting studies. Excludes: initial payment of one percent of capital for the Stamps Tax, annual levy as percentage of capital required by the Province of Buenos Aires, stamp tax to register accounting books. d/ Assumes a cost of $70 to obtain certified copies of the business' by-laws and rental contract or real estate ownership deed e/ Includes the cost of making tax advances the for the Gross Income Tax 0f Includes an estimate of the prices charged by accounting firns for this type of professional consulting and the costs of depositing the amounts deducted and in paying contributions. * Request qualification (habilitaci6n) of the business. The entrepreneur must present the actual application for business qualification. This involves presenting a form completed in triplicate, another two copies of the plan (one of which must be certified), another copy of the rental contract (which must also be certified), a book of minutes and registration proof with national tax authorities. There is a charge for another stamp. Health and advertising. The entrepreneur must request authorization of health conditions and advertisement (billboards) policies, and make the necessary fee payments. In the city of La Plata, these municipal fees vary with business size and type of activity. In short, these procedures require providing information, often-identical information, at different places, using different forms and supporting documents. In the situation described above, the entrepreneur would have to visit three municipal offices, submit documents at six customer windows, including six copies of the floor plan and two copies of the rental contract. Because municipal authorities do not orient entrepreneurs to facilitate and ensure compliance with all the required steps, most entrepreneurs resort to hiring an agent (gestor), who is better informed and has informal connections with bureaucrats, to travel through the jungle of municipal formalities. 29 4.1.3 Registration for Tax Purposes In Argentina, businesses are required to pay a wide variety of taxes at the national, state, and local levels of government. Thus, a single taxable act or condition can be taxed several times. Before initiating its operations, a business and its partners must register with the national tax authority. It must fill out a form for the single value-added tax (impuesto al valor agregado) and another for the income tax (impuesto a las ganancias). The Public Administration of Public Income (Administraci6n Federal de Ingresos Publicos - AFIP) centralizes the collection of national taxes. Tax declarations are made monthly, stating the amount to pay for each category. Although the forms for monthly declarations. are simple, AFIP's regulatory bureaucratic structure and the constant changes in the tax system leave no choice but for businesses to hire a professional economist to fill out the forms and keep track of the legal implications of the existing body of legislation. Businesses that make purchases to suppliers above a certain amount ($1,200) must also register as holding agents (agentes de retenci6n)l for each type of operation (fee payments, payment of services to suppliers and input purchases). The complexity of the regulatory structure does not allow entrepreneurs to easily determine under which conditions the law requires them to register as holding agents. Moreover, the procedures for retaining and making deposits as a holding agent are cumbersome. Once again, the entrepreneur must rely on a professional economist specialized in tax matters. The Simplified System to Pay National Taxes - The Monotributo In 1999, the Argentine government established a simplified tax system for small businesses to reduce transaction costs involved with tax compliance. This system, called monotributo unified the Value Added and Income taxes. Although the initiative is objectively grounded, in practice it has been only relatively successful for the following reasons: * Restrictive eligibility criteria. Eligibility criteria imply that a business cannot sell more than $575 per working day and that it cannot be established as a company. As a result, the monotributo system targets two kinds of businesses: very new and very small, or very small with little growth opportunities. * Overall transaction costs of tax compliance remain high. The monotributo system provides simple registration and payment procedures for national level taxes, but does not eliminate the complexity of provincial and municipal taxes. * Fixed nature of the tax. Because the monotributo is paid monthly based on a fixed amount, some businesses may face a very high tax burden in the "low" months. Anecdotal evidence of a small pasta producer in the province of Santa Fe shows that the tax burden could reach 8 percent of gross sales,'8 which is extremely high compared to the maximum tax rate of 2.5 percent of the Small Taxpayers Regime in Mexico.19 To finance the social security (retirement and health) a SME paying the monotributo has to pay $80 monthly for each work and a variable contribution for labor insurance of about 19 percent depending on the region. Smaller firms (less than 20 workers), which pay an average salary of $229, face substantially larger firms tax rate than larger firms. There are about 63,000 monotributistas paying $80 '7 As a holding agent the business retains the corresponding earning tax of their suppliers. '8 Information collected from interviews with small-scale producers in the province of Sante Fe during December 2000. '9 Under this regimen, small businesses (with annual incomes below $225,000) can deduct three times the minimum wage in their localities. The tax structure is progressive, reaching a 2.5 percent tax on cash incomes above US$100,000. Sales receipts are only required whey they exceed US$5. Micro businesses are required to keep a register of daily incomes. See Sanchez (2000). 30 monthly per worker. The SEPyME is advocating for a reform of the monotributo system to allow seasonal contributions. 4.1.4 Labor registration In Argentina, there are three layers of legal regulations governing worker-firms relations. They are: (a) Labor Contract Act (Ley de Contrato de Trabajo - 20. 744) and the rest of the general labor legislation; (b) Centralized Collective Bargaining at the "sector level," operating as a second tier; (c) firm level contracts (which, if they exist, can onily Figure 4.2 Labor Registration: Firm's Trimites build upon the previous two). I Labor ReRistmlion l Labor regulations, the social security fienersiRegime | l lsystem, (including social coverage, labor- IRciswr a EntloycrIW Rewistimrtwv"rke ct P related risks, health coverage and _1CUIL | _CUSS obligatory life insurance) and procedures fFami nia Fyaints lo related to the tax situation of workers QLabordn uc Lonrt farmrI originate a large set of business Soroidvworkc l iSociei rk l sociworic formalities (trdmites) to register labor qsvo i msvo i xsvo l relations within businesses. -Iincont tax { lro eta Commeem Rime Labor registration involves formalities in Union -1 Won jUnion registering the firm as an employer, hiring Ufe insornon CCT l Ufn insur a: CCT I Lie insurarr CCT l Hu& eti.-emen in umnre | tib insur --I -& Qtirnti e | new workers, and making social security and other employer contributions. After SVO: Mandaota (1f9i99 becoming an employer, each new labor contract imposes an array of administrative procedures, such as providing information about the business and its employees to five different agencies. Tax regulations add another layer of business formalities. Finally, the collective labor agreement requires all businesses to interact with three different official bodies. In short, hiring a new employee means that the employer must take 10 different steps and complete 16 mandatory forms, many of which must be presented in the corresponding government office along with documents that certify the veracity of the information. Overall, approximately 500 pieces of information are submitted, of which roughly 18 refer to the new labor relationship. Hence, more than 96 percent of the information submitted refers to the business, the new employee, and his relatives. Table 4.2 presents a conservative estimate of the overall explicit cost of formalizing labor relations (five percent of the wage bill for SMEs), which can be considered a monthly cost. Complex labor registration procedures may partially explain why the share of informal workers has increased for all firm sizes since 1980s (Figure 4.2). Other factors may also explain this phenomenon, including changes in the employment structure, recent empirical studies point towards an increase in evasion of labor registration requirements (Maloney and Arango, 2000; and Fiel, 2000). Argentina's labor registration system invoives about 12 different government agencies, which collect information about firms, workers, and work relations. Information is fragmented and without mechanisms to verify consistency of information among the various government agencies involved, the quality of the information is poor. Interacting with numerous government agencies adds a dimension of uncertainty in terms of inspections and compliance. 31 4.2 Periodic Costs of Business Formalities Once a business is registered, firms must make periodic costs to remain formal. These include compliance with tax laws and labor regulations. The explicit costs related to tax payments at the national, provincial, and municipal levels, and labor-related obligations amount to 2.7 percent of the firm's average monthly value added. Therefore, firms spend Figure 4.3 Periodic Cost of Business Formalities about 10 percent of the amount that ' - .'-.'., !should be collected in value added tax on bureaucratic expenses. Periodic explicit costs of regulatory compliance have a perverse effect on firms with less than 25 workers. The periodic costs of regulatory compliance are very high for micro 2-5 1 2 6 I O 11-25 2 .40 41 w 51 An. enterprises (less than 5 workers). Busi/ness ze (number of warSers) Although the monotributo simplifies l____________________________________________________ ;the bureaucratic procedure for Source: Authors' Calculations national taxes, micro enterprises still need to bear the cost of the general system. As shown in Figure 4.3, they spend about 3.9 percent of the firm's average monthly value added. Using information from the Encuesta Permanente de Hogares (EPH), the administrative burden of tax and labor regulation is much higher, reaching 7 percent of the firm's monthly earnings (about $1,335 on average).2 4.3 Labor Market Rigidities21 In Argentina, workers have historically enjoyed strong job rights, including the right to advance notice of layoff and the right to severance payments. Moreover Argentina has a very highly centralized collective bargaining agreement system. All types of firms, small and large and independent of their geographical location and predicament are subject to the same agreement negotiated by the heads of the union (cupulas). Clearly, that adversely impacts SMEs not providing them with the flexibility they need to address their particular and individual predicament. Some attempts to provide SME with additional flexibility have been tried but very limited practical success (see Guasch 2000). That legislation, which could be protective under certain economic circumstances, became biased against the absorption of labor once market-oriented policies where implemented in the 1 990s, exposing domestic firms to foreign competition. In addition to a cumbersome and costly labor registration system, labor regulations also introduce other distortions. The Workers Statute (Ley de Contrato de Trabajo) introduces specific job security provisions in the form of expensive costs of dismissal. It also restricts hiring by limiting try-out periods and provides generous sick leave, vacations, and pregnancy provisions. Similarly, firms must contribute to union- sponsored health programs. Finally, firms have limited flexibility in the type of labor contracts used. The most common contract used is the lifetime contract, which enjoys the highest degree of protection. The 20 The EPH is a labor survey carried out three times a year in the main urban areas of Argentina. According to the survey about 420,000 individuals are owners of urban microenterprises. The EPH registers the owner's earnings, and thus is an indirect measure of the firm's value added. 21 This section is based on Guasch (2000) and Mondino and Montoya (2000). 32 system is also highly litigious with decisions contested, adding to the normal labor costs. That latter fact impacts very seriously SMEs since they seldom have the financial cushion to absorb those non-normal costs. It is also usual that a single court case Table 4.3 Structure of Labor Taxes can bring bankruptcy to a small and medium TomLbrTe firm. The reform of 1991 and 1995 introduced (% of gross wages) Sector National some flexibility at the margin with the . Agreement Average introduction of new types of fixed-term Workers' Contributions 20.1 17.0 contracts, which were legislated away since the Legal contributions 17.0 17.0 beginning of 1999. The new 2000 labor market Special contributions 3.1 reform was the most ambitious since it tried to Employers' Contributions 25.1 18.1 address particularly the key issues of highly Legal contributions 20.4 18.1 centralized collective bargaining agreements, Special contributions 4.7 and the damaging clause of ultractivity (any Total 45.2 35.1 agreement remains in place until a new one is Source: Guasch (2000, Table 12) signed by both parties) A/ The collective agreement provides for a variety of employee benefits that go beyond the mandatory ones. Labor market institutions present various shortcomings that are detrimental for job creation in the formal sector and for the ability of the system to respond to adverse shocks. The main obstacles, aside from the labor registry issue flagged above, that affect SMEs' ability to create formal jobs are: * A highly centralized collective bargaining system that hinders individual firms from adapting their contractual arrangements to geographical, sub-sector or firm-specific conditions. In spite of individual firms and workers having strong incentives to revise their contracts, the regulations make such revision illegal. Since all firms are treated equally, restrictions on renegotiation of contracts do the most damage to relatively smaller and more remote firms, and to workers with the least say in negotiations. As shown in Table 4.3, the collection bargaining agreement in the commerce sector, a sector crucial for employment creation, raises firms' labor costs by 10 percentage points over the national average. These surcharges over and above the required mandatory contributions increase labor costs and dissuade formalization of the labor force. On the side of the employees too, these contributions represent a sizeable deduction from their nominal salaries. The last labor law reform in 2000 has made some improvement to that framework but it does not go sufficiently far for it to be effective. * The prevailing practice of ultra-activity22that allows expired collective bargaining agreements to remain in force, even after their period has expired as long as a new one has not been signed. It is suppose to be phased out as the Ministry sets up a case-by-case schedule, but so far little progress has been made. * Costly severance payments regime riddled with legal uncertainties that increase costs significantly and that coexist with an unemployment insurance scheme that all together induces an 8 to 10 percent increase in labor costs, and much higher to SMEs. In short, labor regulations in Argentina are excessive, resulting in substantial costs for all firms, particularly SMEs. According to a recent study on the effect of labor regulation on employment decision in Argentina, tightening of labor regulations reduces the number of jobs while increasing hours worked of those workers that kept their jobs, which is the opposite effect sought by regulators.23 22 Ultra-activity makes collective bargain agreements enforceable even after expiration. 23 Mondino and Montoya (2000). 33 4.4 Logistics and Infrastructure Issues Cost related to logistics is amajor factor impacting the way of doing business in Argentina. Table 4.4 shows the excessive level of logistics costs in Argentina, that adversely impact firms' competitiveness. Table 4.5 illustrates the distorted procedures and taxation for the importing of semi finished goods, again impacting unit costs. As mentioned, these numbers are averages, and typically SME numbers are about 35 percent higher than the averages. That premium is driven by economies of scale, lumpiness, risk factors, volume discounts and internalization of transaction costs. In summary, then, logistics costs are a significant obstacle to the development of SMEs, reducing productivity, increasing unit costs and rendering a number of them- and of new activities, noncompetitive, particularly when considering markets abroad. An example of the scale or lumpiness effects is presented in Table 4.5 where the costs of importing physical capital are presented. It illustrates the high costs due to procedures levies post operation and taxes, it nears doubles the cost of the good -- a much higher value than in most LAC countries where the average value is near 150 -- and claims for revisiting the need and current levels. Table 4.4: Logistic Cost as % of Product Value per Country Country % of GDP Country % of GDP Peru 34.5 Taiwan 13.0 Argentina 29.6 Demark 12.8 Brazil 27.3 Portugal 12.7 Mexico 19.2 Canada 12.0 Ireland 14.2 Japan 11.3 Singapore 13.9 Holand 11.3 Hong Kong 13.7 Italy 11.2 Germany 13.0 England 106 U.S. 10.5 Source: FIEL (2001) and Bank staff Table 4.5: Direct Import Costs Concept . - P. Price FOB price 100.0 Freight (% FOB) 7.0% 7.0 Insurance (% C&F) 1.0% 1.1 Import tariff (CIF) 14.0% 15.1 Statistical rate (CIF) 0.3% 0.3 Taxable value 123.5 VAT 21.0% 25.9 VAT at 9.0% 11.1 Tax on Earnings 3.0% 3.7 Port Handling Costs (% CIF) 3.1% 3.3 Storage Costs (% CIF) 2.9% 3.0 Freight Forwarder (% CIF) 1.5% 1.6 Letter of Credit (% CIF) 1.5% 1.6 Damages and Losses 1.7% 1.7 Financial Costs 6.5% 6.5 Total 181.9 Source: Sgut (2000) al Withholding on VAT Base 34 4.5 The Consequences of a Complex Business Environment Among other factors, shortcomings of the regulatory framework for business formalization have important implications for private sector competitiveness and efficiency, particularly among SMEs. These include: * Business Informality. Not surprisingly, given the onerous compliance costs associated with business formalities, a significant share of SMEs' activities is performed outside the laws and regulations. The regulatory framework is biased towards smaller firms. First, the costs of registering a business, such Figure 4.4: Share of Informal Workers in as those documented here are much higher, Buenos Aires relative to the firm's value added for smaller firms. Second, the monthly costs of go1 M complying with tax and labor regulations so -- - ------------------------------ .. -. ........... amount to a staggering 3 to 7 percent of the TO so ...firm's value added for firms with less than I :: ' ' 25 workers. Thus, labor and tax regulations go40 .---------- are biased against smaller fimns. The amount mso0 . 11 X -.-----r 1 ............ -.that they must spend on bureaucratic red it 20 1--- l 0 --------; ---------- l tape is almost equivalent to what they must o [ | ;t _0 1 pay in income tax and the value added tax 1 2.5 *-25 1-lW 50U 0 sm. through the monotributo, and other local So e: sparbll(2000) 8-... S. (...b., W-k-..taxes. Thus, growth and development of SMEs are hindered. It is not only the individual firm that loses, but society as well, since the benefits, such as fiscal impact, social coverage, access to formal financial markets and access to the contract system, originate from compliance with laws and regulations (de Soto, 1989). Business formality can be viewed as a necessary input to production, which firms decide how much to purchase along a continuum (Levenson and Maloney, 1998). * Rise in Labor Informality. The high bureaucratic cost of'labor regulations may explain, among other factors, increasing labor informality. Argentina, for being a middle-income country, has one of the lowest compliance index on social security registration, not exceeding 50 percent. As shown in Figure 4.4, during 1980-1998 the share of informal salaried workers increased for all firm sizes, especially among smaller firms. This obviously has serious implications for the costs and coverage of social programs, especially health insurance, pensions, and income support during bouts of unemployment. * Reduction of Business Innovation. Although governments have good reasons to regulate economic activities, the Argentine cumbersome business regulation system imposes a constraint to business innovations, especially among businesses in the manufacturing and commerce sectors. For example, a cafe shop (negocio de gastronomia), which is an activity typically dominated by small businesses, will require municipal authorization to offer expresso coffee services, as well as to approve of the location of the coffee machine and auxiliary units. * Disincentives for geographical expansions. To sell products or services and to buy supplies from businesses in other provinces, a business must register with the Convenio Multilateral (Multilateral Agreement). The multilateral agreement is a mechanism for transferring the corresponding part of Gross Income Tax to other provinces when a company generates income from a sale or makes payment for a purchase of goods in a province other than the one in which it is located. * Uncertainties about Compliance and Enforcement. An excessive amount of red tape creates losses for society and is highly susceptible to the discretion of government authorities and corruption while leaving plenty of room for errors and creating judicial insecurity. For businesses with operations in 35 different municipalities; the three-tiered system of business formalities adds additional uncertainties during municipal inspections because definition criteria of economic activity vary widely across municipalities. Labor registration inspections are very ineffective due to lack of good information. Table 4.6: List of Steps (Trcmites) to Create and Register a Business I Firm constitution (e.g., statutes, etc.) and accounting study 2 Certification of the firm's official signatures 3 Payment of Stamp Tax at Banco Provincia de Buenos Aires (one percent of the firm's capital) 4 Payment for "urgent" formality at Banco Municipal La Plata 5 Present the firm's constitution documents to DPPJ 6 Publish the firm's registration act in Boletin Oficial de la Repubtica de Argentina 7 Deposit 25 % of the firm's capital in a public bank to show availability of capital (e.g., Banco de la Provincia de Buenos Aires) 8 Presentaci6n de documentaci6n en la DPPJ 9 Withdraw deposit of 25% of the firm's capital 10 Purchase of books required by law and seal books at Banco Municipal II Request seal of books required by law to DPPJ and pay stamp taxes 12 Purchase of book to register salary expenses and request sealed from Sub-secretary of Labor Tax Registration 13 Registration of shareholders in Registro en el Regimen de Trabajador Aut6nomo- AFIP 14 Registration of the firm in AFIP 15 Obtain temporary CUIT for the firm 16 Receive permanent CUIT from AFIP Cperational Permits and Licenses at the Municipal Level 17 Payment of stamp tax at Pago Banco Municipal to certify the floor plan of the establishment 18 Pedido autenticaci6n planos en Direcci6n Obras Particulares 19 Payment of stamp tax for a technical inspection by the fire department 20 Request technical inspection to the fire department 21 Purchase of fire extinguisher required by law 22 Get form to request authorization to use the business premises from the Secretary of Commerce and Industry of the municipality 23 Pay stamp tax at Banco Municipal para presentaci6n al IMPE 24 Solicitud de localizaci6n ante el IMPE 25 Request registration from the Secretary of Commerce and Industry of the municipality 26 Payment of fees for tasa de habilitaci6n, tax advances, and stamp tax 27 Deposit required documentation at the office of health 28 Deposit required documentation at the "Mesa de Entradas" Tax Requirements (municipal) 29 Registration with office of Impuesto sobre Ingresos Brutos - DPR 30 Payment of tax advances of Impuesto sobre Ingresos Brutos 31 Solicitud de formularios para inscripci6n de productos - Delegaci6n Municipal 32 Pago timbrado y derechos 33 Inscripci6n de productos - por unidad y analisis qulmicos y bacteriol6gicos 34 Inscripci6n en el Registro Industrial de la Naci6n Source: Fundes-BID (2000, Table 2, p. 13) Assumptions: Capital=25,000 and number of employees = 5 Enterprise producing fresh pasta IMPE: Instituto de Producci6n y Empleo. CUIT: Clave de Identificaci6n Tributaria. AFIP: Administracion Federal Ingresos Publicos. DPPJ: Direcci6n Personas Juridicas de la Provincia de Buenos Aires. DOP: Direcci6n de Obras Particulares * Inefficient use of government and firm resources. Argentina's cumbersome labor registration system imposes substantial costs on businesses. According to recent estimations, businesses spend about $700 million per year, representing 1.5 percent of overall salary costs (Giordano et al., 1999). For government agencies, the labor registration system entails duplication of activities and collection of information, and thus expenses. 36 * Reduced productivity and Increased Unit Costs. The increase transaction and operating costs due to that unfriendly business environment, with a large number of both fixed and variable costs-rule driven- adversely impact productivity and unit costs to all firms but particularly to SMiEs Reduced Number of New Economic Activities. The increased cost, direct and indirect here mentioned, renders a number of (new) activities, particularly for SMEs not financially viable, so they are not undertaken, with its obvious adverse impact on employment and income. Figure 4.5 Creating a Business in Argentina Is an Unduly Complex Process NO< e R - ~. Q~~fJ .LI~~ZI -LIZ ZZi-IR W l ! T A3,F ,, JP I 1 ,E'(3 Source: Giordano and Colina (2001)~da ___________ _ id3 ~ I I_..- 3 fE~-a-d1E wt S 019.. . '^ * M_4 d Source: Giordano and Colina (2001) 37 5. Financial Markets and SMEs In Argentina, among policymakers and economists, there is a wide perception that SMEs face severe financing shortages that limit their growth opportunities. Resolving the problems of both access to finance and the high cost of financing have become the main objective of many government programs. There is the perception that SMEs have more difficulty accessing external sources of finance than large businesses or other types of borrowers. According to a recent study, most SMEs state that limited access to external finance as an important barrier to growth (Fundes, 1999). Managers of SMEs often complain of not being able to borrow enough capital at reasonable rates (Batra and Stone, 2000). This chapter addresses whether or not financial markets are biased against SMEs. Comparing the contribution to output of SMEs and distribution of bank loans, we find severe imbalances and a need for policy action to deepen financial markets for SMEs. Small and medium enterprises receive, on average, about 9.8 percent of bank credit during 1998-2000, but they contribute to about 61 percent of Argentina's value added and to 69-78 percent of employment. Other countries present similar findings, such as Peru where SMEs receive 15.1.percent of bank's loans, while generating 42 percent of Peru's GDP. 5.1 Participation in Credit Markets Contrary to the widespread perception that SMEs lack access to finance, a recent survey shows that a substantial share of SMEs have credit. As shown in Figure 5.1, firms paying interest -our measure of access to finance hovered Figure 5.1 SMEs' Access to Finance around 58 percent in 1998 and l9 199S8* 1999 1 1999 among SMEs in the city of Buenos Aires. The l~ 9,100 manufacturing sector presents 81 - the highest rate of access to ------------------------- . - ----- ---------------- --- --- -- - 75 finance w hile the service sector -. 42 9 = , l 0 C O presents the lowest rate. As _ I - .. expected, access to finance is r - l - - - - . _ - - 25s ....... positively related to firm size. ,~ *fFirms with more than 50 _ _1_ _11.80 01.100 101.200 N.,.,0.d,,,1n0 ~ ~ TOW employees not only are more S -10 tl, O5 61 - i0 101-.200 t nublultn Sns". Comml- Tobl NuAbtreflmployees likely to have access to S__e__A__t__b_wed_on_Him_S____y_4__C__y_of__u__A_ finance, but also their access improved between 1998 and 1999. Smaller SMEs, with 6 to 50 employees, have lower rates of access and their access worsened in recent years. Available surveys with indicators on access to bank loans refer to industrial SMEs. The evidence shows that access to bank loans may not be an important constraint compared to micro enterprises and SMEs in other countries. According to various surveys, about 61 - 75 percent of industrial SMEs have received bank loans, compared to 54 percent of SMEs in the US (Fundes, 1999; Fiel, 1996; Berger and Udell, 1998, p. 635). Not surprising, larger firms, in terms of number of workers, present better access to bank loans. Fiel (1996) shows that about 91 percent of large firms, with more than 500 workers, have bank loans, compared to 77 percent of small and medium enterprises. 38 Using data from the 1999 survey of industrial SMEs,24 CEDES (2001) finds that a large share of SMEs have a demand for loans (81.2 percent), including those firms that applied for bank loans (70.2 percent) and 1 percent of firms that auto-exclude themselves out of the market and decide not to apply for loans. The most frequent reason for not requesting bank loans is high level of interest rates (14.4 percent). Some SMEs perceive borrowing as a risky endeavor (10 percent), perhaps due the volatility of their earnings, and the fear of losing their collateral. Finally, a small group of SME is too leveraged to incur additional borrowing (2 percent) (Figure 5.2). Overall, about one in four SMEs with a demand for bank loans can be classified as credit constrained, having an unsatisfied Figure 5.2 Access to Bank Loans among Industrial SMEs in 1998 demand for loans. Those industrial SMEs that had RBcelvod bank loans _ _ _ _ __ _ __ _ _l 61 4 their loan application RBJected 8.9 rejected (about 8.9 percent) Hlgh cost of credit 14 4 are clearly credit High leverage 11 1.7 constrained. Under some High leverage 1.7 assumptions the group of Bank debt [l too risky 9.9 auto-excluded SMEs can be Does not meet bank's requirements 11.0 considered as credit 10 20 30 40 5- s 70 constrained, having a %ofindusbialSMEs demand for current loan Source: CEDES (2001, Table 6) products, but knowing that banks will reject their loan applications, they do not bother to apply. Using data, the CEDES study Figure 5.3 Financial Structure of SMEs analyzed the probability of (M Equity * Liabilities 0 Bank debt receiving a bank loan using a 'IO ._____ ___ _ ______ _ probit model. The empirical 7e _____________ _ results using micro data from a so _ | | _ _ = survey of industrial SMEs 4 o0 - 1| | _ _ r reveal that the SMEs more 30 - likely to receive bank loans 20 _ __ __ _ __ _ __ _|___ _ ___are: more profitable in terms of o - ! z l return on assets, less liquid, Algenena-94 Argenina48 Arntins4il US4J Espah4a-3 Iblia49 Japon4 AI3 Srure l ft-. - r m VUrg.(iss7)ULnA(rs9s) Furld(1ss9} arldBoruala U(1s). ThadrWAq0 and mTre leveraged. I994 tsssared1sssmawruberompartub Interesting, the length of banking relationship does not affect the accessibility to loans. 5.2 Capital Structure Although industrial SMEs seem to enjoy a good degree of access to bank loans, these loans amount to a small share of total assets. Compared to other countries, Argentine SME sector is largely self-financed (Figure 5.3). The largest source of finance for Argentine small businesses is equity finance, which accounts for 53 to 74 percent of small businesses' assets. This may not be surprising as in bank-based economies such as Argentina, debt-equity ratios, even among large firms, are consistently lower than in 24 The Argentine Industrial Association carries out a yearly survey of industrial SMEs. 39 market-based economies (Schmukler and Vesperoni, 2000). By relying on mostly internal sources of finance, the investment behavior of SMEs may be adversely affected. Many SMEs expand their working capital and investment needs on the basis of past profits, subjecting them to economic fluctuations. Among industrial SMEs, accumulated profits finance about 50 percent of expenses in equipment and machinery acquisitions, 20 percent by bank loans, and 18.5 percent by suppliers (UIA, 1998). 5.3 Bank Lending to Small and Medium Enterprises In recent years, the Argentine financial system has undergone a number of changes. In particular, the Tequila crisis of 1994-95 sparked a banking crisis that accelerated the transformation of the Argentine financial system. Policies to deal with the banking crisis included: privatization of about half of all Table 5.1: Number of Financial Intermediaries provincial banks, liquidation and/or Type of Intermediary Dec-96 Dec-97 Dec-98 Dec-99 Sep-00 consolidation of unsound Banks 120 113 102 92 90 entities, bank Public banks 22 20 16 16 15 restructuring, and Federal banks 3 3 3 2 2 tightening of regulation Provincial & 16 15 13 14 13 Bank, 1998; Dages, municipal banks Bn,19;Dgs municipal bank93 Goldberg, and Kinney, Private banks 98 93 86 76 752000; Clarke and Domestic 62 55 43 36 33Cull,1999). Although Cooperative banks 8 6 4 2 2 credit to private sector Foreign banks 28 32 39 38 has grown, reaching 23 Non-bank financial 27 25 23 24 24 percent of GDP by June entities 2000, from 20 percent in Financial companies 18 16 15 17 18 1995,25 these levels are Credit funds 9 9 8 7 6 very low compared to - other countries in the Total of financial 147 138 125 116 114 region.26 Mostly through mergers and acquisitions, Source: Argentine Central Bank and ADEBA (Association of Argentine foreign banks have Banks) played an important role in this process. By August 2000, 52 percent of all financial sector assets were under foreign owned banks, up from 20 percent in end-1994 (BankWatch, 2000). The restructuring process reduced the number of financial entities, dropping from 206 in 1994 to 114 in September 2000, and increased bank concentration. Despite the consolidation process, there are still a large number of very small banks in Argentina. Since 1998, there have been 18 merger or acquisitions, ten of which involved foreign banks. Although these changes have strengthened the banking system, the geographical coverage of the banking sector has weakened. Argentina experienced a reduction in the number of financial outlets from 4,286 to 4,077 between 1994 and 1997, but in recent years banks have expanded their distribution network reaching _ outlets by March 2001 (World Bank, 1998, p.19). The geographical distribution of bank outlets is largely concentrated around the metropolitan area of Buenos Aires. For example, the Federal 25 The combination of foreign bank entry and the privatization of banks seem the driving force behind this growth (World Bank, 1998). 26 Credit to private sector amount to 38 percent, 68 percent, and 59 percent of GDP in Brazil, Chile, and Bolivia, respectively. 40 Capital and the Province of Buenos Aires and the provinces of C6rdoba, Santa Fe, and Mendoza together account for, respectively, over 53 percent and about 25 percent of bank outlets in 1998. The regions are home to 46 percent and 21 percent of the total population, showing a biased distribution. The concentration of bank outlets is paired with concentration of loans. Whereas the Metropolitan area of Buenos Aires presents a loan to GDP rate of about 83 percent by September 1998, other regions, including the manufacturing provinces of C6rdoba, Santa Fe, and Mendoza have an average ratio of less than 20 percent (IAMC, 1999, p. 67). These findings suggest that Argentine banks seem to follow a low- risk lending strategy towards economically dense regions. 5.3.1 What are the Determinants of SME Lending by Banks?27 SME lending is not a main line of business for Argentina banks. Overall, SME lending accounted for around 20.9 percent of banks' loan portfolio in 2000, a decline from 23.6 percent in 1998, partly explained by the economic recession (Table 5.2). Although the volume of lending to this sector declined 4.5 percent during 1998-2000, it did so at a lower rate than loans to large firms (5.8 percent). This may raise concerns among supervisory authorities as SME lending is riskier than other types of lending. Lastly, loans to the non-financial public sector grew about 125 percent over the same period. Table 5.2: Commercial Debt Balances with Banks by Firm Size: 1998-2000 Firm Size Jun-98 Jun-99 % Jun-00 Micro v 862 1.8 858 1.7 873 1.7 Small and Medium 11,536 23.6 11,117 21.5 11,017 20.9 Small b/ 4,905 10.0 4,820 9.3 4,875 9.2 Medium c 6,631 13.6 6,297 12.2 6,142 11.6 Large d/ 31,535 64.5 32,050 62.1 29,720 56.3 Public Sector 4,930 10.1 7,588 14.7 11,139 21.1 Total 48,863 100.0 51,613 100.0 52,749 100.0 Source: BCRA (2000, Table 4.23) v total debt less than $50,000 b/ total debt between $50,001 and $500,000 C/ total debt between $500,000 and $2.5 million d/ total debt above $2.5 million 27 This section is based on Clarke, Cull, Martinez-Peria, and Sanchez (2001). 41 Figure 5.4: Estimated Share of Lending to SMEs and Growth Rates of SME Lending by Bank Size for Foreign and Domestic banks in Argentina 70% 60% -------- 50%-< 40% [- . - ------ - -- - -: ,- -~r~- -- -- -- -- -- -*----I.. .. ..--..----.. ......................-. ; .. ... .. ------ 30% .... ..... ... ........ . . ...... .... .. . --- ... ... ... ... . ..... ......... l0% 20% - ^ a .^-.#r,.;> :r -,.i 10% . \ 0%. 0 3000 6000 9000 (millions of pesos) -| - Domestic (non-linear + interaction) - Foreign (non-linear + interaction) Estimated share of lending to SMEs 40% 310% - - - -- - - -- - - -- - - - -- -0-- - - -- - - - --- -- - -- 00- -- - -- - -------0 0 -- ---0 ------- 0 30% .. 50% . -10% -4-iO0 -2000) - ----- 3000- - --- 4000- - -.--6000S - -- ~-6{000 - --- -7000- - - -8100 _ *_ _ !90 00O -20% .- --- ------ -...... --- ------ . . . . ---..-----.v... -30% - -- -- ----- - ^ -40% __... -50% -60% -70% (millions of pesos) -Domestically Owned Banks - Foreign-owned Banks I Estimated growth of lending to SMEs Source Clarke et al (2001). Note Estimated shares using regression coefficients shown in Table 5.3 and median values of continuous variables for each bank type Estimates are for the final period. 42 Embedded within a larger process of trade and financial liberalization, many Latin American countries have removed the barriers to direct foreign participation in the banking sector. In Latin America, excluding latecomers Brazil and Mexico, foreign control of banks rose from 13 percent to 45 percent of bank assets. Whether or not increasing foreign bank participation will be beneficial for small business lending is an issue of discussion in many countries. Argentina is a case in point. Bleger and Rozenwurcel (2000) indicate that foreign bank participation in Argentina is associated with a reduction of bank lending to SMEs from around 20 to 16 percent of total lending between 1996 and 1998. In contrast, a recent study conducted by the Argentine Central Bank finds that despite their lower propensity to lend to SMEs, foreign banks have increased both their propensity and their market share of lending to the sector between 1998 and 2000. Using a rich data set on Argentinean business debtors in December 1998, Berger, Klapper, and Udell (2000) find that large banks and foreign owned banks are less inclined to extend credit to smaller firms, which are more likely to be informationally opaque. While Berger and others do not explore dynamic issues, the remaining studies base their conclusions on cursory examination of bank level information on small business lending rather than on careful econometric analysis. Using bank level data, Clarke, Cull, Martinez-Peria, and Sanchez (2001) examine the impact of foreign bank entry on the share and growth rate of lending to small businesses, while controlling for other factors in Argentina, Chile, Colombia, and Peru. In the regression analysis, Clarke et al. include as explanatory variables bank size indicators, financial health and performance indicators (return on assets and the ratio of administrative expenses to total assets) and bank ownership indicators while controlling for mergers and acquisitions.28 Consistent with evidence from the U.S., the authors find that bank size has a negative, but decreasing impact on bank lending to small businesses (as a share of total lending), suggesting that small business lending is a small bank business. As shown in Figure 5.4, on average, foreign banks in Argentina lent less to small business (as share of total lending) than private domestic banks (at least by the end of period). Because a decline in. the share of small business lending could arise from small business loans increasing or decreasing at a slightly slower rate than all types of lending, Clarke et al. also analyze the impact of foreign bank entry in the real growth rate of small business lending. As shown in Table 5.3, they find that small business lending by domestically owned banks appears to be growing faster than lending by foreign-owned banks in Argentina. Moreover, the authors find that bank size is positively correlated with growth of small business lending with a large effect among foreign banks. This indicates that small business lending was growing faster for large foreign banks than for small foreign banks (see Figure 5.4). Lastly, one of the most interesting findings of the authors is that Argentine public banks have not come to the rescue of this sector and do not appear to surpass private banks in the extent to which they lend to small businesses. Thus, arguments saying that eliminating government resources invested in public banks or privatizing public banks would hurt small business lending do not have strong empirical evidence. 2S The authors control for three kinds of foreign banks: those that have been operating in the country for a number of years before the period of analysis, de novo foreign banks that start operating at some point in the sample, and banks that within the sample acquired other domestic and foreign banks. 43 Overall, there is evidence that Argentine banks are changing their attitude towards SME lending (Bleger, 2000). Lured by higher margins, a large market, and the possibility of offering other financial services to SMEs and their employees,2 some private banks had adopted measures to increase SME lending, such as offering specific lines of credit for SMEs, and creating specialized SME departments and organizations to promote SME development. For example, the Association of Banks (ABPPRA) cooperates with SEPyME to offer training to banks to foster SME lending. Given the current economic crisis, Given the This phenomenon is not unique to Argentina, which is considered a latecomer to the SME lending arena. Chile and Bolivia present evidence of the entrance of banks intro micro-enterprise and SME lending (Christen, 2000). 29 Banco Rio de la Plata plans to increase SME lending precisely because of the possibility of also offering financial services to SMEs' workers, exploiting economies of scope. BankBoston has a special department that offer credit products marketed for SMEs. 44 Table 5.3: Determinants of SME Lending among Argentine Banks Share of SME Lending [P/(1-P)] Growth of SME Loans Explanatory Variables Coeff. t-stat Coeff. t-stat Coeff. t-stat Coeff. t-stat Bank Size Log of real assets -0.115 (-4.73) *** -2.276 (-5.29) * 0 .056 (5.81) 0* .020 (1.73) * Log of real assets)2 .080 (5.19) * Log of real assets)*(Foreign bank) .156 (3.22) ** .108 (5.36) *** kLog of real assets)2.(Foreign bank) _0.079 (-3.24) ** Financial Indicators Retum on assets 0.026 (-0.61) 0.038 (-0.94) 0.151 (2.84) *** 0.131 (2.42) ** Administrative expenses/assets . 0.422 -2.79 0* .354 (2.54) ** 0.104 (1.81) * 0.098 (1.72) * Bank Ownership and Time Trend Public bank dummy -0.606 (-8.21) *** 0.652 (-8.31) 0 .004 (0.1) 0.048 (1.23) Time trend .063 (0.96) .092 (1.43) Foreignbankdummy 0.129 (-0.86) 14.611 (-3.2) 0 .131 (-3.46) *** 1.563 (-5.61) *** (Time trend)*(Foreign bank) 0.149 (.1.42) -0.192 (-1.74) Mergers & Acquisitions Domestic acquisitions dummy -0.265 (-2.66) -0.142 (-1.3) -0.206 (-3.95) - 0.188 (-4.01) '* Domestic acquisitions* Age .255 -3.11 0 .053 (0.54) Foreign M&A dummy 0.200 (-2.61) - 0.250 (-2.37) -0.037 (-0.72) -0.057 (-1.51) ForeignM&A*Age .111 (1.01) .152 (1.15) De novo foreign entry dummy 8.417 (-69.33) *** 8.327 (-35.65) ' De novo foreign entry)*Age 1.138 (-3.64) . 1.004 (-3.48) * Foreign purchase dummy 1.212 (8.97) **+ 1.220 (8.96) ** 0.095 (-2.74) ** 0.084 (-2.53) ** Foreign purchase*Age .373 (1.63) .393 (1.67) * Constant -0.018 (-0.05) 14.407 (4.86) - 0.699 (-5.68) -0.259 (-1.74) * $ of observations 1388 1388 548 548 R-squared 0.32 0.34 0.07 0.10 *,**,*** significant at 10,5, and 1% P = proportion of loan portfolio directed to SMEs Source: Clarke, Cull, Martinez-Peria, and Sanchez (2001, Tables 3-5). 5.3.2 What Explains the High Interest Rates of SME Loans? Argentine SMEs cited as important constraints the high level of interest rates of bank loans. Several factors may explain the high levels of interest rates paid by SMEs: * Credit Ouality. Differences in credit quality may explain why SMEs pay higher interest rates on bank loans. According to the BCRA study, SME loans present substantial higher levels of non-performing loans than large firm loans (Table 5.4).3° Since the onset of a slowdown in economic activity in 1998, banks have witnessed a surge in the ratio of non-performing loans to total gross loans in the private sector from 19 percent to 25 percent. The quality of SME loans has deteriorated less than that of large firms, suggesting that only higher quality SMEs are getting loans. The ratio of non-performing loans to gross loans for SMEs, which was three times higher than that of larger firms (37 percent versus 12 percent) in June 1998, decreased to 1.7 times by June 2000. 30 Non-performing loans include irregular and non-recoverable loans. 45 Table 5.4: Non-performing Loans by Type of Firms (%) Large Firms SMEs Jun-98 12 36.4 Jun-99 13 41.4 Jun-00 28 47.2 Source: BCRA (2000, Table 24.c) Includes: irregular and non-recoverable loans * Type of Products. When SMEs borrow from banks it is via credit cards and overdraft current accounts that carry interest rates much higher than rates of other types of loans, ceteris paribus (Sepyme's web p). In 1999, about 42 percent of SMEs with access to the banking system were using overdrafts in their current account (decubiertos). Moreover, the UIA's survey of 2000 found that about 50 percent of industrial SMEs use overdrafts, paying on average an interest rate of 2.7 percent per month or 38 percent annually.3' * Administrative Costs. The transaction cost of lending to SMvEs is high in relation to the size of loans, making such lending in many cases unprofitable. Indeed, it cost about US$ -?_ to prepare a credit dossier (IAMC, 1999, p.48). The lack of documentation available makes it difficult to obtain a clear picture on the financial situation and prospective success of such businesses. Most banks use lending technologies based on standardized credit policies and standard information. Informational distance contributes to increase the costs of generating borrower-specific information. * Collateral. Collateral may reduce the cost of intermediation. Banks may find it easier to assess collateral value than to evaluate project creditworthiness and in case of default, expected losses may be lower.32 Various studies have found that SMEs' loans present a higher fraction secured by collateral. This seems to indicate that the legal framework for secured transactions present serious shortcomings. Using a database of business debtors as of October 2000, a team of researchers from CEMA and the BCRA assessed the determinants of the interest rate paid by large firms on their bank overdraft account, which averaged 3.1 percent monthly.33 After controlling for various explanatory variables, the econometric results show that firms that pay lower interest rates are larger, in terms of assets, have better ratio of sales over assets, and have a lower leverage ratio. Moreover, firms with poor credit quality and with loans with lower guarantees pay higher interest rates. 5.3.3 Relationship Lending Given the paucity of information about SMEs, creditor-borrower relationships may enable banks to generate information on the risk characteristics of small firms. A number of theoretical and empirical studies have sought to assess the importance of borrower-lending relationships for the terms and availability of small business lending. Among theoretical models there is little consensus about the impact of closer relationships on the cost of bank loans. On one hand, the bank capture effect predicts that due to the emergence of informational monopoly, price and non-price credit conditions will not improve over time. Banks would appropriate some rents because customers may be "informationally captured" (Sharpe, 1990, Rajan, 1992, Fisher 1990, Greenbaum et al 1989, Mayer 1988). On the other hand, the information production effect predicts that decreasing interest rates over time. Firms with long-lasting lending 31 There are little differences between firm size 32 BCRA regulation recognizes these features of collateral use, requiring lower provisioning rates. 33 CEMA and BCRA (2001). 46 relationships and/or concentrated borrowing patterns will incur lower costs of capital, and/or will have better access to external finance, including lower collateral requirements. Empirical evidence from the US, Germany, and the UK shows that firms maintaining long-standing relationships with banks have to pledge collateral less frequently (Berger and Udell, 1995; Harhoff and Korting, 1998; Cowling, 1999). However, the effect bank-borrower relationships have on interest rates is less conclusive. In the US, Peterson and Rajan (1994) find that the firm-lender relationship is not correlated with small business loan rates, while Berger and Udell (1995) present evidence of the information production effect. In Italy, DAuria and Foglia (1997) and Angelini, Salvo,.and Ferri (1998) find that lending rates tend to increase with the length of the relationship, presenting evidence of the bank capture effect. In the case of Argentina, the empirical evidence is mixed. Powell, Broda, and Burdisso (1997) analyze the peculiarly huge variation in interest rates across banks and regions, showing that banks develop relationships with specific types of clients who have poor information, such as SMEs and firms outside the metropolitan area of Buenos Aires. Because information generated in those relationships can not be easily transferred to other creditors, the study finds that those clients become captured and are charged monopolistic interest rates by banks. Interest rates present huge variations across banks and types of customers, but little variation over time and across lending instrumen'ts. CEMA and BCRA (2001) examine the impact of bank-borrower relationship on the interest rate paid on overdrafts accounts of large firms. The following variables measure the degree of bank-borrower relationships: (a) the share of outstanding debt with the bank institution, (b) the number of banks with which the firmn has loans, and (c) the average number of bank services the firmn is using: The economic results show that firms with better relationships with banks enjoy lower interest'rates after controlling for other variables, providing support to the positive effect of information production generated through bank-borrower relationships. 5.3.4 Borrowing Concentration The concentration of the firm's borrowing across lenders, measured by the number of institutions from which the firm borrows, is an indicator of the closeness of a borrower to its lenders. SMEs tend to have exclusive dealings with a single bank with which they have a strong relationship. Exclusive dealings may be an indication of market power, or limited competition in credit markets. Multiple banking relationships may result from: i) borrowers' efforts to avoid exploitation, ii) credit rationing practices by banks, and iii) credit crunches. Few studies examine borrowing concentration by SMEs in Argentina, illustrating that most small firms have exclusive borrowing relations with banks and that larger firms borrow from various banks. According to Fiel survey, on average, SMEs tend to have 2.9 differen't banks with the smallest firms having just one lender while the largest firms having 14 lenders. The Central Bank of Argentina also finds a positive relationship between firm size and the number of banks a firm borrow from. By June 2000, large firms had loans from 7.6 banks, compared to small and medium sized firms, which had 3.2 banks. 47 Table 5.5: Borrowing Concentration among Debtors In 1998 Borrowing Concentration Firm Size No. Firms Multiple Single banks % Bank % All firns 61,295 26,939 44.0 34,356 56.1 Bottom 50 % (Loans < $33,600) 30,657 6,521 21.3 24,136 78.7 Top 50 % (Loans > $33,600) 30,638 20,418 66.6 10,220 33.4 Source: Berger, Kappler, and Udell (2000, Table 3 and Table 4) Note: The set of firms excludes about 19,500 firms that have a state-owned banks as their primary bank. a/ Each entry shows the number of finns that have at least one loan from a foreign bank, a large bank, or a distressed bank. Larger firms are more likely to borrow from multiple sources. About 49 percent of SME work with a single bank while the largest firms this proportion drops to 22 percent (Fiel, 1996, p.l83). Looking at the December 1998 data of business debtors, Berger, Kappler, and Udell (2000) find that about 56 percent of firms have a single bank, compared to 33 percent of the top 50 percent (Table 5.5). Beside firm size, another factor that may explain borrowing concentration is macroeconomic conditions. Figure 5.5: Evaluation of Credit Services by Exporting 5.3.5 Financing Exports SMEs When exporting most SMEs operate with banks, but only 14 percent of 100 4 soo% exports were financed with bank credit --8-0" eo% . _ (Fundes, 2001). Handicaps in credit 60% markets hinder SMEs' ability to deep - 40% L L L their integration in international 20% markets. As reported by Fundes (2001) few SMEs use government programs, . . * Pn sing Cnd"Unft Colatsml InrnmaRaet suggesting room for improving. -lIdshadtOMy - 38% '>2 6%- . 3% 19% ;ffamZ66i 364% 6f%-.. -64% 81% 5.4 Leasing and Factoring S @,X~Id~'l2OOi. . _ -! ; _ Leasing may become an important instrument for SME lending for the following reasons. Leasing offers many advantages over traditional loans: encourages investments, broadens competition in financial services, expands access to financing for capital equipment and new technology, and can finance a higher percentage of the capital cost of equipment than bank borrowing. For all these reasons, governments all over the world have started promoting leasing with tax incentives and other instruments, considering also that the development of the leasing industry can help in broadening competition in financial services and in stimulating financial innovations. In Argentina, the leasing industry has been growing very rapidly. In 1997, leasing contracts reached $350 million, but by 1999 this amount had almost tripled to $970 million. With the new leasing law, govemment authorities expect that the leasing industry will continue its growing trend. 5.5 Risk Capital Investment- Private Equity Investments Although a limited number of SMEs may benefit from risk capital investment, it is important to present a description of current activities. Not all equity funds operating in Argentina target small companies. Risk capital investment or equity financing is a form of financial intermediation through which companies receive financing primarily from investment funds, which is turn are financed by investors seeking high returns from their investments. According to recent studies, only eight equity investment funds target SMEs reaching $169 millions with $50 invested. These funds have different minimum investment 48 requirements, preference for economic sectors, and internal rate of returns. On average, these funds plan to invest for a period of five years. Box 5.1 Main Shortcomings of the Legal Framework for Risk Capital Investment in Argentina In Argentina, the main shortcomings of the legal framework for risk capital investment are related to evaluation/due diligence, structure of investments, management of investments, and exit strategies. Evaluation/Due Diligence. The lack of accounting conformity makes it difficult to carry out due diligence and valuation. Moreover, equity investors find it difficult to assess contingent liabilities by local companies in the context of acquisitions. Moreover, there are uncertainties regarding the treated of tax audits. Structuring. Inflexible corporate laws limits investors' ability to create certain types of securities or instruments mechanisms commonly used in private equity transactions to increase or restrict their equity stake in a target company based on future performance or corporate events. For example, under restricted circumstances preemptive rights can be waived in Argentina. Moreover, companies may redeem their share under limited circumstances, subjected to local courts' interpretations, adding another layer of uncertainties. Another limitation to structure risk capital investment is the prohibition of stock option plans under corporate law. In Argentina, the dual concept of issued share capital and authorized share capital does not exist: all share capital must be paid for when issued and any issuance of shares must be preceded by a share capital duly approved by the valid vote of the company's shareholders. Investors have used off-shore mechanisms to deal with these structuring problems, raising the costs of investments. Managing Investments. Once investments are realized, investors who take minority positions have a hard time getting their rights protected. This induces investors to take majority positions, which requires more time and efforts to manage the companies. Moreover, Argentine corporate law restricts the ability of companies to buy back their own shares under limited circumstances. These include situations where redemption is done to cancel shares prior to a reduction in share capital or where it avoids a great harm to the company, although what constitutes a great harm may not be consistently applied by the local court system. Prohibitions against buying back shares, all other things being equal, affect SMEs more because in may instances this may be the only viable exit mechanism for investors. Exiting Investments. Risk capital investors face greater uncertainties with respect to the enforceability of rights associated with the three main exit strategies: IPOs, trade sales, and exercise of put options. Demand registration, piggy back, tag-along and drag-along rights and put options are the legal and contractual mechanisms associated with those exit strategies. Although the Argentine statute does not contain statutory prohibitions against them, investors are uncertain about potential problems with the enforceability of those rights. These legal uncertainties and the reliance on trade (strategic) sales as exit strategy have resulted in adverse consequences for SMEs. Fund managers have the incentive to evaluate companies for their future strategic value for strategic buyers, focusing in a narrow list of companies at the expense of SMEs. Source: Mayoral Hemandez and Baptista (2001). Inadequate laws and regulatory regimes have had the unintended consequence of detracting risk capital investment, especially for SMEs (see Box 5.1). Nonetheless, investors have relied on the use of offshore investment vehicles to facilitate foreign risk capital investment and the creation of voting trusts to overcome regulatory hurdles. As a result, these shortcomings increase transaction costs for evaluating, structuring, managing, and exiting equity investment operations. Because investment size and firm size are related, these higher transaction costs have a disproportionately negative effect upon investments in SMEs. To encourage risk capital flow, especially into SMEs, government authorities should examine closely the need to change their legal frameworks. Limitations in legal regimes raise the risk profile for investors. Potential solutions to address the problem areas include: the revision of existing laws, the adoption of new laws, the codification of contractual rights, expedition of the judicial process to enforce exit rights, and 49 the creation of specialized administrative agencies as an alternative to court resolution of some shareholders' disputes. 5.6 Reciprocal Guarantee Societies The reciprocal guarantee societies (RGS Sociedades de Garantia Reciprocas)' constitute another instrument to increase SME access to bank loans. Based under the assumption that the main impediments of access to bank credit are high interest rates and lack of collateral, the Argentine Government promulgated the legal framework for RGS in 1995. The RGSs provide guarantee certificates to improve the creditworthiness of SMEs when requesting loans from banks. In addition, they offer advisory and technical assistance to its members, and foster partnerships between large enterprises and SMEs. The philosophy behind this system is that closer links between the interest groups in SME-dominated industries, such as the craft and retail sectors, can help improve banks' knowledge of a company's standing and of the markets in which it operates.. This in turn alloWs a better ex ante evaluation of the project by the lender and.allows credit risk to be considerably reduced. The fact that the mutual guarantee organizations trust the enterprises and their projects, and express this trust by offering the lending bank a partial guarantee, makes it easier for the latter to grant loans. Recently, the International Finance Corporation initiated a pilot project to finance SMEs using "closed" RGS to screen potential borrowers. A closed RGS groups a large firm like Edenor (Empresa Distribuidora y Comercializadora Norte, S.A) and its suppliers, offsetting adverse incentives for strategic default. Regulation of Reciprocal Guarantee Societies To open a RGS several requirements need to be satisfied. The RGSs are required to have minimum variable capital of $240,000 and 120 partners (about $1,020 per SME). The SEPyME supervises the RGS, keeping a registry (InfoPyMEs, 1999). They have two types of partners: (a) protector partners (socios protectores) that can not receive, credit guarantees and can provide up to 49 percent of the total RGS's capital; and (b) participant partners (socios participes) that can own up to 5 percent of capital. To cover guarantee granted to participant partners, the RGSs are required to create a risk fund (fondo de riesgo). Contributions from protector partners and interest income from the investments of the fund will contribute to resources to the risk fund. To ensure the soundness of the-RGS system, various prudential norms are considered in the law. First, a RGS should create reserves with a minimum value of 20 percent of its social capital. Second, a single partner can not receive in guarantees more than 2.5 percent of the total value of the risk fund (fondo de riesgo). According to the regulation, the risk fund should equal at least 25 percent of total guarantees offered, implying that a single partner can not receive more than 0.625 percent of total guarantees offered. This is a significant reduction from the initial requirement of 5 percent. Third, a single creditor can not receive more than 20 percent of the total guarantees offered. Fourth, to reduce the variability of RGSs' capital, protectors partners should maintain their capital contribution for at least 2 years and face some restrictions to withdraw their funds. Fourth, partners are prohibited from belonging to more than one RGS. Despite a special legal framework and tax exemptions, only five RGSs have been created, having a limited outreach (Farhi, 2000). Table 5.6 presents summary statistics of the RGS system. Garantizar, which operates under the wing of Banco Nacion, is the largest RGS with 757 members. As of September 1999, about 465 credit guarantees have been issued to SMEs, even though some government programs finance SMEs' contributions to RGS! This may not be surprising in light of the above-mentioned studies on the lack of dense inter-firm and inter-organizational networks. Moreover, financial incentives to participate in a guarantee program are limited. The evidence from Argentina, and even advanced industrialized countries, shows that such incentives are insufficient to encourage learning. If the state 50 sticks with financial guarantees (bigger tax breaks or guarantees) it runs the risk of increasing its exposure via a higher loan-loss rate. The lowering of this risk depends on the public institution being well embedded at the local level to create effective decentralized, monitoring mechanisms of public or private delegates to manage the guarantees and project evaluations. Table 5.6: Sociedades de Garantia Reciprocas (September 1999) Fund Guarantees Number of S.G.R. Partners (S millions) (S millions) Guarantees Operating Since GARANTIZAR (BNA) 757 18.8 15.5 77 Sep-97 Compaflia Afianzadora de 232 40.0 158.2 361 Aug-97 Empresas Siderirgicas (CAES) AVALUAR 124 30.7 10.5 24 Sep-98 MACROAVAL 112 4.6 0.2 1 Dec-98 AFIANZAR Total 1,225 94.1 184.4 463 Source: IERAL (1999, p.132) and Plata (1999) 5.7 Under-developed Financial Infrastructure 5.7.1 Information Problems Economist theorists have long recognized the importance of information in the operations of credit markets. Information asymmetries between lenders and borrowers lead to problems of moral hazard and adverse selection (Stiglitz and Weiss, 1981). Information problems are particularly acute for SMEs. The most important characteristics defining small business is informational opacity or lack of credible information (Berger and Udell, 1998). Many of the smallest firms do not have audited financial statements that can be used by banks. Moreover, they maintain parallel accounting books to circumvent govermment regulations, rendering audited financial statements useless to assess creditworthiness.34 For some banks, knowledge about a client is more important for lending decisions than cash flow analysis from financial statements. Hence, lack of detailed micro data on small businesses is likely a major reason why banks have not developed large portfolios, as SMEs often cannot credibly convey their risk quality to banks without offering collateral. Credit information registries can reduce the extent of information asymmetries in credit markets, thus improving the allocation of resources. By making a borrower' credit history available to all lenders, they allow lenders to discern between good and bad risks, reducing adverse selection problems. Moreover, credit registries may provide additional incentives for timely repayment because they allow borrowers to create a reputation in credit markets as good credit risk. The ability of SMEs and his owners to build "reputation collateral" may increase their participation in credit markets by reducing screening costs for lenders as well as the borrowers' costs of signaling their creditworthiness to potential lenders. Considering that Argentine law does not prevent banks from disclosing information about their borrowers, Argentina's Central Bank embarked in a process of data collection and dissemination of information about each bank debtor since 1994 (World Bank, 1998). What are the main problems of the Argentina credit information registries? The evolution of credit registries in Argentine has resulted in fragmented consumer or firm credit histories. Private credit 34 The government is debating whether to simplify the tax regime to reduce tax evasion, allowing SMEs to present more sound financial statements to potential creditors. 51 information registries exist, but the quality of data is uneven. Because many of them operate at the provincial level, data on borrowers' credit history is fragmented among various agencies. This fragmentation increases transaction costs of creditors to enter the market, raise the costs of delivering credit, and ultimately restricts the number of consumers who will receive loans and the amounts they borrow (Barron and Staten, 2000). The wide access of the Central de Deudores raised privacy concerns, prompting the enactment of a privacy law, to become effective on or before April 30, 2001.35 Argentina's privacy law covers the collection and disclosure of personally identifiable information. According to the law, credit information registries are permitted to retain information for a maximum of five years, but for not more than two years after a debtor has settled an obligation. This fragments borrowers' credit histories, reducing the quality of information for credit scoring purposes. The law regulates the credit information registry in Argentina, but the actual enforcement of the law remains uncertain. 5.7.2 Regulation Problems To promote lending to small businesses, the Argentine Central Bank allows banks to treat commercial loans up to $200,000 as personal loans for purposes of information and provisioning requirements, simplifying the approval process for banks. Loans in this category amounted to six million dollars in June 2001. Some aspects of the banking regulation have an impact on the cost of lending, and thus may reduce access to loans, especially by SMEs. They include risk classification of loans, provisioning requirements, and minimum capital requirements. * Risk classification criteria differ between commercial and personal loans. For personal loans, the risk classification depends just on the repayment performance of the borrower, while for commercial loans, it depends also on an analysis of the financial performance and business prospects of the borrower. The regulation requires banks to classify their loans in one of five risk categories. Category I (normal) has the least risk, and category 5 (uncollectible) has the most risk. * Provisioning requirements are equal for commercial and personal loans, and loans with preferred guarantees carry lower provisioning rates. These requirements have a substantial effect on break-even interest rates for loans classified in category 3 (problematic) or higher. For example, the break-even interest rate for a secured (unsecured) loan that moves from normal to problematic doubles (triples), mostly because provisioning requirements increase from one percent to 12 (25) percent (Schreiner and Colombet, 2000). Because SME loans present a worse performance than other loans, this may explain why banks charge higher interest rates on SME loans. Moreover, these norms also have perverse incentives on banks, which may be very cautious on assigning a risk classification of 3 because of the higher costs. Central Bank authorities may consider creating a large number of risk categories with provisioning rates increasing smoothly according to expected loan losses.36 * Argentine regulators require banks to keep a minimum level of capital to act as a buffer against unexpected losses related to credit risk. Besides the bank's rating according to the BCRA and the capital adequacy ratio, minimum capital requirements for loans depend on two factors: the guarantee and the interest rate. Because SME loans usually carry higher interest rates, regulators may consider 35 Argentine Personal Data Protection Act, Law 25.326 enacted October 30, 2000. Argentina introduced a Habeas Data Law in 1994. 36 Using data from the Central de Deudores, Falkenheim and Powell (1999) assessed the adequacy of provisioning levels in Argentina. They found that current provisioning requirements for category 2 may be too low and too high for categories 4 and 5. 52 creating a special table to classify SME loans according to interest rates. The overall effect of these regulations is to increase interest rates by 0.6 - 3.5 percentage points (Schreiner and Colombet, 2000). * Loan Documentation. Banking regulation regarding loan documentation for credit dossiers is another aspect of banking regulation that may influence SME lending. For prudential purposes, bank regulators specify minimum documentation for credit dossiers to support credit analysis based on repayment capacity. This may include: audited financial statements, monthly sales, debts with banks, business' bylaws, proof of registration with tax authorities (DGI and CUIT) and of payment of "obligaciones previsionales." SMEs are particularly affected as it is very costly for SMEs to comply with these requirements, especially when their financial statements are less sophisticated than those of larger businesses are. Moreover, excessive laws and regulations also induce firms to misrepresent their financial performance for tax purposes. * In Argentina, the legal framework requires financial institutions to deny loans to employers that have arrears with social security contributions for their workers (Law 14i499, art. 12). Thus, banks must verify compliance by searching a database of social security debtors and firms must submit a notarized declaration. However, deficiencies in the database, such as outdated and inaccurate information partly due to inadequacies to incorporate government forbearance programs and disputes among various tax jurisdictions, may exclude potential creditworthy SMEs from accessing bank loans. 5.7.3 Legal Framework for Secured Transactions Problems in the legal framework for secured transactions may explain the low usage of liens to back loans. A World Bank study identified several shortcomings in the creation, perfection, and enforcement of security interest in movable property, which may explain the high cost and difficulty of obtaining credit for SMEs (World Bank, 1996). * Argentine law restricts the types of propertv that may be used to create a security interest (particularly movable property and intangibles), the transactions that may be secured, and the type of lenders that can accept such security interests. * Registries seem expensive and inadequate for registering security interests and for searching existing encumbrances. Each geographical area has its own pledge registry, requiring re-registration of goods that move between jurisdictions or against debtors that have multiple business locations. Moreover, each local registry operates as a monopoly charging very high prices. The indexing system of the registry databases allows lenders to check whether a particular movable asset has been pledge, but not whether a particular debtor has ever pledged a movable asset. Finally, registries operate manually and data can only be transmitted in person because fax or mail is not allowed. * The execution of guarantees of movable goods is lengthy and cumbersome because the process is treated as a judicially litigated case. It takes about two years to repossess and sell movable collateral in Argentina. Moreover, associated court fees are high, reaching between 4 to 6 percent of the total claim depending on the province (Cristini, Moya, and Powell 2001). The Argentine Government has taken several steps in reforming the environment for secured transactions, partly improving some areas. The main changes in the legal framework for secured transactions are: a Ley de Prendas (1995), creation of an unified registry for motor vehicles (1996), ratification of the law Facturas de Credito (1997) that had no impact since its use was mostly voluntary, sanctioning of a new leasing law (2000). However, without a good registering system of security interest and an expedite recover of pledged assets, SME lending has not increased. This suggests that the Government may consider a full reform of its system, instead of piecemeal reforms. 53 6. Knowledge Constraints There is an important difference between finance and business environment on one side and knowledge on the other as factors impacting SME growth. The special nature of knowledge of markets, technologies, and supplies, and the fact that the knowledge could be "readily" available but would still require substantial effort on the part of the firm to assimilate it, puts much greater weight on demand factors rather than on supply factors. Whereas any SME would demand good business environment and credit which do not require sophisticated capabilities to consume and assimilate (any one can spend money or benefit from simpler regulations), advice of consultants is only as good as the firm's ability to assimilate and act on it. Consequently, weak SMEs do not often understand the nature of their problems. They tend to list finance constraint as the major problem whereas in most cases it is conspicuous lack of marketing, managerial, and technical capabilities. Yet studies continue to show that Argentine SMEs, notably in manufacturing, lack access to new technology, have poor management practices, and fail. to invest in and produce organizational and technological innovations, particularly when compared to large firms operating in Argentina.37 For instance, INDEC's study of Argentine industrial firms reveals that while only 13 percent of SMEs maintain formal R&D units and 35 percent maintain "informal" R&D units, 30 percent and 50 percent of large Argentine firms have, respectively, formal and informal R&D units. Another Argentinean survey reports that during the decade of the 1990s, 34 percent of firms did not invest anything on technology, another 34 percent did invest but had little to show for and only 33 percent did invest with positive results. Likewise Argentineans SME knowledge of the value and usage of ICT and Internet appears highly limited. A recent survey by FUNDES (2001) among Argentinean exporting SMEs illustrates this knowledge issue, as shown in Figures 6.1-6.4. Figure 6.1 shows the limited knowledge about electronic commerce operations and opportunities. Figure 6.2 shows that near 70 percent of those SME do not perform any transactions electronically bypassing the benefits it provides. Figure 6.3 illustrates the reasons given for that poor usage of the Internet by SME; most of the perceived reasons are knowledge related. Finally Figure 6.4 shows the main obstacles to the broader use of ICT in Argentina. Thus despite the significant benefits reported by those SME extensively using ICT, we find, Argentinean firms are highly oblivious or ignorant. The problem is even more severe that this data indicates, since, the sample is only among exporting SMEs, which tend to be among the most dynamic and better run SMEs than the rest. 37 The most comprehensive studies on these issues in Argentina are FIEL (1995), IERAL (1999), SOCMA (1997), Kosacoff(1993), Elgue (1999), Kantis (1998), and FUJNDES-UNGS (1999.) 54 Figure 6.1: Surveyed Argentine SMEs Knowledae Figure 6.2: SMEs that Perform Electronic on E-Commerce Subjects Transactions Does Not Perform Any 0~~~ M Does Not Know~~~~~~~~~~~~~~~ ElectronicTransactions o, _e Does Not Knowws 0. i Knows | Only Performing Purchases o. - a Only Performing Sales 0o __ 1 t_n _ - ~ 0 Perform Purchases and Sales EC Possibilities EC EC ' Does Not Know/Not Answered Off.,ed Opeations Reulations by EC Figure 6.3: Reasons Why Surveyed Argentinean Figure 6.4: Obstacles to the Development of E- SMEs do not Perform Electronic Transactions Commerce in Argentina T.-_iesliCod = s Distrust of the Medium Figh Impowleetaon Co= nCo Inadequate Pou ovneeuePrall C-fir lf i Lack of Trained Personnl Did Not Consider the Posibility- W Do Not Consider it Neessa._ DNot Know How to Implemen It _ C_l i Others M 0% 10 20% 30% 40% 50 60% 60%' Source: Fundes (2001) 55 From the policy angle the main issues to be addressed to facilitate and speed up the increase usage and development of ICT as a productivity-enhancing instrument are listed in Box 6.1. Box 6.1: Legal Aspects to be Considered for the Development of Electronic Trade . The jurisdiction of the firm that realizes the electronic sale and of the buyer to determine the legal regulations within which these operations will be legislated . The establishment of a system or institution that would let settle conflicts arisen for the collision between domains or internet addresses and registered trades, as well as all disputes related to those domains or addresses. * The establishment of security measures and its evaluation mechanisms * A norm framework to deal with banking issues of ET, considering that the financial transactions can be carried out directly on-line. * Methods of payment that can be used to realize ET transactions * Method and place where potential conflicts and disputes will be settle between parties involved in the transaction * The legality and irrevocability of contracts and documents sent by electronic means * The establishment of conceptual regulations and laws on the digital firm . The establishment of agreements or protocols related to systems and required levels of security and privacy allowed to make possible the electronic transactions, i.e., security system that include some kind of "encriptaci6n" of sent information. . The development of some specific technologies and methodologies that will effectively control the activities that could attempt against the author rights. * The establishment of certain measures, such as the application of internal fees and customs duty, procedures that would not carry major problems for non-digitalized goods. Source: Fundes (2001) 6.1 Low-level equilibrium trap as a result of the knowledge constraint Conspicuously low demand for knowledge sets the stage for a low-low level equilibrium trap - a vicious circle of mutually reinforcing lack of basic understanding of where the problems are, erroneous actions following the wrong diagnosis and resulting poor performance. Poorly designed SME programs can easily reinforce this vicious circle and be strongly counter-productive if, for instance, they pump credit resources to firms whose main limitation is on managerial and marketing issues. Weak knowledge capabilities imply that the credit would be poorly utilized which would put both the SME in question and the bank in jeopardy. Latent and often non-existent demand for knowledge implies that the major function of SME is to enhance demand for knowledge with support schemes, which will be the focus of discussion in Chapter 7. One should avoid primitive view of demand-driven SME support in the sense of providing services SMEs most often ask for. 6.2 Knowledge constraints common to all SMEs Strong and vibrant international value chains, such as suppliers for exporters and large firms, as recent evidence form a large number of countries strongly indicates (Guasch 2001) are important for SME development. However, compared to other countries, Argentine SMEs shows limited integration through out the supply chain and even less success in exporting, save rare exceptions. . They sell abroad about 10 percent of production, compared to 18 percent in Uruguay and 31 percent in the United States. The UIA study shows that only 16 percent of industrial SMEs made exports systematically between 1991-1996. Barriers to exports arise from various sources; including high tariff barriers and lack of information about business opportunities in foreign markets hinder SMEs willingness to export (UIA, 1999). 56 Isolation rather than size constitute the major problem that affects SME growth and access to finance and business development services (BDS). Business linkages, such as marketing, managerial, technical, financial and design, both with large firms and other SMEs are primary determinants of learning and productivity for SMEs. The evidence to date suggests that Argentine SMEs lack both kinds of linkages. First, while the Menem reforms have allowed trade to expand greatly for Argentina, they have also created incentives for Argentine exporters, dominated by a few large firms, to shift their focus toward lower value-added goods. For instance, although exports to Europe and the US have grown considerably, these exports are increasingly in primary goods and simple assembled goods. At the same time, while manufactured and processed goods are a large component of exports to the Mercosur, the most important market for Argentine firms, this market is relatively weak in terms of competition and demands for innovation and quality improvements. (Toulan and Guillen 1997, Guillen 2001). For instance, in the important automotive sector, a recent study by CEPAL (1999) shows that final assemblers are importing more complex, higher value-added components while using Argentine suppliers for simpler parts. Argentine SMEs are not becoming embedded in high-end value chains, as exporters and foreign investors find it less risky to invest in or procure domestically simpler products. On the other hand, large foreign and domestic firms will be less likely to change this strategy to the extent that they find Argentine supplier less reliable and innovative than in other markets.38 By most accounts, Argentine firms in general, and SMEs in particular, work in isolation and lack dense networks. For instance, a recent study by FUNDES-UNGS (1999) found that only 5.1 percent of the firms surveyed consider that ties with other organizations (firms and institutions) are important, while 58.6 percent and 36.3 percent operate in, respectively, high and. medium levels of isolation. Even among successful SME exporters (Fundes, 2000 study discussed in Chapter 3), the formal cooperation agreements with other firns exist in only about 30 percent of the firms. 6.3 Differentiation of knowledge constraints according to capabilities of firms Effective demand being the major knowledge constraint implies wide and crucial differentiation between various classes of firms. As Chapter 3 on heterogeneity of SMEs, showed empirically, knowledge constraints differ greatly between successful SMEs (which grow and export consistently) and failing and barely surviving SMEs. For the successful SMEs, the constraint is high-quality high intensity marketing, managerial, and technical knowledge, which is often available through global export-oriented value chains, such as the garment value chain for instance. The role of public policy is to facilitate the plugging-in of these firms into such global private networks of knowledge. In contrast, weak failing firms do not realize that knowledge capabilities are indeed their major constraint. The role of public policy is then facilitation of demand for simple basic knowledge which the stronger SME take for granted, such as book keeping, market diagnostics and the like. To address the needs of these two widely different types of clients, SME programs has to be designed appropriately. All too often, however, SME programs do not have clear picture of this wide differentiation of knowledge needs and constraints of their clients, and, as below, end up at best irrelevant, and in some cases, counter-productive. Despite the growth in the 1990s of technical assistance and training programs in Argentina, the existing evidence reveals that SMEs have low access to or low motivation for such services. For instance, as mentioned, above the FUNDES-UJNGS study shows that very few SMEs access training and consulting 38 The resent work on globalization and FDI argues that firms locate complex investments not simply based on cost but on the quality of the host country's local institutions and production networks. See for instance, Garrett (1998), Reich (1990), Nelson (1993), Guillen (2001). 57 services (16 percent). It also demonstrates that a staggering 78 percent of firms surveyed used no instrument of SME support (e.g., training, technical assistance, credit, export promotion) during the two years prior to the survey (1998), which adds extra concern as the Argentine economy was rebounding in 1996-97. The following section will discuss promising attempts to create SME programs and policies to alleviate the knowledge constraints. The remainder of this chapter draws on the IERAL (1999) study, unless indicated otherwise. The study first analyzes the organizational structure of national and sub-national agencies and then turns to an analysis of specific types of assistance programs. 6.4 Agencies and programs to alleviate knowledge constraints: lessons from evaluation As can be seen in Tables 6.1 and 6.2, the study gives two groups of data. The first group is the budget size and substantive orientation of the assistance services provided by the agencies (i.e., technological development, best practices, export promotion, etc.). The second group concerns IERAL's performance criteria: * Market Orientation - the extent to which market prices and the preferences of users are determine the activities of the agencies; * Operational Decentralization - the level of development of a network of local institutions that is used to provide services and facilitate user access; * Private Sector - the extent to which non-government and private actors participate in the management and provision of services. The results of IERAL's evaluation of the national agencies, summarized in Table 6.1 are striking. Not only do'all but one score relatively poorly, but also budget size has little in common with performance. IERAL notes that a key reason for their lack of integration into local economies can be found in the way these agencies were construicted from the beginning - very centralized. To a certain degree, over centralization is a product of the efforts by the central government to reorganize and streamline the management of existing programs and agencies. For instance, FONTAR was created in 1996 to provide financing and technical assistance to firms engaged in developing and adapting new technologies. Moreover, FONTAR was to have a decentralized operating structure, with provincial offices working closely with the relevant provincial organs, which also provided representatives for project evaluation and approval committees of FONTAR in the respective provinces. However, this approach was obstructed in the manner that both FONTAR was initiated and the Ministry of Education was reorganized. The ministry created the Agencia Nacional de Promoci6n Cientifica Tecnol6gica to integrate and coordinate that various existing (SECyT, CONICET, INTI) and new initiatives (FONTAR and FONCYT). Concerned with duplication and turf battles, the ministry's actions led to a centralization of decision-making within the Agencia and the other organs. When SePyME was born, its first objective was to catalogue and consolidate many of the existing, redundant and sometimes conflicting SME assistance programs in various ministries. As turf battles grew, centralization increased, which hindered any efforts by SePyME to develop local offices. 58 Table 6.1 National Organizations of Business Development Support Programs Private Size Market Operative Sector Organization (Millions of S) Objectives Orientation Decentralization Role Agencia-Fontar-Foncyt 69.9 MteI-FI Average Low Low Sepyme 54.4 SDE2 Average Low Null INTI 18.6 ET Low Average Low SEGEMAR 4.0 ET Low Low Low Export-Ar Foundation 3.0 PE' Average Average Lw Promex 1.7 PE Low Low Null Network of Enterprise 2.84 SDE High High High Development Centers (IDB-UlA)_ _ __ _ Source: IERAL de Fundaci6n Mediteffanea (IERAL, 1999, p. 106) Notes: I-Technological modernization, innovation and institutional strengthening; 2-Enterprise Development Services, 3-Technological Extension; 4-Export Promotion. Table 6.2 Sub-National Organizations of Business Development Support Size Operative Organization (Millions of S) Market Orientation Decentralization Private Sector Role IDEB 12.0 Average High Average Min. de la Prod. de Santa 26.8 Low F6 Ss. de Com. Ext. 1.1 Low Low Null C6rdoba UPE Mendoza 2.0 Average High Average Pro Mendoza 0.5 Low High Considerable Ctro. De PyMEs 2.5 Low Null. Neuquen U.O.Dir.lnd. Misiones N/d Low Null C.P.P.A. Misiones N/d Low Null Ss. de Cel Misiones N/d Low Null SICTyT del GCBA 3.2 . S.P. de Rosario 1.1 Null Ag. Des.Ec. de C6rdoba 1.0 High High High Source: IERAL de Fundaci6n Mediterranea, 1999, p. 107 These problems may subside with time. Indeed, the new De La Rua administration has recently passed a new law for SMEs that includes funds for SePyME to develop a public-private network of service centers. Currently, a debate is growing about how to proceed. One coalition (oriented around IAMC) advocates converting the BNA into an institution exclusively for SMEs and thus using the BNA branch system (the largest in the country) as the basis for the new SePyME network of service centers. Another coalition. (from the UIA) advocates a de novo system of centers created with local governmental and private actors. Such actions would demand however substantial changes in the way that services are created and in the level of cooperation with provincial and municipal governments. Considering the poor record of the BNA and the difficulties of both restructuring existing BNA activities and integrating SePyME ones, the BNA proposal looks unfeasible. If FONTAR can follow the lead of SePyME, though cooperation between national agencies and programs is usually very poor (IERAL 1999 IAMC 1999, FUNDES- UNGS 1999), the UIA proposal can be evaluated in two ways. First, one must evaluate the quality of the sub-national governments: are they already strong partners or do they need restructuring themselves? However, IERAL offers additional data that illuminates some of 59 the conditions for better or worse performance of agencies. The first, most obvious revelations are that simply being local does not determine success and that dramatically different agencies under the same provincial government may exist. Sub-national governments suffer from many of the same institutional and governance problems that national governments do. In many ways, these problems are exacerbated by a history of both centralism and center-periphery disconnect. This history leaves sub-national governments with a baggage of clientelism and weak institutions of accountability as well as a lack of resources and know-how to' design and implement new private sector support programs. (McDermott 2000, Tendler 1997). In contrast, those sub-national agencies, such as IDEB of the Province of Buenos Aires and the Agencia para el Desarrollo Econ6mico of C6rdoba, that do score high according to the IERAL criteria show a set of common organizational characteristics that combine the use of the market and a rich inter- organizational network (see Box 6.2). First, both agencies focus their programs on support for SME training and technical assistance in process and organizational best practices, rather than grants or training in narrow area, such technology acquisition or exports. Second, they have consciously constructed centers and incubators that appear to have participatory governance structures with a variety of public and private actors. In turn, the agencies can become nexus points for a network of diverse, previously disconnected private and public organizational actors, be they private consultants, trade association, banks, or universities. Such networks not only help the agencies tap into existing human and organizational resources for new uses, but they also promote learning and monitoring by facilitating a flow of information among the participants and allowing the participants to have a strong voice in the design, execution, and evaluation of services. While some participants contribute their knowledge and experience with SMEs but lack resources, others that lack experience with SMEs learn to utilize their institutional resources in new ways. Third, the participation of both supply and demand-side actors helps the agencies develop demand awareness instruments, such as low-cost preliminary consulting and workshops for SMEs. Third, both agencies use co-financing and not full subsidies to use market prices and demand as complementary governance and evaluation mechanisms. In light of this evidence, the SePyME plan for a decentralized network of SME support centers will depend largely on its ability to learn from more successful sub-national partners and to apply these lessons to regions lacking institutional and organizational coherence. The concluding section of this paper will discuss the issues of the institutional conditions for a national body to facilitate decentralized monitoring and inter-regional learning - issues that SePyME officials appear pay minimal attention to. However, in employing several people who managed the relatively successful UIA-IDB Centros de Desarrollo Empresarial, SePyME has the potential knowledge in designing centers. A closer look at the CDEs exposes several strengths and weakness of this model. In late 1996, the IDB (via the MIF) and the Union Industrial Argentina (UIA) launched three SME support centers - in San Rafael (Mendoza), Mar del Plata (Buenos Aires), and Rafaela (Santa Fe). Nexus Associates recently completed an external, positive evaluation of the design and existing results (i.e., finances, customer satisfaction, demand growth, etc.) of the CDE in Rafaela, which covers a few "departments" in Santa F6 and neighboring C6rdoba. (IDB-Nexus Associates 2000) Approximately 2,554 firms were operating in this region during the 1990s, 99 percent of which were classified as SMEs with under 100 employees. Similar to the above-mentioned agencies, the Rafaela CDE combined the use of the market with an inter-organizational network. On the one hand, the charge for services allowed market prices and demand to shape the supply and to discipline the CDE, which must become self- sustainable within a few years. On the other hand, the CDE was designed by combining strong and weak organizational and professional ties: the CDE partners and co-governors are a variety of regional public agencies, educational institutions, and trade associations that also contribute a small, experienced full time staff; the CDE, in turn, has indirect, extended ties to a broad network of private consultants, who are contracted to provide services. This participatory approach helped solidify the commitment of existing actors with resources and local knowledge and facilitated inter-organizational specializations and 60 synergies. Moreover, this type of network helped the CDE, from its inception, identify the key needs of the potential customer base, generate demand awareness instruments for SMEs, and diversify services from general courses on best practices to smaller, intensive workshops for certain types of business operations. This knowledge is supplemented by periodic market surveys and focus groups. A second similar characteristic is that CDE programs focused on training and technical assistance in mainly core business processes, such as strategic planning, product development, marketing, and sales, order fulfillment, and support processes. Programs focused also on specific outcomes and practices as well as behavioral and motivational issues of SMEs and their employees. One, however, notices three critical limitations of the Rafaela CDE model. First, the Nexus report notes that the Rafaela CDE may face serious self-sustainability problems. Currently, the CDEs have a subsidy from the MIF, which covers, for instance about 25 percent of variable costs, including consultant fees and direct project expenses. Nexus recommends that the Rafaela CDE expand and diversify its customer base and services. (Note that the Rafaela CDE targeted 800 firms in the region in need of their services, and by late 1999 616 firms had participated in one or more CDE programs.) But from a network view, even this strategy may be insufficient. As several scholars on network dynamics have argued (Rowley, Behrens, and Krackhardt 2000, Kogut 2000, McDermott 2001, Locke 1996), over time different types of network structures can become rigid, stagnant, and grounds for self-dealing. In turn, expansion of services and the customer base can be limited by the boundaries of the network itself. The issues of financial self- sustainability and expansion of market lead to the two other problems, the second being knowledge development. As mentioned, the Rafaela CDE relies on contracting outside consultants, not only to tap into a network of experienced and capable actors but also keep down costs. Continued use of consultants on project-by-project basis can act as a barrier to institutional development and knowledge transfer. The CDEs, in turn, must explore ways to select certain consultants as full time employees while strengthening the network ties to other sub-contracted consultants. The third related problem is one of self-selection. Anecdotal evidence reveals that the other two CDEs are not functioning as well as the Rafaela CDE, with one in large disarray. Much of the performance differences, between the CDEs as well as between the providers in different parts of Santa Fe and C6rdoba, are due to the fact that the regions have distinctly qualitative differences in their inter-firm and inter-organizational networks. For instance, Rafaela has been the subject of several studies (e.g., Kantis 1998, 1999, Kantis et al. 1999, Casaburi 1999, FUNDES-UNGS 1999) that argue that its dynamic regional economy and SME sector is determined by the existence of rich socio-economic networks. The Rafaela CDE is relatively successful because is tapped into and helped re-focus these network relationships. Box 6.2 Promising Business Development Organization: El Instituto para el Desarrollo Empresarial Bonaerense (IDEB) IDEB was created by a provincial law in 1996. The board of directors is composed of five members of the private sector and five of the public sector. IDEB's mission is to create and promote SMEs. Institutional construction is its main goal. The view of the IDEB is that a stronger institutional frame is key for the development of the Micro- and SME sectors and for economy in general. The organizational structure of IDEB was designed to minimize the amount of hierarchical levels. The whole of the structure has around 110 people, of which 60 are technicians. Companies have access to IDEB's services and programs through around 35 CENTROS IDEB. These centers are non-profit associations where all the managerial forces and the municipalities (less than 50 percent) must be represented. Universities, if there are any, must also be represented at the IDEB CENTER. The centers are active in out-reach and educating demand. IDEB offers a limited selection of products and services to maximize economies of scale. These products and services have a specific manual. Universities give courses on quality, training, management, technology, and environment. Commercial and Associative Programs are given by an IDEB center. Source: IERAL (1999, p.108) 61 7. Toward the Reform of SME Promotion: Current Policies and Lessons from Best Practice The Argentinean government over the years has put in place a number of assistance programs to enhance SME development and executed a number of policy actions for the same purpose, having allocated a considerable amount of resources to that objective. A brief summary of many of those measures is presented here as well as the normative arguments for intervention in that sector. 7.1 Why Foster SME Development? An essential issue is the justification for government interventions and their form. Usually the case for government intervention is often made on the generic basis of government or market failures that adversely impacts SMEs or on equity considerations, considering the impact of those types of firms, and even more so microentreprises, on employment and poverty. Based on the Donor Committee Guidelines for SME policies, summarized in Hallberg (2000), this section presents a brief summary of the economic rationale for SME interventions on efficiency grounds. The previous chapters presented a strong empirical case of substantial difference between SMEs and large firms. A number of those differences or components of, are driven by mainstream economic arguments, preferences, ability to bear risks and so on and are expected and appropriate and, in principle, do not merit intervention. Other are driven by responses to market and government failures and those indeed merit attention and intervention to correct them if feasible. The industrial organization theory, as summarized by Acs, Carlsson, and Karlsson (1999), pinpoints three main groups of factors explaining the empirical size distribution of firms and differences among various firm sizes. First, market power, segmentation, and distortions in input and output markets determine cost differentials between large and small firms. Having dissimilar flexibility to cope with these distortions, small and large firms present cost differentials that affect their growth potential. Many of these give an advantage to larger firms: for example, fixed costs and transactions costs associated with regulations and business processes. Second, the conventional approach explains that technical and locative efficiency determines the minimum efficient scale of production. Economies of scale of production, of scope, along with diseconomies of scale of organization technology, determine ranges of efficient firm size. Long-run average cost is minimized in this model, making SMEs vulnerable to squeeze driven by internal or external shocks. Finally, the transaction cost approach views the firm as an alternative to the market-a mechanism of allocating resources and structuring transactions (contracting, bargaining, etc.). Transactions for which the market is a highly costly form of governance are withdrawn from the market and internalized by the firm, thus increasing the size of the firm. The nature and size of transaction costs can change over time. For example, new communications technologies may lower the costs of transacting with suppliers, leading firms to outsource activities previously handled internally. Transaction costs and market imperfections, and how to manage them, are key determinant-among others- affecting productivity differentials. Some of these factors explaining the size distribution of firms and the differences between SMEs and large firms are influenced by policy and institutional factors and hence become an economic rationale of policy intervention. In particular, those factors related to transactions costs, such as fixed costs of contracting like in the case of financial markets, segmented and distorted labor markets due to institutional rigidities, the degree of competition, and the environment for business operations. Then there are public goods and externalities related factors, such as information and coordination issues, and (firm) networks development, jump stars and start up issues that indeed can merit intervention. Obviously, factors such as scale economies fall outside the realm of public policies. Finally there is 62 always the equity angle, driven by the impact of SME on employment, poverty, and inequality. Overall, a combination of all those arguments is what drives the government support for SMEs. 7.2 Current government policies and programs Argentina has a myriad of specially designed programs to offer training and technical assistance as well as financial support to small and medium-sized enterprises. According to the Secretary of Small and Medium-Sized Enterprises (Secretaria de Pequenas y Medianas Empresas, SEPyME), created in 1997 under the Presidency but moved in 2000 to the Ministry of Economy, Argentina has approximately 300 programs and lines of credit to support SMEs (IAMC, 1999, p.23). This section reviews the status of government interventions in three main areas: enyironment for business transactions, financial markets, and business development services. 7.2.1 Government Efforts on Business Deregulation During the past decade Argentine has experienced a radical change in the organization of its economic institutions. From a vastly over-regulated and state led model, which has driven the country to 25 years of economic decadence, the 1990s witnessed a structural reform emphasizing modernization through deregulation and market oriented policies. Important changes in the monetary system, public sector (mainly rapid privatizations), and liberalization of financial transactions and liberalization of international and domestic trade have taken place. The gains of privatization, deregulation and opening up of the economy resulted in a dramatic improvement in terms of productivity. Over 1991-1998, before the onset of the current economic slowdown, total factor productivity (TFP) grew at 2.1 percent per year suggesting important reduction in production costs for the economy. However, the distortionary effects of inappropriate regulations have resulted in high labor cost in the formal sector due to over-regulation and high taxes and high transaction costs due to pending transformation of the public agencies and the tax system, both at federal and sub- national levels. The Argentine government has shown interest in fostering private sector competitiveness by reducing the costs of business regulations. Current deregulation efforts are focused on deepening the process initiated in the 1990s. The main activities in this regard are: Competitive plans. Competitive plans (planes de competitividad), which are sector specific contracts between all government levels and private agents to reduce the distortionary labor regulations and taxes. These plans are the only coordinated effort between federal and provincial authorities addressing tax and labor tax distortions at all levels of government. However, the enforcement of commitments detailed in the competitiveness plans is weak because provincial governments are independent of the federal government. * One-stop-windows. "Sistema de Ventanillas Unificadas," or one-stop-windows, were created by the municipality of Olavarria under the auspices of an IADB-FUNDES pilot project. This is an isolated activity led by the local government (not coordinated at any level) to simplify the interactions between private agents and the government. It is too early to assess the performance of the one-stop- windows. Its most obvious limitation is limited geographical coverage. * Simplification of tax and labor regulations. Efforts to simplify the (i) tax system (reduce the number of taxes, see below), (ii) the registration procedures for labor hiring (reduce the number of procedures) and (iii) the modernization of the collecting system of federal and social security revenues (use of the bank system). 63 Simplification of Labor Regulations Many elements call for reform to give firms a legal framework to hire and fire workers more efficiently. The three main areas calling for reform are (i) high dismissal costs; (ii) high labor cost; and (iii) the centralized collective bargaining agreements, which are in force even after expiration, imposing high non- wage benefits which drive labor costs up relative to the cost of capital. In the 1990s, the Argentine government has introduced some flexibility, at the margin, to labor costs. A new type of labor contract, contratos de modalidades promovidas, was created to reduce or (in some cases) eliminate dismissal costs and labor taxes. These contracts were applicable only to entrant workers in new jobs. It was prohibited to change the nature of contract for existing workers in a given job. They were set in 1995 but abolished at the beginning of 1999, period during which these contracts accounted for 40 percent of new jobs and 5 percent of total workers. At the end of the past decade, the labor law returns remain inflexible as before. But same positive aspects have remained, such as the existence of a one-month trial period (in which there is no severance payment nor labor taxes) and a lower level for labor taxes. Last year, the new administration sanctioned a labor law reform which includes: i) an extension of trial period from one month to three, extensible to six through collective bargain; ii) lower labor taxes for new employees under 25, above 45 and women; iii) elimination of the concept of ultra-activity in collective bargaining; iii) a set of economic penalties for employers who after firing a worker cannot certificate the prior formalization of that worker.39 Very recently, the Argentine government has produce a major change consisting in the elimination of labor taxes through regarding them as payment-in-advance of the value added tax. Simplification of Tax Regulations In the 1990s, the Argentine government changed the revenue sharing system betWeen the federal and provincial governments. Initially, the reform was targeted to eliminate provincial taxes in exchange for higher shares of the national taxes. The number of taxes was reduced and all agencies in charge of collecting taxes were re-organized under a single agency: the Public Income Federal Administration (Administraci6n Federal de Ingresos Publicos, AFIP). Almost 56 percent of provincial revenues (and up to 90 percent for many small provinces) come through this sharing system. But, this higher level of revenue sharing does not impose an adequate scheme for fiscal responsibility. Moreover, provinces continue to levy their own taxes, which in many cases overlap with national taxes, contributing to increase the regulatory burden. Responding to an increasing fiscal deficit, the Argentina govemment reversed the simplification process by introducing some new distortive taxes, such as the gasoline tax and the firms' debt tax. Fiscal problems tightened in 2000, and the government raised the income tax for high salary earners while imposing some unpopular budget cuts. The objective of the government is to have only two types of taxes: value added tax and income tax with the view of eliminating all taxes that affect the price of factors of production. The main challenge is to reach an agreement with provincial and municipal governments to eliniinate all duplication of taxes. In 39 This is a perverse tool. The penalty consists of an extra payment over the severance payment to the worker. Thus, workers have no incentive to be formalized because they can earn additional money when fired if they were not formalized. To this perverse incentive, it is necessary to consider the high cost of formalization and the complexity of its procedures. The situation turns unfair for firms who hire workers. 64 this regard, the government has taken a serious of steps through the planes de competitividad and the decree that establishes that labor taxes will be considered as a payment-in-advance of VAT. 7.2.2 Financial Markets Throughout the world, governments have intervened extensively in financial markets. The belief that financial markets do not provide adequate funds for SMEs is one of the rationale for government interventions in financial markets. The extent of intervention has ranged from indirect measures aimed at improving the policy environment (for example, by addressing incentive problems and regulating financial intermediaries) to direct steps to increase or supplant credit provided by private lenders. Many countries eager to channel funds to under-served clients, such as small farmers, established government- owned development banks receiving concessional loans and on lending to customers at below-market interest rates, or directed private lenders to make concessional loans. These types of interventions were directed toward the symptoms rather than the causes of inadequate financial intermediation. Moreover, the poor design and performnance of most state-owned development banks, and their access to concessional funds and frequent bailouts have discouraged private, for-profit financial intermediaries from engaging in financial intermediation for under-served clients. Argentina has not been immune to this approach. The Argentine government has a strong history of using directed credit and subsidized lending to increase SMEs' access to credit services. By designing credit lines with subsidized interest rates, the Argentine Government assumes that banks face liquidity shortages and that SMEs cannot support high funding costs. In a recent examination of why banks do not lend in Latin America countries, which includes the case of Argentina, de la Torre et al. (2001) find that real credit growth has been substantially lower than banks' lending capacity. Banks could increase their lending supply to private agents, but choose not to so because of prudence reasons. Moreover, the authors find that the interest rate elasticity of credit demand is very low, indicating that private borrowers are not that concern about high interest rates but rather about other conditions of the availability of fund. The majority of programs for SME financing are at the national level. Several provinces, such as Neuquen, Cordoba, Buenos Aires, Santa Fe, Mendoza, and Misiones, have SME credit programs but there are virtually no analyses of them. National credit programs for SMEs are directed to general financing, working capital, and investment goods. These programs are executed by the SEPyME, the Banco de la Naci6n Argentina (BNA), the Consejo Federal de Inversiones (CFI), and the Banco de Inversion y Comercio Exterior (BICE), among other institutions. Another example of traditional interventions in credit markets is debt-refinancing or debt-forgiveness programs. Recently, the Argentine Government has implemented debt-refinancing programs for micro-, small-, and medium-enterprises, affecting mainly the BNA. A close examination of these programs illustrates perverse consequences in financial markets. These programs foster strategic default by rewarding defaulters, thus generating financial incentives to default. Moreover, these programs hinder the private sector supply of credit to SMEs by tainting the reputation of SMEs as "good" credit risks. 65 Box 7.1 Credit Guarantees The purposes of credit guarantee schemes are to increase the availability of credit and to improve credit conditions for SMEs, stimulating financial institutions, anrd in particular, commercial banks, to lend to small businesses with viable projects (Levitsky and Prasad, 1987). Whether credit guarantees schemes is beneficial to reach those objectives an issue of controversy. Proponents of credit guarantees argue that they can help SMEs to overcome collateral constraints and thus reducing the risks of lending (Gudger, 1998). Credit guarantee schemes aim to share risks with lending institutions so that the lending institution will be compensated for all or part of the loss on a loan default. Credit guarantees have advantages over traditional forms of collateral: easy to collect, value verification, and safekeeping. By requiring collateral, lenders reduce their provisioning requirements, lowering interest rates. Also, it is argued that guarantee funds, especially member-based guarantors that work closely with the target group, can generate a sufficient amount of information that gives them a comparative advantage to assess SMEs' creditworthiness. Moreover, guarantees may allow for a rapid expansion of credit, as they can leverage the guarantee fund various times. Finally, proponents assert that banks would generate relationship information about their SME guarantee clients, bridging the information gap. In contrast, those opposing credit guarantees argue that they are not directed against structural problems, such as high administrative costs or inadequate lending technologies, but rather towards credit risk, which is a borrower attribute (Rhyne, 1988). One possibility is that banks may be correct in that it may be too risky to extend large amounts of credit in the absence of the guarantee, then significant loan losses may occur by disbursing guaranteed loans which banks are not truly concemed with recovering. By guamntying loans that the banking sector finds too risk or too costly a credit guarantee fund is simply absorbing the risks (and ensuing losses) that the private sector now refuses to take. The guarantee fund may go bankrupt. Even if SME loans are not that risky, a credit guarantee induces banks to cut back in their efforts to screen applicants and to enforce credit contracts because they may transfer the credit risks to the credit guarantee fund while saving on the costs of rigorous credit screening and aggressive enforcement (moral hazard). This negative incentive arises because a credit guarantee absorbs the largest portion of the losses incurred when borrowers default on loans. In some cases the guarantee converts an appropriately risky loan into a risk-free asset by providing 100 percent coverage with no deductible for the participating bank. This is the case of state-owned credit guarantees in Argentina for example. Another common argument against credit guarantee schemes is that they present financial sustainability problems, imposing significant credit losses and high administrative costs to the Govemment. This concem is because credit guarantees imply significant risks and additional costs in the process of allocating credit without removing the underlying reasons why Argentine banks do not lend to SMEs. With high administrative expenses and inadequate pricing, which does not reflect the avemge default risk of guarantee recipients, credit guarantee funds often need to be re-capitalize (Levitsky and Prasad, 1987). Most guarantees charge a flat fee, giving the incentive to the lender to recommend loan guarantees to high-risk borrowers. Finally, opponents argue that credit guarantee schemes induce opportunistic behavior among lenders and guarantee recipients (moral hazard). A mechanism to reduce moral hazard is by requiring lender to take at least 30 percent of the potential losses. However, even by insuring only 70 percent of any given loan, credit guarantee schemes may not remove these negative incentives. Participating banks may avoid losing the 30 percent of the loan that is not guaranteed (deductible) by simply requiring a compensatory deposit for an equivalent amount from the borrower. In case of default, the participating bank may avoid all losses by exercising the guarantee from the credit guarantee and cashing the compensatory deposit made by the borrower. Hence, moral hazard on the part of participating bank is not removed. On the other hand, if the deductible (not insured portion of the loan) is too large to be avoided by a compensatory deposit of reasonable size, then the guarantee would not encouMge lenders to extend loans to most individuals whom they.would, otherwise, judge to be unworthy of credit. In this case, banks would include in the guarantee program mostly those borrowers to whom they would have lent in the absence of the guarantee, anyway. These elements imply limited additionality in terms of access to credit (Mudger, 1998; Levitsky and Prasad, 1987; Llanto and Orbeta, 1999). At most, credit guarantees (with a significant deductible) may induce banks to approve few credit applicants who would had been "barely" rejected in the absence of a guarantee. After reviewing the results of evaluations of guarantee schemes for small enterprises in developing countries, Meyer and Nagarajan (1996) concluded that most credit guarantee schemes are not financially sustainable due to poor design, poor management, and risky investment strategies. Most guarantee schemes present high administrative costs, reaching up to 15 percent of the guaranteed amount in various countries in Europe (Mudger, 1998). As a result these schemes are heavily depended on govemment subsidies to cover their operations and to avoid capital losses. In short, the above risks do not seem to be matched by sufficient benefits in terms of increased access to credit with respect to the situation of no guamntee available. Source: Prepared by the authors 66 Guarantee credit programs respond as an effort to correct inefficiencies in the financial market facing SMEs, particularly high interest rates, and lack of collateral. To foster credit guarantees, the Argentine Government has intervened through two main instruments, with apparently mixed results. The Government intervenes throughout direct interventions, such as state-owned credit guarantee schemes and by instituting a legal environment to foster private sector provision of reciprocal guarantee societies (Sociedades de Garantia Reciproca, RGSs). Guarantees are backed by a guarantee fund (i.e., money to be used in case a loan is not paid back in time and the guarantee is called). Loan guarantees provide lenders with assurance that the guarantee fund will make up any difference between a given guaranteed loan payment and an agent's actual loan payment. Overall, in Argentina these interventions have not been effective in lowering the total cost of credit for borrowers and have not resolved problems of asymmetric information, partly explaining their low penetration. It is important to point out that credit guarantee programs, in and of themselves, do little to encouraging banks to learn about SMEs and build relationships with them, creating another layer of transactions (see Box 7.1). 7.2.3 Business Development Services40 In the 1990s, the number of business development program for SMEs grew substantially. These programs offer services, such as training and technical assistance, human resource management, technologies, and export capacities. Overall, these government programs are characterized by lack of coverage, coherence, and follow-up, raising the questions Figure 7.1 BDS Programs for SMEs/GDP about the elements of success and the cost-effectiveness of those programs 0.12%- (Liberman, 1997). 0.10% - < -7The geographical distribution of the 0.08°/ - programs in Argentina presents two 0.06% noteworthy characteristics. First 0.04% - i 8 # _ national funds are about 2.5 times 0 04%/o - -A e ( _ . greater than those at the provincial and 0.02__ __ - municipal level, raising raises the 0.00% I- question about the institutional capacity Brazil Chile Argentina of the sub-national governments to Source: FUNDES (2001) develop, fund, and manage such programs. This dominance of national programs makes the issues of decentralization and national-sub-national collaboration all the more important for the evaluation of nationally sponsored programs. Second although not all programs from sub-national governments are included, the estimation considered the most populated provinces of Argentina, such as Buenos Aires, C6rdoba, Santa Fe, Mendoza, Misiones, and Neuquen and the government of the Federal Capital (Buenos Aires). These sub-national jurisdictions account for 80 percent of the country's SMEs, generating 64 percent of total employment, and 87 percent of the national GDP. One of the few systematic studies of non-financial support programs for SMEs in Argentina estimates that approximately US$222 million is now being spent annually by national, provincial, and municipal 40 Business development services (BDS) include a wide variety of non-financial services such as labor and management training; extension, consultancy, and counseling; marketing and information services; technology development and diffiusion; and mechanisms to improve business linkages through subcontracting, franchising, and business clusters. 67 governments and non-governmnt organizations for such programs.4' (IERAL, 1999, pp.97-99) This figure is equal to about 0.08 percent of the GDP and 0.26 percent of total public sector spending of Argentina and translates into about US$2,800 per SME. Despite the growth in the 1990s of technical assistance and training programs in Argentina, a small number of SMEs use non-financial programs. The existing evidence reveals that SMEs have low access to or low motivation for such services. For example, the Fundes study shows that only 38 percent of SMEs received consulting services in 1997-1998. Public institutions (including municipalities) provided only 16 percent of all consulting services. Also, about 62 percent of SMEs received training services. More importantly, Fundes study also shows that 30 percent of firms surveyed used no instrument of BDS support (e.g., training, technical assistance, export promotion), which adds extra concern about their growth potential. Many of these programs present serious shortcomings regarding their design and effectiveness in reaching target clientele, while entailing substantial costs. IERAL's evaluation of the national agencies presents striking evidence: most of them score relatively poorly in terms of outreach and integration into local economies, but also budget size has little in common with performance. The lack of integration into local economies of these programs can explain by a very centralized organizational structure. To a certain degree, over centralization is a product of the efforts by the central government to reorganize and streamline the management of existing programs and agencies. The most successful SME programs in Argentina in terms of efficacy and efficiency share some of the best practice principles. As presented by IREAL (1999), these programs show a strong market orientation where market prices and demand are used to shape and discipline services and programs. At the same time, they allow the service user (e.g., SME) to choose the instructor or relevant consultant, who is a public employee of the providing agency. To monitor quality of services, these programs have built-in mechanisms of evaluating customer satisfaction and preferences ex post. By attending these programs, SME clients become aware of the necessity to reorganize the firm and improve management practices, creating demand for additional services. This requires that good SME programs customize their services to clients. Moreover, good programs help users and producers of technology build close ties. Finally, a good program allows participating firms develop their own rules and forms of cooperation rather than imposing them on the firms. The principle is to provide guidance and not a straight jacket The data call forth several observations. First, although better-rated programs generally tend to come from sub-national agencies, one can find "good" programs within federal agencies. Such examples are the PRE program within the SEPyME and the Consejerosfiscales and Credito fiscal programs within FONTAR. One should note, however, that the evaluations are based more on program design than actual execution and ex-post results. The question remains whether the existing agencies can identify their quality service/training models and can provide an organizational structure that allows them to flourish and be duplicated or adapted to other types of programs. Second, programs that focus on training and assistance for improvement in firm organization and technology integration score better than those focused on teaching a narrow skill, like export promotion. Third, the IERAL analysis reveals that the characteristics of better-designed programs have much in common with the characteristics of better-designed agencies - extending the market-cum-network models of agency organization to the creation, teaching, and learning methods of specific programs. For instance, poorly rated programs tend to be standardized and rigid with little demand orientation and use of outside consultants or organizations. Not only do these aspects do little to address the needs of SMEs, but 4' These estimations do not include support programs for the fishing and agricultural sectors. 68 they also lead to mistaking information provision for knowledge creation, which demands greater interaction among various types of providers and users. Moreover, the provision of fully subsidized services and credits or tax breaks alone to stimulate the use of certain technologies or services not only eliminates demand sensitivity and market discipline but also makes the grave and often costly mistake of substituting monetary incentives for demand stimulation and user awareness. That is, as studies have shown about SME service and credit programs in other countries (Basch et al. 1995; Brugger et al. 1995), actors such as SMEs often are in a low equilibrium trap because they do not know how to get out. In turn, provision of services or supports must be combined with training and technical assistance to educate the user about how to identify his or her strengths and weaknesses and the adequate solutions. Better-rated programs not only use the market (via co-financing or matching grants) in having the customer cover some of the costs and chose from a variety of instruments or consultants, but they also develop networks of external organizational actors and potential clients to clarify the specific needs of SMEs and the alternative methods or forms of providing the services. In addition to conventional methods of market surveying, which can often identify generalities or the lowest common denominator, several of the programs of IDEB, Mendoza, and Cordoba gain knowledge about substance and forms of training and assistance from two main sources. These include partnerships with existing relevant public and private organizations (on both the demand and supply side) and the use of specific "demand support or awareness" methods, such as low-cost diagnostics and benchmarking of the practices of potential customers and low-cost assistance for the development of business plans. Also, these programs tend to combine approaches that focus on linking certain practices with measurable results and those that help firm and employee behavior and motivation. 7.3 Towards reform of SME programs and policies: lessons of best practice A large number of SME interventions have failed, both in Argentina and worldwide. Few have succeeded. An efficient and efficacious SME program is even more rare. This crucial observation brings us to central policy issues: the identification of factors of success of efficient and efficacious SME initiatives and their adoption in the design a successful SME programs and related institutions in Argentina. Evaluation studies in Latin America and Argentina (Oldsman, 1999, 2000, FUNDES 2001, TERAL 2000) allow discerning the following factors of success of good SME programs and institutions: * Development of a standard set of metrics to measure performance of SMEs and SME programs. For example SME growth and productivity rates, repayment rates, start-up survival rates, export growth, employment levels, linkages levels or index etc. This is essential to monitor effectiveness and adjust assistance and design. * Performance orientation. The successful programs are loaded and driven by incentives and induce self selection among SME firms, helping those that help themselves, illustrating how they can and should help themselves and suggesting exit to those that are not capable for internal or external factors to improve themselves. * Critical importance of entrepreneurial management. Successful organizations supporting micro, small and medium-sized enterprises (MSME) are often initiated by social entrepreneurs, individuals with unusual problem-solving and management skills and motivation for MSME development. In the examples of Foundation Chile and ProChile, success of MSME agency is predicated on such entrepreneurial manager at the top of the organization (Box 7.2). Successful MSME organizations tend to evolve from a reliance on top key individual to a robust organization with efficient corporate governance. * Cost recovery. Successful MSME organization should target for eventual full cost recovery. For example, if most MSMEs learn mostly from each other and that by demonstrating that MSMEs can 69 improve their business using business development services, some MSME organizations have:been able to secure a stable demand for new business development services. Matching grant subsidies to accelerate the demonstration effect can be justified until such demand develops. * Extensive use offollow-up techniques (benchmarking). Successful programs helps-through benchmarking indicators- diagnose where firms are, what do they need to do to improve and alternatives for those unlikely to survive. * Facilitate and reward linkages within the supply chain. Inform and facilitate opportunities to link with relevant supply chains and provide coordination assistance * Leverage their effectiveness with use of ICT. Develop SME internet portals, internet immersion institutes, improve access to the internet; provide access to ICT instruments, train firms accordingly and inform about the benefits. A number of these initiatives should be developing at the state level and coordinated at the federal level. * Participation of clients in the design of the programs. Clients not only need to pay for the services; they need to participate in the design and evaluation of programs. To assure that they do, SME programs shall never be run by the governments (whether federal or sub-national), rather an autonomous private management contractor working in cooperation with the government but independently from it, shall run the programs. * Facilitating collective learning of SMEs. The government (or central authority in general) orchestrates decentralized learning by monitoring and assuring information sharing between local experiments. Central authority does not direct--it facilitates information sharing, encouraging improvisation, and experimentation. Information exchange through continuous networked interaction) is the underlying principle of the decentralized learning in many areas of SME performance. Box 7.3 presents an example. * Incentive driven components and private sector participation. Successful programs are incentive drive and managed by the private sector. For example one of the most widely used are limited time matching grants to first movers. First movers that develop new ways of doing things, such as breaking at non-traditional export markets, are critical agents for economic development. Because SMEs learn most productively from each other, the success of a first mover provides powerful examples to follow and thus has positive spillover effects. Although some public subsidy to first-movers is justifiable because of the spillover effect, the main question for governments is how to deliver such subsidy. To avoid capture and resulting misallocation of resources, government shall not be involved directly in delivering such subsidy. Rather, a firm-level management governance scheme has proved to be a promising solution in various countries. This scheme included the creation of a temporary public fund managed by a private management service provider. To offset certain fixed start-up costs of breaking into export markets the government cover costs of knowledge (consultants' costs, travel to trade fairs), usually on a 50/50 basis. Once eligibility criteria are met, funds are provided on a "first come, first served" basis. The fund is temporary with a typical duration of 3 to 4 years because once exporting ceased to be innovative activity, thereby not leaving any learning spillovers, there is no reason for the public fund to exist. The private management service provider administering the fund operates under a time-bound contract with performance benchmarks and acts as global network information broker (by providing relevant information to firms), markets the scheme to the firms, provides up-front, free support in preparing firm's programs and acts as an impartial administrator. The relevant knowledge (about export markets and on the upgrading of the firm) is highly specific and can be provided only by independent service providers. * Gradualism and Assistance Linked to Performance. Not all SMEs are created equal, as we have seen. Successful programs link assistance with performance and gradual assistance as SMEs improve 70 themselves. A touch of realism assisting SMEs to identify their possibilities and potential is essential, followed with assistance on exit for those not viable. 71 Box 7.2: Performance Oriented SME Organizations: Factors of Success Example 1: ProChile For a country with impressive mineral and natural resource wealth, agro-businesses appeared to be a The first example is the Export Promotion Fund, promising route to follow; yet institutional infrastructure PROCHILE, a government foundation supporting export (leasing companies, demand detection, market projects proposed by groups of firms in the same sector. information etc.) was lacking. Agro-processing was The program was one of several efforts to expand dominated by huge enterprises created by Allende, now exports launched by the Pinochet government after the privatized but still monopolistic. Bottom-up facilitation country's economic collapse in 1981-82. Individual of new agri-processing value chains became the task of firms were ineligible for the program, under which the Foundation Chile. This Foundation is a hybrid government financed 50 percent of each eligible project. organization combining the features of project In addition to financial subsidies, PROCHILE supported development facility (to identify promising opportunities approved projects, and of course the firms clustered at the world market), technology incubator (to adapt around them, with specialized services such as the promising agro-industrial technologies and establish collection information on foreign standards and market firms to pilot the new technology) and seed venture fund trends, and the organization of study visits to the (to finance the firns and realize an upside gains once the facilities of foreign competitors. Current and potential firm is sold to private investors). Foundation Chile is exporters therefore had strong incentives to form sector- widely credited for the ensuing surge of Chilean exports specific committees that in turn defined individual in reared salmon, tomato paste, table grapes, and other projects that might qualify for the financial and other export products. subsidies. Social returns of Foundation Chile are much higher than The projects typically aimed at some combination of private returns (upside gains upon selling the successful improved quality to meet international standards and firms) because of demonstration effect of new SMEs development of new products for world markets. The (bringing new technology up to scale in a number of first sectoral committees grouped producers of processed firms) and technology transfer effect. Hold no illusion: fruits and vegetables, salmon, and other processed Foundation Chile is a rather untransparent institution and seafood, fresh fruit, furniture, textiles, wine, and paper not surprisingly, it made many mistakes by investing products. By 1988 there were some 65 committees into technologies that later proved unsuccessful. It is consisting of six to 14 enterprises each, and including all untransparent both in its origin (it is an arranged told over 700 firms. marriage between ITT corporation which put 50 percent Example 2: Foundation Chile - Incubating New of an initial endowment in exchange to restitution of its Enterprises assets nationalized by Allende and the government of Chile) and its operational transparency (a hybrid of In the 70s, following bold macroeconomic reforms, project development facility, technology incubator and Chile found itself with liberal and stable macroeconomic venture capital fund). But these caveats make climate yet with sluggish export growth. Foundation Chile all the more interesting: it is a second- best response to a highly imperfect institutional environment which was largely successful because of clear performance orientation and visionary management: results on the ground - success of SMEs it helped rather than writing reports for SMEs (usual focus of SME agencies). 72 Box 7.3: Government as Facilitator of Private Learning: SMEs Adopt ISO-1400 Standards Through Innovative Supplier Development Program In 1996, eleven large companies (domestic and The success seems to have depended on four principal multinational) in Guadalajara signed a two-year, factors. First, most SMEs indicated that the invitation voluntary agreement with Mexico's Secretaria de Medio from the mentor company, rather than simply from the Ambiente, Recursos Naturalesy Pesca to mentor small government or university, was a vital source of suppliers in implementing environmental management motivation and cooperation. While all participating firms systems (EMSs). Each company invited two to three had some sense of "ownership" in the project, the small suppliers to participate in the pilot. The large presence of a large company "champion" provided companies and the World Bank provided the SMEs important assurance that their dedication was not funding for EMS training and implementation support. misplaced. Second, the use of the consultant network A team of consultants - from two local universities and provided resources otherwise unavailable to SMEs and a foreign environmental management-consulting firm - even many large firms. The network in effect provided a delivered the services. The aim was to implement the rapid response system to local and firm conditions ISO 14001 EMS model originally designed for large through a diverse group of experts who drew on one firms for the SMEs and to evaluate the applicability of another's wide range of skills and knowledge. Third, the model as well as the sustainability and replication of although they attended the pilot's sessions mainly as the pilot partnership. observers, representatives of local and national environmental authorities focused managers' attention As of late 1998, the pilot was encouraging, with virtually on the project, and prompted many of them to learn all participating SMEs making major advances in the about the benefits and drawbacks of different standards implementation of ISO 14001, the reduction of and enforcement actions. Fourth, beyond the training in pollutants, and their ability to use general management technical issues, the use of benchmarks and an iterative, systems. Moreover, the national government has used collaborative review progress both demonstrated new this experience to develop the substance and forms of cooperation to participants at all levels and implementation methods for new environmental improved the information basis on which they could protection legislation. build the next round of efforts. Argentina is by no means peculiar in its challenge to re-design private sector promotion system from top- down to bottom-up on the basis of incentive-driven principles outlined above. All Western economies, and most recently Mexico, Brazil, El Salvador and the EU latecomers such as Ireland (see Boxes 8.1 and 8.2 in the following chapter) had to mastermind a similar transformation. 73 8. The Path Ahead: Policy Recommendations In the midst of an economic crisis, the Argentine Government faces the challenge to shoulder the economy to higher levels of growth. Small and medium-sized enterprise (SME) advocates argue that SMEs could and should be instrumental in delivering the promise of job creation and strong export performance while raising productivity and quality. To accelerate their response the government has put in place a plethora of SME assistance programs. But, by and large, the impact of these programs has been less than satisfactory even though the budget allocated to those programs is comparable to countries with vibrant SME sectors. Thus, to ensure an efficient allocation of scarce resources Argentina needs to review and overhaul its SME support system. Addressing and correcting the main obstacles for SME development must be the main priority in this reform agenda. This report identifies three blocks of issues that need to be addressed: (i) reducing the shortcomings of the business environment, such as institutional rigidities originating from licensing and registry of firms and labor force, lengthy and cumbersome processes and labor regulations, including the regime of severance payments and labor taxes, and the tax regime; (ii) increasing the availability of competitively priced financial services and better focused financial support to SMEs; and (iii) promoting a market for business development services compatible with SME needs and capabilities. Overall, the Argentine Government needs to revisit their SME support program, load it with incentives, induce self-selection, and invest in a monitoring system of SME programs, including consolidating budget allocations and keeping track of SME dynamics. In sum, it is not an issue of a need of additionality of resources but the better use of the existing budget allocation. The objective of this chapter is to propose policy recommendations for the SME sector, based on best- practice experiences and sound conceptual principles. As this report argued, Argentine SMEs are suffering from an incomplete adjustment to economic liberalization where a complete adjustment requires policies that facilitate the transition from inward-looking firms to learning-oriented firms looking to exploit emerging opportunities. The political will to secure an appropriate incentive framework involves amelioration of both government and market failures. The chapter outlines the need for a dramatic reform of SME promotion system and discusses the issues of reform (Section 8.1 on a general strategy, Sections 8.3 - 8.5 on business environment, finance and business development services). We also address the all- important 'how to' question of implementation of reform, both from immediate perspective (Section 8.2 on crisis as a window of opportunity) and from a longer-term perspective (Section 8.6). 8.1 General Strategy This section presents a strategy to improve the incentive framework for SMEs based on three blocks. The first block is the so-called first generation reforms that eliminate government failure in deregulating markets, in promoting widespread economic liberalization, and in providing a better (de) regulatory environment for businesses and investors. As these reforms do not require strong institutional capabilities to implement and follow-up, they are called 'stroke-of-the-pen' reforms. The second block of reforms includes initiatives to increase the availability of competitively priced and untargeted financial services, and non-distortionary initiatives targeting the SME sector. Deepening financial markets requires stronger credit supply and demand for fihancial services. Finally the third block of assistance refers to the offering of discriminatory and matched business development services (second-generation agenda) encompassing a set of initiatives, to ameliorate market failure and accelerate positive changes in business' behavior. The latter includes schemes such as matching grants, business incubators, private-public seed funds, and facilitation of private-to-private collaboration through business associations and information centers and ICT assistance. 74 Table 8.1 Summary of Market-Oriented SME Interventions Policy agenda 11 - Building Institutional Capacity and Measuring the Delivery of SME Policy agenda - Benchmarks and Policy agenda I -Deregulation Investment in Public Goods Performance Indicators Business Environment Business Registration a Evaluate whether it is worthwhile to expand the pilot project Monitor progress on simplification of business registration in all provinces by * Streamline business registration procedures by reducing initiated in May 2001 by SEPyME surveying firms on the investment climate. Items to monitor are: licensing and registration requirements and . Take advantage of information and communication . Business registration process - cost (for all areas, cost includes in time administrative fees technologies to facilitate interaction between govemment and formal and informal types of payments and contributions, including * Eliminate labor registration requirements beyond the and SMEs bribes), procedures required, delays social security one . Institute systematic monitoring of SME performance and . Business licensing - number and type required, cost, time and payments * Agree on cost-benefit analysis to eliminate the impact of policies and interventions required regulations that accrue little or no net benefit for the * Improve access to information about markets, standards, and * Acquisition of business premises - formalities, costs, constraints, and economy technologies) delays Coordinate business registration between national and a Evaluate the experience of "competitive plans" to see * Import, export and customs regulations - formalities, costs, and delays local authorities whether deregulation agenda can work in that context . Product, input, and equipment certification - types, incidences, • Disseminate information about registration requirements formalities, costs, and delays and procedures. Create process of innovation where Tax administration - requirements, constraints, costs, number of taxes firms' feedback is incorporated to simplify business and forms registration . Business inspections - types (and agency responsible), costs, number, and process followed Labor regulation reform: . Quality of infrastructure and services * Review labor taxes to identify which taxes do not bring * Labor relations direct benefit to the govemment, firms, and the labor Conflict resolution and legal environment force * Number of collective agreements signed at the SME firm level * Allow SMEs firms to sign decentralized-at the firm a Amount reduced of tax wedge level, collective bargaining agreements * Reform the severance payment regime to reduce its costs and make it more transparent and less prone to litigation a Consider substituting the severance payment regime for individually capitalized accounts Tax reform: * Simplify the tax regime and compliance procedures 9 Improve commercial transactions law 75 Policy agenda H - Building Institutional Capacity and Measuring the Delivery of SME Policy agenda - Benchmarks and Policy agenda I -Deregulatlon Investment in Public Goods Performance Indicators Financial Markets Improve legal environment for secured transactions. * Creation of a registry for lien on movable property allowed * Monitor number of security interest registered in the registry for liens on With WB assistance the GOA drafted a secured under a new secured transaction law movable property transaction law within the SSAL that could be submitted * Promote innovation in loan products, lending methodologies, * Reduction of direct lending through public financial institutions to Congress. delivery mechanisms, and risk assessment methodologies * Reduction in volume of subsidized credit lines . Eliminate shortcomings of the legal framework for risk * Develop a credit-scoring tool for the SME market in . Share of SME lending and growth rate of SME loans capital investment to promote equity investments in Argentina using information available in all commercial a Quality of SME loan portfolio SMEs banks * Promote and allow information sharing of all positive * Build institutional capacity in private banks by and negative credit information histories from all types disseminating best-practice on SME lending by commercial transactions to eliminate fragmentation of credit banks histories * Reduce public sector involvement in SME lending * Improve efficiency of database of social security debtors * Prepare a phase-out program of subsidized credit lines and * Improve legislation and monitoring of reciprocal interest rates subsidies. Redesign interest rate subsidies guarantee societies programs to avoid perverse incentives and guarantee additionality of funds. Rethink government participation in guarantee funds by allowing greater involvement of private sector Business Development Services (BDS) * Target subsidies for market development to specific * Facilitate innovation in delivery mechanisms * Cost recovery indicators for publicly-provided or subsidized services market failures . Develop performance and impact indicators * Coverage (number of SME clients), in particular outside major urban * Structure assistance on a gradual mode as SMEs repond * Improve training and technical assistance to private BDS areas to incentive structure providers. * Decentralization (extent of participation of municipal and provincial * Facilitate poor performer exit mechanisms * Limit long-term subsidies for BDS to public goods (e.g., govemments in BDS programs funancing and design) information, labor and management training) Reliance on private provision of services (rather than in-house service Improve management and cost control in public BDS delivery) institutions * Characteristics of program management (extent to which remuneration of Accelerate access, usage of ICT for SMEs. management or BDS services is tied to the performance of the programs) Condition budgetary allocations to the achievement of impact Reduce duplication across agencies in services provided Decentralize service provision to provincial level Introduce pilot initiatives combining deregulation and provision of business development services at a provincial level 76 This reports argues that the solution to the SMEs quagmire is not additionality of resources. Argentina already spends in that sector amounts comparable to countries with buoyant SUE sectors, such as Spain and Italy. The challenge for SME development is correction of policy failures of the business environment to improve incentives to perform and succeed, better and different delivery of business development services, and deepening financial markets. Institutional and knowledge development is at the heart of SME policy agenda. 8.2 Reform of SME promotion system: Crisis as a window of opportunity Evidence presented in a previous chapter strongly suggests not just isolated SME programs but the whole Argentine system of micro enterprise and SME promotion needs radical overhaul. From centralized, government and supply driven, top-down system it needs to evolve into decentralized, private sector and demand driven bottom up system. More specifically, a required decentralized system of promotion envisions new roles of federal, state, and local authorities and civil society. The challenge is daunting but so are the institutional impediments for the reform. Business deregulation was blocked for years by entrenched interests. Likewise, a reform of SME promotion programs is far from obvious because a single agency controls them. SMEs are scattered in more than a dozen federal agencies, suggesting a strong need for a 'reform champion': a top-ranking government official with broad powers to promote deregulation and change the established programs. As documented in Kuznetsov (2000) and elsewhere, crisis provides a unique window of opportunity to embark on broad-range reform. Institutional details of how to take advantage of this window opportunity are necessarily to be developed by stakeholders in Argentina. The following three observations could be useful inputs in designing a specific institutional solutions. 8.2.1 The growth pay-off to dramatic reform of SME promotion system are both significant and immediate Boxes 8.1 and 8.2 provide evidence about the crisis (drought in the Brazilian province of CEARA, and a protracted recession in Ireland) prompted a drastic reform of SME promotion with immediate tangible yet somewhat controversial results. Continents apart, these experiences share common features such as critical importance of reform champion, focus on results (measured by employment creation) and willingness to experiment and innovate. Success of reform seems to hinge on the ability to a deliver a reform package based on (i) aggressive deregulation opening new opportunities to SMEs and (ii) facilitation of access to business development services and finance allowing to act on new opportunities. 77 Box 8.1 Irish Regional Partnerships As a response to a severe recession of the beginning of the 90's, Ireland embarked on an innovative experiment to reduce the incidence and mitigate the effects of unemployment while further encouraging the development of an open, competitive economy. At the core of that experiment are 38 Area-Based Partnerships (initially 12) in urban and rural communities created by the Irish Govermment and the Structural Fund of the European Union (EU), beginning in 1991. The task and opportunity of these partnerships is to reconsider the problems of un- and under- employment within their home jurisdictions and devise effective responses to them that the central government alone could not discover, but to which it may refer in reforming its own administrative structures and, above all, in improving the connection between these and local communities. Legally partnerships are independent corporations under Irish company law. Their boards group representatives of local community interests, including the unemployed, representatives of the national social partner organizations of labor and business, and local or regional representatives of the national social welfare, training, or economic- development administrations. Through this structure, the partnerships often have de facto authority over a significant share of the local activities and expenditures of core agencies of the national government. In addition, they have the right to provide services and build institutions not contemplated by the statutory bodies. They thus pursue, simultaneously, area-based economic development and the local, integrated implementation of national programs connected to it; and they do so in a way that blurs familiar distinctions between public and private, national and local, and representative and participatory democracy. The preliminary results of this deliberate effort to foster development and welfare through new forms of public and private local co-ordination are extraordinarily promising, if still inconclusive. In eight years of operation, urban partnerships have developed innovative techniques for retraining and placing the long-term unemployed and building potentially self-sustaining firms that provide both training and jobs for them. They have also established new programs to help early school leavers and single mothers, and to encourage community policing and the management of housing estates by their tenants. Rural partnerships have found ways to increase employment opportunities for under- rather than unemployed groups, and to rebuild communities depleted by outmigration. Together these innovations may form the foundation of a new model for transferring marketable skills to vulnerable groups and communities, unexpectedly providing the opportunity for them to participate in the kinds of activity broadly characteristic of the modem sector of the economy, from which they are normally excluded. These innovations, moreover, are accompanied by local proposals for adjustments to the rules governing eligibility for social welfare benefits whose purpose is to make participation in the new programs broadly affordable and attractive, and to remove the disincentives that often deter the most needy from exploring their possibilities. All of this activity grows out of and reinforces an exchange of views and proposals between public and para-public agencies and the persons who use their services that is subtly reshaping their shared understanding of which local problems to address and how. One result of this mutual dedication to an urgent, common task is remarkable care in the use of resources and, so far, avoidance of the self-dealing that might jeopardize the reform project. 78 Box 8.2 Unleashing Private Dynamism In the Brazilian Province of Ceara In 1987, the newly elected reformist governor of Ceara, Tasso Jereissati, initiated a set of reforms that would radically improve this poor Brazilian province's fiscal health and economic growth by 1993. A' key set of reforms included innovative, decentralized programs in preventive health, public procurement from informal-sector producers, and a large emergency employment-creating public works program. Ini addition to decentralizing operations and integrating more private sector participants into the programs, the provincial government utilized two key methods to improve the governance of the programs. First, the provincial government used its political authority and financial power to break the traditional hold over program management by clientelistic stakeholders while empowering other, less.powerful local groups. Second, the provincial government defined the rules of program management to promote participatory monitoring by both service providers and users. These two methods were at the core of a demand-driven procurement program that fostered sustainable backward linkages between large customers (in this case government procurement agencies) and vibrant networks of SMEs. The most celebrated achievement was the transformation of the backward district of Sao Jao do Aruaru (SJA). The impact on the towni was startling. When the program began there were only four sawmills in the town, with just 12 employees. Five years later there were 42 sawmills, with about 350 workers, and a further 1,000 people directly or indirectly employed in the woodworking industry. Importantly, the customer base has been diversified over time, so that now over 70 percent of output goes to the private sector. The CearA government first broke the hold of large out-of-province suppliers and of stakeholders, like large banks, in supply-driven SME programs by having the public procurement departments, the Departments of Agriculture and Education, seek SME suppliers with the aid of two traditional SME support agencies. At the same time, rather than using subsidized SME credit schemes and long-term procurement contracts, the procurement departments had the support agencies offer a 50 percent advance on each contract with a supplier for a particular order, with full payment based on customer satisfaction (criteria of quality, delivery time, and price). In so doing, the government not only provided cheap initial capital to firms that they would otherwise not receive but also offered them a reliable customer base if the firms could meet procurement criteria. Success with a relatively simple order brought larger, more complicated orders. Moreover, the support agencies contracted not with individual SMEs but rather sought out existing SME associations and aided the formation of associations of potential suppliers located in one place. Over time, these associations became an organized political group that could pressure the government to expand the program and facilitate the development of new banks and training resources aswell as counter the power of traditional large suppliers interested in re-cooping their patronage. The next crucial success factor was the government's development of a collective monitoring system that fused information sharing, risk sharing and learning. First, the government separated customers from suppliers, using performance-based contracts. The procurement departments contracted with the SME support agencies to supply goods and services and supply SMEs with technical assistance. The support agencies received a 5 percent commission on each contract, the revenue from which became an increasingly source of financing for the agencies. The agencies, in turn, contracted with SME associations, which became responsible for finding the relevant member to manufacture the product and meeting the terms of the contract. Each subordinate had more detailed knowledge than the superior of adequate suppliers and operations and thus simplified the monitoring tasks. Each superior had the right to select, reward, and penalize the subordinate and all had revenue sharing incentives. Source: XXX 8.2.2 Rely on 'reform czar' rather than 'coordination committee' Consolidation, rationalization and avoidance of duplication of SME policies and programs has the order of the day for decades, both in Argentina and world-wide. As Boxes 8.1 and 8.2 on Ireland and Brazil indicate, the task is not impossible but the solution is not bureaucratic top-down coordination between relevant ministries, which has been practiced in Latin American countries for years. Inter-ministerial 79 Committees on SMEs policies and programs can be useful device to pool information and SME policies and programs but they can not initiate and deliver reform. The top-ranking reforrn chanpions appears to be indispensable to jump-start and sustain the reform of SME promotion (Box 8.3). Box 8.3 Deregulation as Entry Point: Mexico's Deregulation Czar In 1988 the president of Mexico appointed a "deregulation czar". Each month this official reported directly to the president and his economic council of ministers. Every business in Mexico, large or small, was promised an equal access to the czar's office to complain about burden associated with government rules and regulations. When the deregulation office received a complaint, it was obliged to find out why the rule existed, how it interacted with other regulations, and whether it should continue in effect. The office operated under a strict timetable: if it did not act to maintain, revise, or abolish the disputed rule within forty-five days, the rule was made void automatically. The work of the deregulation czar over the first four years of his tenure is widely credited with greatly accelerating Mexico's reform program. It provided struggling private business-people with an effective, responsive champion at the highest level of government. The factors behind this success included: . Unequivocal presidential support, signaling to both bureaucrats and citizens the need to comply with the czar's decisions . The fact that his decisions could be overruled only at the highest level of government * The setting of tough penalties for officials who failed to implement the rulings * The time limit, which ensured quick and visible results * The czar's staff, who were skilled in the economic consequences of regulations, in understanding of complicated interactions within the regulatory field other, and their administrative requirements -- no single person can effectively carry out a government-wide program of deregulation Finally, it was critical that the czar won credibility with both officials and the public by giving a fair hearing to the powerless and the influential alike, and setting a consistent record of impartiality. Source: World Development Report 1997, p. 73 8.2.3 Combine top-down down reform with bottom-up SME initiatives at a provincial level The key to a successful strategy for private sector development is in establishing the correct mix of top- down (macro-level) and bottom-up (enterprise-level) reform approaches. Policy or regulatory measures alone will not be sufficient to restore the growth of the real sector, institution building at the enterprise level and in the public sector through appropriate privatization and restructuring will also be necessary. These enterprise reforms will help create and empower constituencies to support further policy, regulatory, and institutional changes. Inadequate corporate governance practices and intrusive government controls are the source of the real sector problem, and the solution cannot be simply based on further policy reforms. They have to be complemented by changes in public sector administration and governance, and other institutions that shape incentives of enterprise managers and government officials. In this vein, we propose a policy package of massive top-down deregulation and a bottom - up pilot SME initiatives in two-three Argentine provinces. Top-down drastic deregulation shall be accompanied by performance metrics to measure a progress of deregulation through periodic enterprise survey of the regulatory burden. The areas of reform and performance metrics shall include (yet not limited to): 1) Business registration - cost (for all areas, cost includes in time and formal and informal types of payments and contributions, including bribes), procedures required, delays; 2) Business licensing - number and type required, cost, time and payments required; 80 3) Acquisition of business premises - procedure, costs, constraints, delays; 4) Import, export and customs regulations - procedures, costs, delays; 5) Product, input and equipment certification - types, incidences, procedures, costs, delays; 6) Tax administ*ation - requirements, constraints, costs, number of taxes and forms; 7) Business inspections - types (and agency responsible), costs, number, process followed Bottom-up initiative can include a pilot program combining matching funds for SME development with aggressive deregulation at sub-national level in two to three provinces. This pilot program would be focused on improving the investment environment. Under such pilot initiative, the Federal Government, Provincial Government, and the private sector will receive incentive to jointly design and implement innovative perfornance-based projects to improve state-level investment environment for micro enterprises and SMEs. For the purposes of this initiative, investment environment includes: (a) regulatory environment, as documented by transaction costs of entry, exit and doing business; (b) human capital endowments of SMEs requiring investments into vocational and on-the-job training; and (c) business development services to small and medium-size enterprises. As part of the required local solutions, the Government should examine where its actions are unduly distortionary and finish its deregulation agenda. Using a federal matching grant system, the Federal Government could convince provincial governments to undertake provincial or local level deregulation, especially in areas where it has been difficult for the federal government to convince state or local-level counterparts. The limited pilot would allow testing of central facilitation and monitoring-capabilities for local implementation, prior to rolling out the approach nationwide. Such a pilot can be established on the basis of successful and/or promising local private sector development agencies (Box 8.4). Box 8.4: Factors of Success of Sub-National Private Sector Development Agencies Sub-national agencies, such as IDEB of the Province of Buenos Aires and the Agencia para el Desarrollo Econ6mico of C6rdoba, that score high according to the IERAL criteria show a set of common organizational characteristics that combine the use of the market and a rich inter-organizational network. First, both agencies focus their programs on support for SME training and technical assistance in process and organizational best practices, rather than grants or training in narrow area, such technology acquisition or exports. Second, they have consciously constructed centers and incubators that appear to have participatory governance structures with a variety of public and private actors. In turn, the agencies can become nexus points for a network of diverse, previously disconnected private and public organizational actors, be they private consultants, trade association, banks, or universities. Such networks not only help the agencies tap into existing human and organizational resources for new uses, but they also promote learning and monitoring by facilitating a flow of information among the participants and allowing the. participants to have a strong voice in the design, execution, and evaluation of services. While some participants contribute their knowledge and experience with SMEs but lack resources, others that lack experience with SMEs learn to utilize their institutional resources in new ways. Third, the participation of both supply and demand-side actors helps the agencies develop demand awareness instruments, such as low-cost preliminary consulting and workshops for SMEs. Third, both agencies use co-financing and not full subsidies to use market prices and demand as complementary governance and evaluation mechanisms. Source: IERAL (1999) In light of this evidence, the SePyME plan for a decentralized network of SME support centers will depend largely on its ability to learn from more successful sub-national partners and to apply these lessons to regions lacking institutional and organizational coherence. The concluding section of this paper will discuss the issues of the institutional conditions for a national body to facilitate decentralized monitoring and inter-regional learning - issues that SePyME officials appear pay minimal attention to. However, in employing several people who managed the relatively successful UIA-IDB Centro's de Desarrollo Empresarial, SePyME has the potential knowledge in designing centers. 81 The details of the reform package are provided in Sections 8.3-8.5. 8.3 Eliminating Government Failure: Business Deregulation There are a number of unnecessary or inefficient procedures at the entry, operating and exit levels for SMEs that increases costs, direct and indirect that should be eliminated, reduced and revisited, as reported on chapter 4, where the details for reform are presented. To reduce those costs of business registration and of operating and exiting businesses, the government may consider the following activities: * Overhaul business registration procedures independently of a broader deregulation effort. Eliminate the obvious and unnecessary business formalities (tramites) that after using cost-benefit filters represent little or no net benefit for the government.42 * Design efficient and timely entry and exit (bankruptcy) procedures * Take advantage of information technologies to reduce the costs and the time of business formalities. * Create one-stop windows, for which it may involve the private sector. The SEPyME initiated in May 2001 a pilot project of ventanillas tunicas in three municipalities. Yet its impact needs to be assessed. * Eliminate or simplify the requirement of labor registry. Registration with social security should be enough * Allow, SMEs to sign decentralized-at the firm level, collective bargaining agreements. * Review the labor taxes and reduce those components that do not bring a direct benefit and are use mostly for fiscal reasons or highly inefficient provision of services to the labor force. * Reform the severance payment regime, reduce its costs, and make it more transparent and less prone to litigation. Start with a special regime for SMEs. Consider substituting the severance payment regime for individually capitalized accounts. * Simplify the tax regime and compliance procedures, at least for SMEs, for all levels of government. The results of the monotributo have not been promising. * Facilitate the development of brokerage markets for transport and logistic services. * Implement the multimodality law and the provision of insurance-across modes-component. * Custom reforms and simplify procedures. Managing the Process: Deregulation for growth. Short, medium, and long-term horizons To overcome pervasive institutional rigidities, the government should consider implementing a three- prong approach to deregulation, addressing short-term, medium-term, and long-term issues. In the short- term and in environment of crisis, one may consider idiosyncratic solutions, such as a "deregulation czar discussed in the previous section. In the medium-term, from the experience of other countries, an institutional solution to support regulatory management has been a consultative mechanism to generate credible commitment between the business and the government. Long-term approach to regulatory management has emerged in OECD countries, which advocate the implementation of "regulatory impact analysis." Regulatory impact analysis is a decision-making tool to inform policy officials of the benefits 42 Regulatory impact assessment was developed in many countries as a response to the growing volume and complexity of government regulations in the 1980s. 82 and costs of proposed or existing regulations and to ensure that the benefits of government action will justify the costs. 8.4 How to Deepen Financial Markets? Increasing the availability of credit and reducing the cost of credit, main objectives of the Argentina government, requires stronger credit supply and demand. In the short-term, the focus of the strategy should be to augment the supply of credit to SMEs by increasing the risk-adjusted returns on SME lending. An increased supply of credit paired with higher competition would translate into lower total cost of borrowing, including lower effective interest rates and transaction costs. By improving the credit conditions of SME lending, SMEs' demand for loans would increase. The standard presumption is that credit is a necessary input for micro-enterprise or SME growth and that without credit, these enterprises cannot grow. However, it is important to point out that credit enables growth, but it is not responsible for growth. Although there is the perception that finance is a binding constraint, there is little empirical evidence to verify this situation in Argentina. Finance will be a binding constraint only when all other elements are in place, but only finance is missing. As shown in Chapter 5, the condition under which credit is available, such as high interest rates and collateral requirements seems to be the more fundamental constraint. Although there may not be a simple solution to increase access to finance, the Government has various types of instruments available to deepen financial markets, not only for SME lending, but for all types of lending to private economic agents. In particular, the Government could: * Contribute to the reduction of information asymmetries in credit markets that are particularly acute for micro firms and SMEs. Lack of detailed micro data on firms is likely a major reason why banks have not developed large portfolios of small firms loans. Developing credit information registries, which provide rapid access to standard information on borrower behavior from a variety of sources, such as lenders and utility companies and making a borrower' credit history available to all lenders, allows sorting potential borrowers at low costs and reduces adverse selection problems (Galindo and Miller, 2001). Credit registries may provide additional incentives for timely repayment because they allow borrowers to create a "reputation collateral " in credit markets as good credit risk. Using credit scoring technologies banks can take advantage of such information to assess the risk and profitability of SME loans. * Reform the legal and institutional environment for using movable assets to secure financial transactions. Movable assets (equipment, machinery, inventories, livestock, and accounts receivable), which constitute the bulk of SMEs' capital, had very limited capacity to carry debt. Although the government has introduced some changes, present legislation excessively restricts the registration and enforcement of security interests. An overhauled of the system will improve the working of credit markets in Argentina. * Reform the legal framework for risk capital investment. As presented in Chapter 5, present legislation increases the costs of investments' due diligence, structuring, managing, and exiting. * Strengthen retail lending operations. Successful financing to SMEs will require the ability of the financial sector to supply financial services to SMEs profitably because the large majority of potential clients are small and make jointly business and family decisions. Government support may be justified to provide technical assistance to private lenders to accelerate their understanding of the SME sector, of how to evaluate their activities, and how to apply different evaluation methods. 83 * Assist within SME program, in the training and preparation of business plans and financial projections. * Promote innovation in specialized lending technologies that reduce the administrative costs associated with credit application, monitoring, and payment * Eliminate all subsidized credit lines. The international experience shows that directing and subsidizing credit in the face of high costs and macro economic distortions will only perpetuate the existing problems by crowding out private lenders (see Chapter 7). However, financial assistance or matching complements to financial institutions with a significant SME lending portfolio, to reduce transaction costs of the processing of small loans should be considered. Likewise funding for the jump-start of financial institutions programs to expand their lending to SMEs can also be considered. * Reconsider government direct interventions in credit markets. As shown in Chapter 5, econometric results show that public banks lend less (as a share of total lending) to SMEs than domestic banks. Thus, privatization of public banks may not be detrimental for SME lending. 8.5 SME Assistance Programs: Institutional Design and Directions for Reform Beyond the elements that an effective SME's assistant program ought to have identified in Chapter 7, the findings of this report suggest the following three set of measures: 8.5.1 Consolidation of existing programs The first priority is critical evaluation of the existing programs: discontinuation of small and less useful programs and consolidation of those small programs which valued by its SME customers into larger and more visible programs. This puts evaluation of the existing efforts at the forefront of the policy agenda. The remaining programs need to be decentralized and become more demand-driven and more client- driven. Good SME prograrn is a private-public partnerships with private beneficiaries in the driving seat in terms of program design and possibly financing. 8.5.2 Piloting of new demand-driven decentralized SME programs Preference shall be given to programs with well-defined objectives, types of interventions and set of clients. Among such programs are: * Linkage promotion and supplier development programs. In such programs large buying acts as a mentor to its small potential and actual suppliers. Public sector plays a limited role of a broker to bring together small and large companies (see Box 7.3) * Program to facilitate access to IT services and usage of Internet * 'Light touch' technical assistance to local economic development agencies, SME agencies and private-public cluster initiatives. This is particularly important for less-developed provinces with a depressed productive sector and prudent public management (see also Box 8.4) * In all initiatives, the critical factor of success will be a performance orientation of SME agencies - a link of the remuneration of the agencies' staff and funding with the performance improvement of the assisted SMEs (see Box 7.1). 8.5.3 Design and nation-wide discussion of a new system of business development services to SMEs. Such a system needs to be re-invented on three levels: federal, sub-national and of civil society. 84 At the federal level-the Government should engage in continuous monitoring, evaluation, and technical assistance to state-level initiatives. The following are the activities necessary for a successful that should be at the center of the actions of the federal government: * Consolidation of existing programs. * Making Federal programs driven by the demand of the provinces and private sector clients * Making Federal budgeting flexible, capable of adapting to demand and building upon experience, and mixing its funds with those of other sources as outlined below * Building in mechanisms for evaluation, for the identification of best practices at the state level and for collecting, evaluating, and disseminating international experiences * Built-in feedback from the state in a process of continuous improvements of their tools * Attracting top caliber talent to promote and facilitate the planning process, to administer the operating program and to guide experimentation and evaluation At the provincial level, the Government should pilot a set of innovative private sector driven initiatives (entry points). * Grown out of discussion and debate within the civil society * Drawn upon active participation of the private sector in financing, conception, and operation * Built-in mechanism for evaluation and improvement in the light of the state's own experience, experience elsewhere in Argentina, and experience abroad At the level of the civil society, the Government could initiate a national process of vision- building and construction of shared change agenda with the civil society. 8.6 Longer-term options to reform the SME promotion system Section 8.2 outlined immediate priorities to jump-start the reform of the SME promotion system in order to deliver growth. The current crisis was viewed as a window of opportunity to initiate the reform. This section addresses a longer-term options to create a new cooperative framework between federal, state government and private sector to develop and implement effective SMEs assistance programs. 8.6.1 Transition at the Federal level New Role. The new role of Federal government is threefold. First, the federal government should promote the process of discussion, debate, and discursive problem solving at the state and local levels. This process will result in private-public initiatives - entry points to unleash private sector dynamism. Second, it should support that process with technical assistance and the mechanism of evaluation, continuous improvement, and program innovation. Most importantly, it must draw into this process a well-trained and committed group of men and women of the caliber of those that in the past supported state-led import substitution and, in the last ten years, devoted their talent to macro-economic management, privatization, and the negotiation of international treaties. Third, the federal government should facilitate the national process of deliberation and sharing of knowledge about promising state and sectoral initiatives involving all strata of civil society, private sector, and the government. Program Consolidation. The first step to reorganize federal programs is to consolidate federal expenditures so that all programs are grouped together in a single budget. Ideally, that budget should be 85 administered by one single agency, however, administrative consolidation may not be possible initially. Nevertheless, the policy making process should be organized around a consolidated budget. The second step in the reform of promotion policy would be to divide the consolidated budget into three broad functional categories: planning, operations, and administrative support for program evaluation, technical assistance, and promotion. The planning function and budget would be almost completely decentralized to the provinces. Operational funds would be allocated to the provinces through a voucher- like system that would enable them to purchase program components from Federal agencies. The Federal Government would retain responsibilities for Program Evaluation, technical assistance, and promotion, and set up new mechanisms to accomplish these tasks. 8.6.2 Institutional design of the new decentralized system The following three elements form the basic institutional architecture of transition to a new decentralized system: a) National Advisory Council acting as a distinct regulator and financier at the third level (tercer piso). Ideally, a private-public council with substantial private sector participation would be in charge of defining the rules of the game (Statement of Policies and Operating Procedures), and entrusted with ex-post monitoring control to ensure the well-functioning of all operating procedures. b) An independent management unit operating at the second level (segundo piso). This management unit will ensure that there is no conflict of interest regarding how to disburse funds. The unit responsible for administering resources and intermediating between supply and demand ideally should be credibly independent and follow transparent operating procedures for resource allocation. The resource allocation procedures should be designed to minimize any tendency for corruption and clientelism, and could range from (i) first-come first-served, to attempt to minimize discretion in the allocation of subsidies to (ii) a more elaborate contest where entrants (enterprises and associated public and private service providers) would have to explain their accomplishments to date and how new activities could strengthen themselves, existing institutions and the economy. c) A diverse array of national and international private service providers operating at the first level. These suppliers are generally private sector consultants, but could also.include public sector service providers as long as they charge for their services and compete on an equal basis with other private providers. As a guarantee of the relevance and need of any service to be provided to the enterprises themselves, it would be pre-requisite that the enterprises contribute a strong element of their own private financing, typically at least 50 percent towards the costs of fees and associated expenses of external support services. Such a local investment environment program can be designed with an option of the following three guiding principles: the matching grant principle, the business-climate ranking system, and competition for open-ended projects. 86 The Matching Grant Principle. Under this principle the federal government simply agrees to match every peso, up to a certain limit, that provincial governments dedicate to economic-development projects decided in collaboration with private actors, on condition that those actors match the state contribution as well. The idea is simply that if the regional government and economic actors are willing to put their own money at-risk to finance the projects they define together, then the federal government can assume that their choices are well considered, and add its backing as well. The advantage of this method is to impose some discipline on project selection with almost no increase in red tape. The economic actors, public and private, at the regional level are given an incentive to sort through their priorities-start looking for trouble-and the federal government limits itself to ratifying their provisional decisions as they emerge. One shortcoming of the program is that it foresees next to no sharing of experience at the national level, and little formalization of the experience shared in arriving at projects regionally or locally. Word of success presumably gets around-our official is not the only one on horseback these days--but the process of deliberation about what projects to choose, and which are in fact working, remains a black box from the official point of view. So political and bureaucratic meddling is discouraged, but at the price of a kind of willful ignorance of what the actors might be learning. The second drawback is simply that the framework funds those who are on the verge of funding themselves. It thus comes close to violating the fairness requirement even though its operation is nothing but even-handed in the way it treats (the fortunate) likes alike. The Business-Climate Ranking System. * The second framework also aims to discipline project selection while holding bureaucracy and the politics of clientelism generally in check. But it does this not by ratifying the actors decisions, but rather by providing information on economic performance that cause them to reflect on their possibilities in new ways. The crucial information intervention could be the creation of a league table of regional economic performance Beginnings of such a ranking already exist in the comparative rating of regional performance established through the joint IBRD-IDB provincial public works program (see Box 8.5). 87 Box 8.5: Argentina: Co-investment and Business-Climate Ranking as Criteria for Funding Provincial Public Works Elements of the first two models have come almost In practice, the PIU is responsible for monitoring the unintentionally to inform a successful program, under fiscal performance and projects of the provinces via the which the IBRD and then the IDB provide low interest, PAFIs as well as presenting recommendations to the long-term loans to provincial governments in Argentina. Creditors Committee on the approval or rejection of The loans fund administrative and fiscal reforms or PAFIs and projects. The PIU currently has a staff of public works. Three general and conventional rules set about 100 provincial and technical/functional experts in by the multilateral agencies govern the allocation of areas such as accounting, contracting, architects, funds among competing provinces. The first imposes an engineers, who PIU provides free technical assistance to extensive requirement to co-invest: provinces must offer the provinces. a share of its coparticipacion tax revenues as collateral to the general loan fund, co-finance (50%) of any public Because of its central position, and ample staff, and works project, and finance and staff the provincial because the line between monitoring and assistance is implementing units. The second establishes thresh-hold often blurred, the PIU is often more engaged in requirements for macro-economic performance: to benchmarking and teaching than monitoring of contract access loans for public works (new or in progress), compliance. For instance, when a province is unable to participating provinces must meet two strict criteria - a meet the above-mentioned fiscal criteria, the PIU assists current account fiscal surplus and a level of debt service the province in reorganizing its finances and identifying that is less than 15 percent of current revenues. solutions to, say, improve the revenue base, and thus Performance on both accounts must be reported to a aids in the development and implementation of a new federal Project Implementation Unit (PIU) by the PAFI. Third, the PIU organizes and co-finances (with province in its Financial Action and Investment Plan the provinces) training programs for provincial (PAFIs). If a province fails to meet these two criteria, it managers and, occasionally, relevant civil servants of the must develop and implement with the PIU a new provincial administrations. financial plan (including debt refinancing, improved The central role of the federal PIU unit has thus come to local tax revenues, restructuring of expenditures, etc.) be more that of facilitating learning between provinces before a loan is granted. The third criterion concerns the and improving both its monitoring and technical utility of the projects. Thus projects to build or improve assistance methods. In addition it acts as an effective schools, hospitals, sanitation, or other infrastructure, interlocutor between the provinces, multilateral lending must demonstrably augment the capacity of provinces to agencies, and the national government have been in the take on the programs that are being decentralized to innovations in the governance structures and the them from the national government. Fourth and finally, incoration in ang. In ang moe successful applications document how the proposed benchmarking methods, the PIU conducts periodic project will meet current standards for accounting, strctured cmpionso the andu betwee financial~ ~ maaeet idn,adcnrcig structured comparisons over tbme and between provinces. In so doing, the PIU learns how to improve It is the innovative role of the federal PrU, financed its own project and PAFI evaluation procedures and mainly by a small percentage of the loans, that makes technical assistance as well as helps the provinces learn the loan program distinct in a way that illustrates the from one another and adopt preventive strategies. As it potential use of co-investment and benchmarking as demonstrates its ability to monitor and assist the instruments of developmental learning. Formally the provinces, the PIU, in turn, has built up reputational PIU acts as an agent of multilateral lending institutions capital vis-a-vis private creditors (i.e., in mediating for and as a monitor and service provider for both the provinces for new non-program projects) and the national and provincial governments. IBRD/IDB (i.e., in presenting proposals for changes in project criteria and monitoring rules as well as new provincial-level programs). A creative example of such benchmarking is a Mexican example of a three-tier approach to lending. By law, sub-national governments in Mexico cannot borrow from foreign lenders. A new regulatory regime for private banks links their reserve requirements to the risk rating of their prospective public-sector borrowers. Loans to the least risky public authorities require a reserve of 20 percent of the loan amount, 88 whereas the reserve requirement is raised to 115 percent of the amount lent to the most risky customers. To ensure that all relevant loans are reported to the federal authorities and available to the outside agencies that actually rate creditworthiness, all unregistered loans are assigned a prohibitive reserve requirement of 150 percent. The upshot of the system is to match borrowers of all levels of creditworthiness to the type of lending institution best able to serve them. The lowest-risk public entities are reasonably the province of domestic private banks, which have neither the competence to provide technical assistance to distressed borrowers, nor the political power to resist the machinations of local governments that fall behind in their debt service. Higher-risk borrowers, with reserve ratings of up to 100 percent, are served by federal development banks, which have experience in dealing with vulnerable but well intentioned borrowers. Finally, sub-national loans with risk weightings of more than 100 percent are allowed if the loan package contains a technical assistance component funded by an international development bank. In this way parties-and potential parties-on both sides of the transaction know what they are getting into, and take precautions accordingly. In time, furthermore, the benchmarking criteria can be refined in the light of lending experience, and the technical assistance programs can themselves become open-ended frameworks that use the refined benchmarks as reference points for creating additional programs with vulnerable SMEs. As the second, benchmarking-based type of framework program develops, adding technical services and becoming self-reflective about benchmarking criteria, it naturally grows into a third type of program that we will call contest-based. Contests for funds to unleash provincial private sector development This type uses benchmarking both as a way of selecting projects and as a way of forcing discussion about the criteria of project selection. Very generally it operates on two levels: At the "top" a benchmarking committee of the relevant government entities and qualified private actors, domestic and foreign, collaborates with potential users to establish the initial substantive and procedural criteria for participation, and defines the initial metrics by which applications are to be ranked. At the "bottom" project groups compete to present projects that score highly under the emergent criteria. After each round of projects the selection criteria and benchmarks are adjusted to reflect improved measures of performance and a richer understanding of success. At least two variants are imaginable, depending on whether projects concern provinces or other broad jurisdictions of government on the one hand, or groups of firms typically operating in many different political jurisdictions on the other. Consider as an illustration of the first possibility a national program to improve the business environment along, roughly speaking, the seven dimensions listed above. As the provinces by their particular regulations and practices define distinct business environments, collaborative elaboration of benchmarks and selection criteria will necessarily entail ajoint effort by representatives of the provinces and representatives of national institutions, augments by the relevant private-sector actors- Project funds are then allocated to provinces according to the criteria and actual rankings; provinces will then allocate the funds they have received to individual projects, applying the national procedures and criteria where appropriate and augmenting these with local ones as required. Competition for funds drives efforts at improvement. Provinces that do not rank well initially-and projects within high ranking provinces that fare poorly in early selection rounds-will be eligible for technical assistance according to criteria established again through national and local collaboration. The types of technical assistance provided, and the level at which they are funded, will be adjusted in a rhythm synchronized with the revisions of the performance metrics and selection criteria. A match may emerge between certain types of service providers and certain groups of clients along the lines of the banking example introduced a moment ago; but the creation of such links will be a result of the operation of the framework, not a design decision. 89 The supplier-improvement project works in a similar way, except that there are no block grants to the provinces or any other sub-national jurisdiction as such. But this does not mean that the public sector is excluded from the program. On the contrary: groups of firms will presumably partner with various government entities as well as with each other in applying for support directly to a national selection committee. For its part that committee makes (improvement of) the breadth and depth of public-private relations in this regard one of its principal selection criteria. The disadvantage of this third type of open- ended program regards govemability. The constitution and operation of the criteria-selection committee creates opportunities for abuse that are avoided in both the matching-grant and once7and-for benchmark schemes. The dangers can be in part mitigated by emphasis on procedures: making the use of transparent methods for selecting program officers and project members a formal criterion for selection, and then using the information that these procedures produce to identify suspect operations. And again the nature of the projects themselves and the information that they produce about interim results reduces the threat of corruption or cronyism. Active involvement of outsiders with incentives to denounce collusion can help as well. But all these are palliatives, not cures, and it may be that open-ended, competitive frameworks are best developed on the basis of the experience accumulated in various combinations of the less burdensome types. Box 8.6 summarizes experience of contests in the World Bank projects. 90 Box 8.6: Contests for federal funds between sub-national governments: Lessons from the World Bank projects Russia Fiscal Reform Loan: Regional Reform Ukraine Private Sector Development: Adaptable Lending Fund. Program In the project, resources from the Regional Ukraine Private Sector Development Adaptable Lending Reform Fund will be awarded on a competitive program envisions matching funds support to enterprise basis to pre-qualified regions that demonstrate restructuring in oblasts (sub-national analogue of Argentinean improved fiscal performance and progress in provinces) which are selected competitively. The contest implementing a program of regional fiscal reform. between oblasts is based on their progress in monitorable These programs have been elaborated with the indicators of regulatory environment. These indicators are help of the Regional Fiscal Technical Assistance measured bi-annually in all 26 oblasts through an enterprise Project. A detailed approach to measurement of survey. Every two years of the adaptable lending program, two progress has been developed by the Ministry of new oblasts are added to the program and these are precisely Finance in conjunction with the the oblasts recorded as the largest improvement in regulatory environment for business. A more controversial feature of the Bank. In the first tranche the Bank will agree on contest is exit criteria. The loan appraisal document envisions the basic operating procedures. Subsequent exit from the projects of the oblasts with the worst performance tranches will call for refinement of selection in terms of regulatory environment. It remains to be seen criteria and implementation of the Regional whether both the Bank and the Federal Government of Ukraine Reform Fund satisfactory to the Bank. will be able to adhere to this strict exit criterion. 8.7 Conclusion In summary, for SMEs to realize their potential, the business environment can and should be improved with relatively little cost and effort, the financial sector needs to be tinkered with at least at the margin with better instruments and lower transaction cost and that will require some effort. Finally SME assistance programs need to be rethought and redesigned for increased effectiveness. The principles are incentive driven components, discriminatory approaches adapted to capabilities, self-selectivity and gradualist support linked to performance. The benefits to be derived are large both in terms of employment and productivity-increased economic activity, leading to a buoyant SME sector. 91 9. References ABA (2000), "La Evoluci6n del Sistema Financiero en la Decada del Noventa," http://www.aba- argentina.com/Entrada.html Abramson, Patty (2000), "Best Practices for Financing Women-Owned Businesses: The Women's Growth Capital Fund and WomenAngels.net," Presentation for the 2nd OECD Conference on Women Entrepreneurs in SMEs: Realizing the Benefits of Globalization and the Knowledge-based Economy. Acs, Zoltan and David Audretsch (1990), Innovation and Small Firms, Cambridge: MIT Press. Acs, Zoltan and David Audretsch (1993), Small Firms and Entrepreneurship: An East-West Comparison. Cambridge: Cambridge University Press. Acs, Zoltan, J. Bo Carlsson, and Charlie Karlsson (1999), Entrepreneurship, Small and Medium-Sized Enterprises and the Macroeconomy, Cambridge: MIT Press. Alston (2000), "Foreign Privacy Laws and Proliferate: New Laws in Argentina and Australia have Extraterritorial Application," http://www.alston.com/ Angelini, P., R. Di Salvo and G. Ferri (1998), "Availability and Cost of Credit for Small Businesses: Customer Relationships and Credit Cooperatives," Journal of Banking and Finance, 22 (6-8), 925-54. Arango, Carlos and William F. Maloney (2000), "Employment, Informality, and Poverty in the Argentine Labor Market," Background Paper No. 3 in Argentina Poor People in a Rich Country. Report No. 19992-AR. The World Bank. Araujo, L. and L. Easton (1996), "Networks in Socioeconomic Systems," A Critical Review, In lacobucci, D (Ed.) Networks in Marketing, Newbury Park, Cal.: Sage. Arrigoni, Miguel Angel (2000), "ABA 2000 La Banca Informal,"ABA's web-page. Audretsch, David (1995), Innovation and Industry Evolution. Cambridge: MIT Press. Avery, Robert B., Raphael W. Bostic and Glenn B. Canner (1998), "CRA Special Lending Program," Journal of Banking and Finance 22, (8), 1019-61. Avery, Robert B., Raphael W. Bostic, and Katherine A. Samolyk (1998), "The Role of Personal Wealth in Small Business Finance," Journal of Banking and Finance 22 (6-8), 1019-61. Banco Central de La Republica Argentina (2000), "Las MIPyMEs y el Mercado de Credito en la Argentina," unpublished manuscript. BankWatch (2000), "Argentine Banks - Results for FY99 and lHOO." Barajas, A., R. Steiner and N. Salazar (1999), "The Impact of Liberalization and Foreign Investment in Colombia's Financial Sector," unpublished manuscript. Basch, Miguel and Camilo Morales (1995), Expanding Access To Financial Services In Latin America. Baltimore: The Johns Hopkins University Press. Batra, Geeta and Andrew Stone (2000), "The Business Environment in Argentina: Results from 'a Private Sector Survey," Unpublished manuscript, The World Bank. Bergam, Paul (2000), "Argentine SMEs: The Uneven Scale, http://www.trading-news.com/ingles/7- 00/smes.htm 92 Berger, A., A. Saunders, J. Scalise, and G. Udell (1998), "The Effects of Bank Mergers and Acquisitions on Small Business Lending," Journal of Financial Economics, 50 (2), 187-229. Berger, A., R. De Young, H. Gency, and G. Udell (2000), "Globalization of Financial Institutions: Evidence from Cross-Border Banking Performance," Brookings-Wharton Papers on Financial Services, No 3. Berger, Allen N. (1998), "The Effects of Bank Mergers and Acquisitions on Small Business Lending," Journal of Financial Economics, 50 (2), 187-229. Berger, Allen N. and Gregory F. Udell (1998), "The Economics of Small Business Finance: The Roles of Private Equity and Debt Markets in the Financial Growth Cycle," Journal of Banking & Finance 22(--), 613-673. Berger, Allen N., Leora Kappler, and Gregory F. Udell (2000), "The Ability of Banks to Lend to Informationally Opaque Small Businesses," Unpublished manuscript, The World Bank. Berlin, Mitchell and Loretta J. Mester (1998), "On the Profitability and Cost of Relationship Lending," Journal of Banking and Finance, 22 ( 6-8), 873-97. Berry, Albert (1996), "Small and Medium Enterprise (SME) Under Trade and Foreign Exchange Liberalization: Latin American and Canadian Experiences and Concerns," Canadian Journal of Development Studies, 17 ( 1), 53-72. Berry, Albert (1997), "SME Competitiveness: The Power of Networking and Subcontracting," Sustainable Development Department, Inter-American Development Bank, Washington, D.C. http://www.iadb.org/sds/SME/publication/publication_168 27_e.htm Bisang, R., G. Burachik and J. Katz (1995), Hacia un Nuevo Modelo de Organizaci6n Industrial. El sector manufacturero argentino en los afios 90. Alianza Editorial/CEPAL/United Nations University Press. Bleger, Leonardo (2000), "El Proceso de Concentraci6n y Extranjerizaci6n del Sistema Bancario Argentino durante los 90," Separata: Boletin Informativo Techint 301. Bleger, Leonardo and Guillermo Rozenwurcel (2000), "Financiamiento a las Pymes y Cambio Estructural en la Argentina, Un Estudio de Caso sobre las Fallas de Mercado y Problemas de Informaci6n," Desarrollo Econ6mico: Revista de Ciencas Sociales, 40 (157). Brewer, Elijah., Hesna Genay, William E. Jackson, and Paula R. Worthington (199x), "How are Small Firms Financed? Evidence from Small Business Investment Companies," Federal Reserve Bank of Chicago Economic Perspectives, 2 (18). Brugger, Ernst and Sarath Rajapatirana (1995), "New Perspectives On Financing Small Business In Developing," Countries. San Francisco: ICS Press. Casaburi, Gabriel (1999), "Dynamic Agroindustrial Clusters: The Political Economy of Competitive Sectors in Argentina and Chile," New York: St. Martin's Press. Casaburi, Gabriel and Pablo Angelelli (1999), Estrategias de Apoyo a la Micro, Pequefna, y Mediana Empresa. IERAL-PNUD, Buenos Aires. Caves, Richard E. (1998), "Industrial Organization and New Findings on the Turnover and Mobility of Firms," Journal of Economic Literature, 36 (4), 1947-1982. CEDES (2001), "Determinants and Consequences of Financial Constraints Facing Firms in Argentina," unpublished manuscript prepared for the IADB. CEMA and BCRA (2001), "Credit Constraints Facing Firms in Argentina." unpublished manuscript prepared for the IADB. 93 Central Bank of Argentina (2000), "Las MIPyMEs y el Mercado de Credito en la Argentina," unpublished manuscript. Gerencia de Investigaci6n, Area de Economia y Finanzas. Central Bank of Argentina (2000), "Texto Ordenado Actualizado de las Normas sobre Clasificaci6n de Deudores", Comunicaci6n "A" 2950. CEPAL (1999), "Hacia un mejor enterno competitivo da la producci6n automotriz en Argentina," PROYECTO ADEFA - CEPAL (Bernardo Kosacoff, Project Director). Cheques Sin Alas, Nuevos Instrumentos, Los Bancos del Mundo Christen, Robert P. (2000), "Commercialization and Mission Drift: The Transformation of Microfinance in Latin America," Consultative Group to Assist the Poorest, Washington, D.C. http://www.cgap.org/assets/images/LAC_Commercialization l .pdf Chudnovsky, D., A. Lopez and F. Porta (1996), Los Limites de la Apertura, Liberalizacion Industrial y Medio Ambiente. Buenos Aires: Editorial Alianza. Claessens, S., A. Demirguc Kunt and H. Huizinga (2000), "How Does Foreign Entry Affect the Domestic Banking Market?," Policy Research Working Paper 1918, The World Bank: Washington, D.C. Clarke, George R. G., Rorbert Cull, L. D'Amato and A. Molinari (1999), "On the Kindness of Strangers? The Impact of Foreign Entry on Domestic Banks in Argentina," Policy Research Working Paper No. 2158, The World Bank: Washington, D.C.. Clarke, George R.G., Robert Cull, Laura D'Amato and Andrea Molinari (1999), "The Effect of Foreign Entry on Argentina's Domestic Banking Sector," Policy Research Working Paper No. 2158. The World Bank, Washington, D.C. Cole, Rebel (1998), "The Importance of Relationships to the Availability of Credit," Journal of Banking and Finance 22, 959-977. Cole, Rebel A., John D. Wolken and R. Louise Woodburn (1996), "Bank and Nonbank Competition for Small Business Credit: Evidence from the 1987 and 1993 National Surveys of Small Business Finances," Federal Reserve Bulletin, 82 (11), 983-95. Cole, Rebel A., John D. Wolken and R. Louise Woodburn (1996), "Bank and Nonbank Competition for Small Business Credit: Evidence from the 1987 and the 1993 National Surveys of Small Business Finances," Federal Reserve Bulletin 82 (11), 983-995. Commission of the European Communities (1996), "Enterprises in Europe." Consejo Coordinador Empresarial (2000), "Calidad del Marco Regulatorio en las Entidades Federativas Mexicanas: Estudio Comparativo 1999," http://www.cce.org.mx/ Cowling, Marc (1999), "The Incidence of Loan Collateralization in Small Business Lending Contracts: Evidence from the UK," Applied Economics Letters, 6 (5), 291-93. Cr6dito Accesible para Pymes? Nuevos Instrumentos. Los Bancos del Mundo Cristini, Marcela and Ramiro Moya (1999), "El Credito para las Microempresas: Una Propuesta de Institucionalizaci6n para la Argentina," Documento de Trabajo No. 61. Fundaci6n de Investigaciones Econ6micas Latinoamericanas, Buenos Aires, Argentina. Cristini, Marcela, Ramiro Moya and Andrew Powell (2001), "The Importance of an Effective Legal System for Credit Markets: The Case of Argentina," Research Network Working Paper #R-428, Inter-American Development Bank, Washington, D.C. Cuevas, Carlos and Steven Schonberger (1998), "Argentina: Rural Finance Improving Access to Financial Services by Small Producers," mineo. Washington, D.C. 94 Cunningham, Wendy (2000), "Mexican Female Small Firm Ownership: Motivations, Returns, and Gender" LCSPR - Gender. The World Bank. Dages, Gerard B., Linda Goldberg and Daniel Kinney (2000), "Foreign and Domestic Bank Participation in Emerging Markets: Lessons from Mexico and Argentina," FRBNY Economic Policy Review, Federal Reserve Bank of New York, 17 (36). De Soto, Hemando (1989), "The Other Path: The Invisible Revolution in the Third World," Harper and Row Publisher's New York. Deeg, Richard (1 999),"Finance Capitalism Unveiled: Banks and the German Political Economy." DeYoung, Robert (1998), "De Novo Banks and Lending to Small Businesses: An Empirical Analysis: Comment on Goldberg and White," Journal of Banking and Finance, 22 (6-8), 868-72. DeYoung, Robert, Lawrence G. Goldber and White (1999),"Youth, Adolescence, and Maturity of Banks: Credit Availability to Small Business in an Era of Banking Consolidation," Journal of Banking and Finance, 23 (2-4), 463-492 Duca, John V. (1998), "Comment on: The Importance of Relationships to the Availability of Credit," Journal of Banking and Finance, 22 ( 978-981). Elgue, Mario CGsar (1999), Globalizaci6n, desarrollo local y redes asociativas. Buenos Aires: Ediciones Corregidor. Elias, Victor J. (2000), "An Econometric Cost-Benefit Analysis of the Young Training Program in Argentina," Unpublished manuscript. Elizondo, Nestor (1992), "Illegality in the Urban Informal Sector of Mexico City," In Victor Tokam, Beyond Regulation: The Infornal Economy in Latin America, Lynne Rienner Publishers, Boulder & London. European Network for SME Research (1994), "The European Observatory for SMEs." Evans, Peter (1995), Embedded Autonomy: States and Industrial Transformation. Princeton: Princeton University Press. Fahri, Diana de Montalban (2000), "Sociedades de Garantias Reciprocas, " Revista Electr6nica de Derecho Societario, No. 1. Abril. http:/www.societario.com/revistas/l/doctrina/sgrdiana-i.htm FIEL (1996), Las pequenias y medianas empresas en la Argentina, Buenos Aires, Argentina: Fundaci6n de Investigaciones. Econ6micas Latinoamericana. FIEL (2000), La Economia Oculta en la Argentina, Fundaci6n de Investigaciones Econ6micas Latinoamericana, Buenos Aires, Argentina. Fiess, Geoffrey (1998), "Global Markets and National Politics: Collision Course or Virtuous Circle?" International Organization, 52 (4), 787-824. Fiess, Norbet M., William Maloney and Rashmi Shankar (2000), "Informality, Wage Rigidities, and Real Exchange Rates," Unpublished Manuscript (do not cite without author's permission) Fuchs, M. and Kosacoff, (1992), "Balance del Comercio Internacional de Manufacturas de Argentina: Las tendencias al incremento del comercio intrasectorial 1974-1990," Working Paper No. 47 (LC/BUE/L. 129), Economic Commission for Latin America and the Caribbean (ECLAC), Buenos Aires. Fundes (1999), "El Problema del Entorno de Negocios: Desarrollo Competitivo de las PYMES Argentinas," Fundes-Argentina and Universidad Nacional General Sarmiento. Fundes (1999), Boletin de la Empresa Media. 7 (99), Mexico, D.F. No. 4 95 Fundes (2001), "Ventajas Competitivas Dinamicas: Las PYMES Exportadoras Exitosas Argentina," Buenos Aires, Argentina. Fundes-BID (2000), "Proyecto Estimulos para la Formalizaci6n de Microempresas," www.fundes.org/argentina Gatto, F. (1994), El desafio de la intemacionalizacion y el MERCOSUR para las Pymes argentinas. (LC/BUE/R.216), Buenos Aires: Economic Commission for Latin America and the Caribbean (ECLAC). Gatto, F. (1995), "Las exportaciones industriales de pequenas y medianas empresas," in B. Kosacoff, (1995), Hacia una nueva estrategia exportadore: la experiencia argentina, el marco regional y las reglas multilaterales. Buenos Aires: Universidad National de Quilmes. Gatto, F., V. Moori-Koenig and G. Yoguel (1993), "Reflexiones Sobre la Competitividad de las Empresas Pymes en el Nuevo Escenario de Apertura e Integraci6n: la Situacion de las Firmas Metalmecanicas," Integracion Economica Series, Working Paper No. IE/03, Secretaria de Programacion Economica. Buenos Aires. Gatto, F.Y. and G. Yoguel (1993), "Las PyMEs Argentinas en una Etapa de Transici6n Productiva y Tecnol6gica'in Kosacoff," in El Desafio de la Competitividad, La Industria Argentina en Transformaci6n. Buenos Aires:' CEPAL/Alianza Editorial. Gatto, Francisco and Carlo Ferraro (1997), "Consecuencias Iniciales de los Comportamientos PYMES en el Nuevo Escenario de Negocios en Argentina Vol. I" CEPAL, Buenos Aires, Argentina. http://www.eclac.org/publicaciones/BuenosAires/2/lcbuel 162/indice.htm George R., G. Clarke, and Robert Cull (1997), Why Privatize? The Case of Argentina's Public Provincial Banks. The World Bank, Washington, D.C. George, R., G. Clarke and Robert Cull (1997), The Policital Economy of Privatization: An Empirical Analysis of Bank Privatization in Argentina. The World Bank, Washington, D.C. Gemot, Grabher and David Stark (1997), "Restructuring Networks in Post-Socialism: Legacies, Linkages, and Localities," Oxford: Oxford University Press. Giordano, Osvaldo and Jorge Colina (2001), "Los Costos de la Formalizaci6n para las Micro, Pequefilas, y Medianas Empresas en la Argentina," background note prepared for this report. Giordano, Osvaldo, Alejandra Torres, Mara Bettiol, Jorge Colina, Pablo Mazzino, and Gustavo Martin (1999), "La Simplificaci6n de la Registraci6n Laboral: Relevamiento de Antecendentes." Goldberg, Lawrence G.and Lawrence J. White (1998), "De Novo Banks and Lending to Small Businesses: An Empirical Analysis," Joumal of Banking and Finance, 22 (6-8), 851-67. Grabher, Gemot (1993), "The Embedded Firm: On the Socioeconomics of Industrial Networks," London: Routledge. Guasch, J. L. and J. Kogan (2000), "Logistic Costs and Competitiveness in Latin America," World Bank Manuscript. Gudger, Michael (1998), "Credit Guarantees: An Assessment of the State of Knowledge and New Avenues of Researh," FAO Agricultural Services Bulletin 129. Rome: FAO. Guillen, Kristin "A Market-Oriented Strategy for Small and Medium-Scale Enterprises," Intemational Finance Corporation Discussion Paper No. 40. The World Bank, Washington, D.C. Guillen, Mauro (2001), The Limits of Convergence: Globalization and Organizational Change in Argentina, South Korea, and Spain. Princeton: Princeton University Press. 96 Guillen, Mauro and Adrian Tschoegl (1999), "The New Conquistadors: Spanish Banks And The Liberalization Of Latin American Financial Markets," Unpublished Manuscript, management Department, Wharton School. Harhoff, Dietmar and Tim Korting (1998), "Lending Relationships in Germany--Empirical Evidence from Survey Data," Journal of Banking and Finance, 22 (10-11), 1317-1353. Haynes, George W. (1996), "Credit Access for High Risk Borrowers in Financially Concentrated Markets: Do SBA Loan Guarantees Help?" Small Business Economics, 8, (--),449-46 1. Haynes, George W. (1996), "Credit Access for High-Risk Borrowers in Financially Concentrated Markets: Do SBA Loan Guarantees Help?" Small Business Economic, 8(--), 449-461. Helper, Susan., John Paul MacDuffie, and Charles Sabel, (2000), "Pragmatic Collaborations: Advancing Knowledge While Controlling Opportunism," Industrial and Corporate Change 9 (3), 443-488. Hernandez, Rafael Mayoral and George A. Baptista (2001), "The Need for Legal and Regulatory Reforms in Argentina, Brazil, Chile, El Salvador and Mexico to Promote Risk Capital Investments in Small and Medium Size Enterprises," report presented at the Workshop on Small and Medium Sized Enterprises' Access to Risk Capital. Inter-American Development Bank. Herrigel, Gary B. (1989), "Industrial Order and the Politics of Industrial Change: Mechanical Engineering," in Industry and Politics in West Germany, edited by Peter Katzenstein, Ithaca: Cornell University Press. Holden, Paul (1997), "Argentina's Financial System: Policy Note," Unpublished manuscript. Holden, Paul "A Description and Analysis of the FUNDES Guarantee Scheme for Lending to Small and Medium Sized Enterprises," Unpublished manuscript. Hughes, Alan and David J. Storey (1994), Finance and the Small Firm. New York: Routledge. IAMC (1999), Politicas para las Pequefnas y Medianas Empresas: Evaluaci6n y Propuestas. Buenos Aires: Instituto Argentino del Mercado de Capitales. IDB-Nexus Associates (2000), "Evaluation of the Experience of the Centros de Desarrollo Empresarial: Lessons Learned from CDE Rafaela-Esperanza in Argentina," Advisory Groups for SME, IDB. IDB-SME (2000), "IDB Group Activities Supporting SMEs (1990-1999)" Sustainable Development Department, Inter-American Development Bank, Washington, D.C. IdePyME -Set of various documents downloaded from the web IERAL (1999), Estrategias de Apoyo a las Micro, Pequefias y Medianas Empresas. Buenos Aires: IMF (2000), "The Role of Foreign Banks in Emerging Markets," International Capital Markets: Developments, Prospects, and Key Policy Issues, IMF, September. Inter-American Development Bank (1998), "Microfinance Guarantees: Is There Another Model?," Microenterprise Development Review, 1(2). Jansson, Tor and Santiago Sedaca (1999), "Formalizing Small Businesses: International Best Practices and the Case of Colombia," Microenterprise Unit, Sustainable Development Department, Inter- American Development Bank, Washington, D.C. Jayaratne, Jith and John Wolken (1999), "How Important Are Small Banks to Small Business Lending? New Evidence from a Survey of Small Firms," Journal of Banking and Finance, 23 (2-4), 427-58. Jenkins, Hatice (2000), "Commercial Bank Behavior in Micro and Small Enterprise Finance," Development Discussion Paper No. 741. Harvard Institute for International Development, Harvard University. 97 Jenkins, Hatice (2000), "Commercial Bank Behaviour in Micro and Small Enterprise Finance," Development Discussion Paper No. 741. Harvard Institute for International Development- Harvard University. Kanengulser, Martin (2000), "Crecimiento Explosivo del Empleo en Negro," La Naci6n, Lunes 7 de febrero de 2000 Kantis, H. (1998), "Capacidades Estrategicas de Las PyMES Argentinas: Elementos Conceptuales y Evidencias Empiricas," In Kantis, Desarrollo y Gesti6n de PyMES: Aportes Para un Debate Necesario" UNGS. Kantis, H. (1999), "El caso de Rafaela: redes y desarrollo local," in M. Elgue, Globalizaci6n, desarrollo local y redes asociativas. Buenos Aires: Ediciones corregidor. Kantis, H. (1996), "Inercia e Innovacion en las Conductas Estrategicas de las Pymes, Elementos Conceptuales y Evidencias Empiricas," Working Paper No. 73 (LC/BUE/L. 156), Economic Commission for Latin America and the Caribbean (ECLAC), Buenos Aires. Kantis, H. (1998), Desarrollo y Gesti6n de PyMEs: Buenos Aires: Universidad General Sarmiento. PROMETEO. Kantis, H. and G. Yoguel (1989), "Eslabonamientos Productivos y Restricciones para el Desarrollo de las Pymes Metalmecanicas del Gran Buenos Aires," Working Paper No. 22 (LC/BUE/R. 163), CI:I- CEI'AL Programme, Economic Commission for Latin America and the Caribbean (ECLAC), Buenos Aires. Kantis, H., R. Ascuia, and N. Mahlet (1999), "Redes Empresariales y Desarrollo Local: el Caso de Rafaela," Paper presented in the I Jomadas de Investigaci6n organised by the Universidad Nacional de General Sarmiento, June 14 y 15 1999. Kantis, H., V. Moori-Koenig and G. Yoguel (1991), Exportaciones Pymes de Manufacturas de Cuero: Primeros Resultados del Trabajo de Campo. (LC/BUE/R.213), Buenos Aires: Economic Commission for Latin America and the Caribbean (ECLAC). Kantis, H., V. Moori-Koenig and G. Yoguel (1993), "Las Empresas Medianas y Pequenas de Bienes de Capital Ante la Apertura del MERCOSUR. Perfiles Empresariales y Escenarios Para Disenar Acciones y Fortalecimiento de su Competitividad," Working Paper No. 34 (LC/BUE/R.138), CFI- CEIPAI. Programme, Economic Commission for Latin America and the Caribbean (ECL.AC), Buenos Aires. Kao, R.W. Y. (1983), "Entry Barriers and New Venture Strategies," Journal of Small Business, 1 (2), 37- 42. Keeton, W. (1996), "Do Bank Mergers Reduce'Lending to Businesses and Farmers," Economic Review, 81 (3),.63-75, Federal Reserve Bank of Kansas City. Keeton, W. (1997), "The Effects of Mergers on Farm and Business Lending at Small Banks: New Evidence from the Tenth District States," Federal Reserve Bank of Kansas City, mimeo. Keeton, W. (1995), "Multi-office Bank Lending to Small Businesses: Some New Evidence," Economic Review 80 (2), Federal Reserve Bank of Kansas City; 45-57. Kennedy, Richard M. (1998), "Case Study on the Operation of Three Romanian Business Centres," UNIDO PDS Technical Working Paper No. 1. Kirzner, I. (1973), Competition and entrepreneurship. Chicago: University of Chicago Press. Kirzner, I. (1997), Entrepreneurial Discovery and the Competitive Market Process: An Austrian Approach, Journal of Economic Literature, 35, (--),0-85. 98 Kogut, B. (2000), "The Network as Knowledge: Generative Rules and the Emergence of Structure," Strategic Management Journal, 21, (--), 405-425. Kolari, J. and A. Zardkoohi (1997), Bank Acquisitions and Small Business Lending. Texas A&M University. Kolari, J. and A. Zardkoohi (1997), The Impact of Structural Change in the Banking Indstry on Small Business Lending. Texas A&M University. Kosacoff, B. (1996), "Estrategias empresariales en la transformacion industrial argentina," in Bolentin Techint No. 288, 25-47. Kosacoff, B. (1992), "El Comercio Intemacional de Manufacturas de la Argentina 1974-1990: Politicas Comerciales, Cambios Estructurales y Nuevas Formas de Insercion Internacional," Working Paper No. 51 (LC/BUE:/I.. 133), Economic Commission for Latin America and the Caribbean (ECL.AC), Buenos Aires. Kosacoff, B. (1993), "La Industria Argentina: un Proceso deRreestructuracion Desarticulad, in B. Kosacoff (1993), El Desafio de la Competitividad: La Industria Argentina en Transpormacion. Buenos Aires: Alianza Editorial. Kosacoff, B. (1993), El Desaflo de la Competitividad. La Industria Argentina en Transformaci6n. Buenos Aires: CEPAL/Alianza Editorial. Kuznetsov, Yevgeny N. (2000), "Resolving Mexico's Growth Paradox: Toward Decentralized Incentive Framework for Private Sector Development," The World Bank, Washington, D.C. Levenson, Alec R. and William F. Maloney (1998), "The Informal Sector, Firm Dynamics, and Institutional Participation," Policy Research Working Paper No. 1988, The World Bank. Latin America and the Caribbean. Poverty Reduction and Economic Management Sector Unit. Levine (1999), Foreign Bank Entry and Capital Control Liberalization: Effects on Growth and Stability. University of Minnesota. Levonian, Mark and J. Soller (1995), "Small Banks, Small Loans, Small Businesses," Federal Reserve Bank of San Francisco Economic Review. Levonian, Mark E. (1996), "Explaining Differences in Farm Lending among Banks," Federal Reserve Bank of San Francisco Economic Review, 0(3), 12-22. Levy, Brian (1993), "Obstacles to Developing Indigeneous Small and Medium Enterprises: An Empirical Assessment," The World Bank Economic Review, 7, 65-83. Levy, Brian (1994), "Technical and Marketing Support Systems for Successful Small and Medium-Size Enterprises in Four Countries," Policy Research Working Paper, No 1400. Lieberman, Ira W. (1991), "Small and Medium Enterprises in Argentina: A Mission Report," Unpublish manuscript, The World Bank, Washington, D.C. Llisterri, Juan Jose, Jesus Puente, and Jorge Rivas (1998), "Inter-American Development Bank Group Activities Supporting Small and Medium Sized Enterprises (1990-1998)," Infrastructure and Financial Markets, Sustainable Development Division, Inter-American Development Bank, Washington, D.C. http://www.iadb.org/sds/IFM/mainpublication_151_e.htm Llorens, Robert van der Host and Iffigo Isusi (1999), "Compilaciones de Estadisticas de Pymes en 18 Paises de America Latina y el Caribe," Grupo Asesor sobre Pequefna y Mediana Empresa, Departamento de Desarrollo Sostenible, Banco Interamericano de Desarrollo. Locke, R. (1995), "Remaking The Italian Economy: Local Politics And Industrial Change," In Contemporary Italy, Ithaca, NY: Cornell University Press. 99 Lora, Eduardo, Patricia Cortes, Ana Maria Herrera (2001), "Los Obstaculos al Desarrollo Empresarial y el Tamafio de las Firmas en Am6rica Latina." Inter-American Development Bank, paper presented at the annual meetings in Santiago, Chile.Maloney, William F. (1998), "The Structure of Labor Markets in Developing Countries: Time Series Evidence on Competing Views," Policy Research Working Paper No. 1940, The World bank. Latin America and the Caribbean. Poverty Reduction and Economic Management Sector Unit. Martin, Fernando M. (1999), "Financiaciones por Actividad Econ6micas." Notas Tecnicas.No. 5, Central Bank of Argentina, Buenos Aires, Argentina. Martinelli, Cesar (1997), "Small Firms, Borrowing Constraints, and Reputation," Journal of Economic Behavior and Organization, 33 (1), 91-105. McDermott, Gerald A. (2000),a. "Reinventing Federalism: Governing Decentralized Institutional Experiments in Latin America," Policy Paper for the Finance, Private Sector, & Infrastructure Division, World Bank; Documento de Trabajo No. 19, Depto. de Humanidades, Universidad de San Andres. McDermott, Gerald A. (2002), Embedded Politics: Industrial Networks and Institutional Change in Post Communism. Ann Arbor: Michigan University Press (Forthcoming). McDermott, Gerald A. (2001), "Network Restructuring and Firm Creation in East-Central Europe: A Public-Private Venture," In The New Entrepreneurs of Europe and Asia, edited by Victoria E. Bonnell and Thomas B. Gold. New York: M.E. Sharpe (Forthcoming). McDermott, Gerald and Michal Mejstrik (1992), "The Role of Small Firms in the Industrial Development and Transformation of Czechoslovakia," Small Business Economics, 4 (3), 179-200. McVary, Mary (1999), "Performance Measurement for Business Development Services to Micro and Small Enterprises: A Revised Framework and Guide to the Preparation of Case Studies," Microenterprise Best Practices Project, USAID Mester, Loretta J. (1997), "What's the Point of Credit Scoring?" Federal Reserve Bank of Philadelphia Business Review 0 (0), 3-16. Mester, Lorreta (1999), "Banking Industry Consolidation: What's a Small Business to Do? Federal Reserve Bank of Philadelphia. x(January/Februaryx) Meyer, Laurence H. (1998), "The Present and Future Roles of Banks in Small Business Finance," Journal of Banking and Finance, 22, (--1)109-116 Meyer, Richard and Geetha Nagarajan (1996), "Evaluating Credit Guarantee Programs in Developing Countries," Department of Agricultural, Environmental, and Development Economics, Occasional Paper No. 2322, Columbus, Ohio. Michael Field, Rob Hitchins, and Marshall Bear (2000), "Designing BDS Interventions as if Markets Matter", Microenterprise Best Practices Project USAID. http://www.mip.org/pubs/mbp-def.htm. Mikkelsen, Lene (1999), "Good Practices in Marketing for Micro and Small Enterprise Products: Cases from Latin America," Inter-American Development Bank, Washington, D.C. Miller, Margaret and Gerard Caprio (1997), "Role of Credit in Small and Medium Enterprise Exports: The Case of Mercosur," Country Department 1, Latin America and the Caribbean Region, The World Bank, Washington, D.C. Mini, Federico and Edgard Rodriguez (1998), "Are SMEs more Efficient? Revisiting Efficiency Indicators in a Philippine Manufacturing Sector," An OED background paper to "World Bank Support for Small and Medium Industries in the Philippines: An Impact Evaluation," (NEED TO SEARCH FOR BOOK) 100 Molano, Walter (1998), "Financial Reverberations: The Latin American Banking System During the Mid-1990s," Working Paper, Social Science Research Network. http://www.ssrn.com/ Mondino, Guillermo and Silvia Montoya (2000), "The Effect of Labor Market Regulations on Employment Decisions by Firms: Empirical Evidence for Argentina," Reseach Network Working Paper # R-391. http://www.iadb.org/oce/pdf/R-391.pdf Moori-Koenig, (1997), " Algunas Caracteristicas del Financiamiento Bancario a las Exportaciones Industriales," Working Paper No. 75 (LC/BUE/L. 158), Economic Commission for Latin America and the Caribbean (ECLAC), Buenos Aires. Moori-Koenig, V. and Yoguel, G. (1996), "Dinamica reciente de las exportaciones industriales de las pequenias y medianas empresas," Boletin Techint No. 288, 49-64. Motta, J. (1996), "El Resurgimiento de las Pymes: Espejismo, Fenomeno Transitorio o Nueva Tendencia de Organizaci6n Industrial?," Instituto de Economia y Finanzas, Universidad National de Cordoba, Cordoba. Mo.tta, Marialisa (2001), "Enhancing E-Govemment in Mexico: work plan for a preliminary analysis," unpublished manuscript. Muent, Holger and Francesca Pissarides (1997), "Impact of Collateral Practice on Lending to Small and Medium Enterprises," Focus on Secured Transactions - EBRD. Nelson, R. (1993), " National Innovation Systems: A Comparative Analysis," New York: Oxford University Press. Nelson, R.and S. Winter (1982), An evolutionary theory of the firm. Cambridge, MA: Belknap Press of Harvard University. Nordio, Raul, Jorge Motta, and Sonia B. Roitter (1995), "Micro y Pequefla Empresa de la Industria Manufacturera Argentina: Situaci6n, Desarrollo Reciente, y Perspectivas (1985-1994)," Unidad Coordinadora del Program Global de Credito para las MyPEs. Observatorio Permanente de las Pymis Argentinas (1997), Informe a las Empresas sobre el Resultado de la Primera Encuesta Estructural. Buenos Aires: IDI-UIA. Observatorio Permanente de las Pymis Argentinas (1999), El mapa territorial sectorial de las pymis. Buenos Aires: IDI-UIA. OCDE (2000), "Reforma Regulatoria en Mexico." OECD (1997), "Regulatory Impact Analysis. Best Practices in OECD Countries," OECD Publications, Paris. OECD (1999), "Regulatory Reform for Smaller Firms," Organization for Economic Co-operation and Development. Science Technology Industry. Oldsman, Eric (2000), "Making Business Development Work: Lessons from the Enterprise Development Center in Rafaela, Argentina," Sustainable Development Department, Inter-American Development Bank, Washington, D.C. Ozorio, Anna Luiza de Almeida, Leandro Alves, and Scott E. M. Graham (1995), "Poverty, Deregulation and Employment in the Informal Sector of Mexico," ESP Discussion Paper Series No. 54. Education and Social Policy Department. Human Resources Development and Operations Policy. The World Bank. Peek, J. and E. Rosengren (1996), "Small Business Credit Availability: How Important Is Size of Lender?," in Financial System Design: The Case of Universal Banking, edited by A. Saunders and I. Walter. Burr Ridge: Irwin Publishing, 628-55. 101 Peek, J. and E. Rosengren (1998), "Bank Consolidation and Small Lending: It's Not Just Bank Size That Matters," Journal of Banking and Finance, 22, 799-819. Peek, J., E. Rosengren, and F. Kasirye (1999), "The Poor Performance of Foreign Bank Subsidiaries: Were the Problems Acquired or Created?" Journal of Banking and Finance, 23 (2-4), 579-604. Peek, Joe., Eric S. Rosengren (1998), "Bank Consolidation and Small Business Lending: It's Not Just Bank Size That Matters," Journal of Banking and Finance, 22(6-8), 799-819. Peek, Joe., Eric S. Rosengren (1998), "The Evolution of Bank Lending to Small Business," Federal Reserve Bank of Boston New England Economic Review, 0 (0), 27-36. Petersen, Mitchell A. (1999), "How Important Are Small Banks to Small Business Lending? New Evidence from a Survey of Small Firms: Comment," Journal of Banking and Finance, 23 (2-4), 459-62. Petersen, Mitchell A. and Raghuram G. Rajan (1994), "The Benefits of Lending Relationships: Evidence from Small Business Data," Journal of Finance, 49 (1), 3-37. Piore, M. and C. Sabel (1984), The Second Industrial Divide. New York: Basic Books. Plata, Nestor Ruben (1999), "Programa SGR, Argentina la Ley de SGR en Latinoamrrica," Presentation in the IV Foro Iberoamericano de Sistemas de Garantias para las Micro, Pequefias, y Medianas Empresas. http://www.nafin.gob.mex/eventos/garantia Possas, M. (1996), "Competitividad de factores sistemicos y politica industrial. Implicaciones para Brasil," in Castro, Possas and Proens, Estrategias empresariais na industria brasileira. Discutindo mudancas (Rio de Janeiro: Forense Universitarias). Powell, Andrew (1996), "On Central Banks and Their Lender of Last Resort Function: 'Constructive Ambiguity' and 'Cheap Talk," Central Bank of Argentina. Powell, Andrew and Veronica Balzarotti (1996), "Capital Requirements for Latin American Banks in Relation to their Market Risks: The Relevance of the Basle 1996 Amendment to Latin America," Paper presented at Jornadas Monetarias y Bancarias, Central Bank of Argentina, August 21-22. Powell, Andrew, Andrea Broda and T. Burdisso (1997), "An Analysis of Lending Interest Rates in Argentina: A Panel Interpretation of a Search Model with Bargaining," Central Bank of Argentina. Pyke, F. and W. Sengenberger (1992), Industrial Districts and Local Economic Regeneration. Geneva: Institute for Labour Studies. Pyke, F., G. Becattini, and W. Sengenberger (1990), "Industrial Districts and Inter-Firm Co-operation in Italy, Geneva: ILO." Quintar, A. (1993), "Pequefios y Medianos Empresarios Industriales Metalmecanicos de Rosario," Working Paper No. 38 (IC/HUE/R. 187), CFI-CII'AI. Programme, Economic Commission for Latin America and the Caribbean (ECLAC), Buenos Aires. Quintar, A., R. Ascua, F. Gatto, and C. Ferraro (1993), "Rafaela: Un Cuasi.Distrito Italiano a la 'Argentina'," Documento de Trabajo.N. 35. CEPAL. Raffin, Marcelo L. (1999), "Una Nota sobre la Rentabilidad de los Bancos Extranjeros en Argentina," Notas Tecnicas No. 5, Central Bank of Argentina, Buenos Aires, Argentina. Rajapatirana, Sarath (1998), "From the Local to the Global Market: The Challenges for Small and Medium Sized Enterprises." 102 Rappaport, A. and R.W. Wyatt (1993), "An Overview of the Securitization of SBA Guaranteed Loans Originated by Commercial Banks," Small Business Economic, 5(--), 215-220. Reich, Robert (1990), "Who is Us?" Harvard Business Review, January. Rosen, Harvey (1998), "The Future of Entreprenurial Finance," Journal of Banking and Finance 22, 1105- 1107. Rosen, Richard (1999), "Comment on: Youth, Adolescence, and Maturity of Banks: Credit Availability to Small Business in an Era of Banking Consolidation," Journal of Banking and Finance, 23 (2- 4), 493-96. Rosen, Richard J. (1998), "Banking Consolidation and Lending to Small Business: Comment," Journal of Banking and Finance, 22 (6-8), 846-49. Rowley, T., D. Behrens, and D. Krackhardt (2000), "Redundant Governance Structures: An Analysis of Structural and Relational Embeddedness in the Steel and Semiconductor Industries," Strategic Management Journal, 21, 369-386. Sabel, Charles and Jane Prokop (1996), "Stabilization through Reorganization? Some Preliminary Implications of Russia's Entry into World Markets in the Age of Discursive Quality Standards," In Corporate Governance in Central Europe and Russia, Volume II: Insiders and the State, edited by Roman Frydman, Cheryl Gray, and Andrzej Rapaczinski. Budapest and London: Central European University Press. Sabel, Charles F. (1996)a. "A Measure of Federalism: Assessing Manufacturing Technology Centers," Research Policy 25, 281-307. Sabel, Charles F. (1993), "Constitutional Ordering in Historical Perspective" In Games in Hierarchies and Markets, edited by Fritz Scharpf, Westview Press. Sabel, Charles F. (1994), "Learning by Monitoring: The Institutions of Economic Development," In The Handbook of Economic Sociology, edited by N. Smesler and R. Swedberg. Princeton: Princeton University Press. Sabel, Charles F. (1996)b. "Local Partnerships and Social Innovation: Ireland. Dublin: OECD". Sabel, Charles, John Griffin, and Richard Deeg (1994), "Making Money Talk: Towards a New Debtor- Creditor Relation in German Banking," In Relational Investing, edited by J. Coffee, R. Gilson, and L. Lowenstein. Oxford University Press. Sanchez, Susana M. (1998), "Gender Earnings Differentials in the Microenterprise Sector: Evidence from Mexico." Saxenian, A. (1994), "Regional Networks: Industrial Adaptation In Silicon Valley And Route 128," Cambridge, MA: Harvard University Press Schreiner, Mark (2000), "Microfinance in Rural Argentina." Working paper. http//www.gwbweb/wust l .edu/users/schreiner Schreiner, Mark and Hector Horacio Colombet (2000), "Microfinance, Regulation, and Uncollateralised Loans to Small Producers in Argentina," Working paper. http//www.gwbweb/wust l .edu/users/schreiner SECOFI (2000), "Regulatory Reform in Mexico". http://www.cde.gob.mx/tmp/frmcentro.html SOCMA (1997), "Las pymes, Clave del crecimiento con equidad". 103 Spicer, A., G. McDermott, and B. Kogut (2000), "Entrepreneurship and Privatization in Central Europe: The Tenuous Balance Between Destruction and Creation," Academy of Management Review, 25 (3), 630-649. Stiglitz, Joseph and Andrew Weiss (1981), "Credit Rationing in Markets with Incomplete Information," American Economic Review, 71, 393-410. Stone, Andreq., Brian Levy, and Ricardo Paredes (1992), "Public Institutions and Private Transactions: The Legal and Regulatory Environment for Business Transactions in Brazil and Chile," Working Papers 891. Strahan, P. and J. Weston (1996), "Small Business Lending and Bank Consolidation: Is There Cause for Concern?" Current Issues in Economics and Finance, Federal Reserve Bank of New York, 2, 1-6. Strahan, P. and J. Weston (1998), "Small Business Lending and the Changing Structure of the Banking Industry," Journal of Banking and Finance 22, 821-45. Strahan, Philip E. and James P. Weston (1996), "Small Business Lending and Bank Consolidation," Federal Reserve Bank of New York, Current Issues in Economics and Finance, 2 (3). Strahan, Philip E. and James P. Weston (1998), "Small business lending and the changing structure of the banking industry, Journal of Banking and Finance, 22 (6-8), 821-45. Tendler, Judith (1997), "Good Government in the Tropics," Baltimore: Johns Hopkins University Press. The Enterprise Research Institute, "SMEs and Public Procurement: Lowering Transaction Costs to Increase Participation." Tokman, Victor E. (1992), "The Informal Sector in Latin America: From Underground to Legality," In Victor Tokam Beyond Regulation: The Informal Economy in Latin America.Lynne Rienner Publishers, Boulder & London. Toulan, Omar and Mauro F. Guillen (1997), "Beneath the Surface: The Impact of Radical Economic Reforms on the Outward Orientation of Argentine and Mendozan Firms, 1989-1995," Journal of Latin American Studies, 29, 395-418. UNIDO, "Promoting Enterprise Through Networked Regional Development." US-SBA (1995), 'The Changing Burden of Regulation, Paperwork, and Tax Compliance on Small Business: A Report to Congress." http://www.sba.gov/advo/laws/archiveflaw_brd.html Walraven, N. "Small Business Lending by Banks Involved in Mergers," Finance and Economics Discussion Series 97-25, Board of Governors of the Federal Reserve. World Bank (1996), "Argentina: The Framework for Secured Transactions and Access to Credit in Agriculture," Report No. 15456-AR. Country Department 1. Environment, Natural Resources and Rural Poverty, Latin America and the Caribbean Region, Washington, DC. World Bank (1998), "Argentina: Financial Sector Review," Report No. 17684-AR. Argentina, Chile, and Uruguay Country Management Unit. Latin America and the Caribbean Region, Washington, DC. World Bank (1998), "Mexico: The Guadalajara Environmental Management Pilot," Report No. 18071- ME, September 8. World Bank (2000), "Information and Communication Technology (ICT) Sector Strategy Paper: Bridging the Digital Divide," November 22, 2000, Draft. World Bank (2000), "Argentina Labor Markets in the New Millenium," Report No. 19996-AR, Finance, Private Sector and Infrastructure Sector Management Unit, Latin American and the Caribbean Region, The World Bank, Washington, D.C. 104 World Development Special Issue (2000) Yoguel, G. F. Bosherini, y M. Lopez (1998), "Sistemas Locales de Innovaci6n y Desarrollo de la Capacidad Innovativa de las Firmas: el Caso de Rafaela," unpublished manuscript, UNGS. Yoguel, Gabriel "Integraci6n Regional y PYMES: El Caso de Brasil y Argentina," CEPAL, Buenos Aires, Argentina. http://www.cefir.org.uy/docs/dtl 1/1 7yoguel.htm Yoguel, Gabriel (1999), "The Responses of Small and Medium-Sized Enterprises," in XXXX Yoguel, Gabriel and A. Boscherini (1996), "La Capacidad Innovativa y el Fortalecimiento de la Competitividad de las Firmas: El Caso de las PYMEs Exportadoras Argentinas," Trabajo No. 71, CEPAL, Buenos Aires, Argentina. http://www.eclac.org/publicaciones/buenosaires/4/lcbuel 1 54/indice.htm Zardkoohi, A. and J. Kolari (1997), "The Effect of Structural Changes in the U.S. Banking Industry on Small Business Lending," Texas A&M University, unpublished manuscript. 105 IMAGING Report No.: 22803 AR Type: SR