69474 Recent Evidence on Private Sector Capabilities: Trade Liberalization, Foreign Direct Investment, and Deregulation Are Not Enough Daniel Lederman Marcelo Olarreaga Office of the Chief Economist for Latin America and the Caribbean The World Bank First Draft: June 19, 2006 Abstract: This paper summarizes new empirical findings concerning the magnitude of market failures affecting industrial performance and innovation, firm dynamics after economic policy reforms, and the effectiveness of popular government interventions designed to promote exports and foreign direct investment (FDI). The new evidence suggests that market failures are significant. Case studies of firm dynamics suggest that in Costa Rica and Mexico market failures might affect both the introduction of new export products (pineapples in Costa Rica) and certain chemical industries (detergents in Mexico). New evidence concerning the adjustment of apparel industries in Latin American and Caribbean (LAC) ecnomies that benefitted from the CBI also suggests that the quality of the industrial restructuring can be enhanced through government interventions focused on the provision of public goods, namely education and broad indicators of infrastructure coverage and energy availability. Regarding the effectiveness of public subsidies and targeted services, new evidence suggests that innovation policies targeting both patentable and non-patentable innovations might yield higher returns in terms of innovative activity if they are broadly diversified across products within broad industrial categories, and revealed comparative advantage by itself provides little guidance as to what these innovative sectors might be. In general, efforts to promote exports and FDI seem to be effective on their own terms when the returns to public expenditures in these areas are measured by their effects on the value of exports and FDI inflows. However, there is significant heterogeneity both across regions and across types of expenditures. Overall the recent research suggests that trade reforms and deregulation are not enough to sustain long-term development. 1 I. Introduction: When Free Markets Are Not Enough Economists, since at least Adam Smith, often argue that the best policy is laissez faire. The market, with the help of the invisible hand, ensures that productive resources are efficiently allocated so as to maximize value of economic activities and economic growth. Policy makers would need large amounts of information, knowledge and resources to be able to match the performance of well functioning markets. In spite of the alleged virtues of free markets, the invisible hand by itself does not always work well: the existence of economic externalities (or actions by one agent that impact the economic, social or other returns of others), coordination failures, and asymmetric information regarding prices, products and production possibilities often lead to market failures characterized by under or over-investment in certain activities, or missing markets.1 In any of these cases government intervention may be justified, in order to correct inefficiencies. Nevertheless, government shortcomings can easily worsened market failures. First, un-productive rent-seeking activities may appear and lobbying may move the government away from its original objective. Second, taxation may be necessary to finance intervention and this may lead to distortions with side-effects that are deleterious for economic growth and social welfare. Third, government interventions can crowd out private-sector investment. Fourth, signals emanating from government planning can increase risk and economic volatility by instigating herd behavior by private agents. Fifth, government interventions can be rendered ineffective by the emergence of informal markets that circumvent public interventions. The free-market reforms undertaken in Latin America over the past decades were based on the assumption that in the region government failures were larger than market failures. But the perception of disappointing growth in the region after reforms has raised some doubts regarding this assumption, perhaps due to market failures in the region that may be significant after all and government failures that may not be as severe as expected. The World Bank’s Office of the Chief Economist for Latin America and the Caribbean has been funding novel research to assess the potential developmental gains that can be attained through selective public interventions to aid private-sector investment. Previous research efforts in this area focused on the rationale and effectiveness of educational and science and technology policies (e.g., De Ferranti et al. 2003; Maloney 2005) and on the complementary development agenda for trade reforms (Lederman et al. 2005; Jaramillo et al. 2006). The research summarized herein is about identifying the existence and economic magnitude of some of these market failures in the region, with a special focus on externalities, asymmetric information and coordination failures. We also explore the potential for government intervention to correct them. The new evidence suggests that market failures are significant. Case studies of firm dynamics suggest that information and knowledge spillovers across firms within and across sectors affect firm dynamics in 1 Also, monopolisitic or oligopolisitc markets, irrational individuals that fail to maximize their long-run utility and insurance or future markets which by definition cannot be perfect are all sources of market failures. 2 the sense that they affect both the growth and structure of industries, and trade and FDI are important sources of knowledge diffusion. Nevertheless, combined with the econometric evidence concerning the existence of market failures, the qualitative case studies support the view that trade, FDI, and deregulation are not enough to sustain private sector development characterized by both innovation and diffusion of commercial ideas. In general, efforts to promote exports and FDI seem to be effective on their own terms when the returns to public expenditures in these areas are measured by their effects on the value of exports and FDI inflows. However, there is significant heterogeneity both across regions and across types of expenditures. Overall the recent research funded by the World Bank suggests that trade reforms and deregulation are not enough to sustain long- term development. The rest of this paper is organized as follows. Section II reviews new evidence on the existence and magnitude of two types of market failures, namely coordination failures affecting the growth vertically integrated industries and informational spillovers affecting non-patentable export innovations. Section III reviews qualitative evidence concerning these two types of market failures in Costa Rica and Mexico, two economies that have undergone dramatic structural changes during the past decades. Section IV examines the adjustment paths of textile and apparel industries in selected Latin American and Caribbean (LAC) countries and links the quality of the adjustment path to policy- sensitive variables, such as female secondary school enrollment, energy availability, and telephone penetration rates. Section V concludes by discussing empirical evidence concerning the effectiveness of certain types of government interventions aimed to stimulated private-sector innovation, export growth, and FDI inflows. II. Searching for Market Failures: New Evidence Gawande and Sauer (2006) focused on the role of coordination failures in vertical firm relationships. The premise of these models is that the impediments to growth are due to thresholds arising from externalities or coordination problems in the production of intermediate goods. Before crossing such a threshold, the economy is trapped in “poor� equilibria where private investment is not profitable unless it is undertaken simultaneously by upstream and downstream firms. Breaking the development trap requires crossing the threshold in order to move into “good� equilibria. The logic of the models is compelling, but there has been so far only a loose empirical basis motivating these models. Gawande and Sauer look at the relationship between downstream industries’ real wages and their use of upstream intermediate goods. They found that in 80 percent of developed country and Asian industries wages start reacting to the availability of intermediate goods only after a certain threshold. That is, the nature of the relationship between wages and intermediate goods depends on whether intermediate goods and labor are complements or substitutes. For Latin American economies, they found evidence of threshold effects in only 50 percent of the industries, namely food, beverages, textiles, apparel, footwear, furniture, paper, printing, petroleum, miscellaneous products of petroleum and coal, 3 plastics, pottery, and iron and steel. In contrast, for Africa the authors found evidence of threshold effects in only two industries. The finding of fewer industries subject to threshold effects in Africa and Latin America than in developed countries probably signals that in many of these industries Latin American and African industries have not yet reached their threshold level, making the identification of threshold effects for the researcher virtually impossible.2 Note that the solution to this type of coordination failure does not necessarily require heavy government intervention. Probably the most straightforward way of providing intermediates to final producers is by having an open trade and investment regime, which takes us back to the Adam Smith view of the world. However, getting rid of government induced barriers to trade and foreign investment may not be enough, in the presence of information or knowledge externalities. Klinger and Lederman (2006a) search for evidence of information and knowledge externalities that may lead to under-investment or missing markets, which can in turn exacerbate coordination failures. In a sample of 73 developed and developing countries the authors found that imitation of first-movers into export markets by followers may be leading to fewer cases of non-patentable inside-the-frontier innovation (measured by the counts of new export products introduced during 1994-2003) than would be the case otherwise. It is noteworthy that this finding is based on the observation that countries with more severe barriers to entry (based on indicators from the World Bank’s Doing Business database that measure the costs of starting new business and the labor hiring and firing costs) tended to have a higher numbers of export discoveries than similar countries with a given level of export profitability. Hence these findings suggest that while deregulation might be advisable to spur entry and thus imitation, which brings about the social gains from innovation, it might have the undesirable effect of deterring private sector innovation as the expectation of imitation by competitors will be higher in unregulated environments. While Hausmann and Rodrik (2003) attribute this imitation specifically to the learning of production costs which are not predictable ex ante, Klinger and Lederman analysis does not identify the particular channel. The spillover may not come from learning production costs, but instead from learning the characteristics of foreign demand (Vettas 2002), free-riding on investments to cultivate foreign demand (Mayer 1984), learning the redesigns needed to meet foreign safety standards (Granslandt and Markusen 2000), or some other unidentified channel. Regardless of the particular channel, the findings provided by Klinger and Lederman (2006a) suggest that there are spillovers benefiting firms that learn from another firm’s demonstration that a new export product is economically viable. While this indicates that public support for experimentation in new sectors and activities may be warranted, especially in open and deregulated industries and economies, especially when the new product is not patentable, it is important to note that it does not mean that increasing barriers to entry through changes in regulations is advisable to enhance private-sector innovation. Indeed, we cannot overstate that imitation is desirable as this is the channel through which the returns of inside-the-frontier innovation are socialized. Consequently 2 Measurement error could also explain these findings, given data quality issues in developing countries. 4 direct support for non-patentable innovations in general, and perhaps for new exports in particular, seems justified by this new empirical evidence. Some of these issues and tradeoffs are also apparent in the case study evidence discussed in the following paragraphs. III. Firm dynamics in open economies – Case Studies The World Bank research project commissioned two case studies of firm dynamics in contexts where trade reforms and other factors were associated with notable changes in economic structure. The first focuses on the process of discovery of new export products, namely the successful case of pineapple exports by Costa Rica. The second focuses on the role of foreign investment and free trade agreements in solving coordination failures between upstream and downstream producers. Guevara (2006) focuses on the successful export discovery case of pineapple exports in Costa Rica. Ten years ago there were 20 small and medium size farms in Costa Rica growing pineapple, and exports were only US$26 million. Ten years later, there are more than 1500 small and medium size farms producing pineapple and exports are 15 times larger making Costa Rica the world leading exporter of fresh. It has 76 percent of the United States market. The process was initiated with the discovery of a successful production possibility in Costa Rica that was repeatedly adapted to innovations made available in other countries at the beginning of the discovery process, taking advantage of earlier legal failures by inventors to patent their agronomic innovations. However, the latest genetic developments (Honey, Honey Gold, or MA-2 pineapple) were undertaken by Costa Rican scientists in their own country, and were patented by the United States Patent and Trademark Office in March of 2006, with the expectation that this will help Costa Rican exporters preserve their growing world market share. A crucial element in the successful adoption of pineapple technology in Costa Rica lied in the country’s human resources that allowed for successful (illegal) introduction of new varieties made available elsewhere, as well as providing large growers with employees that could quickly adapt to new production techniques. The strong link between the innovator (PINDECO) and the National Institute for Learning (Instituto Nacional de Aprendizaje) helped throughout the discovery process. Their partnership aimed at training PINDECO’s employees in agricultural mechanics, electricity, welding and industrial maintenance. A second element that facilitated the early stages of the discovery is that the innovator already had shipping facilities with refrigeration for their banana exports. They only needed to modify the temperature to make it suitable for pineapple, and weekly shipments were already in place, thus providing an existing and well functioning distribution channel for importers, which benefit from a good reputation that would have been otherwise very difficult to establish. These externalities across products were 5 internalized by the producer, but one could imagine cases where producers may not be able to benefit from externalities across products as production and distribution facilities belong to different firms. Javorcik, Keller and Tybout (2006) explore the impact of trade and investment liberalization on the soap/detergent/surfactants (SDS) industry in Mexico. They chose Mexico as a case study because unilateral commercial policy reforms took place in this country during the mid-1980s, and they were followed by the elimination of restrictions on FDI. Thus the Mexican case provides an opportunity to study 20 years of changing degrees of integration. They have chosen the SDS sector because it is characterized by a significant multinational presence in Mexico, it conducts R&D, it exhibits significant cross-border trade, it is closely linked to the domestic retail sector, and the quality of its products and its overall productivity relies heavily on input providers. The authors found that throughout the liberalization period Mexican detergent exports have gained market share in the United States. Falling trade barriers and the growth of the Latino market in the United States have contributed, but perhaps more importantly, the establishment of Walmex (Wal-Mart of Mexico) led to important changes within the Mexican SDS industry. Once Walmex (Wal-Mart of Mexico) was established, it changed the way that SDS producers (and other suppliers of consumer goods) interacted with retailers in several ways. Earlier SDS producers and retailers had a cozy business relationship based on social interactions and bribes that made entering the market by new firms quite costly. Incentives to innovate and invest were quite small. Walmex changed this. By exercising its bargaining power, it squeezed profit margins among the major brands, offering them higher volumes in return. It also engaged the most efficient small- scale local producers as suppliers of store brands, thereby creating for itself a residual source of SDS products that could be used in bargaining with the major (multinational) branded suppliers. Those local firms that were not efficient enough to meet Walmex’s terms lost market share, and many failed. On the other hand, the limited set of producers that survived grew, and with prodding from Walmex, became more efficient and innovative. In addition to the technology diffusion that Walmex brought to the retail sector, other types of induced innovation are traceable to Mexico’s opening. The presence of SDS multinationals in Mexico has led to demonstration effects among the smaller firms, which generally wait for the major brands to experiment with new products, and then imitate them if they are successful. The multinationals themselves rely on their headquarters abroad for basic product development, and focus their research efforts on testing and adapting new products for the local market. Smaller firms also absorb new technologies through their intermediate input suppliers, who come to them with new compounds and processes, making access to imported intermediate inputs a crucial element of the transformation of the Mexican SDS industry. In a nutshell, both case studies provide rich stories concerning the process of technology and knowledge diffusion both across firms within sectors and across sectors. In both cases, innovation and imitation went together, and trade reforms and inter-firm linkages 6 were important elements for firm upgrading. Although the case studies do suggest that trade reforms might be an element of the innovation agenda, they also suggest that they might not be enough, especially when the case studies are considered in conjunction with the previously discussed econometric evidence concerning the existence of market failures. It is also noteworthy that both Costa Rica and Mexico enjoyed preferential access to the U.S. market through the Caribbean Basin Initiative (CBI) in the case of Costa Rica and through the North American Free Trade Agreement for the case of Mexico. Consequently it is prudent to assess industry dynamics when these types of preferences are eroded through trade reforms of the major importing market, namely the United States. Likewise, it might be important to establish the public interventions that might help industries affected by preference erosion to adjust by moving up the product quality ladder rather than through declines in unit prices of the affected exports. The following section examines the adjustment paths of the apparel industries in several LAC countries. IV. Preference Erosion and the Process of Adjustment – What is the Role of the Public Sector Ozden (2006) analyzes the impact of the unilateral preferences on the performance of apparel exporting countries in Central America and the Caribbean. We mainly focus on the prices received by the exporters as they are better measures of the quality of the products and competitiveness of the exporting firms in global markets. The countries we analyze are beneficiaries of the CBI of the United States, which consists of a series of programs first initiated in 1983 that grant duty and quota-free market access. Apparel preferences are the most valuable and heavily used preferences due to high trade barriers imposed by the U.S. against the exports from the rest of the world. The empirical analysis and the exporters' performance is complicated by the two changes experienced by the U.S. trade policy, especially with respect to apparel imports, since the implementation of the first CBI program. First was the implementation of NAFTA which led Mexico to emerge as a main competitor to the CBI countries and to become the largest apparel exporter to the US until 2003. Second was the relaxation and eventual removal of the MFA quotas (at the beginning of 2005) that historically restrained the exports of South and East Asian countries. As a consequence, Chinese apparel exports reached almost $17bn in 2005, almost triple the level of second-ranking Mexican exports and 25% of total US imports. The performance and reaction of the CBI exporters to the preferences and to these developments are crucial for a variety of reasons. Among the main goals of unilateral preference programs are the integration of developing countries into the world trading system and long-term economic growth through international trade. Many unilateral preference programs fail to deliver the promised gains due to a variety of reasons (see Hoekman et.al. [2003] for a review). However, CBI programs provide significant advantages compared to other programs, such as the inclusion of apparel sector, and are considered a success based on the rapid growth of exports from the beneficiary countries. 7 Nevertheless, if the eligible exporters do not capture the full benefits of preferential access and the benefits can not be maintained in the absence of preferential access, then these policies can not be deemed successful. When the barriers faced by excluded countries are lowered, as in the case of implementation of NAFTA or the removal of MFA quotas, the value of the preferences to the beneficiaries erode considerably. The recipient countries should not rely on preferences to deliver long-term rents but use them as a transition stage to an environment where trade flows are determined by comparative advantage, rather than preferential access. The trade data reveals that the preference recipients performance were rather different. Costa Rica and the Dominican Republic took advantage of the initial preferences to increase their export volume. With the implementation of NAFTA and removal of MFA quotas, these countries moved to higher priced/quality exports and actually had their total exports decline. El Salvador, Guatemala and Honduras did not exhibit structural changes in their apparel industry but simply increased their production and exports at the same quality/price level. Nicaragua and Haiti were new entrants to the apparel markets and their exports increased dramatically. However, under competition from Asian countries, they actually moved down the quality ladder to lower priced/quality exports. The prices received by the CBI exporters naturally depend on their own tariffs and the tariffs paid by excluded countries, as well as many other product and market characteristics. One of our innovations is that we are able to use country, industry and year fixed effects in a GLS estimation to isolate the effects of other factors, such as exchange rates, transactions costs and other market characteristics. Our main finding is that the CBI exporters capture around 2/3rd of their preference margin and this translates into around 9% increase in the relative prices they obtain. Eligibility for preferences requires compliance with Rules of Origin Requirements which entail significant administrative and production costs. For example, the use of yarn and fabric from the U.S. instead of the generally cheaper inputs from third countries is required. When these are taken into account, the net benefits are likely to be much lower. We also find that the lower import charges (such as freight and insurance) faced by the CBI exporters due to geographic proximity are another source of price advantage. Lower import charges generate around 1.5% relative price increases. Finally, the quotas faced by the Asian countries also led to price advantages for the CBI countries in the order of around 1.7% which disappeared on January 1, 2005. Preferential market access, geographic proximity and restrictions on competitors lead to significant price advantages to the firms in CBI countries. However, we still observe significant variation in terms of relative prices across countries and over time (Ozden and Sharma 2006). Thus it is important to inform policy debates about the types of public policies that might help improve the quality of the adjustment process, namely by producing adjustment paths characterized by improvements in unit prices of apparel exports, rather than by export efforts in products with declining prices. Ozden finds that policies that create stable macroeconomic environment, better educated labor force, and more energy efficient economy positively influence the export prices. These results seem 8 robust to the inclusion of country-specific characteristics that did not change during the period covered by the study, namely 1989-2005. V. Designing and Evaluating Proactive Policies for Innovation, Exports and FDI In section II we summarized the evidence concerning the existence and magnitude of coordination failures and knowledge and information externalities that lead to under- investment, and missing markets, which may hinder the development process. In this section, we summarize the research findings regarding the effectiveness of certain types of government interventions that may be warranted, as well as an assessment of the role played by export promotion agencies (EPAs) and investment promotion agencies (IPAs). Klinger and Lederman (2006b) explore whether government support in the presence of market failures associated with under-investment in sectors subject to positive knowledge and information externalities (on and off the frontier innovations, and export booms) should be concentrated or diversified across industries or sectors of economic activity. They provide evidence on this by assessing the performance in terms of on-the frontier innovations (patenting), off-the frontier innovations (export discoveries) and export booms across countries with different degrees of concentration in these two types of innovations and export booms. They find a clear negative relationship between concentration and performance: the most successful countries in terms of on and off the frontier innovations and export booms are those where the search processes are spread across a broad range of industries, rather than concentrated in a few sectors. The search for export booms exhibits the least amount of sector concentration and path-dependence. These findings suggest that public support for these processes need not be focused in a narrow range of sectors, especially in the case of export booms. Lederman, Payton and Olarreaga (2006) provide an assessment of export promotion agencies in a sample of 122 developing and developed countries for which data was collected for this project. Contrary to earlier assessments of Export Promotion Agencies (EPAs) such as Hogan, Keesing and Singer (1991), Lederman and his coauthors found that on average EPAs are an effective tool for export promotion. Their econometric estimates, which control of the endogenous emergence of EPAs in countries with fast export growth (i.e. the estimates control for selectivity biases) suggest that each dollar invested in EPA budgets produces on average an additional $490 dollars of exports in LAC, $227 in Asia, $ 160 in OECD countries, $137 in SSA and $96 in MENA (note that the last two estimates are not statistically different from zero).3 On average exports increase with EPAs’ budgets, even though our estimates suggest that at levels about $3 per capita, the marginal efficiency tends to decline. In terms of what works and what does not, their estimates suggest that EPAs the most effective agencies are those that have a large share of the executive board in the hands of 3 Note that the social return may be much smaller as these calculations do not take into account the additional production cost associated with the export increase. 9 the private sector, but with also a large share of public sector funding. In other words, purely private EPAs may not be the most efficient. Also, a single and strong EPA should be preferred to the sometimes observed proliferation of agencies within countries. Results also suggest that EPAs should focus on non-traditional exports or have some broad sector orientation (agriculture, manufacturing, services, tourism, etc.), rather than promote overall exports. It is noteworthy that these results do not contradict the empirical findings of Klinger and Lederman (2006b), which were discussed above, because the econometric evidence suggests that More efficient EPAs are those that focus on large firms which are not yet exporters, rather than on small or already exporting firms. There are also some characteristics that seem to be particularly important for developing countries, but do not matter as much in the full sample. In developing countries, the export promotion activity of the agencies should be shared with other activities such as investment promotion and/or export financing. Similarly, they should focus their expenditure on on-shore export support services rather than country image or marketing and market research activities. Also, representation offices abroad seem to hurt exports in developing countries, suggesting again a focus on on-shore activities. Harding, Javorcik, and Sawada (2006) provide a similar assessment of Investment Promotion Agencies (IPAs) after conducting an international survey of IPAs for 99 developing and developed countries. They found that the presence of an IPA on average leads increase foreign direct investment. On average a 10 percent increase in the budget of the IPA leads to a 5 percent increase in FDI, but again there are diminishing returns for very high budget levels. As in the case of EPAs, a quasi-governmental institutional structure with a private sector oriented executive board and public funding seems to be the most efficient IPA structure, performing better than IPA which depend directly from a Ministry or sub-unit of a Ministry, or any other institutional structure including fully privately run and financed IPAs. The quasi-governmental structure provides IPA with sufficient funding, while they are not subject to government rules regarding its functioning, procurement and hiring and firing of staff , or political influence (as long as budgets are multi-year). In terms of allocation of funds across IPA activities, investment generation and investor servicing activities seem to be the most likely to lead to high FDI flows. Country image activities are never associated with higher FDI flows, and policy advocacy are weakly associated with higher FDI flows when they consider the full sample, but not when restricting the sample to LAC countries. This is in an area where LAC agencies may want to rethink their strategy. LAC agencies tend to devote a larger share of resources to image building than their counterparts in other parts of the world. Agencies in countries, which have already been quite successful in bringing FDI inflows –such as LAC— may want to shift the scarce resources to investment generation and investor servicing activities. 10 VI. References Gawande, Kishore, and Christine Sauer (2006), “Variety in intermediates and growth in manufacturing wages: evidence�, mimeo, University of Texas. Ganslandt, M., and James Markusen (2000). “Standards and Related Regulations in International Trade: A Modeling Approach.� In Quantifying the Impact of Technical Barriers to Trade: Can It Be Done?, pp. 95-135. Studies in International Economics, Ann Arbor: University of Michigan Press. Guevara, Porfirio (2006), “Case studies in Self Discovery. Costa Rica: Pineapple�, mimeo, Costa Rica. Harding, Torfinn, Beata Javorcik and Naotaka Sawada (2006), “Investment Promotion: Why, What and How Lessons for Latin America and the Caribbean�, mimeo, The World Bank. Hausmann, Ricardo. and Dani Rodrik (2003), “Economic Development as Self- Discovery.� Journal of Development Economics 72: 603-633. Klinger, Bailey and Daniel Lederman (2006a), “Diversification, Innovation, and Imitation inside the Global Technological Frontier�, Policy Research Working Paper #3872, The World Bank. Klinger, Bailey, and Daniel Lederman (2006b), “Diversification in the search for innovation and export booms�, mimeo, The World Bank. Lederman, Daniel, Lucy Payton, and Marcelo Olarreaga (2006), “Export promotion: what works, what doesn’t?�, mimeo, The World Bank. Mayer, Wolfgang (1984), “The Infant-Export Industry Argument.� Canadian Journal of Economics 17(2): 249-269. Ozden, Caglar. 2006. “Caribbean Basin Initiative Beneficiary Countries and the Apparel Sector: Same Preferences, Different Responses.� Mimeo. The World Bank. Ozden, Caglar and Gunjan Sharma [forthcoming], �Price Effects of Preferential Market Access: Caribbean Basin Initiative and the Apparel Sector,� World Bank Economic Review Vettas, Nikolaos (2002), “Investment dynamics in markets with endogenous demand.� The Journal of Industrial Economics 48(2): 189-203. 11