From the Local to the Global Market: The Challenges for Small & Medium sized..(1998)

----------------------------------------------------------------------
This document has been made available in electronic format
by the International Co-operative Alliance (ICA)
----------------------------------------------------------------------

			January 1998


SARATH RAJAPATIRANA, Phd
Economic Advisor The World Bank Group

From the Local to the Global Market:  The Challenges for Small and Medium Sized Enterprises (1997)



	One of the most remarkable features of the global economy in the last three decades is the sustained growth in world trade.  Trade has grown faster than world output.  It shows little signs of waning, even though within this rapid growth in trade and globalization, many changes have taken place. (1) Among them, new export success stories emerge, as those of the East Asian countries.  Meanwhile, exporters from other parts of the developing world are also entering the global market, such as those from South Asia and Latin America.  Even more significant than the increase in world trade and the emergence of new exporters, is that for the first time, many small and medium sized enterprises (SME) are participating in globalization in much greater numbers than ever before. 
	The increase in world trade and the participation of SME in the process is of great interest to policy makers, economic analysts and businessmen alike.  Where there is growth there is opportunity.  And where there is opportunity there is a challenge to make use of it.  The issues this paper attempts to describe and analyze relate to these challenges.  The phenomenon of greater participation of SME in world trade and production, what they are and what they do, the opportunities for SME for greater participation in world trade, and the constraints that these enterprises face to make use of the new opportunities. 
	The paper examines the experience from different countries to bring out the main analytical issues and reaches few conclusions.  Its main conclusion is that removing constraints to SME through general policy reforms will allow them better opportunities to participate in global trade and production than selective promotion.  Support at the firm level must be incentive-based and use a light touch rather than excessive intervention.  
	The paper is divided into six sections.  Following the introduction, Section II describes what SME are, what they do and why policy makers are concerned about them. Section III analyzes the opportunities that SME face to increase their participation in the global market.  Section IV examines the constraints they face in entering the global market; Section V analyzes ways to addressing constraints on SME-how and by whom; and Section VI gives the conclusions.


II.  Small and Medium Sized Enterprises: Role, Activities and Policy Focus
	By convention SME in developing countries are those enterprises that employ 1-49 employees and 50-200 employees, respectively.  There is no strict definition of SME.  Different countries use different employment yardsticks.  It can be readily seen that the small enterprise category could include what are called "micro" enterprises (those with less than 5 employees), and the size of medium term enterprises can vary in different country definitions from 50 to 500 employees. (2)
	Nor is size defined solely by the amount of labor used.  Many countries use capital as a criterion and some use both criteria. (3) The problem with the capital or fixed assets criterion is that because of fluctuations in exchange rates, inflation, and capital depreciation, it is difficult to make international comparisons.   On the other hand, capital can be measured in constant prices and exchange rates to yield an internationally homogenous measure.  Differences in labor quality, unfortunately, are more difficult to measure.  There is also no one-to-one relationship between the amount of capital used per unit of output and labor per unit of output. This depends very much on the activity in which the SME is engaged, the choice of technology, and the phase of the business cycle, at which the measurements are made.  Here, the paper is concerned with SME that are engaged in manufactures, which are the largest number of SME. 
	Some analysts use technology as a measure, by comparing  units of labor used per unit of capital and the source of power, whether it is conventional or modern technology, such as between the handloom industry in textiles or the use of power looms.  In the later case, it is measured in terms of horse power.  Another such  measure, is that of technology used by SME in one country measured against state of the art technology used by SME elsewhere.  Of course, there cannot be any presumption that all countries use the same technology.  However, there is an increasing tendency at present for differences in technology to decline over time as the developing countries adopt increasingly superior technology to raise productivity, particularly as labor begins to become more expensive.  This is what has happened in the East Asian countries.
	 SME use less advanced technology compared to large enterprises. Over a wide range of countries and activities, SME use more labor per unit of capital.  However, the valid measure is capital to labor ratios for the same activities in the developed and developing countries.  Developing countries utilize more labor per unit of capital than developed countries since they produce goods that use more labor than capital.  This is the basis of trade between them.  That countries will export goods more intensive in the factor that they have in abundant supply and import goods more intensive in the scarce factor has been well  understood. (4)
	SME in the developing countries are engaged in various activities. They range from textiles, clothing, garments, to furniture,  auto parts, leather products, chemicals and cleaning solutions, electronic equipment to metal work.  The list is wide.  But more generally covered by categories of goods defined as manufactures (i.e. those covered by SITC items 5, 6, 7 and 8-chemicals, machinery and equipment, manufactures and sometimes, category 68, which is non-ferrous metals).  As diverse as SME are, there performances also differ according to the country environments they operate, the type of activity and the type of enterprise.  Such diversity is inevitable in SME given their wide range.  However, some common themes and issues arise, which is why they are of interest analytically. 
	Research on SME reveal the diversity and nature of their production and resource use.  A study done on India confirms that SME are in varied activities from textiles, food processing, metal work, printing, leather work to soap making.  Indian SME produce largely for the domestic market, while those of Hong Kong, Singapore, South Korea, and Taiwan produce almost exclusively of the global market, given the small domestic market size of the latter and their ability to compete globally. In East Asia, many of the small firms operate in vertically integrated sectors, meaning that they produce intermediate goods for large firms, which ultimately export.  In other countries, there is horizontal integration, goods are purchased from SME in final form marketed by a central firm.  This is very much the case in Taiwan, as opposed to South Korea where there is vertical integration.  Finally, other countries such as India, Colombia, and other countries outside East Asia, exhibit little integration.  Hence, SMEs produce i
	 Policy makers are concerned about SME for many reasons.  Some of these concerns are translated into policy actions to support SME, directly targeting incentives on to them.  Others prefer a more neutral approach where SME are expected to benefit from an overall incentive reforms.  There are very good reasons that policy makers should be concerned about SME.  First, they are large in number, particularly in middle income countries.  They have an important product contribution to the economy since their output constitutes a sizable part of manufactures output.  Their growth supports overall growth in the economy.  Second, SME are thought to be the seed bed for new enterprises, both future SME and large enterprises.  To the extent they prosper, more capital accumulation, entrepreneur training and improved efficiency takes place.  Third, improvements in the lot of SME could translate into more employment, better distribution of income, and as some have argued also to contribute to political stability.  Finally
	The preceding indicates that trade liberalization implies globalization, since it links domestic economies to the global market at the input and output ends.  Consequently, trade liberalization is an act of globalization of an economy.  SME are thus exposed to opportunities and challenges that they have not fully found or for that matter appreciated in a restricted local market.  Thus the act of trade liberalization is the rite of passage from the local to the global market.  It has also a deeper psychological impact on domestic entrepreneurs.  Because of the restrictions that have existed in terms of access to inputs and the low returns to exports due to appreciated exchange rates, domestic entrepreneurs and policy makers alike develop an "export pessimism."  It has been shown to be eminently unwarranted as seen from the tremendous export success of East Asian countries.
	The increasing globalization of production has led to deeper integration, by which is meant a movement from trade in goods and services to international production.(10) Firms and activities have become closely linked globally and these links are increasing.  The driving forces for the trends can be identified readily.  First, because of increased communications through TV, satellite transmissions and the print media, there is a growing cross national convergence of consumer preferences.  Consumer cultures converge rather than collide.  Every child knows and wants Nike shoes.  These products are characterized by identical designs and similar standards, wherever they are produced.   Second, due to growing unbundling automation, information and telecommunication technologies, production processes can be "finely spliced" and quality control and production control can take place with low costs around the globe.  Third, the liberalization of goods and capital markets has eased cross border movement not only of go
	Those constraints can be classified into two groups:  those that are economy wide, and those that exist at the enterprise level for SME.  The domestic policy environment is the dominant economy wide constraint.
	The domestic policy environment itself can be divided into three parts:  the macroeconomic environment, incentive policies, and institutions.  In most Latin American countries, macroeconomic instability was endemic, even before the debt crisis that began in 1982. Argentina, Brazil, Mexico, Peru, and Venezuela encountered instability, leading to unsustainable balance of payments deficits and to hyperinflations in the case of Argentina, Brazil and Bolivia.  Even some of the more stable economies of the past, such as Chile, Costa Rica, Paraguay, and Colombia, have encountered high inflation and current account deficits, which led to appreciated exchange rates, imposition of import restrictions and created general policy uncertainty.  These aspects of the macroeconomic environment were harmful to all private enterprise, but particularly to SME.  Given their size, they have fewer options to defend against instability compared to large firms which have access to foreign exchange resources even under exchange cont
	The institutional environment is no less important for SME than for the large enterprises.  The existence of good contracting law, dispute settlements mechanisms between labor and management, laws relating to bankruptcy and those relating to collateral also help to keep transactions costs down and SME more able to compete globally.  These institutions have become more important than in the past due to the globalization of production.  They are the foundation for property rights, must be such to give confidence to foreign firms to write contracts, accept bids to provide inputs, and allow legal recourse to be cost effective so that the contracts can be enforced.  The more certain, predictable and simple these institutions are, the more global SME could become.  They are more important for SME since they provide good protection from arbitrary rules of governments, help to balance power between employers and employees, and also reduce uncertainty.
	The combination of macroeconomic instability, distorted incentives and weak institutions, create a difficult external environment for SME, particularly in their quest to participate in the increasing globalization in production.  These adverse features of this environment have to be addressed to create the best opportunities for the SME.  As would be apparent, without addressing these general constraints, little or no success could be achieved in supporting SME.
	Constraints that are at the firm level include the lack of information about global markets, access to and adoption of new technology, management skills, viable quality control, labor skills, and access to capital. 
	Information has a price.  SME have to compete with large firms to get information on new markets, technologies, sources of supply of raw materiel and intermediate products, and resources such as credit.  SME are put at a disadvantage given that the initial costs would be a large proportion of their total costs.  Often the information has to be gathered through foreign travel, participation in trade fairs, access to domestic sources of  information such as customs data, data on the registration of competitive enterprises, and the like.  In some countries, SME attempt to pool resources through associations to solve the information constraint.  Their success depends very much on the nature of the organization in the country.  Many times, SME  association would use their power to get concessions and to limit competition.  However, opportunities to do this in the global market is virtually non-existent.  Foreign buyers visit countries to establish their own cheap sources of supply of inputs and goods, sub-contra
	Technological constraints can also be alleviated through tie-ins with large firms both domestic and foreign.  Technological change ultimately takes place through investment.  Improvements in the environment for foreign direct investments helps to diffuse technology more efficiently than through the previous patterns of buying technology outright.  In many countries the access to technology have been through the purchase of second hand machines.  This was very much the case of Japan in the last century and many developing countries in the present century.(14) In the garment industry, technological support has come automatically in many countries, from exporters who wanted access to unfilled the export quotas extended by USA and EU under the MFA.  While the MFA is itself a highly distortive arrangement, it has served the purpose of diffusing technology of garment manufacture widely, from Bangladesh to Peru.  Collective public support for  technology has been provided in many settings.  Korea and Japan have pr
	Many of the surveys of SME show that capital is not a strong constraint for these enterprises.  Personal capital and those borrowed from family and informal sources generally allow SME to start up.  After a year or two of operations, they can access financial institutions on the basis of a good track record and on the basis of orders received from clients, which could be discounted as collateral for loans in some environments.  However, in the past, the repression of these financial markets has limited access for SME, due to the high costs of  lending to them.  Financial intermediaries cannot recover these costs because of interest rate ceilings and other restrictions.  Public funds earmarked for SME have a rather poor record of success.
V.  Ways to Address Constraints on SME
	By now there is an appreciable body of knowledge on SME-on what they do, their role in increasing globalization, and the main constraints to their success. This section  describes the ways to address these constraints from the lessons drawn from experiences of  East Asia and parts of  Latin America.
	In the body of knowledge built up on SME on the last two decades, there is evidence that SME have been discriminated against, due to unstable macroeconomic environments, distorted incentives and weak institutions. Correcting these imbalances, distortions and uncertain property rights is the key to alleviating the main constraints on SME.  Beyond this, there has been a discourse between two schools of thought that the state must intervene positively to help SME to enter global markets and those that favor a more neutral approach.  The former school feels that general incentive reforms are necessary but not sufficient to foster the growth of SME.  The East Asian experience is quoted by some as the example to follow, even though the interpretation of that experience remains ambiguous.(16)
	Of the former, some have been worried that increasing competition at home by trade liberalization was detrimental to SME and that they should continue to be protected from imports with exports.(17) In the meantime, evidence from surveys as reported above reveal that promotion of SME exclusively is not consistent with achieving higher GDP growth, improving equality, and increasing international competitiveness.  To be sure, trade protection, distorted investment incentives, credit controls, and the promotion of public sector enterprises have all discriminated against SME.  But equally, selective promotion of SME may have its own disadvantages, such as their lack of scale of economies, limited validity to the belief that they intrinsically use resources efficiently, that they are more labor intensive, and innovative than larger firms. 
	If policy makers get the overall environment right, the SME that emerge in that environment may indeed have had the choice to select the optimal size.  In this regard, the importance of globalization is that it forces countries to recognize that what is good for the whole economy is good for SME too.  This leads to the conclusion that policies should not discriminate by size.  In fact, when SME are given special concessions and when some activities have been reserved for them as in India, their performance has been poor.  These SME did not emerge as the front runners in output growth.  Moreover, the agencies that have been put in charge of SME promotion have dismal records; this is revealed when one looks at the results of their promotion activities and in the views expressed by SME in various surveys.
	However this is not to say that SME should be subject to benign neglect.  Once the general imbalances, distortions, and weak property right issues are addressed, there are areas in which SME could benefit from support more than large enterprises.  By implication, these are mostly at the enterprise level, since both SME and large firms face the same overall incentive environment.
	The access to information has been identified as a constraint.  To address this constraint, two approaches or a combination of them would be used.  The first is the public provision of information.  Unfortunately, unless the SME themselves create collective institutions, publicly owned institutions are notoriously incapable of providing information on time and in a form useful for decision making within a SME.  Public policy for the acquisition of information has been effective when matching funds arrangements have been made available to the SME.  PROEXPO in Chile is a good example.  It provides matching resources to SME for participation in trade fairs and business trips abroad on a limited basis, to make initial contacts.  Some argue for "light touch" intervention, citing KOTRA, the Korean export agency, which has developed an elaborate export information system, multiple opportunities for promoting national exporters through trade fairs, and through an international network.  On the other hand, similar o
	As mentioned above, capital has hardly been a formidable constraint to SME.  They have relied on own funds and family funds to start new enterprises.  When financial markets are liberalized in these countries and interest and portfolio ceilings are lifted, more access to credit could be available.  However, there is also the possibility of providing funds to some financial intermediaries through a one time subsidy to defray the search costs of new SME clients.  While the theoretical argument for such intervention has been around for some time, its implementation has been problematical.  Nevertheless a one time insertion of funds for export promotion should not be ruled out.  This basically the approach taken by Chile in its PROEXPRO program.(19)


VI.  Conclusions
	SME are engaged in diverse activities in the developing countries.  In addition to their size, other common characteristics of SME include that they use more labor per unit of output than large firms and they use less advanced technology.  They cannot realize economies of scale, but instead are generally more flexible in their decision making. 
	Policy makers have been interested in SME because they are prominent in manufacturing, particularly in middle income countries; and many countries have promoted SME believing greater output growth among SME could contribute to overall output growth, increase employment and promote efficiency.  Moreover, SME are also thought to be more efficient, open to innovation, and are considered a seed bed for new entrepreneurs.  While these attributes may hold in many settings, there is also contrary evidence that has been found in India.  On average, in the sectors under study in India, it was found that many of the assertions regarding SME as to their labor use, efficiency, ability to adopt new innovations may not be wholly justified.  The deficiencies found in SME in India may in fact be the result of a poor macroeconomic, incentive and institutional environments. 
	SME could play an important role in the global market if many of the constraints on them could be alleviated or removed.  There has been rapid growth in trade and internationally linked production.  This phenomenon has been described as globalization.  Other aspects of globalization include: the cross national convergence of preferences, leading to identical designs and similar standards of production located in different parts of the globe; the splicing of production in that many goods can be produced in production chains that cross borders; and the liberalization of goods and equity capital that has led to rapid diffusion of technology.  Moreover, local markets have become global to the extent import competition has increased and prices are determined by world supply and demand.  The completion of the Uruguay Round gives even greater opportunities for globalization, since market access for all, including SME, has improved.
	Of the constraints to SME, domestic policy environments have been identified as dominant.  Bringing about macroeconomic stability, adopting better incentive regimes and strengthening institutions can help SME most.  In other words, reform of the overall policy environment is much better for SME than the heavy handed and targeted promotion of them.  Even though some protagonists of specific promotion of SME cite the East Asia model, there is no agreement that the success of East Asia-and SME in that part of the world-was a result of selective intervention. 
	This is not to suggest that the SME should not have supportive measures.  But the support must be generalized and not specific to any sector group or enterprise.  Those measures must assist the SME to alleviate constraints that are found at the enterprise level.  These constraints have been seen as access to information, technology, management skills, quality assurance, and labor skills.  Most of the findings of surveys of SME do not find capital as a constraint.  The reform of the financial system could help SME access capital more efficiently compared to a repressed financial system.
	The ideal way to alleviate the constraints by SME is to address those at the enterprise level (information, technology, management skill, quality assurance, and labor skills). Tie-ins with foreign enterprises through sub-contracting, production sharing, and franchises are cost effective and incentive compatible.  Since products have global reputations, the principals would help SME to address these constraints at the firm level.  Where such tie-ins are not available or feasible, collective action by SME as group can help to alleviate these constraints.  Information, technology acquisition, management and labor training, and quality assurance services have been provided by collective agencies in Korea and Japan, with some support from public funding, such as municipalities in Japan.  Public promotional institutions have a poor record in many other countries, from India and Indonesia to Argentina and Colombia. 
	Some "light touch" promotion may be warranted at the firm level by getting firms to put in some resources of their own and provide support through matching grants as done by PROEXPO in Chile.  These funds would help set up institutions that respond to SME directly.  One time basis support to financial institutions, to laboratories to be accredited to certify NSO 9000 quality assurance, and to hire experts to diagnose the problems of individual enterprises are some of the measures. 
	In summary, reform of the overall policy environment is a powerful way to support SME.  Some assistance to enterprises at their own initiative may be warranted to help them acquire information, upgrade skills and assure quality.  The increasing globalization of production will continue to give SME new opportunities. They can take advantage of them with right overall policies and with light touch support from governments based on cost sharing on a one time basis initiated by SME.  This way, winners can pick the program they want, rather than bureaucrats picking winners.


Bibliography

Berry, Albert "Small and Medium Enterprise (SME) Under Trade and Foreign Exchange Liberalization:  Latin America and the Canadian Experiences and Concerns," Canadian Journal of  Development  Studies, Vol XVII, No: 1, 1996.
Cortes, Mariluz, Albert Berry and Ashfaq Ishaq, "Success in Small and Medium Scale Enterprises:  The Evidence from Colombia, Oxford University Press, 1987.
Holden, Paul and Sarath Rajapatirana, "Unshackling the Private Sector:  A Latin American Story," The World Bank, Washington, D.C, 1995.
Krueger, Anne O., "Foreign Trade and Economic Development: Liberalization Attempts and Consequences, National Bureau of  Economic Research, Ballinger Press 1978.
Krugman, Paul "Growing World Trade: Causes and Consequences" Brookings Papers on Economic Activity 1, pp 327-377, 1995.
Levy, Brian et al, "Technical and Marketing Support for Successful Small and Medium Size Enterprises in Four Countries," Policy Research Paper 1400, World Bank, 1994.
Little, I.M.D, "Small Manufacturing  Enterprises in Developing Countries," The World Bank Economic Review, Vol 1, No:2, pp 204-235, 1987.
Rajapatirana, Sarath, "Policy Recommendations for Export Promotion," Estudios de Economía, Universidad de Chile, Facultad de Ciencias Económicas y Administrativas, Vol. 20, No: 1, June 1993.
Rhee, Yung Whee and Therese Belot "Export Catalysts in Low Income Countries," A Review of Eleven Success Stories," World Bank Discussion Paper, 1990.
Schmidt, Klaus-Dieter, "Small and Medium Sized Enterprises (SME) in International Business: A Survey Recent Literature," Kiel Working Paper No: 721, Kiel Institute of World Economics, 1996.
Stone, Andrew, Brain Levy, Ricardo Paredes, "Public Institutions and Private Transactions:  The Legal and Regulatory  Environments for Business Transactions in Brazil and Chile," Policy Research Working  Paper No: 891, World Bank, 1992.
Stone, Andrew, "Business in Peru:  Transactions Without Trust (mimeo), World Bank 1993.
World Bank,  "The East Asian Miracle," Washington, DC, 1993.
World Trade Organization "Trading into the Future," Geneva, 1995.


Notes:
(1) Globalization is a process  that "integrates the international value added activities of the firm ... in such a way the prosperity of one firm is inextricably bound up with that of its foreign production and marketing activities," Dunning (1995).  Firms are multinational in the sense they sell and buy goods and services internationally and production and marketing chains are inextricably linked across borders.  The growth in intra-firm and industry trade has been more rapidly than those of arms length final goods sales. 

(2) Japan and Korea define SME as those with less than 300 employees, China as those with less than 200 employees, and Latin American countries as also those less than 200 employees, as is the case in Argentina, Colombia, and many others.  

(3) India defines SME by fixed assets, less than Rs. 6 million, Nepal less than NRs 10 million, Bangladesh as 30 million Takas, and the Philippines uses both employment and capital criterion 10-99 as small, 100-199 as medium and capital up to Pesos 5 million as small and up to Pesos 20 million as medium sized enterprises. 

(4) This is the well known Heckscher-Ohlin Theorem in international trade. Empirical tests of the theorem adopted for multi-commodity and multi-country trade have established the validity of the theorem.

(5) Little, I.M.D, "Small Manufacturing  Enterprises in Developing Countries," The World Bank Economic Review, Vol 1, No:2, pp 204-235, 1987

(6) Holden, Paul and Sarath Rajapatirana, "Unshackling the Private Sector: A Latin American Story," The World Bank, Washington  D.C 1995

(7) In surveys, SME are asked to identify their competition who then provide names and addresses of other SMEs that are outside the formal sector.

(8) They are said to be less bureaucratic, able to change and adopt new technology and able to cope with the growing  complexity of production by taking on the production of components, which require more
engaged and detailed access to new inventions.  Cited in Schmidt, Klaus-Dieter, "Small and Medium Sized Enterprises ( SME) in International Business: A Survey Recent Literature," Kiel Working Paper No: 721, Kiel Institute of  World Economics, 1996.

(9) The concept of a bias against exports was introduced and analyzed by Anne O. Krueger in "Foreign Trade and Economic Development: Liberalization  Attempts and Consequences," National Bureau of  Economic Research, Ballinger Press 1978.  

(10) This has been recognized by leading  trade theorists like Paul Krugman, who believes international trade has peaked but that international investment and non-equity involvements such as sub-contracting and licensing have grown much faster than trade.  See Krugman, Paul "Growing World Trade: Causes and Consequences" Brookings Papers on Economic Activity 1,  pp 327-377, 1995

(11) World Trade Organization "Trading into the Future," Geneva, 1995

(12) Time and cost of registering new businesses ranged from 3.2 months for machine tools in Brazil to 4.4 months for garments in Peru. On the other hand  Chile had lower costs, 2.0 months for machine tools. Stone, Andrew, Brain Levy, Ricardo Paredes, "Public Institutions and Private Transactions:  The Legal and Regulatory  Environments for Business Transactions in  Brazil and Chile," Policy Research Working  Paper No: 891, World Bank , 1992, and also Stone Andrew, "Business in Peru:  Transactions Without Trust (mimeo), World Bank 1993.

(13) Levy, Brian et al "Technical and Marketing Support for Successful Small and Medium Size Enterprises in Four Countries," Policy Research Paper 1400, World Bank, 1994.

(14) The case of Colombia is well documented in  Cortes, Mariluz, Albert Berry and Ashfaq Ishaq , "Success in Small and Medium -Scale Enterprises:  The Evidence from Colombia," Oxford University Press, 1987.

(15) ibid , Cortes et al  (1987)

(16) Rhee, Yung Whee and Therese Belot "Export Catalysts in Low Income Countries:  A Review of Eleven Success Stories,"  World Bank Discussion Paper, 1990.   Also, the "East Asian Miracle,"  the World Bank, 1993.

(17) Berry, Albert "Small and Medium Enterprise (SME) - Under Trade and Foreign Exchange Liberalization:  Latin America and the Canadian Experiences and Concerns," Canadian Journal of  Development  Studies, Vol XVII, No: 1, 1996.

(18) The rationale for such a tax credit is the presence of an externality-without it, entrepreneurs may not have the incentive to train workers since once trained they may leave, giving entrepreneurs less of a chance for cost recovery from the difference in changes in productivity and wage increases.

(19)  The design of such a program is discussed in  Rajapatirana, Sarath,  "Policy Recommendations for Export Promotion," Estudios de Economía, Universidad de Chile, 1993