Claude Baissac
(Presented at the WEPZA Round Table on EPZs, Vienna, 4/15/96)
No. 2 Ver. 1
Copies of this booklet may be obtained from:
WEPZA Secretariat
P.O. Box 986
Flagstaff, Arizona, U.S.A.
Tel: (520) 779-0052
Fax: (520) 774-8589
E-Mail: wepza@aol.com
WEPZA Website: http://www.wepza.org
© The Flagstaff Institute, 2004
Introduction
The impact of export-processing zones on their "host" economies has constituted a consistent problematic for those who have approached the phenomenon since the setting up of the first zones in the 1960s. Policy makers tempted by their neighbors' apparent successes at attracting foreign direct investment through the deployment of such zones, economists whose curiosity has been excited by the emergence and subsequent proliferation of a still relatively new economic phenomenon, sociologists and anthropologists who have rapidly studied the age and gender particularity of export-processing zones' labor force, and many others, have been producing many studies on zones. Among them, those concentrating on their economic characteristics have been the most numerous.
As a significant part of the economic literature on export-processing zones has focused its attention on describing the phenomenon - mostly through case and comparative studies - another has been concerned with providing more operational analysis. Among the numerous analytical tools used to evaluate export processing zones, both from a pre- and a post-facto perspective, one has commanded particular attention, although its actual practice has been limited. Because of its supposed capacity to reveal economic price and gains, cost-benefit analysis is a much appreciated tool in economic studies. It is widely believed that such analysis can provide policy makers with an efficient and objective evaluation of programs. Under such assumption, it has been practiced on export-processing zones.
In this article I question the validity of using cost-benefit analysis as a central policy instrument to assess export-processing zones, using the results of specific case studies to generalize the conclusion of the concept. Following a brief introduction of the instrument I will examine the problems of applying it to the evaluation of export-processing zones, undermining the assumption that cost-benefit analysis is a feasible, practical and legitimate evaluation device for their study. I will then review a 1989 article published in The World Bank Research Observer under the title "Export Processing Zones: The Economics of Enclave Manufacturing", by Peter G. Warr. An assessment of its theoretical and methodological frameworks and its choice of sample will demonstrate how they strongly pre-define the scope of the study and how such scope has limited relevance to an overall evaluation of the concept of export processing zones.
Principles of cost-benefit analysis
The principle behind the cost-benefit analysis is of bewildering simplicity. To know the value of a particular program the analyst simply operates the difference between the benefits that have resulted from its implementation and its cost. If the result of such computation is negative, the program has cost more than what it has brought. In this situation, following the reasoning behind the method, the program needs to be amended or terminated so that it stops being a financial drain.
Behind a technique marked by such inescapable logic lies the assumption that the primary value of a program submitted to cost-benefit analysis is, fundamentally, economic in nature. As such, the tool can be of significant importance for the appraisal of a particular program. Cost-benefit analysis is appropriate primarily where and when profitability must be achieved to insure the continuity of the program.
As stated by Heineman (1990) and many others, the method applies where quantitative data are available. It deals easily with tangible facts. When intangibles are considered, significant problems appear. How, indeed, to translate these non-quantitative realities into computable data, and ultimately into monetary units, the ultimate expression of any "valid" analysis? Generally the method used to resolve such problem consists in creating tables where intangibles can somewhat be translated into quantitative - and therefore useful - data. However, the simple "conversion" of non-quantitative data into figures that can enter the calculation is of limited relevance since so much depends on the subjective personal, political and cultural perspective of the translator. Cost-benefit analysis loses merit in the domain of public policy where intangibles and long term or multiple objectives constitute an important part of the considerations.
Cost-Benefit Analysis and the Policy Objectives of EPZs
With this in mind, it must be clear that using cost-benefit analysis as the primary too[ for the evaluation of export-processing zones is based on the assumption that they are primarily economic instruments and, therefore, submitted to the necessity of financial efficiency.
A problem with this view lies in the definition of the financial efficiency -or profitability - of export-processing zones. Indeed, most EPZs that have been so far submitted to publicly available cost-benefit analysis (and, significantly, those chosen by Warr) are public zones. If export-processing zones are indeed created for economic purposes, they are above all public policy instruments with social and political purposes and not exclusively limited to purely economic considerations, EPZs are designed to provide the "host" economy with a range of gains difficult to measure. A United Nations publication of 1985 listed the most common objectives underlying the establishment of zones as following: (i) generating foreign exchange earnings; ( ii ) creating employment; iii attracting foreign capital and advanced technology; ( iv ) acquiring labor management skills; and ( v ) creating linkages between EPZ industries and the domestic economy." (Page 10). The World Bank, in a 1992 study, states a more limited range of objectives, comprising the earning of foreign exchange and the creation of "employment through exports of manufactured goods." (page 1). If the list of objectives retained by the United Nations is accepted as somewhat reflective of reality, significant difficulties face cost-benefit analysis.
To make these difficulties manifest, consider a hypothetical public export processing zone designed to result in the five benefits listed by the United Nations' study of 1985 and setting a clear agenda for the realization of these objectives. A cost-benefit analysis based on that agenda would have to break down the benefits separately:
The calculation of foreign exchange earnings could be done - a priori -without problem. Once calculated, the result could be compared to the figure on the agenda. The same operation would be required with employment creation, attraction of foreign capital and linkages between EPZ industries and the local economy (considering, for example, the total value of purchase). For these four benefits to be subject to proper evaluation under the cost-benefit analysis method, the only requirement would be the existence of comparable data from the analysis to the original plan setting the objectives to be attained by the zone. The actual benefits resulting from the operation of the zone simply would have to be compared to the expectations.
Obstacles would arise when the appraisal of technology transfers is undertaken. How could technology be expressed to fit into the framework of the analysis? How could the value of technologies not yet invented be calculated? How could planners foresee the types and value of the technologies that will be attracted to the zone? The same problem appears with labor management skills: how could labor management skills be defined and measured? How to identify and put a figure on the improvement of skills of management and workers without automatically reducing these gains to clich6s? In other words, how to define skills in a wide enough sense? Moreover, how to include in the benefits unexpected consequences whose types and depths, being far from evident are nevertheless worth considering and could even constitute the bulk of benefits? Restraining the analysis to the benefits planned bears the extremely serious risk of underestimating the real impact of a zone just because of political and methodological views.
The consideration of the cost may seem free from problems of this kind. Indeed, this calculation would be based on the addition of the costs incurred by the necessary investments for the planning-construction of the zone and the necessary serving infrastructures, the cost of maintenance, administration and promotion of the EPZ, the cost of the different incentives and subsidies offered by the government to investing companies, etc. The cost calculation appears to be rather straightforward. Is it this simple-in reality? A deeper review of the cost component of the analysis raises important questions. Indeed, conveniently enough, externalities or induced effects, such as the environmental cost of building a zone or the social cost of factory work are generally not taken into consideration. Without any doubt, these actual costs are too difficult to take into account for many reasons. Therefore, the consideration of cost must be seen as limited since it is restricted to the evaluation of the direct cost of the zone to the government.
In fact, objectives that export-processing zones are to attain are never stated clearly. Never does a government create precise plans, not only because they are unrealistic, but also, and maybe more substantially, because they are self-defeating. Indeed, the political risk involved with generating extensive plans where precise figures or projected results are put down could be too easily used as a political weapon against its designers. In the real world of policy making, general directions are given, general objectives, rough estimates and projections, are formulated. Our hypothetical example was therefore a simple illustration of some of the most serious practical limits that the application of cost-benefit analysis to EPZs may encounter.
Export-processing zones are an instrument of public policy given a mission with multiple objectives that are difficult to assess. The underlying assumption of the application of cost-benefit analysis - that it applies where and when financial efficiency, or profitability, are the end - makes such application undesirable, even counter-indicated. Asserting that cost-benefit analysis constitutes a core tool for the assessment of policy instruments is based on an over-simplified conception of the process having resulted in the implementation of a zone. Indeed, public EPZs have not been created for the sole purpose of generating direct revenue for the government, either through rental or selling of industrial property (the bulk of revenue for private export processing zones), or through direct taxing of investments, activities or income of EPZ investors, managers, technicians and workers.
Cost-benefit analysis cannot be considered to provide an accurate evaluation of public export-processing zones. Serious problems arise when it is to be decided what constitutes a cost and what constitutes a benefit. More significantly, the validity of the use of cost-benefit analysis to determine the quality of a public policy instrument such as EPZs, with their multiple, overlapping and sometimes unclear missions, is to be seriously questioned, particularly when, as Warr intends, it is conceived as a central appraisal tool.
Peter Warr's Cost-Benefit Analysis: Theoretical Frame-work and Methods
Warr bases his analysis on the necessity to reveal the evolution of the economic welfare of citizens of the host country" induced by the existence of a zone. Welfare is not explicitly defined by the author. It can be assumed, however, that it is understood as relating to the level of revenue per capita and must be in line with the "Pareto optimality" principle. It is therefore a narrowly economic understanding of a concept having also sociopsychological and political aspects. By narrowing the implicit definition of welfare, Warr narrows the perspective of his analysis. It also makes his conclusion approach, detailing his model of the export-processing zone: Consider (a) the flows of goods and services and the financial flows between the EPZ and the rest of the world and (b) the flows between the EPZ and the host country. The essence of the enclave approach is that the flows in (b) are relevant for evaluating the welfare impact of the zone, but that the flows in (a) are not.
Under such a model, the EPZ is not a "normal" component of its national economy. A country establishing a zone hosts a foreign enclave. The flows between this enclave and the world are external to the national economy and the latter is solely concerned with welfare gains or losses in the flows taking place between it and the zone. Therefore, a cost-benefit analysis will be limited to these flows and will not consider the former relevant. What are these flows which are supposed to impact on the welfare on the "host" economy's citizens? They are: profits and losses, foreign exchange earnings, employment, technology transfer, purchase of domestic raw materials and capital goods, use of electricity, domestic borrowing, taxes and development and recurrent costs.
For each of these flows Warr carefully explains what is going to enter his calculation. For none of them is data readily available. Each necessitates processing because each presents challenges to consideration in a cost-benefit analysis. The author is consequently constrained by available data and must produce information either through interpretation or through limiting definitions.
Once the flows have been defined, Warr operates his analysis. Benefits of zones include, significantly, employment and foreign exchange as Warr defines them. For the four zones, employment accounts for more than half of the benefits, while local supplies and tax revenues are marginal. Costs includes the infrastructure development, sometimes high enough to outweigh the benefits listed above. Subsidized utilities constitute another major cost that can also outweigh benefits.
Policy Recommendations and Cost-Benefit Analysis, or the Legitimization of a Constricted Approach
Peter Warr's analysis of export-processing zones certainly constitutes a worthwhile effort to bring a new perspective to the phenomenon. An assessment of this effort should be, however, primarily concerned with its theoretical and methodological frameworks in order to put in the foreground the body of presuppositions that condition the direction of the study. I have demonstrated that such study is by no means neutral and is conditioned not only by its scientific foundations, by also by the sampling, and the technical difficulty that translating the scope of benefits and costs into computable data represents. Another considerable limit of the approach is that most of its conclusions are dependent on the period when it is realized. Indeed, Warr studies four public zones at a given moment of their existence, at a time when the concept of EPZ has characteristics that may evolve. His verdict on the validity of developing export-processing zones is contingent on the particular shape of the concept in 1982. Since then, zones have evolved considerably, adding new categories of activities to the traditional manufacturing base and, of central significance to us, being increasingly developed by private capital. Such evolution has tremendous repercussion on the welfare impact of zones.
Certainly, the most contestable aspect of Warr's cost-benefit analysis is its last part. There, the author briefly attempts to generalize his observations to the whole space of export-processing zones. And his conclusions are somewhat radical is we remember that they are based on only four zones featuring very distinct attributes, that his framework conducts him to formulate specific assumptions that can be contradicted, that the nature of the cost-benefit analysis results in an inevitable simplification of what can be introduced in it, and that the same analysis is inevitably short-sighted because of the difficulty to operationalize some of the costs and benefits. Warr's main argument against export-processing zones resulting from his analysis is that the benefits they bring to the local economy are rather limited. An expansive instrument, their beneficial impact on welfare is not proven whereas they represent a significant financial drain on government budgets. EPZs are therefore diverting scarce financial resources that should instead by channeled toward nationally-based deregulation. The author advances that export processing zones could even have a negative welfare effect and result in delayed liberalization reforms. In fact, zones would be beneficial only in the early stages of development and would significantly lose value as host countries would climb the development scale. Warr indicates that, for example, the Republic of Korea gives less value to its zone in Masan than during the 1960s and 1970s.
Warr's conclusions are far-fetched. A close examination of his analysis demonstrates that the basis for an investigation of "the policy implication of EPZs" is far too thin to safely allow extensive policy recommendation. In particular, the conclusions reached about the problem that EPZs pose in relation to economic liberalization seem to indicate the preexistence in the author's approach of a structured thought that may have oriented the result of his cost-benefit analysis prior to its operation.
Conclusions: Revising the Basis of Export Processing Zones Analysis
Export-processing zones have multiple facets. Set in a complex environment, they are objects of multidimensional aspects, are at the center or the periphery of multiple strategies, have multifarious benefits and multifarious costs. None of these are evident, appearing spontaneously through the lens of a single analytical tool. The main benefits can be long hidden or can be revealed only through careful analysis applying diverse and contradicting methods. Benefits can be expressed in ways impossible to render through simple computation. The same goes, obviously, with costs. Most important, zones can be overall beneficial in a given country and counter-indicated in an other. What this indicates is that one should be extremely careful about the generalization of one's localized observations. One should also remain deeply conscious about the implications of a ready-made recipe for economic development.
The Export-Processing Zone is above all a public policy instrument dedicated to improving, among other instruments, the economic situation of the country implementing it. On a case by case approach, where benefits and costs are difficult to assess, the concern of the policy analyst should be to come up with a study as complete as possible, revealing through data analysis and extensive interviewing of the actors concerned the multiple dimensions of costs and benefits. Such approach requires a different model of the export processing zone, a model at the cross-roads of the major disciplines of social science; economics, political science, sociology, geography. Informed by diversified perspectives, concerned with diverse issues, the policy analyst would present to the policy maker a picture of the zone that would be less caricatural, without any doubt, but also more complex.
REFERENCES
General Economic Analysis
Sectoral Economic and Policy Analysis
Case and Comparative Studies