HLE COPY SWP669 Investment Incentives for Industry Some Guidelines for Developing Countries Alice Galenson WORLD BANK STAFF WORKING PAPERS Number- 669 WORLD BANK STAFF WORKING PAPERS Number 669 Investment Incentives for Industry Some Guidelines for Developing Countries Alice Galenson LNTE11NfATIONAL MONETARY FUND JOINT LIBERARY C Y 2 9 198A1 INTERiNATIONAL BANY FOR RECO.iSThUCT0I1i AND DEVELOPUENT WASHNINYCIC. D.C. 2031 The World Bank Washington, D.C., U.S.A. Copyright ~(() 1984 The International Bank for Reconstruction and Devclopmcnt/THE WoRLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing September 1984 This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The views and interpretations in this document are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting on their behalf. Any maps used have been prepared solely for the convenience of the readers; the denominations used and the boundaries shown do not imply, on the part of the World Bank and its affiliates, any judgment on the legal status of any territory or any endorsement or acceptance of such boundaries. The full range of World Bank publications, both free and for sale, is described in the Catalog of Publications; the continuing research program is outlined in Abstracts of Current Studies. Both booklets are updated annually; the most recent edition of each is available without charge from the Publications Sales Unit, Department T, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from the European Office of the Bank, 66 avenue d'l6na, 751 16 Paris, France. Alice Galenson is an economist in the Western Africa Projects Department of the World Bank. Library of Congress Cataloging in Publication Data Galenson, Alice, 1948- Investment incentives for industry. (World Bank staff working papers ; no. 669) Bibliography: p. 1. Investments--Africa. 2. Investments--Developing countries. I. Title. II. Series. HG5822.G35 1984 332.6'732'091724 84-17418 ISBN 0-8213-416-X ABSTRACT Developing countries frequently offer special incentives to investors as a way to promote industrialization. Many of the incentives offered, however, and many of the criteria used to select the benefi- ciaries have actually tended to encourage a structure of production that is inefficient, ill suited to the pattern of comparative advantage and sometimes inconsistent with the original objectives of the incentive system. This paper describes the incentives most frequently offered (with the exception of protection and other trade incentives) and shows how they can lead to distortions and reduced efficiency in the economy, while proving costly to the government in terms of foregone revenue. It then reviews the evidence on the impact of investment incentives and presents some general principles that could be used in designing incentive systems. ACKNOWLEDGEMENTS An earlier draft of this paper was used for a series of seminars on industrial policy in Sub-Saharan Africa offered by the Economic Development Institute of the World Bank. I am grateful to the following people for useful comments and criticisms: Sri-ram Aiyer, Paul Ballard, Michel Beguery, Joel Bergsman, Gordon Billington, Robert F. Conrad, Serge Guetta, Tariq Husain, Donald Keesing, Alexander Keyserlingk, Jacob Levitsky, Wilfred Lewis, Stanley Please, Neil Roger, William Steel and Edward Tower. Michael van Bemmelen compiled background material on incentives in West Africa. Abrege Desireux de stimuler leur industrialisation, de nombreux pays en developpement s'efforcent d'attirer les investisseurs en leur offrant des incitations particulieres. En realite, les mesures proposees et les cri- teres de s6lection des ben6ficiaires favorisent souvent l'instauration d'une structure de production inefficace, mal adapt6e a l'avantage compa- ratif du pays et allant parfois meme a l'encontre des objectifs recher- ches. Le pr6sent document decrit les incitations les plus courantes (a l'exception des mesures de protection et autres incitations commerciales) et explique comment elles peuvent cr6er des distorsions dans 1'6conomie nationale, reduire son efficacit6 et entrainer un important manque a gagner pour l'Etat. I1 examine ensuite quel est, dans la pratique, l'im- pact de ces encouragements a l'investissement et presente quelques prin- cipes generaux applicables a 1'elaboration des systemes d'incitations. EXTRACTO Frecuentemente los paises en desarrollo ofrecen incentivos especiales a los inversionistas como un medio de promover la industrializaci6n. Sin embargo, muchos de los incentivos ofrecidos y de los criterios empleados para seleccionar a los beneficiarios en realidad han tendido a estimular una estructura de producci6n ineficiente, poco apropiada en cuanto a la tendencia de la ventaja comparativa y a veces incongruente con los objetivos originales del sistema de incentivos. En este trabajo se describen los incentivos que se ofrecen con mayor frecuencia (con excepci6n de la protecci6n y otros incentivos comerciales) y se muestra c6mo pueden conducir a distorsiones y a una disminuci6n de la eficiencia de la economia, al mismo tiempo que resultan onerosos para el gobierno desde el punto de vista del ingreso a que este renuncia para ofrecer dichos incentivos. Luego se examinan las pruebas del efecto de los incentivos a la inversi6n y se presentan algunos principios generales que se podrian usar al crear sistemas de incentivos. TABLE OF CONTENTS Page I. Introduction ............................................. 1 II. Objectives of Investment Incentives ...................... 2 III. Types of Incentives ..... ................... ......... . .. . 4 A. Measures to Enhance Investor Confidence ............... 4 B. Trade Policies ....................................... ............. 5 C. Administered Prices ........ . ......... 6 D. Investment Codes .. ................................... . 7 E. Infrastructure and Services ........................... 12 F. Employment Incentives ................................. 14 G. Interest Rates ............ .. .......... ........ . 17 IV. Some African Investment Incentive Systems ................ 18 A. Togo's Investment Code as a Typical Case .............. 18 B. Ivory Coast: The Evolution of an Incentive System .... 21 C. Nigeria's Industrial Incentive System .......... .... *.. 22 D. Administration of Investment Incentives ............... 26 V. Impact of Investment Incentives .......................... 28 A. Impact on the Economy. ...... o ......................... 29 1. Regional Comparisons: Sub-Saharan Africa ... ... 29 2. Ghaa.....an.....a$....0........ 30 3. Sudan ...32 4. The Philippines .. .. . . . . . . . .... .. . . ... 32 5. Thalnd........se*#......d.. 35 6. Co omba........................ 35 7. Mexce.......... i.............. 36 8. Puerto Rico .... *...0.......0...0..0 . 36 B. Impact on Investors...t...s..................... ....... 37 VI. Summary and Conclusions: Some Principles for Incentive Systems ................ .......... .....*.. . . 40 I. INTRODUCTION The general economic health and degree of political stability in a country, along with the presence of resources and markets, are probably the primary determinants of the climate for industrial investment. Given the satisfaction of these basic conditions, investors may still have a wide phoice of locations to choose from. In order to influence this choicd, governments provide a wide variety of incentives aimed at promoting both investment in general and investment in selected sectors or locations or with certain characteristics. Accelerated investment in indusftry is usually one of the primary goals of incentive policies. Many African countries package their incentives in investment codes or laws which specify the criteria for awarding benefits, the nature of the benefits, and the obligations of the beneficiaries. Other, sometimes more powerful, incentives may arise from the countries' policies in the areas of trade and tariffs; credit allocation and interest rates; price, wage and labor regulations; and government investment in infrastructure or productive projects. These factors can influence the price of an activity's inputs or outputs and can thereby influence the investment decision in a positive or negative way. Taken as a whole, this set of factors, including the negative ones, will be referred to as a country's incentive system. This paper will discuss the objectives of investment incentive systems, the types of incentives offered, their administration and the impact that they are likely to have on the level and nature of investment and on the economy. In many cases, the incentives offered by a country, when taken together, are inconsistent, contradictory or redundant. This paper will describe such situations, as well as those where appropriate incentives have been used. Most of the examples will be drawn from the African experience, but evidence from other regions will be presented as well. A broad range of commonly used incentives will be dicussed, but trade policies, which may be the most important, have been covered extensively in the literature and will be included here only in so far as they form part of the incentive package offered to investors. Section II will -2- discuss the reasons and justification for offering incentives. Section III will review a number of incentives and their likely impact. Section IV will present some typical African incentive systems, and Section V will review the evidence on the impact of such systems. Finally, Section VI will draw some lessons from the experience with incentives and reiterate some general principles for incentive packages. This final section can be read on its own for a broad summary of the paper's main conclusions. II. OBJECTIVES OF INVESTMENT INCENTIVES Some incentives are directed simply at raising the general level of investment in the country. While these may be aimed in part at increasing the rate of domestic savings and investment, they are in large part intended to attract foreign investment. In addition, most incentives try to some extent to direct resources into certain priority areas. Many investment codes specify their objectives, either explicitly or implicitly through the criteria for admission. These objectives may be as broad as consistency with the country's development plan, or more specific, for example to increase domestic value added and employment, improve the balance of payments, or promote the development of particular regions. Sometimes the government chooses desired sectors or subsectors for investment. Industrialization is usually a priority, under the implicit assumption that it will speed the country's economic development, both through its direct effects on output and employment and through the externalities associated with it. In addition, diversification out of primary production is expected to lessen the disruptive impact of price fluctuations and adverse trends in the terms of trade. A national industrialization strategy often serves social and political, as well as economic, ends. For example, the location of some key projects in a particular region might demonstrate the government's commitment to improving the status of a certain ethnic group or to improving income distribution. Strategic considerations might require the development of a few basic industries or the attainment of self- sufficiency in some key consumer goods. - 3 - Although the objectives cited above have merit, they do not always justify the use of special incentives. In freely functioning markets, prices play a key role in allocating resources efficiently, ensuring that the structure of production in a country will be consistent with the resources available to it. If an investment is profitable, then it will be undertaken without further incentives; if not, then it should not be,undertaken. If capital is scarce relative to labor, for example, it should earn high returns, which alone should suffice to attract capital. In the real world, however, the market is already distorted by numerous forces (e.g. minimum wages, administered prices, or tariffs). The most direct way to deal with distortions is to eliminate them, but this is not always feasible nor desirable, particularly in the short run, and compensating measures in the form of incentives may be justified. Externalities, or spillover benefits from an investment, may also justify the use of incentives. For example, the creation of a skilled labor force may benefit the country as a whole beyond the profits to the firm providing the training, so society should bear some of the costs. A product used for national defense serves the general public as well as the firm that produces it. Finally, economies of scale and the need for a significant learning period in an industry create the case of an infant industry which becomes profitable only in the long run, but which would not be undertaken at all in the short run by a private investor. If the investor could foresee the long term gains and obtain finance for the interim period, incentives would not be necessary, but imperfections in the capital market are likely to preclude such a solution. Although infant industries may be difficult to identify, they do represent a valid case for intervention. For all of these reasons - existing distortions, externalities and market failures - an investment that is economically profitable for the country as a whole might appear unprofitable to an individual entrepreneur. Appropriately designed incentives can narrow the gap between public and private returns and induce the entrepreneur to undertake the investment. The use of incentives may entail losses as well as gains. They impose both financial and administrative burdens on the government and they can lead to a distortion of the country's productive structure, - 4 - rendering it less efficient. For example, the promotion of investment in industry (i.e. by maintaining high prices) may result in disincentives to agricultural production, thus reducing agricultural incomes, increasing rural-urban migration, and exacerbating the unemployment problem. Policies to promote regionalization, to improve income distribution or to ensure national defense may impose similar costs, which must be balanced against the expected benefits. III. TYPES OF INCENTIVES Once the decision is made to use incentives, a wide choice of instruments is available. The more direct the instrument, the less likely it is to create costly side effects, and conversely. For example, tariffs used to encourage local production by restricting imports raise the price of the protected product, thus discouraging consumption and creating a perhaps unnecessary welfare loss for society. A more direct production subsidy (financed by a neutral tax) would cause fewer distortions in the economy. If the goal is employment creation, a direct subsidy to labor use could be used. Instead, one of the incentives most frequently used. to attract capital (tariff exemption) tends to encourage its use in relatively capital-intensive activities. Similarly, some methods used to promote import substitution, save foreign exchange and improve the balance of payments tend to promote an industrial structure heavily dependent on imports and unable to export. This section will discuss the probable impact of the benefits found in most investment codes, as well as a much broader range of interventions that affect investment decisions. These include the provision of subsidies or services (e.g. training, infrastructure, or direct subsidies to production, employment or capital) and factors such as price controls or minimum wages. More important than any of these factors, however, are the assurances and guarantees offered to reduce the risks faced by investors. A. Measures to Enhance Investor Confidence Surveys show that among the most influential factors in the location decision for foreign investors are political stability, favorable terms for the transfer of profits and the repatriation of capital, absence of discrimination against foreign ownership and control, and freedom from burdensome regulations (Diamond and Diamond, 1983, 1/ Vol. 1, pp. 6-7; Robinson, 1961, pp.4-5). Predictability of conditions and lack of arbitrariness may be the most important assurance that can be offered to investors, who seem able to adapt to practically any conditions as long as the rules are clearly established in advance and followed subsequently (Frank, 1980, pp.111-12). Provision for compensation in the event of nationialization of a business also falls into the category of necessary assurances, without which all the economic incentives together may have no impact at all. The presence of such factors reduces the risk and uncertainty associated with investing abroad, thereby reducing the rate of return required by the investor. 2/ B. Trade Policies In many developing countries trade and exchange rate policies are among the most important determinants of industrial investment. An open trade policy can enhance a country's development in line with its comparative advantage, and substantial evidence exists of more rapid growth among outward oriented countries than among inwardly oriented ones. 3/ On the other hand, high levels of protection encourage production for the domestic market, even though it may be inefficient and ill suited to conditions in the country. Protection can lead to the establishment of inappropriate industries. For example, high tariffs or quantitative restrictions instituted to discourage the consumption of luxury goods may also have the unwanted result of encouraging production of these goods. Resources attracted to protected industries might otherwise be directed into more efficient uses, such as export industries, 1/ This two-volume work provides country-by-country information on legal requirements and incentives for investors. 2/ Section V. B. below will present additional evidence from surveys of investors. 3/ The evidence for this assertion can be found in a number of published sources. These include Balassa, 1982 B; Balassa, et al, 1982; Krueger, 1978; and Krueger, 1983. which, in the absence of subsidies, would be forced to be efficient to remain competitive. Wide differentials in the level of protection to different industries, and the systematic bias against export industries, can create a structure of industrial production that is inefficient and inconsistent with a country's comparative advantage. Trade policies and their impact have been well covered in the literature; this paper will refer to them only insofar as they form part of the incentive package offered to investors. C. Administered Prices Protection of domestic production may reduce the pressures that would otherwise keep prices at competitive levels. If only one domestic firm produces a good, the absence of competition from imports may enable it to charge monopoly prices. As a result, many countries that restrict trade in order to encourage domestic industry have found it necessary to establish price controls. Reasons often cited for their use are to moderate inflation, to keep prices of basic consumer goods low, and to allow firms to achieve "reasonable," but not monopolistic, profits. Cost-plus pricing is one way of administering prices. In Cameroon cost-plus pricing of a wide range of products permits an ex- factory price that incorporates a margin of 12% on top of the cost of all current inputs, services, wages, indirect taxes, depreciation and finance charges. Prices need to be certified before they can become effective, and the claimed expenses must be documented. This method has several drawbacks. It reduces the incentive to lower costs and improve efficiency, provides an incentive for firms to bias their cost estimates upwards, and ignores the principle of comparative advantage by allowing identical profit margins for all activities, regardless of efficiency. To the extent that controls keep prices below their free market level, they encourage increased consumption while discouraging production; this has happened particularly in the case of agriculture. Administration of a price system presents additional problems. If done carefully, with independent price checks, it requires substantial manpower. It also allows ad hoc and inconsistent decisions. If margins are not fixed, but rules simply call for "fair" prices, this danger is even greater; as a result, foreign investors will come in to some countries only with the government as a partner so as to ensure the enterprise reasonable treatment. Long delays in getting changes approved can cause serious problems for firms, especially in a situation of rapid inflation. In Cameroon, manufacturing industries must apply individually for an authorized price for each product, and a slight change in design requires a new official price. The same is true of an increase in the price of raw materials or any other input that the manufacturer wishes to pass Qn to consumers. In practice, delays of six months or longer can occur. D. Investment Codes Investment codes are a common feature of African incentive systems. They describe the benefits offered, the eligibility criteria and the obligations of the government and the investors. Investment codes generally provide, tariff and fiscal concessions to firms that meet certain conditions, often related to size, choice of sector, employment creation, location, and use of domestic raw materials. Although investment codes usually apply equally to domestic and foreign investment, they are aimed primarily at the latter. In addition to the benefits they offer, inv,estment codes can be an important element of the investment climate. They signal the government's interest and desire for a stable environment for investors, and many of them spell out guarantees to investors in areas such as repatriation of profits and equality of treatment for foreign firms, thus enhancing investor confidence. At the same time, investment codes, by establishing fixed rules, can relieve pressures on the government for ad hoc granting of concessions. Tariff concessions are an integral part of many investment incentive packages. Material and equipment inputs to production, and sometimes raw materials, are frequently exempt from import duties for a period usually lasting 5 to 10 years, but sometimes as long as 15 to 25 years for major projects (e.g. in Togo and Guinea). Mauritania extends an exemption on raw materials and spare parts for 12 years to firms located outside of the main cities, and Niger gives exemptions of up to 15 years to small enterprises. The exemptions often apply only to goods not produced domestically. - 8 - This benefit may be costly not only in terms of government revenue foregone, but also in terms of productive efficiency, through distortions in the allocation of resources. Exemption of duties on capital equipment favors the use of capital over labor, since labor not only is usually taxed, but is also often subject to a minimum wage greater than its opportunity cost. Thus the exemption will tend at the margin to increase the capital intensity of the production process over what might have been the case without intervention. Highly sophisticated imported technology could also increase dependence on expensive expatriate technical staff, due to the nature of some capital intensive processes. Furthermore, the duty exemption increases the rate of effective protection, sheltering inefficient firms from competition, and tends to discourage domestic production of the exempted goods. Thus while the system of protection fosters import substitution in final goods, it could actually increase the country's dependence on imports. Export incentives are sometimes also offered. Exemption from export taxes is frequently offered to priority industries under investment codes. Other export incentive measures include export development funds, tax credits for duties paid on imported materials or supplies, preferential tax treatment of income from exports, export subsidies, free trade or export processing zones, concessions on domestic sales in return for export performance, foreign exchange credits from export earnings (in countries where foreign exchange is constrained), and export insurance. These measures are, however, usually outweighed by those favoring production for domestic markets. Tax concessions are one of the most widely used incentives for industrial investment and growth. 1/ Tax policy can be used to create a favorable climate for investment or it can be used to promote particular industries, regions, or types of investment. The following discussion takes the existing tax system as given; a more comprehensive treatment would analyze the impact of the entire system (in both host and home 1/ More material on tax incentives for industrial investment can be found in Heller and Kauffman, 1963; IMF, 1981; Lent, 1967; David Lim, "Taxation Policies," Chapter 5 in Cody, Hughes and Wall, 1980; and Usher, 1977. countries), which might already be biased. Furthermore, investors may prefer moderate taxes coupled with fair enforcement to highly concessional short term rates followed by high taxes imposed in a discriminatory or arbitrary manner. Indonesia recognized this fact when it eliminated special tax incentives in favor of lower tax rates. The tax holiday offers full or partial exemption from income and other taxes for a period generally ranging from 5 to 10 years0 Most African countries fall into this range, with some exceptions, such as the Ivory Coast, which grants up to 25 years of holiday and Niger, up to 15 years. In some cases, the length of the period of exemption varies directly with the size of the investment, while in others, mainly outside of Africa (e.g. Malaysia and the Philippines), it varies with the number of jobs created. Other criteria may also be used, including location, production of "priority" goods, and use of local raw materials (e.g. in Ecuador, Malaysia and Morocco). Tax holidays allow firms to recover their capital more quickly and maintain greater liquidity in their early years, thus reducing the risk. However, they have limitations. A tax exemption is worthless if there is no tax liability, and many firms, particularly in infant industries, earn little profit in their early years. Some countries solve this problem by letting the period of exemption begin not in the first year of operation, but in the first year that profits are earned. Others, including Nigeria and Malawi, allow losses incurred during the holiday period to be written off against profits earned later. As with tariff exemptions, the most obvious cost of a tax holiday is the revenue foregone by the government, although this is not a cost if the investment would not have been undertaken without the tax incentive. Such causation is difficult to determine, making this cost hard to measure. Another possible cost is through the impact on resource allocation. Depending upon the existing tax system and on how the tax holiday is applied, it could influence the capital intensity of investment - 10 - and might encourage short term investments, designed to earn profits quickly, closing at the end of the holiday period. 1/ A lengthy period of exemption from taxes can be very costly to the treasury, while it is likely to be less important to the enterprise concerned. Given the uncertainties attached to investing in a developing country, and the resulting short time horizon over which he will seek profits, the private entrepreneur's discount rate (the rate of interest at which he discounts future earnings) is likely to be high compared to that of the government. The benefits offered and profits expected in the early years of a project are likely to have the greatest impact, and the more extended the period for which incentives are offered, the greater their cost will be to the government relative to their value to the firm. One way to ensure against excessive revenue loss from firms that become profitable quickly would be to set a ceiling on their benefits, e.g. by allowing a tax credit only up to a specified amount. Senegal restricts exempt earnings to 100% of the original investment, and Liberia, to 150%. Other solutions have been found by Guatemala, which suspends the exemption if annual earnings exceed 20% of the invested capital, or the Philippines, which does so if annual earnings exceed 30% of the costs of production. Such rules can be difficult to enforce, however, particularly given the poor accounting practices followed in many cases, and the required monitoring could prove costly. The attractiveness of tax concessions to foreign investors depends heavily on the tax systems in their home countries. France has signed agreements with its former colonies in Africa that allow tax credits for taxes paid in the host country at the corporate level to be deducted from individual income taxes at home. Japan, the United States, Great Britain and Germany permit foreign income taxes paid to be credited against income taxes on income from foreign sources. However, this simply eliminates double taxation; it does not prevent the home country from taxing the income that is exempt from tax in the host country. If the tax rate is at least as high in the home as in the host country, then tax 1/ Such a case is described in Section V. A. 8. holidays will have no effect on a firm's tax liability, and will in fact merely transfer revenue from the host to the home country. For this reason, some countries have "tax-sparing" agreements, whereby the investor's home country gives credit for taxes that would have been paid in the host country if no tax exemption had been given. Such treaties are negotiated on a bilateral basis and are obviously of key importance in determining the usefulness of tax holidays to investors. Tax sparing agreendents were implemented widely during the 1970's, but some countries, notably the United States, have not adopted them. (Robert Antoine, Chapter 2 in Hellawell, 1980, pp. 69-71). Tax holidays should not be offered in the absence of such agreements. Many of the francophone countries in Africa stabilize the fiscal regime applying to important projects at its existing level for extended periods. The members of UDEAC (Cameroon, Central African Republic, Congo and Gabon) offer a stable fiscal regime for 25 years for large projects, as do Chad, Guinea, Mali and Togo. Mauritania, Niger and Upper Volta permit stabilization for 15 to 20 years, and many other countries offer 10 year stabilization conventions. As with lengthy tax exemptions, the cost of this benefit to the Government in the later years may far outweigh its value to the entrepreneur, and it also greatly reduces the government's flexibility in responding to new economic conditions. Another commonly used type of tax incentive is accelerated depreciation, which allows a firm to write off the cost of its capital equipment against its gross revenue in a short period. Like tax holidays, this benefit increases liquidity and reduces risk. However, it also reduces the cost of capital and rewards firms in proportion to the size of their investment, rather than their profitability, thereby tending to encourage larger and more capital intensive industries or methods of production. An incentive frequently used to encourage expansion of firms is the reinvestment allowance, which exempts reinvested income from the corporate income tax. Another way to encourage reinvestment is to tax the portion of tax-exempt corporate profits that is distributed as dividends; Sierra Leone has such a provision. However, this could reduce the supply of equity funds. - 12 - E. Infrastructure and Services Inadequate infrastructure can prevent the establishment of new firms, particularly in the more remote areas of a country. Roads, railroads, and ports are necessary both for obtaining inputs and for selling the output. In some countries, enterprises are forced to provide their own power, or at least a back-up generator to fill in during frequent power interruptions, to avoid the costs associated with loss of production or shutting down and restarting machinery. Inadequate telecommunications systems can slow work, also reducing returns to investment. Finally, adequate water supply and disposal are indispensable for some industries. For these reasons, except for those activities attracted to sources of raw materials or specific markets, modern industries tend to concentrate in major urban and port centers. Infrastructure in many developing countries has deteriorated in recent years. As public funds have been channeled increasingly into direct investment in industrial and commercial enterprises, many of which have become a steady drain on the budget, and into prestigious new projects, the funds allocated for maintenance of existing networks have been inadequate. Many governments offer incentives to firms that will locate outside of the major centers, in order to improve the distribution of income, slow the rate of rural-urban migration, and avoid the negative externalities associated with large urban agglomerations. Tax incentives may be ineffective, however, in the face of inadequate infrastructure, and import duty exemptions might simply encourage greater use of imports, increasing the advantage of locating near a port. Building and maintaining infrastructure could remove much of the disincentive attached to more distant locations, while at the same time meeting various social or equity objectives, such as improved access to schools and hospitals, wider availability of consumer goods, or rural electrification and sanitation. The costs and benfits of such investments should, of course, be analyzed like any other. Non-economic goals may be taken into consideration, but the costs should be recognized explicitly, since government resources can be used in alternative ways to promote social objectives. - 13 - Many countries create industrial estates where firms can buy or lease land furnished with access to some or all of the following: transportation, water, power, sewerage, telecommunications, common service facilities, and factory buildings. However, industrial estates often incur high costs and experience low occupancy rates. As an added incentive, sites are sometimes provided at subsidized prices. For example, the Industrial Parks Authority of Cameroon (MAGZI) develops and rents iand granted to it by the government at rates well below the market, thus artificially stimulating the demand for land. Industrialists ask for and get more land than they need, sometimes then subleasing it at higher rates. Adequate infrastructure is necessary to attract factories, but they should be required to pay commercial rates. Industrial estates have also been used to promote regional dispersion of industry. In order to correct regional imbalances and reduce urban congestion, Japan provided factory sites with water, transportation, communications and electric power supplies, as well as housing, water supply, sewerage, educational and entertainment facilities for the workers. In addition, local governments offered tax incentives to enterprises locating in the new towns. Despite all of these incentives, entrepreneurs still preferred to invest in the larger, more congested urban areas. The experience in India has been similar; estates near the metropolitan centers have been relatively successful, while those in less developed areas have low occupancy rates (Datta-Chaudhuri, Chapter 8 in Cody, Hughes and Wall, 1980 pp. 253-4). Kenya, on the other hand, has had some success in developing local industry in small rural towns through the use of mini-estates, smaller and less costly, and with fewer services than the larger models that have so often proved unsuccessful. Another service frequently provided by governments is assistance in identifying, financing, implementing, and managing projects. A government or other institution might carry out pre-investment studies of promising projects, so as to provide information on markets, availability of raw materials and supply of infrastructure to interested investors. The Caribbean Project Development Facility, for example, is a regional organization that identifies projects, arranges market studies, assists with feasibility studies and locates interested investors and donors. A - 14 - fund could be established to finance studies, to be paid back out of the projects that materialize. Funds or development banks are often established to lend, guarantee, subsidize or invest funds in promising projects; these methods are often geared particularly to promoting small enterprises. Once an enterprise is in place, the government or other organization might provide advice on production processes and marketing techni,ques, or assistance with training. These services should be paid for, at least in part, by the beneficiaries. The imposition of a fee helps to ensure that they will be used efficiently. F. Employment Incentives 1/ Employment creation is an underlying, if not always explicit, objective of most industrialization programs. The instruments used to seek this goal vary widely, ranging from general incentives for increased investment to subsidies specifically aimed at the use of labor. Many incentives seek to attract capital, which is scarce, in order to employ more labor. However, since many manufacturing processes can be carried out using a range of combinations of capital and labor, an incentive that reduces the price of capital may increase the capital-labor ratio. The range of activities undertaken in a country could also be influenced by distortions in relative factor prices. In a labor surplus economy, wage rates should be low enough to encourage labor-intensive industries with no further incentives. Neutral incentives to promote investment and growth generally, while leaving the relative costs of capital and labor unchanged, would be preferable from the point of view of productive efficiency. Several arguments might be offered, however, for targeting the incentives directly at labor use. First is the presence of distortions in the market, frequently caused, for example, by minimum wages that fix the cost of labor above its opportunity cost, thereby discouraging its use, or by guaranteed employment or high severance pay, which have the same effect. Many African countries have minimum wages that make wages high relative to those in other regions, and some have restrictions on retrenching workers. In a recent survey of 1/ For a fuller discussion of this topic, see ILO, 1972. - 15 - multinational investors, several companies asserted that restrictions on their right to dismiss employees had encouraged them to use more capital- intensive methods of production than would otherwise have been the case (Frank, 1980 p. 87). Removal of the distortions would be the most efficient solution. In Mauritius, for example, firms specialized in export are exempt from the indemnification of laid-off workers. However, such measures might not be politically palatable. A wage subsidy or tax incentive related to labor use could reduce the distortions, although monitoring of payrolls might prove difficult. A second argument for employment incentives is the presence of externalities. For example, if high wages result from a lack of trained workers, rather than minimum wages, the use of labor in an enterprise could benefit the economy as a whole by helping to create a trained labor force. If workers are free to change jobs, the firm that bears the cost may not capture all of the benefits, and training subsidies would be appropriate. Another reason sometimes offered for employment promotion is a social one, that employment creation is necessary to improve income distribution, raise incomes and reduce urban unemployment. While income distribution might initially be improved by employment incentives, the promotion of labor use beyond its economically efficient level could well reduce the growth rate of income, thereby worsening conditions in the longer run. Furthermore, creation of more jobs in urban areas might increase the rate of rural-urban migration and lead to even higher unemployment. Finally, it must be recognized that the cost of creating jobs in industry is higher than in other sectors and that given the size of the industrial sector in most African countries, it will be many years before it can have a significant impact on the severe unemployment that exists today. Far more important in the short and medium terms are policies toward agriculture, particularly the maintenance of producer prices at a level adequate to retain the existing agricultural labor force. In an effort to promote employment, incentives are sometimes awarded on the basis of the number of jobs created or the ratio of labor to capital. In Senegal, for example, enterprises can qualify for - 1 6 - investment code benefits by investing 200 million CFA Francs and creating at least 50 jobs for Senegalese, or simply by creating 100 jobs, and in Benin a minimum of 15 jobs is required to qualify even for the regime for small and medium scale enterprises. Niger allows the creation of 500 jobs to substitute for other criteria relating to size of investment and value added in awarding the convention regime to large enterprises. However, if the concessions themselves reduce the cost of capital relative to labor (e.g. exemption from import duties on equipment), this can dilute the impact of the incentive on employment. Alternatively, a firm might temporarily employ more labor than it really needs, without changing its production process, or use part time employment in order to qualify for the concessions; monitoring would be difficult. Employment subsidies are the most direct way to promote the use of labor. 1/ The difficulty in subsidizing employment for many countries lies in the fact that, unlike tax concessions, subsidies must be explicitly accounted for and financed in the budget. 2/ Tax measures adopted to finance the subsidies might have offsetting effects on employment. For example, in some countries, the combined burden of payroll taxes on wages, apprenticeship and social security can be as high as one-quarter of the payroll (Lent, 1967, pp.159-60). 3/ The effect of these taxes on the allocation of capital and labor depends on their incidence. If the tax can be shifted to the employees, it will not affect the demand for labor (it may, however, affect the supply of labor, as well as the distribution of income). The use of general tax revenues would eliminate this problem, but might be politically unpopular. Finally, subsidization of labor in qualifying firms ignores the employment created 1/ In Canada, for example, the Regional Development Incentives Act of 1969 included grants which could reach Can$30,000 per job for new investment or expansion of an existing business. 2/ This could actually be seen as an advantage, since it requires explicit recognition of the cost of the program. 3/ Lent found combined ratios of these taxes to the payroll in 1969 of over 25% in Algeria and Guinea, 15-20% in the CAR, Congo, Ivory Coast, Mali, Togo and Upper Volta, and around 10% in Burundi, Chad, Gabon, and Senegal. Payroll taxes alone ranged from 2.5 to 8%. - 17 - indirectly by the activity through its purchases of raw materials and intermediate goods. Estimating and subsidizing indirect employment would probably be administratively impossible (and expensive), and yet the indirect effects might be at least as important as the direct ones (Usher, 1977, p.135). Alternatives to a subsidy could be to reduce the level of payroll taxes or to allow an extra tax deduction or credit based on wages or on training costs. In the Philippines, for example, the total cost of direct labor and half of training expenses (up to a limit) are tax deductible. A final way to promote employment might be to eliminate the bias against small and medium scale enterprises (SMEs) created by the fact that they are often excluded from the full range of investment code benefits. Smaller enterprises are often more labor intensive than larger ones, and equal treatment could reduce the capital-labor ratio of the industrial sector as a whole. Although SMEs are often eligible for some benefits, these are usually less favorable than those awarded to larger enterprises, and the administrative requirements alone might discourage SMEs from applying. G. Interest Rates 1/ Interest rate subsidies are frequently used to promote certain sectors or types of investment. For example, many central banks set lower discount rates for investment in agriculture or small scale enterprises. However, low interest rates reduce the price of capital relative to labor and might encourage more capital-intensive investment than would otherwise be the case. Furthermore, low rates make lending less attractive to banks and may lead to an excess demand for and rationing of credit. More broadly, low or negative real interest rates in some countries may have reduced the level of savings or prompted savers to place their money abroad. Higher interest rates might have as much effect on increasing the availability of capital for investment as many of the other incentives presently being employed. 1/ Interest rates are of key importance in many countries. They are, however, beyond the scope of this paper and are therefore mentioned only briefly. - 18 - IV. SOME AFRICAN INVESTMENT INCENTIVE SYSTEMS A. Togo's Investment Code as a Typical Case 1/ Togo's investment code illustrates a number of the ideas discussed above. Adopted in 1978, it shares many features with other African investment codes, and many of the following comments could apply equally to the others. (Some differences will be noted.) The code begins by guaranteeing investors the right to transfer capital and profits out of the country and also assures foreign investors that they will not be treated less favorably than Togolese nationals (except under the SME regime). These are important elements in attracting foreign investment. The benefits awarded under the code fall under five different regimes distinguished mainly on the basis of the size of the proposed investment. The Togolese code is quite clear about the types of firms that qualify for benefits. It lists the eligible sectors, leaving somewhat vague only the requirement, shared by a number of other countries, that any project concur with the country's development plan. The commercial sector is omitted from the Togolese code (as it is explicitly in a number of other countries). Some countries have additional rules: Benin requires a minimum of 50% local value added in total output, Mali requires 40%, and Liberia, 25%. The economic rate of return is rarely mentioned as a criterion. In many countries the selection criteria are less precise. Cameroon does not limit the sectors that are eligible, requiring only that enterprises receiving the higher range of benefits be of "special importance" to the country. Upper Volta's code lists size of investment, training of local personnel, domestic value added, and positive impact on the balance of trade as criteria, but without specifying any amounts, leaving the way open for case-by-case decisions during implementation. In the Sudan, approval for industrial licensing can be obtained under any one of six criteria: defense or strategic importance, use of local raw 1/ Togo is in the process of revising its investment code. The code of 1978, however, provides a convenient illustration of the principles discussed in this paper. - 1 9 - materials, earning of export revenue or substitution for imports, employment generation, furthering of economic cooperation or integration with Arab and African states, and contribution to national income. Any activity would probably satisfy at least one of these conditions and thus they are of little value as guidelines for the selection of industries, leaving the decision up to the committee in charge (Acharya, 1979, p.29). Zambia's Industrial Development Act of 1977 lists the following six criteria: maximum utilization of domestic raw materials, production of intermediate goods used by other domestic industries, diversification of the industrial structure, creation of employment, improvement of domestic industrial skills, and promoting rural industries. To qualify as a "priority" industry, a project must satisfy two of the first three criteria and two of the last three. The criteria are not well defined and leave much to the discretion of the administrators of the system. Making the rules and benefits clear in advance is valuable to foreign investors, who may not even apply if they are uncertain of the outcome. It also saves administrative manpower in countries where this is often scarce and reduces the scope for arbitrary and ad hoc decisions. Like many others, the Togolese code implicitly assumes that bigger is better. Larger investments are awarded more benefits. Early African investment codes sometimes excluded small enterprises altogether, but later revisions corrected this imbalance. Nonetheless, small firms are often treated less favorably than larger ones. There is, however, no a priori reason to assume that larger investments will be more beneficial to the economy. In fact, large firms probably tend to be more capital intensive, and the use of this criterion may well encourage a larger and more capital intensive scale of production than is suitable to a developing country's factor endowment. Larger enterprises are more likely to be owned and managed by expatriates and may discourage the growth of small scale enterprises which are more likely to be owned by nationals and which are in many countries one of the most efficient and rapidly growing parts of the economy. Since the purpose of Togo's code appears to be to raise the level of investment within some broad guidelines, it applies to investment for expansion of existing firms as well as to new projects. Restriction of - 20 - promotion to the latter might be justified if the purpose were to assist "infant" or "pioneer" industries which were the first in their field in the country and it would reduce the cost of the program, but it would risk creating monopolies. On the other hand, if benefits are awarded to a second firm that enters the market several years after the first, and if these benefits extend beyond the time when those of the first firm expire, then the second firm may gain a competitive advantage and the first will undoubtedly demand an extension of its benefits. A third alternative is to grant benefits to additional firms, but extend them only until the expiration of the first firm's concessions; this method is used in Mexico. One of the major benefits extended under the Togolese code is exemption from import duties on machinery, equipment and raw materials. This benefit tends to encourage a more capital and import intensive production process than might otherwise be the case, reinforcing the tendency created by using the size criterion to award benefits. This is particularly the case in Togo, since the exemptions are renewable after the initial period and the firm thus has little incentive to adopt a production process more consistent with the country's factor endowment (i.e. with the situation without incentives). Exemption from export taxes (combined with duty-free import of inputs) helps exporting firms compete in world markets, but there is no particular reason to restrict this benefit to firms accepted under the investment code. A second major benefit, the tax holiday, is potentially costly to the government and of questionable value to investors, and it may also discriminate with respect to the investment horizon. In particular, it provides less incentive to infant industries, which may not earn profits in their early years, although this is the type of investment for which incentives are the most easily justified. Awarding lengthier exemptions for location in less developed areas, as does Togo, may contribute marginally to a more equitable distribution of income, but if location in a remote area reduces profitability, the tax exemption will not be worth a great deal. More direct incentives - subsidies or provision of infrastructure - would probably be more effective. - 21 - Another Togolese benefit, accelerated depreciation, is biased toward the use of capital and is redundant when combined with the income tax holiday unless it can be applied after the holiday has ended. Carry- forward of losses compensates firms that do not benefit from the tax holiday, but it may also be redundant if its 3 year limit runs out while a tax holiday is still in effect. The fixed tax base and rates for from 10 to 20 years for larger projects reduces the flexibility of fiscal policy, while probably providing little incentive to investors beyond the first 5 or 10 years. B. Ivory Coast: The Evolution of an Incentive System 1/ Since independence, the industrial sector has been one of the most dynamic elements in the Ivorian economy. Between 1960 and 1980, starting from a very small base, manufacturing activity grew at the rate of 13 percent per year in real terms, and its share in GDP rose from 4 percent to 14 percent. Since 1980, however, the industrial sector has been going through a period of stagnation. Industrial value added decreased by more than 3 percent between 1980 and 1982, while investment in non-oil industrial enterprises declined by 20 percent in real terms. Until the beginning of the 1970s, Ivory Coast went through an initial phase of industrialization based on the continuous expansion of the domestic market and on the development of the West African Economic Community (CEAO) regional markets. This growth took place in the context of a generous Investment Code, similar to those adopted by other African countries, and a moderate tariff schedule that applied fairly uniform protection across various branches. Increasing wage costs which followed the rapid growth of the 1960's and the completion of the easiest phase of import substitution led to demands for somewhat higher protection, which was granted through a reform of the Customs Code in 1973. The acceleration of domestic inflation in 1975 and the overvaluation of the CFA franc which followed the coffee and cocoa boom in 1977, further eroded the competitiveness of Ivorian enterprises, and led to increased protection through the use of quantitative restrictions. 1/ Based on recent analysis by the World Bank Western Africa Region. The experience of the Ivory Coast illustrates the linkages between trade policy and other investment incentives. - 22 - While the production of manufactured goods for sale on the domestic market expanded in the framework of a distorted system of protection, there were no equivalent incentives for exports to counter the anti-export bias inherent in the existing system of industrial incentives. Until the second half of the 1970s exports of manufactured products grew rapidly, essentially within the protected CEAO zone, and the need for an adequate export incentive system was therefore not directly perceived. The development of domestic production of intermediate products has been systematically discouraged by the system of industrial incentives in the Ivory Coast. This is due essentially to the high degree of tariff escalation according to the degree of processing, tariff exemptions on intermediate inputs for priority firms under the Investment Code, and duty-free entry, under the temporary admission scheme, of certain intermediate inputs which could be produced locally at a comparative advantage if the existing system of incentives were modified. In response to the deterioration of industrial sector performance, the Ivorian Government has decided to undertake a major reorientation of its industrial policy, based on a comprehensive reform of the system of industrial incentives and of the institutions in charge of industrial promotion. The importance of this reform is that it includes all elements of the incentive system (tariffs, export subsidies, quantitative restrictions, investment code), thus establishing a consistency rarely found in incentive systems that have evolved piece by piece over the last twenty years. C. Nigeria's Industrial Incentive System 1/ The economic climate for manufacturing in Nigeria has been conditioned by a range of factors which have had a considerable effect on the sector's structure and performance and entrepreneurs' investment and production decisions. In addition to direct public investment, government intervention includes trade and exchange rate policies, fiscal incentives, 1/ Based on recent analysis by the World Bank Western Africa Region. - 23 - and credit allocation and controls; regulations concerning indigenization, expatriate quotas, foreign investment and business permits; and other measures, such as assistance for SMEs, provision of infrastructure and services on industrial estates, and manpower training programs. General trade policies are beyond the scope of this paper; here we will concentrate on tariff concessions granted under "approved user status," fiscal incentives and the regulatory environment. "Approved user status" (AUS), established in 1958, grants manufacturers substantial concessions or exemptions on import duties for certain raw materials, components, spare parts, machinery and equipment. Typical concessions are a reduction of the import duty on steel bars and rods from 33 1/3% to 5%, and on certain electrical materials from 66 2/3% to 5%. Rates may vary depending on the location of the enterprise. The scheme has been extended to include waivers permitting imports of goods under ban or restrictive import license. AUS status is granted on a case- by-case basis by the Federal Ministry of Industries. In practice, firms are eligible for AUS when the item in question is unavailable locally, as indicated by a "letter-of-release" from local manufacturers stating that they cannot supply the product in question. AUS is valid for one year and can be renewed annually. The most important investment incentive measures in Nigeria's tax system are capital consumption allowances and tax holidays granted to activities with "pioneer industry status." Any trade, business or manufacturing activity can claim an initial allowance for depreciation of an asset in the first year of its use in addition to the normal annual capital depreciation allowance. The initial investment allowance ranges from 5% for building (other than industrial) expenses to 25% for plantation expenditure. The annual depreciation allowance ranges from 10% to 15% and covers buildings, equipment, mining, plantation and transport expenditures. Tax holidays of 3 to 5 years on corporate income are granted to pioneer industries, defined as any industry "not being carried on in Nigeria on a scale suitable to the economic requirements of Nigeria or at all" or any industry "for which there are favorable prospects of - 24 - further development." 1/ The government is also empowered to declare any other industry a pioneer if it is in the public interest to do so. Pioneer firms presumably face above average risks and make above average contributions to economic development, both of which could justify special concessions. However, real pioneers are not always easily identified. Markets may have been developed by importers or by close substitutes already in production, so that the new domestic producer faces little risk. The incentives described above have been of varying value to firms, but taken together, and combined with substantial protection from import competition (tariff rates range up to 500% and quantitative restrictions are common), they have undoubtedly had a significant impact on the level and structure of industrial production. By 1980, AUS covered a broad range of goods and benefitted more than 200 firms, mainly large and medium scale. It resulted in the creation of a complex two-tiered tariff system. It provides a significant additional incentive to imported input-intensive consumer goods industries that already benefit from protection from competing imports. At the same time, it may discourage the domestic production of intermediate goods, thus biasing the industrial structure toward final products. Small firms, although eligible for AUS have not used it; either they are not fully aware of its benefits or else they find the procedures too cumbersome. Export promotion in Nigeria does not compensate for the protection accorded production for the domestic market. A customs drawback scheme providing for duty rebates on imported raw materials used in the manufacture of goods for export apparently does not function effectively and, in any case, firms producing for local markets can receive the same benefit under the AUS scheme, in addition to protection of their output. The Nigerian Export Council has created an Export Development Fund and an Export Guarantee Fund and Insurance Scheme, but these have yet to have a major impact. 1/ Industrial Development (Income Tax Relief) Decree of 1971. - 25 - The capital consumption and depreciation provisions in the company income tax appear to cause further distortions in the incentive system, since different assets, and therefore probably different activities, are taxed at different rates. The system is apparently biased in favor of investment in buildings and structures and against investment in equipment and transportation assets, and biased in favor of investment in longer-lived equipment as against short-lived assets (e.g. autos, trucks). Furthermore, given inflation, the existing depreciation provisions may be inadequate. The importance of the tax holiday for pioneer industries depends on how quickly the firm becomes profitable. Since the holiday is granted for 3-5 years, and manufacturing profitability is usually lower in the initial years of operation, the incentive value may be very limited. Furthermore, the criteria for pioneer status are not clearly defined, and delays occur in processing applications, adding to the investor's uncertainty. Ease of evading taxes may also reduce the value of the tax holiday. Whatever the reason, in the two years ending in 1981, no firms applied for pioneer industry status. The impact of the depreciation allowances and tax holiday may be far greater when considered in combination with import protection. When excess profits are earned through protection, a tax holiday will permit all of these profits to be retained, thus increasing the value of the protective system and reinforcing its effect on resource allocation. The regulatory environment has an important influence on the climate for private sector investment. In many respects, Nigerian commercial and tax legislation closely follows the patterns found in other market-oriented economies, while many of the regulations resemble similar requirements in other developing countries. Although many of the laws and regulations apply equally for all private investment, more of them apply to foreign investors, for foreign exchange control and tax reasons, as well as to channel foreign investment into desired areas and encourage Nigerian entrepreneurship. The principal regulations involve company incorporation requirements, exchange control regulations on foreign investment and repatriation of dividends, Nigerian participation in foreign equity investment (reserving certain areas exclusively for - 26 - Nigerians and others for foreign-Nigerian partnerships), quotas on the use of expatriate personnel, and tax legislation. Although the substance of these laws has apparently not discouraged investors, their administration is another matter0 D. Administration of Investment Incentives The administrative requirements and procedures for obtaining investment incentives can have as great an impact as the incentives themselves. Stringent requirements and cumbersome and lengthy procedures can reduce or even cancel out the positive impact of the incentives. They can introduce uncertainty, possibly discouraging potential investors, and, depending on how the benefits are awarded and cut off, they can enhance or reduce the competitiveness of the activities concerned. Inadequate monitoring and enforcement can minimize many of the gains to the economy that the incentives are designed to bring about. In some cases many different agencies are involved in regulating or conferring benefits on enterprises. The greater the number of people involved in a decision, the longer it is likely to take (sometimes well over 6 months), particularly in the cases where the decision must be made by one or more ministers. In Cameroon, for example, as many as 60 copies of an application for investment code privileges are required, and 4 ministers are involved in the decision. Togo requires 35 copies, and Benin, Congo and Gabon, 20. If nothing else, preparation of so many copies of the application is a burden. Furthermore, the more agencies involved, the greater the chance of contradictory or confusing signals and rules. Some countries have maximum statutory periods for a decision, but these are not always observed in practice. Administration of industrial incentives and regulations in Nigeria has caused problems through frequent reviews of applications (including annual renewals), heavy verification and monitoring needs, lack of ready access to the relevant legislation and regulations, inconsistencies in the laws and regulations, and overlapping jurisdictions among agencies which may result in conflicting advice or contradictory decisions. The Government has taken two important steps to ameliorate the situation. A White Paper published in 1980 outlines government industrial policy and the principal incentive programs and legislation; this will - 27 - make the information easily accessible to potential investors. Then in 1981 the Industrial Development Coordination Committee was created within the Federal Ministry of Industries to coordinate and expedite the processing of applications for incentives and other required permits and approvals for new investment. This type of one-stop service could improve the ease of investment in other countries as well. The degree to which the decision on conferring benefits is discretionary is very important. In Africa, the French legal influence has led many countries to enumerate the industries that will be approved for investment; the Togolese investment code is an example of this, and it also specifies clearly the size of investment required for each regime. In other countries, the decision is left up to the agency implementing the law; the approval of pioneer industry or approved user status in Nigeria is such a case. Both methods have their advantages and disadvantages. The discretionary method allows the government to limit the number, and thus the cost, of the concessions awarded, but its success depends strongly on the quality and integrity of the people administering it, and it is costly in terms of trained staff, often a relatively scarce commodity. It also adds to the uncertainty facing the potential investor, who may not know what criteria will be applied to his application. Automatic qualification of all firms meeting clearly specified requirements is much easier to administer, but may be more costly in terms of revenue foregone, since the number of qualifying firms is likely to be higher. Many investment codes list the information required from firms applying for benefits. The requested information is often voluminous and is not always related in any obvious way to the awarding of benefits. Furthermore, it often omits what should be the key variable - the economic profitability of the firm or the information necessary to calculate it. In Togo, for example, investors are asked to submit their projected costs and profitability both with and without the desired regime. This is certainly a step in the right direction, since it demonstrates the extent to which the firm could be profitable without the concessions. However, it does not reveal the profitability of the project to the economy, since the prices used still reflect distortions caused by the general policy - 28 - environment (e.g. minimum wages, tariffs). Furthermore, neither this nor other requested information (e.g. the firm's training program) appears to be used as a criterion for granting concessions or distinguishing among regimes. Effective follow-up and sanctions are necessary to evaluate a project's impact on the economy and to enforce the conditions imposed on firms. In Togo the state may withdraw the agreement if the enterprise does not conform to the information furnished to justify its application or fulfill obligations agreed to, e.g. in realizing its investment program, providing training, supplying domestic markets, or reinvesting. However, monitoring of enterprises may be time consuming, and the greater the number of conditions, the more difficult the monitoring process. In some cases, the necessary information may be difficult to obtain or verify independently of the enterprise involved. A study of tax exemption in Puerto Rico concluded that the problems of administering the program were so substantial that they would eventually make it "unenforceable to a marked degree, even with the highest level of efficiency and diligence in administration" (Taylor, 1957, p.86). The types of sanctions available are also important, since complete withdrawal of all benefits may be too drastic; most codes do not appear to allow for modified sanctions. V. IMPACT OF INVESTMENT INCENTIVES In order to judge which incentives are most effective it would be necessary to measure the individual impact of each incentive on the level of investment, economic growth, and the structure, characteristics and size distribution of industry. Unfortunately, neither the costs nor the benefits of the individual measures or even of the system as a whole are easily measured. One major measurement problem is that of redundancy, or how to judge whether an investment (or an equally productive alternative) would have been made even without any incentives, making the revenue foregone an unnecessary cost. In any case foregone income tax revenue may be very difficult to estimate, since reporting (e.g. tax returns) is often inadequate. A second cost, also difficult to measure, is the loss of efficiency in the economy from incentive-induced distortions in resource allocation. - 29 - Much of the work on the impact of investment incentives has concentrated on the effects of trade and exchange rate policies, and in fact the impact of those variables may well outweigh that of all other incentives combined. This section presents some of the findings of studies that shed light on the question of the effectiveness and economic impact of non-trade incentives. The evidence is of two types: analyses of the impact of the incentives and induced investment on individual economies and surveys of firms to determine which factors are the most influential in foreign investment decisions. Part A will summarize some of the evidence on impact in a number of countries and Part B will describe the results of some surveys. All of the evidence appears to support the conclusion that while tariff protection may be an important investment incentive in some cases, other incentives are not a significant factor in investment decisions. A. Impact on the Economy 1/ 1. Regional Comparisons: Sub-Saharan Africa A recent paper by Balassa (1982 A) analyzes the policies adopted to adjust to the external shocks of the 1973-78 period, investigates the relationship between these policies and the rate of economic growth, and examines changes in economic growth rates, incremental capital-output ratios, and savings rates in 19 low and middle income sub-Saharan African countries. Balassa distinguishes between market-oriented and interventionist countries, where the former place greater reliance on the price mechanism and market forces than do the latter in areas such as relative incentives to manufacturing and primary activities, the determination of exchange rates, the extent of price and wage controls, the setting of interest rates, the degree of credit rationing, the licensing of private investment, the choice of public projects and their share in total investment, and the size of the public budget deficit. Private-market oriented economies include Botswana, Cameroon, Ivory Coast, and Mauritius, while the economies exhibiting the greatest degree of 1/ Additional evidence is discussed in Lent, 1967, pp. 291 - 308; and in Shah and Toye, "Fiscal Incentives for Firms in Some Developing Countries: Survey and Critique," Chapter 12 in Toye, 1978. - 30 - public intervention and state ownership include Benin, Ethiopia, Ghana, Madagascar, Mali, Tanzania, and Zambia. An intermediate group consists of Kenya, Malawi, Niger, Senegal, Sudan, Togo and Upper Volta. While the economies studied and the political environments differ markedly, even within groups, and although the period of comparison is short, the results are nonetheless interesting. During the 1973-79 period, the first group experienced average annual GNP growth rates of 6.8%, while the intermediate group averaged 3.5%, and the interventionist group, only 1.5%. These differences are explained in terms of differences in incremental capital output ratios (ICORs, measuring the efficiency of use of resources 1/) and in savings ratios. In the same period, ICORs were lower in the market-oriented countries, and domestic savings rates higher than in the other groups. Differences in ICORs and domestic savings ratios reflect the policies followed to adjust to external shocks. Interventionist countries applied protection, export taxes and price controls, and maintained large margins in the parastatal trading companies, thus discriminating against exports and agriculture. Their real exchange rates appreciated, discriminating against both exports and import substitution. The licensing of private investment and implementation of costly capital- intensive public projects further reduced the efficiency of resource allocation. Market-oriented countries, on the other hand, showed greater openness, greater price flexibility, more realistic exchange rates, freer choice of private investment, and fewer public projects. Public savings were higher (deficits were lower) in market-oriented countries, and higher private savings reflected real interest rates that, while negative, were less so than in the interventionist countries. 2. Ghana Another study (Ingram and Pearson, 1981) examines in a limited way the impact of investment incentives in Ghana, which grants income tax holidays, exemptions from import duties, accelerated depreciation, 1/ The higher the ICOR, the more capital is required to generate the same output, i.e. the lower the efficiency of its use. - 31 - remissions of export duties or excise taxes, refunds of sales taxes, exemptions from registration and stamp fees, reductions in property taxes, income tax deductions for expenditures on research, and special privileges for repatriation of capital and transfer abroad of profits. In addition, during the period under consideration, government trade and price policies (import tariffs, quantitative restrictions and administered prices) maintained high prices for manufactured goods, and subsidized credit was available. A few policies, such as minimum wages and taxes on material inputs, acted as disincentives. The study estimates the effect of the various interventions on the profitability in a single year (1972) of seven firms benefitting from concessions. 1/ It finds that only two of the seven would have been privately profitable without the concessions, whereas the incentives allowed all but two to generate positive private profits. Four of the firms were enabled to earn excess profits, with actual rates of return ranging from 23% to nearly 600%. On the other hand, six of the seven firms generated large economic losses (for the country), totalling nearly 60% of their gross output in economic prices. Protection of output provided the greatest incentive, contributing about 20% more than the investment concessions. Access to subsidized credit was also important. Ghana's investment concessions sought the following objectives: to enhance the efficient use of the country's resources, to earn foreign exchange, to promote interindustry linkages, to employ and train Ghanaians, and to promote investment in rural areas. The first two objectives were clearly not met by the firms in the sample, since only one was economically profitable, and, taken as a whole, they operated inefficiently and lost foreign exchange. As for the other objectives, there was little evidence of linkages between the seven firms and other Ghanaian firms (most of the seven imported their inputs), the firms were 1/ These seven firms - in food products, textiles, footwear, nonmetallic mineral products and fabricated metal products - constituted less than 2% of the 398 large-scale (i.e. with at least 30 employees) manufacturing firms in Ghana in 1972, but produced 7.3% of the total ouput of large-scale firms. In terms of equity, they made up about 20% of the investment receiving concessions during the period 1966-71. - 32 - mostly capital intensive, and only one of them was located in a rural area. Thus the concessions, while presumably costly to the Government, did not meet any of their objectives. 3. Sudan Sudan's incentive system is typical of many African countries. It includes quantitative restrictions and tariffs on imports, tax holidays, carry over of all losses during the exemption periods, exemption from import duties on machinery, equipment, spare parts and raw materials, exemption from excise taxes, and subsidized land, freight and electricity. All industrial activities qualify, and no systematic economic profitability analysis is carried out. An analysis of this incentive structure found evidence that it provides "considerably greater inducement for allocation of resources to industry than to agriculture. So these policies work against the efficient exploitation of Sudan's oft- diagnosed long term comparative advantage in agriculture" (Acharya, 1979, pp. 66-7). 4. The Philippines 1/ Postwar trade policies in the Philippines focussed on import substitution as an industrialization strategy. Import restrictions, overvaluation of the currency and later a highly protective tariff system led to a manufacturing sector biased toward import substitution in consumer goods and heavily dependent on imported materials and capital equipment. Despite a labor surplus in the economy and a relatively rapid growth of manufacturing output, the share of manufacturing in the labor force stayed constant throughout the 1950's and 1960's. Because of their dependence on imported inputs, factories tended to locate near Manila, the principal port, which was also the only place where tax and credit favors were obtainable. There is evidence that the incentive system did not provide an allocative mechanism rewarding efficient industries and penalizing the inefficient; in fact in many instances the reverse was the case. 1/ The experience of the Philippines summarized here is described in detail in Bautista, Power & Associates, 1979. See especially pp. 32-3, 62. - 33 - When opportunities for import-substituting consumer goods industries fell in the late 1950's, manufacturing growth rates also fell. The Investment Incentives Act of 1967 was enacted to promote an accelerated pace of industrialization. It offered, for eligible enterprises, accelerated depreciation, net operating loss carryover, duty exemption on imported capital equipment, tax credit on domestic capital equipment and for withholding tax on interest, deduction for expansion reinvestment and half of training costs, and low-interest government loans. In addition, "pioneer" enterprises were exempted from all internal revenue taxes but the income tax and could receive tariff protection. Exporters were given additional concessions: tax credit for duties paid on imported materials and supplies, double deduction of shipping costs and promotional expenses for exports, export tax exemptions, and deduction from taxable income of direct labor and local raw material costs used in export production (up to a limit). Export processing zones were established, and various steps taken to promote exports abroad and simplify export procedures. Floating of the domestic currency in 1970 also benefitted exporters. For a number of reasons, industrial growth in the early 1970's, while accelerating, was not remarkable. The incentives apparently had little impact; in 1970, only 131 firms were receiving benefits under the Investment and Export Incentives Acts. Nontraditional manufactured exports did grow rapidly, but this was at least in part because of the de facto devaluation of 1970. An examination of the fiscal incentives granted by the government indicates significant potential effects on the rate of return, user cost of capital, and employment in the enterprises concerned. Assuming a typical rate of return without incentives of 10% and an asset life of 10 years, the rate of return of a firm registered under the Investment Incentives Act could gain an estimated 2 percentage points from accelerated depreciation, 4 percentage points from the tax exemption on imported capital equipment, and, for exporting firms, an additional 8 percentage points from the deduction from taxable income of labor and raw material costs, for a total rate of return of 24%. Pioneer firms gain 8 percentage points from the expansion reinvestment allowance, while others - 34 - get a 3 percentage point increase. For new projects of firms under the Export Incentives Act, the tax exemption of imported capital equipment and tax deduction of labor and local raw material costs results in a 14 percentage point increase in the rate of return. 1/ Despite these striking results, further analysis indicates that when combined with estimates of the impact of effective tariff protection, the other incentives made very little difference to the biases already present in the system, i.e. in favor of consumer goods and against exports. Whatever the above results indicate about the impact of the incentives on the level of investment, they do not shed light on the impact on relative factor intensities. Calculations based on the assumption that firms seek to maximize the present value of net revenue after taxes indicate that for a new project with a 20-year lifespan, the Investment Incentives Act leads to a reduction in the user cost of capital by 14% due to accelerated depreciation, by 15% due to the tax exemption for imported capital equipment (18% due to the tax credit on domestically produced equipment), and by 10% due to the tax credit for withholding tax on interest on foreign loans, for a total reduction of 39-42%. In addition, the expansion reinvestment allowance implies a reduction of 30% for expansion projects for pioneer firms and 15% for non-pioneer firms. For exporting firms, the user cost of capital is reduced by 15% for new projects, and 30-45% for expansion projects (assuming imported capital equipment). The cost of capital is further reduced by 9-35% from subsidized credit to priority sectors. In contrast, labor costs are reduced by only 4% from the deduction of training expenses and, for exporting firms, by 18% from the tax deduction of labor and local raw material cost. These results show that the incentive system, despite its attempts to encourage employment, subsidizes capital to a greater extent than labor, presumably inducing a more capital-intensive type of investment than would otherwise be the case. 1/ Different assumptions about asset life reduce the magnitude of the effects but still indicate a substantial benefit. - 35 - 5. Thailand In Thailand concessions include exemptions and reductions of import duties on imported machinery and materials and of business taxes on domestic machinery and materials, tax holidays and carry-forward of losses beyond the tax holiday period. As we have seen, these incentives tend to favor capital-intensive processes and imports. Capital-intensive activities and large size are encouraged further by minimum investment requirements for concessions. The data show that in fact promoted firms have an investment cost per job more than five times that of all new investment, a result that conflicts with the government's stated objective of providing increasing industrial employment (World Bank, 1980). During the 1970's, the government introduced a number of measures to promote manufactured exports. However, exemptions from import duties and various taxes on imported materials do not fully neutralize the burden of these taxes, in part because of the cumbersome requirements and procedures and the difficulties of calculating the amounts involved. These incentives have apparently had little impact; the share of exports in total production of firms awarded incentives is no higher than the average for all industrial firms. 6. Colombia Another study (Bilsborrow and Porter, 1972) examines the effects of Colombia's 1960 tax reform offering exemption from income taxes for up to ten years to firms in basic industry. A total of 100 firms were studied, out of 12,000 manufacturing establishments, representing 10% of total manufacturing equity. The average before tax profit rate for promoted firms was 21.1%, compared to 11.7% after taxes, so the tax exemption was clearly of some importance. However, the average before tax profit for all of manufacturing was 20.1%, implying that the tax exemption may have been redundant, i.e. that the firms being promoted would have invested in any case, and that projects with initially low profits due to externalities or market imperfections were apparently not encouraged by the tax exemption. The cost of the exemption to the government in 1965 and 1966 was more than 6% of total corporation and excess profits tax receipts. - 36 - 7. Mexico A study of Mexico in the 1950's (Ross and Christensen, 1959) found tax exemption to have been only a minor factor in stimulating industrial investment, particularly when compared to tariff protection (which made possible the profits to which the tax exemption applied). Of the 150 companies that were denied exemption for a total of 160 products during 1951-55, all of them proceeded to produce the products anyway. Ross and Christensen concluded (pp. 101-2) on the basis of their work in Mexico: "There appears to be no instance in the recent past when tax exemption was the decisive factor in an investment decision in Mexico; in fact, there is probably no situation in which it was even a decisive factor." Furthermore, an estimate of the contribution of investment by the exempt sectors to the country's per capita income and to employment over sixteen years (1940-1955) indicates that the direct contribution was small (Heller and Kauffman, 1963, pp. 116-19). 8. Puerto Rico Puerto Rico initiated its tax holiday program in 1949 and is often cited as the most successful example of tax incentives to attract industry. The program gives a full exemption from income, property and municipal taxes for a minimum period of 10 years. The start of the holiday can be delayed for up to two years to cover periods with losses. Most of the firms studied (Bond, 1981) were subsidiaries of U.S. firms, and special treatment of Puerto Rico in U.S. tax laws made them exempt from U.S. taxes as well during the holiday period. This study looked at the effect of the finite limit to the holiday on the behavior of firms in the brassieres, corsets and allied garments industry, which is the largest Puerto Rican industry in terms of employment, was established after the tax holiday program started, and, being low wage and labor intensive, has relatively low entry and exit costs. Empirical results from a sample of 152 firms show a peak in the number of closings at 12 years of age, which is consistent with the idea that firms close once their tax holiday expires. Further analysis strongly supports the hypothesis that the tax holiday results in a rapid turnover of firms, with newly exempt ones driving out those whose holidays have expired. While this result does not completely negate the usefulness - 37 - of the holiday, since while they were operating, the firms created both employment and foreign exchange, it does imply a distortion due to increased turnover of firms and, perhaps, reluctance of firms to invest in training. B. Impact on Investors Survey data, while subjective, are more plentiful than actual measures of the costs and benefits of incentives. Although some of the surveys are quite dated, there is no reason to believe that their results are not still valid today. The surveys question investors (generally foreign investors) as to the most important factors influencing their choice of location. A survey was carried out in 1961 of 205 companies covering 365 investments made in 67 countries around the world and of 20 governments, each of which evaluated the incentives offered foreign investors (Robinson, 1961). The differences between governments and investors as to which factors are believed to be important is striking. The governments rated the five most important incentives offered to foreign investors as tax relief, equality of treatment with domestic enterprises, progressive domestic climate, favorable terms for transfer of profits and repatriation of capital, and government-sponsored credit institutions. The investors, on the other hand, while agreeing that equality of treatment and transfer of profits and repatriation of capital were important, considered the other three most important government policies to be establishment of and firm adherence to a national development program (as a measure of the government's dedication to economic development), nondiscrimination against foreign ownership and control and freedom from detailed or burdensome regulations on organization, ownership and management. The most important non-policy factor mentioned was the opportunity to earn a profit and, given this, the desire to expand or maintain sales by entering a new market or preserving an established one in the face of tariff or exchange barriers. The factors most often mentioned with respect to the host country were not government policy, but political stability, followed by a favorable government attitude toward private enterprise and the profit motive and economic and financial stability. - 38 - In another survey of 247 American corporations with foreign investment, 57% listed currency inconvertibility as a factor discouraging investment, 39% mentioned instability of the host country, 38% national discrimination and 26% a limited market or source of supply. Only 10% expressed any interest in favorable foreign taxes as a condition of investment (Barlow and Wender, 1955 quoted in Lent, 1967, pp. 309-10). In a small survey in Mexico, Ross and Christensen asked 24 firms whether they would have started business without the availability of the tax exemption; 14 answered yes, 9 probably yes, and only 1 probably no. Investors interviewed stressed the importance of the potential for profit, the free convertibility of pesos for dollars, and the government's general attitude toward private investment. Tariff policy was universally mentioned as the most important government policy (Ross and Christensen, 1959, pp. 101-3). Interviews of 45 firms in Jamaica asking which factors influenced the decision to begin operations in Jamaica found that two factors predominated: inertia (for Jamaicans wishing to invest at home) and the assurance of a guaranteed market by protective trade policy (Chen-Young, 1967, p. 298). In a survey in Costa Rica, most firms mentioned tax relief as a factor in the investment decision, but two-thirds said they would have invested anayway (Lent, 1967, pp. 302-3). In Puerto Rico, tax exemption was claimed to have been the most important factor in decisions by American investors (Taylor, 1957, pp. 122-4), 1/ but Puerto Rico is a special case, with lower than usual risks as part of the U.S. judicial system, with assured export markets and a tax sparing agreement. Other important factors mentioned were wage rates, efficiency of machinery and equipment, attitude of the community, present plant site, and adequacy and dependability of electric power, water supply and transportation facilities. Market size and potential, i.e. the opportunity to earn profits, is a recurring theme. Multinationals questioned in the late 1970 's stressed this factor and played down the importance of special incentives. The latter are believed to increase a firm's visibility and vulnerability and are considered to be too volatile or transitory. Tax concessions are often seen as illusory, with holidays given in the early 1/ The survey covered 44 firms. - 39 - years when profits are the least likely, and their value greatly dependent on the tax system in the investor's home country (Frank, 1980, pp. 49-50, 95-96). 1/ After reviewing the evidence, Lent concluded that "in general, developing countries need not be concerned about matching the tax benefits of other countries in order to attract foreign capital for new industries. Tax considerations typically play a role subordinate to more basic economic factors in the location of industry. Competitive bidding among countries by the offer of more and more generous tax concessions tends to reduce revenues from foreign investment without increasing the total flow of capital..." While tax incentives are relatively more important for export industries which do not benefit from protection, "the primary factors determining industrial location are the opportunities for profits based on comparative costs (such as those for labor, transportation and power), availability of raw materials, and size of markets, as well as the political and economic risks..." Tax costs could affect the location decision if all the other factors were comparable among the alternative countries (Lent, 1967, pp. 310, 315). Interviews with over 30 multinational companies covering 74 projects in four industries - automobile, computer, food processing and petrochemical - found that incentives did influence the location decision in two-thirds of the cases studied to the extent that firms would not have located in a particular country if it had eliminated its incentives while other countries retained theirs (Guisinger, 1983). In one-third of the cases, commodity protection - through tariffs or government procurement policies, for example - was the key incentive. Incentives other than protection were not at all important in the location of domestically oriented production, but were more important for the location of firms within common markets (where their impact is, however, reduced by the strong regional similarities of incentives offered by competing countries.) Within countries incentives were sometimes found to have had some locational impact, but here too decisions were often based more on 1/ This survey covered 90 multinational companies with 402 subsidiaries. - 40 - the location of inputs or markets. Many companies reported that incentives were frequently not even considered in feasibility studies and simply made already attractive investments more attractive; in this case they were unnecessary transfers from the host country's taxpayers to the corporate investors. Investment decisions were made in the first instance on the basis of economic or strategic considerations, often long-term, concerning inputs, production costs and markets. In several industries, performance requirements had a fairly strong negative impact on locational decisions. This was particularly true for computers, where corporations were unwilling to share ownership and technologies, but it was also true to some extent of local content laws, employment or export targets, or limitations on transfer of funds. Other important disincentives were price controls, profit ceilings and bureaucratic red tape. In the food industry the negative effect of these factors apparently outweighed any positive impact of the incentives. VI. SUMMARY AND CONCLUSIONS: SOME PRINCIPLES FOR INCENTIVE SYSTEMS Developing countries have used a wide variety of incentives to attract industrial investment. One of the most important incentives is protection from competition, or guaranteed markets, generally accomplished with tariffs and quantitative restrictions that prohibit or limit competing imports. Trade policy is not covered in this paper, but it must be noted that its influence on both the level and the structure of investment is probably greater than all other economic incentives combined. While protection of infant industries may be justified for a limited period, protection can perpetuate industries whose true cost to the country is greater than their benefits. It also permits costs to rise and reduces the level of efficiency. Administered prices, which have been adopted in many countries to restore the control over prices that would normally be exerted by competition, lead to further inefficiencies and distortions in the structure of production. A number of incentives have been adopted over the years in an effort to compensate for the inefficiencies and distortions stemming from - 41 - shortcomings in various policies (such as those dealing with trade and exchange rates) and institutions. Although the best way to deal with the distortions would be to remove them, this might not always be possible, and an argument can be made for using incentives to compensate, for example, for high labor costs resulting from minimum wages and social legislation. In addition, incentives can be used to increase the rate of return to investors in cases where some of the returns from the investment would flow to the country as a whole and to compensate for market failures (e.g. to support infant industries). However, the incentives commonly used in many countries, and the criteria by which they are awarded, often lead to more, rather than fewer, distortions. The criteria used by different countries to award incentives fall into four categories (Usher, 1977, pp.128-30). The broadest possible method is to grant concessions to all new investment. This method might increase total investment in the country, since it would raise the rate of return to all projects, pushing some marginal ones over the minimum required without discouraging others. The drawback to this method is its high cost to the government, unless taxes raised from the induced investment after the concessions end make up the difference. Experience indicates, however, that the typical incentive packages offered to investors have had little or no effect on the level of investment, implying that the lost tax revenues are in fact a real cost to the economy. A second selection method is to restrict the concessions to a limited number of industries believed to be of particular importance to the country. However, promotion of particular sectors or industries implies the relative discouragement of others (usually agriculture). In making such a choice, a government implicitly assumes that it can identify the areas of a country's comparative advantage. This is not an easy task, particularly in an economy where previous interventions have so distorted prices that they cannot be used as an indication of value, and there is a risk that areas with significant potential may be omitted from the list to be promoted. Even in a country like Japan there is no agreement on the extent to which the government has been successful in picking winners. - 42 - The third method used to award investment incentives is that of the committee with only vague guidelines, such as the "good of the country," or consistency with the national development plan. This method might be the least costly in terms of foregone revenue, in that it might be able to restrict concessions to those investments that would not be made otherwise, but the information and skill requirements for such a committee to succeed, and the possibilities for abuse, would be enormous. The fourth method is to screen projects by certain characteristics, such as size, job creation, increased domestic value added, or improvement in the balance of payments. In fact, these criteria really all represent the desire for economic growth in a way that uses the country's resources (including human resources) as efficiently as possible. Many incentive packages, however, fail to list efficiency as a criterion for awarding concessions. Even in promoting non-economic, i.e. social or political, goals, it is still in the country's interest to achieve these goals as efficiently as possible. The easiest way to identify economically efficient investments would be to use the economic rate of return as the main criterion for selecting projects that are to receive special incentives. Entrepreneurs base their decisions on the financial rate of return to their investment, but the Government is more interested in the project's economic rate of return, or its impact on the economy. The economic rate of return might differ markedly from the financial rate of return due, for example, to minimum wages, training, administered prices, or tariffs, and incentives could be used to bring the signals to investors more closely in line with the interests of the country, i.e. to reduce the disparity between the financial and economic rates of return. Most other criteria are simply a way of guessing what the economic rate of return of an investment would be. Thus the employment criterion implicitly tries to compensate firms for wages that are higher than their opportunity cost (although the instruments used often subsidize capital rather than labor). Instead, shadow wages could be used to calculate the rate of return, and appropriate compensation awarded. The size criterion assumes that the largest investments will have the greatest positive impact on the economy, but this will not be the case if they do not use resources efficiently. - 43 - Small enterprises, often ignored by the incentive system, are probably the most efficient resource users in many developing countries. The value added criterion implicitly assumes that domestic value added is the best way to measure economic growth, but it is rarely specified that value added should be calculated at world prices. If domestic prices are used, a high value added might simply reflect high levels of effective protection and could mask a low or even negative economic value to the country. Concessions might in this case be awarded to the very firms that already have the greatest incentive and that are least suited to the country's resource endowment. Use of an economic rate of return criterion would reveal such cases, and it would also make explicit the economic trade-offs involved when using other criteria, including social or political ones, for project selection. Given the decision to promote investment, the choice of instruments is critically important to a country's welfare. In many countries the concessions offered to investors have encouraged a type of investment that is ill suited to the country's comparative advantage and sometimes inconsistent with the government's objectives. In countries endowed with abundant unskilled labor and scarce capital resources, investment codes often offer exemptions from import duties on capital inputs, accelerated depreciation and other concessions that lower the price of capital relative to that of labor. They favor large projects, relatively intensive in the use of capital and imported inputs, and thus in relative terms discourage the use of labor and the production of intermediate goods. On the labor side, interventions by the government, in the form of minimum wages or labor legislation, increase, rather than reduce, labor costs. Many of the industries that have evolved in response to these distorted price signals can survive only as long as the concessions continue. Governments thus find themselves supporting on a permanent basis industries that are economically inefficient and that may even represent a net drain on the economy. More broadly, the policies that encourage investment in industry often do so at the expense of investment in agriculture, creating further distortions in the structure of production. This is a high price to pay for industrialization in countries where the very scarcity of resources argues forcefully for their use in the most productive manner possible. - 44 - A second cost of incentive packages is the revenue foregone by the government when it awards tariff and tax reductions, exemptions and holidays. This cost is not always recognized, since it does not appear in the budget, and it is difficult to measure, since to the extent that the incentives induce investment that would otherwise not have taken place, there is no loss. However, it is nonetheless important, since it represents government expenditures foregone. If the government's objective is to stimulate investment in general, this can best be done not through the investment codes typically found today, but by broad fiscal, monetary and exchange rate policies that will improve the climate for savings and investment, while remaining neutral among sectors or types of investment (e.g. industry vs. agriculture, export vs. import-replacing, intermediate vs. final goods, capital vs. labor intensive). Failing broad policy improvements, investment codes should offer neutral incentives, such as value added subsidies or tax credits. More specialized incentives can be geared to overcoming existing distortions or market failures or to account for externalities, i.e. to ensure that economically viable projects have financial rates of return high enough to attract investors. Creation of new distortions as side effects of the incentives can be minimized by targeting the incentives as directly as possible. Thus, measures to promote employment should do so directly (e.g. by a training subsidy or tax credit for employment to correct for wages above the market level), not by promoting the use of capital in labor-intensive industries (e.g. by allowing accelerated depreciation or duty-free equipment imports as a function of the number of jobs created). At least one African country is revising its investment code to eliminate the most distortionary incentives, such as import duty exemptions for intermediate inputs, and substitute a value added subsidy based on the wage bill for local labor. Similarly, incentives for investment in less developed areas could include provision of infrastructure or training subsidies, which attack the constraints directly, rather than duty free imports, which may actually encourage location near a port. Incentive systems should be as simple as possible. Most investment codes allow some discretion, both in the approval of - 45 - applications and in the size of the benefits awarded. The lack of automaticity implies a good deal of time spent in decision-making and may cause confusion, rather than interest, among investors. Furthermore, a substantial amount of monitoring would be required to ensure that enterprises fulfill the conditions imposed on them. The amount of trained manpower necessary to carry out these tasks adequately is a severe burden in countries where this type of expertise is scarce and could be used far more productively in other capacities. One single concession, whose value to the investor could be easily calculated, might prove the most effective means of attracting investment. A subsidy or tax credit based on criteria that reflect the government's objectives would serve this purpose. The final and most important point to bear in mind is that perhaps the main determinants of investment are those beyond the direct and immediate control of the government. These include political and economic stability, adequate markets, and availability of inputs. Other important incentives, over which the government has greater control, are non-discriminatory treatment for private, including foreign, investment, favorable terms for the repatriation of profits, and freedom from burdensome controls. These factors depend upon broad economic and legislative policies, not on special inducements offered to selected investors. They establish the basic investment climate, and without them all other incentives are likely to prove fruitless. - 46 - REFERENCES Acharya, Shankar N. "Incentives for Resource Allocation: A Case Study of Sudan." World Bank Staff Working Paper No. 367, December 1979. Balassa, Bela. "Adjustment Policies and Development Strategies in Sub-Saharan Africa, 1973-78." World Bank Development Research Dept. Discussion Paper, Report No. DRD41, November 1982 (A). Balassa, Bela. The Newly Industrializing Countries in the World Economy. N.Y.: Pergammon, 1982 (B). Balassa, Bela, et. al. Development Strategies in Semi-Industrial Economies. Baltimore: Johns Hopkins Press, 1982 (B). Barlow, E.R. & Ira T. Wender. Foreign Investment and Taxation. Harvard Law School, International Program in Taxation. Englewood Cliffs, N.J.: Prentice Hall, 1955. Bautista, Romeo M., John Power & Associates. Industrial Promotion Policies in the Philippines. Manila: Philippine Institute for Development Studies, 1979. Bilsborrow, Richard E. & Richard C. Porter. "The Effects of Tax Exemption on Investment by Industrial Firms in Colombia." Weltwirtschaftliches Archiv, Vol. 108, 1972. Bond, Eric. "Tax Holidays and Industry Behaviour." Review of Economics and Statistics, Vol. LXIII, No. 1, February 1981, pp.88-95. Chen-Young, Paul L. "A Study of Tax Incentives in Jamaica." National Tax Journal, Vol. XX, No. 3, 1967, pp. 292-306. Cody, John, Helen Hughes & David Wall, eds. Policies for Industrial Programs in Developing Countries. N.Y.: Oxford University Press, 1980. Diamond, Walter H. & Dorothy B. Diamond. "Capital Formation and Investment Incentives Around the World." 1983. Frank, Isaiah. Foreign Enterprise in Developing Countries. Baltimore: Johns Hopkins Press, 1980. Guisinger, Stephen. "Investment Incentives and Performance Requirements: A Comparative Analysis of Country Foreign Investment Strategies." July 1983, draft. Hellawell, Robert, ed. United States Taxation and Developing Countries. N.Y.: Columbia University Press, 1980. - 47 - Heller, Jack & Kenneth M. Kauffman. Tax Incentives for Industry in Less Developed Countries. Cambridge, Mass: Harvard University Law School, 1963. Ingram, William D. & Scott R. Pearson. "The Impact of Investment Concessions on the Profitability of Selected Firms in Ghana." Economic Development and Cultural Change, Vol. 29, No. 4, July 1981, pp. 831-9. International Labor Organization. Fiscal Measures for Employment Promotion in Developing Countries. Geneva: 1972. International Monetary Fund. Taxation in Sub-Saharan Africa. Washington, D.C.: 1981. Krueger, Anne 0. Foreign Trade Regimes and Economic Development: Liberalization Attempts and Consequences. N.Y.: Ballinger, 1978. Krueger, Anne 0. Trade and Employment in Developing Countries. Chicago and London: NBER, University of Chicago Press, 1983. Lent, George E. "Tax Incentives for Investment in Developing Countries." IMF Staff Papers, Vol. 14, July 1967, pp. 249-321. Robinson, Harry J. The Motivation and Flow of Private Foreign Investment. Menlo Park, Calif.: Stanford Research Institute, International Development Center, Investment Series No. 4, 1961. Ross, Stanford G. & John B. Christensen. Tax Incentives for Industry in Mexico. Cambridge, Mass: Harvard University Law School, 1959. Taylor, Milton C. Industrial Tax Exemption in Puerto Rico. Madison: University of Wisconsin Press, 1957. Toye, J.F.J., ed. Taxation and Economic Development. London: Frank Cass, 1978. Usher, Dan. "The Economics of Tax Incentives to Encourage Investment in Less Developed Countries." Journal of Development Economics. Vol. 4, No. 2, June 1977, pp. 119-40. World Bank. Thailand: Industrial Development Strategy in Thailand. Washington, D.C.: World Bank, 1980. World Bank The Construction Industry: Employment and Development Issues and Strategies in of Small Enterprises Publications Developing Countries David L. Gordon, coordinating of Related Ernesto E. Henriod, coordinating author author Examines the potential role of the Interest Presents a profile of the construction World Bank in encouraging developing industry. Points out that construction countries to assist small enterprises work represents 3 to 8 percent of the and suggests that efficient substitution gross domestic product of developing of labor for capital is possible in a countries. Fostering a domestic capa- broad spectrum of small-scale manu- bility in construction, therefore, is im- facturing and other activities that are A Brief Review of the World portant. Discusses problems and con- able to absorb a rapidly growing labor Lube Oils Industry straints of the industry and formulates force. A. Ceyhan, H. Kohli, L. strategies for future actions. Draws Sector Policy Paper. 1978. 93 pages (in- Wijetilleke, and B.R. Choudhury heavily from the experience of the cluding 3 annexes). World Bank in supporting domestic Stock Nos. BK 9060 (English), BK 9061 This report assesses the structure, construction industries over the past (French), BK 9062 (Spanish). $5. background, and outlook for the world ten years. Useful to contractors, engi- lube oils industry. Presents the histori- neers, and administrators in construc- Estmti Total Fat cal and projected lube oils demand tion industry. Esmatng o actor and trends in manufacturing technolo- 1984 120 Productivity Growth in a gies and production capacity and pro- Developing Country vides an indicative assessment of the ISBN 0-8213-0268-X.Stock No. BK 0268. Anne 0. Krueger and Baran economics of lube oil production with $5. Tuncer detailed market and economic data. Energaindustrkesa ReponorSies N a. Cost-Benefit Evaluation of Staff Working Paper No. 422. 1980. 64 Energy Industries Report Series No. 1. LDC Industrial Sectors Which pages (including references, appendix). 1982. 48 pages (including 13 annexes, ret- H Foi Ownership Stock No. WP 0422. $3. ISBN 0-8213-0054-7. Stock No. BK 0054 Garry G. Pursell Financing Small-Scale Industry $3. Staff Working Paper No. 465. 1981. 45 and Agriculture in Developing pages. Countries: The Merits and Capital Utilization in Stock No. WP 0465. $3. Limitations of "Commercial" Manufacturing: Colombia, Policies Israel, Malaysia, and the Development Finance Dennis Anderson and Farida Philippines Companies Khambata Romeo M. Bautista, Helen Examines the role of development fi- Staff Working Paper No. 519. 1982. 41 Hughes, David Lim, David nance companies as major mechanisms pages (including references). Morawetz, and Francisco E. for assisting medium-scale productive ISBN-0-8213-0007-5. Stock No. WP 0519. Thoumi industries, assesses their potential for 3 aiding small enterprises in meefing so- 3 The authors surveyed 1,200 manufac- turigfirns in four developing coun- coeconomic objectives of developing Fostering the Capital-Goods tries to establish actual levels of capital countries, and discusses the evolution Sector in LDCs: A Survey of utilization. The information coflected of World Bank assistance to them. Evidence and Requirements was the first and remains the only Sector Policy Paper. 1976. 68 pages (in- Howard Pack data base available for the study of cluding 7 annexes). Staff Working Paper No. 376. 1980. 64 capital utilization. It was found that Stock Nos. BK 9040 (English), BK 9058 pages (including references). capital utilization is not as low as had (French), BK 9041 (Spanish). $5. been supposed. The study is con- Stock No. WP 0376. S3. cerned with factors that cause differ- Empirical Justification for Incorporating Uncertainty into ences in levels of capital utilization Infant Industry Protection . Planning of Industrialization and the policies that might be used to Larry E. Westphal Strategies for Developing increase it. Oxford University Press, 1982. 288 pages Staff Working Paper No. 445. 1981. 38 Countries (including bibliography, index). pages (including references). Alexander H. Sarris and Irma LC 81-9526. ISBN 0-19-520268-6, Stock Stock No. WP 0445. $3. Adelman No. OX 520268. $22 hardcover. Staff Working Paper No. 503. 1982. 58 pages (including appendix, references). Stock No. WP-0503. $3. Industrialization and Growth: Macroeconomic Implications of Managing State-Owned The Experience of Large Factor Substitution in Enterprises Countries Industrial Processes Mary M. Shirley Hollis Chenery Howard Pack Discusses efficiencv of state-owned en- Staff Working Paper No. 539. 1982. 38 Staff Working Paper No. 377. 1980. 67 terprises. Gives the nature and size of pages. pages (including bibliography). this sector, including industrial and ISBN 0-8213-0097-0. Stock No. WP 0539. Stock No, WP 0377. $3. commercial firms, mines, utilities, $3 8 transport companies, and financial in- termediaries controlled to some extent Industrial Prospects and M- .4fi Arb by government. Tells how to increase Policie in th , .Mae Develo Jarralc I .ae.nJmaica the sector's efficiencv bv defining ob- jectives, controlling without interfer- Countries X p 5 ence, holding managers accountable Bela Balassa for results, and designing managerial Staff Working Paper No. 453. 1981. 30 skills and incentives. Includes bar pages (including appendix). graphs and charts of information for Stock No. WP 0453. $3. 24 developing and developed coun- tries. Industrial Strategy for Late Staff Working Paper No. 577. 1983. 116 Starters: The Experience of pages. Kenya, Tanzania and Zambia ISBN 0-8213-0241-8. Stock No. WP 0577. Ravi Gulhati and Uday Sekhar $5. Staff Working Paper No. 457. 1981. 63 Made in Jamaica: The Manufacture of Heavy pages (including references, annex). Development of the Electrical Equipment in Stock No. WP 0457. $3. Manufacturing Sector Developing Countries Korean Industrial Competence: Mahmood Ali Ayub Ayhan (;ilingiroglu Where It Came From This book, the first detailed study of Analyzes growth and competitiveness, Larry E. Westphal, Yung W. Rhee, Jamaica's manufacturing sector, pro- comparing prices and costs with those and Garrv G. Pursell vides a comprehensive assessment of in the intemational market. Staff Working Paper No. 469. 1981. 76 tse tmprtanfit chsatructure It relatesh The Johns Hopkins University Press. 1969. pages (including references). the development of the sector during Stock No. WP 0469. $3. the past two decades, describes the ex- LC 76-89962. ISBN 0-8018-1097-3, $5.50 tent of protection provided to the sec- paperback. tor in 1978, and examines the pros- Spanish: Fabricaci6n de equipo electrico pe- NEW pects for growth of manufactured sado en los paises en desarrollo. Editorial exports during the coming years. Pol- Tecnos, 1971. $5.50 paperback. Location Factors in the icy recommendations are made on the T M Decentralization of Industry: A basis of this analysis. The Minng Industry and the Survey of Metropolitan Sao The Johns Hopkins University Press, 1981. Developing Countries Paulo, Brazil 144 pages. Rex Bosson and Bension Varon Peter M. Townroe LC 80-27765. ISBN 0-8018-2568-7. Stock An overview of the world's nonfuel No. IH 2568, $6.50 paperback. mining industry, its structure and op- Focuses on decisionmaking procedures 'eration, and the major factors bearing for industrial companies that are estab- Managerial Structures and on them. lishing new plants or relocating their Practices in Manufacturing Oxford University Press, 1977; 3rd print- plant sites. Some 581 industrial com- panies in Brazil participated in a 1980 Enterprises: A Yugoslav Case ing, 1984. 304 pages (including 12 appen- survey to determine companv motives Study dixes, bibliography, index). for seeking a new site or building. Ap- Martin Schrenk LC 77-2983. ISBN 0-19-920096-3. Stock pendixes include nine detailed tables Staff Working Paper No. 455. 1981. 104 No. OX 920096, $29.50 hardcover; ISBN useful to industrial planners and com- pages (including 4 appendixes). 0-19-920099-8, Stock No. OX 920099, pany plant developers. Stock No. WP 0455. $5. $14.95 paperback. Staff Working Paper No. 517. 1983. 112 French: L'industrie miniere dans le tiers pages. monde. Economica, 1978. ISBN 2-7178- ISBN 0-8213-0005-9. Stock No. WP 0517. 0030-1, Stock No. IB 0538, S14.95. $5. Spanish: La industria minera y los paises en desarrollo. Editorial Tecnos, 1978. ISBN 84-309-0779-3, Stock No. IB 0521, $14.95. Occupational Structures of Policies for Industrial Progress Oxford University Press, 1980; 2nd print- Industries in Developing Countries ing, 1982. 325 pages (including bibliog- Manuel Zymelman John Cody, Helen Hughes, and raphy, index). Eighty-four tables profile the occupa- David Wall, editors LC 79-24786. ISBN 0-19-520176-0, Stock tional composition of industries in Analysis of the principal policy issues No. OX 520176, $24.95 hardcover; ISBN each of twenty-six countries. Data that influence the course and pace of 0-19-520177-9, Stock No. OX 520177, show the structure of employment by industrialization in the developing S9.95 paperback. sectors and industries for each coun- countries. The text, organized along try; cross-classify 120 occupations with lines of govemmental administrative Pollution Control in Sao Paulo, fifty-eight industries; and provide in- responsibility for various industrial Brazil: Costs, Benefits, and formation about productivity (value policies, includes chapters on trade, fi- Effects on Industrial Location added per person engaged), energy nance, labor-technology relations, tax- Vinod Thomas consumption per person engaged, and ation, licensing and other direct pro- employment. duction controls, public enterprises, Staff Working Paper No. 501. 1981. 127 1980; second printing, 1982. 2'l pages. infrastructure and location, industry- pages (including annex, references). ISBN 0-8213-0126-8. Stock No. BK 0126. agriculture linkage, and the intema- Stock No. WP 0501. $5. S20 tional environment. The Process of Industrial Development and Alternative The Planning of Investment Programs Development Strategies Alexander Meeraus and Ardy J. cesses of relevance to fertilizer produc- Bela Balassa Stoutjesdijk, editors tion and a systematic description of Staff Working Paper No. 438. 1980. 42 Series comprising three volumes (to the planning problems that need to be pages (including appendix). date) that describe a systematic ap- addressed during the project idenfifi- Stock No. WP 0438. $3. proach to investment planning, relying cation phase. primarily on mathematical program- The Johns Hopkins University Press, 1980. Public Subsidies to Industry: ming techniques. Includes both gen- 320 pages. The Case of Sweden and Its eral methodological volumes and stud- LC 78-8436. ISBN 0-8018-2138-X, Stock Shipbuilding Industry ies dealing with specific industrial No. JH 2138, $25 hardcover; ISBN 0- Carl Hamilton subsectors. 8018-2153-3, Stock No. IH 2153, $15 pa- Examines the reasons for the high Volume 1: The Planning of perback. govemment subsidies given to the Industrial Investment NEW Swedish shipbuilding industry during Programs: A Methodology the recession period of the 1970s. Swe- David A. Kendrick and Ardy J. Volume 3: The Planning of den's approach to the shipbuilding StoutJesdijk Investment Programs in the problem is compared with the adjust- Steel Industry ment made by Japan when it faced a The analysical approach with special A. l n dry similar situation. Concludes that a sta- emphasis on the complications arising David A. Kendrick, Alexander bilization policy is important in achiev- from economies of scale; a helpful in- Meeraus, and Jaime Alatorre ing the objective of full employment. troduction to linear and mixed-integer As a supplier of both capital equip- Staff Working Paper No. 566. 1983. 52 programming, facilitating understand- ment and materials for further proc- pages. ing of subsequent volumes in the se- essing, the steel industry has a sub- ISBN 0-8213-0196-9.Stock no. WP 0566. nes. stantial effect on the cost structure and The Johns Hopkins University Press, 1979. competitiveness of other economic ac- S3. 144 pages (including index). tivities. Its own cost structure, how- LC 78-8428. ISBN 0-8018-2139-8, Stock ever, depends to a large extent on the Restructuring of No. JH 2139, $18.50 hardcover; ISBN 0 efficiency of past investments. Manufacturing Industry: The 8018-2152-5, Stock No. IH 2152, $12 pa- Provides an overview of the technol- Experience of the Textile perback. ogy of steel production, and the prob- Industry in Pakistan, French: La programmation des investisse- lems of investment analysis in this in- Philippines, Portugal, and ments industriels: methode et etude de cas. dustry, and contains an application of Turkey Economica, 1981. (Combines translation of investment analysis to the Mexican Barend A. de Vries and Willem this book with that of the case study of the steel industry. Introduces a new eco- Brakel fertilizer industry in Volume 2, below.) nomic modeling language, GAMS, ISBN 2-7178-0328-9, Stock No. IB 0544, which decreases the time and effort re- Views the restructuring and moderni- $12. quired to construct and use industrial zation of manufacturing from the per- sector models. spective of World Bank assistance in Volume 2: The Planning of . . the textile industry. Evaluates the roles Investment Programs in the 3T2h8 Johns Hopkins University Press 1984 of government, the financial system Fertilizer Industry 328e. 1 T and the private sector in restructuring. Armeane M. Choksi, Alexander LC 83-18722. ISBN 0-8018-3197-0, Stock World Bank Working paper No. 558. Meeraus, and Ardy J. StoutJesdijk 8018-3198-9, Stock No. vH 3198, $15 pa- 1983. 59 pages. Discusses the main products and pro- perback. ISBN 0-8213-0151-9. Stock No. WP 0558. _ $3. Small Enterprises and Ste h NEW Development Policy in the h _sa Philippines: A Case Study a IWflutW Technological Change and Dennis Anderson and Farida T hS1ICa Industrial Development: Issues Khambata and Opportunities Staff Working Paper No. 468. 1981. 239 Frederick T. Moore pages (including bibliography, annex). Identifies principal issues relating tech- Stock No. WP 0468. $10. nological change to growth in indus- trial development. Draws upon the Small Industry in Developing theoretical and empirical literature for Countries: Some Issues an economic analysis of effective pro- Dennis Anderson gram designs. Projects underway in the engineering and capital goods in- Staff Working Paper No. 518. 1982. 77 dustries suggests methods for revising pages (including references). . policies and promoting new technolog- ISBN 0-8213-0006-7. Stock No. WP 0518. ical information in industry. $3. State Manufacturing Enterprise Staff Working Paper No. 613. 1983. 96 in a Mixed Economy: The pages. Small-Scale Enterprises in Turkish Case ISBN 0-8213-02574. Stock No. WP 0613. Korea and Taiwan Bertil Walstedt $3. Sam P.S. Ho Traces the historic roots of "etatism" Staff Working Paper No. 384. 1980. 157 and reviews the performance of six Transition toward More Rapid pages (including 4 appendixes). major state industries in Turkey. and Labor-Intensive Industrial Stock No. WP 0384. $5. The Johns Hopkins University Press, 1980. Development: The Case of the 354 pages (including appendixes, index). Philippines LC 78-21398. ISBN 0-8018-2226-2, Stock Barend A. de Vries NEW No. JH 2226, $30 hardcover; ISBN 0- Staff Working Paper No. 424. 1980. 32 8018-2227-0, Stock No. IH 2227, $13.50 pages (including references, 12 tables). Sources of Industrial Growth paperback. Stock No. WP 0424. $3. and Structural Change: The Case of Turkey Merih Celasun Considers the role of Turkey's public and private sectors in the industrial transition since the 1950s. Compares Turkey's trade prospects in the 1980s with growth in earlier periods of de- velopment as well as growth in other semi-industrial countries. Staff Working Paper No. 614. 1983. 188 pages. ISBN 0-8213-0283-2. Stock No. WP 0614. Prices subject to change without notice $5. and may vary by country. The World Bank Publications Order Form SEND TO: YOUR LOCAL DISTRIBUTOR OR TO WORLD BANK PUBLICATIONS (See the other side of this form.) P.O. BOX 37525 WASHINGTON, D.C. 20013 U.S.A. Date Name Ship to: (Enter if different from purchaser) Title Name Firm Title Address Firm City State Postal Code Address Country Te lephone I ) City State - Postal Code Purchaser Reference No. Country 'lUephone Check your method of payment. Enclosed is my El Check C1 International Money Order O Unesco Coupons O International Postal Coupon. Make payable to World Bank Publications for U. S. dollars unless you are ordering from your local distributor. Charge my O VISA rI MasterCard 0 American Express L Choice. (Credit cards accepted only for orders addressed to World Bank Publications.) Credit Card Account Number Expiration Date Signature rO Invoice me and please reference my Purchase Order No. Please ship me the items listed below. Customer Internal Stock Number Author/ Title Routing Code Quantity Unit Price Total Amount S All prices subject to change. Prices may vary by country. Allow 6-8 weeks for delivery. Subtotal Cost $ Thtal copies Air mail surcharge if desired ($2.00 each) $ Postage and handling for more than two complimentary items ($2.00 each) $ 'Ibtal S Thank you for your order. I BRD0063 Distributors of World Attnl: Mr Giancarlo Bigazzi Attn:Mr.J.M. Hernandez Bank Publications 50121 Madrid ARGENTINA Florence SRI LANKA AND THE MALDIVES Carlos Hirsch, SRL, JAPAN Lake House Bookshop Attn: Ms. Monica Bustos Eastern Book Service Attn: Mr. Victor Walatara Florida 165 45 piso Attn: Mr. Terumasa Hirano 41 Wad Ramanavake Mawatha Galeria Guemes 37-3, Hongo 3-Chome, Bunkvo-ku 113 Colombo 2 Buenos Aires 1307 Tokyo SWEDEN AUSTRALIA, PAPUA NEW GUINEA, KENYA ABCE Fritzes Kungl. Hov.bokhandel FIJI, SOLOMON ISLANDS, Africa Book Services (E.A.) Ltd. Attn: Mr. Eide Segerback WESTERN SAMOA, AND Attn: Mr. M.B. Dar Regerngsgatan 12, Box 16356 VANUATU P.O. Box 45245 S-103 27 Stockholm The Australian Financial Review Nairobi SWITZERLAND Information Service (AFRIS) KOREA, REPUBLIC OF Librairie Pavot Attn: Mr. David Jamieson Pan Korea Book Corporation Attn: Mr. Henri de Perrot 235-243 Jones Street Attn: Mr. Yoon-Sun kim 6, rue Grenus Broadway P.O. Box 101, Kwanghwamun 1211 Geneva Sydney, hSW 20001 Seoul TANZANIA BELGIUM MALAYSIA Oxford University Press Publications des Nations Unies University of Malaya Cooperative Attn: Mr. Anthony Theobold Attn: Mr. Jean de Lannoy Bookshop Ltd. Maktaba Road, P.O. Box 5299 av. du Roi 202 Attn: Mr. Mohammed Fahim Htj Dar es Salaam 1060 Brussels Yacob THAILAND CANADA P.O. Box 1127. Jalan Pantai Baru Central Department Store, Head Office Le Diffuseur Kuala Lumpur Attn: Mrs. Ratana Attn: Mrs. Suzanne Vermette MEXICO 306 Silom Road C.P. 85, Boucherville J4B 5E6 INFOTEC Bangkok Quebec Attn: Mr. Jorge Cepeda Thailand Management Association COSTA RICA San Lorenzo 153-11, Col. del Valle, Attn: Mrs. Sunan Libreria Tre'os Deleg. Benito Juarez 308 Silom Road Attn: Mr. Hugo Chamberlain 031 Mexico, D.F. Bangkok Calle 11-13, Av. Fernandez Guell MIDDLE EAST TUNISIA San Jose Middle East Marketing Research Societe Tunisienne de Diffusion DENMARK Bureau Attn: Mr. Slaheddine Ben Hamida Sanfundslitteratur Attn: Mr. George Vassilou i Avenue de Carthage Sanfundslitteratur Mitsis Bldg. 3 Tunis Attn: Mr. Wilfried Roloff Makarios fili Avenue Rosenderns Alle 11 Nicosia TURKEY DK-1970 Copenhagen V. Cvprus Haset Kita vpCi A.S. EGYPT, Arab Republic of (Branch offices in Bahrain, Greece, Attn: Mr. Ezzet Izerel Al Ahram Morocco, Kuwait, United Arab 469, Istiklal Caddesi Al Galaa Street Emirates. Jordan) Bevoglu-lstanbul Cairo NETHERLANDS UNITED KINGDOM AND MBE BV NORTHERN IRELAND FINLAND Attn: Mr. Gerhard van Bussell Microinfo Ltd. Akateeminen Kirjakauppa Noorderwal 38, Attn: Mr. Roy Selwyn Attn: Mr. Kari Litmanen 7241 BL Lochem Newman Lane, P.O. Box 3 Keskuskatu 1, SF-00100 NORWAY Alton, Hampshire GU34 2PG Helsinki 10 Johan Grundt Tanum A.S. England FRANCE Attn: Ms. Randi Mikkelborg World Bank Publications P.O. Box 1177 Sentrum UNITED STATES 66, avenue cl'lena Oslo I The World Bank Book Store 66,16 Pavenuerdis na Os A 600 19th Street, N.W. 75116 Paris PANAMA Washington. D.C. 20433 GERMANY, Federal Republic of Ediciones Libreria Cultural Panamena (Postal address: P.O. Box 37525 UNO-Verlag Attn: Mr. Luis Fernandez Fraguela R. Washington, D.C. 20013, U.S.A.) Atn:-Ver.JachmKrueAv. 7, Es ana 16 Baker and Taylor Company Simrockstrasse 23 Panama 'one 1 501 South Gladiola Avenue D-5300 Bonn I PHILIPPINES Momence, Illinois, 60954 HONG KONG, MACAU National Book Store 380 Edison Way Asia 2000 Ltd. Attn: Mrs. Socorro C. Ramos Reno, Nevada, 89564 Attn: Ms. Gretchen Wearing Smith 701 Rizal Avenue 50 Kirby Avenue 6 Fl., 146 Prince Edward Road Manila Somerville, New Jersey, 08876 Kowloon PORTUGAL Commerce Georgia 30599 INDIA Livraria Portugal C g UBS Publishers' Distributors Ltd. Attn: Mr. Antonio Alves Martins Bernan Associates Attn: Mr. D.P. Veer Rua Do Carmo 70-74 9730-E George Palmer Highway 5 Ansari Road, Post Box 7015 1200 Lanham, Maryland, 20761 New Delhi 110002 Lisbon Blackwell North America, Inc. (Branch offices in Bombav, Bangalore, SAUDI ARABIA 1001 Fries Mill Road Kanpur, Calcutta, and NMadras) Jarir Book Store Blackwood, New Jersey 08012 INDONESIA Attn: Mr. Akram Al-Agil Sidnev Kramer Books Pt. Indira Limited P.O. Box 3196 1722 P Street, N.W. Attn: Mr. Bambang Wahvudi Riyadh Washington, D.C. 20006 JI, Dr. Sam Ratulangi No. 37 SINGAPORE, TAIWAN, BURMA United Nations Bookshop Jakarta Pusat Information Publications Private, Ltd. United Nations Plaza IRELAND Attn: Ms. Janet David New York, N.Y. 10017 TDC Publishers 02-06 Ist Floor, Pei-Fu Industrial Attn: Mr. James Booth Building VENEZUELA 12 North Frederick Street 24 New Industrial Road Libreria del Este Dublin I Singapore Attn. Mr. Juan Pericas Dublin I Singapore Avda Francisco de M iranda, no. 52 ITALY SPAIN Edificio Galipan, Aptdo. 60.337 Licosa Commissionaria Sansoni SPA Mundi-Prensa Libros, S.A. Caracas 1060A HG 3881.5 .W57 W67 NO.669 C.4 GALENSON, ALICE, 1948- INVES,MENN INCENTIVES FOR The World Bank Headquarters European Office Tokyo Office H 1818 H Strcet, N.W. 66, avenue d'16na Kokusai Building Washington, DC. 20433, U.S.A. 75116 Paris. France 1-l Marunouchi 3-chonie Telephone: (202) 477-1234 Telephonc: (1) 723-54.21 Chiyoda-ku, Tokyo 00 lapa Telex: WUI 64145 WORLDBANK Telex: 842-62062 TIelephone: (03) 214-5001 RCA 248423 WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTONDC ISSN 0253-21 15/ISBN El-82 13-04 16-X