* I * 03 a.0. -* / UNITED STATES GENERAL ACCOUN WASHINGTON, D.C. 20 INTERNATIONAL DIVISION B-173240 SEP 24 1!3”$% lllllllllrllllllllllllllllllllllllll LM095514 Dear Mr. Mills: The General Accounting Office has made a study of the .development._impact.of i---:L- U.S. investment in less developed countries under the investment insurance program formerly administered by the Agency for International Development and ' now by the Overseas Private Investment Corporation (OPIC). ' Investment insurance is the most significant of several pro- grams now being administered by OPIC for encouraging the flow of new development-oriented U.S. investment to less developed countries. Our interest in the development aspects of this program stems from the stress placed by the Congress on the need for the activities undertaken by OPIC to be development oriented. The recent legislation establishing OPIC states that OPIC is II to encourage and support only those private invest- ment; in less developed friendly countries and areas which are sensitive and responsive to the special needs and require- ments of their economies and which contribute to the social and economic development of their people; . . ." Our study relates to insured investment in extractive and manufacturing industries in the less developed countries. Insured investment in these industries has accounted for about 77 percent of total insured investment. In our study we focused on the impact of insured investment on employment, efficiency, exports, and stimulation of the growth of local industry. While these are not the only ways of measuring developmental impact, they are important factors in deciding whether insured invest- ment may benefit less developed countries. 50 TH ANNIVERSARY 1921.1971 . 1 \ l ,The results of our study, although based upon limited available information, did indicate that the investment insur- ance program had had varied impact in stimulating development in the less developed countries. We are attaching, for your information and consideration, a summary of what we found, with the hope that it will assist OPIC in the efforts it is report- edly making to develop sensible and workable criteria for assessing the development impact of the investment insurance program. We would be happy to discuss any of these matters further with you or your staff if you wish. We recognize, of course, that investors are primarily motivated by considerations of profitability which may not coincide with development considerations and that the U.S. Government has less influence over privately financed projects than those which it finances. Hence, we can appreciate the views of OPIC officials that their choice is not whether a proposed investment offers maximum development impact to the less developed country, but rather to decide whether the par- ticular mix of returns-- to the less developed country, to the investor, and to the enterprise concerned--on balance offers significant constructive development impact in and of itself. We wish to thank you and your staff for the cooperation extended to us during the course of our review. Copies of this letter are being sent to the Administrator, AID (Chairman, Board of Directors of OPIC), the Director? Office of Management and Budget, and the Foreign Operations and Govern- ment Information Subcommittee of the House Government Operations, ,. Committee. Sincerely yours, f ~QJA 3,ia c\<- \ o-. $, Oye V. Stovall Director Attachment Mr. Bradford Mills President, Overseas Private ,. Investment Corporation -2- . . Attachment Some Factors For Consideration In Assessing Development Impact of U.S. Insured Investment in Less Developed Countries EMPLOYMENT The recently published Pearson Commission Report concluded: "The failure to create meaningful employment is the most tragic failure of development. All indica- tions are that unemployment and under-utilization of human resources have increased in the 1960's and that the problem will grow even more serious. I! h/ As an example of the magnitude of the employment problem, the Economic Commission for Latin America has estimated that about 30 percent of the labor for (economically active popu- lation) is un- or underemployed. -59 The incidence of unemployment is chiefly on unskilled labor. While unemployment of unskilled labor is large and increasing, there is at the same time a shortage of skilled labor in less developed countries. One of the major reasons for increased unemployment has been that industrial growth in the LDCs has been concentrated in the more capital intensive industries. a/ For this reason, GAO has evaluated the employment impact of previously-insured investment in terms of whether investment went into activities which are relatively capital and skilled- labor intensive (which will be referred to as capital intensive) or into activities which are unskilled labor intensive (which will be+?referred to as labor intensive). For purposes of this evaluation, capital intensive industries are defined to be those industries where value added (value of production minus cost of inputs purchased from other industries) per employee exceeds the average (median) value added of the entire sector ( i.e., manufacturing, mining, or manufacturing and mining combined) in the United States and labor intensive industries 11 Pearson, L.B., Partners in Development, Report of the Commission on International Development, Praeger Publishers, New York, 1969, p. 58 a Economic Survey of Latin America, 1968, Economic Commission for Latin America, 1969 (Dot. E/CN.12/825). 2/ Dziadek, F., Unemployment in the Less Developed Countries, AID Discussion Paper No. 16, June 1967. 'are those where value per employee is less than the average (median) value added per employee for the entire sector in the United States. This methodology is based on a recent study of LDC manufacturing exports which shows that value- added per employee is a good measure of capital intensity and that there is an international similarity in the ranking of industries by degree of capital intensity. 11 The results of our investigation are shown below: Percentage of insured investment relatively Insured investment labor intensive Mining and Manufac- Mining and Manufac- Year manufacturing turing only manufacturing turing 1961 $ 41.6 million $ 12.9 million 1 1 1965 181.5 million 159.5 million 21 14 1969 458.1 million 244.1 million 7 9 Itis clear that insured investment has gone into the more capital intensive industrial activities. It would therefore appear that employment returns from U.S. investments have been limited. 21 EFFICIENCY Industrialization is a key to improving the economic status of many of the less developed countries, and consequently a number of strategies have been adopted by them to promote the establishment and expansion of industrial activities. A major problem, however, in fostering industrialization that has plagued many industrial promotion programs is that the industries esta- blished may be economically inefficient; that is, the goods pro- dueed by these industries may be significantly more costly than if they could be purchased in the world market. L;/ Lary, H.B., Imports of Manufactures from Less Developed Coun- tries, National Bureau of Economic Research, New York, 1968. 2/ CA0 recognizes that the measure used is only one, albeit, important indicator of the employment impact of U.S. foreign investment. -20 It might be argued that in the context of a less developed country, cost considerations especially those of domestic resources are not important. Apparently, this argument is based upon the characterization of a less developed country as a labor surplus economy. However, it ignores the fact that production in less developed countries involves a significant use of scarce ( i.e., expensive) factors of production. These include imports, domestic capital and skilled labor. The generalization that labor is plentiful most certainly does not apply to skilled labor. The efficiency of insured investment projects may be gauged by comparing the cost of production associated with insured invest- ment projects in less developed countries with the cost of produc- tion elsewhere. Unfortunately, the data for maki;$ s;;trzs;;s.ments of insured investment projects were unavailable. have not been any studies covering a large number of LDC's, on the comparative cost of U.S. investment (i,e., insured and non-insured). It is not possible to determine from the available evidence whether insured investment has been efficient (i.e., costs of production from insured investment in less developed countries are reasonable relative to costs of production elsewhere). The evidence, however, does suggest that efficiency requires special attention by OPIC. A recent study of industrial development in seven LDC's concluded that there has been high cost, as well as efficient foreign investment in these countries. 2/ We examined a number of studies that compare costs in LXX's to foreign costs of production. These studies reveal that many industries in LDC's are high cost relative to the world market. A/ L/ Responses to OPIC's new questionnaire may permit such assess- ments of proposed insured investments. 2?./ I. Little, et. al., Industry and Trade in Some Developing Coun- tries, OECD, Oxford University Press, New York, 1970, pp. 58-59. 2/ Bela Balassa and Associates, The Structure of Protection in Developing Countries, preliminary study sponsored by the International Dank for Reconstruction and Development and the Inter-American Development Bank; airgram for Turkey AID Mission dated 10/31/68 discussing the cost of Turkish industry; and Little, op. cit. -30 One study covers the manufacturing sector in some detail in five LDC's for which there are insurance programs - Brazil, Chile, Pakistan, the Philippines, and Western Malaysia. J./ With the exception of one country, it appears that insured manu- facturing investment in these countries has located in industries which are high cost. 21 The following schedule summarizes each country's performance: Percent of insured manufacturing Country Brazil 85 Chile 71 Pakistan 42 Philippines 87 Western Malaysia 0 Balassa, op. cit. The cost data for Brazil, Chile, Pakistan, the Philippines, and Western Malysia are for 1967, 1961, 1963/64, 1965, and 1965, respectively. The product groups used in the study are broader than the industry categories by which investment is classified. Therefore, the correspondence of the investment figures with their respective product groups is only approximate. Thus, the level of aggregation in the study's data may con- ceal differences in relative costs for individual industries and companies. Some insured investment could not fit into any of the product groups identified in the individual country chapters of the study. We therefore omitted such investment. Except for the Philippines, the amounts excluded were relatively small. -4- During the course of our review, the Agency for Inter- national Development retained a consultant to propose criteria for determining whether potential insured investment is likely to lead to high cost production. It is our understanding that his proposed guidelines have been adopted by OPIC. EXPORTS Exports are important for the development of LDC's for several reasons. They permit a country to earn foreign exchange for purchasing vital imports of goods and services which may not be able to be produced domestically in sufficient quantity and at reasonable cost. They are a good indicator of the extent to which domestic goods are produced efficiently; if a signifi- cant proportion of a country's output is sold abroad, it is likely that domestic costs for products it exports are reason- able because exports generally must be priced competitively with those of other countries. Also the quality of domestic output is likely to be high because exports in general must compete in quality as well as in price with the exports of other countries. Replies to a joint AID-GAO sample survey of insured investors show that exports of insured U.S. manufacturing affi- liates amounted to 10 percent of total sales in 1969-70. However, if intra-company exports, which frequently represent transactions at non-market prices are excluded, the export proportion of insured investment declines to 5 percent. The survey also shows that only 25 percent (12 percent excluding intra-company exports) of insured investment was in manufacturing activities which generated exports amounting to 10 percent or more of total sales. Therefore, it is doubtful that U.S. insured investment in manu- facturing has a satisfactory export orientation. L/ There were few responses to the survey by insured extractive industry affiliates. As a proxy, we used Department of Commerce data, which cover both insured and uninsured extractive investment I/ There appears to be an improvement in export orientation between 1957 and 1966. Department of Commerce data shows the export proportion of U.S. manufacturing affiliates in Latin America, including intra-company sales, increasing from 4 to 10 percent. Since the 1957 data do not separate intra-company sales, a comparison net of intra-company sales is not possible. -5- in Wtin America for 1966. Much of U.S, direct extractive industry investment in Latin America appears to be insured. Moreover, insured extractive industry investment in Latin America is a large proportion of total insured investment in this sector in all LDC's. In 1966, exports of U.S. extractive affiliates in Latin America were 65 percent of total sales. 11 If company sales are excluded, the export proportion declines to 41 percent. Therefore U.S. investment in extractive industries has a good export orientation. STIMULATION OF THE GROWTH OF LOCAL INDUSTRY U.S. foreign investment may make an important contribution to the development of less developed countries by stimulating the growth of local industry. One way in which it can do this is by spending on locally produced goods and services. Replies to the previously mentioned joint AID-GAB sample survey of insured investors indicate that insured U.S. manufactur- ing affiliates have spent a significant amount on locally produced goods and services. In 1969 - 70, local purchases amounted to 44 percent of their total sales. There were few responses to the survey by insured extractive affiliates. As with exportss as a proxy we used Department of Commerce data, wh%ch cover both insured and uninsured extractive investment in Latin America in 1966. Although the relative amount of local purchases are lower than for manufacturing affiliates, we conclude that U.S. extrac- tive affiliates have spent a significant amount on locally pro- duced goods and services. In 1966 local purchases of materials, supplies and services of U.S. extractive affiliates in Latin America amounted to 29 percent of their total sales. t/ These results include manufacturing as well as the mining aspects of petroleum industries. Only the former category is eligible for AID insurance with limited coverage avail- able for the exploration and production aspects of petroleum mining investment. -6- *U.S. foreign investment may also stimulate the growth of local industry by encouraging local participation in the manage- ment of the enterprises into which U.S. investment enters. Local participation in the management of U.S. foreign affiliates is also important because it helps to reduce nationalist sensi- tivities to foreign investment. One measure of the extent to which insured investment has encouraged local participation is the proportion of local equity in the total equity of U.S. imsured affiliates. A tabulation of the usable results from the joint AID-GAB sample survey shows the following: % Local Equity Number of Insured Projects 0 29 1 - 24 5 25 - 49 I.6 50 & over 22 zf It is clear that local equity participation in U.S. financed ventures has been low. A salient feature is the relatively large number of firms reporting no local equity participation. In this connection, it is interesting to note that the President's task force on International Development recommended that QPIC make greater use of its guaranty programs to encourage local equity participation. -7-
Development Impact of U.S. Investment in Less Developed Countries
Published by the Government Accountability Office on 1971-09-24.
Below is a raw (and likely hideous) rendition of the original report. (PDF)