Sectoral Policy Preferences of the Peruvian Government, 1946-1968


The economic policies that have been implemented in Peru since 1968 by the military government have been the object of considerable attention. [1] By contrast, evaluation of Peruvian economic policy in the preceding two decades has received rather limited attention. It is our objective in this article to put into perspective the export bias of important elements of Peruvian economic policy in the period following World War II and to outline some of the consequences of that bias.
By the economist's conventional aggregate measures, Peru's economy grew more rapidly than that of most of the other Latin American countries in the period since the end of World War II. The growth propellant for Peru had been exports, [2] the dollar value of which grew at the average annual compound rate of 8.3 per cent between 1946 and 1968. Whereas severe foreign exchange shortages at times represented a constraint of some consequence in the process of development in Argentina, Brazil, and Colombia, such difficulties were largely absent in Peru, except in the immediate post-war years of 1946 to 1949.
The documents of the United Nations Economic Commission for Latin America and the United Nations Conference on Trade and Development have for some time proclaimed the dim prospects for the exports of the primary producing countries within the existing international economic framework. Simultaneously, academic observers have reported that the granting of extensive privileges to the export sector by developing countries is a thing of the distant past, having ceased when the race to industrialize got under way. Both of these generalizations must be abandoned when examining the Peruvian experience in the two decades following the War, when exports did grow and substantial inducements were provided to many firms producing for export, particularly foreign-owned mining firms.
In this paper we shall apply an approach inspired by the theory of sectoral clashes, developed by Markos Mamalakis, [3] in evaluating the performance of the Peruvian economy: a mid-century economy in which the export sector was 'dominant'. Our concern will be particularly with the relationships between the Peruvian government and the export sector. It is our belief that emphasis on the sectoral nature of public policy aids indispensably our understanding by broadening our perspective from that of economic analysis to that of political economy.

Sectoral growth patterns

The relative contribution of the various export sub-sectors to Peruvian growth cannot be unambiguously determined by an examination of sectoral income and product accounts. In Peru the contribution to gross output of the export sub-sectors is intertwined with the value of production for the domestic market in the data for agriculture, fishing, mining, and industry. Agricultural output includes the production of export crops such as sugar, cotton, and coffee as well as food crops for the domestic market and for subsistence consumption. Industrial output includes the processing of sugar, coffee, anchoveta (to produce fishmeal) and metallic ores. Fishing and mining may be regarded largely as export sectors, but sectoral data tend to understate their growth and relative importance owing to the industrial processing of the output of both sectors. Consequently, a comparison of changes in sectoral shares of national product or of sectoral growth rates does not provide a clear picture of the significance of changes in value added in export production in Peru.
Mining, fishing, and industry all increased their relative share of gross product in the period between 1950 and 1968 (Table 1). The quinquennial growth rates (Table 2) show that mining and fishing experienced the most rapid growth during the decade ending in 1960, while industry grew most rapidly from 1961 to 1965. Agriculture has been a lagging sector. Although the real value of agricultural output grew more rapidly than population, [4] it was not adequate to meet the growth of demand (as estimated by adjusting initial food expenditure levels not only by population growth but also by increases in per capita income and income elasticity of demand for food). The expenditure elasticity would have had to have been no greater than 0.22 for agricultural growth to have been adequate during the period from 1950 to 1965. In fact, however, one estimate has placed elasticity at 0.69. [5] The deficit was met by food imports, the annual dollar value of which tripled between 1950 and 1965. [6]
In his analysis of policy patterns, Mamalakis makes the distinction between ‘dominant’, ‘neutral’, and ‘suppressed’ sectors, depending on whether a sector has been favoured, overlooked, or penalized by government policy. [7] Dominance may or may not be associated with high rates of sectoral growth, depending in part on the effectiveness of the policies employed, as well as on the situation of the favoured sector. Rapid growth rates do not necessarily imply that a sector is dominant in the sense of having obtained preferential treatment from the government. Fishing, for example, grew largely in response to external demand. The fishing sector was a neutral one in that no special privileges were extended by the government to the sector until 1960, by which time the fishmeal boom was already well under way. The dominant sector in Peru throughout the 1950's was mining, while from 1960 mining had to share its dominance with industry. In the discussion which follows,


Peru: Sectoral Shares of Real Gross National Product (at 1963 prices),
1950 & 1968

Per cent

Agriculture and forestry


our attention will be directed primarily to the policies that provided dominance and to the mechanisms of dominance.


Peru: Annual Compound Sectoral Growth Rates of Real Gross National Product (at 1963 prices), 1951-55, 1956-60, 1961-60, 1961-65, 1966-68

per cent

Agriculture and forestry

Notes and Sources: as Table 1.

Export dominance

The rapid growth of Peruvian exports (and of the Peruvian economy) since 1949 is frequently attributed to Peru's liberal foreign exchange policy. The belief is reinforced by sophistic comparisons between Peru's experience in the 1950s and 1960s and that of her neighbours, as well as that of Peru itself between 1946-49. Exchange policy is, however, only one area in which generalized export dominance was manifested; tax, credit, and tariff policy were also relied on to varying extents to advance the fortunes of particular export commodities.
Exchange Policy . Peruvian exchange policy was characterized largely by avoidance of exchange control and periodic recourse to devaluation, with the resultant gains consistently accruing to the export sector. Only during the administration of Josè Bustamante y Rivero (1945-48) was a concerted effort made both to prevent devaluation and to allocate foreign exchange on the basis of government-established priorities. [8] Even Bustamante capitulated before his overthrow by varying the percentage of exchange which had to be surrendered at the official rate, a measure equivalent to a devaluation of the export rate. The ‘exporter-politicians’, as Bustamante called them, [9] could not be satisfied with a half-way measure; they wanted the abolition of the official rate.
Exporters found the administrations of General Manuel Odría (1948-56) and Manuel Prado 0956-62) more obliging. Within thirteen months after ousting Bustamante y Rivero from the Presidency, the Odría government abolished the official rate and set the sol free to fluctuate. This was the first of three periods in the 1950s during which the sol was allowed to fluctuate. In practice, the fluctuating rate was a convenient device for facilitating depreciation of the sol, and consequently met with the approval of the export sector. Depreciation of the sol was tolerated and occasionally even assisted by the Central Bank. In contrast, when appreciation of the rate began to occur, the Central Bank was quick to intervene to peg, the rate before the favoured situation of the exporters was eroded. A comparison of monthly exchange rates, Central Bank foreign exchange holdings, and (infrequent) drawings from the International Monetary Fund (IMF) amply demonstrates the export orientation of the Peruvian monetary authorities. [10]
The devaluations of 1953 and of 1958-59 both followed declines in the prices of agricultural and mining exports. The response of the Central Bank to cyclical difficulties in each instance was to release the sol to speculative pressures rather than to defend it. When recourse was made to the facilities of the IMF, either through the negotiation of a stand-by agreement (line-of-credit) or through drawing (borrowing), it was after most of the damage had been done. While a stand-by agreement was established in the month preceding the 1967 devaluation, the Central Bank quickly capitulated in the face of the export interests’ widely-held belief that the sol was overvalued.
No attempt was made to recapture through taxes or other means the windfall gains, which in some cases were quite sizable, that accrued to exporters as a consequence of devaluation. For example, as a Consequence of the export rate devaluations of September and December 1948, the sol value of cotton exports increased by 126.5 per cent in 1949 relative to 1948, even though their volume and dollar price changed very little. [11]
While devaluations may have brought gains for the export sector, they brought temporary setbacks to consumers and to other producing sectors. Investment levels and imports of capital goods either fell sharply or stagnated after each devaluation. The ratio of gross fixed investment to gross national product fell by six percentage points in the year following the 1953 devaluation, by five percentage points following the 1958 devaluation, and again by six percentage points following the 1967 devaluation. [12] The stimulus to domestic agriculture and industry resulting from the higher post-devaluation local prices for imported substitutes was offset by the increased cost of the imported investment goods necessary for expanding productive capacity. Imports of capital goods for agriculture, in particular, were strongly affected by the devaluations of the 1950's and of 1967. The dollar value of capital goods imports for agriculture in 1959, for example, were equal to less than one-half the value of those of 1957. [13] The same was true of capital goods imports for agriculture in 1968 relative to 1967. [14]
The gains from devaluation were concentrated; the losses, diffused. The periodic devaluations slowed short-run growth by depressing investment and may have moderated the rate and distorted the pattern of long-run growth by further concentrating income in few hands. The obverse of increases in the concentration of income is a limitation on the dispersion of purchasing power. If market size is small and growing slowly, new firms are more likely to be high-cost, low-volume operations. The blame for inefficient industry must surely be shared widely, but one must not overlook the negative contribution of an exchange policy which reflected export dominance in Peru.
Credit. The Central Reserve Bank of Peril, through its discount operations, effectively favoured to an increasing extent one group within the export sector: cotton producers. The Central Bank has generally provided virtually unlimited credit at rates which, at their maximum, did not exceed four per cent per year to the publicly owned Agricultural Development Bank. [15] The Agricultural Bank, in turn, increased the share of its total loans going to cotton producers from one-third in 1945/46 to a maximum of almost one-half in 1964/65. [16] By 1964/65 the value of new lending for cotton production was equivalent to 36 per cent of the value of (1965) cotton exports. [17] These loans were extended to producers who accounted for one-half the area planted in cotton.
Bustamante, writing in 1949 after being ousted from office, referred to the Agricultural Bank as a specialized agency of the commercial banks. [18] Nineteen years later, despite changes in the Bank's organic law, Bustamante's statement had a truer ring than when it was originally written. Loans by the Bank to cotton producers were secured, short-term crop loans, and were extended mainly to those who would probably have experienced no difficulty in obtaining commercial bank credit. A World Bank-Food and Agriculture Organization mission to Peru recommended in 1959 that the large-scale cotton producers be encouraged to borrow from the commercial banks. [19] As of October 1968 little had been done to implement that recommendation.
While the Central Bank had lent freely to the Agricultural Bank and other public development banks (the Industrial Bank and the Mining Bank), its operations with the commercial banks had been limited. At the ends of the financial years from 1946 to 1968, the Central Bank's outstanding loan balances to commercial banks as a percentage of the outstanding loan balances of commercial banks only once exceeded 10 per cent (1953), while in 1964 it fell to 0.2 per cent. [20] On the other hand, the indebtedness of the Agricultural Bank with the Central Bank was equal to one-third of the former's outstanding balances in 1964. [21] Moreover, in every year from 1954 to 1967, except 1957, the indebtedness of the Agricultural Bank to the Central Bank exceeded that of the entire commercial banking system.
The Agricultural Bank's preference for lending to cotton producers meant that less credit was available for food crops at a time when the growth of the agricultural sector was not keeping pace with the growth of domestic demand. While Agricultural Bank lending to cotton producers increased in relative importance both in relation to the Bank's portfolio and to cotton production, the value of cotton exports increased from $68.0 million in 1950 to a peak value of only $97.1 million in 1962. The imports of foodstuffs, however, increased from $38.7 million in 1950 to $114.4 million in 1965, while the corresponding value of all agricultural imports grew from $59.4 million to $167.0 million, the latter value being double that of 1965 exports of cotton . [22]
Thus the neglect of non-export agriculture was reinforced by the behaviour of one of the very institutions which was ostensibly created to promote balanced agricultural development.
Taxes. From 1950 to 1959 only one group of export producers had de jure tax preference: producers of all minerals (except petroleum, which is covered by separate legislation). The Mining Code of 1950 specified that the mining sector was to receive special tax treatment for a twenty-five year period. [23]
This included the provision of an allowance for depletion and exemption from the stamp tax, the prodesocupados tax of two per cent of taxable profits, and the excess profits tax (which existed until 1959).
Preferential treatment was extended to all existing mining enterprises regardless of whether they undertook new investments. No quid pro quo was asked by the government in exchange for the tax revenues which it was proposing to forgo. In contrast, a 1955 Chilean tax law imposed on large-scale copper mining a fifty per cent profit tax rate plus a surcharge of twenty-five per cent which could be increased or decreased depending on the relation of the level of output to that of a designated base period. No allowance for depletion was permitted in Chile. [24] In Peru all mining firms were subject to a twenty per cent maximum tax rate on profits net of depletion from 1950 until 1964, when the maximum was increased to thirty per cent. [25] The maximum was again increased in 1965 to thirty-five per cent [26] - still well below the Chilean rate. Moreover, the Peruvian depletion allowance, which is calculated as fifteen per cent of the gross value of production, could at its maximum reduce the already low tax liabilities by one-half.
The Mining Code also provided for the negotiation of special contracts with mining firms considering major new investments, but narrowly limited the range of obligations which could be demanded by the government from the contracting party: the maximum profit tax which the government could demand was twenty per cent of profits net of the depletion allowance. In the agreement signed by the Odría government with the Southern Peru Copper Corporation in 1954, under which the company agreed to develop the porphyry deposits at Toquepala, Cuajone, and Quellaveco, greater restraints were placed on the acts of the government than on those of the company. The ,crovernment prohibited itself under the terms of the agreement from placing restrictions on either imports and exports made by the company or on future earnings of the company. [27]
The special tax treatment accorded Southern Peru Copper Corporation was to remain in force for that period of time necessary for the sum of annual profits net of taxes, depreciation, and depletion to equal the amount of invested capital. The Peruvian National Planning Institute estimated that for the Toquepala investment alone (the only one of the three undertaken as of October 1968) the method designated for computing amortization would have served to maintain the system of privilege for eighteen years. [28]

Shared dominance : The Industrial Promotion Law of 1959

The Industrial Promotion Law enacted by the administration of Manuel Prado in 1959 [29] represented an attempt by the Peruvian government to elevate industry to dominance both through tariff protection and tax credits for investment. Paradoxically, the law, inter alia , extended export dominance by providing special treatment to the industrial processing of sugar, coffee, and fishmeal. Under the terms of the law, sugar mills and fishmeal plants located, as most of them are, in coastal provinces other than Lima and Callao, could invest amounts of up to sixty per cent of net profits free of tax. It seems reasonable to believe that some integrated firms, combining both 'extraction' and processing, reported depressed earnings for extraction and inflated earnings and claims of investment for processing in order to minimize tax payments. This belief appears to be supported by the fact that in the quinquennium 1961-65, the annual growth rate of real GNP generated in fishing was significantly lower than the growth rate of the dollar value of exports of fish and derivatives: 7.8 as against 30.0 per cent per year. [30]
Tax incentives are a blunt instrument which provides for the forfeiting of tax revenues from all those who fall into a legal category regardless of whether they would have increased investment without the tax concession. In Peru, the legal categories were extremely broad and hence incentives proved to be a particularly blunt instrument.
The granting of tax privileges to segments of the export sector and to industry resulted in a significant redistribution of the tax burden in Peru. It is usually thought that one manifestation of economic 'progress' is an increase in the relative importance of direct taxes (on income and wealth) as a source of government revenue. In Peru, however, the share of central government revenue accounted for by indirect taxes (on transactions) increased from 34.0 per cent in 1950 to 58.3 per cent in 1968. [31] As a consequence of export and, later, industrial dominance, the share of government revenue represented by profit tax receipts fell steadily from 46.3 per cent in 1950 to a low of 15.4 per cent in 1966. [32] Profit taxes paid to the government as a share of the taxable profits of business fell from 30 per cent in 1960 to a low of 17 per cent in 1966. [33]
To offset the revenue lost through concessions, taxes on non-dominant activities such as agriculture and commerce were increased, as were those which fell more heavily on consumers, thus tending to reduce the rate of growth of market demand. With the redistribution of tax obligations, sectoral clashes, whose existence analysts might previously have inferred, tended to become overt. In 1964, for example, cotton producers argued that a proposal to increase taxes on cotton was a direct result of earlier tax exemptions granted to the fishing industry. [34]
Tax burdens could not, however, match tariff levels as a generator of inter-sectoral polemic and of overt clashes. In 1950 the Peruvian National Industrial Society complained that the recently enacted tariff law (Law 11048, 1 July 1949) did not 'correspond in even the most remote form to the aspirations of Industry'. [35] Not only had tariffs been set at low levels, but both mining and agriculture were subsequently granted the right to import many of their required inputs duty free. [36] Ten years later, after the enactment of the Industrial Promotion Law, the situation had changed: an industry-government coalition was in its ascendancy, with industry bidding, at a minimum, to share with the dominant export sector the favours of an indulgent government.
The guarantee of 'adequate protection' to industry made by the Industrial Promotion Law brought an adverse response from the producer associations representing agriculture, mining, and fishing. The National Society of Fishing was probably the most outspoken, charging (in 1966), for example, that tariff protection was excessive and was leading to the establishment of anti-economic, parasitic industry which would increase the cost of living, benefit only a few, and would live 'on the effort and efficiency of other firms'. [37]

Mechanisms of dominance

The representation of interest groups in policy formation has been a well-established practice in Peru. Someone not familiar with Peru might have thought it curious that the National Agrarian Society, representing primarily the large coastal agriculturalists, had complained of not being consulted prior to the 1969 Agrarian Reform Law. In the Peruvian context, the accustomed procedure would not have been mere consultation, but rather participation in policy formation. In fact, such participation had been one of the major vehicles for facilitating sectoral dominance. The National Agrarian Society, for example, appointed members to the directorate of both the Central Reserve Bank and the Agricultural Bank. The National Society of Industries participated in the commission which reviewed the 1959 Industrial Promotion Law prior to its enactment. Similarly, the National Society of Mining and Petroleum co-operated in the framing of the 1950 Mining Code. The preceding are but a few examples of the widespread participation of interest groups in economic policy-making in Peru.
It was well recognized by producer groups in Peru that representation on public bodies was a key to obtaining preferential treatment for a sector. Associations representing fishing, mining, and agriculture had all sought unsuccessfully to gain membership on the Superior Council of Industries, a body which guided the application of the Industrial Promotion Law. The National Society of Fishing announced in its biennial report of 1962-63 that it was working to gain for fishing its ‘own direct access to the central nuclei in the economic direction of the country, such as the Central Reserve Bank ...’ [38]


The 1950s were a decade of relatively unchallenged dominance of the export sector (and, in particular, mining) in Peru. Before 1950, the government of Bustamante y Rivero had unsuccessfully attempted to limit the privileges of the export sector. In the 1950s, de facto export dominance became de lure dominance for mining, while the enactment of the Industrial Promotion Law by the government of Manuel Prado in 1959 extended privileges to a second major sector of the economy. The reformist government of Fernando Belaúnde Terry (1963-68) attempted to honour that part of the dominance pattern inherited from Odría and Prado which was represented by tax concessions. Belaúnde’s programme to promote development through public expenditure, agrarian reform in the Sierra and the construction of the Marginal Highway of the Selva had to be financed by a marked expansion of both government budget deficits and external public debt. [39] Pre-existing privileges remained intact until action was forced by the growing financial crisis.
Perhaps as significant as the policies themselves were the growing structures for overt and direct representation of interest groups in the formation of economic policy. Representatives of producers' groups were being involved to an increasing extent both in the framing of development plans and in the determination of short-run policy. What effect these evolving forms might have produced can now only be a matter for speculation. Since the military coup of October 1968, new policies have been enacted and new relationships are emerging. Those mechanisms of dominance which we have described have been largely swept aside. For example, representation of interest groups on the boards of public banking institutions has been replaced by representation of the relevant government ministries and public bodies . [40]
In the two decades preceding the 1968 coup, wide-ranging policies of a general nature (devaluation) and of a specific nature (credits, tax incentives, tariff exemptions) were used to stimulate export growth. Market forces were generally relied upon to spread the benefits of export growth to the rest of the society. Today, export promotion remains one of the foci of economic policy, but now mechanisms have been established (in the form of state marketing and/or ownership) whose objectives are to ensure that the gains of export expansion are put to the service of the community.

[1] For a review of a number of books dealing with Peruvian policy since 1968, see Colin Harding, 'Peru: Questions of Revolution', Latin American Review of Books, No. I (Spring, 1973), 185-90.
[2] Erik Thorbecke and Apostolos Condos concluded, for example, that Peruvian private investment ‘is determined completely exogenously through the changes occurring in the export sector’. Thorbecke & Condos, 'Macroeconomic Growth and Development Models of the Peruvian Economy,' Irma Adelman and Erik Thorbecke, eds., The Theory and Design of Economic Development (Baltimore, The Johns Hopkins Press, 1966), pp. 189-99.
[3] Markos Mamalakis, 'The Theory of Sectoral Clashes,' Latin American Research Review , IV (Fall, 1969), 9-46, and Mamalakis, 'The Theory of Sectoral Clashes and Coalitions Revisited,' Latin American Research Review , VI (Fall, 1973), 89-126.
[4] The value of the contribution of agriculture and forestry to real gross national product grew at an annual average compound rate of 2.8 per cent; population, at a rate of 2.6 per cent.
[5] J. S. Coffey, 'Income Elasticities for Peru - A Preliminary Analysis ' (Lima : University of North Carolina Mission, c. 1965).
[6] Banco Central de Rcserva del Perú, Cuentas nacionales del Perú ; 1950-1965 (Lima, 1966), pp. 54-55.
[7] Mamalakis, ‘The Theory...’ (1969), pp. 10-17.
[*] Includes electricity, gas, water, housing income, government, commerce, transportation, services, and banking.
Sources: Banco Central de Reserva del Perú, Cuentas nacionales del Perú 1950-1967 (Lima, 1968), p. 24 and ibid. 1960-1969 (Lima, 1970), p. 24.
[8] The legislative record of the period is contained in Jorge Eugenio Castañeda, Control de cambios (Lima, Casa Nacional de Moneda, 1949).
[9] José L. Bustarnante y Rivero, Tres años de lucha por la democracia en el Perú (Buenos Aires, 1949), pp. 235-36.
[10] Myron J. Frankman, Export Promotion and Developmental Priorities in Peru , 1946-1965 (Austin, University of Texas Ph.D. dissertation, 1968), pp. 192-93. The principal exchange rate was pegged following appreciation in October 1954 and in October 1959. In both instances increases in Central Bank official reserves preceded and followed the fixing of the value of the exchange rate.
[11] Banco Central, Boletín, various issues.
[12] Banco Central, Cuentas..., p. 26 and ibid., 1960-1969 (Lima, 1970), p. 12.
[13] Ibid., 1950-1965, p. 55.
[14] Ibid., 1960-1969, p. 44.
[15] Frankman, op. cit. , pp. 105-107.
[16] Banco de Fornento Agropecuario del Perú, Memoria, various issues.
[17] This was exceeded during the 1967-68 drought when the volume of cotton production and exports fell markedly.
[18] Bustamante y Rivero, p. 194.
[19] International Bank for Reconstruction and Development and United Nations Food and Agriculture Organization, The Agricultural Development of Peru . I. General Report (Washington, August 1959), p. 42.
[20] Banco Central, Boletín, various issues.
[21] Ibid. and Banco de Fornento Agropecuario, Memoria, various issues.
[22] Banco Central, Cuentas . . . 1950-1965, pp. 54-55.
[23] Law 11357, 12 May, 1950.
[24] Organization of American States and Inter-American Development Bank, Joint Tax Program, Sistenias tributarios de América Latina : Chile (Washington, Pan American Union, 1964), p. 35.
[25] Law 14920, 27 Feb. 1964.
[26] Law 15221, 16 Nov. 1964.
[27] El Peruano (13 Nov. 1954), p. 1.
[28] Instituto Nacional de Planificación, Análisis de la realidad socio-económica del Perú;. III. Análisis de los sectores económicos (A) (Lima, July 1963), pp. V-72.
[29] Law 13270, 30 Nov. 1959.
[30] Banco Central, Cuentas ... 1950-1967 (Lima, 1968), pp. 16, 44.
[31] Ibid., p. 34 and ibid., 1960-1969, p. 34.
[32] Ibid.
[33] Ibid., 1960-1969, p. 14.
[34] Sociedad Nacional Agraria del Perú, Memoria 1963-1964 (Lima), p. 74.
[35] Sociedad Nacional de Industrias, Memoria 1950 (Lima), p. 3.
[36] Frankman, op. cit. , pp. 224-34.
[37] Sociedad Nacional de Pesquería, 'Comunicado', La Prensa (Lima), 30 May, 1966, p. 3.
[38] Sociedad Nacional de Pesquería, Memoria 1962-1963 (Lima), p. 7.
[39] Peru had one of the highest growth rates of public external indebtedness in Latin America during the Belaúnde government. Inter-American Development Bank, Economic and Social Progress in Latin America, Annual Report 1972 (Washington, D.C.), p. 407.
[40] The powerful National Agrarian Society, though a shadow of its former self after the 1969 Agrarian Reform Law, was legally abolished by the government in 1972.