- Learning from Experience
- Alternatives to Current Development Strategies
- If Not Import Substitution Industrialisation, What Alternative Strategy?
- There Are Policy Alternatives – Examples
Development strategies experimented in Latin America are currently classified as those that promote development "desde dentro" "from within" or that are "outward looking". The former means that the engine of development comes from domestic demand sufficient to fully mobilise domestic resources through the full exploitation of domestic resources. The latter is based on full, unrestricted participation in the global trading and financial system to allow comparative advantage trade financed by external resources to act as an engine of growth. In the past, Latin America has experienced both types of development strategy. The 19th century was characterised by open trade and financial flows, while the period since the 1920s has emphasized development from within based on what eventually came to be called import substitution industrialisation. While the former policy was successful in the latter part of the 19th century, it failed miserably in the first quarter of the 20th and eventually collapsed under the weight of developed country trade restrictions and recession in the 1930s. On the other hand the latter policy appears to have been as successful as a response to the Great Depression and the war years, reaching it highest point in the 1950s and 1960s. Similar to the reconstruction strategies employed to rebuild Europe after the Second war, Latin America also experienced "economic miracles" of extremely rapid growth relative to current experience. For the 21 years from 1950 to 1970 the eight major Latin American countries (Argentina, Brazil, Chile, Colombia, Mexico, Peru, Uruguay and Venezuela) grew at an annual average rate in excess of 5.25 per cent with per capita incomes growing at over 2.36 per cent. In Brazil the increase in per capita incomes was over 3 per cent.
Alternatives to Current Development Strategies
There are thus at least two alternative strategies to evaluate. However, somewhat paradoxically the debate in the last quarter of the 20th century has proceeded as if there were no development strategies. Or better said, the current strategy is that the best development strategy 2 is no development strategy. This may sound a bit extreme, especially since most everyone believes that current official development strategy is driven by the Washington Consensus, or now Washington Consensus Mark II.
My point is based on John Williamson’s affirmation that the policies contained in the Consensus were based on "ideas that had long been regarded as orthodox so far as OECD countries are concerned". The Consensus was meant to counter what Williamson characterises as "a sort of global apartheid which claimed that developing countries came from a different universe."1 This implied that the difficulties faced by Latin American economies were largely the result of the belief that Latin America required policies that were specific to the context of the region. The implication of the Consensus was not only that there was no need for a specific approach to the problems of development in Latin America, but that there was no need for a special theory of economic development – traditional neoclassical theory and policy were all that was required.2
But in the late 1980s this position was not radical. Indeed, it has a long history. Anyone knowledgeable in the 19th century history of Latin American will recognize a similar approach to development policy , supported by business and financial circles and widely supported by national governments who called upon developed country economists to apply their theories in the region.
A similar approach was also the basis for early development policy in the United Nations: "most of us …felt that economic problems are economic problems wherever they happen to be. You wanted to know the facts of a particular country and situation, of course, but detailed knowledge of a particular country wouldn’t make much difference as to exchange rate policy or the investment situation or trade regulations and so on."3
Although UN economists such as Raul Prebisch, Singer and Lewis would eventually provide an alternative point of view, the dominance of this position was short-lived. Almost a decade before the pronouncement of the Washington Consensus, Depak Lal had attacked what was by then considered the traditional "development economics" of authors such as Hirschman for being "theoretical curiosities" based on a perversion of standard economic principles which are held to be applicable in developing economies only, and "falsely deny the universal applicability of rational economic behaviour and the existence of marginal substitution possibilities on which standard economic theory relies for its results."4 Even worse, they argue that the "case for free trade is invalid for developing economies." Or simply that development economics was not economics.
Lal was providing, with journalistic bombast, an extension to academic development theory of the monetarist "counter-revolution" linked to the work of Milton Friedman, and supported by the political revolution led by Margaret Thatcher and Ronald Reagan.5 Set off by Little, Scitovsky and Scott’s OECD study of trade and industry in developing countries that provided strong crisiticsm of the impact of protection on growth, the counter-revolution was soon extended to a call for minimal state intervention, free trade and free markets as the basis for eliminating the errors and inconsistencies in government regulations in support of development. Since one of the mistaken policies that Williamson singled out, "import substitution industrialisation", was generally believed to be little more than extended protection, Latin America obviously was ripe for counter-revolution..
Williamson notes that he did not create the Washington Consensus, he was only providing a faithful rendition of the new policy approach then accepted by Washington development circles. This is certainly correct for almost a decade earlier the US representative to the Asian Development Bank had declared that "the United States completely rejects the idea that there is such a thing as development economics."6 Thus, not only did Latin America not require a special theory of development, there is no such thing as development economics from which policies could be derived.
It is also important to note that it was not only in downtown Washington that these policies were commonly accepted. They had been proposed by influential Latin American business and commercial interests since the late 1960s.
In Mexico Coparmex,7 representing the financial banking elite and Monterrey business interests, presented its own seven-point programme for the new Eceverria government to adopt as early as 1971. Although the recommendations were not implemented they provided the framework for the policies that were introduced in the late 1980s. The programme included :
$ reprivatisation of the state controlled enterprises;
$ an end to all state subsidies to state firms;
$ an increase in direct foreign investment;
$ an end to all land reform programs and a "reorganization" of the countryside.
In Brazil, Jânio Quadros, successor to the the populist Kubischek government in 196_ received a set of suggested economic policy measures (Sugestões para uma politica nacional de desemvolvimento) from Conclap that reaffirmed the primary role of the private sector and foreign investment in the development process, the reduction of the role of the State, the control of populist tendencies, the redefinition of the role of the state, control of inflation and a improvement in the public administration.8 When Quadros resigned he was replaced of Joao Goulart, who had defeated Quadros=s running mate in the election. The rapid shift back towards populist policies indicated by Goulart soon led to a military takeover, which did not proceed with the implementation of the suggestions.
In 1976 the Consejo Economico Argentino produced a comprehensive economic programme for the new military government, the Proceso de Reorganización Nacional 9 It was based on three points:
$ redefining the role of the State and reducing its role in the economy,
$ the liberalisation, opening and modernisation of the economy,
$ monetary stability.
As in Brazil and Mexico, the program was not instantly accepted by the new military government, but became the background for policy discussion and eventually became the basis for government policy when democratic governments returned.10
Thus, there is no existing theory for which I can provide an alternative. What needs to be done is to provide a rationale for a development theory.
If Not Import Substitution Industrialisation, What Alternative Strategy?
Not only did the Washington Consensus suggest that no special development theory was required, it represented a return to policies that had failed repeatedly in the 19th century. However, it did reflect the changing political environment facing developed countries in the1980s concerning the role of government and market mechanisms. Thus if we are to find an alternative theory it must be one that builds on the idea that developing countries face special conditions and require a special theory. It is interesting that this is precisely the approach that had been adopted by a wide range of economists in work done in the 1950s and 1960s.They include a number of Latin American economists at CEPAL, Raul Prebisch and his collaborators such as Celso Furtado, but was also characteristic of the work of Gunnar Myrdal, Albert Hirschman, Thomas Balogh, Paul Streeten, Arthur Lewis and Dudley Sears. I will concentrate on the approach of Prebisch, first since his work has had the greatest influence on the positions taken in the United Nations, and in particular in UNCTAD. Second, because some of these economists built on Keynes’s theory to reject neoclassical theory as the appropriate basis for any economic policy and presumed that Keynes’s theory could be generally applied as neoclassical theory had been before. Prebisch on the other hand considered neoclassical theory as inappropriate for understanding the problems facing developing countries in Latin America and developing policies to remedy them, but was also critical of certain aspects of Keynes’s theory that did not reflect the reality of the region. For Prebisch, theory evolved from an understanding of the real problems and conditions facing Latin America.
Let us start with neoclassical theory. First, neoclassical theory starts from individual equilibrium, aggregates to market equilibrium for goods and factors, and then to the general equilibrium of output and employment for the system as a whole without ever having to mention international trade and the role of government. These are usually added only as an afterthought and then within the same framework. While neoclassical theory deals with the benefits from opening an economy under autarchy to trade, the economies of post-conquest Latin America have always been open to trade.11 While neoclassical theory deals with distortions that non-profit maximising actions of government may introduce into a perfectly competitive system, the economies of post-conquest Latin America have for most of their history had their production and price structures imposed by colonial, populist or military governments. It was the fact that developing countries were colonised precisely in order to provide the raw materials for the industrial structures of developed countries that led to international division of labour that caused their concentration in primary commodity production. As Marx12 pointed out, and Bulmer- Thomas13 has emphasised, hardly any of the primary commodities which now dominate production and trade between developing and developed countries were indigenous to Latin America. Coffee and sugar are particularly important examples of what are called "non-traditional" commodity exports. Metals, cacao and petroleum seem to be the major indigenous primary materials and the former were quickly exhausted and the latter was only discovered late in the development process.
It is interesting that extensions of Keynes’s theory in its neoclassical synthesis form also started with firms and households and only subsequently adjusts for the impact of international trade and government on aggregate demand. But, in the Prebisch approach the most appropriate theory would be one that starts analysis on the basis of trade and government action.
The second facet deals with the fact that equilibrium in neoclassical theory requires diminishing returns to prevail if equilibrium is to be achieved in utility and profit creation, and in its analysis of the gains from trade requires that all countries have access to the same production function so all have access to the same set of technical knowledge. But, as the theory of the declining terms of trade points out, if the former were the case, then Malthus and the other "dismal scientists" would have been right and population growth would have long ago outstripped the growth of agricultural production. There must be either increasing returns or technical progress in the production of primary materials that has allowed substantial increases in productivity in agricultural production. Indeed, this was the basis of the idea of the declining terms of trade – the point was not the differential behaviour of the prices of primary products and industrial goods, but rather the differential potential for increasing productivity and the fact that the increased productivity in agricultural goods produced by developing countries did not accrue to domestic workers in the form of increased real wages, but was transferred to industrial workers in developed countries. This increased purchasing power then became the basis for rising demand for industrial products that through economies of scale further increased their productivity. It was the international distribution of the productivity gains from increasing returns to scale and technical progress that was the important issue. As Prebisch pointed out, this was not only a problem faced by developing countries.14 Developed countries also experienced difficulties in creating an equitable distribution of the fruits of technical progress between agricultural and industrial producers. In the early stages of development in one such country this produced a civil war over a distribution that was based on slavery, as it had been in many other countries. However, the problem was subsequently resolved by creating substantial income transfers to agricultural producers in the form of production subsidies and income supplements. Absent an advanced domestic industrial sector, developing countries could not avail themselves of this remedy. Absent an international mechanism to return the gains of their technical progress, the only possibility was for developing countries to build their own industrial sector in order to benefit from increasing returns and technical progress. Indeed, Prebisch argued that their was a symbiosis between the two sectors that caused productivity to rise more rapidly if a domestic industrial sector were to be developed. The point has also been argued by the agricultural economist J.K. Galbraith, "a purely agricultural coutnry is likely to be unprogressive even in its agriculture," since industrialisation creates technology that can be applied to agriculture, while the opposite is hardly ever true.15
Prebisch also recognised that the only way developing countries would have access to the existing technical knowledge embodied in the neoclassical production function was by importing the technology in the form of capital goods. Not only was the existing industrial structure to produce these technologically advanced goods lacking, imports would allow developing countries to skip the long process that have been required by developed countries to reach the existing level of technology.
The imperative of domestic industrialisation was to reap increasing returns to generate higher productivity in order to raise domestic real wages and build a strong internal market that led to the importance of the external constraint. If developing countries were to buy the capital goods imports they needed to build a domestic manufacturing sector they would have to produce enough exports to pay for them. This would be a Sysiphisian task if developing countries only exported primary materials subject to declining terms of trade.16 Thus, the objective of a more equitable global distribution of the fruits of technical progress led to the emphasis on domestic industrialisation that depended on promotion and development of exports.
In this regard it is important to note another difference between Prebisch and traditional theory. While he supported the central role of the market in development, there were particular tasks for which he noted it was ill-suited. Primary was the elimination of the extreme inequalities of the distribution of income17. Just as the operation of the free market would fail to reverse, indeed would reinforce, the deterioration in the terms of trade between agriculture and industry in a developed country, he applied the same reasoning to the distribution of income between developing and developed countries. In Harrod’s terminology we would say that the dynamic operation of the international price mechanism produced centripetal forces driving the global economic system towards greater and greater disparity as developing countries specialised in primary commodities and developed countries in manufactured goods. External intervention would be necessary to reverse the cumulative forces of the market mechanism. Similar views can also be found in post-war discussions of uneven development and dualism in Europe such as Francois Perroux’s concept of "pôles de croissance"18, Hirschman’s19 concepts of "polarization" and "trickling down" in his identification of "Growing points" and "Lagging regions" and Myrdal’s ideas of "spread" and "backwash".20 They are also part of Kaldor’s rejection of the concept of equilibrium and its substitution with the idea of cumulative causation.
The centripetal operation of market forces is also based on the integration of micro and macro factors. The basic cause of the deterioration in the terms of trade and the unequal distribution of the fruits of technical progress was due to the way former colonies were integrated into the international division of labour. Latin America served as an integral part of the global production system of the British Empire. England produced manufactured goods using inputs of developing countries’ primary materials. It lent Latin American developing countries the finance to buy its capital goods exports and developing countries paid debt service by exporting primary commodities to England. In this international division of labour the periphery is dependent on the industrial centre.
But market structure was also important. The fact that commodity markets were international and competitive while manufactured goods markets were largely domestic and oligopolistic meant that increasing output in the former led to lower prices and income and the associated "pig cycle" phenomenon, while in the latter increasing prices lowered costs and increased real wages and margins.21 Again, the market forces were centripetal, with falling incomes and prices in the former and rising profits and wages in the latter, leading to the lack of incentives for investment in the former and stimulus to investment in the latter. It is interesting to note that the recent report by the group of experts convened by the Secretary General of UNCTAD suggested that a new oligopolistic form of market for commodities had eliminated competition in final product markets and at the same time created monopsony in the producers markets, so that prices to developing country producers continue to fall, but the transfer of the fruits of technical progress now go to profit the transnational companies that act as intermediaries in these markets, rather than increasing the purchasing power of developed country consumers.22
These are not market imperfections, but the natural operation of dynamic market forces similar to those noted by Marshall in his discussion of the irreversibility of downward sloping supply curves representing decreasing costs and gave a pricing advantage to firms who first exploited them. As noted, for Prebisch the solution was not to make the markets function better, but to get on the right side of the unequal distribution mechanism by developing a manufacturing sector. "Industrialisation is not an end in itself, but the principal means at the disposal of those countries of obtaining a share of the benefits of technical progress and of progressively raising the standard of living of the workers"23.
While it is true that domestic industrialisation by its very nature implies a degree of state intervention as well as a degree of import substitution – this was precisely the point of breaking out of the 19th century international division of labour, it had to be accompanied by export promotion and the creation of market size big enough to prevent state or private monopolies. As Prebisch points out24, isolation and protectionism were never part of this approach. On the other hand, the Great Depression and the Second World War meant that Latin American countries faced forced domestic industrialisation if they were to continue to have manufactured goods available, and the conditions of the 1930s, with riding protectionism and falling international trade, were hardly propitious to developing export industries. Indeed, Prebisch was critical of the fact that this industrialisation, when it occurred, tended to be concentrated in consumption goods destined to high-income consumers, rather than in capital goods, and thus had little potential to provide increasing productivity or a dynamic domestic market. This had a further drawback in creating a national consumption pattern more appropriate to developed country real income levels and produced consumption aspirations that were detrimental to national savings and created an divergence between nominal and real income growth.
Indeed, by the 1950s Prebisch and others were supporting the types of policies that were eventually introduced in the Asian newly industrialising economies such as Korea and Taiwan. A reading of the Report of the Secretary General to the First United Nations Conference on Trade and Development makes this position very clear, and it has recently been reiterated by Alice Amsden.25 What Amsden’s position makes clear is that few Latin American economies managed to follow this approach and Prebisch was among the first to criticise governments for their continued concentration on import substitution in the post war period when global markets were expanding and international trade was growing more rapidly than global income.
Although Prebisch was an early student of Keynes’s General Theory his alternative approach was independent of it and he was critical of Keynes’s adoption of the "classical" mode of arguing in terms of equilibrium. As noted above, his approach was dynamic and much closer to the ideas of cumulative causation associated with the work of Myrdal and Kaldor. For this reason Prebisch considered the multiplier the most important part of Keynes’s approach. This is also the part of Keynes’s theory that has been adopted as the core of the theories of development that emerged in the post-war period based on Harrod’s blending of the multiplier and accelerator and Domar’s stable growth conditions.26 Unfortunately, both deal with closed, developed economies.
The aim of Keynesian theory is to show how developed countries can achieve steady growth with full utilisation of productive capacity by using government spending to supplement the shortfall in private consumption and investment. However the emphasis on the saving-investment balance as the central element in determining the growth in industrialised economies may not be appropriate for developing countries without developed manufacturing sectors and excess unskilled labour supplies working at near-subsistence wages. Here the basic problem is to increase the share of manufactured goods, and thus the retained share of technical progress, rather than to increase aggregate demand through higher investment. Indeed, building a domestic manufacturing sector requires increased investment by both the private and the public sector, but an increase in investment will be accompanied by an increase in imported capital goods, without any automatic mechanism leading to a compensating increase in export earnings. A Keynesian recommendation to increase income growth through increased investment to achieve potential growth rates in conditions of stable macroeconomic balance (X+I+G=M+S+T) would then be associated with an external disequilibrium (XT) and/or (I>S).27 Further, any attempt to increase exports to balance the deficit would also be associated with an increase in imports required by the investment to increase capacity. Thus, in the absence of a balanced increase in exports, the multiplier is neutralised by a one for one import leakage of demand abroad. And the increase in exports required to reverse this result is made more difficult if they are primarily composed of commodities subject to declining terms of trade.
The problem is aggravated if unequal distribution of income leads to a high proportion of imports of luxury goods. Prebisch was particularly concerned by the tendency for consumption patterns in Latin America to follow those in developed countries, increasing the demand for technologically advanced consumption goods.28 José Serra has noted that the dominance of international companies in the technologically advanced sectors of developing countries gives them control over the sectoral pattern of production and dominant presence in the most dynamic sectors. In particular they determine the pattern of domestic consumption with strategies aimed at new products and product differentiation whose sales depend on consumers with incomes close to developed country levels which can only result from a highly inequitable distribution of income. As a result, attempts to improve income distribution through higher wages are seen by these companies as producing an increase in their production costs without any impact on their sales29. Samir Amin in Accumulation on a World Scale also argued that the value of the multiplier in many African developing countries could be less than one due to the leakages of expenditures into luxury imports from developed countries.
Celso Furtado was among the first to note the inapplicability of Keynesian multiplier analysis to developing economies, noting that in an economy that depends on both imported capital goods and imported labour as in Brazil’s early stages of development, increased investment expenditures leaked abroad in the form of imports of slaves as well as through the import of equipment and building materials. His conclusion was that in such an economy there is no prospect for growth based on internal demand "giving rise to a self-generated developmental process. …the slave economy was dependent for all practical purposes on external demand".30 An equivalent argument holds when the labour force receives a subsistence wage since the secondary income effects of an increase in investment only produce the automatic balancing of saving and investment if saving is positive. With all saving thus left to entrepreneurs who spend on imported investment or consumption goods the multiplier can have little impact.
It thus appears as if the failure of investment to generate growth and the prevalence of the external constraint is due to lack of savings or excessive government expenditure when it is due to the absence of a domestic manufacturing sector and the associated inequality in the distribution of income created by the failure to appropriate the increased productivity in the form of increased domestic purchasing power.
Not only does such an approach place the difficulty on lack of domestic savings, it implies that it is the failure to finance this external imbalance with foreign resources means cutting back on the investment required for development of a domestic manufacturing sector and to achieve full resource utilisation. The result is the argument that developing countries lack domestic resources for development and require external resources in the midst of extensive unemployment of domestic resources – a paradoxical equivalent of Keynes’s complaint about poverty in the midst of plenty. But the problem is not a lack of resources, the problem is the structural imbalance in production represented by the absence of a developed manufacturing sector and an appropriate access to technology. The problem cannot be resolved by external financing in the absence of a plan to build domestic industry to substitute for the import of capital goods and restricting the import of consumption goods.
In this respect it is interesting to note that Prebisch was never an advocate of using external financing to pay the costs of domestic industrialisation. Indeed, in the Report to UNCTAD I the major recommendation for external finance is the creation of a compensation fund that would reimburse developing countries for the decline in the terms of trade suffered. In fact, he cites the difficulties that have been caused by private capital flows in creating what are now called negative net resources transfers. This is probably the result not of the similar difficulties that would eventually emerge from the failure of the Alliance for Progress to increase net flows from to the region, but from the experience of the region in the 1920s and after. In the UNCTAD II Conference he limits the call for external financing to developed countries meeting the UN goal of 1 per cent of GDP, implying a 0.3 per cent goal for private financing.
This is another area in which Prebisch departed from the dominant theory based on the idea that the lack of development was based on a lack of domestic savings that could be supplemented by international capital movements if restrictions on international investments were lifted. First, note that while such a position might have been justifiable in what are now called the "least developed countries", it would have conflicted with the reality of the major Latin American economies, many of which had already reached a degree of industrialisation sufficient to raise the question of the appropriate political development to support conditions required for them to reach the stage of developed democratic economies.31 It also raised the question of the international distribution of resources via international capital markets or through investments by transnational corporations. As noted above, foreign direct investment by foreign corporations had been the dominant form of delivery of external financing since the beginning of the process of externally led, trade based development in the early 19th century as well as in the period of development "from within" in the early 20th century. According to Victor Urquidi, "since World War II there has been a renewed inflow of foreign capital, not only in the traditional form of direct investment by foreign enterprises, but in the form of long- and medium term credits for importing machinery and equipment, granted by official and private foreign banks and by international agencies. It is estimated that in the twelve years from 1947 to 1959 there was a net inflow of both private and public foreign capital to Latin America of about $10 billion, of which nearly $7.5 billion was in the form of direct foreign investments and the rest were mainly official loans …. Of the $16.6 billion worth of direct foreign investment in Latin America in Latin America in 1959, it was calculated that more than $8,2 billion, or 60 percent, was United States capital"32
In this regard, the policies to attract transnational investment only serve to maintain the major difficulty faced by developing countries, the high import content of investment and exports; the problem is aggravated by the creation of tax free assembly zones in which imported semi-finished goods are imported free of duty to be assembled into final goods.33
There is another difficulty caused by the application of Keynesian theory designed to deal with developed countries to developing countries – the formulation of stabilisation policy. In traditional Keynesian theory inflation is the result of excess demand, usually a fiscal deficit that pushes domestic absorption above domestic productive capacity and rising prices. In conditions of exchange rate stability this leads to overvaluation of the currency, a deficit on external account and falling reserves. It can also be the result of nominal wages growing more rapidly than productivity causing domestic purchasing power to exceed productive capacity and bringing the same result with a more rapid rate of inflation. In short, both inflation and external disequilibrium are caused by a flow imbalance between income and output.
However, for Latin American developing countries, as seen above, an external deficit may be the result of a number of factors that have nothing to do with excessive domestic demand such as a decline in the terms of trade or investment increasing the import of capital goods or an increase in the cost of external debt service. Thus policies such as those used in the Washington Consensus may mistake the source of internal inflation and external disequilibrium as coming from excess demand and thus mistake the policies to eliminate them by means of reducing domestic incomes. First, consider a decline in the terms of trade. It is the result of relative price changes and market structure. Reducing domestic demand will have little impact on the external disequilibrium. Second, consider the impact of increased investment to build a domestic manufacturing sector. It is the result of the asymmetric domestic production structure and reducing demand will make it even more difficult to build domestic purchasing power and market size required to reap the benefits of increasing returns to scale on domestic real wages. Finally, consider the impact of rising international interest rates or appreciation of the currency of denomination of foreign borrowing. Again domestic demand has no impact on either of these factors. Thus, the use of expenditure reducing policies to remedy the deficiencies of the domestic production structure will simply make domestic incomes lower and the development of domestic industry harder.
But it also explains why Latin American developing countries so seldom achieve the full employment of their potential resources since domestic demand should always be restrictive, or at best neutral, according to Washington Consensus policies. It is important to note that these policies, employed under the conditionalities of structural adjustment policies by the IMF are based on the Keynesian adjustment models developed in the IMF research department to deal with balance of payments crises in developed countries and allow no other source of external disequilibrium than excess domestic income creation.
A second part of the Consensus that leads to mistaken policies because of the failure to recognise the specific conditions facing Latin American economies is the unilateral opening of the economy to foreign trade. This policy is recommended as a means of increasing domestic efficiency and productivity and thus as a substitute for domestic demand reduction to reduce inflation and excess demand. However, the theory that supports the benefits from opening an economy to trade is based on a comparative static equilibrium argument comparing an economy that is closed with one that is fully open, and presumes that the economy is on its production possibility curve before and after the change. None of these conditions apply in Latin America where trade has played a central role in the economy, even under import substitution industrialisation, and underemployment of resources is chronic.
In an economy assumed to be on a production possibility curve that reflects the same technology frontier available to all other economies with which it trades, with its consumers’ preferences the same as those of the country with which it trades and with constant returns to scale in the absence of factor mobility, the main message of the "factor price equalisation" theorem is that trade can provide a substitute for factor mobility. For developing countries the argument is expanded to the idea that trade can compensate for the lack of savings and capital resources. According to the theory by opening to trade a developing country with excess unskilled labour and insufficient capital resources will increase production and export of labour intensive goods, increasing the demand for unskilled labour and raising its wage. It will import capital intensive goods, reducing demand for domestic output and the skilled labour it requires, lowering its wage. Since there is more unskilled than skilled labour the end result will be an increase in average wages and incomes in the developing country. Comparative advantage trade can thus offset a developing country’s deficiency in capital, technology and skilled labour, improve income distribution and increase income per capita without the distortion that would be produced by the protection required for domestic industrialisation.
But both the composition of domestic output available for trade and the assumption of full employment are important to this result. First, as long as there are unemployed resources, using them to substitute for imports has no resource costs, and indeed may be a benefit if maintaining unemployed workers has a cost. It may be globally inefficient, but so is unemployment. Second, it is perhaps useful to remember that when 15th century European tourists set off on their voyages to the East (sometimes in the wrong direction) they had virtually nothing of value to trade with the civilisations they visited and thus had to rely on threat of violence and invasion to procure the spices they desired.34 – the early equivalent of forcing an improvement in the terms of trade. It is also important to recall the point initially made by Marx and noted by Galeano35 that virtually none of the non-traditional primary commodity exports from developing countries were not indigenous to those countries so their specialisaton in them could not have produced by comparative advantage.
Aside from the obvious failure of the assumptions to reflect the characteristics of developing countries, in particular to reach their production possibility frontier, most developing countries have discovered that both capital and labour are indeed mobile. And in particular that as a result of labour mobility through emigration what occurs in that both skilled and unskilled workers emigrate to take advantage of the higher wages in developed countries, but while this has little positive impact on the domestic wages of unskilled workers in developing countries it does prevent skilled workers remuneration from falling and preserves income inequality. The failure of more open trade to provide narrowing of differentials can be seen in the large and rising flows of emigrant remittances, which are largely the result of low skill emigration, and the large flows of foreign domestic investments in developing countries that are the result of expatriate investments, largely the result of high skilled emigration. This latter phenomenon has been especially important in India and China. At the same time it has little impact on skilled wages in developed countries, while it does reduce wages of the unskilled, again preserving income inequality both within and across countries.
Again, traditional neoclassical theory developed for industrial economies36 suggested that opening to trade can improve per capita incomes, increase productivity and reduce income inequality and provide a supplement to aggregate demand without the need for a special theory to introduce active policies to create an industrial sector or to rely on external resources to offset savings or resource gaps.
Opening to trade was also meant to provide for the efficient pricing of developing countries’ scarce resources. But, this argument only carries weight if international markets are competitively efficient. A number of analysts of globalisation have suggested that the opening of international markets has created a level international competitive playing field that will improve the conditions of both developed and developing countries. But there is no evidence to suggest that opening domestic markets to encompass international markets will improve pricing efficiency since the international markets for industrial goods are increasingly dominated by a small number of transnational corporations. It is well known since Coase that firms represent the antithesis of market relations, so that if globalisation is represented by an expansion of the operations of transnational firms across different national markets this may mean a reduction in the operation of the market in determining prices. The visible differences in pharmaceutical prices for products produced by the same multinational company across different national markets suggests that international markets are far from fulfilling the law of one price.
Although multilateral financial institutions continue to promote the unqualified benefits of more open international markets and freer trade, they have also extended this argument to mobility of factors, or better, the mobility of one specific factor, capital, perhaps not recognising that in the presence of the factor-price equalisation theorem this is redundant. Thus, developing countries have not only been encouraged to unilaterally open their goods markets, they have also been urged to open and deregulate their financial markets. Again the argument is based on comparative static efficiency: "The basic argument for international investment of capital is that under normal conditions it results in the movement of capital from countries in which its marginal value productivity is low to countries in which its marginal value productivity is high and that it thus tends toward an equalization of marginal value productivity of capital throughout the world and consequently toward a maximum contribution of the world’s capital resources to world production and income."37 Thus, the justification for the belief that globalisation should be beneficial for developing countries is based on the gains from trade and the optimal allocation of global resources through free global capital flows.
But, in the present period these flows have been primarily via foreign direct investment. The reduction in developing country trade restraints following the Uruguay round, along with the revolution in communications and transport technology has led to the global integration of production with transnational firms engaging in geographical diversification of the production process in order to place each stage of the production process in its lowest cost environment. This has led to a rapid increase in manufacturing output in many developing countries, as well as an increase in their exports of manufactured goods from "in bond" production zones. However, this has not had the beneficial impact on development that Prebisch had anticipated. The horizontal geographical division of production means that most processes are simply the assembly of imported semi-finished goods, leaving only the local wage content to increase domestic value added. Further, to attract these investments, best known as maquilladoras in Mexico, countries have to keep their wages low, thus making the export of embodied domestic unskilled labour the equivalent of primary commodity exports – and subject to declining terms of trade. The data presented in the 2002 Trade and Development Report38 for example, shows that while Mexico increased her share of global exports of manufactures by 2 full percentage points between 1980 and 1997, her share of global value added from manufactured goods exports fell by 7 tenths of a percentage point. This is the equivalent of the declining terms of trade for commodities, but now in globalised manufactured goods trade.
The impact of foreign investors was also to have brought greater access to technology and create Hirschman-type forward and backward linkages. However, there has been very little evidence that this is the case. Indeed, these effects were greater under import substitution since multinational firms were investing to produce to the local market behind protective barriers and thus had a greater incentive to build domestic supply and technical capacity.
Finally, the sharp recent rise in foreign direct investments has brought an additional aspect to the external constraint faced by developing countries in the form of potential repatriation of earnings. Although this has recently become a problem it is not a new one, for as already mentioned, there were substantial foreign direct investment flows to Latin America in the 19th century and then again during the period of import substitution industrialisation after the second world war. Neither were the risks unknown, "Another point to remember is that the country receiving foreign capital incurs debt that must be paid back with interest or dividends; therefore, it should accept such resources from abroad only if it can utilize them to advance its development and improve its future balance of payments. Foreign capital ought to make a positive contribution to the country’s growth and create the means, through increased exports or reduced imports, with which to amortize it and pay the interest on past borrowing.39 Evidence presented in the 2003 Trade and Development Report suggests that the recent experience of the major Latin American developing countries has been that foreign direct investment by transnational companies has increased imports and foreign borrowing and debt service more than it has increased earnings of foreign currency and has thus made the external constraint on growth more binding.
There Are Policy Alternatives – Examples
If we accept that there is a case for a theory of development, and that there is a case for a theory appropriate to Latin America, it remains to formulate alternative development policies. It is clear that this cannot be a simple return to domestic protection for import substitution industrialisation. Indeed, this would be inconsistent with the entire approach based on Prebisch’s work. As already suggested, it appears that Asia has been better able to adapt this approach to policy measures, but this does not necessarily provide the appropriate example for policies that take into consideration the particular conditions in Latin America. Further, a large number of policies that were applied in the Asian experience may no longer be viable in the post-Uruguay round trade environment. On the other hand, the negotiations under the Doha Agenda aiming to introduce a development dimension to the trading system may provide a new opportunity.
The first point that we have to propose is the necessity for a development policy and then to determine the priorities of that policy. It is also necessary to note the differences that have taken place since Prebisch was writing.
There are three fundamental characteristics of Latin American development. The first is the failure to convert technical progress into gains in real incomes. In difference from Prebisch’s time this point now applies not only to primary commodity production, but also to manufacturing. The second is the inequality in the distribution of income. This is also now the result of the way manufacturing has been introduced into many developing countries. The third is the underemployment of domestic resources that leads to the paradox of borrowing foreign resources when available domestic resources remain underutilised or not utilised at all.
Of course, these three factors are linked, for the deterioration in the terms of trade for commodity and manufactured exports drain domestic demand abroad, leading to the failure to provide market size sufficient to development domestic industry and fully employment domestic resources while the pressure to retain foreign investments puts downward pressure on domestic wages and purchasing power, leading to a further gap between wage and profit incomes. The result is a more binding external constraint that is met by policy actions to reduce domestic incomes, and in particular wages incomes, reinforcing the three characteristics. In a global financial system that is still subject to asymmetric adjustment international institutions also contribute to this result.
A new policy approach thus should start at home with policies to fully utilise domestic resources. When necessary this would also include an employer of last resort programme to provide for public works and public services. This would provide an fiscal policy that gives the automatic stabilisation that corresponds to Lerner’s fiscal federalism. This could be supplemented by the introduction of capital account budgetting for the government in which consumption expenditures are separated from capital projects when calculating current expenditure levels. Keynes’s original proposal was to keep the current budget in rough balance, ensuring that the provision of government services was not subsidised in order to prevent moral hazard, while the capital account would be used in a countercyclical manner. This culd easily be coordinated with the employer of last resort function. But this only refers to the expenditure side. On the tax side, Prebisch eventually came to accept Kaldor’s idea of expenditure tax40 with a highly progressive rate structure which could exceed 100 per cent at the maximum. This could be supplemented with a negative income tax as well as a luxury good sales tax.
Currently governments hold excessive foreign reserves, and these are potential resources available for development. The original purpose of the IMF was to provide for a pooling of reserve funds in a way that would not tie up resources. If the IMF no longer serves this purpose it should be reformed to do so, or countries should form regional institutions that do so.
If domestic resources are fully utilised, the rationale for external capital is greatly reduced. Since it is now clear that there is no market mechanism to ensure that the size and temporal distribution of these flows is compatible with sustained external equilibrium some regulation will be required. This should first be on a prudential basis to ensure the solidity of the domestic financial system, and secondly to fit the desired industrial structure. This will mean that there will have to be selection of foreign direct investment projects in accordance with the medium term national development plan. As Urquidi noted above, foreign investment must be structured so as to be self-financing in terms of its foreign debt service obligations.
Finally, as Eric Reinert41 has argued, no successful developed country has been able to bypass the "mandatory passage" of domestic industrialisation via the introduction of increasing returns industries and by keeping those productivity gains in the form of rising domestic real wages. Here the Asian experience is instructive in two respects. The first is that it explicitly rejected specialisation in trade based on comparative advantage and supported domestic industrialisation based on what Alice Amsden has terms "getting prices wrong".42 Second, it was based on a process of using a low wage base for the replication of existing developed country technologies in manufacturing based on domestic industry.43 Despite extremely low national savings, Korea was able to use its domestic financial system to fund these investments and generate income growth which eventually produced extremely high savings rates.44
Thus, we can conclude not only that Prebisch was right when he noted the need for a theory of development based on the reality of Latin America, and that we can build on that theory a set of policies that are alternatives without having to embrace the problems of import substitution. The important point made by Prebisch is that while the answer is to be found in domestic policies and actions, the international trading system needs to be revised to allow Latin America to earn its own development, not to borrow it from developed countries.
* Este artículo se encuentra bajo la licencia Creative Commons. Are There Alternative Development Strategies for Latin America?. En publicacion: Alternativas á globalização: pôtencias emergentes e os novos caminhos da modernidade. Jan Kregel UNESCO, Organización de las Naciones Unidas para la Educación, la Ciencia y la Cltura. 2005. Acceso al texto completo: http://bibliotecavirtual.clacso.org.ar/ar/libros/reggen/pp03.pdf
1. John Williamson, "Did the Washington Consensus Fail?" Outline of remarks at the Center for Strategic &monoeconomics" in his Essays in Trespassing, Cambridge, 1981.
3. John Coppock, a US State Department economist from the UN Oral History Project, cited in John Toye and Richard Toye, The UN and Global Political Economy, United Nations Intellectual History Project Bloomington: Indiana University Press, 2004 p. 26 and note 47.
4. This attack was made, not in an academic article, but in a political pamphlet, The Poverty of Development Economics, London: Institute of Economic Affairs, Hobart Paper, # 16.
5. See John Toye, Dilemmas of Development, 2nd edition, Oxford: Blackwell, 1993. He refers to a famous article by Harry Johnson heralding Friedman's theory as a "counter-revolution" against the Cambridge (UK) dominance of macroeconomic and monetary theory.
6. Cited in Toye, op. cit. p. 94.
7. Coparmex had 15,000 affiliated members in 1986. The businesses controlled by members employed 2 million workers – that is, 24 per cent of the full time private sector jobs in Mexico, cf. James Cypher, State and Capital in Mexico – Development Policy since 1940, Boulder, San Francisco and Oxford, Westview Press, 1990, p. 100.
8. See Rene Armand Dreifuss, 1964: A conquista do Estado – Ação, Poder e Golpe de Classe, Petropolis, Editora Voces, 1981. 5
9. See Jaime Fuchs and José Carlos Vélez, Argentina de rodillas, Tribuna Latina americana, Buenos Aires, 2002, pp. 102ff.
10. See Los costos del Estado Regulador ,Buenos Aires: Ediciones Manantial, 1988, produced by the Fundacion de Investigaciones Econmicas Latinoamericanas a business interest group provides a sector by sector analysis of the inefficiencies created by government regulation and interference in the economy.
11. This appears to be true not only of the Aztecs, but also in the precolonial period where government directed international trade has been identified as early as the Olmecs, cf. Ross Hassig, Comercio, tributo y trasporte-la economia politica del Valle de Mexico en el siglo XVI, Mexico: Alianza, 1990, p. 125, note 91.
12. Quoted in Eduardo Galeano, Open Veins of Latin America: Five Centuries of the Pillage of a Continent, New York: Monthly Review Press, 1973.
13. Victor Bulmer-Thomas, The Economic History of Latin America, Cambridge Latin American Studies, 1994.
14. Recognising this aspect of the declining termes of trade in Prebisch's approach highlights the fact that the similarities with Arghiri Emmanuel's Unequal Exchange London: New Left Review, 1972 are closer than Emmanuel's discussion of Prebisch allows. See also John Spraos, "Deteriorating Terms of Trade and Beyond," UNCTAD Review, No. 4 Winter, 1982, p. 102.
15. See quote in Gunnar Myrdal, The International Economy,New York: Harper, 1956, p. 227.
16. The point is also made by Nurkse, "To push exports of primary commodities in the face of an inelastic and more or less stationary demand would not be a promising line of long-run development." See Problems of Capital Formation in Under Developed Countries, Oxford: Blackwell, 1953, pp. 101. See also P. Rosnstein-Rodan, "Problems of Industrialisation of Eastern and South Eastern Europe," Economic Journal, June-September, 1943.
17. See Raul Prebisch, La crisis del desarrollo argentino: de la frustracion al crecimiento vigoroso, Buenos Aires, El Ateneo, 1986, p. 50.
18. "Note sur la notion de 'pôle de croissance' in Matériaux pour une analyse e la croissance économique, Cahiers de l'Institute de Science Economique Appliquée, Série D. No. 8, 1955.
19. See Chapter 10 of his The Strategy of Economic Development, New Haven: Yale University Press, 1958.
20. Gunnar Myrdal, Economic Theory and Underdeveloped Regions, London, 1957.
21. Compare Paolo Sylos Labini, Oligopoly and Technical Progress, Harvard Economic Studies, 1962.
22. Nicholas Kaldor, in "Stabilising the Terms of Trade of Under-developed Countries", written for ECLA and published in the Economic Bulletin for Latin America in March 1963, puts the point simply: "Sellers of primary commodities suffer from two important handicaps … first … primary producers are "price takers". A fall in demand for manufactured goods leads directly to a reduction of output: any reduction in prices occurs only indirectly and incidentally, depending on the extent to which producers are induced to lower profit margins. A fall in demand for primary commodities, on the other hand, leads directly to a fall in prices; it leads to a restriction in output only indirectly, in so far as the decline in prices causes producers to lower their output. The second handicap it that, whereas the benefits of technical progress in manufacturing are largely retained by the producers (in the form of higher real wages and profits), the benefits of technical progress in the primary production are largely passed on to the consumers, in the form of lower prices, leaving little benefit to the producers in the form of higher real income." A note at this point adds, "This is partly due to the prevalence of imperfect competition in manufacturing and partly to the fact that organisations of industrial workers are in a position -unlike workers in the agricultural sectors of under-developed countries- to ensure that industrial wages increase at least as fast as industrial productivity." (Reprinted in N. Kaldor, Essays on Economic Policy – Volume Two, New York: Norton, 1964, pp.121-2.) Those familiar with the discussions between Richard Kahn, Piero Sraffa and Keynes over the former’s Fellowship Dissertation will recognize a similarity in the argument concerning the reaction of firms to declining output.
23. Raul Prebisch, The Economic Development of Latin America and its Principal Problems, UNECLA, 1950, p. 1.
24. See Raul Prebisch, La crisis del desarrollo argentino: de la frustracion al crecimiento vogoroso, Buenos Aires, El Ateneo, 1986, p. 101 .
25. See Alice Amsden, "La sustitucion de importaciones en las industrias de alta tecnologia: Prebisch renace en Asia"Revista de la CEPAL, No. 82, Abril 2004, pp. 75-90.
26. See Hirschman, op. cit., chapter 2, pp. 29 ff. "Growth Models-Help or Hindrance?" and Keith Griffin, Underdevelopment in Spanish America, Cambridge: MIT Press, 1969, p. 179 ff.
27. This has led to the argument that the external imbalance is either due to deficient saving or to excessive government expenditure, rather than being a structural characteristic of an economy without a developed manufacturing sector.
28. The international dissemination of patterns of consumption of technically advanced goods from developed to developing countries before the latter have the capabilities to supply the goods domestically is one of the reasons that Prebisch gives for building on externally imported technology and capital goods. The process might be done internally but that would require 250 years and exclusion from the rest of the world.
29. See J. Serra, "O desenvolvimento da América Latina – Notas Introdutórias" in América Latina – Ensaios de interpretação econômica, Rio de Janiero: Paz e terra, 1976, pp. 21-22.
30. See Celso Furtado, The Economic Growth of Brazil, Berkeley, University of California Press, 1963, p. 52-7. 15
31. The "Introduction" to F.H. Cardoso and E. Faletto's Dependencia y Desarrollo en America Latina (Mexico City: Siglo Veintuno Editores, 1969, p. 3) that raised the political dimension of development opens by noting that "At the end of the second world war some Latin American countries appeared to have reached a level of industrialisation capable of providing the economic transformation required for self-sustaining development." The problem that remained was the political "strengthening and modernisation of the State" "to organise national decision-makers in such as way as to make then concentrate on the development problems of their countries" (p. 6). According to José Serra, (op. cit. p. 16) after the Second World War "high rates of industrial growth were accompanied by a considerable diversification of the production structure" with the major countries in the region introducing "an industrial complex based on metal mechanic and electrical material producing machine tools and equipment, transport and durable consumption goods".
32. Victor Urquidi, The Challenge of Development in Latin America, New York" Praeger Paperbacks, 1962 (1965) p. 47. Urquidi bases these estimates on statistical work done by ECLA.
33.The impact of trade liberalisation on the share of imports required to reach a given growth rate is more fully discussed in UNCTAD Trade and Development Report 1999, Part II.
34. Cf. Griffin, op. cit., p. 33-5. He notes the words of the Chinese Emperor that "our celestial empire possesses all things in prolific abundance" and thus had no need to exchange tea for English exports. The problem was resolved by forcing the Chinese to demand opium from British India.
35. Eduardo Galeano, Open Veins of Latin America: Five Centuries of the Pillage of a Continent,
36. Although it does not provide a good description of trade in these economies either, given that most of the increased trade in developed countries has been intra-industry trade rather than increased specialisation.
37. See Jacob Viner, "International Finance in the postwar World," Journal of Political Economy, 55, April, 1947, p. 98.
38. Table 3.5, p. 81
39. See Urquidi, op. cit., p. 45.
40. Prebisch, 1986, op. cit., pp. 90-91.
41. See. "How Rich nations got Rich: Essays in the History of Economic Policy," University of Oslo SUM Working Paper 2004.01.
42. See Alice H. Amsden, Asia's Next Giant: South Korea and Late Industrialisation, New York: Oxford University Press, 1989 25
43. She notes that unlike the UK, Germany or the US, Korea did not, indeed could not, build on the basis of technical progress produced by entrepreneurs developing and introducing industrial innovations. Rather its success was based on its ability to organise the reproduction and adaptation to local conditions of methods of production already in place in the developed countries. This is what she calls industrialisation by "learning".
44. Indeed, Korea started its industrialisation process with extremely low savings rates (in 1962 personal savings were negative), but that the success of its development process provided a rapid increase in personal savings to over 10 per cent by the 1970s. Total savings rose from 12 per cent of GDP in 1962 to 28 percent by 1984 SeeAmsden, op. cit. 1989, p. 75).
Jan Kregel