Avi BRAM (2016), London School of Economics and Political Sciences, Msc Political Economy of Late Development
Development economics emerged as an applied field of economics (in Hirschman’s words, a ‘sub-discipline’, Hirschman, 1981) in the late 40’s and early 1950’s. Its aim was to bring rapid development – defined as increase in per capita output – to the undeveloped parts of the world, many of which had recently become or were becoming independent after years of colonisation. However, by the 1980s most of the developing world still lagged far behind the Most Developed Countries (MDCs) on a range of economic indicators. This essay will argue that, although over-ambitious and simplistic in its approach, the Early Development Economics (EDE) was not the failure that has been claimed by its critics. The reason it fell out of favour was as much due to changes in the wider context as to the success or failure of its policies.
EDE diverged from the ‘orthodox’ neo-classical economics of the time, positing that the rules of economics applicable to MDCs did not apply entirely in Less Developed Countries (LDCs). Furthermore, EDE focused on dynamism and transformation in the economy (i.e. how a country could change the industries and sectors that it specialised in), as opposed to the static, marginalist approach that dominated neo-classical economics. EDE was not an entirely novel departure: it took from Keynesianism the idea that different laws of economics might apply in different situations, and from the early classical economists the focus on transformation and structural change (Adam Smith’s Enquiry into the Wealth of Nations, though a foundational text for the neo-classicists, was after all a historical inquiry into how some nations became more developed and productive).
Although the EDE sub-discipline was diverse and multi-faceted in its approach, two ideas were particularly important it marking out its approach: rural underemployment and late industrialisation (Hirschman, 1981). Rural underemployment was believed to be a common feature of LDCs: in fact, Arthur Lewis reckoned that in some cases the marginal product of labour in traditional agricultural is zero. Late industrialisation refers to the specific challenges of industrialising when there are already-industrialised countries exporting industrial goods to the LDCs. Theories of comparative advantage posit that the LDC will not be able to compete in industrial goods as the industrialised countries have an unassailable advantage in capital, know-how and infrastructure: the laws of the market dictate that the LDC will not adopt modern industry. However, the EDE theorists believed that a planned, guided effort (usually state-guided) could overcome this disadvantage.
What’s more, they believed that producing modern industry was a crucial step for LDCs to become developed: industry required skills, linkages and technology, but once operational it provided momentum for further learning, investment, and the establishment of related industries. EDE advocated heterodox economic policies, often interventionist (distorting market signals): such as tariffs, quotas, credit and foreign exchange rationing. The following sections consider the ideas of EDE in more depth.
Amartya Sen points out that although the diagnosis of rural underemployment has been challenged, it is often correct. Studies have shown marginal product in agriculture is very low for many LDCs (Sen, 1983) and therefore there is great potential for growth from ‘mobilising’ rural workers to move to industry jobs (as occurred in China and South Korea).
A separate critique of the rural underemployment thesis is that it lead to incomplete policies: rural workers were mobilised to the urban areas, but the ‘modern’, capital-intensive industries advocated by the EDE practitioners could not provide enough jobs to absorb them. Harris and Todaro viewed unemployment in LDC cities as a sign that modern industry was paying inappropriately high wages (due to minimum wage laws or other interventions), and was attracting rural newcomers who were then assigned to unemployment. In their model, newcomers keep arriving until the expected wage in the city is the same as in the countryside – i.e. citydwellers face periods of unemployment when their income is lower than agricultural workers, and are only just compensated for it by periods of employment.
However, empirical studies have challenged this view: for example, a study of Jamaica in the 1950s found that the ‘unemployed’ in the cities were largely engaging in casual work, family income-sharing or informal work, and had a standard of living higher than agricultural workers. The large informal sector found in most LDC cities is typically “economically efficient and profit-making, though small in scale and limited by simple technologies” according to an ILO survey. In East Asian countries, formal and informal employers often pay similar wages. These studies suggest that the rural mobilisation advocated by EDE did raise productivity and encourage sectoral transformation in LDCs, albeit not to the extent that early practitioners envisaged.
EDE policies often focused on capital accumulation. One reason was the ‘low-savings trap’: it was believed that countries where most of the population had subsistence-level income could not save much, and therefore couldn’t invest in modern industry. A UN report in 1951 noted that in fast-growing countries net capital formation was “at least 10% of national income”, but in most under-developed countries it was stuck below 5% (Meier, 1984). To remedy this ‘savings gap’, aid agencies and multi-lateral institutions injected large amounts of investment capital into LDCs. A direct relationship between capital and growth was assumed, often using the Harrod-Domar model as an analytical justification (even though this model was devised as a model of short-run business cycles in developed countries, not long-run growth).
This approach has been criticised for producing white elephants (such as dams and factories that were subsequently abandoned). Bill Easterly finds that there is no relationship between investment and subsequent growth but his approach is not very robust: he looks at investment in a 4-year period and growth in the following 4-year period, and then switches over to within-country relationships between investment in a given year and growth the following year (Easterly, 2002). This choice of time periods is arbitrary and may well be too short-term to capture the effects of investment. On the other hand, there is some evidence that investment is associated with growth – Sen notes that the fastest growing low- and middle-income countries in 1960-1980 were those with the highest share of GDP in investment, and those with the lowest share were the slowest growing (Sen, 1983). The share of GDP measure comes from 1980, which is the end of the period in question, which means that it cannot be used to show that investment caused growth. But at least there remains some theoretical and empirical support for investment being a necessary condition for growth.
EDE advocated using state power to intervene in the market, in order to direct investment or promote the growth of certain sectors. Neo-classical economists objected to measures which they considered to distort price signals and lead to inefficient allocation of resources. Deepak Lal for example, argued that central planners lack the information to make investment decisions better than thousands of decentralised agents (i.e. consumers and entrepreneurs). He believed that LDC governments especially were often inept or corrupt and therefore would be unlikely to implement controls in such a way that would benefit the economy.
However, this critique ignores the fact that a central planner (i.e. the state) need not know more about the economy than all other economic actors in aggregate in order to direct the economy in a beneficial way. Owing to the problem of collective action and the presence of externalities, government action may be required to move between market equilibria. Singer’s thesis gives one example of this: without government action an LDC may specialise in raw material extraction, where it has a comparative advantage. However, a long-term movement in the terms of trade against raw materials will leave producers in the LDC locked-in to a ‘bad equilibrium’. Even if terms of trade do not worsen, raw material producers have a low-value product which will leave them worse off than producers of high-value, industrial goods in MDCs. Market forces alone may not provide sufficient direction for LDC entrepreneurs to invest: when raw material prices are high then there is little incentive to invest in industrialisation, and when raw material prices are low then there may be an incentive to diversify into industry, but there is very little surplus (from commodity sales) available to invest.
The example of the East Asia ‘miracle’ economies (Singapore, South Korea, Hong Kong and Taiwan) is instructive. Cristobal Kay’s paper comparing East Asia to Latin America shows that both regions employed EDE approaches. South Korea, for example, used state procurement to keep food prices low, capturing agricultural surplus for investment in industry and encouraging rural-dwellers to move to urban areas. It received large quantities of aid from abroad (especially the USA) and later, foreign investment. The state intervened in credit and foreign exchange allocation, and imposed tariffs in order to encourage key industries that were unable to compete freely on world markets. However, the East Asian approach to industrialisation was export-led, unlike the Latin American ‘import-substitution’ approach which concentrated on the home market. South Korea and Taiwan also had a radical land reform imposed on them after 1945, which meant most of the rural land became self-owned farms. As well as improving equality and work incentives in rural areas, this removed the class of landlords that might have blocked future reforms. The state invested in modernising agriculture, so that productivity rose in agriculture as well as in industry. In contrast, Latin American land reform came after industrialisation, where it occurred at all, and was often resisted by large landowners. Kay’s conclusion is that the competence of the state, land reform, and interactions between agricultural and industrial policy all matter as to whether EDE policies succeeded.
The wider political and ideological context
EDE thinking was challenged in the late 70s from the Right, by the neo-liberal movement and from the Left, by the dependency theorists who argued that development policy served only to embed LDCs in economic subjugation to the developed countries. Although development economists were refining their approach to address these critiques, they were fatally weakened by the Third World Debt crisis. As LDCs were defaulting on their debts and their industries suffered, it gave opponents of EDE the leverage to impose a radical change of approach: the structural adjustment plans of the 80s and 90s. In fact, crisis owed much to factors unrelated to EDE policies. The rise in oil prices in the 70s forced LDC oil-importers to borrow money in order to maintain imports at their previous level, which meant that repayments became unsustainable when world interest rates rose in the 80s.
Did Early Development Economics fail?
This essay has shown that there were some flaws in the thinking of the early development economists: the faith placed in capital accumulation without careful consideration of how this capital could be best used. A second failing was to treat LDCs as having essentially similar economies, that could be guided into rapid economic growth by a set of similar policies imposed by outside experts. However, overall there were many aspects of their analysis that were borne out, such as understanding of the dual economy, the opportunities for growth afforded by industrialisation, and the need for state guidance in order to accelerate development. Most LDCs experienced reasonable economic growth in 1960-1980, and the experiences of East Asia showed that competent governments implementing locally-suitable policies could achieve a lot. The fact that developmentalist policies fell out of fashion in the 1980s owes as much to external factors as it does to any inherent flaws.
Easterly, William, The Elusive Quest for Growth (MIT Press, 2002)
Kay, Cristobal, “Why East Asia overtook Latin America: agrarian reform, industrialisation and development”, Third World Quarterly, vol. 23, 6 (2002)
Hirschman, Albert O., “The rise and decline of development economics” in Essays in Trespassing (1981)
Lal, Deepak, “The misconceptions of ‘development economics’”, Finance and Development, vol. 22, 2 (June 1985)
Meier & Rauch, Leading Issues in Economic Development, 7th edition (OUP, 2000)
Meier & Seers (eds), Pioneers in Development (1984)
Ray, Debraj, Development Economics (Princeton University Press, 1998), ch. 10
Rostow, Walt Whitman, The Stages of Economic Growth: a non-Communist Manifesto (1960)
Sen, Amartya, “Development: Which Way Now?”, Economic Journal, vol 93, 372 (1983)
Singer, H., “The distribution of gains between investing and borrowing countries”, American Economic Review, vol. 40, 2 (1950)