1915-1991
William Arthur Lewis was quintessentially Caribbean. Born on January 23, 1915, in colonial St. Lucia of immigrant Antiguan parents who were both schoolteachers, he spent much of his life living in Britain or the United States but working on the problems of the Caribbean, Africa, Latin America, and Asia. In 1933, on a government scholarship, Lewis went to the London School of Economics (LSE) to do the Bachelor of Commerce degree. He graduated with first-class honors in 1937 and received a scholarship to do the PhD in economics, which he completed in 1940. In 1979, Lewis shared the Nobel Prize in economics with Theodore Schultz for pioneering research into economic development, in particular the problems of developing countries.
Lewis’s great achievements were in economics, but he also had a distinguished career as an administrator. He was UN Economic Adviser to Kwame Nkrumah, the prime minister of Ghana, deputy managing director of the UN Special Fund, vice-chancellor of the University of the West Indies (1960-1963), and founder of the Caribbean Development Bank (1970-1974). By his account, this experience taught him more about policy than economics and motivated his publications on development planning. Scholarly enrichment came from his travels (Lewis 1994, p. xlviii).
Lewis published extensively on industrial economics after 1937, then published Overhead Costs in 1948 and dropped the subject. In 1948, at age thirty-three, Lewis became a full professor at the University of Manchester. There, he tackled development economics, addressing: (1) the fundamental forces determining the rates of growth of agricultural and industrial countries or sectors; (2) the relative price of agricultural and industrial products; and (3) distribution—the adjustment of the real wage rate and wage share of output as capital accumulates. As Lewis tells it, one day in August 1952, on the road in Bangkok, he saw the common solution to these problems: Use the classical assumption of "an unlimited supply of labor" available to a capitalist sector from an indigenous noncapitalist sector "at subsistence wages" (Lewis 1992). Unlimited supplies keep the wage low and fairly constant and generate a high rate of profit and a low relative price of agricultural output. They also allow creation of new capital with technological progress, a rising rate of profit and savings, and expanded employment and output in the capitalist sector without raising consumption per worker. This initially causes faster growth in industrial countries because they reap all the benefits of technological progress. However, as industrial countries face rising wages, they export capital and encourage immigration, thereby allowing the growth rate of surplus labor countries to catch up. How much growth these capital exports generate for surplus labor countries depends partly on their natural resources, capital stock, demand in industrial countries, and the terms of trade. Lewis published this solution in his classic 1954 article, "Economic Development with Unlimited Supplies of Labour."
In 1963, Lewis took up a new professorship of political economy at Princeton University, a position he held until his death on June 15, 1991. At Princeton, Lewis used the growth channel of his "unlimited supplies" model to clarify the evolution of the international economic order since 1870. Trade is the principal link through which growth of industrial countries causes growth of the labor-surplus countries. Moderated growth in industrial countries causes slower growth of surplus-labor countries, unless an alternative growth engine is found. One option is a greater share of the markets of industrial countries, but this is unreliable. Preferable is growing trade among surplus-labor countries. This alternative trade engine could arise if a sufficient number of the large surplus-labor countries, such as India and Brazil, can end trade dependence on industrial countries, achieve self-sustaining growth, and provide alternative markets to the others. Lewis published these findings in his book The Evolution of the International Economic Order (1978) and his 1980 Nobel Lecture, "The Slowing Down of the Engine of Growth." By his logic, the current explosion of growth in excess of 7 percent per annum in Brazil, China, and India should provide much stimulus for the development of other surplus labor economies.
Lewis’s Bangkok inspiration was rooted more firmly in influences of economic historians of the British nonrevolutionary socialist Fabian Society. Leading earlier members Barbara and J. L. Hammond (1917) had argued that the English industrial revolution was possible mainly because the emerging bourgeoisie could attract a large redundant rural labor force into urban industrial activity at low and constant wage rates. Lewis’s firsthand knowledge of the Caribbean and observations of Egypt and Asia convinced him that a similar exodus could also underwrite an industrial revolution in those and similar countries where many nonindustrial subsistence workers contribute a zero or negative marginal product. However, an attendant condition was active trade unions and an emerging class of entrepreneurs or an active government operating democratically and dedicated to making and accumulating capital. Lewis (1949) summarized the main arguments in a Fabian pamphlet, foreshadowing the arguments of the 1954 paper.
From the start, economists questioned the updated classical assumption, the assumption that surplus labor only creates new capacity within the capitalist sector, along with Lewis’s pessimism about the ability of surplus-labor countries to win a greater share of developed-coun-try markets. Schultz (1964) used neoclassical data, competitive conditions, the assumption of rationality, and evidence from Egypt and Asia to argue that the marginal product of labor is neither zero nor negative. With transformation of traditional agriculture, rural wages could fall low enough relative to the marginal product in agriculture to enable full employment. Lewis’s views amounted to assuming either an uncompetitive market or a distorted preference for leisure among agricultural workers.
Lewis himself viewed peasants as hardworking and thrifty and held that the main problem was a capitalist sector that is too small or disinclined to transform agriculture by accumulating capital (Lewis 1936, 1949, 1954, 1955, 1968, 1979). Amartya K. Sen (1966) also objected to Schultz, citing contrary evidence for India and observing that Lewis’s assumption really means that the marginal product in the capitalist sector is greater than the average product of the subsistence sector. Nevertheless, Lewis’s essential difference with Schultz on the capitalist wage is that its floor is set by the average productivity of the self-employed.Subsistence workers were household and community members who could not be easily dismissed without damaging established family and social norms. They could opt for the average benefits of the family operation rather than a lower capitalist real wage and thereby cause a breakdown of Adam Smith’s law of demand.
Perceptive critiques of Lewis also came from the Caribbean. In three major proposals on Caribbean industrialization, Lewis proposed attracting multinational corporations to lead the process (Lewis 1950, 1951; UN 1951). Lloyd Best lampooned the strategy, naming it "industrialisation by invitation" and calling Lewis an "Afro-Saxon" (Best 1999). Multinational corporations would continue to import labor-displacing capital, re-create surplus labor, repatriate high rents, avoid the push for good governance, and promote repression of radical labor and neglect of the peasantry, as Lewis himself had complained (Lewis 1936). More important, the subsistence sector included viable entrepreneurs and collaborators producing and accumulating capacity and trading with the capitalist sector, not as pain and trouble but for the pleasure of building an alternative to the legacies of slavery and strict indenture. The market for labor was tighter than Lewis implied and rising real wages were inevitable (Best 1975). Once import substitution started in the Caribbean, this became evident, but Lewis found it perplexing (Lewis 1958, 1968).
Yet Lewis’s formulations can accommodate the Caribbean critique. His geometry of the marginal product of labor features diminishing returns to capital scarcity and an implicit production function that may not be algebraic and is dynamic, taking into account net capital accumulation, technical progress, increasing returns, and an underlying multiplier linked to the capitalist employment rate and the ratio of the marginal product of labor to its maximum. It allows residual profit at a rate different from the marginal product of capital and an independent distribution identity, and can be updated to account for rents on imported capital and production and accumulation of domestic capacity. With deliberate domestic capacity creation and supporting policy for their indigenous sectors, abundant externalities and stabilized rentals on imported capacity can replace Lewis’s cheap constant wage as the driver of the savings mechanism, and surplus-labor countries need not be pessimistic about winning higher market share in industrial countries over time in the context of rising real wages. These adjustments would still be consistent with Lewis’s view of how to achieve balanced growth, clarified in his 1949 note on "Colonial Development" in United Nations (1951) and in Lewis (1954, 1964). So, Lewis’s independent marginal product function and his associated price, distribution, and growth theories represented major steps in clarifying how surplus-labor countries grow and the wider world economy evolves. They hold as limiting cases, providing an updated classical alternative to the increasingly dominant neoclassical theory of growth and trade, and they still apply to many countries of the world today. Viewed in this light and his time, Arthur Lewis was among the best of his generation and one of the greatest economists of all time.
