Ferdinando Targetti [1]
Economic Development and Change, 2005.

1. Introduction

Nicholas Kaldor made seminal contributions to many fields of modern economic theory (for an overall view see Targetti, 1992; Thirlwall, 1987), but often these have not been properly acknowledged. It is surprising how many crucial elements of modern theories have roots in or are consistent with ideas that Kaldor developed in seminal papers. The ‘learning by doing’ ideas that underlie the technical progress function of Kaldor’s models of growth (see Kaldor, 1957; 1961) are one of the origins of the theory of endogenous growth; Kaldor’s ‘A Model of the Trade Cycle’ (1940), the first cyclical model based on non-linear functions of saving and investment, has been amply developed in the subsequent literature, including in the 1970s when it was applied to modern theories of catastrophes; and Kaldor’s work on regional and national divergences (see Kaldor, 1970, 1981) anticipated the late 1980s literature on international trade based on increasing returns. Moreover, Kaldor’s theory of endogenous money supply has influenced modern neo-Keynesian macroeconomics.

Also Kaldor’s intellectual contribution to the literature on economic development has not been properly appreciated, despite the fact that this work was ample and covered several fields. Although the term globalization was not yet in use when he died, many contemporary issues can be illuminated by Kaldor’s ideas on growth, development, and international trade. This review of Kaldor’s intellectual legacy seeks to redress these oversights and to highlight his contribution—a contribution which appears to be consistent with the present reaction to neo-liberal and neo-classical thought on economic development.

Kaldor’s contributions to development economics have two origins: his writings on taxation and on growth. During the 1930s, Kaldor was already a respected economist for his theoretical work on capital, speculation, and the trade cycle. In the 1950s, his long-lasting interest on growth theory developed. Following his appointment by British Chancellor of the Exchequer Hugh Gaitskell to the Royal Commission on Taxation and Profits in 1951; Kaldor disagreed with the Commission’s conclusions and wrote An Expenditure Tax (1955), the book which made him famous as an expert on taxation. From 1956, when he became an advisor to the Indian government (see Kaldor, 1959), Kaldor spent twenty years as a taxation advisor to a number of developing countries, and this led him to be deeply involved in development issues. As far as growth theory is concerned, from 1966 Kaldor’s interests shifted from the pure theory of growth to the development of ‘growth laws’—hypothetical explanations of ‘stylized facts’.

Kaldor’s analysis of development hinges on four fundamental concepts: (1) increasing returns in the manufacturing sector; (2) effective demand-constrained growth; (3) the agriculture-industry relationship; and (4) internal-external market relations. In terms of development policies, Kaldor believed that: (1) economic development requires industrialization; (2) this in turn presupposes an ‘agriculture revolution’; (3) entering into the global market with a temporary stage of protection for newly established industries; (4) this must be accompanied by export-led growth policies.

This paper is developed as follows: part 2 illustrates the Law of Manufacturing as the Engine of Growth and the connected Law of Increasing Returns in the Manufacturing Sector; parts 3 and 4 illustrate Kaldor’s Agriculture-Industry Growth Model, which in his analysis performs the double task of explaining the importance of the complementarity of agriculture for the manufacturing sector and the stagflation bias of the world economy; part 5 illustrates the so-called Kaldor-Thirlwall Law, which determines the maximum rate of growth of an open economy; part 6 shows how, on the basis of these laws, Kaldor argued that countries can be locked into vicious and virtuous circles; part 7 attempts to integrate Kaldor’s explanation of the falling-behind forces with that of the catching-up forces which are also working in the global economy; parts 8 and 9 contrast the policies for development based on the orthodox paradigm with those emerging from Kaldor’s analysis and provides an evaluation in the light of the recent literature on globalization.

2. The First Two Laws: Manufacturing and Increasing Returns

Kaldor developed four ‘empirical laws’: the first is the Law of Manufacturing Sector as the Engine of Growth’. Maddison (2005) has recently argued that modernity did not start with the Industrial Revolution, which at the beginning of nineteenth century was the engine of what Rostow calls the economic ‘take off’. For Maddison, the roots of modernity should instead be located in the long apprenticeship during the previous three centuries when at least two revolutions occurred: the first initiated by improvements in shipping, with the consequent expansion of international trade, and the second as a result of Gutenberg’s invention, which allowed much more rapid diffusion of scientific knowledge. Nevertheless, as he expressed it in three lectures at Cornell University, Kaldor (1967) thought that before 1750 there were no great differences in the income and income per capita among regions and countries, while after that period a group of countries in Western Europe experienced an annual or decennial rate of growth which was thirty or forty times higher than they had achieved in the previous seventeen centuries. He sought the explanation in the development of the manufacturing sector. He found evidence that manufacturing industry is the engine of growth for every country at every stage of growth. Consequently, in his inaugural lecture at the University of Cambridge, Kaldor (1966) argued that the decline (at that time) of the British economy can be explained in terms of a premature decline in the manufacturing sector (he would not have expected that his country could have achieved the high rate of growth which it has recently experienced through the development of the financial and service sectors).

The second law, called Kaldor-Verdoorn’s Law, provides an explanation of the first law: the more the output of the manufacturing sector grows the greater is the increase of productivity in the system as a whole. This law takes the form of a relation between the growth of productivity p and the growth of manufacturing output g: p = a + b g, with parameter a having a value of around 1 and parameter b having a value of around 0.5. This implies that every increase in manufacturing output of one per cent gives rise to an increase in productivity of one-and-a-half per cent. The explanation for this relies on several factors that are external and internal to the sector. There are three external reasons: first, because the growth of manufacturing provides capital goods and hence technical advances embodied in them as input for other sectors; second, because an increase in output and employment in the manufacturing sector reduces the employment in agriculture but not its output; third, because greater activity in the manufacturing sector produces greater turnover per worker in the distribution sector. Internally, the explanation of the second law is provided by the existence in the manufacturing sector—and only in that sector—of static and dynamic returns to scale. The economies of scale that Kaldor has in mind stem from the discovery of new processes, from increasing differentiation, and from new subsidiary industries—that is, from general industrial expansion and not just from the expansion of a particular firm or industry. Nowadays, this process has been complicated by the development of information technology, which has affected both the manufacturing and service sectors; however even in this hybrid sector, the economies of scale, taking the form of the net economy, play a role of utmost importance. Increasing returns to scale, which Kaldor learned about from Adam Smith and Allyn Young, played a central role in his growth analysis long before the principle was rediscovered by the modern theory of endogenous growth. [2]

       The first two laws imply that capital accumulation is self-generating as output increases. The limit on growth of the capital good sector (the manufacturing sector) has consequently not to be found in some supply constraint, not even in the shortage of labour (which was his original idea in 1966), but in some demand constraint. In other words, the growth of industrial output must be induced by autonomous demand, which derives from outside the sector, either from the agriculture sector (the agriculture-industry model which I call the Kaldor’s Third Law of Growth) or from the rest of the world (the Kaldor-Thirlwall Law of Growth).

3. The Third Law, the Industry-Agriculture Relation, and the Two Sector Model

The third law refers to the assumption of disguised unemployment in the economy (at the early stages in agriculture and subsequently in services), which together with the hypothesis of rigid wages in the industrial sector exceeding agriculture wages leads to an elastic supply of labour for industry. The reproduction of the industrial labour force depends, à la Marx, upon the real purchasing power of wages in terms of agriculture goods, which in turn depends upon the agriculture techniques of production. In the latter sector, decreasing returns prevail and output can be raised only by an exogenous progress (technical or social as it derives from land reforms). This agriculture revolution has taken a variety of forms: it occurred in England because the landowners expropriated the peasants; in France because the peasants expropriated the landlords; in Japan because the Meiji revolution imposed a fiscal system capable of raising large sums in direct taxes imposed on the agricultural sector (plus an extensive rural banking-credit system that was able to harness small savings from this sector in the heavy industry for investment purposes, something that even today is often lacking in less-developed countries). South-East Asia and Latin America, which did not undertake fully-fledged agrarian revolutions (or were not allowed to undertake such a revolution because as colonies had been kept as agrarian economies or transformed into a plantation economy), were hampered in their industrial revolutions. In Kaldor’s opinion, the connection between agriculture and industry emerges not only because, as in the Lewis model, the agriculture revolution creates the extra wage goods for the growing urban population, but also because it creates the conditions for autonomous demand in the manufacturing sector.

The development of the agriculture-industry relationship was examined by Kaldor over a decade, beginning with his 1975 paper ‘What is Wrong with Economic Theory?’ and concluding with his 1984 Raffaele Mattioli Lecture in Milan (Kaldor, 1995). His insights have been formalised by others in two-sector models (Targetti, 1985; Thirlwall, 1986). The two sectors are manufacturing, which produces investment goods, and agriculture which produces consumption goods. The rate of growth of agriculture depends on the volume of investment which is obtained in exchange for a given quantity of agriculture goods sold to the manufacturing sector; that is to say, the curve representing the growth of agriculture, ga, is downward sloping in relation to the terms of trade (price of investment divided by the price of consumption); and vice versa, the curve representing the rate of growth of the manufacturing sector, gi, is upward sloping with these terms of trade. The equilibrium growth rates of the two sectors g* is given by the intersection of the two growth curves, and it occurs at a certain terms of trade p*. Every sector is constrained by the demand of the other. If the ga curve shifts to the left because of decreasing returns to scale the overall rate of growth will decrease and the terms of trade of agriculture tend to improve relative to industry (Ricardo-effect); if the curve shifts to the right because of land-saving technical progress the terms of trade tend to worsen (Prebish-effect) and the overall rate of growth improves. The same happens if the gi curves shifts to the right because of the growth of productivity in industry.

4. The World Economy

The two-sector model describes the case of a single developing country, but Kaldor used it to describe the trade between less-developed countries exporting agricultural products and more developed countries exporting manufactured goods. He could not envisage, but would have welcomed a third wave of globalization during which countries with nearly half of the planet’s population would enter into world trade by exporting industrial goods and then growing at a pace previously unknown. Furthermore, he was writing at the time of the two oil shocks and hence in a period of stagflation of the world economy; in Kaldor’s view (1976), international trade relations give the world economy a deflationary bias because when there is a surplus in primary production the fall in these prices leads to a reduction in the purchasing power of these countries which, as we have seen, is a demand constraint of the output of advanced countries; by contrast, if there is a shortage of primary products, their prices increase, money wages increase, inflation increases and anti-inflationary policies will reduce output and employment at the world scale. This is the reason why Kaldor (1964c) called for a reform in the international monetary system that would offset this bias. He argued for the issuance of a new international reserve currency, similar to Keynes’ ‘bancor’, that would be backed by a great deal of major commodities and would operate as a buffer stock: a boom in the output of primary commodities would increase the creation of this international money, which would then be spent on buying industrial output; conversely, a shortage of agricultural output would reduce the creation of this international money, which would reduce demand from the industrial sector instead of creating inflation.

5. The Fourth Law

As growth continues, the key source of external demand for the manufacturing sector shifts from agriculture to that deriving from outside the country, namely exports. In an open economy, if the endogenous component is not only consumption (à la Kahn), but consumption and also investment (à la Hicks) and if the autonomous component, instead of being investment, is export, then Keynes’s multiplier is transformed into Harrod’s foreign trade multiplier. Export growth then sets the pace for economic growth.

Furthermore, in an open economy a constraint on capital accumulation does not derive only from the demand of its output, but also from the supply of its input, namely imports. A country, above all if it is developing, that has little access to international credit requires exports to pay for imported goods. The income elasticity of demand for imports is greater than unity, which requires that exports grow faster than the domestic economy. Trade flows are determined by the relative prices and by the relative income elasticities of demand. In Kaldor’s opinion, the second factor is more important than the first in explaining the surplus or deficit in the balance of payments. A country cannot afford an external deficit unless it receives a flow of investment and/or of aid, but this flow amounts to a growth in the stock of foreign debt, which after a certain point causes a debt-income spiral according to the same unpleasant mathematics of internal debt unsustainability and has even worse economic consequences. Hence in the medium-long run the growth of exports must equal the growth of imports and the growth of output cannot exceed this rate. The maximum rate of growth of output is then given by the rate of growth of export divided by the elasticity of imports relative to income. Thus if the world demand for a country’s exports is growing at 3 per cent per annum and this country has an income elasticity of 2, the maximum growth rate for that country will be 1.5 per cent per annum, which may be well below the potential rate of growth (McCombie and Thirlwall, 2004). This is known as Thirlwall’s Law, and Kaldor himself added it to his three laws.

The constraint identified by this law bites in different ways according to the position of the country within the international financial system: the US economy is enjoying a sort of seignorage right, because it has been growing beyond the limit of external constraints for many years. Its growing trade deficit is offset by acquisition of financial assets denominated in dollars by the rest of the world. Since 1989 the US has been accumulating net foreign liabilities which now exceed 3,000 billion dollars—24 per cent of American GNP. On the other hand, Brazil has an external debt much lower in relation to its national income but is compelled to reduce its rate of growth below its potential and consequently runs the risk of a debt crisis. European countries have reduced the individual constraint on their balances of payments by creating a large monetary area which could, under certain circumstances, put the entire area into a similar condition as that of the United States.

6. Vicious and Virtuous Circles of Economic Growth

By integrating Harrod’s foreign trade multiplier with Kaldor’s second law, we have a model of virtuous and vicious circles of economic growth and an explanation of divergent growth among countries. Rapid growth of demand and output leads to an increase in the growth of productivity due to increasing returns to scale, which increases capital accumulation. In an open economy, this leads to competitive advantages, and consequently to faster growth of exports, which in turn contributes to the growth of demand and to a virtuous circle in a process of cumulative causation à la Myrdal. [3]

It is therefore international relationships that condition the rapidity of a country’s growth and explain divergences in growth and the factors that are responsible for the rich getting richer and the poor getting poorer. In Kaldor’s opinion (1981), the pure theory of international trade, from Ricardo to Mill and from Marshall to Hecksher-Ohlin, has been unable to provide a convincing explanation for this process. According to the classical and neoclassical theories of international trade, free trade benefits every participating country by re-allocating resources in a way that every unit of labour can contribute more than previously to the national product. The reason for this is the assumption of constant returns to scale. Furthermore, if it is also assumed that a production function is homogeneous for all commodities and equal for every country (the reason for trading being based only on differences in resource endowment), it is well known after Samuelson that, even if the factors are not mobile, trade in commodities equalizes the price of factors. Hence, all countries benefit from international trade and the poor more than the rich.

In Kaldor’s model (1970), the result is the opposite. If a South country mostly produces goods subject to decreasing returns, where land and not labour is the limiting factor, and if a North country produces mainly manufactured goods subject to increasing returns, the opening of international trade brings the price of manufactured goods in terms of agricultural products so low in the North country that such activity practically disappears in the South country. The opening of international trade for the South country results in increased exports of agricultural products but a lower national income, because the income earned from the greater quantity of agricultural exports (a sector producing at decreasing returns) cannot compensate for the loss of income resulting from the demise of the manufacturing sector (a sector producing at increasing returns). International trade makes the South country poorer, with lower employment and lower national product, and the North country richer. The role of increasing returns to explain processes of localization of industries in countries entering into international trade has been acknowledged in recent times by specialists on international trade (see Helpman and Krugman, 1985) though for these authors, contrary to Kaldor, international trade based on economies of scale will be advantageous even for those countries that have lost the increasing returns sectors.

A methodological comment is worth adding here. For Krugman, the structure of international trade produced by economies of scale is not predictable and history and randomness play a role. Several years previously, Kaldor developed an interpretation of growth and development which reached analogous but broader conclusions. In his 1983 Okun Memorial Lectures, Kaldor (1985) stresses the inconsistency of increasing returns with any orthodox concept of equilibrium and tries to substantiate a case for evolutionary growth, harnessing biological and ecological metaphors. Economic development should not be analysed from an equilibrium perspective, he suggests, but from an evolutionary one.

7. Globalization, Convergences and Divergences, and an Integration of Kaldor’s Model

Since Kaldor’s death in 1986, the world has experienced an acceleration of the process of globalization and new evidence has emerged about the relationship between development and international relations. The literature about this process is now immense and it would be impossible to review it all here. However, I would point out some recent evidence (Collier and Dollar, 2001; Glyn, 2004) and test it against Kaldor’s thought.

In the last twenty years:

- the composition of exports from the South changed in favour of manufacturing (between 1980 and 1999 the share of developing countries’ manufacturing exports as a proportion of their total exports increased from 25 to 80 per cent);

- as far as economic growth is concerned, the world can no longer be divided into the developed industrialized North with high rates of economic growth and the developing agricultural South with slow rates of economic growth, but rather into fast globalizing countries experiencing high rates of growth and slow globalizing countries experiencing slow rates of growth;

- the income per capita gap between countries (differences in the average income per capita) is increased if every country has the same weight, but it is reduced if countries are weighted by population because a group of countries—among which are China and India, whose combined populations are nearly half of the world’s total—enters into the global economy and experiences a rate of growth higher than every country in the past;

 - foreign direct investment flows (FDI)—and the process of outsourcing some phases of manufacturing production in the North—assume an importance in the international relations between South and North even greater than it had during the late nineteenth century and unknown until twenty years ago; the great majority of FDI are directed toward very few countries experiencing a high rate of growth and do not go toward the poorer and slower growing, even if this is a case of a cumulative process in motion.

This evidence cannot be explained either by the pure neoclassical theory of growth and international trade nor by Kaldor’s model as we have summed it up here. The neoclassical theory is strong in explaining the process of diffusion of economic activity and consequently of the convergence processes of factor prices and income per capita, but it is weak in explaining the polarization processes which are still in motion: China has entered into a cumulative process and consequently reduced its gap with the developed world, while Africa has increased its gap. Kaldor’s model, by contrast, is able to explain the divergences which cumulative processes put in motion, but needs some further element to explain convergences. The leaps from one club into another is explained in terms of the capacity for a less-developed country to implement the industrialization policies that Kaldor advocated (see part 9 below).

In order to be able to contemplate the processes of convergences and divergences, Kaldor’s model has to be altered in two main ways: one involving wages and the other productivity. The assumption that industrial wages are fixed and spent only on agricultural goods should be dropped. Even if a country of the North experiences returns to scale higher than those of a country of the South, the difference in wage dynamics could more than offset the differences in the dynamic of productivity, and the consequent lower labour cost per unit of industrial output in the South could induce trade of a large variety of manufacturing goods and intermediary goods from South to North. The diffusion effect of wage equalization could offset, or even more than offset the divergent process of income growth per capita produced by increasing returns.

As for productivity, its source is not only endogenous (learning by doing and increasing returns to scale), but also exogenous (invention or import of capital goods embodying technical progress). An important cause of growth of income per capita is identified by the theory of ‘catching up’, which was developed initially by Gerschenkron (1952) and more recently by Abramovitz (1986). It states that the larger the gap between a country’s (or group of countries’) level of income per capita and that of the leading country (or countries) at an initial point in time, the higher its (their) rate(s) of growth in income per capita. The follower importing capital goods or receiving FDIs makes an investment which embodies the latest generation of technical progress; the ratio of new to old capital stock is higher in the economy of the follower than in that of the leader, and consequently the rate of growth of output per person is also higher. As time goes on and the follower catches up with the leader, the gap between incomes per capita declines and the two rates of growth converge.

The trends of economic growth in developed and developing countries shows that the catching-up process took place between European countries and the US for two or three decades following the Second World War and has been occurring for the last one or two decades between Asian and OECD countries. The model is not able to account for cases of countries starting too far behind the leaders (for example, many countries in Africa), nor can it explain some Latin American cases. For this reason, an integration of this model with that of Kaldor’s could shed some light on events (Targetti an Foti, 1997).

8. Policies of Growth and Development in the Late Twentieth Century.

The case of Latin America introduces the debate over the reforms of the 1980s that occurred between North American economists and policy technocrats around the world—a debate that Kaldor would have strongly criticized. The nature of the controversial policy can be summarized under the triple commandments ‘stabilize, liberalize and privatize’ and it has been called the Washington Consensus. Latin America was the region which adopted this policy agenda most fully, but similar reforms were adopted in sub-Saharan Africa and in many Asian countries. In just a few years most of these developing countries eliminated quantitative restrictions on imports, lowered tariff barriers, and reduced the dispersion of tariff rates. Furthermore, capital was free to move and budget deficits were reduced. If growth was hampered by high inflation generated by populist macroeconomic policies and by statist protectionism, the elimination of these obstacles should have generated a boom in the private sector of these countries. But that did not happen. The rates of economic growth for countries that quickly adopted the triple commandments were lower than before and lower than those of other countries which were slower and more reluctant to adopt the reforms. Africa’s poor performance can be attributed to the ravages of civil wars and disease. But Latin America’s poor performance during the 1990s in terms of output and growth in productivity in relation to the past cannot be easily explained. The ‘mean’ period 1950-80 of import substitution outpaced the ‘glorious one’ of free trade (1980 to the present).

During the last two decades, China has grown at 9 per cent per annum, India has doubled its growth rate, and other countries like Vietnam have also performed very well. It is certainly true that these countries have adopted market-oriented strategies and it is equally true that their acceleration in growth rates has coincided with their entering into global markets, but the market reforms followed a time pattern which often was not at all quick and the specific ways they undertook reforms have been shaped by their specific institutions. China added a market system to a planned one that is being dismantled very slowly; private property goes together with enterprises owned by local governments; free trade is limited to special economic zones and a protectionist trade regime is slowly being reduced following the country’s entry into the WTO. India started its take off in the 1980s, a decade before the liberalization of 1991, and the Vietnamese model is similar to the Chinese. This policy of shaping international integration on the basis of national institutions is not a novelty for Asia because the ‘gang of four’ followed a similar path in the previous two decades and Japan even earlier. Only the peculiar town-economy of Hong Kong is a case of successful economy which rigorously adopted the ‘Washington consensus’, but it is an exception, not a rule. Also on the grounds of finance and regulations the ‘Washington model’ was not a success (Stiglitz, 2004).

The positive experience of growth in a number of Asian countries led Rodrik (2004) to state that there are some first order principles of economic policy that all successful countries have more-or-less adhered to: (1) integration into the world economy but preserving a high level of protection against imports (remaining for a long period outside GATT-WTO rules); (2) implementing effective policies to push exports and to attract investment; (3) achieving macroeconomic stability to prevent high inflation and debt; (4) providing protection for foreign investors in terms of property rights; and (5) maintaining a certain degree of social cohesion, solidarity, and political stability. These principles, however, do not coalesce directly and uniquely into a specific policy agenda; in fact, a growing number of academics (North, 1994; Rodrik, 2003) are aware that sound institutional arrangements have large elements of indeterminacy and are country specific; as a consequence, policy experimentation becomes a necessary component of institutional development.

9. Kaldor’s Development Policies

As a consequence of his laws of growth and development, Kaldor advocated development policies that were primarily based on industrialization. But industrialization is not a natural phenomenon which takes place once markets are ‘free’, and thus it requires wise policies based on land reform (since otherwise industrialization will be hindered by the growth of agriculture supply) and the protection of ‘infant industries’. If he was still alive today, Kaldor would find confirmation of his theory in China’s industrial take off, which was preceded by the solution of the historical problem of feeding its population and by several years of export-led growth before deciding to join the WTO.

It should be underlined that Kaldor believed that protective measures should be modest, discriminatory, and temporary. As the internal market broadens and industry moves out of its ‘infancy’, there should be a progressive relaxing of tariffs. The import substitution policies of Latin America, which were not accompanied by land reform or by the liberalization of agricultural imports, led to structural inflation due to the increase of wage good prices, to an excessive share of profit on national income, and to consequent over-expenditure on luxury imported goods (Kaldor, 1964b). To counter structural inflation and trade deficit, the IMF’s policies of restrictive monetary policy and currency devaluation are, in Kaldor’s opinion, both ineffective: the first leads to income deflation, the second to price inflation. A single equilibrium exchange rate between the national cost of production and international prices does not exist for developing countries. Balance of payment equilibrium can therefore only be achieved by reducing industrial costs relative to agricultural prices. The remedy is a system of dual exchange rates: a fixed managed rate should apply to the export of raw materials and to essential imports (including the goods necessary for industrialization), and a free-floating rate should apply to other (unnecessary) imported goods and to the export of manufactured goods (Kaldor, 1964a). Unnecessary import growth depreciates the free rate; since it does not produce wage inflation (because wage goods are imported under the fixed rate of exchange), this depreciation lowers the price of manufacturing exports in foreign currencies and serves as a stimulus to industrialization. The same effect can be produced by a system of export subsidies and import duties.

As was noted at the beginning of this paper, Kaldor was a tax advisor for many developing countries (India, Ceylon, Chile, Mexico, Ghana, British Guyana, Turkey, Iran, and Venezuela). It is difficult to sum up the wide range of ideas he suggested to these countries (see Kaldor, 1980). Taxation was in his opinion a crucial instrument for development because it both provides incentives for industrialization and, above all, financing for education, health, communications and infrastructure (human capital accumulation together with increasing returns are the two basic principles of the modern theory of endogenous growth). The aim of the optimal tax system should be to tax the ‘taxable surplus’, the difference between the actual consumption and the minimum consumption of the population. This taxation potential depends not so much on the level of income as on the degree of inequality in the distribution of income and wealth, on the structure of land ownership, on the pressure of population on land and on the proportion of national income going to non-residents. Only if taxes are levied on this taxable surplus can inflation or the contraction of private accumulation be avoided. On several occasions, Kaldor pointed out that the developing countries are unable to tax adequately (at that time, thirty years ago, the average tax/income ratio was 15 per cent for the developing countries and more than twice as much for the developed ones) and that the tax system was progressive in theory, but regressive in practice. To counter these detrimental characteristics, he proposed a system generally based on a land tax, a VAT tax (India’s Congress Party is trying to introduce one now with lots of opposition!), a personal family-income tax with moderate rates flanked by a personal wealth tax with very low rates, and a transfer of property tax proportionate to the wealth of the recipient and not to the value of the transfer. An important fiscal instrument is the abandonment of the secrecy and anonymity of property and the struggle against corruption. As one can see, redistributive aspects play a major role. This was the main reason why his proposals were often accompanied by fierce opposition and even riots in the streets by the powerful lobbies that would be hindered by these measures.

10. Conclusions

Orthodox economic theory is unable to correctly identify and encapsulate the kind of processes that lead to sustained development—i.e., it is unable or unwilling to come to grips with increasing returns in the manufacturing sector and, therefore, it is irrelevant as a framework in which to discuss development processes. By contrast, increasing returns are pivotal in the Kaldorian vision of development; indeed, successfully emerging countries have ignored the precepts of orthodox theory and taken seriously the implications of increasing returns in manufacturing (broadly conceived), the relevance of ‘the extent of the market’, and the cumulative virtuous circle that comes about by the interaction between the two.

Kaldor’s thought on development and on the policies for development is very modern. His ideas correspond substantially to Rodrik’s (2004) ‘five first order principles’ and encourage the idea of setting into motion experimental policies based on specific institutional arrangements. In particular, Kaldor argued that the only way a country can escape the spiral of increasing impoverishment is by industrializing, and one of the necessary conditions for successful industrialization is the entry of developing countries into the world market. A failure to industrialize and to enter the global market leads to widening gaps among per capita country incomes. The chief reason for the economic backwardness of many countries does not lie in capitalism as such (today we would say ‘globalization’), but rather is due to the economic policy errors made by governments and international institutions, and, in the majority of cases, to the short-sightedness of national ruling classes that are unwilling to sacrifice their immediate interests for the future well-being of the community. Examples include failures to reform agriculture, public administrations and tax systems which, because of the inadequacy of the revenues they generate, cannot finance the infrastructure necessary for industrialization nor the education necessary for the formation of human capital. The successful developing countries are not characterized by adopting a single economic model imported from abroad, but by having a ruling class able and willing to shape economic and social institutions supportive of economic growth.


Abramovitz, A. (1986) ‘Catching Up, Forging Ahead and Falling Behind’, Journal of Economic History, 46(2):385-406.

Collier, P. and D. Dollar (2001) Globalization, Growth and Poverty. Building an Inclusive World Economy. A World Bank Policy Research Report. New York: Oxford University Press.

Gerschenkron, A. (1952) ‘Economic Backwardness in Historic Perspective’, in B. Hoselitz (ed), Progress of Underdeveloped Areas. Chicago, IL: University of Chicago Press.

Glyn, A. (ed) (2004) ‘Globalization’, Oxford Review of Economic Policy, 20(1):1-156.

Helpman, E and P. Krugman (1985) Market Structure and Foreign Trade. Increasing Returns, Imperfect Competition and the International Economy. Cambridge, MA: MIT Press.

Kaldor, N. (1940) ‘A Model of the Trade Cycle’, Economic Journal, 50(197):78-92.

Kaldor, N. (1955) An Expenditure Tax. London: Allen and Unwin.

Kaldor, N. (1957) ‘A Model of Economic Growth’, Economic Journal 67(268):591-624.

Kaldor, N. (1959) ‘Tax Reform in India’, Economic Weekly (Bombay) 15(January):195-98; rptd. in R. M. Bird and O. Oldman (eds) Readings on Taxation in Developing Countries, revised ed., pp. 273-80. Baltimore, MA: Johns Hopkins University Press, 1967.

Kaldor, N. (1961) ‘Capital Accumulation and Economic Growth’, in F. A. Lutz and D. C. Hague (eds), The Theory of Capital, pp. 177-222. London: Macmillan.

Kaldor, N. (1964a) ‘Dual Exchange Rates and Economic Development’, Economic Bulletin for Latin America 9(2):215-23.

Kaldor, N. (1964b) ‘Economic Problems of Chile’ in N. Kaldor Essays on Economic Policy. Volume II, pp. 233-87. London: Gerald Duckworth.

Kaldor, N. (1964c) ‘The Problem of International Liquidity’, Bulletin of Oxford Institute of Economics and Statistics 26(3). Rptd in N Kaldor, Further Essays on Applied Economics, pp. 593-67. London: Gerald Duckworth, 1978.

Kaldor, N. (1966) Causes of the Slow Rate of Economic Growth in the United Kingdom. Cambridge: Cambridge University Press.

Kaldor, N. (1967) Strategic Factors in Economic Development. Ithaca, NY: Cornell University Press.

Kaldor, N. (1970) ‘The Case for Regional Policies’, Scottish Journal of Political Economy 17(3):337-48.

Kaldor, N. (1975) ‘What is Wrong with Economic Theory?’, Quarterly Journal of Economics 89:347-57.

Kaldor, N. (1976) ‘Inflation and Recession in the World Economy’, Economic Journal, December 86(344):703-14.

Kaldor, N. (1980) Reports on Taxation, vols. 1 and 2. London: Gerald Duckworth.

Kaldor, N. (1981) ‘The Role of Increasing Returns, Technical Progress and Cumulative Causation in the Theory of International Trade’, Economie Appliquée, 24(4):593-617.

Kaldor, N. (1985) Economics Without Equilibrium. The Arthur M. Okun Memorial Lectures.  Armonk, NY: M.E. Sharpe.

Kaldor, N. (1995) Causes of Growth and Stagnation in the World Economy. Cambridge: University Press.

Lucas, R. (1988) ‘On the Mechanics of Economic Development’, Journal of Monetary Economics 22(1):3-42.

Maddison, A. (2005) Growth and Interaction in the World Economy. The Roots of Modernity. The Henry Wendt Lecture Series. Washington, DC: American Enterprise Institute.

McCombie, J. S. L. and A. P. Thirlwall (2004) Essays on Balance of Payment Constrained Growth. Theory and Evidence. London: Routledge

Myrdal, G. (1957) Economic Theory and Underdeveloped Regions. London: Gerald Duckworth.

North, D. C. (1994) ‘Economic Performance through Time’, American Economic Review 84(3):359-68.

Rodrik, D. (2003) In Search of Prosperity. Analytic Narratives on Economic Growth. Princeton, NJ: Princeton University Press.

Rodrik, D. (2004) ‘Rethinking Economic Growth in Developing Countries’. Luca d’Agliano Lecture in Development Economics, Turin, Italy (October).

Romer, P. M. (1989) ‘Increasing Returns and New Developments in the Theory of Growth’, NBER Working Paper 3098 (January).

Stiglitz, J. E. (2004) ‘Capital-Market liberalization, Globalization and the IMF’, Oxford Review of Economic Policy, 20(1):57-71.

Targetti, F. (1985) ‘Growth and the Terms of Trade: a Kaldorian Two Sector Model’, Metroeconomica 37(1):79-96.

Targetti, F. (1992) Nicholas Kaldor. The Economics and Politics of Capitalism as a Dynamic System. Oxford: Clarendon Press.

Targetti, F. and A. Foti (1997) ‘Growth and Productivity: A Model of Cumulative Growth and Catching Up’, Cambridge Journal of Economics 21(1):27-43.

Thirlwall, A. P. (1986) ‘A General Model of Growth and Development on Kaldorian Lines’, Oxford Economic Papers 38(2):199-219.

Thirlwall, A. P. (1987) Nicholas Kaldor. Brighton: Harvester Wheatsheaf Books.

[1] I would like to thank Prof. Vela Velupillai for his thoughtful, stimulating, and detailed comments and an anonymous referee for her/his helpful comments. The usual disclaimers apply.

[2] Lucas (1988) and Romer (1989). Note, however, that only the latter acknowledges Kaldor’s papers of the 1950s for the concept of stylized facts and he pays only a lip service to Allyn Young for the concept of increasing returns. (This has been pointed out to me by Prof. Velupillai.)

[3] See Myrdal (1957), who in turn acknowledges that this idea originated from Wicksell’s ‘cumulative process’.