No. 119

ECONOMIC GROWTH AND INCOME INEQUALITY: THEORETICAL BACKGROUND AND EMPIRICAL EVIDENCE

Cesar Gallo

2002 Working Paper No. 119 ISSN 1474­3280

ECONOMIC GROWTH AND INCOME INEQUALITY: THEORETICAL BACKGROUND AND EMPIRICAL EVIDENCE

Cesar Gallo

2002

Development Planning Unit University College London 9 Endsleigh Gardens London, WC1H 0ED [email protected]

2 ECONOMIC GROWTH AND INCOME INEQUALITY: THEORETICAL BACKGROUND AND EMPIRICAL EVIDENCE

CONTENTS

1 Introduction 4

2 The Theoretical Background 5 2.2 Inequality 5 2.2.1 Inequality in Income Consumption Or Wealth? 6 2.2.2 Definition of Income 7 2.3 Economic Growth 8 2.3.1 The Lewis’ Model 10 2.4 Income Distribution 12 2.4.1 General of Income Distribution 12 2.4.2 Theories Of Size Income Distribution 14

3 Economic Growth and Income Inequality: The Empirical Evidence 17 3.1 The Kuznets’ Hypothesis 17 3.2 The Debate About The Inverted­U Hypothesis 18 3.3 The Effect on 19 3.4 The “U­Turn” In Developed Countries 21 3.5 Income Inequality As a Cause Of Economic Growth 22

4 Conclusion 25

References 27

Endnotes 33

3 ECONOMIC GROWTH AND INCOME INEQUALITY: THEORETICAL BACKGROUND AND EMPIRICAL EVIDENCE

1. INTRODUCTION

All sciences should be aimed at providing the expense of one of the two productive humanity with a better . The factors, thus yielding economic classical economists were concerned with inefficiency. The concern was mainly with this objective of Economics as a science. the factor distribution of income, as it was According to (1884), no in . The division of can be flourishing and happy, of national income between wages and which the far greater part of the members profits tells nothing directly about how are poor and miserable. incomes are distributed among people, placed the distribution of income at the which is usually understood as the centre of his thought when he stated that problem of income distribution. Political Economy should be aimed at The 1990s witnessed the re­ determining the laws that rule the incorporation of income distribution into distribution of income (Bigsten, 1983; the main body of economic analysis, Ferrán, 1997, Atkinson, 1997). Hartwell although economic growth still remained (1972) goes further in arguing that the ultimate objective. According to the “Economics is, in essence, the study of 1998­1999 Inter­American Development poverty” (p. 3). Bank report (IADB, 1998­1999), the However, this primary objective central argument behind this is that high seems to have been lost many times along inequality in income distribution can slow the history of economic thought. For down the pace of the accumulation of much of the last century, the subject of physical and human , which are the income distribution has been absent in the main sources of economic growth. Thus, agenda of economic and policy the main concern is that a poor income (Atkinson, 1997). Recent history has distribution harms economic growth. If witnessed changes in perception about the common welfare were the ultimate ultimate nature of the economic activity. objective, economists should be During the 1970s, in the developed world, concerned also with the effects of there was a growing concern with the quality of life, and which was manifested in economic growth on income distribution. protests against the consequences of An undesirable economic growth is one economic growth, such as pollution and which increases income inequality and depletion of natural resources. In the widespread poverty leading to social developing world the main concern was tension and political discontent that focused on the relationship between jeopardises the well­being of society. economic growth and income distribution, The elimination of widespread since many countries that had poverty and growing income inequality are experienced growth rates above their two of the main problems of development, historical standards realised that such and should be the principal objectives of growth seemed to have negatively development policy. However, the affected the income distribution, leading to problem of the distribution of income has increased inequality and a failure to relatively received little scientific study. eliminate, even reduce, the level of This paper reviews the theories that have poverty (Todaro, 1994). been developed on the subject and The revival of neoclassical discusses the empirical work on the economics during the 1980s, in a context relationship between economic growth and of a debt crisis that particularly affected income distribution. Section 2 is devoted Latin American countries, once again put to introducing the theoretical background, economic growth at the forefront of the while Section 3 presents a literature debate. It was becoming the ultimate review on what the empirical evidence objective of economic activity, despite the says about the relationship between fact that inequality was increasing and the growth and inequality. Section 4 poverty problem was worsening. summarizes a set of conclusions. According to the neoclassical perspective, any redistribution would be at

4 2. THE THEORETICAL BACKGROUND with social welfare. In this common usage the term embodies some sort of value There are three main concepts involved in judgement about fairness which will differ the debate presented in this paper that according to different ethical viewpoints. must be discussed: inequality, economic Debates on subjects such as the growth, and income distribution. relationship between inequality among Inequality is a multidimensional individuals and social welfare, fairness in concept. The discussion given in this the distribution of wealth, levels of paper refers to one of those dimensions: inequality allowed without being offensive the inequality in the distribution of income. to society, can not avoid some ethical Sub­section 2.2 discusses this concept, arguments. the different approaches to measure Naturally, these arguments are also inequality, and the definition of income. present when regarding the measurement Sub­section 2.3 gives a quick overview of issue. According to Sen (1973) there are the theories of economic growth and their broadly two categories of inequality extensions which helps to understand the measures. On the one hand those which possible connections between economic try to catch the extent of inequality from an growth and income distribution. objective point of view, using statistical However, these connections can measure of relative variation of income, only be seen in the light of the theories of and on the other those that try to measure income distribution. How the total income inequality in terms of some normative generated in a society is distributed notion of social welfare which introduces among its members and what determines the complication of ethical valuations. The such distribution have been questions problem is that there is not a clear line placed at the top of the debate in some drawn delimiting the two approaches. periods in the history of the economic "Even if we take inequality as an objective thought and relegated to the background notion, our interest in its measurement during others. The dominant ideas in a must relate to our normative concern with specific period are linked with those in it, and in judging the relative merits of former periods, even in times when different objective measures of inequality, revolutionary changes have occurred. it would indeed be relevant to introduce Therefore, having knowledge of the main normative considerations. At the same ideas about the distributive problems time, even if we take a normative view of developed in the past contributes to a the measures of income inequality, this is better understanding of the current not necessarily meant to catch the totality theoretical debate on income distribution. of our ethical evaluation" (Sen, 1973, p. 3). This is the subject of Sub­section 2.4. But the problem goes even beyond this point. It has been said earlier that inequality is about comparison. What one 2.2 Inequality would want to compare is the living In any context the concept of inequality standard of each individual in a specific refers always to comparability between society. Unfortunately, no agreement has elements. The comparison is usually been achieved among social scientists based on specific characteristics which about what exactly the standard of living of can be measured using adequate indexes an individual means and how to measure or indicators. Thus, the fact is reduced to it. The controversies arise not only from comparisons between quantities and the different ethical points of view of those inequality will relate to differences in these who want to measure the extent of quantities. In these terms, inequality inequality, but also from the difficulties in seems to be a straightforward concept capturing accurately the person's well­ which, as Cowell (1995) states, being. As reflected by Sen, "obviously" suggests a departure from the simple idea of equality, this is, the fact that You could be well off, without being well. two or more quantities are the same size. You could be well, without being able to However, difficulties arise when the lead the life you wanted. You could have concept is framed into the social context got the life you wanted, without being and in connection with economic happy. You could be happy, without having much freedom. You could have a problems. In this context, the concept is good deal of freedom without achieving generally related to differences in income, much (Sen, 1985, p. 1., quoted by consumption or wealth and associated

5 Goodman, Johnson, and Webb 1997, p. are sometimes used to compensate the 11). weakness of income or consumption based measures in capturing adequately Then, it seems quite unlikely to find many aspects of well­being. Some of out a single index able to provide a full those indicators are life expectancy, infant description of living standard as this mortality, nutrition, the proportion of outlined by Sen. The literature on household budget spent on food, literacy, inequality and poverty has often used school enrolment rates, access to health income, consumption, and wealth as clinics or drinking water. Again, the idea is proxies for living standards, but none of to have a standard scale so that different these three concepts takes into account population groups may be compared health, freedom or achievement. They do (Wratten, 1995). not measure happiness unless we assume The problem is that when using a that happiness is directly equivalent to wide range of indicators to describe level of income or consumption. Even inequality the different variables may tell more, in fact, what people regard as conflicting stories. Cowell (1995) argues happiness is influenced by culture and that we often want to answer a question personal preferences, and this varies from such as “has inequality increased or individual to individual. These proxies do decreased?” with a straight “yes” or “no”. not measure the “worth” of an individual. So, making inequality a multidimensional Income, consumption and wealth tell us concept the probability of giving about the command over resources ­ ambiguous answers is pretty high. Also, potential in the case of income and wealth, having a standard indicator is useful and actual in the case of consumption, but because it provides a uniform scale not about welfare per se (Goodman, against which comparisons can be made Johnson, and Webb, 1997). in different populations such as urban and Furthermore, Cowell (1995) discuss rural, populations living in different regions that none of these concepts cover of the country, urban populations living in completely the command over resources different parts of the city, male and female for all goods and services in society. They headed households, old and young, etc. exclude “social wage” elements such as Comparative data are essential in order to the benefits received from enjoying items target resources to specific groups of the such as municipal parks, public libraries, population. This is why, in practice and for the police systems, whose distribution may policy purposes, income, consumption, only be conjectured. According to Lipton and wealth remain the key measures of and Ravallion (1995) measures based on inequality, despite the grave deficiencies person's consumption of goods and of compressing the different features of services are intrinsically limited. They may inequality into a single indicator of well­ reveal nothing about the disutility of work, being which provides only a partial picture the length or health of the life over which of reality. consumption is expected, risk and However, standardised indicators of variability, etc. "Income is a useful living standards based on income, indicator if we want to identify which consumption, and wealth can give rather people are likely to lack the resources to different measures of inequality depending achieve a social acceptable standard of on the specific circumstances to which living. However, it does not measure they are applied. This is worth discussing accurately their capacity to achieve access here. (which may be influenced by other factors such as , information, legal 2.2.1 Inequality In Income, rights, illness, threatened domestic Consumption Or Wealth? violence or insecurity)” (Wratten, 1995, p.13). As it was said earlier, income, Thus, we can see that the living consumption, and wealth are three standard of an individual or household is, concepts associated with command over in fact, a multidimensional concept resources ­potential in the case of income involving, in principle, every aspect of and wealth, and actual in the case of direct consumption along with non­ consumption. The controversy about consumption activities and services and which indicator should be used arises from this makes inequality a multidimensional the discussion on whether we should concept. Supplementary social indicators measure potential or actual enjoyments of

6 goods and services. Let us have a closer life­cycle theories, consumption would be look at this. more or less the same over time. Between these three concepts there However, a rather different story is told are such relationships that variations in with respect to income and wealth. During one yield changes in another, which is the first period he has no income and ends interesting to observe. In any year any up with negative wealth. During the person may receive one (or a second period he has income which is combination) of the following flow of greater than his consumption, because he resources: earnings, transfers (student is paying back debts and . grant when young, pension when old) and Meanwhile, his wealth gradually becomes capital receipts such as legacies and gifts. positive. During the last period he has These flows constitute his incomei . Also, little income and his wealth is decreasing this person may hold some wealth the until his death when it will be zero. value of which might increase (decrease). Therefore, Goodman, Johnson, and Webb This increase (decrease) in value (1997) argue that if a population of three constitutes another flow of resources individuals identical in all respects but age which is also regarded as income is assumed, each one at the beginning of (negative). This person can save part of each of these periods previously his income. Therefore, the difference described, a measure based on wealth or between his income and saving is his income would show significant inequality, consumption. At the moment, let us while a measure based on consumption assume that there is no difference would find no inequality. Even more, the between expenditure and consumption measure based on wealth would rank the which will be discussed later onii . If this individual at the beginning of the third person does not save, his wealth at the period as the best off followed by the one beginning and at the end of the year will at the beginning of the first period, and the remain the same. If he saves, this person one at the beginning of the second period, will be increasing his stock of wealth. If his who would be the best off according to an consumption is greater than his income, income­based measure. his wealth will decrease or he will have to Obviously, the real world does not borrow. work like that. People do not follow that Goodman, Johnson, and Webb rationality. As the authors argue, capital (1997) use the life­cycle theories to markets are not perfect and individuals are explain how these three concepts can give not always able to borrow as much as they rather different measures of inequality. want, they can be risk averse and According to these theories people smooth inherited wealth matters. Also, these their consumption when their income individuals may belong to households in varies. Their consumption decisions are which they share living standards with not only based on their current income, but others of different ages, a case in which, also on their expected income in the again, different stories can be told about future. These authors assume an isolated inequality if the unit of observation is the individual who enters his adult life with household or the individual. neither income nor wealth, and that there But the main problem arises when are no student grants or pensions. Then, considering these indicators as measures he goes through three periods ­education, of command over resources. To see the work, and retirement. "Life­cycle theories differences, based on an example given would predict that he would borrow during by Goodman, Johnson, and Webb (1997), the period of education. In the working let us assume two households (or period he would pay­off these debts and individuals) with identical characteristics, save for retirement. In retirement, he their income included. One of them would receive some investment income spends all its income, whereas the other from his assets but he would also use up decides to save (or can just spend) 50% of his assets. He would die (leaving nothing its income. A measure of living standard as he has no children to inherit from him) based on income will find no difference with nothing" (Goodman, Johnson, and between these two households. In this Webb, 1997, p. 13). case, income is a measure of the potential The main point here is to see what command over resources. A measure is going on with this person's income, based on consumption will tell us a rather wealth, and consumption throughout these different story. It will show a significant three periods of his life. According to the inequality between these two households,

7 since one of them does not access (or cannot achieve access) to all goods and 2.2.2 Definition Of Income services it could have had with the income Income is the simplest and most at its disposal. In this case, consumption fundamental concept in Economics. is a measure of actual command over Despite seeming a straightforward resources. In a more illustrative way, using concept, a clear definition is needed when Atkinson's words, "A miserly millionaire it is going to be used as a measure of who lived on virtually nothing would from living standard. In the broadest sense of one point of view appear rich, and from the the word income denotes what comes in. other, poor" (Atkinson, 1983, p. 37). So, Thus, it may be used in different contexts the choice between income­based and with quite different meanings. The consumption­based measures will make definition widely accepted in Economics of quite a difference to the level of inequality personal income is the amount of money, reported. in a given period of time, that an individual If one is specially interested in can spend in consumption without altering comparing the actual enjoyment of good the value of his/her wealth (Lindahl, 1933; and services by households (or Simons, 1938; Hicks, 1946; Atkinson, individuals) in society as a way of 1983). comparing their living standards, As it has been stated earlier, measures based on consumption seem to personal income is a concept that has to be advisable. However, the information do with command over resources. Simons commonly available is expenditure and not (1938) put this in terms of possession and actual consumption which is what one exercise of rights. It connotes, broadly, really would like to measure. There are so the exercise of control over the use of many goods which are not consumed at society's scarce resources. So, in these the time of purchase as well as so many terms, according to the definition given goods that people consume over a long above, the calculation of personal income period of time. "Current expenditure will implies the estimation of the value of rights overestimate consumption if one has just which the person might have exercised in made a large purchase of a durable item, consumption without altering the value of but will underestimate the amount of his/her store of rights. consumption if they are making use of Therefore, when the value of store many goods for which they have already of rights is unaltered, in a given period of paid in the past" (Goodman, Johnson, and time, the potential command over Webb ,1997, p. 21). Also, some resources equates the actual. In this households produce for home case, income is a good proxy for living consumption which increases the standard. Unfortunately this rarely command over resources, but this type of happens, as it was discussed earlier. consumption is not taken into account Actually, it is the level of enjoyment of when an expenditure­based measure is individuals through exercising their rights used. These are perhaps the main what one would be interested in weaknesses of expenditure as a proxy for measuring and comparing. Then how consumption in measuring the living helpful is the concept of income outlined standard of households or individuals in above in measuring these enjoyments? society. Then, it seems quite unlikely that Fisher (1930) uses the concept of real actual command over resources can be income to clarify this. Real income measured with accuracy. includes all those events such as the use Thus, due to the availability and of food, wearing clothes, going to the reliability of data as much as anything cinema, etc., which contribute to our else, most researchers involved in the enjoyments. However, most of these analysis of inequality from an economic events occur in different ways producing perspective – to which the debate different levels of enjoyment for different presented in Section 3 refers ­ concentrate individuals. Therefore, they are difficult to on income as the primary measure. This measure by using any standard unit. means that they concentrate on potential "They have no common denominator. command over resources. However, it is Even the individual who experiences them worth pointing out that in the presence of cannot weigh and measure them directly. accumulated and wealth, income All he can do is to measure the money he will give only a partial view of potential paid to get them" (Fisher, 1930, p. 47). command over resources.

8 Fisher (1930) identifies three stages economic perspective uses the money in the process that leads to these received in a certain period of time, which enjoyments, which occur in the opposite has been called just income, as the order that they are stated. These stages measure of the potential living standard of are: i) the inner enjoyment which he calls individuals (or households). In doing so, enjoyment income, ii) the events in the there will be a difference between income outer world which give us inner enjoyment and cost of living equal to the amount of which he calls real income, and iii) the savings minus the amount of borrowings iii . amount of money that we have to pay to Thus, only if the person neither saves nor get those events which he calls cost of borrows or if he/she borrows to save living. Thus, the cost of living measures (savings = borrowings), the money the real income which give us our received equals the cost of living. enjoyment income. Here the following question arises: 2.3 Economic Growth does the money that the individual Economic growth can be defined as a rise receives in a certain period of time equate in per capita income and national product. to his cost of living? We have already To increase national product the volume of seen that, in general, the answer in no. investment must be greater than the To understand the difference Fisher amount necessary to replace depreciated (1930) uses the concept of money income. capital. Therefore, the amount of savings "Money income includes all money and investment plays a significant role in received which is not obviously, and in the the process of economic growth This nature of the case, to be devoted to view is central in the classical models of reinvestment ­ or, as the expression is, economic growth such as Harrod (1939), 'earmarked' for reinvestment. In other Domar (1946), Solow (1962). Holding words, all money received and really constant other variables, an increase in available and intended to be used for the rate of savings would accelerate the spending is money income" (Fisher, 1930, rate of economic growth. This is the p.49). simplest statement in the influential Money income is usually less than Harrod­Domar model. cost of living. Cost of living practically But obviously those other variables never equates money income. Many are not constant, they are changing. persons live beyond their money income. Changes in the rate at which capital According to Fisher's definition, money produces output can counteract or add to income can never be greater than cost of the effects of the rate of savings on living. economic growth, as well as the rate at Thus, the money that the individual which the total population is growing, receives in a certain period of time would affects the rate of growth of per capita be money income only in the case in income. If the population is growing at the which this individual spent all the money same rate or faster than that of total received in that period without saving. If output, then the rate of per capita income the individual saves part of the money may be zero or even negative. These received, only that part devoted to be factors have been incorporated in the spent will be his/her money income. In this classical models of economics growth case the money received will be greater which predict correct correlation between than the cost of living. If the individual the rate of growth and variables such as does not save but borrows, all the money savings rate, output­capital ratio, and received will be money income and it will population rate. But these models are not be less than the cost of living. able to predict exact levels of growth rates When working on inequality, ideally, providing specific values of the variables what one would be interested in involved which the models assume as measuring and comparing is the exogenous. The reality is not as simple as enjoyment income of individuals which is that. Not only the growth process affects, not directly measurable. Cost of living in fact, the value of all those variables, but would be the indirect monetary measure of also the effect does not follow a definite enjoyment income, but then there are pattern. The rate of savings, for problems related to the differences example, perhaps the most relevant between expenditures and consumption variable in the models of growth, may be discussed earlier. This is why most of the influenced by the level of per capita research done on inequality from an income in society, as well as by how total

9 income is distributed among the factors. When labour is abundant, few population. It seems logical to think that additional units of capital are required to countries with a low level of income also produce larger amounts of output. In this exhibit small or even negative rates of case, the capital­output ratio is low. As savings. The very poor countries have more capital is incorporated into the limited room for savings. As the economy productive process () grows this room for savings may enlarge. the capital­output ratio rises because But once a certain level of per capita labour becomes a relatively scarce factor. income has been reached, there is not What is at work here is the law of guarantee that the savings rate will keep diminishing returns to individual factors of on growing. Several factors can be production. mentioned. One is that the notion of what The Solow model takes account of the are changes as society the diminishing returns on capital and becomes healthier, which may imply predicts a steady­state of per capita increases in the rate of consumption to income in the long run, in which total such an extent that it may even overcome output will be growing at the rate of growth the savings rate. For example, the USA is of the population. As capital is one of the world's richest countries, but accumulated faster than labour, the also has one of the world's lowest rates of capital­output ratio decreases, reaching a savings (Ray, 1998). Another factor is how point at which no further growth in per the benefits of growth are distributed. If capita income is possible. In contradiction the benefits of economic growth go in with the Harrod­Domar model, in which greater proportion to those groups of constant returns are assumed, the Solow population who are regarded as having a model predicts that the savings rate has high propensity to save, then the overall no long term effect on the rate of growth. rate of savings is expected to rise. But if Capital accumulation itself is unable to growth benefits more the groups with high sustain per capita growth. This important propensity to consume, then the overall conclusion leads to the incorporation of rate of savings would decrease as a result technological in the model as the of growth and would harm subsequent ultimate source of growth. Technical growth. This argument has been used by progress increases the productivity of some scholars in the debate about the labour so that, the model predicts, in the effects of growth on income distribution, as long run increase in per capita income it will be seen in the next section. takes place at the rate of technical Another factor affected by the path progress. of economic growth is . The idea that technical progress Both the death rate and the birth rate tend may drive long­run per capita growth is to be high in countries with low levels of quite acceptable. But the following income, which keep the rate of growth of questions arise: how is technical progress population at low levels and varying within introduced and expanded in the production a narrow range of values (For figures, see process? Does technical progress evenly Ray, 1998, p. 299). Economic growth is increase the productivity of labour? expected to bring a rise in the living According to Ray (1998), the so called standard of the population which makes "New Growth Theories" bring to the arena the death rate fall due to improved the concept of . In short, conditions in hygiene, sanitation, and human capital is regarded as skilled­ medicine, while the birth rate reacts with labour. This is, educated labour that is delay iv. As a consequence, the growth skilled in production, able to operate rate of population initially increases. Once sophisticated machinery, able to improve the birth rate adjusts, the growth rate of methods of production, and able to create population starts decreasing and in the new ideas applicable to economic long run it reaches low levels again. activities. These theories augment the Also the capital­output ratio might Solow model by splitting savings into be an endogenous variable. The value of physical and human capital. The the capital­output ratio depends on the introduction of human capital in the model relative endowments of capital and labour of growth makes it feasible that there may in the economy. This relative endowment be broadly constant returns to physical of productive factors may change with and human capital combined, even though economic growth. Total output is the result physical capital exhibits diminishing of the interaction of these two productive returns. This makes the economy appear

10 to behave as predicted by the simple (who are poor) to vote for high taxation Harrod­Domar model. In this way, the rate aimed at redistributing income. The of savings reassumes its role in the long­ increase in taxation reduces the after­tax run growth. marginal product of capital and, as a However, the effectiveness of consequence, the rate of accumulation savings and investment cannot be taken decreases and therefore the rate of just in aggregate terms. There are key growth. Finally, the instability channel sectors in the economy which may require operates through the fact that a large more investment than others. These are group of impoverished citizens, facing a the sectors which may be able to shoot up small and very rich group of individuals, the growth process, spreading the benefits demand radical changes leading to mass to the whole economy. Models based on violence and illegal seizures of power that dual economies see the industrial sector negatively affect the investment as the engine of growth so that the opportunities vi . balance between rural and urban sectors is of special importance. These models 2.3.1 The Lewis' Model have been in the centre of the debate Lewis (1954) put forward an approach of about economic growth in developing based on the countries, they have also been used to process of transformation of a traditional explain and justify the increasing inequality economy into a modern one. This in developing countries as a result of approach is based on the existence of a economic growth. The most influential dual economy in which a “traditional” model of this type is the Lewis (1954) sector coexists with a “modern” sector. Model treated with some detail in the next The term traditional is associated not only sub­section. with the agricultural sector, which The connection between inequality produces the traditional output of society and growth has been seen not only using old techniques of production which through the effects of growth on income are labour­intensive, but also with all those distribution, but also in how growth is economic activities based on the same affected by income inequality. There is conditions of production. These conditions recent literature connected with the new of production include an organisation theories of endogenous growth that based on a family whose members share establish the links through economic and the overall output. On the contrary, the v political mechanisms . The economic modern sector encompasses all those mechanism emphasises the effect of economic activities that are organised on income distribution on the size of domestic capitalist principles and use new demand and, as a consequence, the technologies which are capital­intensive. potential for , as well as However, it is worth saying that in the real the role of imperfect capital markets. The world this distinction between traditional latter establishes the links through the and modern sector is not that clear. The investment in human capital. In the agricultural sector can operate based on presence of imperfect capital markets, the capitalist principles of organisation and individuals face borrowing restrictions for may use advanced capital­intensive education. The inheritance of each technologies, while there are economic individual determines whether he invests activities organised in capitalistic form, in education or not. Hence, the although using labour­intensive distribution of income determines the technologies. Nevertheless, such a simple aggregate level of investment in human distinction is helpful to explain the main capital and then the level of output (Galor ideas of Lewis' Model. and Zeira, 1993). The fundamental assumption of Alessina and Perotti (1993) highlight Lewis' Model is the existence of a large three political mechanisms. One stresses surplus of labour in the traditional sector of the fact that unequal distribution of income the economy which can be transferred to is linked to large incentives for the poor to the modern sector without affecting the engage in rent­seeking activities or in amount of output in the traditional sector. illegal activities that pose a threat to There are too many workers relative to property rights, thus hindering investment other factors of production, so that and growth. The second mechanism is removing the surplus does not affect the the fiscal channel in which income level of output. This means that the inequality induces the majority of voters

11 marginal product of labour in the traditional (income share criterion). Let us call this sector is zero or very close to zero. "wage" w1. If some labourers are According to Ray (1998), this is a transferred from the traditional sector to narrow definition of the concept of surplus the modern sector an output surplus is labour which may be inapplicable to originated, assuming that output does not reality. His main argument is that, holding fall and "wage" does not rise in the constant other factors of production, there traditional sector. This output surplus is always the possibility of increasing the divided by the number of labourers output by using additional labour with removed equates w1. Since the labourers improved and more intensified techniques. removed from the traditional sector have However, the applicability of this concept now to buy their food, w1 represents their depends on an adequate interpretation subsistence level. Multiplying w1 by the beyond this purely technological definition, terms of trade (relative price) between the Ray (1998) adds. In this sense, even if we two sectors we get the minimum wage (w * assume that zero marginal productivity of ) in the modern sector. Thus, in this phase, labour really exists in traditional activities, we have a perfectly elastic supply of we have to remember that payments in labour. This is the so called phase of these activities are based on income­ economic development with unlimited share criterion (total output divided by the supply of labour, maintaining wages at the number of workers) and workers will not subsistence level. transfer to activities with positive marginal As the process goes on, the output productivity of labour unless this marginal in the traditional sector starts to fall which productivity is greater than their actual leads to a rise in food prices. This is, the payment per unit of labour. Moreover, process moves out of the surplus labour zero marginal productivity is not a phase. As a consequence, wages in the necessary condition for the transfer to go modern sector have to rise above w * to on. It is enough that there exists a compensate this increase in food prices. differential in marginal productivity of However, workers in the traditional sector labour between the two sectors. This still consume the same per capita amount, differential leads to the concept of which means that their "wages" remain at disguised unemployment which has been w1 . This is called the disguised interpreted as an extension of the concept unemployment phase. From this point, the of surplus labour (Ray, 1998). The supply of labour in the modern sector is extension is complemented with the fact not perfectly elastic any longer. As the that removing labourers from the process of labour transfer continues, the traditional activities does not mean total output in the traditional sector keeps removing labour. Actually, the remaining falling and reaches a point in which the workers in the traditional sector may marginal product of labour begins to increase their effort such that total output exceed w1 . Therefore, the wage in the does not fall. This is, the labour input may traditional sector rises above w1 which remain constant so that removing induces a sharper increase in the modern labourers does not affect total output. sector wage to compensate not only for This extension of the concept of rising food prices, but also to be able to surplus labour is fundamental in the attract additional labour in order to sustain description of the process of economic further growth in this sector. This is the growth as a result of the process of labour commercialisation phase. transfer from the traditional sector to the From the demand side, the scenario modern sector described by the Lewis is that, while the economy is in the surplus Model and later extended by Ranis and labour phase, the modern sector demands Fei (1961). labour paying wages at the level of Ray (1998) provides a schematic subsistence and further expansion can go description of how the process takes on, demanding more labour, without place, based on Ranis and Fei (1961), raising the wage rate. Once beyond this which can be summarised as follows. phase, further investment requires higher There are three phases in the process. wages. Labour turns into a scarce factor The first one is the surplus labour phase. of production and food prices rise, In this phase a surplus of labour in the increasing the cost of hiring workers. The traditional sector is assumed in which expansion of the modern sector "wage" is equal to the ratio between total accelerates growth, but this is limited by output divided by total number of labourers the ability of the economy to produce a

12 surplus of food. Thus, the Lewis Model factor of production and factor shares that tells us that the basic requirement to shoot would explain the factorial distribution of up economic growth is transferring labour income. "Most theories conceive the from traditional activities to the modern central problem of income distribution as sector, paying wages at the level of the determination of the level of subsistence. employment and remuneration of the factors of production, usually grouped into 2.4 Income Distribution capital and labour. They differ mainly in The term "income distribution" is usually their assumptions about behaviour coined to "picture" who receives how and the way in which wages and product much income within a specific society. prices are determined" (Ahluwalia and There are two principal concepts of Chenery, 1983, p. 43). But a theory of income distribution encountered in the size income distribution needs to explain literature: the functional and the personal also the distribution of the ownership of or size distribution of income. The factors among households (Knight, 1976). functional distribution shows how much Some theories fail in going further than the income is received by each factor of functional distribution of incomes. production. This is, how total income is The classical period was distributed between land, labour and characterised by focusing only on the capital. The theories based on functional functional distribution. Adam Smith distribution consider the existence of only devoted his work to the causes of wealth three groups (or classes) in society: and discussed the division of what was labourers, capitalists, and landowners, produced between wages, rent, and profit, assuming within­group homogeneity. but he did not develop a theory about the On the other hand, the size determinants of such distribution. David distribution of income shows how many Ricardo was the one who placed the individuals (or households) receive how distribution of income at the centre of his much income. This is, how total income, thought. According to Ricardo, the Political from all sources, is distributed among Economy was aimed at determining the individuals or households. laws that rule the distribution of income Other concepts of income (Bigsten, 1983; Ferrán, 1997). "He was distribution sometimes used in analysis of the first economist to derive a meaningful income inequality are those which make a income distribution theory" (Bigsten, 1983, distinction between urban and rural areas p. 4). as well as interregional or interstate The productive factors are land, differentials. However, the theoretical capital and labour, and total income is debate about income inequality has been distributed according to rent, profits and focused on the concepts of functional and wages. The basic idea in Ricardian size distribution of income. thought is that a differential rent is produced only when less fertile lands are 2.4.1 General Theories Of Income exploited requiring more capital and/or Distribution labour leading to a rise in the price of agricultural products. As a consequence, If some agreement arises from all the the owners of the more fertile lands debates about distribution of income along receive an increased rent. This is why the history of economic thought this is that Ricardo insisted that the increase in rent is there is no agreement among economists not a cause but a consequence of wealth about which are the determinants of (Ferrán, 1997). In the Ricardian system, income distribution. The main reason is distribution is prior to exchange, thus that the distribution of income is the final income distribution does not depend on result of the entire economic process demand for final products (Bigsten, 1983). (Bigsten, 1983), and it is well known that The distribution of total income there is a lack of unanimity of views even works as follows: the surplus over the on general economic issues of that production costs (output value) constitutes process, which makes it logical to expect the rent and the rest is distributed between no agreement on a topic that has been the profits and wages. "The Ricardian system source of ideological wars and political accepts a Malthusian unlimited supply of revolutions (Sahota, 1978). labour at the subsistence wage in the A theory of income distribution long­run; it assigns to 'profits' the residual needs a theory which explains prices of between marginal product of labour at a

13 point in time and the subsistence (or linked to the structure of the production institutional) wage, and attributes to relations. Therefore, a change in the landlord rent the remaining residual in total structure of distribution can only happen output value" (Cline 1975, p. 360). Since as a result of a change in the production land is not unlimited and not equally fertile, conditions. According to Marx, capitalism in the long­run, according to this theory, constantly reproduces its production the share of profits tend to fall while rent relations and therefore the laws that rule and wage shares tend to rise (more labour the corresponding distribution of income. is required), although the real wage is kept For this reason, the economic position of at subsistence level. the working class can only improve when According to Ricardo, this capitalism itself collapses. He saw the distribution in favour of the landlords industrial concentration as the result of the prevents economic growth acceleration tendency for the profit rate to fall leading to and the economy would tend to stagnate, cyclical crisis until a final apocalyptic crisis. since he regarded landlords as (Cline, 1975; Bigsten, 1983). "In the spendthrifts and thought that economic Marxian long­run the system collapses growth was financed by the savings of the due to declining wages (whether declining capitalists. Moreover, he proposed absolutely or only relatively to capitalist reducing restrictions to imports and income is unclear) and intolerable worker introducing technological innovations poverty in the face of capitalistic without increasing the proportion of labour accumulation" (Cline, 1975, p. 361). and keeping wages at subsistence level in Opposite to the Marxian theory in order to achieve rapid economic growth many aspects, the neo­classical or (Gillis, Perkins, Roemer, Snodgrass, marginal productivity theory postulates 1987). This proposition reveals his view that all factors of production are in scarce that economic growth requires not only supply and all of them create value vii . This redistribution from landlords to capitalists, refutes the Marxian view that only labour but also stressing inequality between creates value. The product value is profits and wages. explained by the marginal utility which also The Ricardian theory focuses on the explains the remuneration of factors. They conflict between rent and profits. Later on generalised the marginal productivity as this focus shifts to the conflict between the basis for payment of all factors, profits and wages especially with Marx's eliminating the attribution of residual view. Since regarded the main income to any one factor. This view is class conflict to be between capitalists and based on the principle that the producers workers, he only considered two sources maximise their benefits, and this point is of income: profits and wages. Although only reached when all factor payments Marx recognised that rent, benefits and equalise their corresponding marginal interests were also sources of income, he productivity. Thus, the distribution of assumed that these types of income were income is regarded as part of the general received only by one class: the capitalists pricing process in the economy. In this (Ferrán, 1997). way, the demand of factors depends on Marx uses the Ricardian "labour product demand, this is, the demand of theory of value" to diagnose exploitation of factors is derived from product demand. workers (Cline, 1975). Like Ricardo, he The prices of factors and goods are assumed an unlimited labour supply which assumed to be determined by market allowed the capitalists to hold down the forces. Also, factor substitutability is wages at a subsistence level. But the assumed so that the rise in one factor labour surplus was possible through the supply decrease its relative price. existence of a "reserve army of labour" Therefore, the neo­classical theory of that capitalists were stimulated to maintain distribution is based on production through labour displacing innovations in functions and elasticities of substitution order to achieve the profit rate level that (Cline, 1975; Bigsten 1983; Ferran, 1997). capital accumulation requires. Marx Explaining the functional refused to accept the Malthusian theory distribution, the neo­classical view helps to that demographic forces created the explain the size distribution of income. labour surplus (Gillis, Perkins, Roemer, "The sum of payments to the factors of Snodgrass, 1987). production possessed by an individual In the Marxian theory, the structure determines his income" (Bigsten, 1983, p. of the distribution of incomes is strongly 5). Thus, the changes in the functional

14 distribution and, as a consequence, in the size distribution of income over time can This last proposition shows clearly be explained by changes in relative factor Kaldor's view on capitalists' consumption, supplies, elasticity of substitution between stated above viii . Therefore, according to factors, changes in product demand, and Kaldor's model, there is a specific the capital or labour saving bias in distribution between wages and profits that technological change (Cline, 1975). makes savings equal to investment. An While the neo­classical theory uses increase in investment forces an increase the marginal productivity of factors to in savings to restore equilibrium. Since explain income distribution, the Keynesian capitalists are assumed to have a higher economics bases its view on marginal propensity to save, the equilibrium can propensities to save. According to Ferrán only be restored by increasing profits (1997), Nicholas Kaldor establishes this through an increase in the price level view, stating that the profit share in total (Knight, 1976). This suggests a positive income is determined by macroeconomic relationship between economic growth and forces which assure that capitalists inequality in the factorial distribution of expenditures in consumption produce income. benefits to finance those expenditures. Cline (1975) points out a couple of This is, as quoted by Kurz (1994), basic flaws in Kaldor's model. The assuming that only capitalists are net assumption that there are only two classes savers, a given amount of profits can only makes its materialise if there is a corresponding application extremely restricted, since it amount of net investment and capitalists' becomes undetermined considering three consumption. or more classes. Also, an exogenously Kaldor's model assumes that there given investment is not well justified are only two classes: capitalists and While Kaldor built up his model workers. Each class has a specific focusing on profit share, Michal Kalecki propensity to save ­workers' lower than proposed a model based on Marxian view that of capitalists. In this model, the ratio focused on wage share in which he uses of investment to national income is an monopoly analysis. In his microeconomic exogenous variable which does not analysis Kalecki uses the Lerner definition depend on changes in the propensities to of the "degree of monopoly": µ = (p ­m)/p, save. Then, under full employment where p is product price and m is the condition, equalising saving and marginal cost. "If marginal cost is equal to investment yields the only one possible marginal revenue, µ is equal to the inverse distribution of total income between wages of the elasticity of demand for the product and profits (Cline, 1975; Bigsten, 1983). of the enterprise" (Kalecki, 1951, p. 19). The profit share in total income is In Kalecki's model the components of expressed by the equation variable costs are raw material and labour P/Y = [1/(sp ­ sw)][(I/Y) ­ sw] costs. Aggregating for a closed economy, (2.1) Kalecki shows that labour share varies negatively with the "average" degree of Where P are profits, I investment, Y total monopoly power in the economy (Cline, income, sp and sw are the propensities to 19775; Bigsten, 1983). So, according to save of capitalists and workers the relationship between the degree of respectively, being sw < sp (See Ferran monopoly and the elasticity of product 1997, pp. 248­249). Assuming that both demand stated above, this means that the propensities to save are given, the profit lower the product demand elasticity, the share is determined by the rate of lower the share of wages in value added. investment. The coefficient 1/(sp ­ sw) Thus, this model suggests that economic measures the changes in the distribution growth relied on a growing monopoly of income due to an unitary change in the power in the economy would lead to an investment rate. The smaller the difference increasing gap between wage and profit between sp and sw the greater the impact shares in total income. Kalecki's model of changes in investment rate on income faces the difficulties of building up a distribution. Assuming sw very small, macroecomic theory of distribution based nearly zero, equation (2.1) leads to on microeconomic elements without paying attention to aggregation problems. P = (1/sp)I For example, Ferran (1997) argues that at (2.2) microeconomic level it is plausible to

15 assume a given factor supply at a given large amount that a few individuals can price, while at macroeconomic level some win (Ferran, 1997). The risk averse limitations may arise by factor availability. individual will take less risky choices. As a consequence, a society composed of risk 2.4.2 Theories Of Size Income averse individuals will generate less Distribution inequality than one composed of risk­ The foregoing theories are aimed at taking individuals. This theory suggests explaining the functional distribution of that the rich are risk­taking while the poor income rather than the size distribution of are risk averse individuals. However, income. Therefore, they are of limited Ferran (1997) refers to the work by value in analysing the governmental Stanley Lebergott (1959) who tests action. According to Ahluwalia and Friedman's model using data from USA Chenery (1983), because most of the and finds that the group who takes the wage earners belong to the middle­income most risky investments is not the richest groups is the reason why policies affecting but the small investors expecting a the distribution between wages and profits windfall. Moreover, in most countries mainly concern the upper end of the size (developed or developing), gambling distribution. Although, the neo­classical seems to be supported by the low income theory makes some contribution in groups who expect through a windfall to understanding the determinants of the size improve their economic position. distribution of incomes, differences in Cline (1975) and Sahota (1978) factor endowments seem not to be enough highlight the work by Mincer (1958) among to explain the large inequalities in those economists who support the human developing countries, particularly in Latin capital approach. This approach focuses America. It would be necessary to explain on the explanation of job­earning also how these differences in endowments differentials. "Mincer's human capital were created. model is based on the idea that A number of theories of size income occupations requiring longer training distribution have been developed. periods pay higher earnings to According to Sahota (1978) the set of compensate for the foregone income theories of personal income distribution during training. More precisely, all can be classified into two basic groups. individuals are assumed to choose among One group ranges from those theories occupations such that the present developed by economists who believe that discounted value of earnings is equated income inequalities are largely a among all occupations...all incomes are in consequence of voluntary choice to those reality equal. Observed differences are in which inheritance and play merely statistical illusions stemming from the main role. The other, which Sahota the fact that the high income individual has (1978) calls the "fatalist" group, is been on the job a shorter time than his low constituted by three schools: a) theories income cohort" (Cline 1975, pp. 365­366). based on the premise that incomes are There are two general objections to this distributed among individuals according to model. First, it focuses only on job­earning abilities which are genetically determined; differentials as the main source of income b) theories that postulates that income inequality, disregarding the effects of other inequalities are largely determined by sources such as holding assets. Second, chance, luck and stochastic factors; and c) it overemphasises the impact of schooling the life­cycles theories which give high on earning differentials, but disregards the relevance to the age effect on earning effects of sex, race and family background capacities. on schooling achievement which could The theory presented by Milton limit the range of real choices that Friedman is referred to as the individual individuals have. choice theory in which stochastic Another interesting approach within influences are combined with optimising the context of the job­earning differentials behaviour on behalf of the individuals is the "job competition" model attributable (Bigsten, 1983). According to Friedman, mainly to Thurow (1972, 1975) (Cline, being risk­taker is what explains that a 1975; Bigsten, 1983). According to this small group in society can receive a large model "wages are paid based on the proportion of total income since, as in the characteristics of the job in question and lottery, the amount of money that many workers are distributed across job can lose is small in comparison with the opportunities based on their relative

16 position in the labour queue" (Cline 1975, property incomes. Sahota (1978) and p. 367). Bigsten (1983) argues that the Bigsten, (1983) summarise the view of position of a worker in the queue should inheritance, attributable to Meade (1964, be dependent on the potential cost of 1976), as consisting of a bundle of four training the worker which may be major endowments: genetic make­up, determined using some screening device parental level of education and training, such as education. social contacts, and inherited property The job competition approach itself. The mutual interaction of these refutes the assumption that the labour endowments affect incomes, savings, and market determines the wage level. Wages accumulation, so inequality in these are not set by the marginal product of the endowments explains differences in worker associated with his/her previously income. acquired education level, but by the This brief shows that the marginal product associated with the skills ideal theory able to explain simultaneously he/she learns on the job. As a the determinants of factor prices, consequence, people with identical levels functional shares and the size distribution of education may be paid different wages. of income does not exist. Although the This theory disregards that how rapid theory of inheritance considers the workers can learn skills on the job might property incomes, the missing element depend in some degree on the level of from all those theories is an explicit education previously achieved, which in treatment of the distribution among turn depends on several other factors such households of the various forms of assets as family background. So, according to such as land, privately owned capital, the job competition view, expanding access to public goods, and human education would not tend to equalise capital. Ahluwalia and Chenery (1983) income distribution. The effect of coincide with Sahota's (1978) view that the education expansion would be only to distribution of assets is more unequal than increase the admission requirements for the distribution of earnings. Also, obtaining jobs, while expanding job Deininger and Squire (1997) refer to six opportunities would have a more effective developing countries in which inequality in equalisation impact. the distribution of land is greater than that Abilities and stochastic forces have in the distribution of incomes. Ahluwalia also been used as explanations for and Chenery (1983) associate the inequalities in the distribution of income. variations in income at the lower levels Differences in abilities explains differences with the lack of human skills, physical in worker's productivity and as a capital and access to them. Attanasio and consequence differences in income, while Sékely (1999) document that income other theories regard stochastic forces as inequality in is, to a large responsible for the skewed shape of the extent, a reflection of a very skewed personal distribution of incomes. However, distribution of income generating assets. these views give little help for explaining Therefore, asset distribution cannot be the determinants of income inequality and disregarded when assessing the effects of for policy making aimed at changing the economic growth on income inequality. distribution of income. (For details about these theories, see Sahota, 1978, and Bigsten, 1983,). The theories previously discussed are concerned mainly with the distribution of earnings. They neglect an important source of income which is property. Sahota (1978) points out that property incomes are more unequally distributed than earnings and that inheritance is the major source of property class perpetuation. Hence, a theory of distribution that neglects property incomes will tell only part of the story. The personal­income­distribution theory of inheritance takes account of

17 3. ECONOMIC GROWTH AND INCOME However, some supporters of the INEQUALITY: THE EMPIRICAL hypothesis, such as Anand and Kanbur EVIDENCE (1993), have tried to provide the "adequate" empirical evidence and Economists have been widely concerned formalise the process using mathematical with the effects of economic growth on models. The motive seems to be looking inequality and vice versa. Does economic for explanations of the increasing growth tend to improve, worsen or have no inequality in developing countries. If the necessary effect on income distribution? inverted­U hypothesis holds, then the less Are high levels of income inequality a developed countries (LDCs) are on the left necessary condition for growth side of the curve, thus increasing acceleration? These are the main inequality is justified, and it is just a matter questions around which the debate of time for the "trickle down" to operate. revolves. Perhaps behind this debate is This is, growth itself generates the desire of finding "economic laws" that opportunities for lifting up the bottom end describe the path of the distribution of of the income distribution. Hence, it is income in the course of economic growth. worth having a closer look to Kuznets' The hypothesis put forward by Simon explanations of his "conjecture" and his Kuznets in 1955, well known as the warning about favouring repetition of past inverted­U hypothesis, has been assumed patterns of the now developed countries to have that category of economic law as (DCs) in the present LDCs. the centre of the debate for more that 40 According to Kuznets (1955), years, despite the little empirical support excluding government intervention, there that this hypothesis has received. This is are two forces that explain income why Gary Fields (1988, p.462) called it inequality before taxes: the concentration "one of the greater ironies in the history of of savings in the upper­income groups and thought on economic development". the industrial structure of the income To present the debate I have distribution. The former yields inequality in divided the section into five sub­sections. savings which, all other conditions being Sub­section 3.1 presents the hypothesis equal, has a cumulative effect of launched by in 1955. Sub­ increasing the proportion of income­ section 3.2 is dedicated to the debate on yielding assets in the hands of the upper­ the effects of economic growth on income income groups leading to larger income inequality, as Sub­section 3.3 summarises shares of these groups and their the findings related to "poverty" ix. Sub­ descendants. The other force is the result section 3.4 looks at the debate in the DCs of the process of industrialisation and context, while in Sub­section 3.5 the other urbanisation, this is, economic growth side of the coin is presented, this is, accompanied by the shift away from income inequality as a cause of economic agricultural activities. On one hand, the growth. process increases the urban share in total population, which is assumed more 3.1 The Kuznets' Hypothesis unequal than rural population. On the Kuznets (1955, p. 18) stated that there other hand, since average per capita was "... a long swing in the inequality income of rural population is usually found characterising the secular income lower than that of the urban, Kuznets structure: widening in the early phases of (1955) argues, that this gap in relative economic growth when the transition from mean incomes tends to widen as a result the pre­industrial to the industrial of a more rapid growth of the per capita civilisation was most rapid; becoming productivity in economic urban activities stabilised for a while; and then narrowing than in agriculture. in the later phases". Surprisingly, Kuznets However, despite the cumulative recognises that "No adequate empirical effect of the concentration of savings, evidence is available for checking this eventually these tendencies reverse over conjecture of a long secular swing in time, according to Kuznets (1955), as a income inequality; nor can the phases be result of both government redistributive dated precisely" (p. 19). Even more, he intervention through legislative admits that his paper "...is perhaps 5 per interference and political decisions and a cent empirical information and 95 per cent group of "less obvious" factors which speculation, some of it possibly tainted by characterise a dynamic growing economy. wishful thinking" (p. 26). First, the decreasing proportion of rich

18 families as a result of more family control Kuznets (1955) has been taken as in this income group and the increasing inevitable by believers in trickle down immigration entering at the lower­income whose explanations are almost always levels reduces the share of the top 5 related to the nature of structural change. percent of the population. Second, As Gillis, Perkins, Roemer, Snodgrass, dynamic economies create an atmosphere (1987) discuss, the main argument is of relative freedom of individual based on the similarities between opportunity giving the chance for more conditions observed in LDCs and those rapid growth to younger industries and which prevailed in the industrialised hence expanding the emergent middle countries before the Industrial Revolution classes. Third, in these economies there is The theoretical framework to explain rising a rising trend, shifting workers from lower­ inequality with economic growth is income to higher­income industries. The provided by the Lewis model, discussed last two "less obvious" factors along with earlier, while the Fei­Ranis (1964) the Lewis' view (1954) have been used as extension of the Lewis model explains, arguments by the believers in the trickle according to these authors, that this down, when they attempt to translate tendency toward increasing inequality is successful experiences in DCs to the reversed when all the surplus labour is LDCs’ context. In this respect, Kuznets absorbed into modern­sector employment. himself warned against the temptation of Thus, labour becomes a scarce factor of favouring the repetition of past patterns of production and further growth, which the "now" DCs under the markedly implies an increase in labour demand, will different conditions of the "present" LDCs. push the wages up. So, this rise in the This warning has rarely been cited in the general wage level will bring the downturn literature and surprisingly ignored, a in inequality and poverty reduction. reason why it is worth quoting his own words here as basis for further discussion. 3.2 The Debate About The Inverted­U There is danger in simple analogies; in Hypothesis arguing that because an unequal income The Kuznets' hypothesis at the early distribution in Western Europe in the past stages of economic growth received led to accumulation of savings and support by the first major study about the financing of basic capital formation, the relationship between economic growth and preservation or accentuation of present income inequality by Adelman and Morris income inequalities in the underdeveloped (1973), although they admitted that the countries is necessary to secure the same downturn predicted by the Fei­Ranis result. Even disregarding the implications model is not automatic, this is also for the lower­income groups, we may find challenged by Ashwani Saith (1983) who that in at least some of these countries places emphasis on the differences not today the consumption propensities of only between DCs and LDCs, but also the upper­income groups are far higher and fundamental inter­country differences with saving propensities far lower than were regard to size, historical heritage, the those of the more puritanical upper­income timing of their industrialisation process, groups of the presently developed etc. In this sense, Saith (1983) gives countries. Because they may have proved justice to the Kuznets' warning quoting favourable in the past, it is dangerous to Kuznets' call for national studies: "If argue that completely free markets, lack of emphasis is to be on similarities and penalties implicit in progressive taxation, differences in the basic characteristics of and the like are indispensable for the the process of adoption of the industrial economic growth of the now system ­ for countries distinguished by underdeveloped countries. Under present their size, historical heritage, and the conditions the results may be quite the timing of their industrialisation process ­ opposite ­withdrawal of accumulated assets we obviously need a variety of national to relatively 'safe' channels, either by flight studies" (Kuznets 1954, p. 153, quoted by abroad or into real state; and the inability of Saith, 1983, p. 369). Later on Adelman governments to serve as basic agents in and Fuwa (1994) give relevance to policy the kind of capital formation that is choices made in the course of the indispensable to economic growth (Kuznets development process where the downturn 1955, pp 25­26). in inequality occurs. Despite these words being But Saith (1983) also challenged the completely valid, the process described by methodological aspects. He discusses the

19 fact that the inverted­U hypothesis evidence of experiencing the second acquired a paradigmatic status with the phase of the inverted­U curve. These arrival of the Paukert (1973) and Ahluwalia findings suggest that, in the case of LDCs, (1976) papers which attempt to test the the relationship between economic growth hypothesis on the basis of cross­country and inequality could be better described data x. He argues that these econometric by an inverted­L curve rather than an studies have little in common from a inverted U­curve. However, notice that methodological point of view with Kuznets' this discussion is focused only on the studies which were supported by time­ relevance of the level of growth and/or the series data on the income of the "rich", rate of growth. Also, when conjecturing while these papers concentrate on the about the inverted L­curve, heterogeneity bottom 20% or 40% of the population. In among LDCs is ignored. fact, Kuznets (1955) did not hypothesise Another methodological weakness about the income share of "the poor". of those cross­country studies might be Even more interesting is Saith's criticism the comparability problem which arises as about the estimation of one U­curve from a result of the differences in recipient unit observations on both LDCs and DCs. choices, income definition, geographic Actually, when the sample is restricted to coverage, and variations in household the LDCs group of countries the inverted­ survey design. Sailesh K. Jha (1996) U hypothesis does not hold. Saith (1983) addresses this problem using an objects to Ahluwalia's assumption that the expanded data set up to the 1990s and position of the DCs on the inverted­U finds that most of the variations in income curve reflects the future position of the distribution are due to country LDCs on the basis of two arguments. characteristics and not to data First, that the internal conditions and the comparability problems. Another important international context in which the LDCs finding in Jha's estimates is that the (taken from the sample) existed, were growth rate is not a significant variable, essentially different from those in which indicating that there is not feedback from the DCs (taken from the same sample) the rate of growth to income distribution. developed. Second, that the economic This result is consistent with the findings of and political life of the LDCs' world is not Ahluwalia (1976), who also suggested that independent of the DCs' world. "Thus, the way in which economic growth is different cases in Ahluwalia's cross­ promoted matters for the distribution of country regressions are not really income, and with those of Papanek and independent of each other. Such a Kyn (1986), and Adelman and Fuwa dependence would make the cross­ (1994). sectional exercise even more Gary Fields (1988) assessing the questionable" (Saith 1983, p. 374). lessons learned from studies which have Capital exportation from DCs to used cross­sectional data, inter­temporal LDCs make feasible further economic data, and micro data states that growth in DCs still under unlimited labour considering the two possible conclusions ­ supply conditions, once the "turning point" one that income inequality "must" increase on the U­curve is reached in DCs. So, before it decreases, the other that income one could speculate on whether some inequality may increase or decrease DCs are on the right leg of the inverted­U depending on the "type" of country and the curve at the expense of their relationship policies pursued ­ the latter is certainly with some LDCs or not and if, under these more consistent with the empirical conditions, those LDCs would be able in evidence at hand. Also, the studies that the future to reach that leg of the curve as consider structural and policy factors along well. In respect of this, Saith (1983) re­ with income level or rate of growth, Fields estimates the U­curve using only the 41 (1988) reports, show that the extent of LDC sub­sample, excluding the "outliers" inequality in different countries was (the poorest and the richest countries). He associated with factors such as education, shows that a better fit is provided not by extent of direct government economic the inverted U­curve but rather by an activity, population growth rate, inverted­L curve. Also, Ahluwalia, Carter, urbanisation, and importance of the and Chenery (1979) use a 36 LDCs agricultural sector in total production. "This sample of which 16 countries are beyond provided strong motivation for looking at the "turning point" estimated from cross­ the changes that have taken place within country data and only Taiwan shows some

20 individual countries and the reasons for could be overstated. In respect of this, them" (Fields 1988, p. 467) xi . Janvry and Sadoulet (1996) had made a Later, Fields (1991), using papers contribution to the methodology for produced for 20 countries as part of a analysing the relation between growth and research project launched by the World inequality/poverty. They argue that causal Bank in 1985 and adding information of analysis of this relation has usually been another 35 countries, confirms his done by observing changes in inequality previous finding that there is not a definite and poverty during broad periods of relationship between changes in inequality growth and recession instead of treating and the level or the rate of economic country­specific periods of growth and growth, but that those changes seem to be recession as observation units. For associated with the "pattern of growth". example, in the case of Latin America, Further support to this view is given by they report, the 1970s have been Mátyás, Kónya and Macquarie (1998), associated with growth, mid­1980s with who, using a two panel data set of 47 and recession, and late 1980s and early 1990s 62 countries, find that it is not the GDP per with growth recovery. However, countries capita which explains income inequalities like Costa Rica and Colombia were but rather the specific characteristics of a avoiding recession during the mid­1980s, country such as , political while Brazil and Peru were still in system, and natural resources. recession in the late 1980s and even early 1990s. Therefore, they suggest, the 3.3 The Effect On Poverty analysis needs to allow for this country In relation to poverty, Fields (1991) results heterogeneity by looking at country­ show that poverty is nearly always specific spell of growth and recession. reduced by economic growth, and it is Otherwise, if a negative relationship is more apt to decrease the more rapid is found between growth and poverty, for economic growth. Deininger and Squire example, the sign may derive from (1997) reach similar results, they use an recession, from growth, or from both. So, expanded data set on income distribution if the strongest relation occurs during adding new observations from primary recession, optimistic predictions about this survey data to official statistical relationship could be wrong. Even more, publications and research papers. They government policies aimed at promoting found that periods of growth are an equitable income distribution might be associated with an increase in inequality driving part of those optimistic results, the almost as often (43 cases) as with a reason why Jha (1996) suggests that decrease in inequality (45 cases). In further research should focus on how contrast, they found a strong systematic much of this growth versus equity trade­off relationship between overall growth and is due to policy. growth in the income of the poorest The focus of the research by Janvry quintile, the latter increased in more than and Sadoulet (1996) is Latin America, and 85% percent of 91 cases. This finding is they make another important shared by Morley (1995), Psacharopoulos methodological contribution which allows et. al (1995), and Ravallion and Chen measurement of how much of the impact (1997) who found that growth reduced of economic growth on income inequality "poverty", although not inequality, while is due to the "pattern of growth". They Ravallion and Datt (1996) show that argue that there are qualitative differences aggregate growth is able to reduce between growth before and after the debt "poverty". Also, Jha's (1996) estimates crisis that should be taken into account. prove that the bottom 20% of the Before the crisis, many Latin American population benefits from economic growth economies were still implementing models in the long run, suggesting that trickle of import substitution industrialisation, down seems to operate. accumulating debt and discriminating Therefore, these findings in relation against agriculture. After the crisis, to the poorest groups of the population following IMF­ prescriptions, seem to give support to the trickle­down most Latin American countries started to argument, since there is little apply severe fiscal discipline, restrictive disagreement about the positive effect of monetary policies, devaluation, and trade growth on poverty reduction. However, liberalisation. Janvry and Sodoulet control based on these results we could be for these different "patterns of growth", tempted to make optimistic predictions that creating a set of variables which pick up

21 the "qualitative features of growth" while reaches a minimum and tends to remain another set do the same with the there. This response might be related to "structural features of growth" xii . Their the sensitivity of poverty to inequality findings are very interesting. First, as in changes. Kakwani (1993) found that previous works, they found that growth poverty measures were considerably more itself is not a strong determinant of the elastic for changes in inequality than for change in inequality, while the structural growth rate, so if inequality remains at characteristics of the countries were found high levels, economic growth will be less relevant to inequality. The effect of efficient in reducing poverty. This is economic growth on poverty reduction was confirmed in the case of the USA by found favourable in all periods, which Smolensky et al (1994), who found that reinforces the findings reported above, but ".the effectiveness of growth in reducing they highlight that the main link between poverty depends a great deal on how income and poverty is established through growth is distributed. The rise in inequality recessions. offsets the poverty­reducing effects of It is important to bare in mind this more than decades' worth of growth..." methodological contribution when (Smolensky et al, 1994, p. 222). However, assessing results on Latin American this fact has not been explored thoroughly inequality, since increasing inequality and more research is needed on this might be a misleading finding in terms of subject. the inverted­U hypothesis. In fact, Altimir (1995), Beccaria et al. (1992), Fields 3.4 The "U­Turn" In Developed (1992), Lustig (1995), Morley (1995), and Countries Psacharopoulus et al. (1995) have found The debate is not limited to the LDCs that inequality mirrors the economic cycle. context. The relationship between income This is, inequality rises during periods of inequality and economic growth for recession and falls during recovery, which advanced economies, according to suggests a negative relationship that Kuznets' view, should lie on the right leg of contradicts the inverted­U hypothesis. the inverted­U curve. However, Mátyás, Janvri and Sadoulet (1999), using data for Kónya and Macquarie (1998) report that the 1970­1994 period for 12 Latin the is growing parallel to American countries, do not find support for the GDP in Denmark, Japan, and Sweden. the inequalizing effect of growth in Latin Also, Kirby (2000) surveying evidence on America. They find a very strong growing inequality in DCs finds that in the asymmetry in the effects of growth and USA income inequality increased steadily recession on inequality, whereby one year from the mid 1970s and throughout the of recession can cancel the inequality 1980s. In the UK, income inequality fell up reduction effects of more than one year of to the late 1970s but the Gini coefficient growth. This suggests that recessions in rose by more than 30 per cent between Latin America have been devastating on 1978 and 1991 which, according to Kirby inequality, and growth has been ineffective (2000), is more than double the decline in in reducing it. These findings are in line the UK from 1949­1976. This finding with those of Adelman and Fuwa (1994) coincides with that of Goodman, Johnson, who deal with the effects of policies on the and Webb (1997) who report that from growth­equality trade­off for the poor. They 1977 onwards, in the UK, the level of used income distribution data for LDCs inequality as measured by the Gini has compiled by the World Bank for the 1970s registered a continuous increase which is and the 1980s. Their regression system historically unusual. Also Kirby (2000) confirms the inverted U­hypothesis for reports an upward trend in income both periods, but the most interesting inequality over the 1980s in Australia, and aspect of their results is that the variables increase in inequality in New Zealand portraying the nature of adjustment during the late 1980s. policies are of significant influence on Rati Ram (1991) presents formal income distribution to such an extent that evidence from the USA, which was one of the authors recognise that their results the three countries used by Kuznets (the suggest that during the 1980s the other two were Germany and the UK), policies have contradicting the inverted­U hypothesis exacerbated the growth­equality trade­off associated with income inequality. After for the poor. They find that the share of the Kuznets (1955), this is one of the few poorest groups decline very steeply, studies made using inter­temporal data of

22 a single country. He recognises that using economic growth has been associated data for about half a century might not be with a significant upturn in measured able to capture the full Kuznets structure.x iii income inequality. "As far as the U.S. However, his work has the advantage of economy is concerned, and unfortunately covering a single country and avoids the for the Kuznets thesis, it now appears that problem of dealing with country the inverted U­curve has turned up again" heterogeneity. The use of cross­sections (Harris 1993, p. 69). of states deals with some degree of heterogeneity, but this is much smaller 3.5 Income Inequality As A Cause Of than that in cross­country samples which Economic Growth include countries at various stages of The debate presented so far has been on economic development. According to the the effects of economic growth on income advanced nature of the USA economy it inequality, to which the bulk of the would be "reasonable" to expect, based on literature has been devoted. Let us now Kuznets' view, that the USA economy turn our attention to the reverse of the would lie on the right leg of the inverted­U relationship. According to Colman and curve. The prediction of a monotonic Nixson (1988), the Lewis Model gives decline in income inequality, even at such support to the argument that increasing a high level of economic development, is inequality is not only an inevitable effect of not supported by Ram's results, which economic growth, but also a necessary suggests that the performance of the condition for growth. The basic argument relationship between growth and income is that savings are essential to increase inequality fit better with a U­curve rather xiv productive capacity which leads to higher than with an inverted­U curve . rates of growth, the reason why income Donald Harris (1993) makes an must be redistributed towards the groups important contribution to this debate from that save and invest ­ the rich ­ in order to a Marxist point of view. He regards "the ensure capital accumulation and growth. trickle down idea is analytically invalid and Therefore, an economy with a high historically incorrect". There is no concentration of income by the upper automatic mechanism in a market groups is more likely to grow faster than economy which could guarantee income one with a more equitable distribution of inequality reduction with economic xv income. This is the pro­inequality growth . He opposes the neo­classical argument. and neo­keynesian models' assumption of However, in line with Kuztnets' abstract economies operating at full (1955) warning, in a LDC context there are employment with the "real world" of no guarantees that the higher­income actually functioning capitalist economies groups will save a significant proportion of operating with a margin of unutilised their income in their own country. capacity as well as a reserve army of According to Todaro (1994), unlike the unemployed labour. Harris' (1993) main historical experience of the now developed argument is that, in these economies, countries, the rich in developing countries growth has a cyclical character (boom and are characterised by spending in luxury bust) which produces corresponding consumption (usually imported) and cycles in the pattern of income distribution saving abroad. This author also without any necessary long­term tendency challenges the pro­inequality argument in one direction or the other. In boom pointing out that highly unequal distribution periods, wages and profits rise and wages of income is manifested in poor health, may even rise ahead of profits as the army nutrition, housing and education for the reserve diminishes, Harris argues. But the vast majority of the population, which in tendency of profit rate to decrease leads to turn lead to low levels of productivity and a downturn imposing the limit to which the thereby to a slower­growing economy, income distribution may improve and as while the redistribution of income to the the downturn continues the income lower­income groups would stimulate inequality worsens again. Given the demand for home production, since the pattern of property ownership, these rich tend to spend more of their additional cycles in the factorial distribution translate incomes on imported goods. “Rising to the size distribution. Harris (1993) demand for local goods provides a greater reports the empirical evidence of this stimulus to local production, local cyclical pattern in the case of USA, employment, and local investment. Such showing that over the past 30 years USA demand thus creates the conditions for

23 rapid economic growth and a broader characteristics and structural factors in the popular participation in that growth” relationship between economic growth and (Todaro 1994, p.159) xvi . This is the pro­ income inequality, also it might have equality argument. important policy implications. The authors Empirical works have been suggest that if unequal distribution of conducted to test these arguments. assets is proved to have a strong negative Leightner (1992) did not find evidence to effect on economic growth, then policies support the pro­inequality argument in the that enhance the number of people who case of Korea for the period 1963­1980. are able to access credit markets could His regressions undermine the rationale contribute to accelerating economic for the argument that an increase in growth. private saving causes an increase in The mechanisms through which an investment. In contrast, he finds evidence unequal initial distribution of income might that an increase in consumption causes an affect subsequent growth are addressed increase in investment. This finding gives by both Deininger and Squire (1997) and support to the pro­equality argument that Garcia Cecilia (1994). Garcia (1994) greater equality produces more reviews a group of papers that examine consumption which drives investment and whether there exists a causal relationship growth. Leightner (1992) recognises that between inequality and long­run growth. his findings are surprising in light of According to Garcia, there have been two Korea's rapid growth because the Korean theoretical approaches. One of them government pushed inequality too far and, focuses on political decisions about according to his results, a redistribution of income redistribution through taxes ­ the income would have produced greater political mechanism xviii ­ while the other growth. He also refers to the experience looks at how inequality affects the in Japan with similar, or greater growth opportunities of individuals to invest in rates, while its government purposefully their education and thus its implication on increased income equality. Leightner's growth rates ­the educational (1992) work is specific on Korea and does mechanism xix. The first group shows that not prove the validity of the pro­equality inequality harms growth, while the second argument. What his findings show is that group finds a U­shaped relationship increasing inequality is not a necessary between inequality and growth. So, based condition for growth. on the first group one could speculate that Deininger and Squire (1997) also more egalitarian distributions would tend address the question whether more to favour growth, while from the latter egalitarian countries grow faster or not. group no definite relationship can be Their results challenge the view that predictedx x. According to Garcia (1994), higher inequality accelerates economic there is little empirical evidence on the growth. To deal with the problem of limited political mechanism which gives little availability of data to assess the impact of support to it, suggesting that there are initial income distribution on subsequent other links, while some empirical support growth, they complemented their data with is found to the educational mechanism information on the distribution of land from both the evidence that it is holdings, which provides a better measure redistribution towards the middle class that of initial distribution. To investigate the tends to have the greater impact on effect of initial inequality on long­term growth and from the correlation between growth, they look at determinants of inequality and schooling variables. Similar growth rates for the period 1960­1992. approaches are empirically tested by These authors find that the negative Deininger and Squire (1997) who also impact of the initial income inequality on reach similar results. Also, Alessina and subsequent growth is not very strong xvii . Perotti (1993) find that the evidence is However, inequality in the initial inconclusive on the political channel. distribution of assets, measured by the Forbes (2000) challenges empirical distribution of land, exerts a significant works which have reported a negative negative effect on subsequent growth. In relationship between inequality and a previous work Alessina and Rodrik subsequent economic growth, arguing that (1994) find a similar result. This result is there are a number of potential problems quite relevant, since it gives further with these works such as lack of support to findings of research, reported robustness in the estimates, measurement earlier, about the role played by country errors in inequality, and omitted variable

24 bias. This author argues that when adding 4. CONCLUSIONS explanatory variables or regional dummies, the coefficient on inequality The theories of income distribution, often becomes insignificant. Also, a reviewed in this paper, are aimed only at negative bias in the relationship can be explaining the determinants of the level of introduced by employment of productive factors and their measurement errors from, for example, relative prices, with the exception of the underreported inequality statistics, or from neo­classical theory which makes a little omitting variables such as the level of contribution to explain the size distribution. corruption, which tends to be positively They are of limited value in analysing correlated with inequality and negatively government action and the impact of correlated with growth. So, using an economic growth on size distribution of improved data set, which not only reduces income. But even those theories which measurement errors, but also eliminates a attempt to address the determinants of potential source of omitted variable bias, size distribution are mainly concerned with Forbes' (2000) results suggest that in the the determinants of individuals' earnings. short and medium term, an increase in To explain earning differentials some of income inequality has a significant positive them look at the differences in risk impact on subsequent economic growth. aversion among individuals, while other This result can be interpreted as saying highlight differences in ability and that, as long as one looks within the same stochastic factors. Some theories give to country, increases in inequality promote schooling the pre­eminent role in growth in the short ­ and medium ­ term. explaining the level of income of However, Banerjee and Duflo (2000) individuals, while others reduce this role to report works that find no relationship a simple screening device giving between inequality and growth, but when preponderance to either job characteristics the sample is broken up into poor and rich or the skills that individuals learn on the countries a negative relationship in found job. But these theories, reviewed in this in the sample of poor countries, while a paper, neglect property as a source of positive relationship is found in the sample income, and as a consequence they can of rich countries. This finding is quite tell only part of the story. Although interesting since it suggests that the pro­ inheritance has been considered, an equality argument would be plausible for explicit treatment of the distribution of the poor countries, while in the case of rich various forms of assets such as land, countries the pro­inequality argument physical and human capital, and access to might be more applicable. public goods is missed in these theories. Forbes' (2000) work is focused on However, there has been a the short ­ and medium ­ term relationship recognition in the recent literature that the and the author admits that these results do return on the various assets owned and not directly contradict the previously used in a productive fashion by the reported negative relationship between household members is relevant for the inequality and growth, since the positive household income. The distribution of impact of inequality on growth could either these assets may be affected by the path diminish or reverse in the long­term. It is of economic growth, the reason why interesting to observe also that Forbes' analysing these effects might be a good (2000) work is intended to give empirical step forward in explaining the links support to theoretical arguments such as between economic growth and income "...in more unequal , the medium distribution. voter will elect a higher rate of taxation to If one had to summarise in few finance public education, which will words the main conclusions drawn from increase aggregate human capital and the debate on the relationship between economic growth" (Forbes, 2000, p.1­2). economic growth and income inequality, However, this would be a long­term effect one should say that, despite the literature rather than a short­term effect. In the about the relationship between economic short­term, it is more reasonable to expect growth and income, inequality is huge and that redistributive taxes (higher taxes) the debate has been very long, no definite would reduce investment incentives relationship has been found and that we leading to lower growth. are still far from generalising the channels through which economic growth affects the distribution of income. The analysis of

25 inequality changes that have taken place extension of the Lewis Model is not within individual countries and their causes automatic. On the contrary, there is is perhaps one of the best contributions to empirical evidence which gives support to clarify how economic growth affects the view that the impact of economic income inequality. growth on income distribution depends Savings seem to be one of the key more on the "type" of economic growth variables in the whole debate. It was and the policies pursued than on the level shown in Section 2 that in classical of per capita income or the rate of growth. theories savings are a relevant variable in It seems that the way in which growth is explaining not only the process of promoted is what matters for its effects on economic growth, but also how this income distribution. process relates to the distribution of In the case of Latin­American income. Section 3 has shown that the countries, income inequality mirrors the current debate has been mainly around economic cycle. Qualitative differences the Kuznets' hypothesis, which achieved a have been found between growth before paradigmatic character during the 1970s, and after the debt crisis. One year of its basic assumption being that only the recession can cancel the inequality upper­income groups save, in line with reduction effects of more than one year of Kaldor's view. Despite Kuznets' warning, growth, according to some empirical the argument that the initial inequality in works. Recessions in Latin America have the distribution of income in DCs led to significantly increased inequality, while accumulation of savings which financed growth has been ineffective in reducing it. investment and thus promoted growth has According to many scholars the been translated to the LDCs context. extent of inequality has been found Nevertheless, some scholars have shown strongly associated with education, the evidence that there are not enough extent of direct government economic reasons to think that the savings of the activity, population growth rate, higher­income groups in LDCs will lead to urbanisation, importance of the agricultural acceleration of economic growth as it sector in total production (Fields, 1988; happened in today's DCs. In fact, most of Adelman and Fuwa, 1994; Jha, 1996), the the empirical works are based on cross­ specific characteristics of a country such country analysis which use a mixture of as political system and data from DCs and LDCs, disregarding the natural resources (Mátyás, Kómya and economic and political links between many Macquarie, 1998), and other structural DCs and LDCs. Even more, there are factors such as barriers to entry into high­ some empirical evidence that suggests income jobs, that the sign of the relationship between the structure of capital markets and limited inequality and economic growth can switch access to credits, the existing distribution from positive to negative depending on of property ownership, and the inheritance whether the country is rich or poor, or on system (Harris, 1993). Therefore, it whether the relationship is seen in the seems that looking at the relationship short ­ or long ­ term. between economic growth and that type of Therefore, it is safe to say that the factor, within the boundaries of individual Kuznets' hypothesis has received little countries, is a relevant contribution in empirical support. The process described establishing the links between economic by Kuznets is not inevitable. The growth and income inequality. downturn predicted by the Fei­Ranis

26 REFERENCES

Adelman, Irma and Cynthia T. Morris, 1973, Economic Growth and Social Equity in Developing Countries, Stanford University Press, Stanford California.

Adelman, Irma and Nobuhiko Fuwa, 1994, "Income Inequality and Development. The 1970s and the 1980s Compared", Economic Appliquée, XLVI, No. 1, p. 7­29.

Ahluwalia, Montek. S., 1976, "Inequality, Poverty and Development", Journal of , Vol. 3, No. 4.

Ahluwalia, Montek. S., Carter N., and Chenery H.,1979, Growth and Poverty in Developing Countries, Journal of Development Economics, Vol. 6, No. 3, pp. 299­341.

Ahluwalia, Montek, and Hollis Chenery, 1983, "The Economic Framework", Chapter II, in Chenery, H., Ahluwalia M., Bell, C. Duby. J., and Jolly, R., (eds), Redistribution with Growth, Oxford University Press, London.

Alessina, Alberto and Roberto Perotti, 1993, “The Political Economy of Growth: A Critical Survey of the Recent Literature”, The World Bank Economic Review, Vol. 8, No. 3, pp. 351­ 371.

Alesina, A. and Rodrik, D., 1991, "Distributive Politics and Economic Growth", CEPR discussion paper 565, Cambridge, USA.

______, 1994, “Distributive Politics and Economic Growth”, Quarterly Journal of Economics, Vol. 109, pp. 465­490.

Altimir, Oscar, 1995, Changes in Inequality and Poverty in Latin America, CEPAL, Santiago, .

Anand, S., and Kanbur, S, 1986, Inequality and Development: A critique, 25th Anniversary Symposium, Yale Growth Center.

______, 1993, "The Kuznets Process and the Inequality­Development Relationship", Journal of Development Economics, Vol. 40, No. 1, pp. 27­47.

Atkinson, A. B., 1983, The Economics of Inequality, second edition, Claredon Press, Oxford.

______, 1997, “Bringing income distribution in from the cold”, The Economic Journal, No. 107, March, pp. 297­321.

Attanazio, O. and Székely, M., 1999, An Asset­Based Approach to the Analysis of Poverty in Latin America, Latin American Research Network, Office of the Chief Economist, Working Paper No. R­376, IADB, Washington.

Bacha E., 1979, "The and Beyond: Growth and Changes in Inequalities", in: E. Malinvaud (ed.), Economic Growth and Resources, St. Martin's Press, New York, pp. 52­73.

Banerjee, A. V. and Duflo, E., 2000, Inequality and Growth, Massachusetts Institute of Technology, Department of Economics, National Bureau of Economic Research, working paper No. W7793, July.

Beccaria, Luis, Julio Boltvinik, Juan Carlos Feres, Oscar Fresneda, Arturo León, and , 1992, América Latina: El Reto de la Pobreza, Características, Evolución y Perspectivas, UNDP, Bogota.

Bigsten. A. 1983, Income Distribution and Development: Theory, Evidence, and Policy, Heinemann, London.

27 Bishop, John A., John P. Formby, and W.James Smith, 1991, "Lorenz Dominance and Weakfare: Changes in the US Distribution of Income, 1967­1986", Review of Economics and Statistics, Feb.

Braun, Denny, 1991, "Income Inequality and Economic Development: Geographic Divergence", Social Science Quarterly, Sept.

Chenery, H., Ahluwalia M., Bell C., 1974, et al., Redistribution with Growth, Oxford University Press, Oxford.

Chenery, H., and Syrquin, M., 1975, Patterns of Development, 1975­1970, Oxford University Press, New York.

Chiswick, Barry, 1971, "Earnings Inequality and Economic Development", Quarterly Journal of Economics, Feb.

Cline, William R. 1975, "Distribution and Development. A Survey Of Literature". Journal of Development, Vol. 1, pp. 359­400.

Colman, D.and Nixson, F., 1988, Economics of Change in Less Developed Countries, Second Edition, Philip Allan Publishers Limited, Oxford.

Coughin, Cletus C. and Thomas B. Mandelbaum, 1998, "Why have State Per Capita Incomes Diverged Recently?", Federal Reserve Bank of St. Louis Review, Sept./Oct.

Cowell, Frank A., 1995, Measuring Inequality, second edition, LSE Handbooks in Economics Series, Prentice Hall/Harvester Wheatsheaf, London.

Deininger, K., and Squire, L., 1997, "Economic Growth and Income Inequality: Re­examining the Links", Finance and Development, Vol.34, No.1, pp. 38­41.

Domar, E., 1946, "Capital Expansion, Rate of Growth and Employment", Econometrica, Vol. 14.

Domar, E., 1957, Essays in The Theory of Economic Growth, Oxford University Press.

Fei, J. C. and Gustav Ranis, 1964, Development of Labour Surplus Economy: Theory and Policy, Richard D. Irwin, Homewood, III.

Ferrán de, Lourdes U. 1997, Historia del Pensamiento sobre la Distribución del Ingreso, Fondo Editorial FINTEC, Caracas.

Fields, Gary, 1988, "Income Distribution and Economic Growth", in Gustav Ranis and T. Paul Schultz, (eds), The State of Development Economics, Basil Blackwell, Oxford.

______, 1991, "Growth and Income Distribution", in George Psacharopoulos (ed), Essays on Poverty, Equity and Growth, Pergamon Press, Oxford.

______, 1992, "Changing Poverty and Inequality in Latin America", Public Finances, Vol. 47, pp. 59­76.

Fisher, Irving, 1930a, "Income and Capital", printed from The Theory of Interest, New York, Macmillan, pp. 3­35, in , R. H. Parker, G. C. Harcourt, and G. Whittington (eds.), Readings in The Concept and Measurement of Income, Philip Allan Publisher Ltd., Oxford, 1986.

Forbes, Kristin, 2000, "A Reassessment of the Relationship Between Inequality and Growth", American Economic Review, Forthcoming.

Fosus, Augustin,.K, 1993, "Kutznets Inverted­U Hypothesis ­Comment", Southern Economic Journal, Vol. 59, No. 3, pp. 23­28.

28 Galor, O. and Zeira, J., 1993, "Income Distribution and Macroeconomics: The Human Capital Connection", Review of Economics Studies 60, pp. 35­52.

Garcia P., Cecilia, 1994, "Inequality and Growth: A Note on Recent Theories", Investigaciones Economicas, Vol. XVIII (1), Enero, pp. 97­116.

Gillis, M., Perkins, D., Roemer, M., and Snodgrass, D., 1987 Economics of Development, 2nd. ed., WW Norton and Company.

Glomm, G. and Ravikumar, B., 1992, "Public versus Private Investment in Human Capital: Endogenous Growth and Income Inequality", Journal of Political Economy 100, pp. 818­834.

Goodman, Alissa; Johnson, Paul; and Webb, Steven, 1997, Innequality in the UK, Oxford University Press, New York.

Grupta, K. L., 1970, "Personal Savings in Developing Countries: Further Evidence", Economic Record, June, 1970.

Hartwell, M., 1972, “Consequences of the industrial revolution in England for the poor”, in IEA (Institute of Economics Affairs), The Long Debate on Poverty, Unwin, London.

Harris, Donald, J., 1993, "Economic Growth and Equity: Complements or Opposites", Review of Black Political Economy, Vol. 21, No. 3, pp. 65­72.

Harrod, R. F., 1939, "An Essay in Dynamic Theory", Economic Journal, Vol. 49, No. 1.

Hicks, J. R., 1946, "Income", printed from Value and Capital, Oxford: Claredon Press, second edition, pp. 41­58, in R. H. Parker, G. C. Harcourt, and G. Whittington (eds.), Readings in The Concept and Measurement of Income, Philip Allan Publisher Ltd., Oxford,1986.

IADB, 1998­1999, Facing Up to Inequality in Latin America, Economic and Social Progress in Latin America, Washington.

Janvry, A. and Sadoulet, E., 1996, Growth, Inequality, and Poverty in Latin America: A Causal Analysis, 1970­94, Working Paper No. 784, Department of Agricultural and Resource Economics, Division of Agriculture and Natural Resources, University of California at Berkeley.

______, 1999, Growth, Inequality, and Poverty in Latin America: A Causal Analysis, 1970­94, Inter­American Development Bank, Conference on Social Protection and Poverty organised by the Poverty and Inequality Advisory Unit of Department, Washington D.C.

Jha, Sailesh K., 1996, "The Kuznets Curve: A Reassessment", World Development, Vol. 24, No. 4, pp.773­779.

Kakwani Nanak, 1993, "Poverty and Economic Growth with Application to Cote D'Ivore", Review of Income and Wealth, Series 39, No. 2, June, pp.121­139.

Kaldor, Nicholas, 1956, Alternative Theories of Income Distribution, Review of Economics Studies, Vol. 23.

______, 1980, Essays on Value and Distribution, Duckworth, second edition.

Kalecki. Michal, 1939, Essays in The Theory of Economics Fluctuations, George Allen and UNWIN Ltd.

29 Kirby, Peadar, 2000, Growth with Inequality: The International Political Economy of Ireland's Development in the 1990s, unpublished PhD thesis for the London School of Economics.

Knight, J. B., 1976, "Explaining Income Distribution in Less Developed Countries: A Framework and Agenda", Oxford Bulletin of Economics and Statistics, Vol. 38, No. 10, October, pp. 161­177.

Kurz, Heinz, 1994, "Growth and Distribution", Review of Political Economy, Vol. 6, No. 4, pp.393­420.

Kuznets, S., 1954, "Underdeveloped countries and the pre­industrial phase in the advanced countries", in A.N. Agarwala, and S.P. Singh (eds), The economics of , Oxford University Press, New York.

______, 1955, "Economic Growth and Income Inequality", The American Economic Review, Vol. XLV, No. 1, pp. 1­28.

Lebergott, Stanley, 1959, "The Shape of the Income Distribution", American Economic Review, June.

Leightner, Jonathan E., 1992, "The Compatibility of Growth and Increased Equality: Korea", Journal of , Vol. 29, No. 1, pp. 49­71.

Lessard, D. R. and J. Williamson, 1987, Capital Flight: The Problem and Policy Responses, Institute for International Economics, Washington, D.C.

Lewis, W. Arthur., 1954, "Economic Development with Unlimited Supplies of Labour", The Manchester School of Economic and Social Studies, Vol. 22, pp. 139­191

Lindahl, Erik, 1933, "The Concept of Income", printed from Economics Essays in Honour of Gustav Cassel, London: Allen & Unwin, pp. 399­407, in R. H. Parker, G. C. Harcourt, and G. Whittington (eds.), Readings in The Concept and Measurement of Income, Philip Allan Publisher Ltd., Oxford, 1986.

Lipton, M. and Ravalion, M., 1995, "Poverty and Policy", Chapter 41, in J. Behrman and T.N. Srinivasan (eds), Handbook of Development Economics, Volume 3B, Elsevier Science BV.

Lustig, Nora, 1995, Introduction, in Nora Lustig (ed.), "Coping with : Poverty and Inequality in Latin America", The Brookings , Washington DC.

Mason, Andrew, 1988, "Savings, Economic Growth and Demographic Change", Population and Development Review, 14, March.

Mátyás, L., Kónya, L., and Macquarie, L., 1998, "The Kuznets U­Curve Hypothesis: Some Panel Data Evidence", Applied Economics Letters, 5, pp. 907­912.

Meade, James, 1964, Efficiency, Equality, and The Ownership of Property, Allen & Unwin, London.

______, 1976, The Just Economy, Allen & Unwin, London.

Mincer, J., 1958, "Investment in Human Capital and Personal Income Distribution", Journal of Political Economy, Vol. 66.

Morley, Samuel, 1995, Poverty and Inequality in Latin America: The Impact of Adjustment and Recovery in the 1980s., The Johns Hopkins University Press, Baltimore.

Myrdal, G., 1968, Asian drama, Penguin Books.

30 Papanek, G., and O. Kyn, 1986, “The Effect on Income Distribution of Development, the Growth Rate, and Economic Strategy”, Journal of Development Economics, Vol. 23, pp. 55­ 65.

______, 1987, "Flattering The Kuznets Curve: The Consequences For Income Distribution Of Development Strategy, Government Intervention, Income And The Rate Of Growth", The Pakistan Development Review, Vol. 26, pp. 1­54.

Paukert, F., 1973, "Income Distribution at Different Levels of Development: A Survey of Evidence", International Labour Review, CVIII, Aug.­Sept.

Perotti, R. 1993, "Political Equilibrium, Income Distribution, and Growth", Review of Economics Studies 60, pp. 755­776.

Persson, T. and Tabellini, G., 1992, "Growth, Distribution and Politics", European Economic Review 36, papers and proceedings, pp. 593­602.

______, 1994, "Is Inequality Harmful For Growth? American Economic Review, Vol. 84, pp. 600­621.

Psacharopoulus, George; Samuel Morley; Ariel Fiszbein; Haeduck Lee; and William Wood, 1995, "Poverty and Income Inequality in Latin America During the 1980s", Review of Income and Wealth, Series 41, No. 3, pp. 245­64.

Ram, Rati, 1991, "Kutznets' Inverted­U Hypothesis: Evidence from a Highly ", Southern Economic Journal, April, 57, pp. 1112­1123.

______, 1993, "Kutznets Inverted­U Hypothesis ­ Reply", Southern Economic Journal, Vol. 59, No. 3, pp. 528­532.

Ranis, G. and J. Fei, 1961, "A Theory of Economic Development", American Economic Review, vol. 51, pp. 533­565.

Ranis, Gustav, 1962, "Investment Criteria, Productivity and Economic Development: An Empirical Comment", Quarterly Journal of Economics, May.

Ravallion, Martin and Shaoua Chen, 1997, "What Can new Survey Data Tell Us about Recent Changes in Distribution and Poverty?", The World Bank Economic review, Vol. 11, No. 2. pp. 357­82.

Ravallion, Martin and Gaurav Datt, 1996, "How Important to India's Poor Is the Sectoral Composition of Economic Growth?", The World Bank Economic Review, Vol. 10, No.1, pp. 1­ 25.

Ray, Cadwell L. and R. Lynn Ritternoure, 1987, "Recent Regional Growth Patterns: More Inequality", Economic Development Quarterly.

Ray, Debray, 1998, Development Economics, Princenton University Press, Princenton, New Jersey.

Sahota, Gian, S., 1978, "Theories of Personal Income Distribution: A Survey", Journal of Economic Literature, March 1978, pp. 1­55.

Saint­Paul, G. and Verdier, T., 1992a, "Education, Democracy, and Growth", CEPR discussion paper, 613, Cambridge, USA.

______, 1992b, "Distributional Conflicts, Power and Multiple Growth Paths", CEPR discussion paper, 633, Cambridge, USA.

31 ______, 1992c, "Historical Accidents and the Persistence of Distributional Conflicts", DELTA document 92­07, Paris.

Saith, Ashwani, 1983, "Development and Distribution. A Critique of the Cross­Country ­ hypothesis", Journal of Development Economics, Vol. 13, pp. 367­382.

Sen, Amartya, 1973, On Economic Inequality, Claredon Press, Oxford.

______, 1985, "The Standard of Living: Lecture I ­ Concepts and Critiques'", in G. Hawthorne (ed.), The Standard of Living, Cambridge University Press, Camdridge.

Simons, H. C., 1938, "The Definition of Income", printed from Personal Income Taxation, University of Chicago Press, pp. 41­58, in R. H. Parker, G. C. Harcourt, and G. Whittington (eds.), Readings in The Concept and Measurement of Income, Philip Allan Publisher Ltd., Oxford,1986.

Smith, A., 1884, The Wealth of Nations, (ed. J. S. Nicholson), Nelson, London.

Smolensky, E., Plotnick, R., Evenhouse, E., and Reilly, S, 1994, "Growth, Inequality, and Poverty: A Cautionary Note", Review of Income and Wealth, Series 40, No. 2, June, pp.217­ 222.

Solow, R. M., 1962, "Technical Progress, Capital Formation and Economic Growth", American Economic Review, Vol. 25.

Thurow, L. C. and R. Lucas, 1972, The American Distribution of Income: A Structural Problem (Joint Economic Committee, Congress of the United States, Washington, D.C.).

Thurow, L. C., 1975, Generating Inequality: Mechanics of Distribution in the US Economy, Basic Books, New York.

Todaro, Michael, 1994, Economic Development, fifth edition, Longman Group.

Wratten, Ellen, 1995, "Conceptualising Urban Poverty", Environment and Urbanisation, Vol. 7, No. 1, Institute for the Environment and Development, London. pp. 11­36.

32 ENDNOTES

i This concept will be discussed more formally in the next sub­section. ii Atkinson (1983) considers that income can be disposed in one or a combination of the following ways: expenditure, saving, and capital transfers to others. Thus, the difference between income and saving would be, in this case, equal to expenditure plus capital transfers. iii money received = money income + savings; cost of living = money income + borrowings; money received = cost of living + savings ­ borrowings iv There is an inertia effect on birth rate. In poor countries population is young with high levels of fertility. Also, with limited flow of information, families tend to reproduce their decisions regarding fertility. For a more detailed discussion see Ray (1998, Chapter 9). v Alessina and Perotti (1993) present a critical survey of this literature. vi I will return to this point in Section 3.5 with empirical evidence. vii This theory was worked out in the late nineteenth and early twentieth centuries led by Alfred Marshall. viii In fact, since sp = 1­ cp , where cp is the propensity to consume of capitalists, from P = (1/sp)I we get P = I + cpP ix The definition of poverty varies among the works reviewed in this section. The poorest quintile of the population is used by Fields (1991), Deininger and Squire (1997), and Jha (1996), while methodologies based on definition of a poverty line according a minimum requirement of household income per capita per day is used by Kakwani (1993), Smolensky et al (1994), Psacharopoulos et al (1995), Morley (1995), Ravallion and Datt (1996), Janvry and Sadoulet (1996, 1999), and Ravallion and Chen (1997). x Adelman and Fuwa (1994) report an additional number of works which using cross section regressions confirm the existence of the Kuznets curve. These works are by Bacha (1979); Chenery et al (1974); Ahluwalia, Carter and Chenery (1979); Anand an Kanbur (1986); and Papanek and Kyn (1986, 1987). xi Among the studies examined by Fields (1988) those which considered structural and policy factors are Chiswick (1971), Adelman and Morris (1973), Chenery and Syrquin (1975), and Ahluwalia (1976). xii Qualitative features of growth: predicted growth rate of the Gini coefficient; length of growth or recession sequence; difference in growth of value added between agriculture and , and agriculture and services; inflation; real exchange rate growth; international terms of trade; coefficient of variation of GDP per capita; migration rate; and urban minimum wage growth. Structural features of growth: initial GDP per capita, share of agriculture in GDP, initial level of inequality, initial urban and rural poverty levels, natural growth rates of urban and rural population, initial shares of urban and rural in total population. xiii The time­series observations for the U.S. cover the period 1947­1988, and four cross­ sections of states pertain to the census years 1949, 1959, 1969 and 1979. (Ram, 1991, p. 1113).

33 xiv Fosus (1993) makes methodological objections to Ram's (1991) work referred to the use of inequality measures based on family incomes with no control for family composition. Ram (1993) replies showing that the addition of the family composition variable does not add any difference to his original estimates. Also, he points out that many other scholars such as Braun (1991), Bishop, Formby, and Smith (1991), Coughlin and Mandelbaum (1988), and Ray and Ritternoure (1987), have reported strong evidence that shows either the U­shape for the relationship growth­inequality for the USA or brings out increasing inequality over the 1970s and/or the 1980s. xv It is interesting to notice that Harris (1993) argument is in line with Saith (1983) and Adelman and Fuwa (1994), but from a different point of view and in a different economic context. xvi Todaro (1994) supports his view on research by Ranis (1962), Grupta (1970), Lessard and Williamson (1987), and Mason (1988), and Myrdal's (1968) view. xvii Actually, later Deininger and Squire (1998) question the validity of this negative relationship between income inequality and growth. xviii The argument is that redistributive taxes reduce investment incentives and, as a consequence, the rate of accumulation. xix This is based on the view that lower stock of human capital determine lower growth. xx The works reviewed by Garcia (1994) are: Alesina and Rodrik (1991) and Persson and Tabellini (1992, 1994) for the redistribution approach, and Galor and Zeira (1993), Glomm and Ravikumar (1992), Saint­Paul and Verdier (1992a, b, c), Garcia Peñaloza (1993), and Perotti (1993), for the educational approach.

34