The objective of the book and the criteria for choosing the six economists (Quesnay, Smith, Ricardo, Marx, Pareto and Kuznets)
The objective of my book was to look at how most important, in a way, canonical economists, have thought about income distribution. As I say in the Introduction, there are two parts to the questions that I am interested in and that I am metaphorically asking them. The first is, what are the forces they see as determining income distribution in the place and time where and when they were writing. The second is how they see the evolution of income distribution in the future, again of course taking account the time and the place where they are situated. Mine is, I think, the first book that looks at the canonical economists through this angle only: income distribution. The initial title of the book was “Through the Lens of Inequality”. I thought it was a very good title but the publisher did not like it. I do not look at the economists profiled here in the way that for example Schumpeter or Heilbroner do, trying to cover their entire oeuvre. For example, I do not engage with Ricardo’s and Marx’s labor theory of value. But as the readers will see, distribution is not a negligible part, far from it, of their work.
The choice of the economists was not difficult because the six of them are, as I said, canonical figures generally studied in history of economic thought. Not of course that they have been studied with the objective of looking at income distribution but in principle they were and are studied. But there is also a subjective or idiosyncratic element reflective of my own preferences and as importantly of my own knowledge or lack of it.
All six of them are the authors whom I have read for many years and when I started writing the book there was relatively little, I would say probably not more than ¼ of their writings (and some of them were letters, esp. Ricardo’s and Marx’s), which I have not read before. I have written papers on, and taught, Quesnay, Marx, Pareto, Kuznets; so I was quite familiar with them and that familiarity influenced also my choice. When it comes to the choice of authors I had only two or three doubts which I also mentioned in the Introduction. If I knew more about Latin American structuralists I might have included somebody like Celso Fortado for example. I also thought of including Samir Amin and maybe at the end of our conversation when we talk about more recent developments I would explain why I did not. The same applies to Keynes.
Why is Quesnay important?
I think there are two elements in Quesnay’s work that are quite important with regard to income distribution. Quesnay was the first economist to have introduced explicitly social classes in the economic analysis, and the first to have defined the social surplus. In fact, the two are related. The social classes introduced by Quesnay reflect the pre-revolutionary France with the legally defined social classes. (Quesnay was writing in 1750-60s.) The top class or the elite is called “les propriétaires” and is composed of three subclasses, namely aristocracy, clergy, and government. The three subclasses provide functions that are necessary for the survival of a society: they provide land on which food is grown, moral “sustenance”, and administration of justice and protection from external aggression. But wherefrom are they paid? Wherefrom their income comes? And here is that second very important element: they are paid out of a surplus.
This is the first introduction of the surplus which arises in production not in exchange as the mercantilists believed. This is a development whose importance cannot be overestimated. It is basically the introduction of net income, or, if you will, of GDP. It answers the question, Why we work? Does our work create something more than covering depreciation of instruments and “depreciation” of the labor power? Of course that made Quesnay very attractive for Marx and Marx wrote innumerable pages in his Theories of the Surplus Value on Quesnay. As did other economists later.
The Theory of Moral Sentiments (TMS) and the Wealth of Nations (WoN) according to Amartya Sen and the Chicago school: “leftist” and “rightist”.
I do actually disagree with Amartya and I believe that he has misinterpreted Adam Smith. It sounds like a big claim but let me explain what it is based on. The Theory of Moral Sentiments is written when Adam Smith was a young man and professor of moral philosophy at the University of Glasgow. TMS was (with some exaggeration, since he already held the job) his job-market paper. The book was supposed to reinforce his position as a moral philosopher, that is, in those days, as a protector of religious feelings. Reflecting the intellectual standards and the conventions of the time, and the job he had, TMS is written as a quasi theological treatise. When it comes to income distribution, TMS justifies it in its existing form. It comes close to being Panglossian. Smith never questions the right of those who are on the top to remain there. He never poses the question of the origin of their property. (We shall see that it is all very different in the WoN). He indeed makes fun of the rich, he ridicules their infatuation with babbles, he derides their silly pride and vanity. That’s all.
But does he believe that they have no right to this wealth, that they have to share it, that they should be expropriated, that such manifest inequality is unjust? No, never. Since the book is quasi theological or theosophic it moves moreover very quickly into strong defense of the status quo. It is very interesting how Smith defends the status quo in the TMS. Of course, he observes large inequality in wealth and income. Isn’t God then responsible for this, one wonders? Smith answers: sure, God is not unaware of this manifest inequality in wealth, but he is not indifferent to the fate of the poor. What he does not give them in the form of wealth, he gives them through other advantages: for example, they don’t have to worry and fret about their property because they have none; nobody is going to rob or kill them since there is nothing to steal; they can, as Smith says, “sun [themselves] by the side of the highway [and] possess that security which kings are fighting for” (TMS, Part IV.I.10), indeed a luxury unattainable by the kings who always have to worry about being deposed or attacked. So he defends the existing order by making claim that the God has not forgotten the poor and that God has given them some other advantages which are not income-based but that compensate for their material poverty. It is a pretty heavy handed defense of inequality of property and income.
What is the difference with the WoN? Indeed it is true that the WoN is based on self-interest but there Smith is absolutely scathing about the top classes and in particular the capitalists, who are called in the book the employers or masters because the term capitalist had not existed in Smith’s time. Thus when he lists how the elite (we could say, the top 1 percent) have become rich, there is a litany of misdeeds. They got rich because of the use of monopoly power, plunder of foreign lands, exploitation of the weak, use of slave labor, collusion (which is much more easily done among capitalists than among workers because of the smallness of the former group) and finally through the control of government. When you look at how the top class is treated in the WoN, it is treated as a group of people who got there through violation of either free competition, or the use of compulsion or total disregard of moral norms. There is hardly any justification for their high incomes.
I think this part is very seldom quoted maybe people don’t read the WoN anymore, or, in the selective choice they are presented, they read only parts which are anti-government (which of course exist), but not parts that are anti-capitalist. I listed intentionally in the book on a page and a half all the excerpts where Smith talks, in strongest terms, about how the rich have become rich. Now of course the part that is quite well known is his critique of the East India Company and also his critiques of the merchant states of Venice, Pisa and Genoa. This also brings to the fore his extraordinary negative view about the political power of merchants or capitalists. In some ways, Smith’s critique of “primitive accumulation” is stronger even than that of Marx, but I can explain that later.
Controversial point made in the Marx’s chapter: it is wrong to believe that Marx held the so-called “theory of immiseration” whereby over longer term, in advanced capitalism, incomes of workers will be reduced to the subsistence and incomes of the capitalists would be heavily concentrated, and thus inequality would be extremely high. It is just one of four possibilities that exist in Marx. What are the others and why the “immiseration theory” which is quite popular among the readers of Marx is wrong?
I believe that the theory which is now most commonly associated with Marx about the evolution of income distribution under capitalism is wrong: it is based on almost no textual or analytic evidence that can be found in Marx. Let me focus on wages first because they are the most important part of that story. Wages for Marx are a historical category: they have two components: one that is historical and depends on the level of development of a given society and another which is physiological. (Even the physiological part varies from society to society because, for Marx, subsistence minimum is also a historical category as are our wants, but I will come back to this in a moment.) The historical component, as I said, depends on the level of development and this is very clearly said by Marx many times. He even provides numerical examples on the difference in the real wage between England, France, Austria, Prussia and Russia. Thus development, or what we call today increase in GDP per capita, would be simply incompatible with the real wage staying at the same level because development increases both the historical and even the physiological part of the wage (the latter through the creation of new needs). So both textually and more importantly analytically, Marx accepts the likelihood of the increase in real wage.
Furthermore, all economic categories in Marx are historical. There is not one economic category or phenomenon that is not historical. Subsistence in terms of real goods in Roman Empire is not the same as the subsistence in the 19th century Britain. Simply because certain goods did not exist in Roman Empire and they could not be part of our needs or desires or what we now call, following Bob Allen “the respectability basket”, or even “the bare-bones basket”. So the wage too has to be a historical category. People who believe that the wage somehow would always stay at the subsistence are speaking against a truly fundamental analytical argument in Marx: namely, as I said before, that all categories are historical. And when the citations from Marx are produced in support of the “immiseration theory” they come from only two mentions that can be interpreted in such light: one is in The Communist Manifesto which was written when Marx was 29 years of age and when he basically didn’t know much about political economy and another quote in Capital which can also be interpreted in the same light. But there are no other citations.
The second part of income distribution obviously deals with what happens to the rate of profit in the long term. And there, and without going into all of the details because it is one of the three big issues in Marxist economics that have occupied generations of economists, I think that the arguments that he made about the tendency of the rate profit rate to fall as the organic composition of capital increases are pretty strong. And then if you take these two things together: that the real wage depends on the level of income of the country and that the rate of return to capital declines with the wealth of the country then these two things imply that inequality is likely to be reduced. Income per unit of labor goes up; income per unit of capital goes down. I think this is the most reasonable interpretation, based on Marx’s writings—because, as we know, he never came up with a finished theory of income distribution or, more exactly, of factoral income distribution.
I also take note of other possible interpretations which are essentially derivatives but I don’t think that the case for them is as strong as for the one I discussed. One possibility is to combine the increasing real wage with the greater concentration of capital income because of an argument that Marx also makes, i.e. that capitalists will become fewer in numbers. In that case we would have polarization, a situation which is not very dissimilar from the current situation in the USA: greater top 1% share is accompanied by an increase in real wage. Polarization goes up even if inequality does not.
Marx held that the distribution in advanced capitalism will be characterized by (I) higher real wage, (ii) reduced labor share, (iii) increased organic composition of capital (high K/L ratio or capital deepening) and (iv) lower rate of profit, and (v) concentration of capital incomes.
Wasn’t Marx interested in reducing income inequality under capitalism?
Marx says it quite a few times clearly, Engels even more so (in the Critique of the Gotha Program). The objective of the Communist movement is abolition of classes. Abolition of classes means end of privately owned means of production and thus end of property-based incomes. Only after that point can we meaningfully discuss optimal level of inequality. Prior to that, as Marx writes, it makes as much sense as to discuss political freedoms within a slave-based society. In other words, Marxism is not some kind of leftist social-democracy. Social democracy or Fabianism accept capitalism and try to improve distribution. Marxism rejects capitalism,. That’s why Marxism is simply impossible to absorb within capitalism. It asks capitalists to disappear. They obviously are not keen to do that.
Moving on to Neoclassical Economics: what is attractive in Pareto?
When we study Pareto we have to start with his ideological approach and reasons which led him to believe that he has found an “iron law of income distribution” and then to argue that the coefficient on top incomes (and he knew that they were top incomes because, as nowadays, all his data came from fiscal sources) is stable. The ideological basis was the rejection of Marxism. The rejection of Marxism, joined with his view of the circulation of the elites, meant that for Pareto all societies are indeed divided societies, but not as Marx thought, divided solely by the ownership or not of the means of production, but also by a number of other dividers and cleavages (background, ethnicity, race, other interests). Following Machiavelli, he believed there would be always an elite that rules. The elites are of two different kinds: lions who rule by force, and foxes who rule by guile. Societies may be different but an elite would always exist and rule.
This is what he says in his first big book written against the Marxist system, Les systèmes socialistes in 1902. He thought, and was to a large extent proven correct by the later really-existing socialist systems, that the elite in the new system would be recruited differently and would be an elite of bureaucrats who would have higher incomes than the rest of the population. Just like capitalists under the really-existing capitalism. They would not be the same people as in capitalist societies, of course, but that’s immaterial: they would be an elite.
When he found, to some extent by accident (although he was much helped by his mathematical knowledge) that fiscal data from various European cities and states gave on the double log scale a straight line such that the percentage of income holders having an income greater than x decreases at the same rate throughout as the income threshold, x, is raised, Pareto thought that he had hit the jackpot. He found a validation of his view that regardless of a type of society income inequality would be the same. We know now that the Pareto’s coefficient is not constant at all, that it varies within a given income distribution, obviously even more between the distributions, but he was the first to have defined the power law which is a very powerful tool used not only in income and wealth distributions but in a number of other phenomena from the size of the cities to the size of the floods. And I do think that Pareto’s sociological contributions are not sufficiently appreciated today. Perhaps this would change as we are increasingly, and obviously, governed by the elites.
Kuznets was with Keynes the most influential economist of the 20th century because he was intimately involved in the definition of what is the first moment of income distribution, i.e. the mean income or GDP per capita, and the second moment which is heterogeneity of incomes or variance of incomes and thus income distribution. But the period after Kuznets was the eclipse of income distribution studies.
I’m critical about the contributions made then. I’m critical because I see several developments to have converged to make income distribution studies of much lower relevance and quality than before. I entitle that chapter “The Eclipse” and it covers the period from the mid-1960s to 1990s. By the way, the book ends around 1990, with the fall of the Berlin Wall which represents a symbolic end point not only of Communism but of the forces unleashed by the French revolution too, the era with which the book opens. I deal with post-1990s developments only in a very short epilogue.
Why was there an eclipse of income distribution studies? There are several reasons. The first was political: it was due to the fact that the capitalist system was in a struggle or competition with the socialist system and the socialist system claimed to have abolished classes. Incidentally, the first part of “The Eclipse” chapter which is quite long (about 40 pages) deals with income distribution under socialism. I will not discuss this here but I think it’s really an important part to read because only when you read that part and when you grasp both the political pressure that existed under socialism to claim that social classes have been abolished, and the objective difficulty of defining classes in such a system, can you better understand why an equivalent claim in the US was ideologically necessary. If the USSR claimed that they have abolished classes, the US could also claim they have abolished classes because social mobility is high, the background doesn’t matter, everybody can become a capitalist, basically the Horatio Alger story. In brief: if you claim not to have classes, so do we.
Secondly, there was an independent development in economic theory which saw individuals an interchangeable agents and thus got rid of the class structure too, by emphasizing, or rather claiming, simply that we are all agents maximizing our income under conditions of scarcity and constraints. Consequently somebody with huge capital income and somebody who is very poor and unemployed behave, from an economic point of view, the same. There is no need to distinguish them. They are both just agents. Representative agent further removed differences between people.
The third development was the funding of research by the rich who obviously had an interest to claim that income distribution doesn’t matter. So all of these elements made the studies of income distribution undergo an eclipse.
That does not mean that studies of income distribution did not exist. Actually, there were many: many more than ever before. But most of them, perhaps I would say 98% of them, can be classified into two groups, neither of which I find particularly illuminating. The first group were uber empirical studies of income distribution. They are fine and I have done them myself but I don’t think that other than addressing individual countries and individual situations they analytically or intellectually add very much to our knowledge. Then there were many studies which are entirely theoretical which use arbitrary or unlikely assumptions and I criticize them rather harshly. The example I use is Alan Blinder’s book entitled Towards a Theory of Income Distribution that has as much to do with earthy reality as if he presented income distribution from Mars. I read his book, or parts of the book, in 1975 or 1976 and even then I thought that this was an example of how to write about “income distribution” (note the quotes) while not writing of anything of relevance for the topic or for real people. These were the uber theoretical studies that had nothing to do with reality, present and past.
So we had a sort of bizarre situation with many very empirical studies that never questioned the social order and generally totally omitted the role of capital and politics, and on the other hand we had studies which had no relationship to reality. This is my critical view of that period.
And in terms of what I call early in the book “integrative” studies of income distribution, namely that relevant works should have a strong narrative, theoretical grounding (which can be verbal, not necessarily mathematical) and empirics, none of the work done during the period of the eclipse comes close to it.
If you then look back to the cases of all six authors studied individually, the three desiderata were fulfilled. Take the example of Kuznets: there is a very clear narrative which is based on the theory of modernization, and economic and political theory: you might dislike that theory or like it but it’s very clear and it accords with the fact of the transfer of labor from the agricultural sector to the industrial sector. It also explains why this might lead to an increase and later to a decrease in inequality. There is also the empirical basis for the Kuznets hypothesis, contested indeed, but existent and capable of testing. Is there anything similar that was produced after Kuznets and before Piketty? I think not.
People not included in the book who might be included.
I say in the Introduction that people who have contributed something new during the period of the eclipse were probably Latin American structuralists and neo-Marxists like Samir Amin. I am much more familiar with Samir because I read him exactly at the same time when I read Blinder: actually the two of them are juxtaposed as a negative and a positive example in “The Eclipse” chapter. Of Latin American structuralists I think I didn’t know enough and I didn’t want to discuss people with whose work I was not fully comfortable. Regarding Samir Amin, I came to the conclusion that while his early work was very important his later work was repetitive and did not add much new, even empirically, to what he wrote when he was in his twenties. This is a shame. Why was he important? I think he may still be the most original contributor in the period of the eclipse because he introduced that entirely novel idea that income distribution is not just determined by domestic forces but is driven by the interaction of domestic and international forces. If somebody looks at all the stuff that I’ve said until now he or she would notice that income distribution is always discussed within a single nation. In other words, it is always internal class structure or internal elites or in the case of Kuznets internal transfer of labor that determine income distribution. What Samir Amin said was that the internal power structure depends on the international position of the country, international division of labor, the North-South relationship, that empowers a given class to become the ruling class and to take the lion’s share of total income. It is not that if you study inequality in Egypt that you can just study it without regard to Egypt’s position in the global distribution of production and without looking at the North-South relations of structural inequality or dominance. That means that there is a group of people, call it the comprador bourgeoisie, that is maintained in power thanks to the external factors. This was, I think, quite revolutionary at the time because once you start thinking about the fact that income distribution within a country depends also on external conditions then of course your overall approach changes. You are quickly led to global structural factors and towards studies of global income distribution.
The last person whom I considered adding (and he is in the book but in a very small role) is Keynes. His omission provoked perhaps most questions. I thought of that very much while writing (almost as much as about Amin), particularly since I think I know The General Theory quite well and they are indeed mentions of income distribution in it. I thought that if I were to discuss Keynes I would have discussed it on three or four pages only because Keynes himself, in my opinion, was reluctant to write about income distribution. Now let me be very clear on that. When you read Keynes it’s very obvious that once he introduces the marginal propensity to consume which decreases as income goes up, redistribution is the way to raise the aggregate demand. This is a no-brainer. This is something that goes back to Sismondi, Hobson and even Marx in the parts of Marx’s writings where he takes a quasi under-consumptionist point of view. But Keynes, I think for political reasons, didn’t want to take that route. He didn’t want to be associated with the underconsumptionists who, he thought, belong to the “underworld of economics” and he decided not to use that door to redistribution that he so clearly opened himself. Rather, he decided to argue for increased government spending as a solution for the lack of effective demand. So, I thought that Keynes simply did not want to discuss distribution. Perhaps it was too leftist for him? Or perhaps he was afraid to be seen as a socialist crank?
He does indirectly mention it in “The Economic Possibilities for our Grandchildren” which is just one lecture and where he talks about the world of plenty and affluence, close to Marx’s communism, but where, precisely because it is the world of affluence and few working hours, goods would be shared by all—but objectively under such conditions there is close to nothing one can be saying about income distribution.
The developments in the past thirty years.
The developments there were/are much different and much better. After 1990, the capitalism vs. communism competition disappeared so things could became much more open. There was also a change in the neoclassical approach with the introduction of the heterogeneous agents who own both capital and labor in various quantities, and I think Tony Atkinson contributed to that in the latter part of his life through bringing back capital and property incomes, much more forcefully than before, into income distribution studies. We have had a plethora of studies of wage distribution, returns to skills and the like, which, in my opinion, are not studies of income distribution at all but that’s a different matter. Tony re-introduced capital and that reintroduction was then continued, in a bigger way, by Piketty who reintroduced an element totally missing during the eclipse: politics, class struggle, political fights. What I think is important in Piketty is not only an attempted integration of the theory of production with the theory of distribution but, more importantly, the “localization” of income distribution within the political struggle. It makes no sense to speak of inequality without speaking of politics. Income distribution is more than anything else a political phenomenon. Not only Marx, but Smith and Ricardo made this central to their view of the world. And for Piketty I will not quote only his well-known Capital in the 21st century but his earlier work on top incomes in the 19th century France where even more tightly politics and economics and class ideology are seen as one. This was revolutionary because in the past you had empirical work but now the empirical work was integrated with political or social developments.
There are, to finish, two additional important and positive development which I think will bear fruit in the future. We now have global income distribution studies because we have been able to get the data for most of the world and to compare incomes thanks to the ICP project. Global income distribution poses entirely different problems than within-country distributions. For example, what is the role of the global middle class when you don’t have any political framework within which that class can actually express itself?; is income convergence between countries leading to peace or to war? why some people think that China’s income convergence (a very good development in itself) is a threat to peace?; migration is no longer an exogeneous force but entirely endogenous etc.
Finally, regarding historical work, an increased availability of social tables is very important. This is so because it’s the only source, other than some accidental fiscal data, for the past and because it gives us the social composition of a country. This is something different and superior to the information on the top 1% or the top 5%. When we talk about the top 1% share we get one number only but we have no idea who these people are. In the social tables, we have the social origin of these people; thus you have in one society landlords on the top, in another, aristocracy, in the third, bourgeoisie or the government officials or perhaps the king and his entourage. Moreover, you have that information not only for the top but for the entire distribution. This is much richer than what you get simply by lining up all people according to their incomes and having no idea who is there.