Tuesday, June 3, 2014

Nit-Piketting continued

Debraj Ray’s original review of Piketty’s book was, appropriately, called “Nit-Picketty.” His reply to my critique of his critique, published here, could be called Nit-Picketting II or nit-Picketting big time. It brings no new points and thus leads our discussion to a close. I would organize my own nit-Debrajing into two parts, substantive and stylistic.

Substantive. Debraj has identified a weakness in Piketty’s model on which he spends some nine pages, denying the importance of the key insight of the book, and that weakness is as follows.  r>g is not sufficient to guarantee either an increasing share of capital in total income (over time), nor increasing inter-personal inequality (over time, too) because it does not take into account what capitalists do with the r which they receive as income. They can consume it all, or consume enough of it so that the portion which is reinvested (r*) becomes less than g.  Never mind that historically in the literature r was used, as Piketty does in his book,  to indicate both the rate of return on capital and the subsequent rate of increase of the capital stock (under the simplifying but sensible assumption that capitalists just turn around and reinvest all their capital income). Never mind that all empirical studies show that propensity to save increases with income and that capitalists tend to be rich, and hence save proportionately more, so that their consumption will be insufficient  to overturn the r>g inequality.  Debraj has found a small, even trivial, point, and yes, he insists  r has to be written as r*=r-cc,  where cc is capitalists’ consumption out of their return (all expressed in terms of capital stock), and if Piketty were to  write r*>g  then, yes, everything would be fine, but you see, just writing r>g is all wrong. Without cc, no going.  r>g  “implies nothing”, but r*>g would solve everything.  

Why not go further? Do we really know what capitalists will do with their r? They might give away a fair share of it. Denote the portion  of their capital return which they donate by dd, and then write a new relationship that should be even more satisfying: r-cc-dd>g.  However absurd this game of adding additional ways capitalists can use not to reinvest their full capital income is, adding  dd makes more sense than adding cc.  At least it is empirically justified. While for consumption/income ratio we know that it goes down as income goes up, for charity we know that it behaves exactly in the way to reduce r* as it increases with income. So capitalists will indeed have a positive dd, while for most of the others dd will be zero. Here I have now discovered that Piketty’s conclusions need to be modified in a yet another important way.

Fundamentally, Debraj and I do not disagree.  It is just that I think that his point is (yes, I must say so again: the 2nd time) trivial, and while logically correct,  all empirical evidence about savings, and all our views about the essential features of capitalism,  show that it is not valid.  For it to be valid, either rich people should not be capitalists, or average propensity to save of the rich people should be less than of the poor. Manifestly, neither is true, and Debraj agrees with it. In other words, Debraj wants to drive his small logical point to an end which is in contradiction with everything which we know about the behavior of capitalists and the nature of capitalism.

Stylistic. I personally like Debraj and admire his work. I was extremely impressed by his presentations and writings. He is a brilliant economist. In addition, for those who do not know him, he looks like a Greek demi-God. His thinking is at the same level. But even demi-Gods should get used to mortals disagreeing with them, and even of having to repeat the same thing 97 (or was it 98?) times.  

Monday, June 2, 2014

Where I disagree and agree with Debraj Ray’s critique of Piketty’s Capital in the 21s Century.

Two people whose opinion I hold in very high esteem have told me, “Read Debraj Ray’s critique of Piketty; It is the best written yet.” I have read it before, but only in parts, and have focused on a few rather unimportant points in the first part of Debraj’s piece. So I decided now to read it carefully, and in full. It is a short piece, some 9 pages long. It is very well and clearly written. There are many things with which I agree.  But there are also many with which I do not. Let me start, because it is more fun, with the latter.

Debraj’s s key point is summarized in the last two sentence of his annex: “All I am saying is that it [any growth model] explains why r exceeds g. But the fact that r exceeds g explains  nothing about the rise in inequality.” At face value, it is a pretty damning critique of Piketty. First, r>g inequality is, as Debraj shows, a feature of any model of growth. It has nothing to do with being a “contradiction of capitalism.” Second, if inequality increases it is not because r is greater than g, as Piketty argues, but because capitalists have a higher marginal propensity to save. A bit earlier Debraj calls the much praised r>g inequality a “red herring.”

Let me now explain why I disagree with Debraj. While r>g (or r>=g) may be a feature of all growth models it is still a contradiction of capitalism for three reasons: because returns from capital are privately owned (appropriated), because they are more unequally distributed (meaning that the Gini coefficient  of income from capital is greater than the Gini coefficient of income from labor), and finally and most importantly because recipients of capital incomes are generally higher up in the income pyramid that recipients of labor income. The last two conditions, translated in the language of inequality mean that the concentration curve of income from capital lies below (further from the 45 degree line) the concentration curve of income from labor, and also below the Lorenz curve. Less technically, it means that capital incomes are more unequally distributed and are positively correlated with overall income. Even less technically, it means that if share of capital incomes in total increases, inequality will go up. And this happens precisely when r exceeds  g.

It is indeed a contradiction of capitalism because capitalism is not  a system where both the poor and the rich have the same shares of capital and labor income. Indeed if that were the case, inequality would still exist, but r>g would not imply its increase. A poor guy with original capital income of $100 and labor income of $100 would gain next year $5 additional dollars from capital and $3 from labor; the rich guy with $1000 in capital and $1000 in labor with gain additional $50 from capital and $30 from labor. Their overall income ratios will remain unchanged. But the real world is such that the poor guy in our case is faced by a capitalist who has $2000 of capital income and  nothing in labor and his income accordingly will grow by $100, thus widening the income gap between the two individuals.

So, if K/L ratios were the same along the entire income distribution, r>g would not have any special meaning. I grant that to Debraj. But precisely because the K/L ratio in capitalism is most emphatically not equal along income distribution (and we do not have a single historical example  where it was), but is rising, we do get increasing income inequality driven by r>g.

But, now, Debraj’s answer to that is to point out that this condition is not sufficient for rising inequality.  The requirement for inequality to go up is also that capitalists  should not spend most or all of their income, but rather invest it. Let us be clear: If capitalists were spendthrifts and used all their income on consumption,  there would be indeed no accumulation of capital, we would be in Marx’s simple reproduction, and Debraj would be right. But again this is not the  world we live in.  Not only is the assumption that capitalists save and reinvest 100% of their capital income a standard one in growth models (not because it is fully accurate but because it keeps the main features of the things in sharp relief; remember Marx’s “Accumulate, accumulate, it is Moses and all the prophets”). It also corresponds with empirical data. We indeed know that richer people tend to save greater proportion of their income (and people who receive more of their total income from capital tend to be richer). The other way to state that is to say that propensity to save out of capital income is greater than propensity to save out of labor name. This also tends to be empirically true.

Thus Piketty’s mechanism survives both of Debraj’s attacks. The standard growth requirement that r>g is indeed a contradiction of capitalism because in capitalism, capital owners are private persons and they tend to be rich.  And these rich people save more of their income, and gradually build up more and more of their capital stock, thus creating precisely the process of divergence of which Piketty speaks.

Now, why has Debraj gone astray in his otherwise very lucidly argued paper? Because he essentially assumes that economic processes described by theory of growth are abstract and do not take place in a capitalist environment where indeed the facts are such that capitalists tend to be rich (and not poor), and where such capital owners do not spend all of their income on consumption.

To conclude. If we had a capitalism  where capitalists were poor or a capitalism where capitalists would spend all of their capital incomes on booze and trinkets, yes, Debraj’s critique of Piketty would be right. But it just so happens that these are not the features of contemporary, nor of any other known, capitalism at least over the past 200 years.

This typically neoclassical and ahistorical approach to basically forget that the economic processes occur within a capitalist society, is also the reason why Debraj is wrong to wonder why Piketty calls a mere  identity, namely that the share of capital income in total income is equal to the rate of return  times capital/income ratio a fundamental law of capitalism. Seemingly reasonably, from his ahistorical point,  Debraj exclaims: this is just a definition of capital share in total output! It is an identity. How can it be a fundamental law?

Yes, it can. That definition carries weight in capitalism because it defines how much of total output  will be taken by a special class of population that does not have to work for its income. The more it gets, the less is left for those who have to work. It is not a Fundamental Law of Economics but it is a Fundamental Law of Capitalism.  (I made this point  in my review of Piketty’s book published on the Internet in October last year, forthcoming in the June issue of  the Journal of Economic Inequality). And to realize this, suppose that all capital is socially-owned and that income from capital is distributed equally among all population. Then clearly, capital share would be still the same as before, but this would not be a fundamental law nor a systemic problem since it would not imply unequal distribution of income, nor the existence of a class that does not need to work.

Debraj’s error consists, in my opinion, in not realizing that normal capitalist relations of production (where capitalists tend to be rich) are forgotten when we look at economic laws in an abstract manner. Not doing that is precisely a great virtue of Piketty’s book. Surely, (a) if capital/labor proportions were the same across income distribution; (b) if, more extremely, capitalists were poor and workers rich; (c) if capital were state-owned, all of these contradictions would disappear. But none of (a)-(c) conditions holds in contemporary capitalism. So Piketty’s economic laws and contradictions of capitalism do exist.

Where do I agree wit Debraj? That Kuznets curve cannot be easily dismissed. I am currently working precisely on this idea, alluded to also by Debraj, that we are now witnessing the upswing of the second Kuznets curve since the Industrial revolution. Moreover I believe this is not only the second but perhaps fifth or sixth or tenth such curve in western economic history over the past 1000 years. This is my next project, and there I come very close to agreeing with Debraj.

Does this agreement on Kuznets then, by itself, imply that my defense of Piketty’s mechanism cannot be right or consistent? Not at all. Piketty isolated the key features of capitalist inequality trends when they are left to themselves: the forces of divergence (inequality) will win. But there are also other forces: capital destruction, wars, confiscatory taxation, hyperinflation, pressure of trade unions, high taxation of capital, rising importance of labor and higher wages, that at different times go the other way, and, in a Kuznets-like fashion, drive inequality down. So, I believe, Piketty has beautifully uncovered the forces of divergence, mentioned some of the forces of convergence, but did not lay to rest the ghost of Kuznets inverted U shaped curve.

Sunday, June 1, 2014

Limits of neoclassical economics

When people criticize Piketty for elevating  a mere economic identity that shows that the share of capital income in total income  is x times y (not important for my argument) to a Fundamental Law of Capitalism they show their inability to go back to economics as a social science, or differently, to transcend neoclassical economics.

The share of capital income in total income is not only a reflection of the fact that people with a factor of production B have so much, and people with the factor of production A have the rest. It gives us a measure of the share of the total pie that owners of capital (who are the principal social group in capitalism) are able to claim for themselves without having to work. This is key. We are basically saying: 20% of people of  the richest people claim one-half of national output and they do so without having to work. If it were a question of changing the distribution in favor of factor A (donuts) and against factor B (pecan pies), there would be no reason to be concerned. But here you change the distribution in favor of those who do not need to work, and against those that do. You thereby affect the entire social structure of society. This is where social science comes in, and neoclassical economics goes away.

The entire 100 years  of neoclassical economics was, in part, driven by the attempts to make us forget this key distinction: between having or not having to work for a living. Hence neoclassicists like to treat capital (and labor) as basically  the same thing: factors of production: a donut and a pecan pie. You have a bit more of a donut, I have a bit more of the pecan pie. No big deal. Thence also the attempt to treat labor as human capital. We are all capitalists now: a guy who works at Walmart for less than the minimum wage is a capitalist since he is using his human capital; a broker who makes a million in a day is also a capitalist, he just works with a different type of capital.

The true social reality was thus entirely hidden. But once you bring back the sharp dividing line between those work to earn and those who earn by doing practically nothing, you are back into a social science and you ask yourself questions like, would a society where 20% of non-workers earn 70% of total income be okay? What are the values that such a society would promote?  What would be the consumption patterns it would encourage? What is the  religion or ethical system that says that it is fine that people who do not work should be rich?  (Can we find one?) And what about equality of opportunity as such wealth would be transmitted to their kids.

The principal, and stark, issue then becomes: can a society where most of the rich are non-workers be called a  “good society"?