Thursday, February 4, 2016

Thomas Piketty and Reality of Capital in the 21st Century

This post violates a couple of key principles of blogging - it is very long (so was the book) - as you will see I was not a fan of his work,  which I guarantee will be not read but will be quoted liberally by many on the left.   This originally was posted in my Goodreads account but I have done some edits beyond that.

Ideally a book called Capital in the 21st Century might address two sets of questions which are very current.   The first would be “what is likely to happen to capital in the 21st Century?”  Two prominent economists (Robert Gordon and Tyler Cowan) have argued that the boom in technological advancement is coming to an end - we’ve “picked all the low hanging fruit” (in Cowan’s words).   They both argue that those developments are a problem.  The second question is even tougher.   In all developed societies the last couple of decades (depending on how you count and when you begin) have witnessed a gradual increase in inequality in both incomes and wealth.   Those trends have happened almost in spite of which policies a country chooses to adopt  Inequality has grown in the last seven years in the US, even though the Administration is committed to reducing it.  So what are the causes of those changes and how can we as societies respond to them?

Another possible set of questions might address some issues about capital formation when much of what we are creating is tiny - in health and technology.  Does our understanding of what constitutes capital and how to develop it change when things are digitized and done on the nano level? I think we probably need to do some thinking about that.   But again Piketty fails to consider these questions and instead concentrates on the increasing inequality of income and wealth.   Like many other Malthusians he uses linear logic elegantly.   A lot of what he offers as logical chains is, at best, tautological.

Finally, there is a third topic worth discussing.   At each financial panic, those in the governing class assume that the best way to "smooth" the market is to enact a new law.   So after the Enron fiasco, we adopted Sarbanes-Oxley.   After the bank silliness, we adopted Dodd-Frank.   But there is a lot of evidence that those and other "reform" measures actually did nothing to fix the problem and may have actually created new ones.   So with Sarbox, we got a significant increase in the number of startups wanting to stay private and thus avoid the perils of the accounting rules in the law.    A few years after D-F we find that the number of start-up banks has gone to almost zero.   Both of those laws have changed the way capital is created an allocated in society - but Piketty chose not to look at the issues.

Let me offer ten questions about the book.    #1 - Why I didn’t I read this when it came out?   When the English version first came out I bought it immediately with the intention of letting it sit for a while before I tackled it. I thought that Capital in the 21st Century is likely to become something akin to two other economics standards (The Wealth of Nations /Theory of Moral Sentiments and Das Kapital).   Both are distinguished as being the most quoted and least read books in the field.  While I think most of Kapital is bunk, it has had a significant influence on the field.  These kinds of books enter into discussions but often with distortions from the original that are significant.

I also delayed reading the book, even though I read a lot of commentary when it first came out (and one of the best short critiques was in the Economist magazine - ) because I wanted to be able to take the book in without much noise.  Deirdre McCloskey also wrote a good review for Cato (  Finally Casey Mulligan wrote a review for the Independent Institute which I thought was excellent (  Mulligan's comment was especially funny (in an economist's perspective) he said even though the book had come to him free there was an opportunity cost attached to reading the book.   Indeed, there was.  The point is there was a lot of commentary on the book early on.   By waiting many months it was possible to let the dust settle.

#2 - Does an Economic Model Influence any Economist’s Perspective?  Right after James Buchanan won the Nobel in Economics, he gave a lecture at the University of San Diego.  In it, he described two ways to begin the study of Economics.   The first is to imagine supply and demand curves.   If the is X supply of a good then at some level of demand the suppliers and consumers will produce an optimal quantity to satisfy both.  That simple model can be expanded to all sorts of complexity.  But there is an alternative way to begin the study.  Assume that you produce apples and I produce oranges.   And that I am the better farmer for either crop.   If I concentrate on the product I produce best (oranges), we will both be better off.   In the first model one soon gets into the idea of scarcity - if you Google the word “economics” you soon encounter the term “scarcity.”   As Buchanan argued that night, relying on scarcity instead of beginning with the idea of the benefits from trade drives you in much different directions.  Most of life is better seen in a cooperative light rather than a zero-sum game.   Scarcity, and indeed most of Piketty’s analysis is based on an operating assumption on the left which from my view it is not helpful to understanding complex problems because you ultimately spend more time on allocation issues rather than figuring out how to make everyone better off.

Piketty is a logical successor to Thomas Malthus - whose Treatise had such an influence on much of economic thinking but began with flawed data and linear assumptions.   Malthus, whose work was designed to influence Parliament’s debates on the poor laws projected the most horrible consequences to humanity because he thought food supply would be eclipsed by population growth.   But his careful logic was linear and failed to project for things like the steel plow. All of a sudden a few years after his book came out the plow began to produce almost geometric increases in the food supply because it tilled the soil more effectively.  The problem with most Malthusians is they never seem to recognize that flaws in data and logical models can have consequences.

Right before he goes into his policy proposals, he has a section on sovereign wealth funds and argues that if oil gets to $200 per barrel that the oil exporting countries could soon control a lot of the world’s capital.   And if that does not happen (woe is me!) some of the other funds like China’s will control because of how much stuff they will accumulate.   It is hard not to laugh out loud about these kinds of projections.   I suspect he thinks the projections reinforce his policy proposals.  They don’t.

#3 - Is Piketty Uriah Heap?  Piketty should be given some points for effort.   This is a complex book with a lot of theory.  He tries (I think often successfully) to explain just how complex income deciles and things like GINI coefficients can be.    His book is nothing, if not comprehensive in its view.   But it relies on a series of very shaky assumptions.  The most important is his assumption about a term called “purchasing power parity” (he believes you can make reliable comparisons between times and across borders) but as Diane Coyle points out in her excellent short book on GDP “the large margin of uncertainty around this information should never be forgotten.”   Unfortunately, Piketty concedes the problem and then uses the data to prove his points using the same data.   He fits into the character of Uriah quite neatly.   Even more annoying is his tendency to make historical comparisons which are based on the novels of Balzac, Austin, and Henry James - as if their writing involved a careful analysis of the structure of the economy at the time.  

#4 - Should he dismiss the Kuznets Curve so quickly?   He credits Simon Kuznets for his excellent work on the development of national income data.  He should.  Kuznets was the father of National Income Accounts.   One of Kuznets’ key contributions was also something called the Kuznets Curve which postulates that as the economy goes through technological development cycles economic inequality first increases, then decreases.   A good example of the phenomenon started at the beginning of the industrial age (after the Civil War in the US).   Large fortunes came to those who developed new technologies - the early payoffs were huge.   But as the cycle progressed, income disparities began to be reduced.   Some economists argue that the apparent increased concentration of wealth in many societies during the current period is partially as a result of that same kinds of trends.   Early winners in technology got huge compensation.    

But the current inequality is not just from technology rewards to entrepreneurs.  This period is also influenced by larger compensation to stars (movies, music, sports) - where the value of super-stardom (because it is so easy to replicate a performance) has increased significantly. Caruso in his time got paid a few multiples of what an average opera singer got paid.   Now the difference between a run of the mill tenor and a star is huge.   We also began to pay lawyers and doctors and bankers more than we did in earlier times.  But the question is whether all of these higher wages will be sustained or diffused as the cycle continues. Because it has happened over the last several years (or even decades) does not mean it will continue forever.   The salaries of young law associates and investment bankers have begun to plateau - if that continues those professions will attract fewer of the best and brightest.   Piketty sees the world as an economist who ultimately believes we live in a zero sum world.   There is plenty of evidence that assumption is false.

#5 - Does he offer anything useful or reliable about the long term trends on income/wealth?  The author certainly sees the world in terms of good and evil.  One of the most frustrating discussions in his book is about the growth of what he calls “super managers”  he argues that the US and the UK have led  the growth in inequality as a result of over compensating what he describes as the top of the corporate hierarchy.  But his numbers don’t tell the same story.   He estimates that 5% of the top 1% are actually performers (by that he means stars - my guess is that his number is low) and that 20% are financial types but that the rest are these over compensated super managers.  That picture is simplistic.   We began to compensate lawyers and doctors more generously, beginning several decades ago.   They certainly are not “super-managers” but they are in the 1%.  

Another flaw in his logic is that most of these professionals (highly compensated lawyers and bankers) are in countries which control a large percentage of the global financial deals - that means the US and Britain and not France and most of the rest of Europe.  (The US, UK and Japan are just under 50% of the total capitalization of world financial markets).   In California at least a lot of the very rich have two characteristics.   First, they are relatively young and second they (or their kids) may or may not be in the top brackets in the future.   The payoffs to technological innovators have been huge.   As the Kuznets curve suggests - over time those initial payoffs disburse into the broader society.  Piketty makes some speculations about what might happen as the people who were highly compensated age; but none of his scenarios are convincing unless you believe there is no economic turbulence.  

There is also the discussion of the Modigliani Triangle - which was used to torture Economics students and postulates a life cycle theory of investing; in essence, we accumulate dough when we are young when we can so we can live through retirement.   I am not a fan of how Piketty explains this elegant piece of theory but the book does present some very interesting data on how estates developed and were distributed over time. (again qualified with the complexity of doing inter-temporal comparisons)  If the population is aging and wealth is accumulating - the natural transitions between generations may be disrupted.   I don’t think I agree with Piketty’s conclusions but his data made me think about what might be happening.

One of the odd things later in the book is the author’s lack of understanding of investment returns and the returns of hard work.   He argues that it is impossible to be sure what percentage of large fortunes came from effort, theft or simply good luck (being born with the right parents).   Sure that is true, but so what?   Was the payoff to Bill Gates for his efforts on developing an operating system and other technology innovations justified?   I honestly don’t know.  Piketty wants to make those allocation decisions more “democratic” by which I think he means more political.  From my view, moving those kinds of decisions into the political realm would be a disaster.  In order to make Piketty's theory work, you need to assume there is a lot of luck in anyone's success.  Hard work and having a good idea - need to be discounted.   For anyone else, it is impossible not to attribute luck to some of a person's success.   But from my view, Piketty seems to think it is a lot more important than I do.

My best assessment of his understanding of investment is that he is almost uniquely naive about how investments work.   Without that knowledge, it is hard to take his ideas about what should happen to policies which affect the allocation and return on capital.   That is harsh but I think it is appropriate.

Piketty argues that it is appropriate to force individuals to disclose the amount of their wealth as somehow being more democratic?  I will come back to that issue in the final question,   For now, just why is it a good idea to force public disclosure of income and wealth?  

Many of his arguments - about the long-term structure of capital, about the relative distribution of wealth in a country or across borders are based on two very faulty assumptions.   First, as he admits, his data is uncertain.  But second, they are based on linear assumptions.  Consider an argument made in the late 1970s - if the Japanese economy continues to invest in US assets, by XXXX they will own the entire US economy.   That did not happen because of the dynamic effects of innovation.  Configural logic is a lot harder to work with - but as we found in the Japanese example - it is often better in hindsight.   When the Japanese economy began to implode and the Nikkei, which had been at 36,000, began to fall - we were able to buy back all of those assets they bought for pennies on the dollar.  When writers were expressing opinions about the “Japanese way of managing” Americans were working on technological changes which began to make the organizational structure and technological structure of Japan out of date.   All the smart money was on companies like Sony (Current Capitalization of $27 billion and change) not on ones like Apple (Current capitalization of $565 billion - or a cap value of almost 20X).  He could have benefitted from reading Schumpeter on creative destruction.

#6 Has he bothered to think about the implications of tax theory?   He also seems to not understand the differences between rates and share in tax systems. Even though tax rates were considerably higher in the nineteen fifties, the percentage of GDP paid to taxes has remained about the same and the distribution of burdens has changed toward more progressivity.   There is plenty of evidence from the 1986 Tax Reform Act that, at least in the US, as rates were reduced and the base broadened, the highest income taxpayers paid an increasing share of the total tax bill.   (The Tax Foundation estimates that at this point the top 1% gain about 19% of total income and pay just under 40% of the income taxes in the US.)  So what is a better way to do a tax system - with confiscatory rates and lots of complexity or with low rates and little complexity.   He seems to think that higher rates do not induce complexity - the evidence, at least from the US, is to the contrary.

In one chapter he makes the absurd assertion that a change from 0-30% on taxes on capital the main effect will be reducing inequality.   Take that logic to its extreme and tax all capital at 100% and see how much we improve inequality and, for that matter, capital formation.

A fundamental concept in tax theory is called the substitution effect.  If you tax something, at some point, people will substitute something else for the taxed thing.   Thus, if you tax root beer at 10%, and not Dr. Pepper - one would expect more people to buy Dr. Pepper.  Piketty in all of his argument for a progressive tax on capital never seems to think that the highest income taxpayers would begin substituting capital preferences.   Again, the evidence in the US example is against him.

Complexity, which is mostly an unseen cost in the tax system - decreased through 1986 and then began to increase where we now have moderate rates and labyrinthian complexity.  The evidence from the Economic Recovery Tax Act (1981) and the Tax Reform Act (1986) is that the simplification of of capital structures and the lowering of capital rates offered some huge incentives to the fuel technological boom we experienced over the last couple of decades (which Gordon and Cowan now say is now nearing its end).

#7 - What is the right GINI?  Piketty explains the implications of relying on the GINI coefficient as an indicator of how well we are doing in promoting an equitable society.  I think he is right.    Corrado Gini, an Italian, came up with this formula to express how equal incomes/wealth were in society in 1912.   (Another tool to torture students of Economics)  There is a lot of discussion in the field about what the distribution of wealth and income should be.   A GINI of 0 would have each of the 10 deciles in the coefficient being equal.    Don't we want to reward effort and skill to some degree?  The question is how much is fair.   If you can’t hit homers then you think the distribution of income among baseball players is unfair.   I do not know what the right distribution of wealth or income should be, except in very general terms.   One of the problems with thinking about GINIs is that there is always someone at the bottom.   Piketty does a lot of discussion about using centiles not deciles and that seems like a good idea.   With centiles, you might be able to make more intelligent observations about what is happening in the middle.

#8 - If the percentage of consumption/wealth in the world continues away from the US, is that a bad thing?   He spends some time on the diffusion of consumption in the world. The US is “slipping” from a predominate force in the world.   More people from poorer (or formerly poorer nations) are able to get all sorts of new consumer goods.  The question I always have for someone who thinks this may be bad is “are we worse off with more consumers in the world?”   The answer is no.   A zero-sum analysis is often plain silly.  We should applaud increases in consumption in places which had little.  And we should not be worried if our share of the total diminishes - so long as the pie gets bigger.   He says the pie cannot expand forever and that is true.  But we do not know the limits of pie expansion.

#9 Should we watch out for secondary effects?  Piketty makes a series of proposals to solve to inequalities which he never ultimately defines.  I am pretty sure I like some inequality in incomes and wealth but just not too much.   Piketty seems to live by the same standard - although we would differ on how to achieve equity.   Many of his ideas may a) not solve the problem in the direction he supports (and which most of the American people would oppose) and b) may also create externalities in other areas.  

A clear example, which Piketty would probably not support (based on a gratuitous comment he makes later in the book on public pensions) is what effects a program like social security has on wealth.   A number of countries, including Chile, have privatized their retirement systems.   Workers in Chile are required to pay into a retirement account with a limited number of options but the government role in the system is to set the rules.   What would the effect on capital be if other countries did the same?   In the US we are forced to contribute to a system which does not accumulate wealth although it does have a modest positive effect on incomes of retired persons (which might be even larger if we were to adopt a Chilean system).   So what would be the effect on Capital in the 21st Century if we were to begin to change from the US system to the Chilean one?  (He contrasts the two systems as PAY-GO versus CAPITALIZED pensions.)  

#10 Was the last section written first?  A quarter of the book is dedicated to proposing things to solve the scourges he describes in the first three sections.  

At the outset of the last section, Piketty shows his cards.   He wants more of everything provided by government - he keeps referring to the “social” state.   And he argues that collective action is not possible without taxation.   That is simply not true.   Much of what passes for political debate these days is in that simplistic form.   But Piketty conflates two issues which should always be separated.   First, how much of a good do we want to provide for society?  And second, what is the most efficient way to provide it?   There is a good body of literature on the inefficiencies of governmental provision.  Piketty seems to have missed that.   Government does not always have to be the provider.   One could use his discussion of pensions to prove the point. 

He argues that one reason for keeping the current system of “pay as you go pensions” is that the long-term risks associated with investments cannot be borne by most workers.  Yet he does not seem to concede that the demographics in most developed countries are against sound funding for these programs.  Has he ever bothered to look at the capitalized pension systems that have been adopted around the world which simultaneously increase a country’s stock of capital, assure higher benefits and in the end assure a greater chance that savings in a country will be less concentrated at the highest income levels?  In the first instance, present in many of the “developed” countries of the world - government is the provider; in the alternative or “capitalized” model, government sets the rules for the system to work and mandates how much citizens must save out of their wages.   The pay-go system in almost all instances is lurching toward insolvency.  

He makes some absurd claims including that the American income tax system has become more regressive.   The Tax Foundation numbers show something completely different.  The top 1% earn about 18-19% of the income in this country and pay almost 40% of the tax.   Not sure how anyone could describe that as “regressive.”

His tax discussion, ultimately pushing for a worldwide progressive tax on capital, which even he admits is utopian is a mishmash of jargon and speculation.   For example, he argues that the growth in executive salaries in the US began to grow with the Reagan tax cuts of 1981 and 1986.  He never seems to make a distinction between contingent and salary payments.   A good part of the compensation of what he calls super managers is contingent.  The fault in at least some of those systems may be the point of analysis not the incentives.  But he never gets to those questions.

In this final section Piketty would clearly favor two things - an international or at least multi-national progressive tax on capital and a higher percentage of national income being contributed to what he refers to as “social” (read governmental) spending.   Oddly he goes into some detail to explain how complex it has been for the European Union to agree on some very simple common elements in the fiscal constitution.  Yet, he breezes by those large potholes to suggest how we should move to a system which even the EU cannot seem to agree on.   Did he bother to look at the discussions in several EU countries about whether the country should work toward additional integration?   Getting even the major economic powers to agree on a tax system that would collect and distribute revenue among countries would be very tough.

He makes the claim that the confiscatory rates in the US tax system improved transparency in the US.  But the history does not support his claim.   As the 1920s frenzy in stocks began to grow, several leading academics and professionals on Wall Street began to argue for more clarity in business accounting standards.  Benjamin Graham started his book Security Analysis in the late 20s as a result of his teaching at Columbia.   He published the first edition of his massive and still useful work in 1934, well before the implementation of although the first high rates were introduced under Hoover in 1931. (63%)  The rates above 63% (which is pretty confiscatory) were added later in FDRs term.

Part of his discussion focuses on debt reduction and in exploring in a rudimentary way what the ideal level of debt for a country should be.   What he does not seem to care about is why every mature democracy has significant problems with national debt.   As the literature in Public Choice theory suggests log-rolling on spending is a huge problem.   Everybody gets something like a “bridge to nowhere” and suddenly you have $20 trillion in debt.

Some final comments - In the last section Professor Piketty makes two pitches - one which I agree with and one which I do not.   The professor clearly thinks that we underinvest in the public sector, even though in the countries on which he focuses the percentage of GDP going to government is between 35 and 55%.  He even makes the odd statement that  “Private wealth rests on public poverty.”  Is the question underinvestment or is it faulty allocation?  

At the very end of the discussion he makes an impassioned argument that Economics is not science and he prefers (as I do) to use the term political economy.   He suggests that part of what economists must do is lessen their reliance on numbers and formulas (although he thinks better data helps us understand what is going on in a particular area) and do a bit more thinking about what the proper mix of government and the private sector should be.   In my opinion his argument here is right on target.  Where we surely differ is what the appropriate division should be.

No comments:

Post a Comment