Why I am not persuaded by Thomas Piketty’s argument

My Foreign Affairs review is here.  (Open up “New private window” in Firefox, if need be.)  I won’t attempt to cover all of the review, but rather will rephrase a few of my points for MR readers, in slightly different terminology:

1. If the rate of return remains higher than the growth rate of the economy, wages are likely to rise and quite a bit.  You can find a wonky version of that idea here from Matt Rognlie.   But it suffices to apply common sense, namely that capital accumulation bids up wages.  Piketty suggests we are headed back to something resembling the 19th century.  Well, that was a pretty good time for the average working person in Western Europe, especially once we get past the first part of that century, which had lots of war and a still-incomplete industrial revolution.

Since we today have had some wage stagnation, perhaps it does not feel that kind of favorable outcome is what we will get and many commentators are trading off this mood.  But also realize the (risk-adjusted) return on capital hasn’t been that high lately and it has been falling for decades.  This combination of variables — low returns and stagnant wages — does not refute Piketty but it doesn’t exactly fit into his mold either.

2. The crude seven-word version of Piketty’s argument is  “rates of return on capital won’t diminish.”  Is that really such a powerful forecast?  I say over the next fifty or one hundred years we don’t have a very good sense of which factors will show diminishing returns and which will not.  It is hard enough to make predictions of trend over a twenty-year time horizon.  NB: At many points in the Piketty book he seeks to have it both ways: loads of caveats, but then he falls back into the basic model, and he and his defenders cite the caveats when it is convenient.

3. Piketty’s reasons why rates of return on capital won’t diminish are fairly specific and restricted to only a small share of capital.  He cites advanced financial management techniques of the very wealthy and also investing abroad in emerging economies.  Neither of these covers most capital, and thus capital returns as a whole may not be so robust.  Nor is it obvious that either technique will prove especially successful over the next few decades or longer.  Again, is there any particular reason to think either of these factors will outrace the basic logic of diminishing returns, or for that matter EMH, relative to other factor returns that is?  They might, to be sure.  They also might underperform.  In any case this is pure speculation and Piketty’s entire argument depends upon it.

4. The actual increases in income inequality we observe are mostly about labor income, not capital income.  They don’t fit easily into Piketty’s story and arguably they don’t fit into the story at all.

5. Piketty converts the entrepreneur into the rentier.  To the extent capital reaps high returns, it is by assuming risk (over the broad sweep of history real rates on T-Bills are hardly impressive).  Yet the concept of risk hardly plays a role in the major arguments of this book.  Once you introduce risk, the long-run fate of capital returns again becomes far from certain.  In fact the entire book ought to be about risk but instead we get the rentier.

Overall, the main argument is based on two (false) claims.  First, that capital returns will be high and non-diminishing, relative to other factors, and sufficiently certain to support the r > g story as a dominant account of economic history looking forward.  Second, that this can happen without significant increases in real wages.

Addendum: Still, it is a very important book and you should read and study it!  But I’m not convinced by the main arguments, and the positive reviews I  have read worsen rather than alleviate my anxieties.  I’ll cover the policy and politics of this book in a separate post.  Do read my review itself, which has much more than what is in this blog post.


The question I have not heard is how the long term baby crash is going to effect wages. The lowering of labor force participation and lower birth rates, families are moving more towards the Asian 'tiger' model where the birth rate is lower and the kids are trained to become successful college graduates. (Isn't this happening in a lot in Euorpe and Northeastern US as well as East Asia.)

Will there a long term labor shortage?

American billionaires like Mark Zuckerberg and Bill Gates organize politically to demand more immigration to keep down the salaries they have to pay their workers. It's a spectacularly obvious version of what Piketty is talking about, but of course he doesn't dare bring it up.

Armed with a hammer everything looks like a nail.

Especially a racist hammer

Seriously, what kind of racist nut doesn't trust the motives of billionaire oligopolists who have already been caught red handed engaging in a cartel to limit competition in the labor market. Racist nut.

And how would Bill Gates organize politically anyway? What Foundation is there for this claim? I mean, its preposterous, he has only one vote, just like me and David Rockefeller.

No Steve Sailer has a point. This is really economics of immigration 101, and it fits neatly with Pikettys Theory

It's funny how this book gets everything so wrong, but has stimulated so much good thinking! It's because he's addressed a crucial, neglected, topic, and taken a strong and specific stance.

On the 3rd world: it seems to me there is huge potential there, even just thinking about CHina and India, because of government reforms. I can imagine very high rates of return on capital for a long time--- probably increasing for US investors, in fact, as we catch on to those investment opportunities. This will be bad for US workers who supply the same kind of labor as is common in India and China, though---- the unskilled, and super-smart programmers.

"neglected, topic"

I must have missed something. Which "topic" is that?

"It’s funny how this book gets everything so wrong, but has stimulated so much good thinking!"

Right, the rich getting richer is a very big topic, although r > g sounds a tad over confident. Of course, we can tell if the thinking has gotten very good if we start hearing questions asked such as: "Hey, how come all the rich people, like Carlos Slim and Mark Zuckerberg, want more immigration? Isn't immigration one of those issues on which the rich conspire to impose their political will and self-interest on the public?"

Surprising no one, Steve didn't read the book but is sure it supports his lifelong thesis that brown people are ruining the United States.

I want Tyler to post about the NBA playoffs so Sailer can comment about how darkies like Serge Ibaka and Tony Parker are taking jobs from Americans. I wonder if Dirk Nowitzki or Steve Nash will get a mention?

The economic activity that brings a country like China up to western standards of living will generate enormous wealth for those providing the services and goods. I fail to see how that is a problem that needs fixing.

Maybe Piketty is complaining that his country, France, or those in the US who are echo his complaints are about being on the losing end of comparative advantage.

Yes indeed. If a country has a high fixed cost for economic activity, it will happen somewhere else. And any in that country who benefit from the growth elsewhere will be those with capital, since the high fixed costs price labor out of the market.

Brad DeLong predicts that the real rate of return on capital will be 4.5% even after K/Y ratio more than doubles. Not sure why he thinks that. Looks to me like returns are *already* lower than that. Stocks are probably set to return at most 5% long term going forward, and more global wealth is in bonds than in stocks. What does Brad know that I'm not seeing?


5% before tax = 2.5% after tax.

That was a good article. The last sentence is particularly well written.

Here is libertarian Arnold Kling making the argument for a wealth tax (rather than debt default). He sees no particular problem. Says the Navy will take care of it.


I think that the U.S. government would enact a wealth tax rather than default on its debt.

Other countries that have defaulted have not had the option of enacting wealth taxes. When you are in a banana republic with shaky government finances and you have a lot of wealth, you send that wealth over to the United States, where your government cannot get to it. That "safe haven" motive is what keeps the dollar so strong. Anyway, by the time the banana republic gets around to enacting a wealth tax, all the wealth has fled the country and there is nothing left to tax. So the banana republic defaults.

As the U.S. government's finances deteriorate, it will strengthen its hold on its citizens' wealth. My guess is that you will see tighter laws that restrict your ability to hide wealth overseas and much more enforcement of those laws.

Basically, if a banana republic says to us, "Help us keep the wealth of our citizens," we can say no. On the other hand, if we tell another country, "Help us keep the wealth of our citizens," that country will co-operate. This asymmetry reflects the distribution of military power.

So what I am saying is that the ultimate guarantor against a U.S. government default is the U.S. Navy. Because of the navy, the U.S. government can control the policies of other governments. Because it can control the policies of other governments, the U.S. government is in a position to dictate whose wealth can flee where. Because the U.S. government can stop our wealth from fleeing, the U.S. could enact a wealth tax. Because it could enact a wealth tax, the U.S. is unlikely to default on its debt.

The hypothesis so well stimulates the interest that the model being faulty can be set aside? Sort of like Freud?

It's true that right now, the rate of return on capital is low. But growth and wages are even lower. How does this not conform to Piketty's basic claim that r > g?

r > g uses ex-post r, which omits risk (and the destruction of some people's fortunes therein).

Average return on capital it is greater than the growth rate is because there is a risk premium. If people were not risk adverse in the short term stocks would have the same average returns as bonds for long holding times, but they do not. see http://visualizingeconomics.com/blog/2013/5/21/rolling-real-returns-stocks-bonds-and-bills-since-1928 To the extent that the wealthy can preserve capital for the next generation they have longer time frame than people who save to meet goals like paying for college of retirement means they will earn higher returns

It is difficult to get a man to understand something, when his salary depends upon his not understanding it.

Amen and LOL.

If our host thinks that r >> g will cause wages to rise, exactly where is the extra production to pay the wages going to come from?

Where is the demand going to come from when wages are stagnant or falling and capital is depreciating but not being replaced?

Of course both are occurring because "capital" is seeking higher rates of return so that calls for cuts in labor income and cuts in total capital so the declining quantity of capital will inflate in price thanks to demand for even scarcer capital.

Inflation has been very high over the past five years, but all focused on securities, not goods, not services, not real tangible assets, not wages.

The difference of reception between France and the US is really striking. In general, french journalists liked the book but not to the point of calling it a masterpiece or even the most important book of the year. And I haven't read a single review by a french economist which wasn't skeptical of his main claim (these reviews are really good, too bad that none has been translated into english and that there is apparently not a single american economist able to read french anymore.)

But then, the book was probably conceived specifically for an american readership, otherwise why was it translated barely two or three months after the book was out in french?

If you want some serious debunking of the empirical claim that the capital-output ratio has been increasing over time, try this one : http://spire.sciencespo.fr/hdl:/2441/6d6bmqq2mq9avo75ba1s430vom/resources/wp-25-bonnet-et-al-liepp.pdf
bottom line : if you want to include housing in your capital measurement, you should measure it on the basis of the present discounted value of rents, not on market value, which is uncorellated with the real revenues of the real estate capital and has been overvalued for almost twenty years. When you do that the ratio becomes basically trendless.

Central banker Thilo Sarrazin's book "Germany Abolishes Itself" sold over a million copies in Germany, but I've never heard of a translation into English.

I am curious about any positive reviews by actual trained economists. I have yet to see one that doesn't point out serious flaws.

"The difference of reception between France and the US is really striking."

I didn't realize that there was such a lukewarm reception in France. Perhaps, the Left in France (and the rest of Europe?) is not as desperate as the Left in the US to find some sort of coherent, intellectual framework to justify perpetually growing government? In France, there is already widespread acceptance of social democracy relative to the US, where the emphasis is on individual liberty (egalite vs. right to pursuit of happiness). Since American values tend to place liberty above social welfare, egalitarian concerns alone typically don't justify more government. Hence, the American Left is in constant search for other rationales. The Left seeks government health care not just as an entitlement. Since everything affects one's health in some way, government health care turns every personal decision, right down to how much sugary soda one consumes, into a government interest subject to regulation. Global warming, now climate change, converts every aspect of human life, personal or otherwise, into an externality. Everything affects the environment, right down to one's choice of light bulbs. However, Americans have been resistant to government health care, and global warming's political potency has been weakened by the on-going, extended pause in global temperature. Hence, the Left's latest fetish seems to be income inequality: inequality itself weakens growth, inequality begets more inequality, inequality is self-perpetuating, etc. Piketty's timely tome would seem to satisfy a hunger among the American Left that perhaps doesn't exist in France.

What's your basis for 4? Labor income inequality has neither changed much in the US, nor is it particularly high. Inequality appears to be almost entirely about capital income. I think you might be committing the old fallacy of ignoring the inequality within the top 1%, who happen to own most of the capital--almost all of the increase in inequality has been within that 1%.

"He cites advanced financial management techniques of the very wealthy and also investing abroad in emerging economies. Neither of these covers most capital, and thus capital returns as a whole may not be so robust"

What is the evidence for this? The aggregation of capital into investment funds means fund managers can take advantage of same of the same loopholes as the super wealthy (not least the carried interest exception, though that doesn't directly affect the fund's capital).

And if you're investing in multinational firms aren't you also investing in emerging economies? Coke sells a lot of sugar water in China.

Today most westerners can get returns on capital. Adding insulation to your home can yield some great tax free returns and where would this leave investors in electric companies? Solar might work too. Self driving cars could create a capital return that could put a big hit on investors in insurance companies. Biotech could make it easy to grow fruits in the back yard.

I was going to add that as overhead costs rise relative to direct costs, the ownership of capital that produces for in family consumption becomes a better deal. This also true of debt. As the overhead cost of making and enforcing loan repayment goes up it gets better to keep the lending within the family.

Thank you for citing the law of diminishing returns. It is one of those oft forgotten but main learnings from economics

Looks like Nassim nailed him Piketty got the math wrong

1. Absolute wages do not matter. Our wages may rise but if we are continuously outbid for homes in good neighborhoods, it does not matter. (I present a mostly developed-world/Robert Frank argument; my argument may not hold in the developing world, but just give them time to catch up.)

2.1. It will be difficult to know because policy will likely (hopefully) intervene and then we will have many arguments about why inequity was reduced. To figure all this out, we will need the really, really, long-long-run. In 300 years--after all the policy interventions become footnotes--we still end up with more inequity.

2.2. I do not interpret Piketty to be forecasting as much as pointing to history and saying that "since this has happened in all of the recorded history of capitalism, there is no reason to believe that it will change unless the very nature of capitalism changes."

3. Pikeety's reasons are based on the point of history in which we find ourselves. If he were writing in the 18th century, he would have pointed to slave trading and owning sugar cane or tea plantations. Capitalists will find other methods to rent the system 100 years from now.

4. Hmmm .... I thought we have been reading stuff lately that says that labor income can change dramatically from one year to the next and that the one-percent of labor-income earners is not a stable group of people. Maybe I am wrong.

5. Maybe Piketty does this because there is a history of capitalists buying out the system.

Absolute wages do not matter. Our wages may rise but if we are continuously outbid for homes in good neighborhoods, it does not matter.

Nonsense. First off, the vast majority of things we buy, such as food, are not positional goods and thus absolute wages are hugely relevant. Second, it seems the solution to a lack of supply of homes in good neighborhoods is to either build more homes or more good neighborhoods (such as through the construction of new subdivisions or the gentrification of less desirable neighborhoods). Robert Frank usually uses this argument in conjunction with the desirability of good schools, but the obvious solution here is to attack the linkages between schools and geography such as through vouchers.

Wow, the majority of what we buy isnt positional?! Could have fooled me. My house, my car, my kids' school tuition, my wife's clothes, my new carbon fiber racing bike, my vacation on the beach... all more or less positional.

Just because something can be a positional good for some people doesn't mean that they are for all people. And cars and houses aren't even positional goods in the strict sense because they are non-exclusive (well, most cars and houses, anyway). BMW isn't running out of cars, and you can generally find a builder to build you a house (maybe an uptown apartment or something is more exclusive).

Maybe, the fact that you view your house, car, clothes, bicycle (unless you really care about winning races) and vacation as positional goods is why you think absolute wages don't matter.

Gregory Clark's "A Farewell to Alms" points out how important positional goods have always been in at least Northwestern Europe to getting married. Unlike traditional China, which had near-universal early marriage for women, and therefore had massive famines when the population quickly reached levels that couldn't be maintained during times of disorder, Europe didn't have many famines because Europeans were acting on Rev. Malthus's advice to restrain themselves avant la lettre. Couples couldn't get married until they had achieved the positional goods and status markers appropriate for their class.

Is contemporary America all that different?

That sounds nothing at all like the China I know. While all Chinese women got married, many Chinese men did not, and still do not. This is why every army in Chinese history promised its soldiers wives. Positional goods are incredibly important to marriage in China and have been for as limg as we can document. And Mathusian crisis was not the driver of Chinese famine for the last thousand years at least. This is bad even for you Steve.

I am not so sure about the famines thing. Europe had large famines, approximately every 7 years, until it introduced potatoes as a staple food. Since then, with the exception of the Irish potato famine, the crops were much more reliable.

The Central European chronicles up to the 18th century are FULL of famine records.

Since you mention Malthus you are presumable talking of Europe after the mid 1800's? If so, how can a comparison to China ignore the crucial points of education, literacy and a general enlightenment? Was society in France or Germany in 1890 at all comparable to China?

I'd posit that had a bigger effect on marriage than all the other factors you cite.

Not 1890, 1790.

"Second, it seems the solution to a lack of supply of homes in good neighborhoods is to either build more homes or more good neighborhoods."

I think that's what has happened outside of the densely populated East and West coasts. In the heartland areas, the middlest of middle income people own huge McMansions. "Bidding up" of home prices seems to be confined mostly to the East and West coasts. I think Megan McCardle might have written a blog post a while ago speculating that this might explain the divide between Red States and Blue States. East and West coast folks may be more sensitive to inequality because the bidding up of home prices in these areas affects middle class cost of living.

I was pleasantly surprised by this review after posting that weak data on mobility yesterday.

After reading hundreds of pages of brilliant facts and accounting (unndercut by iffy modeling), it was so dispriting to find the big policy solution on offer to be a hopelessly naive global wealth tax. After the disaster that the climate treaty movement proved to be for the environmentalists, it is amazing how freely such a ridiculous, and impossibly unmessy supranational solution could be latched onto by a major thinker, much less one who has spent most of his adult life in the Eurozone(!).

"wages are likely to rise and quite a bit"

But as you say, "the actual increases in income inequality we observe are mostly about labor income". Wages could rise quite a bit as median wages fall quite a bit. We could have robots and highly paid roboticists while everyone else makes next to nothing.

The scenario you describe is not about the concentration of wealth, it is about adjustments due to technological advances. As well as the growth and median wage improvements occurring elsewhere.

Am I wrong in thinking that the substitution of capital for labor isn't about technological advances?

Is it hard to believe that a concentration of wealth in the future won't exist among an elite which are both the capitalists and the roboticists? Won't the capitalists and roboticists basically be the same people? It's would be more meritocratic than Picketty's predictions, but the results for society would be about the same: massive, 1st century-style inequality.

I just have a general (somewhat academicky) question that has been baffling me and a couple other friends about the book (slightly off the thrust of this thread). In the r-g formulation, why is it GDP growth and not per capita growth? We've spent a couple hours debating this and don't see any reason why GDP growth should be the relevant metric. What are we missing?

"... the success of certain immigrant groups suggests that cultural factors play a more significant role in mobility than does the capital-to-income ratio..." So you disagree with Gregory Clark's biological model then?

Forgive such a naive question, but putting aside his projections that the future will look different than the past, is Tyler arguing that r is, on average actually, not consistently greater than growth? I guess I don't understand exactly how incorporating risk makes r smaller. I totally agree that there are variations in the rate of return across capital and that "sudden reversals and retrenchments are inevitable." But isn't one of the benefits of being wealthy having enough money to be strategic about investments, allowing people to incorporate a lot of hedging in their capital investments? Sure there is risk, but if wealthy investors are smart about it, can't they avoid total bust scenarios and keep their returns larger than g, on average?

Your FA review is the best, and most thoughtful, of those that I have read.

If the rate of return remains higher than the growth rate of the economy, wages are likely to rise and quite a bit.

I'm a bit confused by the seeming implication that we could have high rates of return on capital and high wage growth without high economic growth (assuming that all are in "real" terms).

If I understand it, Matt Ronglie explains that Picketty conflates capital income without depreciation with capital income net of depreciation. The return to capital and the return to wages can both be higher than growth before you subtract for depreciation. (I think that's what Matt and Tyler are driving at here.)

By "both" I meant "when combined".

This isn't quite right. What Piketty conflates (according to Rognlie) is the "elasticity of substitution" in a gross vs net production function.

In other words, as we increase the amount of capital, how much does the return on capital decline due to diminishing returns? The decline in gross return will always be smaller (in percentage terms) than the decline to net-of-depreciation returns. Expressed in math, if r declines by 10 percent, then r-delta will decline by more than 10 percent as long as delta is positive. So if you want to use estimates from the literature you need to make sure you pay attention to which one you are using.

Brilliant review. This is going to be my "go to" review when I discuss the book with people.

"At many points in the Piketty book he seeks to have it both ways: loads of caveats, but then he falls back into the basic model, and he and his defenders cite the caveats when it is convenient." How else is it supposed to work? The word "convenient" makes this sound like an attack, but otherwise this sounds like the model of how to make predictions with humility.

The claim here is that the caveats are only being applied in defense of the argument, but were *not* applied in the positive claims and policy recommendations the argument applies (the opposite of humility). Thus, the policy recommendations are weakened by the caveats.

As I wrote on my blog a few days ago, the rate of growth of capital isn’t homogeneous. There is no one 'r' at any specific period of time. A booming business like Walmart in the 70′s or Microsoft in the 80′s will have a high growth rate of capital but eventually it slows down to the overall growth of the economy, by virtue of the law of large numbers and diminishing returns to scale. Some businesses become stagnant, like home builders in the late 2000′s. So while the Walton family will get richer, they cannot become richer at a rate much faster than the economy unless Walmart expands its business by opening more stores or branching out into different areas of commerce, or Walmart gets more business attributable to a change in consumer preferences.

A lot depends upon the degree of antitrust enforcement. During the middle of the 20th Century, there was a lot of government pressure on businesses not to merge and not to follow winner-take-all strategies. This retarded the development of great fortunes during that era and kept wealth spread out.

It's striking out out-of-fashion antitrust is these days. It barely comes up in all these discussions of Piketty.

All of these points sound very weak to me.

Try adjusting your priors?

My favorite sentence from the review: "what might appear to be static blocks of wealth have done a great deal to boost dynamic productivity."

One might dislike the existence of "undeserving" rich with inherited wealth, but they are often the only ones with enough free time and resources to fund exotic and creative activities that push the boundaries of a society outward. Small comfort for those slaving away in poverty though.


In your article in the last paragraph you wrote:

A more sensible and practicable policy agenda for reducing inequality would include calls for establishing more sovereign wealth funds

Could you please walk us through how that mechanism would "reduce inequality"? Please also include where the funds for that SWF would could from.


P.S. Excellent article by the way!

would could from should read "would come from".

A well-written review and much to think about. Out of curiosity what parts of Piketty's book did you find persuasive ... even in a 'right for the wrong reason' way? You laid out some non-trivial policy ideas at the end of the review (NB: widespread de-regulation is probably more fanciful right now than a global wealth tax), so surely something (even probabilistically) worries you too about inequality. Put differently, if all the reviews had slammed Piketty, what would you as a contrarian have written in his defense?

The agenda of the left in the US today is to undo Reganism. They want to return to 80 percent marginal tax rates and, beyond that, even add a wealth tax. Piketty's book provides intellectual cover for this program hence lavish praise. The actual intellectual merit of the book is irrelevant.

Modern physical capital can network more effectively (100+x) than human capital. This is the big point.

One thing I haven't seen addressed is that taxes and expenses place a huge drag on net investor returns. IMO, it's unlikely that r will be greater than g after inflation, taxes and expenses are subtracted.

Investors pay taxes and expenses on their nominal r not their real r. For example, if nominal r is 8%, taxes on that 8% take at least 120 bp away. Next, investment expenses take another 50-100bp at least for trading costs, fund expense ratios, hedge fund manager costs, etc.

So that original 8% nominal r becomes a 3% real net r once inflation, taxes and investment costs are factored in.

What am I missing?


You really should switch to Chrome.

r > g is indeed both elegant and alluring . . . PIketty obviously knows a lot about capital, but if he hasn't taken out a copyright on "r > g" in preparation for launching a product line of t-shirts, posters and bumper stickers, then he doesn't know the first thing about capitalism.

Like coach Pat Riley copyrighting "threepeat."

When inequality is too low, it's a sign of serious economic problems: http://www.devilsdictionaries.com/1/post/2014/03/inequality-is-good.html

He cites advanced financial management techniques of the very wealthy

Heh, that argument cuts the opposite way he thinks -- HFs don't actually generate higher returns, just larger fees. The only effective means of beating the market is rentseeking, which is why only politicians and their friends beat the market in the long run. Of course, it's totally possible to create a rentseeking HF and do well with it, but that tells us more about the government than about the wealthy.

I did wonder if he's read When Genius Failed, about LTCM's failure.

It's easier today to diversify your wealth so that you aren't wiped out in a crash, as often happened to the Gilded Age rich.

Moreover, lots of rich people in the past managed to spend their fortunes, on Downton Abbey-size houses and the like. It's hard to think of too many rich people these days who have consumed their way out of the rich -- actor Nicholas Cage maybe.

The Bronfman family.

This is why I am very skeptical of Tyler's risk argument about risk.

Just on this point:

“To the extent capital reaps high returns, it is by assuming risk...Yet the concept of risk hardly plays a role in the major arguments of this book."

I'm not sure that makes sense. Risk is the risk to an individual of losing their money. But if Piketty is talking about the wealth accumulated by a class - and particularly if that class is defined by how much wealth they have - then there is almost zero risk. Entrepreneur A "takes a risk" and builds a factory; she loses her money; factory is snapped up by rich person B. The risk was realized, but the wealth stayed within the 1%.

You say that the crude seven-word version of Piketty’s argument is “rates of return on capital won’t diminish.”

Absolutely not. Piketty's argument is that "rates of return on capital don't *have to* diminish." This is very different. If he thought that rates of return on capital couldn't approach growth--as he points out has happened in the past--he wouldn't make the policy proposals he does!

"Overall, the main argument is based on two (false) claims. First, that capital returns will be high and non-diminishing, relative to other factors, and sufficiently certain to support the r > g story as a dominant account of economic history looking forward. Second, that this can happen without significant increases in real wages."

This makes me want to bang my head against a wall. Again, the claim is simply that NOTHING GUARANTEES THAT CAPITAL RETURNS WILL CONVERGE TO GROWTH. Yes, the claim you impute to Piketty is hard to defend--I agree!--but it's not the claim he makes. The upshot is that policy has a role to play. It's a pretty modest claim, actually, amply supported by the work.

As for r remaining higher than g without significant increases in wages, Piketty is not saying that this is theoretically determined, he's saying that this *has happened* and we need ask ourselves if our theories are adequate to the economic reality.

You seem to be in the position of a classical economist telling Keynes that he's wrong about the downward stickiness of wages because that observation doesn't fit the existing model.

James K. Galbraith has a new critical review of Piketty's book: http://www.alternet.org/economy/james-k-galbraith-takes-thomas-pikettys-capital-twenty-first-century?paging=off

"If the rate of return remains higher than the growth rate of the economy, wages are likely to rise and quite a bit. You can find a wonky version of that idea here from Matt Rognlie. But it suffices to apply common sense, namely that capital accumulation bids up wages."

Sure, the employed may have higher wages but there will be less people employed because of higher capital to labor ratio. Meaning that total earnings from wages will be less.

I don't share the common sense that capital accumulation bids up wages. When I think of capital accumulators, I think of Scrooge McDuck types who got rich keeping wages as low as possible. Isn't keeping wages down (by conspiracy if necessary, like some of history's most successful accumulators did in Silicon Valley) one of the best ways to accumulate capital?

"Piketty suggests we are headed back to something resembling the 19th century. Well, that was a pretty good time for the average working person in Western Europe..." Are you serious? Did you write this article on 4/20? Being a worker back then sucked! Horrible hours, low pay, horrible conditions, child labour, unequal pay depending on sex/race/age, etc. etc.

And none of that has anything to do with the level of economic development being far lower in the 19th century? Do you really expect that a return to 19th century economic policy would result in workers standard of living reverting back to those levels despite an order of magnitude higher GDP per capita?

I'm just glad we have brilliant economists like Tyler Cowen to tell policymakers why it's more profitable to hire a dozen ditchdiggers at minimum wage instead of $30K for a backhoe and $17/hr for a single operator. Why, next you'll tell me why the task of milling grain may be accomplished by means of hydraulic power with but a single mill operator to replace two oxen and a stout peasant!

If only economists like Tyler had been around in the 1950's to persuade the US cotton industry why harvesters were a losing proposition. Instead of a few higher-g workers operating wasteful, capital-intensive equipment, we would have teams of efficient darkies! I mean, come on, who wouldn't rather have this than this?!

This misses what I consider to be a large problem, not just in Piketty's book, but in many of the people that are making themselves upset about statistics: the capital/income ratios are computed only within the boundaries of a country. If you are a country like the United States, where some quantity of the labor that drives return on capital is in another country, you will get distorted statistics. Capital growth here is driving wage growth in other countries.

This over-simplification of economics to national borders is rather facile, especially when one wants to make statements about the global economy.

The other major problem is that so many seem to have blindly accepted the premise that inequality is bad, when we should be more concerned about rising quality of life measures. Inequality of results is a good thing. I could be making money right now, but, instead, I am writing a blog comment for free. Does that give me a right to a portion of someone else's wealth?

One thing I wonder about is whether the definition of "return on capital" being used is too narrow. For example, many jobs require specialized skills--for example, a doctor. Those specialized skills are capital, at least in the usual economic definition as I understand it, and a large part of the doctor's income is return on that capital, not labor income. But I don't see that kind of capital appearing in either the online summaries of Piketty's work, or in any of the reviews I've read of his book.

Under the proper expanded definition of "return on capital", it seems to me that things are worse than Piketty predicts in one way, but better in another. They're worse because much of what Piketty appears to be counting as labor income should actually be counted as return on capital, making the inequality even larger than he says it is. But they're better because a much larger fraction of the population can potentially earn return on capital under the broader definition; i.e., what we have is not a contest between capital and labor, but a general rule that it's better to have more skills.

Have you read the book? You don't sound like you have taken in the main points. This is not an exercise in forecasting, this is an exercise in reviewing history: he's either right or wrong, to prove him wrong, bring your own data and/or point out his spreadsheet errors. The accounting framework is carefully explained, discussed, alternatives are presented.

It is coarse to dismiss Piketty with so little to show for.

For what it's worth, my opinion is that this is the most careful study of economic history in the history of humanity in the known universe. I also think it's the most important books since Keynes.

The good news is that your casual dismissal will do nothing for truth and knowledge to be disseminated, eventually, to the world's elite (though naturally Fox viewers and the Republican party and sadly perhaps you will remain locked in their alternative universe).

"this is the most careful study of economic history in the history of humanity in the known universe...though naturally Fox viewers and the Republican party... will remain locked in their alternative universe"

Please don't feed the trolls.

Respectfully, your comments seem equally nebulous.

I am being unfair to make this comment after reading only paragraph 1. But maybe not, as it is fair to assume that what Mr. Cowen writes sets the stage for the mind set throughout the article and review of TP's book.

He says, and I shorten the paragraph: 1. "...But it suffices to apply common sense, namely that capital accumulation bids up wages. ..." First, it is an assumption, based possibly on some analyses, of past practice, that capital accumulation bids up wages. This is only a partial truth with regards to 'equality' as the one who accumulates the capital will not put into the system (wages, etc.) anymore than what he thinks might be a "prudent" amount, that will preserve the increased capital. This is the fundamental thinking that generates the continued inequality. The oligarch takes reaps a profit of say, $100 from the system, will not put ever put it all back into the system; he'll keep and accumulate as much as he can, acquiring as much advantage in both sides of a transaction, buying or selling. The oligarch with greater control always has the negotiating advantage in buying and selling.

The existing accumulation which is growing daily, and exponentially is what has landed us where we are...1% always ends up holding more (greater percentage) of whatever is added to the system.

That's just one simple aspect.

I get very tired of pseudo-intellectuals like Tyler Cowen and the writer with the Alexei Sadeski pseudonym who are all talk and no equation. They talk about laws and corollaries and axioms, but in truth produce crap like the vague, ambiguous and silly ‘Ten Pillars of Economic Wisdom’. They and their ilk remind me of Dostoevsky's Ganya - vain, ambitious and desperate to be original although the epitome of mediocrity.

I'm sorry C Waid, but that seems to be a very vague, ambigious criticism. Although I do like that you managed to fit the word "ilk" in. Any chance you could be a bit more specific and equatious about why you don't like what Tyler and Alexei have to say?

I can only assume that TC is embarrassed to be mentioned in the same breath as a hack like me.

The problem with much of "empirical" economics is that the empirics are an attempt to 'dazzle with nonsense.' The other problem is that these empirics are essentially statist: that which statistics can show will provide some material benefit to society should thus be done. Yet the full consequences of such policies are left unexplored by the author.

This is why we have principles.

Am I incorrect in taking the crude, ten-word version of Tyler's counter argument to Piketty to be "eventually technological innovation will lead to more egalitarian wealth redistribution"? That doesn't seem to be a very powerful forecast either. The movement from a hunter-gatherer society to an agrarian one invovles a huge technological leap, and vastly increases the total supply of wealth, but it can also create massive societal inequalities based on the structures of property ownership. Why should we take it as given that the movement from present day technology to the technology we're going to have in a century will work in the opposite direction? I admire (and to some degree share) Tyler's technological optimism, but I hardly see it as refuting Piketty's overarching argument that we are heading in the direction of a new and relatively stable equilibrium in which those who own and control capital will have a much bigger slice of the economic pie than the rest.

'eventually technological innovation will lead to more egalitarian wealth redistribution'

Prof. Cowen has spent a significant amount of time from his undoubtedly busy schedule to let as many people as possible know about the recently published work "Average Is Over: Powering America Beyond the Age of the Great Stagnation,' which disagrees with that summation (apart from the 'eventually,' as Prof. Cowen is as much a fan of caveats as anyone other economist). In part, the work purportedly explains that a less egalitarian wealth redistribution is what our more immediate (if not the eventual) future portends, in large part due to technological innovation.

I think Cowen's review misses the mark. Reading through the FA article, I noticed that Tyler's comments effectively boil down to the following: "We have reason to believe that economic factors will drive down r - whatever r is". But isn't that the point of Piketty's book - to show that economic factors aren't pulling their weight? That other forces - namely, a sort of economic fatalism - drive economic behaviour?

In essence, Tyler takes the optimistic route. But why should I believe Tyler's arguments when Piketty's data shows that, from a time-series perspective, Tyler is wrong. There are, after all, mechanisms that allow to very wealthy to accumulate the new economic opportunities first. I don't see why r will fall below g based on Tyler's arguments at all.

"iketty suggests we are headed back to something resembling the 19th century. Well, that was a pretty good time for the average working person in Western Europe, especially once we get past the first part of that century, which had lots of war and a still-incomplete industrial revolution."

This incorrect. The average rate of poverty (below a poverty line of "just barely getting by") in late 19th Century Britain was about 30%.

See http://eprints.lse.ac.uk/3913/1/One_hundred_years_of_poverty.pdf at 22.

Exactly. Only 30%.

I would like to see more work done on the following issues (most raised by, among others, Arnold Kling in From Poverty to Prosperity):
1. Emerging "property" is intellectual property (apps, software, inventory control methods or even conceivably a trading strategy) which is largely a function of knowledge and legal constructs and is non-rivalrous -- it is not something that -- like rentiers of the 19th century -- you sit on and live off of.
2. Barriers to entry in an economy built around such property are much weaker -- and, as noted, strictly a function of things like patent laws.
3. In such an economy, it seems to me (but what do I know), who "the rich" are (or seem to be) a constantly changing cohort.

I have always considered my self conservative and/or libertarian especially in terms of economics but it seems that the real issue here is the distribution of capital not just r > g. When capital is concentrated in the hands of a few the result is similar to monopoly. In order for capitalism and free market economy to deliver their promise capital must be spread at some distribution range X or greater among individual and entities in order for competition and other free market mechansims to perform for the greatest good. Communism and Patrimonial Capitalism can both be seen in terms of capital distribution as the same, in one the Communist party holds all of the capital in the other the 1%. But then you say oh no the 1% earned it, but did they? Really?

I recall the recent publication of a Harvard study that supports the first part of the below quotation, that is, the lack of upward mobility in the US. However, I am unaware of any research that has indicated the latter part of the statement, that there are certain immigrant groups more successful than others. Is there empirical data on this point somewhere?

"To be sure, outside the realm of the ultra-elites, the United States suffers from unfairness in terms of who gets ahead in life, and a lack of upward mobility profoundly affects the prospects of lower-income Americans. Still, the success of certain immigrant groups suggests that cultural factors play a more significant role in mobility than does the capital-to-income ratio, since the children and grandchildren of immigrants from those groups tend to advance socioeconomically even if their forebears arrived without much in the way of accumulated fortunes."

Policy Wonks unite..... Are we left with no argument but whether r > ???

Cut it any way you want. The mal distribution of wealth is based on power not some obscure inequality (mathematically that is). Pun intended.

With this power--in the hands of oligarchs or the next 10-20 percent-- those remaining individuals and families are simply not left enuf to make the human and societal investments to sustain such a society for the long term. Take a look at our ridiculously narrow definitions of Capital to those of 'business' or mfg, And a tax system that completely ignores them. What about human capital, natural capital, social capital (law, education, etc), cultural capital.

The business can deduct all the costs--expensed or amortized--of the production/service. Every drop of grease, the transport costs of getting resourses to the 'factory,'
Then he pays taxes on the remainder. Well maybe so maybe not. He owns the congress and the courts.

Then a worker, his'her own capital asset. No fuel--food-- deductions for them, medical only if exceeding 7,5 PCT of adjusted income, no transport to get 'their resouce selves' to work. And then no amortization for their retirement and care until death--other than what they can wrest from their employer. An most important, no 'investment tax credit' for bearing, feeding, educating and medicating the children by which a society can survive into the future. Such people--in passing--who are the very reason and economy exists. It is not the other way around!! Lets put first things first. Morons.

"The actual increases in income inequality we observe are mostly about labor income, not capital income".

What planet do you live on? Ginormous bonuses and raises to financial CEO's, and you call this labor income? You know the economics: these are rents, and not scarcity rents.

I try to use a sovereign wealth fund to even out inequality


it would be the largest such entity on the planet in only a few decades, I am thinking that it could never have very high earnings do to its size

I wholeheartedly agree that the return on capital is, above all, a compensation for risk. This is completely transparent in that book, and the whole analysis seem completely flawed -- at the very least on this respect.

But it is not surprising from Piketty. This guy is a pure product of the socialist party/network in France, which entirely rules academic life in France, and since his teenage years he has lived on it, made his career on it, and got protected by it. To fix ideas, his ex girlfriend is the current French minister of culture, and they broke up after taking a beating from him. Needless to say that Piketty has never set a foot in a business of whatever kind, and that he believes that a CEO is the bad guy whose job is to spoil deserving workers.

Comments for this post are closed