By Jeff Madrick, a regular contributor to The New York Review of Books, and a former economics columnist for The New York Times. He is also the editor of Challenge Magazine, visiting professor of humanities at The Cooper Union, and senior fellow at the Roosevelt Institute and the Schwartz Center for Economic Policy Analysis, The New School. Cross posted from Triple Crisis
The Piketty bubble may be coming to an end. Economists are starting to criticize the heart of his argument. That is not to diminish important aspects of his book. But the most profound of his claims simply may not hold.
Arriving with a fanfare worthy of Caesar, Thomas Piketty’s long book, Capital in the 21st Century was at first welcomed almost uncritically by enthusiastic centrists and progressives both. Why not? As one read the first sections of the book, who wouldn’t have? I am an admirer and remain one. Here was an economist widely respected in the mainstream telling us point blank that the rich earned far more than they deserved, that economic theory regarding labor markets failed, that the most respected economists had little sense of the real world, and that inheritance was a source of persistent inequality.
Most impressive was the quantity and depth of empirical backing. Piketty scolded economists for depending on models with little empirical basis. This needed to be said by someone so respected. Piketty’s remarkably influential work on income inequality with his colleague Emmanuel Saez is what really revolutionized thinking about economics—and paved the way for his enthusiastic acceptance. Using tax records, they showed the remarkable concentration of wealth in the top 1%—basically they counted how many really rich people there were. Their findings about the extreme distribution of income towards the risk were shocking and confirmed anecdotal evidence.
The empirical analysis in the new book went further. It showed that the equality that existed since World War II and began to reverse in the early 1980s had been an aberration. Capital usually grew faster than incomes throughout history. And it would likely continue to do so! Piketty found that this relation in which r, the rate of return on capital, exceeded g, the growth rate of the economy, seemed permanently etched into not merely history but the future.
And he told us that the best way to deal with such a law of inequality was to tax the rich through a global wealth tax.
I’d hate to put a damper on the enthusiasm. Early critics included James Galbaith and Dean Baker. Galbraith was perhaps the first to question his empirical findings, arguing that Piketty mixed up the price of capital with actual physical capital. Even if Piketty’s right about capital, he and Dean Baker argued early on that there were many other way was to keep capital from rising so fast than to levy taxes. These included financial regulations, anti-trust enforcement, and weakened copyright laws.
But others still believed that Piketty was on to something. And they were right. Except in the mid-twentieth century, the return on capital, or r, he found, was usually—if not almost always—empirically higher than the rate of growth of national income. He defined capital as stocks, bonds, housing, plants, and even firewood if it led to returns. And if the return on investment in them would always grow faster than GDP, then inequality would grow inexorably. Had he found the fundamental weakness of capitalism—a quest Marx himself failed at?
It was exciting to find a strong refutation of the long-held mainstream view that the share of capital and the share of labor in GDP were stable. Workers and investors would split the economy’s bounty according to some unknown law of equality. Now Piketty was saying not so. Historically capital did much better. This is darned important.
But to get there, Piketty ironically returns to conventional mainstream economics without quite telling us he shifted. To him, the invisible hand failed in labor markets, but works in capital markets. Piketty hypothesized that the reason r stayed high was that the “elasticity of substitution between capital and labor” favored capital. To oversimplify a bit, it made sense to invest more in capital than to hire more workers.
But was this true historically, and forever after? Martin Wolf in The Financial Times said it was plausible. So did Jason Furman in a piece written from the White House. But none other than Larry Summers had his doubts. More technically, so did Robert Rowthorn of Cambridge University.
The simple question is: Why doesn’t r fall as returns necessarily diminish? Piketty’s mainstream answer is that new technologies, broadly defined, keep creating new profitable opportunities.
This leaves me at a loss. Pure free markets create r that is always greater than the growth rate of GDP? Fortunately, Lance Taylor, formerly at MIT and now emeritus of the New School, shows pretty clearly that the capital proportion can rise, fall or persist under varying conditions.
And, finally, even if capital rises, is it the central cause of income inequality? In the same piece cited above, Jason Furman breaks down the sources of inequality in recent decades. He finds that the rise in the capital ratio to GDP by no means accounted for the lion’s share of growing income inequality. It’s not clear to me, I should quickly add, that Furman not fully adjusted wage income for stock options that were affected by rising capital values. But neither did Piketty.
Most important, and most bothersome, I heard Piketty say at an assemblage about his book at CUNY that we may not agree with his solution in Part 4 of the book, but we may learn a lot in the first three parts. Some of those who criticized Piketty’s single solution of a high global tax on capital seemed to miss the point that the recommendation was a result of his analysis. If a rising capital ratio is inevitable (as his history empirically suggests), and capital markets work the way the neoclassical models says they do, then taxes are the only tool available.
But as Taylor has pointed out in some technical detail, and others have alluded to, there are many reasons other than the invisible hand for r to remain high. Much of this can involve market failure, which Piketty explicitly rejects. See page 424 if you don’t believe me. R greater than g has nothing to do with market imperfections.
James Galbraith opened his review noting that capital meant power. If Piketty’s empirical analyses are right, and r has been persistently high, I’d suggest it reflects the power of the rich, not the natural forces of a free market economy. Higher taxes would help stymie the rich, but so would financial regulations, anti-trust campaigns, and public financing of politics that could minimize their power and privilege.
I have compiled links to all these reviews on the Bernard L. Schwartz Rediscovering Government web site of the Century Foundation.
I hope this doesn’t sound too harsh. Piketty has done a more than admirable job to trace high capital ratios. He lays the groundwork for more analysis and a true attack on general equilibrium theory and its relevance in the real world. But in interviews I’ve read, he defends himself by saying he’s talked about market imperfections and political institutions in the book. But making many broad general comments is not analysis. His central assertion depends on faith in a general equilibrium model. As he has done in some interviews, arguing that people haven’t read a book as large as his fully is not a defense. It could be equally fairly charged that writing such a large book led him to too many inconsistencies.
In the long run, I think Piketty’s work will indeed prove seminal. It will force economists to deal with the remarkably wide range of issues he raises. But he hasn’t replaced Marx with a more well-founded model of capitalism’s unfairness. For me it is not capital that is power alone. Piketty’s persistently high r, a wonderful discovery, is likely a reflection of the power of wealth not of natural economic forces. With his empirical work we can begin to find solutions about how to constrain the power. But let’s follow his example in regard to income inequality and understand more fully the market failures in capital markets. A global tax would be a wonderful addition to the list of potential tools to bring down r. So let the arguments begin.
89 comments
Comments are closed.
I am glad Madrick bought this up. There is a very real danger that Piketty’s empirical work will get lost in an argument about how to measure capital (one that has been running since Marx published Das Capital). What is important is recognize the power differential between a very tiny removed elite and the rest of us. Solutions are necessary because the elite will collapse the economy that the rest of us depend on to feed and clothe ourselves. By definition a successful solution requires a full understanding of the problem (easier said than done in the field of social science, obviously). We need to recognize that capital is power, and it is the power differential that is driving us to economic instability. It is our duty to save the elite from themselves. They are not fit for purpose and endanger us all with their stupid ‘I am better than you games’.
I have to say that Wray’s proposals yesterday really resonated with me. The powerful will always wriggle out of taxes – and it is really boring having to have the same battle with the top 10% over and over again. I say prosecute with vigour those who have committed fraud, change the law so that corporations do not have the status of personhood. I also say, change the law on the rights to unionize. This is a recognition that power is the issue – and amassing loads of resources is the consequence of overwheening power.
Nicely done, and without the mistakes that accompany my endless battles against spell-checkers.
The real problem is that, in his “prescriptive” area, Piketty is silent or amnesiac about unions. I’ll say “amnesiac” as the harder-hitting version, yet still polite. This is a man who’s from a country where unionized farmers have driven tractors to Paris and unions aren’t part of his solution?
He’s nothing more than a left-neoliberal expecting a technocratic answer.
http://socraticgadfly.blogspot.com/2014/05/piketty-liberal-frenchman-ignorant-of.html
“Solutions are necessary because the elite will collapse the economy that the rest of us depend on to feed and clothe ourselves.”
May I suggest a change of tense to “has collapsed” the economy?
The power of the rich to avoid moral hazard.
Bailouts … TBTF (TBTJ).
Superb critique! I wholly agree with the points made.
I think to understand Piketty you have to hear him speak and explain his book:
https://www.youtube.com/watch?v=27oDSki8yGw
This is a discussion (in French) between Thomas Piketty and Emmanuel Todd (one of my favorite French intellectuals) which is extremely useful in setting the context of Le capital au XXIe siècle. Piketty is exceedingly clear and well-spoken. During the interview it becomes obvious his book needs to be understood within a globalization vs. nationalist framework. It is a basically an attempt to convince the public that the latest economic problems in Europe are not related to globalization, but are due to the inherent nature of capitalism. And he insists that solutions can be found that do not threaten the globalization paradigm. Piketty explicitly says he does not want protectionism. Thankfully Todd disagrees and insists that if Europe wants to keep its social model then it needs to impose protectionism at the European level.
Piketty’s basic argument is that capital (wealth) increases at around 5% a year while the economy is only growing at around 1% and therefore it is natural that differences in wealth are increasing. He says this was the case before the two great wars of the 20th century and it was only because these wars destroyed so much wealth; and in the rebuilding process created high annual growth numbers, that we got the illusion that Capitalism had been tamed and that equality was the future. But now that the special circumstances of the post war period are gone, we are back to the pre-war nature of capitalism.
What he fails to do is link the fact that Europe only grows at 1% a year has everything to do with globalization (Europe’s potential growth is being offshored to cheaper markets). Nor does he mention that the late nineteenth / earl twentieth century was a time of globalization or that the post-war period was a period of low immigration and high trade barriers.
Piketty emphasizes that the ideal and civilized solution (for Europe) would be progressive taxes on wealth that would impact the 5% a year growth that capital increases. The goal would be that fortunes do not increase at a rate higher than the global growth rate. He insists this is an ideal solution. The presenter and the two guests half-joke that in the end it will only be another war or revolution that will actually rebalance the scales of wealth. And in this they are of course correct.
Interestingly both Piketty and Todd claim the US is way ahead of Europe in looking for solutions to the increased diversity in wealth levels. They emphasize that while the US is the world champion of income disparities, and of course sports huge differences in wealth, the Americans are more open to finding solutions to these problems than Europeans are. They note that at the time of the French Revolution there were 30 million French people and now there are 60 million. At that time the US was 3 million and now is 330 million. What Piketty is hinting at is that Europe needs more immigrants so that somehow magically wealth will get more spread out. While wealth may be more fluid in the US, it is certainly just as concentrated if not more. I think what they are saying is that there has been more movement getting into and dropping out of the top 1% in the US than in Europe over the last century.
Piketty emphasizes twice that Europe is an extremely rich region and says that the ratio of European wealth to public and private debt is quite high (5-6 times more wealth than debt) and says Europe owns more of the rest of the world than the RoW owns in Europe. At one point he says Europe is the wealthiest region in the world. This is an attempt to justify globalization. What he doesn’t ask is what good is Europe having so much wealth if because of globalization it is concentrated in so few hands? And while wealth is concentrated, debt is socialized so that the masses have to pay back the debt or suffer austerity while the rich keep the wealth. But he is explicit, his job is to defend globalization and so he doesn’t ask these questions.
Todd agrees with a lot of what Piketty says but disagrees strongly on the protectionism issue. Piketty naively asks why it would be better to have local oligarchs instead of international oligarchs. The answer (not stated) is that labor has more bargaining power against a local oligarch if capital is chained to their local labor by border protections than they do against oligarch free to pit local labor against third world labor.
Piketty attacks the Euro for being a currency without a state; Todd disagrees. In the end they both agree that Europe’s main problem is their dysfunctional political institutions. Todd seems to prefer and strong Europe that can impose protectionist policies. Piketty seems to want more powerful European institutions that can impose his wealth tax and better manage the Euro. His basic pitch to elites is that if you don’t accept this wealth tax, Marine le Pen will gain power and take away all your globalization toys.
In any case it is through globalization / nationalism prism that we can best understand this book and why it is so popular in the Anglo Saxon countries because at the most fundamental level it is a strong defense of globalization. Given US global power, globalization is the new nationalism of the US.
You need to read the Lance Taylor critique. There’s no sound basis for Piketty’s contention that capital makes a steady 5%, or perhaps more important, a 4% premium to GDP growth. Pull out a calculator. It does not take that long before capital that consistently showed that much of a return premium would eat the entire economy. Then by definition it would be the entire economy and unable to earn a premium. His r>g as some sort of constant is absurd, tantamount to “trees grow to the sky.”
http://ineteconomics.org/sites/inet.civicactions.net/files/Lance%20Taylor-Piketty%20Paper.pdf
Ben Johannson translated the Taylor paper out of economese earlier in the week:
On Triumph of the Rentier:
1) Taylor makes the point that Picketty’s determinations of the rate of profit and the capitalists’ share of those profits assume a fully employed global labor force due to his use of the neoclassical production function (the one trashed back in the 1950s during the Cambridge capital controversies). This is THE critical error in Picketty’s work, that capital can be aggregated and differences simply assumed away while the reality of effective demand is ignored.
2)The rate of profit and share of net profits will vary over time depending on the business cycles, employment level, monetary policies, technical changes, etc. The neoclassical production function referenced above does not take this into account.
3) The accumulation of wealth at the top is not an autonomous product of “capital”, some natural law of economics which states that it will always produce growing inequality, but rather a product of specific policies which can be reversed. Altering the ratio of output/capital and the share of profits taken by the capitalist class is the better and more easily implemented choice for reducing inequality rather than taxation. In other words rising real wages is more effective in sustaining aggregate demand and attenuating capitalist power, while relying on taxation will fail to address stagnating wages and continue the current trends.
“Pull out a calculator. It does not take that long before capital that consistently showed that much of a return premium would eat the entire economy..”
Now use that same calculator to determine how long before a steady growth rate for an economy as a whole would convert the entire mass of the planet into consumer products. And then extend that calculation a bit further to determine how long before that same steady growth rate (even one as small as say, .01%) would eat the entire universe, eventually even exceeding the speed of light to do so. All sorts of bizarre and improbable results occur when you plug expected infinite exponential growth into models and then follow them to their logical conclusion.
Which does sort of make Picketty’s model, although his is only one example of the reams of absurd models economists use that bear no relation to reality. Someone recently mentioned to me that economists typically spend exactly one half of their time arguing why their ideas/ideologies are right, and the other half of their time explaining why the predictions from their models didn’t match what actually happened, which I thought was pretty insightful. Picketty’s fetish with models shows that he is little more worthy of credibility than any other economist. Karl Marx and Adam Smith were far better (and more eloquent) as economists than the current no-talent crop–who have to use phony maths and elaborate technical phraseologies to obscure the fact that they are completely ignorant and have utterly no idea about anything at all.
If you haven’t already seen this, you might enjoy it:
http://physics.ucsd.edu/do-the-math/2012/04/economist-meets-physicist/
Thanks –whatever JGordon thinks, I liked it a lot.
Yeah, if you really want a grasp on “dismal science” you need to look long and hard on thermodynamics.
I could be very wrong here (I am certainly not an economist) but I think there is a misunderstanding in what Piketty is claiming. He compares a 1% increase in GDP to a 5% increase in the value of patrimoine which would translate into something like total assets, which is the value of all tangible assets plus financial assets within the nation. For example in the US there is a GDP of around $14 trillion but a total assets number of around $188 trillion which means there is a ratio of 13.4 to 1. Presumably (remember I am not an economist) increases in the value of total assets are not (or only partially) captured in the GDP numbers.
So Piketty is claiming on average next year’s GDP would be $14.14 trillion (1% increase) and Total Assets will be $197.4 (5% increase) giving ratio of 13.9 to 1. I suppose he is claiming this 5% increase in the value of patrimoine through observation of historical data. Of course now I am mixing the US and Europe but I believe this is what he is claiming. The GDP totals and patrimoine totals are mostly independent.
When a war comes along patrimoine is destroyed and afterwards the ratios between GDP and wealth become much smaller and so income can compete more evenly with patrimoine.
I guess I will just have to go get his damn book to be sure…
His analysis of inequality does depend crucially on the dynamics of the evolution of patrimonial wealth. But the 5% number that people usually refer to as a value of r is not a rate of increase of any kind. It is a rate of return to capital.
If the annual return to capital is 5%, and the ratio of capital to annual income is 6/1, then the annual capital share of income is 30%. It is possible in principle for all three figures to be stable over time, and they will be if the ratio of the rate of savings to the rate of growth is also 6/1. If however the latter ratio is less than 6/1, both the capital share and the capital income ratio will tend to fall; if it is greater than 6/1, they will tend to rise.
Thank you! I think I understand it now.
You’re welcome.
In the interview you linked to, Piketty provides a concrete example of this 5%. Let’s say you own an apartment worth 100,000 euros. Let’s say you can rent it out for 400 euros per month. Your yearly return on asset is then:
12*400/100000 = 0.048 = 4.8%
Maybe Piketty is comparing 2 things that shouldn’t be compared. Why would this have anything to do with the rate of GDP progression?
Because when the rate r is significantly higher than the rate of national income growth for an extended period of time, that creates conditions for capital income to grow faster than labor income, and that in turn contributes to a growing concentration of wealth given the inequality of capital distribution.
By the way, Piketty’s notion of national income is not identical to GDP. It is equal to GDP minus depreciation plus net income from abroad.
“But the 5% number that people usually refer to as a value of r is not a rate of increase of any kind. It is a rate of return to capital.”
If it is “a rate of return to capital” and it doesn’t ‘increase’, then where does it go? How is it use? Are you suggesting that it is ‘consumed’ and not recycled in an attempt to increase the return to “capital”, i.e., that it is simple interest and not compounded?
A portion of it is saved and reinvested. Lawrence Summers, for some reason, was under the impression Piketty was claiming that it is 100% reinvested. But Piketty doesn’t make that assumption. He assumes that the wealthy save at much higher rates than those who are not wealthy, but nowhere assumes a 100% savings rate on capital income as far as I can tell.
In one scenario he describes for France in the 19th century, where the capital share of income was nearly 40%, he says that a savings rate of only 25% of capital income was sufficient to generate an increasing concentration of wealth.
Thanks for the feedback. I always appreciate your commentaries, insights and particular point of view.
I would agree that the wealthy save at much higher rates. Assuming they don’t work (“income” from ‘return to capital’, i.e., rentiers) and have physical needs, then 100% savings is not realistic. However, if “A portion of it is saved and reinvested”, then their “capital” is increasing. If savings are being reinvested then this “capital” has to be “growing” at some compounded rate. The math of compounding at any rate is basically f(x) =2^x where x is a ‘doubling’ (rule of 72), where “growth” is represented exponentially. If someone’s “private’’ savings are someone’s ‘’private’’ debt, then eventually this deluded exigency has to be reconciled with real world limits. It was suggested to me that if Creditor/Debtor is a marriage of convenience, then compound interest is domestic abuse.
The overall rate of return to capital includes both actual returns in the form of interest, rents, dividends and other cash returns, and increases in value. To see what he is talking about, look at his discussion of the fortune of Liliane Bettancourt on page 525.
That may help explain why it’s not really a problem that the forms of capital, say physical and financial, are mixed in his discussions; remembering that returns include an allowance for depreciation and for replacement.
Let me first clarify that I read as a student, not as a follower. This means searching for context, which leads to observations and questioning. This is why I really appreciate this site and its wide range of POV’s.
“The overall rate of return to capital includes both actual returns in the form of interest, rents, dividends and other cash returns”
These are all “financial” and any discussion about ‘rate of return’ without clarifying the distinction between simple and compound can be confusing and misleading. The difference in effects are huge. “Return ‘ON’ capital” implies the former while “return ‘TO’ capital” implies the latter. Can you see how the seemingly harmless toggling between these two simple words(on,to) can keep most of us confused, while the wealthy use it as a strategy( toggling between equity and leverage)? Notice how they are used on this thread.
“and increases in value” -…the difficulty in pricing value is from my POV what makes bubbles insidious and extremely ‘profitable’ for a few while extremely damaging to the rest of us.
“That may help explain why it’s not really a problem that the forms of capital, say physical and financial, are mixed in his discussions; remembering that returns include an allowance for depreciation and for replacement.”
I don’t accept the presupposition that “finance” ‘IS’ the economy and therefore from my POV the distinction between physical and ‘financial’ seems very important.
But you are missing that the return to capital is more capital!!! Hello!
In year 2, you multiply r by your year 1 capital plus r from year 1. Piketty’s definition of capital is extremely broad (it’s been criticized for that reason). Folks, this is basic compounding, and that’s where the fallacy in his reasoning lies. It’s bloomin’ obvious. That’s why it being a materially higher rates than GDP growth has to break down, and not in a very long period of time either.
And Taylor demonstrates, contra Piketty, that the return to capital can be higher or lower than economic growth generally.
Hi, Yves,
As a math professor, I have to agree with Dan here. But it’s actually not a math issue, its a nomenclature issue. You are confusing the total AMOUNT of capital with the “Annual Income from that capital”, which is the “return” on capital, and which Dan referred to (perhaps a bit confusingly) as the “total capital share”. What he meant was the “share of income that constitutes income “returns” from the Total Capital Stock”, I believe.
It is the latter (a quantity much smaller than the total amount of capital) that Dan and Piketty are comparing (in ratio form) to the “Total Annual Income”, which is (roughly) GDP.
The Total Amount of Capital is already much more than 100% of GDP. But as Dan correctly asserts, the Income from Capital, divided by total Income (GDP), will stabilize (or fluctuate around) some ratio much less than 100% given the kinds of growth rates for r and g that Piketty finds evidence of.
I will also just say that Piketty may not be Einstein, but no economist of his stature could have possibly failed to note the math contradiction you attribute to him–if he were actually comparing the quantities you are comparing. But he’s not, and his and Dan’s math is certainly correct! As for the rest of his analyis, I’ll withhold comment until I read the book…though Dan’s very nice summary may well save me the effort!
No the total capital share in a given year is not the same thing as the rate of return on capital in that year. The total capital share is equal to the the rate of return on capital multiplied by the capital-to-income ratio.
Here’s an example: Suppose a society has accumulated $6 trillion in wealth and that its national income in Year 1 is $1 trillion dollars. Then its income-to-capital ratio is 6/1 or 600%. Now suppose the sum total of all income in Year 1 that comes from profits, rents, dividends, interest, royalties and capital gains – i.e. all income generated by capital rather than labor – is $300 billion, which is exactly 5% of the total accumulated capital stock of $6 trillion. That’s the rate of return to capital. $300 billion is exactly 30% of the total annual income of $1 trillion. That’s the capital share of income. By algebraic necessity, the capital share Yc/Y is equal to the return on capital Yc/W times the capital-to-income ratio W/Y. In this particular case, that equality is instantiated by (0.05) x 6 = 0.3
Dan is again correct, and consistent in his terminology, but I’d just caution the average reader that the nomenclature can be misleading. The phrase “total capital share” has, out of context, at least 3 different interpretations, depending on whether share has an absolute or relative meaning, and on how the words modify one another syntactically. Yves used an interpretation not intended by Piketty, unlike Dan. I thought my explanation was equivalent to Dan’s, but perhaps it carried a different implied usage of “share” w.r.t. relative vs. absolute. Regardless, I’m sure Dan’s usage is most appropriate here, since he’s actually read Piketty, while I’ve only read reviews, including Dan’s very cogent summary. :)
However, given that Yves’s confusion was quite elementary, I think Dan’s follow-up comment just below is therefore a tad too much math overkill. It’s again certainly correct, but Lambert’s bemused comment on it below highlights an important issue.
This is that too much unneeded focus on “involved” math really does dissuade average readers from getting the real issues. I’m a math professor, but other than glaring, elementary errors in the math, one doesn’t need sophisticated math formulations to demonstrate or illustrate the fundamental injustice of our capitalist political economy, nor to propose major alternative conceptual frameworks.
Lambert’s comment is so nuanced that I actually don’t know how (meta-)ironic it is–he and Dan were just the other day disagreeing (amiably, mildly) about MMT (which isn’t involved in the Piketty math issue), with Dan wanting more sophisticated math models from it to generate budgets and predict inflation crisis points. I see no need for that–one doesn’t need a math model to realize one is driving one’s car too fast or too slow. One just steadily presses the brake pedal or accelerator until the speed becomes much more appropriate to the surroundings. Then one stops pressing that pedal.
At any rate, now that Piketty has revealed the previously unknown fact that there is grotesque inequality and oppression under capitalism, it is indeed on to the next steps: creating workplace democracy, empowering labor, and organizing real political opposition to our capitalist overlords. No math required to advocate those things…and what could be more fun! :)
But thanks! And i think you should read the book.
Well, if you want people to read the book, you’re going to have to stop publishing such clear and concise summaries of it… :) For example, your comment below of 5/22 8:38 pm in reply to Yves was extremely well done, particularly because the math was almost all extremely simple and concrete–yet (rather, hence) convincing.
Out of curiosity/laziness, what’s the ballpark current U.S. capital-to-income ratio?
Yves, you cannot treat Piketty’s wealth growth framework as just a compounding problem where the invested stake is continually rolled over. For one thing you are completely ignoring the national savings rate, according to which only a small fraction of national income is typically added each year to the capital stock. You are also leaving out the fact that there are sources of income other than returns to capital. Here is the framework:
We need to employ the following quantities:
Wi – the nation’s accumulated capital stock in year i; it’s wealth.
Yi – the national income in year i
si – the national savings rate: i.e the portion of national income in each year that is added to the capital stock
YCi – income generated from capital in year i
ri – the rate of return on capital: i.e. the ratio of income generated from capital YCi/Wi
g is the rate of growth of national income
Here’s the rule that states how wealth accumulates from one year to the next:
Wi+1 = Wi + sYi
And for income growth we have:
Yi+1 = (1 + g)Yi
The capital-to-income ratio βi for any year i is equal to Wi/Yi
The capital share of income αi in year i is YCi/Yi, which is also equal to ri times βi
Now, assume initial values of 600 for W1 and 100 for Y1. Assume a constant savings rate of 10%, a constant growth rate of 2% and a constant rate of return to capital of 5%, which means that the income form capital YC1 in year one is 30. The initial capital-to-income ratio β1 can be seen to be 6, and the initial capital share of income α1 is clearly 30%. Now do an extended time series computation for αi and βi, using the wealth accumulation rule and the income growth rule. You will see that that neither figure grows without bound, but that αi converges to 25% and βi converges to 5. If we assume a 1% annual growth rate instead of 2%, then αi converges to 50% and βi converges to 10.
Piketty is perfectly aware that that rate of return on capital can go up or down. He describes scenarios in which that happens. One thing that he points out, however, is that once the ratio βi gets sufficiently high, then the capital share αi can continue to rise. He also demonstrates empirically that the actual rate of return to capital is usually between 4.5% and 5%.
I already mentioned in this thread one mistake that Taylor makes right at the outset. He claims that Piketty’s 2nd law, stating that βi converges to s/g in the long run for stable values of s and g, is an identity: i.e. that β is identically equal to s/g. But it is nothing of the sort. It is only a long term asymptotic law. Taylor then claims that r > g follows from this identity and other national accounting identities as a theorem. But again, that is wildly off. There is no conceptual necessity for r > g. Piketty emphasizes in several places that r > g is a contingent empirical regularity that can be observed to have held true throughout almost all economic history. But we can easily conceptualize situations in which it is false.
Still waiting for a human wave of MMT h8ters to complain how complicated this is…. #justsaying.
Is this something Professor Picketty wrote when he was in Italy?
:-) bowwaaahahahahah
It’s easier just to tax their assets off. Nobody will ever understand this if they’re a normal person. Even the people who pretend they understand don’t really understand. ‘Assume a space ship that can travel 3 times faster than the speed of light,, now assume a galaxy 599 light years away, then the space ship can reach the galaxy in less than 200 years or about 10 minutes iif you’e on the ship. this means you can be back for lunch if you leave now. But don’t make the sandwich until you get back because there’s no guarantee the regrigerator will be there
‘[Piketty] also demonstrates empirically that the actual rate of return to capital is usually between 4.5% and 5%.’
Rob Arnott and the late Peter Bernstein found, surveying 200 years of U.S. return data, that ‘the history of dividend growth shows no evidence that dividends can ever grow materially faster than per capita GDP.’ (p. 73)
They add (p. 80) that ‘the real internal growth that companies generated in their dividends averaged 0.9 percent a year over the past 200 years, whereas … the increase in real per capita GDP averaged 1.6 percent.’
http://www.researchaffiliates.com/Production%20content%20library/FAJ_Mar_Apr_2002_What_Risk_Premium_is_Normal.pdf
————
If Arnott and Bernstein’s results are general, there should be no intrinsic problem with capital returns systematically outstripping labor returns.
Hi, Jim,
Check out Nathanael’s comment way below, May 22, 11:44:
‘You have to remember that when he says “capital” he means “wealth”, and when he says “return on capital”, he means “rents collectible by holders of wealth”. ‘
And check out Dan’s immediate (confirming) reply:
‘Yep, he says it includes (at least) profits, rents, dividends, interest, royalties and capital gains.’
The point is that it’s not just dividends that constitute the (alleged) 5%-ish returns to capital, as Piketty defines capital. Doesn’t mean Piketty’s right, but using data from the study you cite to critique Piketty would seem to be comparing apples to kumquats…
Hi, Jim,
Sorry, I think in my comment of 12:02 am I may have addressed the wrong error in your comment… :)
I’m not sure without seeing the details of the account you cite, but Dan’s point is still relevant: the rate of return “r” that Piketty discusses is not a rate of “increase” or “growth”. It is just a rate of return to invested capital–the paradigmatic instance of it is the “average interest rate” on bonds (all bonds, including Treasuries, muni’s, corporate, junk, whatever…)
There are of course other forms of return to capital, but if you keep that example in mind, you will have a good paradigm. Note in particular that the average rate of interest on (all) bonds hasn’t grown much, if any over time. It varies slightly within a fairly stable, bounded range.
But let’s return to dividends. So, for example, suppose 200 years ago the average dividend rate was 4% (of price of stock.) The fact that absolute (real) dollar value of dividends increased at .9% per year (which seems to me to be the only reasonable interpretation of the figure) doesn’t tell us anything about the rate of return (absolute dollar value of dividends divided by the price of stock). The latter depends on how the price of stock changed, in addition to how the dividends themselves changed.
So, the (hypothetical) original rate of 4% (of stock price) return represented by dividends may or may not have increased over 200 years. The relevant figure is the dividend rate itself, not the “growth” in absolute dividends. I have no idea what the dividend rate has been for the last 200 years but it seems to me that for quite a while recently it has probably been in the ballpark of 3 to 4 to 5 %, yes?
And then, of course, my first comment remains true–many other returns to wealth are covered by Piketty’s “r”, besides dividends.
I see little reason to doubt that his overall figure of 5%-ish isn’t ballpark(ishly) true…
And, of course, no math at all (other than the number line) is required to see that massive injustice in income distribution is present in capitalist political economies! All that is required to see that is to consider how people should be rewarded for their work: on the basis of time (average, per worker per type of job), effort, and sacrifice, rather than on the basis of luck or talent.
Pull out a calculator. It does not take that long before capital that consistently showed that much of a return premium would eat the entire economy.
I’m sorry Yves, but your math is simply wrong here. In an economy that maintained a consistent 5% return to capital, a 1% growth rate and a 10% savings rate, the total capital share would converge to 50% of annual income, not 100%. And if the growth rate were 1.5%, the capital share would converge to 33%. You can easily work this out on a spreadsheet.
The Taylor paper commits a pretty big howler at the outset in claiming that Piketty’s so-called “second fundamental law of capitalism” is an identity. It is not. It is a long-term asymptotic law that says that, for fixed savings rate rand fixed growth rate g, the capital-to-income ratio β converges to s/g in the long run. Not only is β = s/g not an identity, but as Piketty points out, an economy can pass through a period of many years during which β = s/g is not even approximately true. And of course savings rates and growth rates never do stay rigidly fixed. The second fundamental law is really a statement about the direction in which the capital-to-income ratio and capiatl share are moving, and how fast they are moving.
Piketty is entirely aware that r can move up and down, and he does not derive r > g as some kind of a theorem. It is a contingent historical regularity whose robustness over time Piketty documents on the basis of historical evidence. He considers possible reasons why it holds so consistently, but comes to know definitive answer and argues that in the end it seems to be the result of a confluence of diverse social, political and cultural factors.
Piketty’ discussion of the CES production function framework is not a whole-hearted embrace of that framework. He claims in the online technical appendix that “it is extremely simplistic to summarize all of the possibilities of substitution between capital and labor (as regards technology as well as consumption patterns) with a sole parameter.” What he is doing in that section is showing that even if you assume that framework, you only need to assume small elasticites greater than 1 to get the result that the capital share is an increasing function of the capital-to-income ration, and thus varies a lot with variations in the capital-to-income ratio varies. His target is defenders of the self-correcting nature of capitalism who hold that the capital share is stable over time.
There is another article by Robert Rowthorn than claims that the empirical evidence does not bear out a recent historical increase in the the capital-to-income ratio. But Rowthorn claims that Piketty implicitly “assumes” that β is equal to the ratio K/Y where K is capital “as conventionally understood”, and also that the capital share α is the “share of profits in national income.” But Piketty denies both of these claims. Piketty defines β as W/Y, where W is a more comprehensive concept of capital as “wealth” which is broader than the conventional concept of capital, and he explicitly argues that profits are only one portion of capital income. So when he then uses historical data to compute elasticizes within the simplified production function framework, these elasticizes are not going to come out to be the same thing you would get if you look only at capital and the capital share “as conventionally understood.” People have claimed Piketty is “confused” about the correct definition of “capital.” But not only is he not confused about these matters, but he is quite explicit in the earliest chapters of the book about how his broader notion differs from the narrower notion; and employing the broader notion is essential to his argument.
But all of these points have to be put in context, because the analysis of the structure of inequality in chapters 7 through 12, and the forces for income and wealth divergence that tend to generate greater inequality, does not depend on a continually increasing capital share. You can have increasing inequality with a stable capital share. You could also – in principle – have a very egalitarian system with a continually rising capital share: that’s what you would get if capital were evenly distributed. The former scenario, however, is much more relevant to actual historical circumstances.
Piketty is perfectly well aware that the tendency for these forces to produce inequality is the effect of policy, and not some indelible law of nature. He thinks that in any system based on private property and private investment, capital will always get a return, and that the size of that return in conjuntion with the social customs of inheritence and gift-giving by which capitalists transfer wealth to their heirs will in most historical circumstances lead to growing inequality unless the public then taxes some of that return away from them. So he argues for taxing it away from them. But he also argues for new forms of “democratic control of capital” which can have the effect you mention of reducing the private return to capital before taxation.
Look what’s going on here. The Marxists don’t like Piketty because they have an analysis that predicts capitalism leads to an endless increase in the capital share, a crisis of profitability, and then possibly revolution. The liberals don’t like Piketty because he is arguing that the problems with actually existing capitalism are deeper than those that can be fixed just by making markets more perfect while building a safety net. Piketty is arguing that capitalism can persist in a relatively stable condition for a very long time where the rich just keep getting richer. And in the end, even after returns to capital fall and this process of increasing inequality levels off, you don’t get crisis or collapse, but a stable neo-feudal system with a flat, but very high, capital share, and with rich people collecting their rents year after year.
I don’t think Yves was suggesting convergence at 100%, but that the result would be sufficient for rapid national ruin. 50% is a reasonable point at which to expect that outcome particularly in a populous nation.
Well, she said it would “eat the entire economy”. And that is not true.
Also, Piketty doesn’t predict a 50% capital share. He produces a simulation which shows a 7/1 capital-to-income ratio by the end of the 21st century, which with a return to capital of 5% would give only a 35% capital share.
In the Beginning God Created Math, and there Was Math, and He Looked Upon the Math, and He got Confused
Holy Smokes you really did read that thing! I do wonder, though, if the economy is growing at 1% how the savings rate could be 10%, unless capital was doing most of the savings from the income it receives from itself through the work of L.
there’s also a theoretical problem with “non-stationarity” of the price structure which makes constants not very constant.
Also, if all the bailouts of capital that always seem to happen didn’t happen, it could be that r C, like in Russia in 1917. That’s an example of non-stationarity.
Otherwise, I do appreciate your brief reports since I’ll never read this doorjam but would kind of like to know what it says! Just in case.
Well the savings rate is a percentage of income. If income falls, savings will fall, but the rate could in principle stay the same.
And yes, Piketty does stress that the wealthy save at rates much higher than the average savings rate. If for example, 20% of capital income is saved and only 5% of labor income is saved, but the capital share of income is 30%, then the overall national savings rate would be 9.5%.
Piketty defines capital very broadly, in fact comprehensively, and from what I have read, does NOT differentiate “savings” which you treat as inert, like a liquidity buffer. As Piketty defines capital, the return on capital in one period becomes capital in the next time period by virtue of being capital-type assets (loans, stocks, investments in more production, etc.)
Knut, who has read Piketty (see his comment below) confirms my understanding:
Piketty’s concept of capital is what the rest of us call net wealth or net marketable wealth. It includes what the classical economists called ‘capital’ but is far more. Housing accounts for almost half of marketable wealth in advanced countries, and government debt a considerable share. The book is about a fact–the wealth/income ratio tends to grow over time, and the distribution of that wealth comes increasingly to be concentrated in the top decile and top centile of the income distribution. Unless someone comes up with comparable data that refute these findings, we have what passes in economics for a hard fact, just as the constancy of the speed of light is a hard fact.
Economists trained in pure theory have a very hard time getting their minds around facts that don’t jive with their theory but are nevertheless facts. I could give any number of referee’s reports written by people whose priors are so strong no amount of evidence can alter their opinion. Piketty’s work is a massive challenge to that mindset. He does not offer an over-arching explanation of his results, but simply reports them with some plausible but not binding interpretation.
It is crucial to keep in mind that Piketty’s concept of wealth is essentially the purchasing power of that wealth, not its productivity.
I believe that is your fix and not present in Piketty. Thus the reinvestment and compounding issue I have raised holds.
As Piketty defines capital, the return on capital in one period becomes capital in the next time period by virtue of being capital-type assets (loans, stocks, investments in more production, etc.)
No, Piketty does not make this assumption. The return on capital is just one component of income, and can take many forms, and only a fraction of that income is added to the capital stock.
As you note, for Piketty “capital” refers to all wealth. Specifically, it is “the sum total of all-nonhuman assets that can be owned and exchanged on some market.” A government bond is capital; and so is a tractor; but so is a can of beans; and so is a hula hoop; and so is a car; and so is toy car. It’s much more intuitive in understanding what Piketty is up to to just read capital as “national wealth.”
Every year the nation produces some of this kind of stuff; and its get some other stuff from abroad while shipping other stuff abroad; and some of its old stuff is lost due to depreciation. (His concept of depreciation applies equally well to the tractor that falls apart or the can of beans that goes bad on the shelf.) The national product plus the net acquisition of stuff from abroad minus the depreciation is what Piketty defines as “national income”.
If the society starts with accumulated wealth W at the start of year one and a national income Y in that year, what is the accumulated wealth at the start of year two. Is it W + Y? No, of course not. Most of what is produced in a year is consumed in that year, and a lot of old stuff is also consumed or goes to waste. So the addition the new national wealth will only be W + sY, where s is some number between 0 and 1. This is the number Piketty calls “the rate of savings” and he argues it is typically around 10%.
So if you’re a subsistence farmer, and you’ve got 100 Bell jars of beans in your cellar at the start of the year, and you grow and jar 100 more jars in that year, and that’s all your income, and you consume 80 jars and 10 jars have to be thrown out because they are too old, then you have made a net addition of 10 jars to your wealth, which is 10% of your income that year, and your stock of bean wealth has grown to 110 jars.
Note that many parts of a nation’s accumulated wealth earns no return at all – probably like the beans, unless someone will pay you to rent them and look at them – but other kinds of wealth earn very high returns. The 4.5% to 5% number is Piketty’s measure of the typical average return on the wealth stock.
I haven’t made up any conceptual innovations here to help out Piketty. This framework is explained with admirable and unpretentious clarity in the very first chapters of the book.
A lot of the economists I have read so far begin their critiques by first translating Piketty’s framework into their own similar but different frameworks, using such concepts as “profits”, “productive capital”, etc. , and then run everything through their own pre-existing model and say, “Hey, this doesn’t work!” But Piketty explicitly rejects these limitations. He stresses that profits are only one component of returns to capital and that so-called productive capital (a term he finds of dubious relevance in the first place) is only one part of capital as he is using the term.
That means that every other concept in the book that depends on these basic concepts must be interpreted accordingly. For example, when Piketty arrives at an empirical measure of the elasiticity of substitution of capital for labor by looking at the historical evolution of capital and labor and applying the CES framework to those numbers, the capital whose substitution elasticity he is trying to measure is Piketty-capital, that is, capital as he has defined it. So if an economist comes along with some alternative and more restricted form of capital, and says they have come up with a lower number for its substitution elasticity, the criticism misfires.
Dan, saving rates can increase in response to declining income.
Yes, I know.
This seems credible for working “middle class” who can afford to, as an emergency buffer…less plausible for the oner’s, who don’t have to worry about modifying lifestyle and/or standard of living… For the vast majority it means prioritizing decisions for survival and can’t even ponder what it means to save.
Seems to me it doesn’t have anything to do with God or Math.
More like labor and tax arb in the case of multi-nationals – if the marbles can roll downhill, then they will find the lowest point on earth.
In the case of domestics, striving towards monopolies always works, along with setting up legal barriers to competition. Union busting labor tactics, or just ensure the labor market gets flooded somehow is always good for a little more profit margin.
Leveraged takeovers, then some asset stripping is always good for a quick buck. Flip it into junk bonds and pocket 30% of your deal on the front end is too good to pass up.
The devil is in the details – but you can make the devil disappear with math. We’ll never know what happened. Good thing I can still remember my working career!
If that’s the case, you need to smoke more reefer! It’s amazing that after some people get rich the only thing they do is trry to get even more rich. I’d settle for just a few million and then I’d do nothing but lay around and waste time. Maybe ride the bus all day long instead of just in the morning. That sounds like a good way to waste time.
If you tax their assets off, then the economists say you reduce their incentive to work. That sounds good, because they work too much as it is! Much of their work, after a certain point, becomes redundant to work they’ve already done and just fools people. It isn’t reallly new work, it’s just old work in disguise. If you tax them, maybe they’ll stop. hahaha
If they stopped, then they’d do less but other people would have a chance to do more. Some people have nothing to do and some people could lay around all week if they wanted, but they just work until they mess things up so badly it’s almost easier to start over. That’s too much work, so the best thing to do is tax their assets off so we don’t have to.
But don’t tax my assets, no way,
Yeah, it’s amazing when you realize all you need to do to get a guy like Michael Milken to quit is to tax his great big ass.
Duh![Simpson head slap]
How come no one ever figures this stuff out? They must all be busy working – like figuring out if spice money proceeded spice money debt or was it visa versa? Or some such thing. But something must be keeping all of ’em busy.
As far as taxing your assets (and mine) – well, that’ll be easy:)
“What he doesn’t ask is what good is Europe having so much wealth if because of globalization it is concentrated in so few hands?”
And what good is Europe having so much wealth if you’re not a European? The argument that Globalization is great because Europe has gained wealth thereby, isn’t nearly as convincing if you live in, say, Africa or South-East Asia.
Last time I looked at Piketty’s World Top Incomes Database, in France, the income obtained by the 10Percenter Class in France was around 33%, whilst in the US it was at 46% for the year 2010.
Income becomes wealth, if you’ve enough of it. The division of wealth from the Domhoff studies of US wealth (see here) indicate an accumulation by the 10Percenters that is even worse than Income at 77% – see PieChart here.
Mathematically this is an “optimization problem”. You seek to optimize C given L with certain constraints: such as distribution of C across L, distribution of income across L, distribution of income (I) and C across regions and geographies, mobility (or lack there of) of both C and L, the conditions in which L operates to create both I and C, regulation of C, etc., market structure in which C can be traded, etc.
However, few can agree as to these factor specificities, definitions, constraints, or even the scale upon which they will be measured. This produces a mathematically intractable problem. One can try scenario analysis through Monte Carlo simulations to entertain and frighten oneself, but one can’t evade the lack of specificity. Nor can one evade the problem of “non staitionarity”, which is that whatever scale you choose. the scale itself shifts over time in ways that render whatever measurement you make serially unstable. In other words, you never have any real information.
These are daunting problems, which is why, as model complexity increases, decision making strategies approach politics. And as model complexity decreases, model output approaches an arbitrary and random superficiality.
It’s hard to make sense of this stuff with math. Not that I can either. I’ll be the first to admit it. Unless somebody has beaten me to it.
:)
post-war period was a period of low immigration
That sure aint true for France. Most north Africans arrived in France in that period specifically to help with the reconstruction. They fully took part in the making of les 30 glorieuses, and are now being thanked by being parked in the dreary banlieues.
Followed Jeff Madrick all the way til the last line, “So let the arguments begin.”. Madrick points out that power and wealth accumulation are so insidiously entwined, that wealth will use its power to frame, manipulate, and control the debate — from all sides. Heterodox economics hasn’t failed in practice, it’s failed to be heard (and understood) due to the power wealth has accumulated.
Madrick’s summary repeats several errors that have appeared in the recent reviews. Fir example, when he says ….
The simple question is: Why doesn’t r fall as returns necessarily diminish? Piketty’s mainstream answer is that new technologies, broadly defined, keep creating new profitable opportunities.
he’s wrong on both the question and the answer. Piketty doesn’t say r won’t fall as new capital is created and an abundance of capital eventually reduces the return on capital. What he says is for a long time the “price effect” of the fall in r can be outweighed by the volume effect of a high capital-to-income ratio. The capital share of income can continue to rise even as the return on capital is falling.
Also, Piketty does not simply appeal to technology to explain the persistence of the r > g inequality. Piketty believes that r > g is a contingent historical fact that does not hold either due to logical necessity or to some single theoretically all-powerful reason, but for a “variety of technological, psychological, social and cultural reasons” – a “confluence of forces” that have historically produced returns on capital of between 4% and 5%.
Following up on the previous point, Madrick says:
This leaves me at a loss. Pure free markets create r that is always greater than the growth rate of GDP? Fortunately, Lance Taylor, formerly at MIT and now emeritus of the New School, shows pretty clearly that the capital proportion can rise, fall or persist under varying conditions.
But Piketty also allows that r can rise, fall or stay the same. The Taylor piece Madrick cites claims r > g is a theorem of national accounting, and that Piketty’s “Second Fundamental Law of Capitalism” is an identity. But both of these claims are false, and no part of Piketty’s argument. r > g is not some kind of axiom, or a theorem deduced from first principles, but an observable historical regularity.
Piketty’s main message here is something like, “What are you going to believe, high economic theory or your own lying eyes?” What Piketty stresses is that r > g is an empirically confirmable historical regularity that has held true throughout human history, with the ratio r/g often being quite high, sometimes by large margins, and and with r dipping below g only during disasters. So if you’ve got some economist with telling you, on the basis of pure theory, that r is bound to fall to a low level relative g, there are thousands of year of actual history telling you otherwise.
On market imperfections, Piketty does not deny the existence of market imperfections, and addresses them in several places, particularly in the section on inequality of labor income. But what he is arguing is that market imperfections will not save the those who argue that inequality is a self-correcting phenomenon in a properly functioning capitalism, which will go away if the market imperfections are eliminated and markets are made to function as textbook-style free, competitive markets, because even under those circumstances capitalism contains forces that tend to increase inequality. In the passage Madrick cites, Piketty is advising the reader that his main narrative is not a story of market imperfections.
As I said, the reviews of Piketty are filled with errors. I have attempted to identify some of those errors and clarify Piketty’s argument in a series of posts:
http://ruggedegalitarianism.wordpress.com/2014/05/15/a-proper-pile-of-piketty-posts/
See especially the top for posts in the list dealing with Summers, Hassett, Cooper and Avent. I have more coming.
Dan:
I’m on my second reading of Piketty, precisely for the reasons you have hopped in on here, in answering Jeff Madrick. And I concur with your rebuttal. As to the other factors which may affect capital accumulation and its rate of return, don’t underestimate the not fully explicated but suggestive implications of the words Piketty uses in Chapter 7, “Inequality and Concentration: Preliminary Bearings,” specifically on page 262: “Indeed, whether such extreme inequality is or is not sustainable depends not only on the effectiveness of the repressive apparatus but also, and perhaps primarily, on the effectiveness of the apparatus of justification.”
Those two words “repressive and justification” clearly tell me that Piketty is quite aware of the power of ideology and politics to affect the shares of wealth and income, although his data maintains that over three centuries capital and its “apparatuses” in these two areas is the usual winner. Robert Kuttner and Thomas Frank and James Galbraith to a lesser extent have rightly pointed out that the great compression, the golden days in the West capitalist economy stemmed from political policy interventions via the New Deal and other egalitarian measures in Europe, not from the more deterministic mechanisms of the Great Depression and World War II’s great destructiveness. I believe Piketty has admitted he may have underestimated the policy interventions of the New Deal in making the US more egalitarian than usual from 1935-1973.
And let’s not forget that in that same Chapter Seven, Piketty says this on page 265, which surely must resonate as one of the most damning indictments of the current state of economic and political affairs in the US in history, even as I have to qualify it by saying Piketty overall gives greater stress to wealth inequality than income, and especially wage inequality: “…and what primarily characterizes the United States at the moment is a record level of inequality of income from labor (probably higher than in any other society at any time in the past, anywhere in the world, including societies in which skill disparities were extremely large)…
I think on the whole Piketty has done a great service for a better economy and it is bound to reinvigorate the left. Especially on the matters of the returns on capital, its nature and composition and the laws that it may or may not operate under, Piketty’s work certainly will invite further commentary and scrutiny from serious Marxist economists, where all these matters and definitions are very contested terrain.
I have said in public written comments that Piketty is like a man sitting with a lighted candle atop an enormous pile of ideological fireworks, attempting to illuminate them but not set off an explosion. Good luck with that. The material he covers, as he has explicitly noted, are historically and politically charged and fraught. Like it or not they will go off in many directions and start little fires, or great ones, of illumination we hope, which may spread and lead to constructive and great changes.
I have also wondered in email discussions whether Piketty knows very well the troubles Krugman has had with the profession over New Deal type interventions into the economy under present ideological conditions in politics and within the economics profession itself. No more dangerous ground exist than calling for a living minimum wage (which is not 10.10, by the way, and Piketty is self-contradictory on the connection between productivity and our current wage on page 313: if we give labor its missing share of productivity gains following his own analysis, we come out between $16 and $22 per hour, not the $9 he mentions) or public employment programs.
If critics needed any proof that the more explicit one gets in matters of left leaning political economy, the worse the reception, just consider not only the troubles Krugman has had with the profession, but that Wendell Berry encountered with his great Jefferson Lecture in April of 2012 – the nation’s highest award in the humanities and easily found online. He ventured into the political economy, and did so admirably along the lines coming to dominate the present moment…yet the intellectual establishment totally ignored his warnings – and the speech, delivered at the Kennedy Center in DC. I can’t prove it, but I suspect Piketty chose not to go down that path. But he has paved the way to make it easier for others to do so. I thank him.
Thanks William. I’m glad you cite Chapter 7, because the story has gotten out there that Piketty does not deal with inequality of labor income, when in fact he has devoted a whole chapter to it.
As far as I’m concerned, the core of the book is Part Three on “The Structure of Inequality”. It consists of 6 chapters and about 240 pages, and yet that part of the book has gotten remarkably little attention from the reviewers.
The ideological fireworks all seem to be taking place over chapter 6, where so many economists have something professionally invested. But the role the arguments in chapter 6 play in the context of the book’s overall argument is widely misinterpreted. The argument about how inherited wealth increases its own share of the capital share is just as important as considerations of how, or even whether, the capital share is increasing.
Excellent work, Dan.
You should get some of those posts published here at Naked Capitalism.
They are all available for re-posting :)
Picketty makes a series of assumptions which tend to undermine his preferred solution. The capital controversies weren’t just about capital aggregation, they established that contra the neoclassical description there are inflection points at which labor and capital will substitute for each other in a counterintuitive fashion. Combined with his reliance on market valuations this creates some serious problems for reliance on how a surplus accumulates and what concentration of inputs is preferred. In his attempt to draw a trend line Picketty assumes that capital intensification is inevitable, a view right out of the mainstream textbooks and one decisively countered by Solow over forty years ago.
To put it simply there’s no way to get there from here following Picketty’s narrative, unless we ignore a few things.
In his attempt to draw a trend line Picketty assumes that capital intensification is inevitable.
No, I don’t think so Ben. He argues on page 233 that:
“it is also possible that technological changes over the very long run will slightly favor human labor over capital, thus lowering the return on capital and the capital share. But the size of this long-term effect seems limited, and it is possible that it will be more than compensated by other forces tending in the opposite direction, such as the creation of increasingly sophisticated systems of financial intermediation and international competition for capital.”
So both scenarios are possible. Nothing is inevitable. It’s more a matter of determining which scenarios are more probably given prevailing trends and the historical record.
“more probable” not “more probably”
When the rate of return on capital exceeds the rate of growth of output and income, as it did in the nineteenth century and seems quite likely to do again in the twenty-first, capitalism automatically generates arbitrary and unsustainable inequalities that radically undermine the meritocratic values on which democratic societies are based.
This can only be accepted if the theoretical foundation on which Picketty bases his analysis is correct. If all capital is equivalent with a unified rate of return (if a dollar is a bond is a stock is a wrench is a factory is a computer) then Picketty’s conclusions may be valid.
If, however, capital must be disaggregated, which was revealed by by the capital debates in the 1950s and 1960s, then Picketty’s analysis is built on sand as it become very problematic, perhaps impossible to determine a rate of return. Picketty made the decision to presume the neoclassical production function correct because there was no other way to conduct the research. His conclusions, based as they are on a flawed theory are therefore suspect.
Piketty does not use the production function framework to predict the rate of return to capital in the 21st century. Piketty takes a squarely empiricist approach to this issue, not a rationalist one that attempts to derive the future from some mechanistic framework. The main discussion of the issue takes place from pages 350 to 361 of the book.
He assigns real output as a function of a capital/labor ratio. That is a production function and he does assign capital, whether physical or financial, common unit value based on his estimation of market valuation. He depends on that valuation to determine the capital/income ratio, again an aggregation of productive with nonproductive wealth. In fact Picketty doesn’t appear to understand the difference, he just assumes it is all the same.
Well sure, I guess if we just give up on the task of attempting to measure the amount of wealth people have, and what kind of return it produces, on the assumption that wealth comes in many different kinds and they are all incommensurable, then we can never know whether the rich are getting richer.
Seems like obscurantism to me. Things have market prices for a reason, and the market prices serve as a decent common measure of how highly people value them.
Hi, Dan,
This is just be a quibble, but as I understand it, the “Cambridge controversy”, as it might apply to Piketty, isn’t saying that different types of wealth are necessarily incommensurable. Rather, it’s saying that IF your measure of commensurability is market value, and since market value of capital depends on expected, real return–think bond prices–and since calculations of (realized) return depend upon original market value, the following statement is true. Any measure of capital value, return, or the ratio of them, upon which your favorite mathematical economic model depends, is not well defined. Rather, such measures employ circular reasoning.
Thus, the (total) real value of housing in 2006 was not its (total) market price at that time. A few house owners at that time could have realized their market price as a real value by selling just before the crash, but not in the aggregate, as no more than a few could have sold until a point was reached where every new seller would have reduced market prices. Of course, this then actually happened during the crash and aftermath.
This failure to have market price reflect real value is because “Bubble pricing” employs circular reasoning: “Everyone’s house will have increasing market value because everyone will continue to want to buy houses in the future. Everyone will continue to want to buy houses in the future because…everyone’s house will have increasing market value.”
But, as I see it, this critique of Piketty will have substantive impact only in proportion to the degree to which the market value of capital is, over time, in “bubble” (including negative bubbles) mode. It is probably the case, however, that over a reasonably long run, such positive and negative bubble effects would mostly cancel out.
And none of this means we can’t see the obvious fact that there is huge injustice in the distribution of income in capitalist political economies, because we don’t need a mathematical model to see that! But of course, we also didn’t really need Piketty to point it out…or, I guess, the fact that many DID need him to point it out reflects how hard it will be to change this (inevitable) feature of capitalism.
Excellent post, and comments as well. The is issue is complicated no doubt, but there is no denying the connection between wealth and power. As Wray pointed out the rich will find a loophole if you let them carve it out. That being said, I don’t think you can simply ignore the impact of wealth and appropriate taxation. I’ll show you an example here in Texas to explain why. Since my initial training/background was in education, I take a rather dim view of two-tiered application of government policy.
Seeing what was happening with our escalating property values (and CAD assessments), I decided to analyze how our CAD is applying the property tax code in terms of “equality and uniformity”. What I found was shocking disparity in application of assessments between middle class homes and more expensive homes.
Between 5 different neighborhoods I looked at 45 homes on the lower end and 45 homes on the upper end. For the 45 lower priced homes, the CAD issued preliminary assessments which totaled more than $16,000 over the combined sale prices. For the 45 high end homes, the CAD issued preliminary assessments that totaled over $3,4 million UNDER the combined actual sale prices!
http://aaronlayman.com/2014/05/crony-capitalism-at-the-fort-bend-central-appraisal-district/
Yet another example of how the middle class is getting taken to the woodshed
Piketty’s concept of capital is what the rest of us call net wealth or net marketable wealth. It includes what the classical economists called ‘capital’ but is far more. Housing accounts for almost half of marketable wealth in advanced countries, and government debt a considerable share. The book is about a fact–the wealth/income ratio tends to grow over time, and the distribution of that wealth comes increasingly to be concentrated in the top decile and top centile of the income distribution. Unless someone comes up with comparable data that refute these findings, we have what passes in economics for a hard fact, just as the constancy of the speed of light is a hard fact.
Economists trained in pure theory have a very hard time getting their minds around facts that don’t jive with their theory but are nevertheless facts. I could give any number of referee’s reports written by people whose priors are so strong no amount of evidence can alter their opinion. Piketty’s work is a massive challenge to that mindset. He does not offer an over-arching explanation of his results, but simply reports them with some plausible but not binding interpretation.
It is crucial to keep in mind that Piketty’s concept of wealth is essentially the purchasing power of that wealth, not its productivity. It’s of course true that if everyone tried to cash in that wealth at the same time, it wouldn’t be worth much, but such panics are rare and generally seem to be quickly repaired. I was surprised that rent-seeking and monoply doesn’t play as big a role in the mal-distribution ad I expected, but he is lookng at a long run of data and exploitation comes in many forms. It is conceivable that the condition r>g is a disequilibrium state that in principle will be asymptotically corrected, but ad Keyned quipped, in the long run wr are all dead, and the adymptote is a very long run. In the mean time those who profit have plenty of time to squirrel away their ill-gotten gains in tax-free municipal bonds that just keep growing and growing.
As I said, economists have a hard time getting around these things, because the modest operandi is to find facts that fit the theory rather than the other way round. I think this book signals the start of a big reversal in how economics will be done in the future.
a comment from the peanut gallery:
the “problem” with the regulatory approach is that it costs a lot of time, money and expertise to carry off coupled with persistent application. meanwhile, the rich are gaming the system and avoiding the regulations on one side, and using their ill-gotten wealth to fight politically for deregulation on the other.
“rich tax” appears easier, but is subject to the same forces (costs of implementation and finding those who creatively avoid paying, plus same political circus).
how about an “all of the above” approach?
“The simple question is: Why doesn’t r fall as returns necessarily diminish? Piketty’s mainstream answer is that new technologies, broadly defined, keep creating new profitable opportunities.”
This is Piketty’s main mistake. In fact, the reason r stays high is that the rentiers *demand* that r stay high, and so they *change the laws* in order to keep r high.
We’ve witnessed that repeatedly in recent years.
On the whole, the rest of Piketty is right.
You have to remember that when he says “capital” he means “wealth”, and when he says “return on capital”, he means “rents collectible by holders of wealth”. It all snaps into sharp focus.
Yep, he says it includes (at least) profits, rents, dividends, interest, royalties and capital gains.
“James Galbraith opened his review noting that capital meant power. If Piketty’s empirical analyses are right, and r has been persistently high, I’d suggest it reflects the power of the rich, not the natural forces of a free market economy. Higher taxes would help stymie the rich, but so would financial regulations, anti-trust campaigns, and public financing of politics that could minimize their power and privilege.”
It just seems to me we don’t need another economists equation (r>g) to explain the phenomenon clearly and simply – criminality, corruption, and the abnegation of the LOSS part of profit and loss for anybody who has more than a billion dollars.
It really is an amazing thing that the entire Federal government, when an economic catastrophe stuck, decided the treasure of the US was best used making damn sure no banker suffered. Astounding, really…..
Piketty’s text is certainly better than nothing, but it does not go far enough. Capitalism is an incredibly flexible ideology. It somehow forces its harshest critics to speak in its own language and ultimately hamstrings them. I would like to see more anthropologists and sociologists discussing reasonable economic models. Too many economists are corrupted by a discipline that is strictly associated with capitalism. Marxists/Socialist economists are no solution either. All the Marxists were/are Capitalists-in-waiting if you will. If alive today, I’m sure Marx would admit this. The only way we can break the incestuous dialogue on the concept of economy is to involve disciplines involved in studying human behavior, and social structures. There aren’t many more ways we can break down and understand Capital. It has shown us all its tricks, there are none left.
Rosario:
I’m curious what you think of Karl Polanyi and his not quite famous book from 1944, the second great book from that year, “The Great Transformation?” Polanyi was a social democrat/socialist who did not sound like a Marxist, did pioneering work as an anthropologist-economist, and influenced Sir M.I. Finley the great classical scholar, when their paths crossed at Columbia University shortly after WWII. I couldn’t help but think of Polanyi’s great work on the lives of the first fully industrialized mill workers in the Midlands, and what their lives had been like before as agriculture workers” when I read Piketty’s confirmation that indeed, the early years of the 19th century were terrible for wages, and they did not improve until the very last decades of the 19th century. Piketty flies high; Polanyi puts you painfully right at ground level for the awful spectacle of what he called the world’s first consciously constructed (by the classical economists of England) society wide “labor market.”
Or what you thought of the work of David Harvey, a Marxist scholar whose home “field” is geography but who is listed at the “Distinguished Professor of Anthropology” at the Graduate Center of CUNY? The Financial Times called his book “The Enigma of Capital” “Elegant,” but I doubt they’ll like his most recent one, ” Seventeen Contradictions and the End of Capitalism,” which has quite an “edge” to it. And I don’t know if the title was an answer to Ralph Nader’s “Seventeen Solutions…”
I do enjoy David Harvey. The Enigma of Capital is on my reading list for the future. I’ve admittedly only read a handful of his articles and viewed a few lectures, but he does fit the ideal model of an economist (at least as they should be). He is comfortable with a heterodox perspective on economics, which is refreshing, and he is not afraid to call BS on either side of the political spectrum. Ideally, economists should be well versed in many social science disciplines and he does a pretty good job. Thanks for the recommending Karl Polanyi. I’ll look into him.
You won’t be disappointed by Karl Polanyi’s masterpiece. Just to whet your appetite, he said no one, even the classical economists who “invented it,” could live with the “pure” free market. It was too ruthless, and as soon as they had shaped it in the early decades of the 19th century in England, the formal Parliamentary and Commission studies triggered the regulatory state to curb its worst excesses. And in Polanyi’s handling of what he calls the “double movement,” the push pull between capital and the forces most at the mercy of unregulated markets – including some businesses themselves, and workers, and farmers, you will read one of the fairest accounts of the tensions within the 19th century political economy that I can recall.
I guess I’m a little bit puzzled by the sentiment here. What do people think would be a “normal” return to capital in a free market system without the impact of abuse, political influence-buying, and other insidious forms of power?
If wealth just is power, then it follows that the power of the rich is one of the “natural forces” of a capitalist economy, not something that exists separately and outside those forces.
.
If that is the only solution, then a solution to Income Disparity is are truly screwed. It is almost Mission Impossible to have it uniformly adopted. We can’t even get war to be outlawed, never mind a functional world-wide tax on capital.
The administration of such a tax is just as mind-boggling as its adoption. (Of course, one benefit would be space-exploration and colonizing – just to escape the earth’s “3WT” (World-Wide Wealth Tax ;^)
And yet, the EU seems to be agreeing on a Tobin Tax, so that’s a small start. The US needs to play catch-up by putting upper-income tax rates back where they were before 1960.
Since most capital is labor-value [transferred from the actual laborer (wage-earner)], it should come as no great surprise the rate of return on capital should surpass the growth in wages [as they enjoy a near-reciprocal relationship].
Yves Smith said “As Piketty defines capital, the return on capital in one period becomes capital in the next time period by virtue of being capital-type assets (loans, stocks, investments in more production, etc.)”
I may be misunderstanding. Piketty says in Chapter 10, page 351,”For example, if g=1%, and r=5%, saving one-fifth of the capital from income (while consuming the other four-fifth) is enough to ensure that capital inherited from the previous generation grows at the same rate as the economy. If one saves more, because one’s fortune is large enough to live well while consuming less than one’s annual rent, then one’s fortune will increase more rapidly than the economy, and inequality of wealth will tend to increase even if one contributes no income from labor.”
“It was exciting to find a strong refutation of the long-held mainstream view that the share of capital and the share of labor in GDP were stable. Workers and investors would split the economy’s bounty according to some unknown law of equality. Now Piketty was saying not so. Historically capital did much better. This is darned important.
But to get there, Piketty ironically returns to conventional mainstream economics without quite telling us he shifted. To him, the invisible hand failed in labor markets, but works in capital markets. Piketty hypothesized that the reason r stayed high was that the “elasticity of substitution between capital and labor” favored capital. To oversimplify a bit, it made sense to invest more in capital than to hire more workers.” From above Madrick: parargraphs 8&9
And why did it make sense to invest more in capital than workers? Short answer for the sake of brevity, productivity. One of the hallmarks of capitalism is to reinvest profits in innovative means to produce more, with the same amount of labor. Labor wages can remain the same, the amount of people working and amount of hours in a day can remain the same. Investment in techniques and or equipment to allow those working to produce more allows you to sell more, make more profits and not share that increase in wealth with the laboring class. Innovations not only allow you to make more with the labor costs remaining constant, the innovations can also allow more, better quality and cheaper products which in turn crush competition or keep competition from entering a market. This of course boost profit taking, allowing for wages to remain constant. This is not the result of an iron law of economic determinism, it is history, or as theoreticians like to say, empirical. But what else shifts the balance from capital to labor?
As an individual worker, you can’t throw more of yourself at the problem of how to get more money, the way the capitalist can throw money at the problem of productivity and get good results. Labor only has 24 hours in a day and a human, like a horse, can only exert so much force of energy measured in BTUs. But continually improving equipment in agriculture or continually innovating management and machinery can exert many more BTUs of productive activity that millions of men and horse can’t hope to match. And furthermore, the dual outcome of crushing competition and keeping wages constant makes labor permanently relegated to an inferior position when the means of production are controlled by people who use your labor at constant rate to watch over an ever improving production juggernaut that becomes a engine of profit that grows year after year in its profit producing capacity. But your laboring power can not grow year after year by 5% or 2% or even 1% to earn you more money within the same 24 hour day. More than likely, as you age, your labor capacity diminishes while capital continually looks for way to not depend on labor and keep all of the profits from ongoing productivity gains.
While I am sure that there are worthwhile reasons to argue back forth in a serious manner so there is some validity to any policy guidance than can rest on this important research, Piketty is not reaching beyond the boundaries of his discipline when he comes up with a global tax on capital, however it may be operationally defined for model legislation to be used internationally. Of course, there are political reasons why the wealthy remain wealthy and pass this wealth down over generations. Sociology talks about social capital. They don’t mean stocks, bonds, luxury items or vast real estate holdings, but the social relationships which guard a person and their families from the shortcomings of the human condition. Grover Norquist is the spearhead the anti-tax ideology that is well funded by the wealthy from the entire spectrum of society’s wealthiest. The super-rich endow an immense counter-culture to the liberal institutions, think tanks and universities to propagate their belief in capitalism and their hostility to any government activity real or imagined, that is not in their interest. Piketty has a globally designed research data base to call upon to use as his unit of analysis, not just one country as if one nation existed economically divorced from transnational capital flows and trade. Of course we also have to talk about other dimensions of analysis, it is almost impossible not to think of them immediately after the economic ideas trail off in their conclusion. But Piketty can only conclude so much from his research. He is not a political scientist or policy expert but the very nature of his research demands that we think in those terms as we read him or read about him, or listen to him on youtube.
This is as much history, a tightly focused economic history, but one that may not demonstrate a causal connection between taxation and political equality. He does not have to do that. He has show the inequality. Does he have to show the mechanism of inequality in order for us to believe what our lying eyes see? There is over a hundred years of social science and human reason to lay bare the inequality. Taxation is an element of an over all set of many, many restraints that must be placed upon profit making market economies by the nations they operate within. The market has to serve the larger social order in which it exists or else the social order is bent to the will of the market, dominated by the wealthy. Just as the military has to be under the control of the civil order, the capitalist order has to be bridled and placed under the control of civil society.
The sad fact and menacing reality are that the U.S. military is not under the control of the civil order, but rather under that of the capitalist order. And it’s not just the military, but all other institutions of government that have been co-opted and corrupted. Fascism has advanced and is now unstoppable until the economy collapses or there’s another world war. Collapse may be our only hope of averting global extinction due to climate change. World war would bolster the economy but would hasten the extinction. In any case, we are all in grave danger.
One area where labor in theory does have some control over wages is in limiting the supply of labor available to capital. A lower supply of labor should mean the unit price of labor goes up. But this is only possible if capital is not allowed to obtain labor (or import it) from elsewhere.
You discuss the super-rich endow an immense counter-culture to the liberal institutions, think tanks and universities to propagate their belief in capitalism and their hostility to any government activity real or imagined, that is not in their interest. . I’m sorry but the super wealthy are not stupid enough to put all their eggs in one ideological basket like that. They also insidiously finance the liberal institutions and steer them towards oligarch friendly policies; for example the idea that to limit the supply of labor is racist.
How else but through oligarchic cooption of liberal institutions could we have gone from Cesar Chavez, Ted Abernathy, and Walter Mondale marching at the Mexican border to stop scab labor from being imported into the United States to one generation later the idea that limiting labor supply is racist?
VERY spirited discussion, all this.
I am of the opinion that Piketty is guesing because some data proves a point he wants to make and the risk of interpretation (of elements) is very great.
Frankly, as regards Wealth (which surely is income evolved into capital wealth), I prefer the work done by GW Domhoff published at his site Who Rules America?. Domhoff makes far less sweeping historical references to the evolution of capital.
Shall we all chew on that bone a bit before moving on to more pedantic matters? ;^)
In all honesty i think Marx had the general idea within reach, but got himself sidetracked by the ongoing debate about value. And so he got himself mired in the whole labor theory of value, because he had to demonstrate that value came from somewhere objectively.
The basic thing is that capitalism can’t escape that workers and consumers are one and the same. As such, trying to squeeze more out of the workers invariably undermines the very consumption that drives capitalism.