Is Piketty’s Data Reliable?

When Thomas Piketty’s Capital in the Twenty-First Century first appeared many economists demurred on the theory but heaped praise on the empirical work. “Even if none of Piketty’s theories stands up,” Larry Summers argued, his “deeply grounded” and “painstaking empirical research” was “a Nobel Prize-worthy contribution”.

Theory is easier to evaluate than empirical work, however, and Phillip Magness and Robert Murphy were among the few authors to actually take a close look at Piketty’s data and they came to a different conclusion:

We find evidence of pervasive errors of historical fact, opaque methodological choices, and the cherry-picking of sources to construct favorable patterns from ambiguous data.

Magness and Murphy, however, could be dismissed as economic history outsiders with an ax to grind. Moreover, their paper was published in an obscure libertarian-oriented journal. (Chris Giles and Ferdinando Giugliano writing in the FT also pointed to errors but they could be dismissed as journalists.) The Magness and Murphy conclusions, however, have now been verified (and then some) by a respected figure in economic history, Richard Sutch.

I have never read an abstract quite like the one to Sutch’s paper, The One-Percent across Two Centuries: A Replication of Thomas Piketty’s Data on the Distribution of Wealth for the United States (earlier wp version):

This exercise reproduces and assesses the historical time series on the top shares of the wealth distribution for the United States presented by Thomas Piketty in Capital in
the Twenty-First Century….Here I examine Piketty’s US data for the period 1810 to 2010 for the top 10 percent and the top 1 percent of the wealth distribution. I conclude that Piketty’s data for the wealth share of the top 10 percent for the period 1870 to 1970 are unreliable.
The values he reported are manufactured from the observations for the top 1 percent inflated by a constant 36 percentage points. Piketty’s data for the top 1 percent of the distribution for the nineteenth century (1810–1910) are also unreliable. They are based
on a single mid-century observation that provides no guidance about the antebellum trend and only tenuous information about the trend in inequality during the Gilded Age. The values Piketty reported for the twentieth century (1910–2010) are based on more
solid ground, but have the disadvantage of muting the marked rise of inequality during the Roaring Twenties and the decline associated with the Great Depression. This article offers an alternative picture of the trend in inequality based on newly available data and a reanalysis of the 1870 Census of Wealth. This article does not question Piketty’s integrity.

You know it’s bad when a disclaimer like that is necessary. In the body, Sutch is even stronger. He concludes:

Very little of value can be salvaged from Piketty’s treatment of data from the nineteenth century. The user is provided with no reliable information on the antebellum trends in the wealth share and is even left uncertain about the trend for the top 10 percent during
the Gilded Age (1870–1916). This is noteworthy because Piketty spends the bulk of his attention devoted to America discussing the nineteenth-century trends (Piketty 2014: 347–50).

The heavily manipulated twentieth-century data for the top 1 percent share, the lack of empirical support for the top 10 percent share, the lack of clarity about the procedures used to harmonize and average the data, the insufficient documentation, and the spreadsheet errors are more than annoying. Together they create a misleading picture of the dynamics of wealth inequality. They obliterate the intradecade movements essential to an understanding of the impact of political and financial-market shocks on inequality. Piketty’s estimates offer no help to those who wish to understand the impact of inequality on “the way economic, social, and political actors view what is just and what is not” (Piketty 2014: 20).

One of the reasons Piketty’s book received such acclaim is that it fed into concerns about rising inequality and it’s important to note that Sutch is not claiming that inequality hasn’t risen. Indeed, in some cases, Sutch argues that it has risen more than Piketty claims. Sutch is rather a journeyman of economic history upset not about Piketty’s conclusions but about the methods Piketty used to reach those conclusions.


'You know it’s bad when a disclaimer like that is necessary.'

And worse when a disclaimer like that is not included, and words like these are used - ' coding errors, selective exclusion of available data, and unconventional weighting of summary statistics.'

Besides, with the following being true, there is no reason to accuse anyone of lying - 'When properly calculated, the average real GDP growth rate for countries carrying a public-debt-to-GDP ratio of over 90 percent is actually 2.2 percent, not −0.1 percent as published in Reinhart and Rogoff.'

In other news, some other guy was wrong one time

Or to put it another way, the last actual major economics issue concerning accuracy and reliability of data was this one, as the actual conclusions of the co-authors turns out to be wrong. It received three posts in one day from Prof. Cowen -

Some economists seem to have considered it quite important. And it was not one guy - the authors of that work are a married pair.

Uh, pretty sure they are not married...

This should be the classic example of whataboutism.

Not that I'm really against whataboutism, but its immediate derailment of the topic at hand is masterful.

Actually, it was in regards to the quote, which was ‘You know it’s bad when a disclaimer like that is necessary.’ and which started my comment.

Because in the case of Reinhart and Rogoff, it seems as if a number of the people were happy to criticize their integrity, and were able to buttress their case quite strongly . In other words, you may know its bad when someone includes such a disclaimer, but it is clearly worse when such a disclaimer is considered unsupported by the facts at hand.

In essence, what about when your critics, pointing out incorrect and ignored data that disproves your basic assertions, don't feel a need to include a disclaimer concerning your integrity?

First of all, so what? What is your point? Just to distract and confuse?

Second of all, no it did not change their conclusion and no their integrity really was not in question.

The minions of the malefactors of great wealth have turned against Monsieur Piketty.

Noah Smith (in 2014): "In short, Tyler Cowen's portion of Marginal Revolution has become your one-stop-shop for anti-Piketty links and analysis. This is interesting, because Cowen himself recently wrote a book called Average is Over, which makes predictions somewhat similar to those made by Piketty.... There is a difference, of course. Piketty forecasts that wealth inequality will go up because of an increase in capital's share of income, while Cowen forecasts that wealth inequality will go up because of increased inequality in labor income. But the basic future foretold by the two is the same.... So why has Tyler turned into an anti-Piketty crusader? Well, Piketty is a popular topic, and his thesis is encountering a huge amount of skepticism, so there is demand out there for a one-stop anti-Piketty shop. But it also seems possible that Piketty has deeply frightened economists who thought that concern over inequality was a thing of the past, and that laissez-faire had basically won the battle of ideas. Piketty's immense popularity might seem, to these economists, to threaten to drag us back into a dark age when radical wealth redistribution was taken seriously, not only by large segments of the public, but by a number of prominent economists as well. Piketty might seem like the vanguard of an onrushing wave of socialist thought that could succeed in turning back the tide of neoliberalism that had been advancing for at least 40 years. So Cowen - and the numerous anti-Piketty writers he links to - may simply be scared of Piketty and what he represents."

Ad hominem attacks for the loss.

Yeah really. Are there actually people who find attacks on others' presumed motives, in lieu of critiques of their data and analyses, interesting or persuasive?

I do. Character is fate, we say.

I find them very interesting. It makes me think the person they are attacking is right.

I find Ad Hominem a very useful heuristic for which side in a debate is mostly likely wrong.

Sorry what is this loss you're referring to?

I think the comment is a lame, weak, losing argument. Basically, it sucks.

Thanks. I agree.

"Ad hominem attacks for the loss."

It's clearly an ad hominem attack, but is it truly a loss? The useful idiots will quote it and others will read it along with large piles of other such drivel. That's the point. Not to persuade critical thinkers with a logical argument, but to buttress true believers with an endless pile of shallow verbiage.

Smith's attack isn't just ad hominem, it's Bulverism!

Prof. Cowen's focus on Piketty has been a running joke here for years. (For the pedantic, I am fully aware that this post is from Prof. Tabarrok.)

Just type 'Piketty' into the search box upper right, and scroll through the results.

This isn't even the first time that Piketty's accuracy has been questioned here, either -

Prof. Cowen’s prior_test3's focus on Piketty GMU has been a running joke here for years.

Prof. Cowen’s prior_test3’s focus on Piketty GMU has been a running joke here for years.

Well, my focus on GMU comes from having studied and worked there. It is not really all that mysterious.

Well then I guess a short-term focus on an extremely widely discussed book should not be a surprise either. Less of a surprise, maybe, considering it did not come out 40 years ago.

Classic Prior - strawman the argument, then switch to an ad hominem tu quoque of the other side. It must have worked at his High School debating soc, so unfortunately he's internalised the behaviour.

Nope, never did high school debate. And did you actually look at the links? Piketty's work is not really all that important or seminal, to be honest, but for a while there, it felt like all Piketty all the time.

And ... it begins to look like Cowen and Tabarrok were definitely right to question Piketty, no?

Sure, apart from the fact that according to Prof. Tabarrok, 'Indeed, in some cases, Sutch argues that it has risen more than Piketty claims.'

In other words, a more accurate methodology would make Piketty's basic point more strongly.

Luckily, you picked up on the proper narrative.

"In other words, a more accurate methodology would make Piketty’s basic point more strongly."

Wrong. Reading comprehension fail

And an impressively incompetent one, given that this is an Alex post. It's doubly irrelevant

In hindsight, in this paragraph Noah Smith explains pretty convincingly why HE bought Piketty's thesis without any skepticism.

Ha! Agreed, Xavier

Smith finds himself out of his depth pretty regularly ime.


However he fits into the media zeitgeist perfectly.

We need a purge.

Yes, and of course the flipside to rayward's comment is that Piketty could have been caught using a random number generator and lots people would brush it off because they like the argument he's making. They want it to be true, because it helps justify stuff they want to do anyway, like, say, redistribute property from the productive to the unproductive.

Ironically, I wanted Piketty's data, if not his theory, to be true.

I wanted a least one honest economist on the Marxist side of the argument. Sigh.

Robin, not Batman

Noah Smith in 2015:

Without peer-review, there is no such thing as science. And in 2017, without replication... Don't underestimate the amount of flawed work out there...

Think of it this way.

The author's criticism doesn't make sense unless you believe in abrupt discontinuities between periods where he claims Pickety is wrong, and where he concedes Piketty is probably correct.

Think of it as driving a car. The car driver would have really had to accelerate fast to get from one period to the next (where the author claims Pikketty was correct), or between period where he did not dispute Pikkety's observations. When you have a time series like this where there are places where someone picks or pokes at a single period within it, but concedes that parts are probably correct, likely correct, or mildly disputed, you have to ask: what is more probable: a car starting at 50, stopping from a speed of 120, accelerating to 150 , and clocking a speed at 180 when you are more confident of your observation.

We are living in world of big data--where data from multiple sources will be combined to be accessed by researchers--so its not surprising that Pikkety's data will be improved. But, it would be surprising to find that there were lurches, screeches, abrupt braking, and quick and sudden accelerations to an end point (or points in between) where there is greater confidence in the data.

And yet, that is exactly what is observed in the good data that we have, which is the point made about the '20s and '30s. And in fact, driving a car is a terrible example because people stop all the time, right?

Actually, no. The 20's and thirty's are confirmed by the Pikketty data. The car is the good example for continuous functions.

You wouldn't know a continuous function if it bit you.

"The values Piketty reported for the twentieth century (1910–2010) are based on more solid ground, but have the disadvantage of muting the marked rise of inequality during the Roaring Twenties and the decline associated with the Great Depression."

Cars maintain the same speed for long periods of time and then the speed suddenly changes very rapidly before stabilizing again. It might as well be a step function.

" I conclude that Piketty’s data for the wealth share of the top 10 percent for the period 1870 to 1970 are unreliable. "

"Piketty’s data for the top 1 percent of the distribution for the nineteenth century (1810–1910) are also unreliable."

I think a rational person would assume that if this criticism has merit, it undermines the credibility of Piketty's conclusion.

How do you connect the points between the data that he claims is unreliable to the data that he doesn't challenge? Notice also that in one case he quotes the top 1% data and the other the top 10% data, which means the critic changes the frame of analysis as well. Or, didn't you notice.

I think a rational person would notice discontinuities and have difficulty explaining how discontinuities exist when there are times he does not dispute Pikketty data in intervening periods.

"How do you connect the points between the data that he claims is unreliable to the data that he doesn’t challenge?"

LOL, why would he have to? You don't have to prove all parts of an opponents argument wrong, just a significant part. If a significant part of the data is incorrect it casts doubts on the conclusion.

"Notice also that in one case he quotes the top 1% data and the other the top 10% data, which means the critic changes the frame of analysis as well. Or, didn’t you notice."

I notice how Sutch points out that in 1 part of Picketty's data he gets the top 1% wrong and that in another part he gets the 10% data wrong. In fact I quoted the specific lines.

"I think a rational person would notice discontinuities and have difficulty explaining how discontinuities exist when there are times he does not dispute Pikketty data in intervening periods."

This comment makes no logical sense. If I point out an error in steps 3 and 5 of a mathematical 5 step solution, would a rational person conclude the answer is still probably correct because I found no errors in steps 1, 2 & 4?

Sorry JW, but but Pikketty presented more data, data that was not disputed, and although there is data that is, the data offering by the disputer doesn't make sense if you view it as a continuous, rather than abrupt continuous function. Second, if the challenger claims that the top .1% is wrong for one period, but says nothing about the top 10% for the same period; and in another sentence says later the top 10%--does it strike you as strange you would change the baseline (i.e., shift from top .1% to top 10%), and also have no discussion of variance around an estimate. Your final comment does not make sense: we are not taking about a mathematical solution, but rather sampling over a large time; you reframed the sampling observation as a mathematical formula operation which it isn't.

Bottom line: I am not impressed by the attack; and if you look at the picking at Pikketty at data point over a continuous period where there is no dispute you have to question the challenge and explain it. Every sample has variance, but when you havre a large collection of data over a long period it all adds up to make a strong case.

" the data offering by the disputer doesn’t make sense if you view it as a continuous, rather than abrupt continuous function."

This makes no sense whatsoever.

"but rather sampling over a large time;"

This is not sampling over time. The data from different periods came from different sources. This is clear from the abstract alone and is detailed in the paper. Which you quite obviously didn't bother to read.

"Piketty used two basic sources to estimate the distribution of wealth in the twentieth century"

"Piketty admits, in the Technical Appendix to his book, that “huge uncertainties exist on [his
nineteenth century] estimates” [Piketty 2014b: 58]. For the United States in the mid-nineteenth
century Piketty has only one real data point, an estimate for 1870 of 32 percent (Figure 2). This
is the proportion of the national wealth held by the top one percent based on the census of wealth
conducted at the time of the 1870 Census of Population [U.S. Census Office 1870]."century and up to 2010 [Piketty 2014: 347].

” the data offering by the disputer doesn’t make sense if you view it as a continuous, rather than abrupt continuous function.”

Actually this makes a little sense, now that I understand your mistake. The is not a continuous data set. It's multiple sets of data covering different time ranges. However, I'm baffled how you could have ever thought that it's continuous. It's crystal clear that Picketty didn't use a continuous set of data for the US.

JWatts, It's an even stronger argument that he uses different data sets over a range all pointing in the same direction.

"JWatts, It’s an even stronger argument that he uses different data sets over a range all pointing in the same direction."

Bill, you should really learn to stop commenting about a subject that you clearly know nothing about and you can't be bothered to read even the abstract written above. The data doesn't all point in the same direction. Hell, that's a major part of Picketty's point. According to Picketty (and most everyone), Inequality was high in the US during 1920's & 30's, collapsed at the start of WW2, remained low for 40 years and started rising during the 1980's.

Here's are some simple charts, (I don't think anyone can get confused looking at them):

Why is that implausible? Evolutionary theory doesn't assume constant gradual change either, but "punctuated equilibrium". Why would economics be different?

Which is more plausble--multiple abrupt discontinuities or continuities with reference points that are not disputed. Remember, we are talking about rankings, changes in relative wealth positions across a large population, not some new species born of a recombinant DNA which produced a bird with a colored beak.

It's not that hard to imagine that events like the industrial revolution, the Internet, the invention of the steam engine might have produced abrupt changes in wealth distribution. The Internet is probably the biggest driver of the current increase. People are becoming billionaires practically overnight.

Hazel, You understand what I am saying. If you want me to do so, I will lay out a series of numbers--beginning, middle, and ending points--which are not disputed; identify numbers which are disputed in intervening stages--and ask you to draw a conclusion.

In your description of events, you forgot to include sunspots.

To be clear, Piketty's explanation for rising inequality (captured in the expression r>g) isn't persuasive to me, but to deny rising inequality is to deny the existence of air. I've observed that excessive inequality is self-correcting, absent intervention by governments and central banks. The proof is the precipitous drop in inequality following the 1929 financial crisis (no intervention) and the failure of correction following the 2007-08 financial crisis (intervention). As to the latter, inequality dropped somewhat but has since risen back to the pre-crisis level. Piketty, being European (French), identifies the destruction of capital as the correction for inequality: two world wars conducted mostly in Europe destroyed much capital, capital that was owned by the wealthy, thus causing a precipitous drop in inequality in Europe. America also experienced a destruction in capital, not a physical destruction as in Europe but a destruction in the value of capital (as the result of the 1929 financial crisis and the Great Depression). Why a high level of inequality would trigger a financial crisis is an interesting topic (inequality in 1929 was about the same as in 2007), one that some economists refuse to consider (because even considering the topic would be an acceptance of the connection between inequality and financial and economic instability). I should point out that rising inequality isn't limited to developed countries such as America, but is being experienced in developing countries too, including in China (which has a higher level of inequality than America). While some economists would refer to view a high level of inequality as a potential social problem (the mob with pitchforks and torches descending on the Hamptons!), I view it as a potential economic problem (financial and economic instability). Of course, financial and economic instability can produce its own social instability along with the populist uprisings and they trigger and the demagogues who exploit it.

I'd encourage you to check out the following papers by Bricker et al (2016):

and Auten and Splinter (2017):

Both present detailed empirical work on SCF (Bricker) & tax data-derived (Auten & Splinter) estimates of inequality in recent years. Both conclude that the posited rise of inequality is substantially less than that claimed by Piketty (Bricker), or barely evident at all (Auten and Splinter). Of all the recent empirical work we've seen on inequality, those coming from some combination of the Piketty-Saez-Zucman team are the only ones showing an extravagant rise since 1980. Everyone else shows a more subdued rise, or little rise at all.

to deny rising inequality is to deny the existence of air

I'm not really sure about that. We're definitely less unequal today than in the 19th century.
Maybe even less unequal than in the 1950s, where there was still plenty of "old money" and class stratification around.

Not looking at data, do I feel more unequal today than in 1990? I don't think so. Do you feel like there's more class stratification?

Without data, I don't think anyone would feel they're being oppressed by Bill Gates' wealth.

Imagine if we didn't have an income tax, and surveys about income were banned.

I really doubt people would be outraged by "inequality." Its only when they read media stories about how they should be outraged.

Its especially interesting that most people seem to accept that movie stars like say, Adam Sandler, or athletes, can make millions and people just accept that as perfectly normal. But if a bank president makes millions, suddenly they're angry.*

My pet theory is that in their mental model, they know they couldn't be that funny or beautiful or that athletic, but running a bank from an office isn't far from their job (they imagine...they really have no idea of the hours, stress, responsibility, etc.) so they balk at that.) Just like most workers and socialists have zero clue how much money is at risk when someone opens a factory, and how much work is done by the suits so that the workers can clock off at 5 and just drink beer and relax.

*I base this off reddit threads. You'll find no one rails against capitalism when its Derek Jeter or Adam Sandler's salary. Its only when its a job they could imagine themselves conceivably doing. I suspect most people also have no idea how much capital is at risk in a business vs. their W-2 job.

Its especially interesting that most people seem to accept that movie stars like say, Adam Sandler, or athletes, can make millions and people just accept that as perfectly normal. But if a bank president makes millions, suddenly they’re angry.*

The compensation granted Sandler etc is an arms-length transaction. Not so the CEO. Peter Drucker, hardly a populist sentimentalist, suggested CEO salaries should not exceed a certain ratio to ordinary compensation.

In 1980, the most handsomely compensated CEO in the United States was a man named Robert Charpie of the Cabot Corporation, who was paid about $3.3 million a year. At the time, mean compensation per worker in private industry was about $16,000 per year, so Mr. Charpie received 200x the mean that year. As we speak, a sum 200x mean compensation per worker in private industry in this country would amount to $16,000,000 a year. There are, as we speak, 171 CEOs receiving in excess of $16,000,000 per year.

The size of the employed workforce in private industry grew by 58% between 1980 and 2015. We do not have 1 or 2 CEOs earning north 200x the mean of employed persons, we have 171. Marginal tax rates are a good deal lower than they were in 1980 as well.

They're very free with their stockholders money. Oink oink.

I should note that the Augustus Gloops consist of high-ed CEOs and some among their camarilla. There are 223,000 corporate CEOs in the United States. Their median cash compensation is $181,000 per year.

Why on earth should we take seriously the notion that CEOs should never earn more that 200x the mean of employed persons?

171 is still a pretty small number, considering there are over 223,000 corporate CEOs in America.
So, even if we take the 200x number seriously, we went from an infintesimal number of corporate CEOs making more than 200x mean, to a tiny fraction of corporate CEOs making more than 200x mean.

In the larger scheme of things, is the average person going to notice the difference?

The shifting CEO-worker pay ratio is certainly interesting on first sight.

BUT.I worry about :

Has mean firm size changed? (Larger firms must have a greater top-bottom split)
Has firm size variance changed? (Fewer medium sized firms relative to small and big ones?)
Does whole-package compensation effect things? (more pensions and healthcare bundles?)
Do payroll taxes effect things? Are worker more expensive, relatively, to employ?
Is it just CEO's taking larger splits? Do we see the effect across boards?
What are the total ratios of CEO's and workers?
Is there more demand for CEO's in the modern era?
Can some of the change be explained by rising cost of high value human capital? IQ is growing more expensive.
Is there a broader trend going back 100 years?

Global inequality has gone down as global wealth and capital has gone up.

Bruh you’re harshing the narrative with your alternative facts.

Sanders 2020!

Next they're going to tell me that Mathew, Mark, Luke, and John misrepresented first-century Judea and Jesus' life in the testament. No thanks. I think I'll decide what facts I'll accept.

Here's an easy exercise: download Jesper Roine and Daniel Waldenstrom's painstakingly collected data on the Swedish wealth and income distributions (, which Piketty cites as his sole source for his Swedish data. Then replicate Piketty's work. You'll find (1) an error in switching between two data sources in 1920 and some very questionable choices to drop or shift the years of late-20th and early-21st century data. The source also ends in 2006, but Piketty continues his graph to the "2010s" with no source given.

This is an easy exercise that an undergraduate could do. To my knowledge, me and Malin were the only ones who did it.

Swedish language writeup:

I've defended Picketty in the past for the odd coding error - inevitable in a large work. But the errors and selection artefacts and opaque coding have rapidly added up.

Excusable mistakes --> Incompetent mistakes --> unconscious bias --> conscious bias / fraud

Currently we're somewhere about the third step....

Let's hope that Piketty et al. are incorrect in their data, and that inequality isn't nearly as high as they have recorded, because if they are correct in their data, we likely can expect more financial crises and economic instability ahead. However, even if Piketty et al. are incorrect, if the perception is that they are correct, we likely face more financial crises and economic instability (perception is reality). I view attacks on Piketty et al. as an effort to change perceptions about inequality and thereby avoid the likely consequences of the perception of a high level of inequality (which includes the mismatch of the supply of capital and investment in productive capital). It's all very Straussian.

A basic reality is that economics is an adversarial system. Studies and counterstudies are produced because of course they are. Everyone knows their tribe.

That said, I think most try to be honest. Most will acknowledge a solid challenge.

In the case of inequality, we know it's there. We know the worldwide reduction is good (the global Gini coefficient has been falling since 2000). The harder questions are contested.

I don't even know if good is better. If the poor get twice as rich and the rich get 3x as rich is that bad or good. You have to hate the poor to say it's bad, but you do get a lot of that.

I don’t even know if reduction is better.....

We do have parallel data, that absolute poverty is decreasing, lifespans increasing. All the Hans Rosling stuff points in the right direction.

It's true that inequality between developed countries and developing countries is falling, but inequality within developing countries is rising (just as inequality is rising in developed countries). That's a distinction that even the always objective Cowen sometimes fails to make.

I don't always get what Tyler is thinking, but I assume he is trying to go somewhere positive.

For yuks though, let's check with Niskanen ...

Hmm. Looks legit.

Many academics and journaists who write about Japan have argued over the past ten years that inequality is becoming greater and greater. That has been true of income but not of the post tax-transfer gini coeffecient, which was was .33 in 2015 and increased slightly around 2003, but essentially no change for ten years. (It did increase in the 1990s) Another myth is that Japan used to be much more equal but again in post tax/transfer terms, the gini ceffecient was lower in the mid 1980s and early 1990s but higher than today in the 1970s.

Here are other countries post tax/transfer gini coeffcients in 2004 and 2014/15, OECD data:

Belgium: .29/ .27; Canada: .32/.31; Finland: .26/.26; France: .29/.30; Germany: .27/.29; Hungary .30/.29; Italy: .31/.33; South Korea: .30/.30; Norway: .29/.27; Poland: .38/.30; Spain: .32/.34 Sweden: .23/.27; UK: .36/.36, U.S.: .36/.39 [but the U.S. was at .38 in 2005 and .39 in 2015]

Sweden's post transfer inequality has increased the most while Italy, Spain and Germany have increased somewhat but either no change or almost no change in France, Japan, S Korea, Canada, the UK and the US over the past ten years.

I see no reason that greater inequality would make a financial crises more likely, the data set is too small to draw such a conclusion from looking at the data.

>One of the reasons Piketty’s book received such acclaim is that it fed into ...

... the fact that 98% of economists yearn for more statism.

Kind of impolite to say, but Niskanen might beat Mercatus because the former doesn't have that rigidity.

Also, it is ridiculous this fixation on income distribution with the average income of Western countries. The difference means that instead of a Sub-zero double-door refrigerator of the people in the first quintile, the lower income people have a normal sized whirlpool. It is not that they starve if it doesn’t rain in a season. Plus, the increase in real gdp if quality was properly assessed would make the issue of distribution even more meaningless. As I stated already in this forum, the Whirlpool of the lower quintile is very likely better than the top of the line subzero of only 15 years ago. You can make the case that for the majority of consumption categories, a working class American live better and longer than Ford or Rockefeller.

Let them eat refrigerators!

They can can eat bread, if they prefer. At the beginning of historical series, in 1914, the price of a bushel of wheat was around 1$. The price now is 4.3$. All these are nominal prices. Considering that the dollar lost 96% of their value since the introduction of the Fed, the real price of wheat went down by 6 times since 1914.

"Indeed, in some cases, Sutch argues that it [inequality] has risen more than Piketty claim"

So where does THAT leave us?

"Sutch is rather a journeyman of economic history ...": it's economic history - surely anyone doing good work is likely to be a journeyman? If you want to be a star do molecular biology.

Soon, Americans will learn that, by supporting ever increasing inequality, they actually sold their children's future to Mammon. Then, Rachel will weep for her children,
refuse to be comforted for her children,
because they are not. But it will be too lare because their silver will have become dross.

What you call supporting inequality, is actually NOT supporting pillage and robbery.

Nobody want to tax the poorest quintile to give to the richest quintile. At the very extreme, in the case of anarchists like me, the call is simply to leave everybody alone eliminating taxes altogether. Not even that is supporting inequality, it is simply let the voluntary market place allocate the resources.

It is allowing malefactors of great wealth to profit from their brothers and sisters' misery.

Love this hybrid rhetorics between Lenin and Saint Francis.

Let me understand again how a free-market enterpreneur can profit of his sisters and brothers misery. Does he put a gun to their temple forcing them to buy his products or services? Does he corrupt some parassite public servant in order to obtain a legal monopoly? Does he regularly put a bomb in his competitors' premises to keep them from offering a better deal to consumers? Does he force them to work for him against their will?

Seriously, you should consider to change your pen-name. Even Captain America in the Winter Soldier has more ethical doubts than you.

Actually, famous American president Franklin Delano Roosevelt warned America about the malefactors of great wealth, moneychangers who have fled from their high seats in the temple of our civilization.
Malefactors of great wealth control the economy and outsource American jobs immiseraring the American worker, who loses everything he has.

Any analysis that impacts a politically charged topic, and doesn't represent a clear declaration against interest, is suspect. The non-specialist can seldom reasonably assess the work.

It seems a strong argument for higher standards of replication, but the incentives for that appear weak.

Perhaps a bounty system? There is some history of software companies paying a bounty for defect reports on their products.

Michael Bellesiles tried the admission-against-interest tack by telling the Chronicle of Higher Education he was a Republican.

The non-specialist can seldom reasonably assess the work.

Interestingly, the data used in a technical analysis can often provide a window for non-specialists to legitimately challenge an author's analytical findings. That's something that we verified during our Examples of Junk Science Series, where at one of the examples we presented (Falsifiability Fail), would not have required any special knowledge of econometric methods to invalidate the author's findings. In this case, simply asking and answering the question "Does this data exclude information that might invalidate the author's findings?" would have been sufficient.

It seems a strong argument for higher standards of replication, but the incentives for that appear weak.

There has been some discussion about replacing or supplementing p-values with a replicability index: P-rep, which would represent the probability that the findings will be replicated. Other proposals have suggested increasing the bar for the p-values themselves (changing the typical threshold to assess significance from .05 to .005), but that would not of itself address the replication issue.

Perhaps a bounty system?

I like the idea, but rather than say paying cash per replication, it might be better for journals to provide full publication credit for those doing replication work, up to a set number of successful/unsuccessful outcomes per originally published paper. My guess is that most of the people who would be engaged in doing that work would benefit more from being able to boost the count of their published works than they would from a bounty.

I like both ideas. We need to move from a culture of novelty in science to a culture of assurance. More credit and reward for the replicators!

Replications / failed replications should be of decreasing value as time goes on as evidence amasses, but the first few attempts should be nearly equal credit with the discoverers.

As with all books that make it into headlines (and with most books in general), my approach to Piketty was: I will wait for ten years after the publication and if people still think it's relevant, I will read it. It's starting to look that this is not a book I have to read in six years.

+1. An excellent filter I adopt myself.

I do the same kind of thing for breaking news stories.

If I wait a good while, the details will be clearer and I can get the story in much less time. If I've forgotten to read up on the story after a couple of weeks, then it probably wasn't that important of an event after all.

You mean someone opened the Black Box and inside was some lint, a rusted penny stuck to the bottom with the residue of soda pop, and an antique piece of chewing gum which had once been in the mouth of Michael Bellesisles?

Criticism of Piketty has evolved over time (the three years since the publication of his book). At the outset there was the charge that he's a socialist and, therefore can't be trusted - and he's French! Unfortunately, Piketty went on the speaking tour and proved himself to be very likeable, always smiling and always polite - and he speaks a very clear and pleasant English in only the way the French do. So critics moved on to attacking Piketty's thesis, r>g. That proved to be counterproductive since it implicitly acknowledged that inequality is rising but simply disputed the cause. The final refuge came in the form of attacking Piketty's data. Sure, Piketty is a great guy (even if he's French!) but he just got his data wrong, an innocent mistake. This approach had the benefit of confirming everyone's skepticism about economists and their data. If violence is the last refuge of the incompetent, then the last refuge of economists is attacking other economists' data - economists manning the circular firing squad.

This is stupid even by your low standards.

He started this thread stupid, and damn it, he's going to finish it that way!

This is smart even by your high standards.

I was thinking the same, Art. Bellesiles redux.

We're not questioning his integrity. He had good intentions. That's the important thing. So what if his work product is shoddy?

Exactly. Good Intentions. Good Intentions. Like the Communists. Good Intentions. Good Intentions. 50 million dead. Good Intentions. Good Intentions. Omelette/Eggs. Good Intentions. Good Intentions. Good Intentions. Good Intentions. Good Intentions. 50 million dead. Good Intentions. Good Intentions. Good Intentions. Good Intentions.

Gosh, it's so hard to keep track of what is relevant when judging the value of a peer-reviewed paper. Let's see, there is whether the author has a PhD, who their thesis advisor was, who they post-doc'ed for, which school awarded them their degree, where they are employed now, what awards they have received, ..., and now Alex claims that we need to understand the "bias" of the journal. I've forgotten what he got his PhD in, was it English Lit. or Women's Studies? I guess the world is just too complicated to expect a paper to stand on its own, and obviously AT doesn't believe that is exactly the purpose of publishing in a peer-reviewed journal. I didn't read Piketty's book, but iirc at the time it was available in the US there were lots of people making the same criticisms (in broad strokes) about it.

I assume the mention there was to alert people that the peer review might not be that robust in this case.

Peer review is now broken. If it ever wasn't.

Inequality is rising, but we're at the stage where world hunger is a distribution problem. We can give out packets of fortified food containing 2800 calories, more than the minimum daily requirement. Then we hear the complaints the poor are fat. With the advancements in solar technology, etc., the poor will be able to sustain and grow wealthier. We've come a long way.

Oh, *that* Richard Sutch. For the non-finance jocks in the crowd, here's what he did 51 years ago:

Innovations in Interest Rate Policy
Franco Modigliani and Richard Sutch
The American Economic Review
Vol. 56, No. 1/2 (Mar. 1, 1966), pp. 178-197

This article is the foundation of modern fixed-income research. Piketty is toast.

FWIW, Piketty, Saez, and Zucman have revised their own data. They still show high inequality, but now they show middle-class incomes rising by 40% from 1980-2014. I'm really surprised that no one is talking about this yet.

Well, the conclusion that the top levels have grown has not changed, nor has the lower half been better off:

Here is from the abstract:

"We estimate the distribu- tion of both pre-tax and post-tax income, making it possible to provide a comprehensive view of how government redistribution affects inequality. Average pre-tax real national income per adult has increased 60% from 1980 to 2014, but we find that it has stagnated for the bottom 50% of the distribution at about $16,000 a year. The pre-tax income of the middle class—adults between the median and the 90th percentile—has grown 40% since 1980, faster than what tax and survey data suggest, due in particular to the rise of tax- exempt fringe benefits. Income has boomed at the top. The upsurge of top incomes was first a labor income phenomenon but has mostly been a capital income phenomenon since 2000. The government has offset only a small fraction of the increase in inequality. The reduction of the gender gap in earnings has mitigated the increase in inequality among adults, but the share of women falls steeply as one moves up the labor income distribution, and is only 11% in the top 0.1% in 2014. JEL Codes: E01, H2, H5, J3."

Well, that's interesting.

Comments for this post are closed