by Tyler Cowen
on May 29, 2014 at 12:28 pm
in Books, Data Source, Economics, History |
There are 4,400 words here, mostly on the FT kerfluffle, and Neil Irwin summarizes it here: “The short version: He doesn’t give an inch.”
As usual, my side correctly identifies the real problem, while the other side erects sideshows and tries to distract from the debate.
Wait, what are we talking about? Oh, never mind.
I suggest that we all chip in to cover all expenses necessary to lock Giles and Piketty in the same room alongside their datasets, Excel and an unlimited supply of fine ale. No contact with the outside world allowed. We just wait for the joint statement (or disjoint statements) and THEN we start writing 10,000 posts on the conclusion they’ve reached.
We are bombarded non-stop by information and, seriously, in this precise case it’s far from clear this is welfare improving.
“No contact with the outside world allowed.”
Color me as one of those who thinks it’s just as useful to see the back and forth on the way to the (dis)joint statement as it is to see the final statement itself.
Yeah… you may be right… I guess I’m one of those who have a hard time processing all this information…
That’s actually why I like to follow as these kinds of debates unfold. I’m not great at simply processing everything in one fell swoop, so the back-and-forth repeatedly exposes me to the arguments until I get them, while flagging the most important by repeating them most often.
Per my assumptions, stereotypes and absolute goddamn certainties, I refuse to believe that Piketty drinks anything but fine wine.
So he makes a lot of plausible but tendentious “decisions.” This is the same way NASA “discovered” a warming trend in the US data that didn’t exist in 2001; every set of adjustments is full of plausible changes that all just happen to confirm the authors’ biases.
And in both cases, the arguments over methodology and data tend to overshadow the fact that the attendant policy prescriptions don’t make any sense, as France recently found out with their Socialist experiment.
I knew there was a reason I liked you.
finally an ungated version of the criticisms.
funny that it is only provided by Piketty.
Yes imagine, a newspaper that dont share everything for free.
Clearly then, the newspaper must be wrong.
it was a blog post, not a newspaper.
it was free, you have to register or some bs.
Though I did not like Piketty thesis, as presented by summaries and critics I have read (since I haven´t read the book), and enjoyed Gilles questions, Thomas has answered it in a proper way. Not that everything was clariefied, doubts about the quality of his data and method choices persist, but he was unequivocal.
So he makes a lot of plausible but tendentious “decisions.”
That’s kind of what I got out of it, too. He’ll use one year as a substitute for another, make decennial averages including the substituted figures, and make averages where different decades have different numbers of entries. Each individual year is of course not a measurement, but a corrected measurement, and the corrections are all self citations.
The irritating thing is that it wouldn’t be too hard to do these things carefully. Say, by using a gaussian weighting function to compute a smoothed yearly average. You could even use a gaussian with a five year width to get a “decadal” average. But it’s all ad hoc, all the way down.
So he has abandoned his data on the U.S. and basically says “substitute in this other study for my data, which is not good.” And it certainly appears that the conclusions are not robust for varying assumptions. But yet somehow he argues that none of the disagreements matter very much. It’s like his conclusions are totally immune to the data. The data is completely uncertain and could be practically anything, but no matter how much the data might change the conclusion is sound apparently.
Since when does ‘some uncertainty’ == ‘could be practically anything’? The point was that he acknowledged that the wealth distribution is difficult to measure for the US because it must be measured through proxies. He strongly defended his analysis — but he went even further. He pointed to a more recent, independent study which supports his conclusions about US wealth ineq even more strongly than his own data and analysis.
You seem to be suggesting that this new and improved evidence is somehow irrelevant to the discussion and this it is just a diversionary tactic. Well the point is what do you believe about wealth inequality in the US in the last 30 years? I see Piketty defending his conclusions by citing the best available evidence.
Also Giles’ analysis is total bunk – P nails it – Giles would have us believe that Britain is equal as sweden?? I think that Giles has a lot of explaining to do here!
No, this is failure of reasoning. There are lots of fields of study where there is uncertainty — it does not follow that the facts or data in that field “could be practically anything.” Historians still debate the details of the Napoleonic wars — they don’t debate whether or not the Battle of Waterloo actually happened. Piketty clearly explains that the U.S. wealth data comes from a mix of estate tax records (which are less complete than in Europe, I guess due to the fact that most households are exempt from estate taxes) and wealth surveys which, like most surveys, have the problem of response bias among wealthier individuals.
You have to actually click the link Piketty provides in his response to learn this but Saez (who has co-authored with Piketty in the past) and Zucman have done the hard work of coming up with a new estimate of wealth distribution in the U.S. by collecting data from individual tax returns on income earned from assets and “backing out” an estimate of wealth for the individual based on reported investment income.
In short, if you are going to question the claim that U.S. wealth inequality is increasing, you can’t just wave your hands and say the data could be anything and expect other people to treat that as an intelligent or informed comment. Instead, you have to clearly explain why the rise in the inequality of investment income reported to the IRS is not due to the rise in the inequality of the distribution of the underlying assets.
In his previous book with Emmanuel Saez and Camille Landais on tax reform they answered to almost all substantial criticisms like that, they even had a website where you can simulate your own tax reform :http://www.revolution-fiscale.fr
Note that he didn’t answer the housing bubble criticism although he must be aware of it since the paper by Wasmer et al comes from a discussion of his book where he was present. Probably harder to answer to that one, it’s true.
The book itself addresses the impacts of housing and other asset bubbles, at least for the national capital time series (haven’t gotten to the inequality section). Reading the online criticisms, you’d think Pikkety just ignores asset bubbles, but this isn’t the case at all.
Well, I don’t think Piketty ever imagined he had much of a chance of convincing those of you who have been so blessed by biodiversity that falsifying the academic consensus climate model can be done in your spare time!
To be fair, anyone who uses the phrase “the academic consensus climate model” has no clue what they are talking about. (Let alone “falsifying” a model)
From the summary:
Indeed, if the survey data cited by The Financial Times is correct, it “would mean that Britain is currently one of the most egalitarian countries in history in terms of wealth distribution; in particular this would mean that Britain is a lot more equal than Sweden, and in fact more equal than what Sweden has ever been (including in the 1980s). This does not look particularly plausible.”
This seems to be put a pretty big hole in Giles’ boat.
Because Downton Abbey?
I have learned that Downtown Abbey is a TV show about aristocrats in Britain but I am still not sure what your comment means. To be more clear on my comment the data Giles uses to attack Piketty on British inequality is almost certainly incorrect. It is so badly incorrect it calls into question pretty much all of his other analysis as well.
“What is troubling about the FT methodological choices is that they use the estimates based upon estate tax statistics for the older decades (until the 1980s), and then they shift to the survey based estimates for the more recent period. This is problematic because we know that in every country wealth surveys tend to underestimate top wealth shares as compared to estimates based upon administrative fiscal data. Therefore such a methodological choice is bound to bias the results in the direction of declining inequality. For instance, as I note in the technical appendix to chapter 10 (available here), the recent wealth surveys undertaken by INSEE in 2004-2010 in France indicate a top decile share just above 50% of the total wealth, whereas fiscal data (inheritance and wealth tax) suggest a top decile share above 60% of the total wealth. The gap seems particularly large for the case of Britain, which could reflect the fact that the “wealth and assets survey” seems particularly bad at measuring the top part of the wealth distribution of the UK. Indeed, according to the latest report by the Office of national statistics (ONS), the response rate for this survey was only 64% in 2010-2012; this is an improvement as compared to the response rate of 55% that was observed during the 2006-2008 wave of the same survey (see ONS 2014, Table 7.1); but it is pretty clear that with such a low response rate, it is hard to claim that one can adequately measure wealth inequality, particularly at the top of the distribution.”
If you’ve spent 5 minutes in both countries, you know it’s not plausible.
The problem, from my perspective, is that all the people jumping on the FT’s criticisms of Piketty would’ve jumped on *any* criticism of Piketty, warranted or not. So it’s not inherently trustworthy.
Of course you could (and should) say the same for the people trumpeting Piketty’s own findings. At the risk of going all Robin Hanson.
Are we seeing Frenchness?
“He doesn’t give an inch.” Would he give a centimetre?
Piketty ¨ argues that The Financial Times’s approach requires apples-to-oranges comparisons, using estate tax data for older periods and then switching to surveys for more recent information. ¨
I think Piketty wins this one and Giles was comparing apples and oranges. The report that is the source of the Giles survey data that the top 10% wealth share is 44% also said based on previous surveys that wealth inequality was increasing. Their numbers for the top 1% wealth share was 17% for the early 90s vs 23% for 2007-8
You wait until the FT claims Piketty was born in Kenya.
Just you wait.
I find his reply adequate, he seems to put to bed all the claims that he made up numbers out of “thin air” and he definitely explains his methodology adequately. I think he’s a bit too abrasive towards his critics, though, and I tend to agree that the methodological disputes are mostly minor and don’t impact the overall argument.
What I’d like to hear is his response to Summers’ or Debraj Ray’s critiques of his theoretical work, which blows far greater wholes in his policy prescriptions and overall argument.
In fairness, Piketty does concede that the data are hard to get… and then says that if we had a wealth tax, we’d have much better data!
So Pikkety refuses to update his priors … which economist doesn’t?
I have no strong view on whether Pikkety is right, I will wait to make up my mind until I read his book. I felt the same about Reinhart-rogoff which has been on my pile for quite some time.
To me, there is a common issue in both where a famous academic result many want to be true is maybe not so robust. I make no particular claims about either, but do think that very few people have been consistent in their views of the two situations. This is hardly an original observation, but I think very few participants in the debate seem willing to seriously engage with it.
Take for example, the common sense argument critics deploy in each instance. Some people “just know” high debt is bad, others “just know” inequality must be rising at the very top.
Difference is that Reinhart-Rogoff actually made mistakes, Piketty didn’t.
That is in dispute, to be clear I am not taking a view. Many defenses of Piketty’s book suggest that any possible mistakes matter little. This was an assertion in the Reinhart-Rogoff debate.
To be clear, on the list of Piketty’s criticisms the interesting ones (to me) are not about excel errors. It’s more the large number of assumptions where reasonable people can differ.
If whether there are (or aren’t) excel errors is of secondary importance.
Other than assuming that Piketty assumes R>G in perpetuity, which he doesn’t, there are no rebuttal arguments that can be offered to his thesis. It’s all really air tight, such the econ text need a thorough rewrite.
So basically Piketty calls shenanigans on Financial Times. Funny how the financial times uses two different datasets to show declining inequality as if I used pounds, then kilograms to show I was losing weight.
I am looking forward to the next post in which you take a position on whether Piketty has successfully refuted the FT’s claim that Piketty’s analysis contained significant errors. Given your multiple previous posts on the subject, it would appear incumbent on you to take a position as to whether there was ever a legitimate claim of error in the FT piece, or, simply a disagreement with Piketty’s legitimate methodological choices that the FT misleadingly framed as a claim of error, as Piketty persuasively contends.
dont hold your breath
Comments on this entry are closed.
Previous post: Assorted links
Next post: Score one for the signaling model of education
Email Tyler Cowen
Follow Tyler on Twitter
Email Alex Tabarrok
Follow Alex on Twitter
Subscribe in a reader
Follow Us on Twitter
Marginal Revolution on Twitter Counter.com
Get smart with the Thesis WordPress Theme from DIYthemes.