Did Inequality Fall During the Great Depression ?


The graph above is taken from Piketty and Saez in their seminal 2003 article in the Quarterly Journal of Economics. It shows that inequality fell during the Great Depression. This is a contention that I have always been very skeptical of for many reasons and which has been – since 2012 – the reason why I view the IRS-data derived measure of inequality through a very skeptical lens (disclaimer: I think that it gives us an idea of inequality but I am not sure how accurate it is).

Here is why.

During the Great Depression, unemployment was never below 15% (see Romer here for a comparison prior to 1930 and this image derived from Timothy Hatton’s work). In some years, it was close to 25%. When such a large share of the population is earning near zero in terms of income, it is hard to imagine that inequality did not increase. Secondly, real wages were up during the Depression. Workers who still had a job were not worse off, they were better off. This means that you had a large share of the population who saw income reductions close to 100% and the remaining share saw actual increases in real wages. This would push up inequality no questions asked. This could be offset by a fall in the incomes from profits of the top income shares, but you would need a pretty big drop (which is what Piketty and Saez argue for).

There is some research that have tried to focus only on the Great Depression. The first was one rarely cited NBER paper by Horst Mendershausen from 1946 who found modest increases in inequality from 1929 to 1933. The data was largely centered on urban data, but this flaw works in favor of my skepticism as farm incomes (i.e. rural incomes) fell more during the depression than average incomes. There is also evidence, more recent, regarding other countries during the Great Depression. For example, Hungary saw an increase in inequality during the era from 1928 to 1941 with most of the increase in the early 1930s. A similar development was observed in Canada as well (slight increase based on the Veall dataset).

Had Piketty and Saez showed an increase in inequality during the Depression, I would have been more willing to accept their series with fewer questions and doubts. However, they do not discuss these points in great details and as such, we should be skeptical.

Did 89% of American Millionaires Disappear During the Great Depression?

Over the years, I became increasingly skeptical of using tax data to measure inequality. I do not believe that there is no value in computing inequality with those sources (especially after the 1960s, the quality is much better in the case of the US). I simply believe that there is a great need for prudence in not overstretching the results. This is not the first time I make this point (see my paper with Phil Schlosser and John Moore here) and I think it is especially crucial for anything prior to 1943 (the introduction of tax withholding).

One of my main point is that the work of Gene Smiley which ended up published in the Journal of Economic History has generally been ignored. Smiley had highlighted many failings in the way the tax data was computed for measuring inequality. His most important point was that tax avoidance foiled the measurements of top incomes and how well they could transposed on the overall national accounts.

More precisely, Smiley argued that the tax shelters of the 1920s and 1930s would have affected reporting behavior. As long as corporations could issue stock dividends rather than cash dividends, delaying the payment of dividends until shareholders were in lower tax brackets, there would be avoidance. Furthermore, state and municipal securities were exempted from taxation which meant that taxpayers could shelter income and end up in lower brackets. All this combined to wide fluctuations in marginal tax rates conspires to reduce the quality of the tax data in computing inequality. Rather than substantial increases in inequality, Smiley found that his corrected estimates (which kept tax rates constant) suggested no increase in inequality during the 1920s and a minimal decrease when you exclude capital gains.

Alongside John Moore, Phil Schlosser and Phil Magness, I am in the process of attempting to extend the Smiley corrections to include everything up to 1941 (Smiley had ended in 1929). As a result, I had to assemble the tax data and the tax rates and I was surprised to see that, even without regressions, we can see the problem of relying on the tax data for the interwar period.

The number of millionaires in the tax reports is displayed below. As one can see, it is very low from 1917 to 1924 – a period of high tax rates. However, as tax rates fell in the 1920s, the number of millionaires quintupled. And then, when the Depression started in synchronicity with the increases in top marginal tax rates, it went back down. It went down by 89% from 1929 to 1941. Now, I am quite willing to entertain that many millionaires were wiped out during the Great Depression. I am not willing to entertain the idea that 9 out of every 10 millionaires disappeared. What I am willing to entertain is that the tax data is clearly and heavily problematic for the pre-withholding era.* This is evidence in favor of caution and prudence in interpreting inequality measures derived from tax data.



I am of those who believe that inequality was lower than reported elsewhere in the 1920s, higher than reported in the 1930s and 1940s. Combined together, these would mean that inequality would tend to follow a L-curve or a J-curve from the 1920s up to the present rather than the U-curve often reported.  I will post more on this as my paper with Moore, Schlosser and Magness progresses. 

Can we use tax data to measure living standards (part 2)?

Yesterday, my post on the differences in per capita income and total income per tax unit caused some friends to be puzzled by my results. To their credit, the point can be defended that tax units are not the same as households and the number of tax units may have increased faster than population (example: a father in 1920 filled one tax unit even though his household had six members, but with more single households in the 1960s onwards the number of tax units could rise faster than population for a time).

The problems regarding the use of tax units instead of households is not new. In fact, it is one of the sticking point advanced by skeptics like Alan Reynolds (see his 2006 book) and, more recently, by Richard Burkhauser of Cornell University (see his National Tax Journal article here).

Could it be that all the differences between GDP per person and income per tax unit are caused by this problem? Not really.

There is an easy to see if the problem is real. Both measures are ratios (income over a population). Either the numerator is wrong or the denominator is wrong. Those who view tax units as the problem argue that the problem is the denominator. I do not agree since I believe that the numerator is at fault. The way to see this is simply to plot total income reported by all tax units and compare this with real GDP. What’s the result?

Even with tax-reported income being deflated with the Implicit Price Deflator (IPD) instead of the consumer price index, we end up with a difference (in 2013) of roughly 3 orders of magnitude between GDP and tax-reported income relative to the 1929 base point. Basically, GDP has increased by a factor of 14.749 since 1929 while IPD-deflated tax-reported income has only increased by a factor of 11.546.


As a result, I do not believe that the problem is the tax unit issue. The problem seems to be that tax data is not capturing the same thing as GDP is!

Can we use tax units to measure living standards?

In the debate on inequality, I am a skeptic of how large a problem the issue is. Personally, I tend to believe that worries of inequality only increase when growth is stagnant. In fact, I also believe that there are numerous statistical biases causing us to misidentify stagnation as rising inequality. Most of the debate on inequality is plagued with statistical problems of daunting magnitudes (regional convergence in income, regional price levels, demographic changes, increasing heterogeneity of preferences, increasing heterogeneity of personal characteristics, income not being purely monetary, the role of taxes and transfers etc.)

One of them centers around the use of tax data. This has been the domain of Thomas Piketty and Emmanuel Saez. I can understand the appeal of using tax data since it is easily available and usable. Yet, is it perfect?

A year or two ago, I would have been inclined to simply say “yes” and not bother with the details. Theoretically, taxes should be an “okay” proxy for the income distribution and should follow average income even if at different levels. Yet, after reading the article of Phil Magness and Robert Murphy in the Journal of Private Enterprise I confess that I am no longer accepting anything as “granted” in the inequality debate. So, I simply decided to chart GDP per capita with the average taxable income per tax unit. Just to see what happens. Both are basically averages of the overall population, they should look pretty much the same (theoretically).  The data for the tax units is made available in the Mark W. Frank dataset based on the Piketty-Saez data (see here) and I deflated with both the CPI and the implicit price deflator available at FRED/St-Louis.

The result is the following and it shows two very different stories! Either the GDP statistics are wrong and we have average stagnation (which does not mean that there is no increase in inequality) or the taxable income data is wrong in estimating the trend of living standards and the GDP are closer to reality (which does not that there is no increase in inequality).  In the end, there is a problem to be assessed with the quality of the data used to measure inequality.

Tax Data