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.

 

taxreports

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. 

On Gentrification, Inequality and Zoning

On the CityLab blog, Richard Florida posted a piece pointing out that gentrification has virtually no effects on homeowners. I can buy that result, especially since I wrote a policy piece for a think tank back in the summer of 2016 on the issue. The important point that Florida underlines (by citing a paper by Martin and Beck in Urban Affairs Review) is that homeowners are not being displaced, but renters are more likely to be. This will probably fuel some people who are concerned about inequality. I disagree.

I want to point out that my interest in the issue is entirely related to the issue of inequality which some individuals have tried to tie to gentrification (sometimes without understanding that causality can run both ways). If you want to tie the two issues together, then you must realize that there are four “types” of gentrification. First of all, gentrification always appear in an area that is poor and it is always a result of a shift in demand for land in that area. However, that area can be largely unoccupied or heavily inhabited. It can also be in a district where zoning is lax or burdensome. In each of these situations, you will different effects with different interpretations for inequality.

  • Scenario 1 (largely vacant, lax zoning laws): in this situation, demand shifts right but there is slack in the local housing market and in any case, supply can adjust easily. In that case, the effects on rents will be minimal and will probably be smaller than the economic gains in terms of local economic activity. In this situation, there is little displacement and there is in fact a reduction in inequality.
  • Scenario 2 (largely vacant, heavy zoning laws): same happens, except that the restrictions on construction and building conversions put a ceiling on the capacity of a local area to adapt. The effect on rents is ambiguous and depends largely on the relative quantity changes (how many people relative to empty units). There are probably small to moderate gains in the area. There are ambiguous effects on inequality.
  • Scenario 3 (heavily occupied, lax zoning laws): in this situation, the influx of individuals creates a temporary surge in rents. This is because, in the short-term, housing supply is inelastic. In the long-run, the supply is more elastic and new units can be added to counterbalance the price effects. So, there is a long-term benefit that comes after a small bump. More individuals will be displaced than in scenario 1. Overall, a reduction in inequality might occur.
  • Scenario 4 (heavily occupied, heavy zoning laws): in this situation, the influx happens in a market where the supply is highly inelastic (short and long-run). In that case, the shift in demand creates a substantial increase in rents. This is where gentrification can hurt and be tied to inequality.

These four scenarios are important because they show something important that some people have to understand. Gentrification can increase inequality. However, that depends on the context and the institutions (zoning) surrounding the area in which it happens. In all cases, gentrification is a normal process that can’t really be stopped but turns sour because of zoning laws. Thus, if you really want to tie gentrification to inequality, it should twice removed since the first parents are zoning laws and construction limits.

What is Wrong with Income Inequality? Five Reasons to be Concerned

I sometimes part ways with many of my libertarian and classical liberal friends in that I do have some amount of tentative concern for income/wealth inequality (for the purposes of this article, the otherwise important economic distinction between the two is not particularly relevant since the two are strongly correlated with each other). Many libertarians argue that inequality ultimately doesn’t matter. There is good reason to think this drawing from the classic arguments of Nozick and Hayek about how free exchange in a market economy can often interrupt preferred distributions.

The argument goes like this: take whatever your preferred distribution of income is, be it purely egalitarian or some sort of Rawlsian distribution such that the distribution benefits only the worst off in society. Assume there is one individual in the economy who has some product or service everyone wants to buy (in Nozick’s example it was Wilt Chamberlain playing basketball), and let everyone pay a relatively small amount of income to that one individual. For example, assume you have a society with 10,000 people all who start off with an equal endowment of $5 and all of them decide to pay Wilt Chamberlain $1 to watch him play basketball. Very few people would object to those individual exchanges, yet at the end Wilt Chamberlain ends up with $10,005 dollars and everyone else has $4, and our preferred distribution of income has been grossly upset even though the individual actions that led to that distribution are not objectionable. In other words, allowing for free exchange precludes trying to construct an optimal result of that free exchange (a basic consequence of recognizing spontaneous order).

Further, these libertarians argue, it is more important to ensure that the poor are better off in absolute terms than to ensure they are better off relative to their wealthier peers. Therefore, if a given policy will increase the wealth of the wealthiest by 10% and the poorest by 5%, there is no reason to oppose this policy on the grounds that it increases inequality because the poor are still made richer. Therefore, it is claimed, we should focus on policies that improve economic growth and the incomes of the poor and be indifferent as to its impact on relative inequality, since those policies are strongly correlated with bettering the economic conditions of the poor. In fact, as Mises Argued in Liberalism and the Classical Tradition, a certain amount of inequality is necessary for markets to function: they create a market for luxury goods that can be experimented and developed into future mass-consumption goods everyone can consume. Not everyone could afford, for an example, an IPod when it first came out, however today MP3 players are cheap and plentiful because the very wealthy were able to demand it when it was very expensive.

I agree with my libertarians in thinking that this argument is largely correct, however I do not think it proves, as Hayek argued, that social justice (understood in this context as distributive justice) is a “mirage” or that we should be altogether unconcerned with wealth or income distributions. All this argument does is mean that there is no overall deontological theory for an ideal income distribution, but there still might be good consequentialist reasons to think that excessively unequal distributions can impact many of the things that classical liberals tell us to worry about, such as the earnings of the poor, more free political economic outcomes, or overall economic growth. Further, even on Nozick’s entitlement theory of justice, we might oppose income inequality if it arises through unjust means. Here are five reasons why libertarians and classical liberals should be concerned about income inequality (note that they are mostly empirical reasons, not claims about the nature of justice):

1) Income Inequality as a Result of Rent Seeking

Certain government policies result in uneven income distribution. For an example, a paper by Patrick MacLaughlin and Lauren Stanley at the Mercatus Center empirically analyze the regressive effects of regulatory policy. Specifically, Stanley and MacLaughlin find that high barriers to entry create barriers to entry which worsens income mobility. Poorer would-be entrepreneurs cannot enter the market if they must, for an example, pay thousands of dollars for a license, or spend a large amount of time getting costly education and certifications to please some regulatory bureaucracy. This was admitted even by the Obama Administration in a recent report advising reform of occupational licensing laws. As basic public choice theory teaches, regulators are subject to regulatory capture, in which established business interests lobby regulators to erect barriers to entry to harm would-be competitors. Insofar as inequality is a result of such rent-seeking, libertarians have an obvious reason to oppose it.

Many other policies can worsen inequality. When wealthy corporations receive artificial monopolies from policies such as excessive intellectual property laws, insulating them from competition or when they gain wealth at the expense of poorer taxpayers through improper subsidies. When the government uses violent policing tactics to unequally enforce drug laws against poorer communities, or when it uses civil asset forfeiture to take the property of the worst off. When the government uses eminent domain to take the property of disadvantaged individuals and communities in the name of public works projects, or when they implement minimum wage laws that displace low-skilled workers. Or, if the structure of welfare benefits discourages income mobility, which also worsens inequality. There are a myriad of bad government policies which benefit the rich and exploit the poor, some of which are a direct result of rent-seeking on behalf of the wealthy.

If the rich are getting richer, or if the poor are stopped from becoming wealthier, as a result of government coercion, even Nozick’s entitlement theory of justice calls for us to be skeptical of the resulting income distribution. As Matt Zwolinski argues, income distributions are not only a result of, pace Nozick, a result of the free exchanges of individuals, but they are also a result of the institutions in which those individuals exchange. Insofar as inequality is a result of unjust institutions, we have good reason to call that inequality unjust.

Of course, that principle is still very hard to empirically apply. It is hard to tell how much of an unequal distribution is a function of bad institutions and how much is a function of free exchange. However, this means we can provide very limited theories of distributive justice not as constructivist attempts to mold market outcomes to our moral desires, but as rough rules of thumb. If it is true that unequal distributions are a function of bad institutions, then unequal distributions should cause us to re-evaluate those institutions.

2) Income Inequality and Government Exploitation
Of course, many with more Marxist inclinations will argue that any amount of economic inequality will inherently result in class-based exploitation. There are very good, stand-by classical liberal (and neoclassical economic) reasons to reject this as Marxian class analysis as it depends on a highly flawed labor theory of value. However, that does not mean there is not some correlation between some notion of macro-level exploitation of the worst-off and high levels of inequality which libertarians have good reason to be concerned about, for reasons closely related to rent-seeking. Those with a high amount of economic power, particularly in western democracies, are very likely to also have a strong influence over the policies set by the government. There is reason to fear that this will create a class of wealthy people who, through political rent-seeking channels discuss earlier, will control state policies and institutions to protect their interests and wealth at the expense of the worst-off in society. Using state coercion to protect oneself at the expense of others is, under any understanding of the term, coercion. In this way, income inequality can beget rent-seeking and regressive policies which lead to more income inequality which leads to more rent-seeking, leading to a vicious political-economic cycle of exploitation and increasing inequality. In fact, even early radical classical liberal economists applied theories of class analysis to this type of problem.

3) Inequality’s Impacts on Economic Growth

There is a robust amount of empirical literature suggesting that excessive income inequality can harm economic growth. How? The Economist explains:

Inequality could impair growth if those with low incomes suffer poor health and low productivity as a result, or if, as evidence suggests, the poor struggle to finance investments in education. Inequality could also threaten public confidence in growth-boosting policies like free trade, says Dani Rodrik of the Institute for Advanced Study in Princeton.

Of course, this is of special concern to consequentialist classical liberals who claim we should worry mostly about the betterment of the poor in absolute terms, since economic growth is strongly correlated with bettering living standards. There is even some reason for these classical liberals, given their stated normative reasons, to (at least in the short-term given that we have unjust institutions) support some limited redistributive policies, but only those that are implemented well and don’t worsen inequality or growth (such as a Negative Income Tax), insofar as it boosts growth and helps limited the growth of rent-seeking culture described with reasons one and two.

4) Inequality and Political Stability

There is further some evidence that income inequality increases political instability. If the poor perceive that current distributions are unjust (however wrong they may or may not be), they might have social discontent. In moderate scenarios, (as the Alsenia paper I linked to argue) this can lead to reduced investment, which aggravates third problem discussed earlier. In some scenarios, this can lead to support for populist demagogues (such as Trump or Bernie Sanders) who will implement bad policies that not only might harm the poor but also limit individual liberty in other important ways. In the most extreme scenarios (however unlikely, but still plausible), it can lead to all-out violent revolutions and warfare. At any rate, libertarians and classical liberals concerned with ensuring tranquility and freedom should be concerned if inequality increases.

5) Inequality and Social Mobility

More meritocratic-leaning libertarians might say we should be concerned about equal opportunities rather than equal outcomes. There is some evidence that the two are greatly linked. In particular, the so-called “Great Gatsby Curve,” which shows a negative relationship between economic mobility and income inequality. In other words, unequal outcomes can undermine unequal opportunities. This can be because higher inequality means unequal access to certain services, eg. Education, that can enable social mobility, or that the poorer may have fewer connections to better-paying opportunities because of their socio-economic status. Of course, there is likely some reverse causality here; institutions that limit social mobility (such as those discussed in problem one and two) can be said to worsen income mobility intergenerationally, leading to higher inequality in the future. Though teasing out the direction of causality empirically can be challenging, there is reason for concern here if one is concerned about social mobility.

The main point I’m getting at is nothing new: one need not be a radical leftist social egalitarian who thinks equal economic outcomes are necessarily the only moral outcomes to be concerned on some level with inequality. How one responds to inequality is empirically dependent on the causes of the problems, and we have some good reasons to think that more limited government is a good solution to unequal outcomes.

This is not to say inequality poses no problem for libertarians’ ideal political order: if it is the case that markets inherently beget problematic levels of inequality, as for example Thomas Piketty claims, then we might need to re-evaluate how we integrate markets. However, there is good reason to be skeptical of such claims (Thomas Piketty’s in particular are suspect). Even if we grant that markets by themselves do lead to levels of inequality that cause problems 3-5, we must not commit the Nirvana fallacy. We need to compare government’s aptitude at managing income distribution, which for well-worn public choice reasons outlined in problems one and two as well as a mammoth epistemic problem inherent in figuring out how much inequality is likely to lead to those problems, and compare it to the extent to which markets do generate those problems. It is possible (very likely, even) that even if markets are not perfect in the sense of ensuring distribution that does not have problematic political economic outcomes, the state attempting to correct these outcomes would only make things worse.

But that is a complex empirical research project which obviously can’t be solved in this short blog post, suffice it to say now that though libertarians are right to be skeptical of overarching moralistic outrage about rising levels of inequality, there are other very good empirical reasons to be concerned.

Canadian Megatrends: Top 1% income share and median age

Statistics Canada just came up with a study on the top income share of the top 1% in Canada. As I have explained elsewhere, my view of inequality is that: a) it has increased; b) not as much as we think; c) a lot of the increase is from desirable factors (personal utility maximization differing from income maximization or international immigration) or neutral factors (demography, marriage); d) that the inequality that is worrisome stems either from birth or government manipulations of the market and; e) that those stemming from government manipulations, direct (like subsidizing firms) or indirect (like the war on drugs which means that a large number of individuals are jailed and then released with a “prison earnings penalty” which stymies their income levels and growth), are the easiest to fight.

The recent Statistics Canada study allows me to make my point again with regards to element C of my answer. As I looked at their series, all I could think was “median age”. A lot of the variations seem to be related to the median age of the population. I went back to the census data I had collected for my book and plotted it against the data. This is what it looks like.

medianage

Why would there be a relation? Well, each year you measure the income distribution, the demographic structure of that population changes. As it grows older, you have more people at the top of their earnings curve relative to those at the bottom. Not only that, but earnings curve seem elongated in recent times – we live longer and so some people work older as witnessed by increased labor force participation rates above a certain age closer to retirement. And the heights of the earnings curve are now higher than ever before while we also enter later into the labor market.

Now, I am not sure how much aging would “explain away” rising inequality in Canada, but there is no point denying that it does explain some of it away. But, I would not be surprised that a large part is explained away. Why am I saying that? Because of this paper on Norway’s age structure. 

In Norway, the median age in 1950 was much higher than it was in Canada back then and today, it is roughly the same as Canada (although Canada has had a steeper increase in inequality). And according to the paper on Norway, adjusting for composition bias in inequality measures caused by aging, eliminates entirely the upward trend in that country. In fact, it may even reverse the trend whereby inequality adjusted for age has actually declined over time. This is a powerful observation. Given that Canada has had a steeper increase in median age, this suggests that the increase in inequality might be simply the cause of a statistical artifice.

The most depressing thing with Chetty et al.

The Chetty et al. paper has been on my mind over the weekend (see Saturday’s post). The one thing that has moved more or less in line with the absolute mobility measure of Chetty et al. has been…the size of government.

I know that as soon as some of you read the last four words on the previous paragraphs, your eyes rolled. However, even from a social-democratic perspective, it is depressing! It is not the first time I make this observation.   In the pages of Essays in Economic and Business HistoryI recently reviewed Unequal Gains (authored by Peter Lindert and Jeffrey Williamson and published at Princeton University Press) and I observed that the “great leveling” they observed from the 1910s to the 1970s had a lot to do with the northward migration of American blacks, the closing of the gender wage gap and the convergence of the southern states. I also observed that the increase in inequality in the United States after 1970 occurred at the same time as an the state grew more in size and scope (see blog post here).

However, as I mentioned elsewhere, I am very skeptical of the tax-based data on inequality in the United States and I am afraid to push that point. However, the Chetty et al. data provides further confirmation: trends in inequality/social mobility deteriorates as the state becomes more active (see the graph below).

sizegov

Now, I am aware that the causality can cut both ways. It may be that inequality (economic mobility) is rising (falling) in spite of increasing state action, it may be that state action is fueling the the rise (reduction) of inequality (economic mobility) or it may be that the state has no effects whatsoever on the evolution. Regardless of which of the three viewpoints you tend to adopt (I lean towards a mixture the second option – see my paper with Steve Horwitz here which is under consideration for publication), the implications are immensely depressing with regards to social policy in the last 75 years.

Sons outearning Fathers in Chetty et al. : working hours should be considered

In response to my post yesterday, my friend and economist/nuclear engineer (great mix) Laurent Béland pointed out that the Father-Sons mobility figures in Chetty et al. are depressing. Yes, at first glance, they are (see below – the red line). fathersons

But, at second glance, it is not as terrible. Think about family structures with the 1940 birth cohorts. The father works and, in most likelihood, the mother is a stay-at-home father. Most of the earnings come from the father who probably works 45 to 60 hours a week.  If my father earns 40,000$ at 60 hours a week or earn 40,000$ at 40 hours a week, the line remains at the same height, but we are not talking about the same living standard in reality. Chetty et al. do not account for hours worked to achieve income.  The steep decline – faster than the baseline of household-size adjusted decline – matches the steep increase in female labor force participation and the decline labor force participation of males (see graph here and Nicolas Eberstadt’s work here) as well as the decline in hours worked by males.

If the question had been “what are your chances of out-earning your father per hour worked”, then the red line would not have fallen like that. Income divided by labor supplied would probably bring the red-line back with the blue-line.

Note: Again, please note that I am not trying to rip apart Chetty et al. (as some have claimed elsewhere). Their work is great and as a guy who does all his research on providing data series regarding economic history, I am never going to rip on someone who does hard data work like Chetty et al. did ! My point is that I am not convinced that the decline is so big. And, in good faith, it seems that Chetty et al. do try to put the “caution” labels where its needed – and its important to discuss those caution labels before some politician or two-cents-pundit goes all Trump on us by saying stuff that this doesn’t say!

A flaw regarding the chance of “out-earning” your parents

When Raj Chetty publishes a paper, it generally comes with a splash. The last one is no exception. His paper (co-authored), picked up by David Leonhardt at the New York Times and Justin Wolfers on Twitter, basically measures the American dream : what are your chances to do better than your parents. The stunning conclusion is that someone born in 1940 had a 90%+ chance of “out-earning” his parents compared with a few points above 50% for those born in the 1980s. I am not convinced. Well, when I am not convinced, I am saying I am not convincing about how big the drop is! I think the drop is smoother (the slope of decline is gentler) and the starting point for the 1940 cohort is too high.  As a big fan of Chetty, I must press this point.

More precisely, I am saying that the bar (income threshold) over which someone had to jump in 1940 is underestimated and overestimated in 1980. Setting the bar too low (high) means very high (low) chances of “out-earning” your parents. To set the bar too low, you must underestimate (overestimate) the income of the parents.  This could occur if household economies of scale are not accounted for.

An income of 30,000$ for 3 persons is not the same as an income of 60,000$ for 6 peoples. On a per capita basis, the income is the same. But, if you adjust for economies of scale in housing and furnitures, there are differences (the simplest is square root).  This gives you income per adult equivalent. Chetty et al. are aware of that and they provided a sensitivity analysis which is not mentioned by those who are relaying the article. Since household size has tended to fall over time, the growth in per capita income is faster than the growth in income per adult equivalent (a better measure). Any correction for this long-term demographic trend would attenuate the slope of the decline of the chance to out-earn your parents. And indeed, once Chetty et al. make the correction, the decline is much more modest (but still present – see below).

size

Simultaneously, Chetty et al. also present other important sensitivity checks. All of them relevant. But, in a strange decision, Chetty et al. decided to isolate each of the sensitivity checks rather than compile them. Taken individual, they all seem minor – except adjusting for family size. But compound this with the other sensitivity check proposed by Chetty et al.: price deflators. Using the well-known bias in the the CPI that overestimates inflation by 0.8%, Chetty et al. find that, by the end of their perod, there is roughly a ten percentage point difference between the baseline uncorrected CPI and the corrected CPI (see below). Compound this with the corrections for family and you still get a decline – but again the slope of the decline is much more modest. If you add panel B from figure 3 in Chetty et al – which includes taxes and transfers – you probably get a few extra points up. There will still probably be a decline, but a moderate one.

pricetaxes

Finally, at footnote 19, Chetty et al. also point out that they do not account for in-kind transfers prior to 1967 (there were some).  And, on page 13, they point out that “one may be concerned that levels of absolute mobility for recent cohorts may still be understated because of increases in fringe benefits, nonmarket goods, or under-reporting of income in the CPS”. Add in all these little extra problems to the family size, the transfers and the inflation correction and I am not sure how big the drop from 1940 to the end of the studied period is. Finally, I would also add that an understudied point in economic history is what the distribution of in-kind payments according to income was. From studying the British industrial revolution, I have generally to see that it is the poorest workers who receive in-kind payments (which are not measured) and the richest receive much fewer of those in proportion of their incomes. One of the few to note that distributional was the hardcore left-leaning scholar Gabriel Kolko who mentioned this issue in Dissent back in the 1950s.  If Kolko is correct, then the income of “poor parents” in 1940 is underestimated. As a result, the bar over which the children of said parents must jump is set mildly too low. If that is the case, the odds for the 1940 birth cohort are overestimated.

Combine all of these things together and I am not sure that the drop is as dramatic as many are making it out to be. I would be very satisfied if Chetty et al. would publish all the corrections they did and do a sensitivity check with hypothetical regarding a sliding-scale of in-kind payments in 1940 according to income (10% of income for poorest to 0% for the richest). I would just like to see how much it matters.