Pissing Down Your Leg

Thoughts on Economics and Economic Policy

Archive for November 2011

The Relationship Between Marginal Tax Rates and the Top 1%

leave a comment »

Over the last (almost) 100 years the United States has had widely different approaches to very high income in its tax code. In 1913 the top marginal tax rate was only 7%. By 1918 it was 77% before falling in the Roaring 20s down to a low of 24% in 1929. The Great Depression lead to another steep increase and the top marginal tax rate never dipped below 63% until the Reagan Revolution in 1982. For much of the 1950s it was over 90%. Now, the top income level to which the tax applied was much higher (especially in real terms) than it is now. But during the strong growth of the 1950s we asked our richest individuals to fork over $0.90 of the last dollar they earned.

At the same time, we have seen widely different shares of wage and salary income going to the top 1 percent of earners over those (almost) 100 years. According to Piketty and Saez, it reached a pre-Depression high of 18.4% in 1928. That fell fairly steadily down to a mere 7.7% in 1973 before returning to 18.3% in 2007 (it has dipped slightly during this Lesser Depression).

Putting these two measures on the same graph makes it look a lot like they are closely related.

The correlation is -0.61. When the tax rate goes down, the share of income going to the top 1% goes up, and vice versa. Is it possible that one is causing the other? Or that there’s something else causing both? On the other hand, is it possible that they aren’t related at all?

Assuming there is some sort of causal relationship at work, I can think of a few possible explanations. Deciding which one is correct is not so easy.

  • Rich people don’t work as hard (or as much) when marginal tax rates are high, reducing their income.
    • This is certainly possible. If I had to pay 90% of my last dollar to the government, I would definitely consider not working quite so much.
    • And this is one argument against high marginal tax rates. It reduces the incentive to work and therefore will hurt overall economic growth.
    • But there doesn’t seem to be any relationship between higher top marginal tax rates and economic growth. In fact, if there is any relationship, it looks like it is positive, at least until you get to about 70%.
  • Rich people hide more of their income from the IRS when marginal tax rates are higher.
    • I’m told that there was evidence of this in the 1950s and 1960s. And again, I’m sympathetic. 90% is a lot of money to hand over to the government on that last paycheck.
    • But one question then is how much they hid, and how much they stopped hiding when marginal tax rates came down. Specifically, was it enough to increase federal revenue?
    • The answer seems to be maybe. There is almost no relationship between Federal tax revenue as a percent of GDP and the top marginal tax rate.
  • Very high marginal tax rates serve as a signal about what society views as acceptable top incomes.
    • This is fairly speculative, but the idea would be that a top marginal tax rate of 90% says to CEOs and finance professionals that we don’t think anybody should earn above a certain amount.
    • So if we set the highest marginal tax rate at $10 million (or $20 million, or $50 million), then we’re saying that there’s no point in having an income higher than that. After all, going from $10 million to $100 million in gross income only increases take-home pay to $19 million.
    • I think the interesting question is what this would do to everybody else’s salaries. Would they fall due to lower economic growth? It doesn’t look like it. Would they increase? I’m not sure.

Written by Liam C Malloy

November 3, 2011 at 9:55 pm

Posted in Uncategorized