Do higher taxes make the rich work less?

Greg Mankiw is an economist whose academic work I greatly admire and respect, but whose economics-related punditry often leaves me gaping in incredulous dismay.

Case in point: Mankiw has an editorial in the New York Times explaining why higher taxes on the rich will make the rich work less. He writes:

An important issue dividing the political parties is whether to raise taxes on those earning more than $250,000 a year. Democrats say these taxpayers can afford to chip in a bit more. Republicans say raising taxes on those who already face the highest marginal tax rates will hurt the economy.

So I thought it might be useful to do a case study on one of these high-income taxpayers. Fortunately, I have one handy: me. As a professor at Harvard and the author of some popular textbooks, I am comfortably in the income range that would be hit by this tax increase. I have been thinking — narcissistically, to be sure — about how higher taxes would affect me.

This is not how economics should be done; anecdotes prove nothing. And anyway, Mankiw's case is a terrible example of what he's trying to prove; he could make a lot more money in the private sector, but doesn't, because the prestige of working at Harvard matters more to him.

But let's put that aside and look at the data. That is what good scientists should do. When people are taxed for supplying something (for example, labor), the amount that taxes make them reduce their supply is called the "elasticity of supply." If this elasticity (or the "elasticity of demand) is high, then taxes hurt the economy by causing people to stop doing whatever productive activity is getting taxed. If the elasticity is low, then taxes don't hurt the economy much, and instead simply move money around from one place to another (which is what we want them to do).

So what is the elasticity of labor supply? How much does income tax cause people to work less? When economists look at the micro-level data, they find that it's about 0.1; raising average income taxes by 10% reduces labor supply by about 1%.

That's not much. The reason it's so low is that most people can't actually decide how much they work. Greg Mankiw may have the chance for side gigs as a consultant or a speaker, but he's a rarity in that regard; most people collect one single paycheck, and can't decide their hours. It's either a full workweek or unemployment. (In labor economics, this is known as the "extensive margin," in case you were wondering.)

This fits perfectly with what we see when we examine the historical record. In the 1970s, the top marginal tax rates were very very high - including state taxes, it was over 90% in some states. And yet working hours have actually declined since taxes were drastically cut in the early 80s. That's right: lower taxes, less work. Now, this is not to say that the lower taxes caused people to work less. The point is that Mankiw's warning about how higher taxes on the rich is going to stop them from working has no basis in historical fact.

These facts heavily imply that Mankiw's opposition to higher taxes on the wealthy is based not on any concern for the efficiency of the economy, but on his personal ideology. In fact, this is not conjecture; Mankiw has repeatedly and explicitly expressed the idea that rich people, having contributed more to society than others, "deserve" lower taxes.

In my opinion, this is not how good economics is done. Economics, I believe, should be about the facts first and ideology second; when Mankiw invokes largely specious arguments about the impact of taxes on labor supply, for the (unstated) purpose of advancing his ideology, he behaves not as a scientist but as a lawyer, disingenuously pushing his case using any means at his disposal. This is not fair to the American people.

Neither, in my opinion, are the policies he supports. This country already faces huge deficits, and spending cuts alone - though absolutely necessary - will not be enough to close the gap, especially as the Baby Boomers retire. Higher taxes are needed to keep our economy solvent. Because the elasticity of labor supply is so low, personal income taxes are a relatively efficient way of raising this money (as opposed to, say, corporate taxes, which are actually way too high in this country). If Greg Mankiw wins his battle to protect the rich from "undeserved" taxes, economic efficiency - which he claims to support - will be swamped under a rising tide of debt. Mankiw sullies his reputation as an economist by ignoring this looming danger.

Update: There's also another reason why income taxes don't discourage work as much as Mankiw claims: income effects.

Update: Ryan Avent makes a bunch more points about why Mankiw's analysis is simplistic.