Thursday, March 18, 2010

Income Tax Distortions

Most economists, and also conservative talk radio hosts, will tell you that raising income tax adversely affects how much people work.  In economics, this is called the labor-leisure decision.  The response FLG often heard from liberals was that most people can't just not work, so the effects ought to be minimal.  He always thought this was a pretty compelling response.  However, he also figured that this meant only that the distortions manifested in other, more subtle, ways.

Until today, FLG had never seen an empirical study on this topic.  The St. Louis Fed has a paper (PDF) on what they call the wage-amenity decision.  (Perhaps this is a common nomeclature in the economics literature, but as he just mentioned FLG has never seen this before.)

Taxes can affect occupational choice by distorting the return to monetary wages relative to non-taxable amenities. In this paper, we study the effect of tax changes on wage-amenity decisions where amenities are defined in a broad and agnostic manner. Non-taxable amenities of a job include both observable characteristics such as health insurance provision and unobservable characteristics such as stress and workplace environment. We introduce a two-step estimation procedure and use compensating differentials as the summary statistic of the nontaxable amenities associated with a job. We believe that this methodology offers a fruitful means of characterizing the amenity decision faced by workers and can be extended to other contexts where occupational choice or non-taxable amenities play a significant role.

We find that when the net-of-tax rate increases, workers move to higher wage jobs, implicitly sacrificing non-taxable amenities. We estimate a statistically significant compensated elasticity of 0.05, suggesting that a 10% increase in the marginal net-of-tax rate leads workers to choose an occupation with a 0.5% higher wage. In related work, Powell (2009b) focuses on the tax elasticity of labor income and reports a mean elasticity of 0.5. Our paper, then, suggests that the wage-amenity tradeoff accounts for 10% of the overall labor tax distortion. While our estimated elasticity is modest, it is possible that the true long-term distortion is larger than this elasticity suggests.
What does that all mean?  Increase taxes and people choose to work in jobs that have tax free benefits, like better health insurance or more flexible work hours.  Their statistical analysis implies that if we lower the taxes people face by 10%, then they will search for even higher wages, but the effect appears to be pretty small.  Yet, they have reason to believe that this effect could be larger over the long-term.

What does FLG's explanation mean in practical terms?  Yes, liberals who argue that raising income taxes won't drive most people to stop working, which seems to be common sense, is correct.  BUT those people will choose jobs that pay less in wages with non-taxable benefits.  This isn't to say that an investment banker is going to give up his job and work at an NGO at a fraction of his former wage just because his taxes went up, but it does suggest that income taxes do distort in ways other than the simple choice of whether to work or not.


Anonymous said...

Regardless of the back-and-forth of academic debate, consider this: Do we think the behavior of the entrepreneurial class is significantly affected by changes in marginal income tax rates? The Editorial Page of the WSJ has been saying "yes" since the late 1970s, but there is literally no evidence that this is true, and much evidence that it is not.

For example, successful entrepreneurs realize their wealth largely through capital gains, not salaries. Ditto venture capitalists. When internal rates of return are sufficiently high, even fairly significant differences in marginal tax rates have a trivial effect on net after-tax income. It's basic math.

No one cares whether a law partner takes early retirement because he's concluded that taxes are too high. Nor is it easy to picture many corporate CEOs who would quit because rates went up -- their real income comes in other forms. And I doubt there is even ONE hedge fund manager who would retire if his profits were taxed as ordinary income. Issues of economic efficiency aside (though they are important), the only people one would not want to see particularly deterred from doing what they do best are job-creating entrepreneurs.* And for these people, marginal rates -- while potentially annoying -- are ultimately trivial.

P.S. Do people arrange their lives to minimize their taxes and maximize their incomes? Ever wonder why guys like Johnny Apple and others in his generation of reporters were mildly rich? Because they spent years overseas with U.S. salaries on which they paid no income taxes; that, combined with lavish expense accounts, kept the ink stains off their bespoke shirts, and allowed them the luxury, in their old age, of being reliable voices for muscular liberalism.
*This is not quite what I mean to say, but I am too lazy to fix it, and the after-tax environment only makes things worse.

FLG said...


What about Gentry and Hubbard?

"While the level of the marginal tax rate does not affect entry into
self employment in a consistent manner across specifications, we find robust results that
progressive marginal tax rates discourage entry into self-employment and into business ownership.
Our estimates of the effects of the convexity of the tax schedule on entrepreneurial entry are
rather large. For example, we estimate that the Omnibus Budget Reconciliation Act of 1993,
which raised the top marginal tax rate, lowered the probability of entry into self employment for
upper-middle-income households by about 20 percent."

Anonymous said...

... we find robust results that
progressive marginal tax rates discourage entry into self-employment and into business ownership.

Fine distinctions matter, and gross generalizations mislead (as is often their purpose).

Most small businesses are created by people who want to a) be their own boss, b) create jobs for their relatives, and/or c) escape the clutches of the tax system. (This is not a controversial claim, and even the WSJ has acknowledged as much.)

What we should care about, as a matter of economic policy, is that tiny fraction of *new* small business owners who are different -- who have created a business that is scalable.

(This is not to say that we should treat other small businesses unfairly, whatever that means, but that we should always bear in mind that the primary motivations for small business formation have little or nothing to do with large-scale job creation.)

P.S.That claim about the effects of the 1993 change in the marginal rate is somewhat inconsistent with what happened over the balance of the decade, isn't it?

Anonymous said...

I'm going to climb into this high-level discussion with a low-level anecdote: middle-to-upper-middle class people in Sweden and other high-tax places do a surprising amount of their own work - painting, plumbing, ditches, etc. For them to get the money to hire it done, from their highly-taxed incomes, and filter it through the taxes to be paid by the guys they would hire - it's not worth it.

As a society, it's not clear the Swedes are better off when their oncologists knock off after 40 hours and go home and paint their living rooms. They've put a lot of money into training their oncologists. Here in USA, we squeeze 60 hours a week out of our oncologists, and laborers can get jobs painting living rooms. dave.s.

FLG said...


That's a damn good point.

Alan Howe said...

Not planning to read the whole discussion but let me point out that Sweden's unemployment rate, even with oncologists doing their own home upkeep, appears to be lower than ours.

Creative Commons License
This work is licensed under a Creative Commons Attribution-No Derivative Works 3.0 United States License.