David R. Henderson  

The Top 0.1 Percent Responds to Incentives

Market Failure: The Case of Or... Sidney Winter's Case for Gover...

Alan Reynolds writes:

One affluent member of the Top One Percent club, Paul Krugman, has narrowed his sights to the even more affluent top 0.1 percent in his new book, End This Depression Now! He claims that, "Recent work by the economists Jon Bakija, Adam Cole and Bradley Heim gives us a good sense of who the top 0.1 percent are. The short answer is that they're basically corporate executives or financial wheeler-dealers."

The phrase "corporate executives" clearly suggests CEOs and other top executives of publicly-traded corporations. Unfortunately, the paper Krugman refers to ("Jobs and Income Growth of Top Earners and the causes of Changing Income Inequality") shows that corporate executives account for a small and declining share of the income reported by the top 0.1%.

This table [Reynolds's post], adapted from that paper, shows that corporate executives in public companies accounted for only 15.5% of the very highest incomes in 2005, down from 21.1% in 1997. If we include capital gains, corporate executives accounted for 16.8% of top 0.1 percent incomes in 2005, down from 23.9% in 1997.

Reynolds points out that the authors of the paper reach an interesting bottom line that accords with other things Reynolds has been writing:
Bakija, Cole and Heim find an elasticity of about 0.72 for the top 0.1 percent "suggesting a high degree of responsiveness to incentives for income-earning efforts (or income reporting) among those with the highest incomes, and a correspondingly large deadweight loss from imposing highly progressive tax rates on these taxpayers."

Comments and Sharing


COMMENTS (11 to date)
Henry Nguyen writes:

One the the problems that has been brought up is the inequality in the tax system. People in the one percent arent fully taxed the 35% because as CEO's dont give themselves much of a salary. A bulk of what they are taxed on is in capital gains. A flat tax proposed by Rabushka would simplify the tax code, eliminate the tax inequality, and allow for advances in technology and gains in specialization. (Given all the tax preparers find different lines of work.)

Tom West writes:

I'm not certain I'm getting the message Alan Reynolds wants me to get out of this.

When I look at the high degree of elasticity, it looks to me that increasing the tax rate will significantly bring down the level of inequality, which is in general a good thing for social cohesion.

andy writes:

"it looks to me that increasing the tax rate will significantly bring down the level of inequality, which is in general a good thing for social cohesion."

Is it? I mean..really?

Even if it is, considering that you do it be effectively forbidding the 'rich' people to work, Chruchill was probably right:

The inherent vice of capitalism is the unequal sharing of blessings; the inherent virtue of socialism is the equal sharing of misery.

Tom West writes:

Is it? I mean..really?

Based on my personal experience with cultures where inequality is far more ingrained, yes.

It seems to me almost inevitable that (in general), our need to justify our wealth when others have little gradually transforms (as the gap widens) into a belief that the other human beings are less worthy, and eventually, less human, and thus not deserving of the same protections as us.

The second part (and one major reason why I am not a libertarian), is that I think it irrational to expect the most powerful in society to *not* try to change the rules in their favor, if it is within their power. A higher concentration of wealth makes it that much easier.

Perhaps I am wrong, I have only my experience to go on, but are there any historical counter-examples at all?

(By the way, I like the Churchill quote - a clear indication that some inequality helps propel growth. But like any graph (growth vs. inequality) that has zero on both ends, the real question is where's the sweet spot in the middle?)

Arthur_500 writes:
which is in general a good thing for social cohesion.

I have never seen any data that social cohesion is improved when income is brought to the lowest common denominator. In fact, this is exactly what brings about the brain drain.

If I were to be paid the same as others my skills and abilities and efforts and care and concern would all be for naught. I as a human want to be appreciated. Either I will depart for greener pastures or I will do nothing.

It appears that even in places like North Korea excellence is rewarded.

Joe Cushing writes:

Why is it that people are not willing to consider that corporations pay taxes before stock holders can collect dividends and that capital gains are based on expected future dividends. The reason for lower tax rates on income from stocks is because this income is taxed twice. For example, I remember reading how Berkshire Hathaway paid for half of the federal budget for a day one year and now Warren Buffet is claiming that he doesn't pay as much in taxes as his assistant----because he doesn't consider taxes paid by his company to be actually taxes paid by him and his shareholders. Corporations aren't people, they are owned by people. When corporations pay taxes, it's the shareholders who pay that tax.

I think I had this stuff figured out when I was in junior high. Why is it so difficult for people to get?

Tom West writes:

I have never seen any data that social cohesion is improved when income is brought to the lowest common denominator.

Well, obviously I should have qualified my statement with what I though would be obvious. Complete income equality is pretty harmful for the reasons you stated.

However, there's a lot of space between the current levels of inequality and the communist utopia :-).

Bob Murphy writes:

In my experience, when Krugman writes "basically" it means, "The following statement is not actually true, but it would bolster my point if it were." I'm being dead serious.

Jeff FIsher writes:

Looks to me like Reynolds has tortured the data quite effectively.

From the abstract of the paper:
"The data demonstrate that executives, managers, supervisors, and financial professionals
account for about 60 percent of the top 0.1 percent of income earners in recent years"

So the authors appear to agree with Krugman.

Reynolds excerpts table 7.
The title of table 7 is "Percentage of national income (excluding capital gains) received by top 0.1 percent, and each primary taxpayer occupation in top 0.1 percent"

So already capital gains are excluded. Table 1 includes capital gains, and looks to be the source of the point made by the authors in their abstract.

But even table 7 shows "Executives, managers, and supervisors (non-finance)" with 3.42% of national income of the 7.43% total for the top .1%, and "Financial professions, including management" with another 1.45%... for a total of 4.87/7.43=65%... so what did Reynolds do to further reduce the number... well other than directing readers away from the "financial professions, including management" bit (including it is an addendum, rather than the main point)?

Right down at the bottom of the table there is some further breakout. "Detail on executives, managers, and supervisors (non-finance)" which breaks out executives vs managers vs supervisors and breaks out "closely held business" vs "salaried". Reynolds decided to toss out the closely held executives, all the managers, and all the supervisors (though the supervisors barely matter). That's how Reynolds gets the number down to 1.14/7.43.

So the papers's abstract is supported by their data and is very close to Krugman's claim, which the word "basically" makes clear is a quick sketch which ought to be generally accurate, but maybe not exactly (the main exact difference one could claim is the use of the world "executives" vs "executives, managers, and supervisors".

Reynolds' calculation is the percentage of the national income excluding capital gains going to the top 1% which goes to non-financial executives not in closely held companies. Really getting quite far from Krugman's (and the paper's) claim.

So, basically, Krugman is correct and Reynolds is ridiculously cherry-picking.

Jeff FIsher writes:

I don't buy the "corporate executives" = "CEOs and other top executives of publicly-traded corporations" claim either.

There are quite a few sizable corporations that aren't publicly traded and I think the vast majority of the population would consider their managers "corporate executives".

But I suppose one could argue that it's a bit sloppy. Would be better to just say "executives", even with the "basically" qualifier.

Alan Reynolds writes:

Jeff Fisher suggests that “Krugman is correct” when he lumps together successful self-employed people and white collar workers (“executives, managers and supervisors”) in his definition of “corporate executives.” Krugman is likewise presumably correct to describe all financial professionals as “wheeler dealers,” and to suggest that corporate executives and wheeler dealers account for basically all income of the top 0.1 percent.

Bakija, Cole and Heim begin with the dubious assumption that “closely-held business income . . . largely reflect[s] labor compensation.” That assumption, plus the inexcusable exclusion of capital gains in this context, allows them to claim that “theories to explain the rising top income shares . . . must largely be about compensation of labor.” However, they contradict themselves by assigning taxpayers to the closely-held business category only if their income from self-employment, partnerships or S-Corporations exceeds their income from salary, bonuses and stock options. All other white-color workers are assumed “likely to be working for publicly-traded corporations” precisely because most of their income is from labor (or capital gains).

Bakija, Cole and Heim conclude that “financial market asset prices, corporate governance, entrepreneurship, and income shifting across corporate and personal tax bases may be especially important in explaining the dramatic rise in top income shares.” But they also find the drop in top tax rates on salary and partnership income alone (ignoring the 2003 drop in tax rates on dividends and capital gains) accounts for much of the rise in top income shares. Within that list of explanations, only “corporate governance” fits Krugman’s description. Yet all relevant corporate governance issues − such as shareholder voting on compensation, or IRS and SEC regulations related to compensation − apply only to executives of publicly-traded corporations, not to shareholders in an S-Corporation, partners, hedge fund managers, or the self-employed. And compensation of corporate executives in public corporations is a miniscule and declining share of the highest incomes, by all of this study’s measures.

Incidentally, Tom West writes that, “When I look at the high degree of elasticity, it looks to me that increasing the tax rate will significantly bring down the level of inequality.” It will bring down the amount of pretax income reported on the top 0.1-1 percent of individual tax returns (as opposed to being reported on corporate returns, unreported or unearned). But the effect on after-tax income is ambiguous if the elasticity is close to unity as most other studies say it is. Because top income shares are inversely related to poverty and unemployment rates, the effect on inequality is also ambiguous.

Comments for this entry have been closed
Return to top