David R. Henderson  

Henderson on Piketty, Part 3

Friday Night Video: Tom Palmer... Fixed Costs and Open Borders...

More excerpts from my recently published review of Thomas Piketty's Capital in the Twenty-First Century.

For those who are worried about growing wealth inequality because their own wealth is not growing, there is a simple solution: save more and invest in stock market index funds. And, to the extent possible, do so with tax-favored 401(k) and 403(b) plans and Individual Retirement Accounts (Roth or non-Roth.) When a friend who studies saving patterns of various ethnic groups in America visited me some years ago, I told him that my wife and I normally save between 15 and 20 percent of our before-tax income. His eyes grew wide. "You're Korean," he said, jokingly. Of course, hitting that saving rate meant that we didn't go to Europe or Asia, didn't buy $40,000 cars or $200 shoes, didn't buy expensive clothes, and didn't drink alcohol when we went to restaurants. What a tough life!

Piketty does not give any space in his tome to making that point. He writes as if he is the central planner making decisions from the top down and essentially disregards the fact that people are individuals who want to deal with their individual situations.

But even as central planner, Piketty fails. The driver of his model is his strongly held assumption that the rate of return on stocks will substantially exceed the growth rate of the economy and the growth rate of real wages. Under Social Security, your benefits will grow at no more than the growth rate of real wages because your benefits are paid by Social Security taxes on current workers. So, wouldn't it make sense to let people invest their Social Security taxes in stocks rather than get only the low rate of return that they get now? Piketty says no. He makes one good argument for this, one I myself have made: the transition problem out of the Social Security Ponzi scheme is wicked. But his other argument is that investing in stocks is "a roll of the dice." What happened to his confidence about the rate of return on stocks?

Given his emphasis on--and distaste for--inequality and his conclusion that owners of capital will get an increasing share of an economy's output, it is not surprising that Piketty favors much higher taxes on wealthy people. He argues briefly that the optimal top income tax rate in richer countries is "probably above 80 percent." He claims that such a rate on incomes above $500,000 or $1 million "will not bring the government much in the way of revenue"--I agree--but will drastically reduce the pay of high-paid people. He also suggests an annual "global tax on capital," with rates that would rise with wealth. "One might imagine," he writes, "a rate of 0 percent for net assets below 1 million euros, 1 percent between 1 million and 5 million, and 2 percent above 5 million." One might imagine many things: I take it, as virtually every reviewer pro or con has, that Piketty is not just "imagining" those taxes, but actually advocating them. He adds that "one might prefer" a stiff annual tax of "5 or 10 percent on assets above 1 billion euros."

But if there is anything we know in economics, it is that incentives matter. An annual tax on capital will reduce the incentive to create capital. With less capital than otherwise, the marginal product of workers will be lower than otherwise. Bottom line: Piketty's proposed tax on capital would hurt labor.

Next: More on how Piketty handles this incentive problem--and Robert Solow's thoughts on the matter.

Comments and Sharing

COMMENTS (20 to date)
Roger McKinney writes:

Great analysis, as usual!

We know what happens when tax rates reach 80% of income. We had such a tax for decades in the US. At such high levels people take their compensation in ways that are not taxed, company stocks for example. Boards looking to reward good managers will simply give them perks, like free use of the company jet, company-paid vacations, etc., like they used to do in the days before the Reagan tax cuts. Piketty doesn't know people and has too much confidence in his simplistic models.

Andrew_FL writes:

The statement that a high marginal rate on the highest earners will not bring in much revenue but drastically reduce high earner incomes, is a frank admission that the objective here is to bring the top down, not lift the bottom up.

After all, if it won't bring in much revenue, there actually isn't any opportunity to significantly lift the bottom up through redistribution-assuming that would work in the first place.

Shouldn't it just be common sense? Would you rather live in a world where everyone is at least as rich as Bill Gates is today, or a world in which no one ever is again?

Apparently the Pikettys of the world would unhesitatingly choose the latter.

Tom West writes:

While I don't agree with 80% tax at a million dollars, I will say that at a certain very high level, the money is obviously no longer the motivation. As CEO, you probably don't decide only to put in $10 million dollars worth of effort...

In other words, I doubt the president of IBM works harder now than he did in 1960, despite being able to keep far more of his income.

The effects of taxation on effort are real, but occur at much lower levels that what we leftists are typically obsessed with (the $20+ million earnings).

What happened to his confidence about the rate of return on stocks?

My guess - stability. Stocks may outperform in the long term, but are not really an option if you must *guarantee* stability every step of the way.

Of course, no-one can literally guarantee stability, but obviously it's unacceptable to build something where you acknowledge at the outset that there's some reasonable chance that the fund won't be able to pay out (i.e. if there's a multi-year stock market plunge).

If you have *unforseeable* circumstance (the government collapsed), then that's okay.

Don Geddis writes:

@Tom West: "there's some reasonable chance that the fund won't be able to pay out". But if Piketty actually believes that, then his whole complaint about "r > g" falls apart. If regular people can't rely on the stock market to provide for their retirement, then why does Piketty have such confidence that wealthy people can manage multi-generational assets in order to claim an ever-larger fraction of ownership of the economy?

You don't get to have it both ways.

Tom West writes:

I guess I wasn't clear. If stocks change by 5% +- 50% and a "risk-free" investment gives 2%, then an investment in the stock market is good for anyone who can survive long enough that the 5% becomes more important than the +/- 50%.

That might be okay over generations, but it's not a risk that many would feel retirement savings should be allowed to take (either because we aren't willing to have pensioners starve because of a bad decision or because starving pensioners *will* have the political clout to get a bail out, depending on your political orientation.)

Andrew_FL writes:

@Tom West-If the CEO doesn't work any harder for his marginal dollars, how come those deciding how to compensate them haven't figured out they could slash their pay with no consequences? That's an enormous foregone profit opportunity!

Vivian Darkbloom writes:

"Under Social Security, your benefits will grow at no more than the growth rate of real wages because your benefits are paid by Social Security taxes on current workers."

That may be the ultimate back stop, but before one gets to that, Social Security benefits are restricted in growth by the CPI-W, a very poor proxy for "the growth in real wages". If the growth in wages is "real", then pretty much by definition they exceed the CPI-W.


"If the CEO doesn't work any harder for his marginal dollars, how come those deciding how to compensate them haven't figured out they could slash their pay with no consequences?"

You seem to have a rather naive belief in the ability and incentive of "those deciding how to compensate them" to "seize those profit opportunities". Are those board members actually aligned with shareholders as much as with the corporate CEO? After all, they vote themselves stock options, too, and many of them are CEO's of other companies. How much CEO's are paid is based primarily on how much other CEO's are paid. It's pretty much a one-way race to the top. How effective, really, is the ability of shareholders (one group, which *in theory*, has the ability to control CEO pay) to seize that "profit opportunity"?

It's kind of like saying that if North Koreans really had an interest in liberty, they'd overthrow their masters, so, because they haven't, where's the problem?

Piketty should spend less time worrying about income redistribution of CEO pay and devote more effort to improving the rights of the owners of capital, that is, shareholders "to seize those opportunities".

Damien writes:

@Roger McKinney: that's not really a problem or an argument against high MTR. The IRS already has to determine what counts and doesn't count as income for tax purposes. So it's no different from saying that taxing income could encourage people to try to get paid in-kind. Except that it's not really happening because in-kind benefits are also taxed. So it's only a matter of enforcing pre-existing rules: the two examples that you list are already considered income for tax purposes.

Piketty is well aware that people have been *arguing* that high MTR = more untaxed in-kind perks. He just doesn't think it's plausible given what we know about total compensation back then: "perquisites would have had to be huge pre-1970 to generate a high elasticity of avoidance through that channel" and "According to Yermack (2006), Grinstein, Weinbaum and Yehuda (2008), and Frydman and Saks (2010), today’s perks are significantly larger than even the total taxable pay of top executives pre-1970s, casting doubtupon the idea that perks could have been even larger pre 1970" (Optimal Taxation of Top Labor Incomes: A Tale of Three Elasticities, p. 249).

The good thing is that this paper explicitly examines optimal tax rates for each competing story about soaring CEO pay: supply-side changes, previous tax avoidance, rent extraction. And they address (on pp. ) the objection that tax avoidance would be a problem

"Piketty doesn't know people and has too much confidence in his simplistic models."

It's a very ironic statement given that it's the opposing side of the argument that is most likely to argue that executive pay is determined by the most simplistic of all models and that rising compensation just reflects the fact that executives have become much more productive.

Damien writes:

" And they address (on pp. 251ff and at other places in the paper)

Tom West writes:

how come those deciding how to compensate them haven't figured out they could slash their pay with no consequences?

One, as Vivian Darkbloom pointed out, board members are usually peers to the CEO. I would not expect a group of plumbers, when evaluating what a plumber was worth, to choose the smallest price they could get away with.

Two, in every low information, high-risk environment I've been part of, pure rationality inevitably fell prey to superstition and "what is everybody else doing?". Also, remember that humans are wired to value things by how much they cost. An overpayed CEO will be valued more highly by observers than an identical CEO who is paid less.

(low information = you don't get to run 20-30 CEOs to see which one works, and outside of terrible CEOs, it's almost always impossible to know how to apportion success/failure to the CEO vs. the myriad of other factors that affect the success of a big company).

Three, overpaying CEOs costs the board almost nothing. If he doesn't work out, the board at least bought the best CEO (as proof by how much they paid). If they paid less for the same CEO, they'd be subject to criticism for trying to go cheap on the CEO and costing the company big time.

Rationality can be expected when you have hard reliable numbers and you are insulated from the human impact of your decision. Expecting perfect rationality in situations like evaluating CEO's salaries is to ignore everything we know about how humans work.

Richard Rider writes:

Just checking: Did not Picketty omit IRA's, 401k's and pensions when considering how much stock a person owns?

Any savvy investor saving for retirement knows that the FIRST place to invest is in such tax advantaged plans -- especially a ROTH IRA if one is under age 50 or so. One would EXPECT investors to put less directly into stocks, and more in such plans (which usually invest in stocks).

In the case of government career (30 year or so) employees who often will get 70%-100% of their salary upon retirement in a pension, it's only logical that they save NOTHING extra for retirement -- spending it all for immediate gratification. Does that make them victims of income inequality?

Andrew_FL writes:

So the answer to my question is:

1. Conspiracy


2. Irrationality

How unexpected. Forgive me if these claims don't pass the laugh test.

Nathan W writes:

Taxes on wealth would be more egalitarian than taxes on income because income can fluctuate enormously compared to wealth.

Find me a person with assets over $5 million who would change their general approach to things in a significant manner in the case of a global 1 or 2% tax on wealth.

At the $5 million threshold, or the $1 billion threshold, do you think these people would decide that they didn't want to play any more? Especially in the top class, there are only rare cases where connections and/or monopoly advantages are at the root of the wealth (even a sizable share of the wealth of Bill Gates can be attributed to monopoly advantage, if you consider that 15 years ago there were at least two, or perhaps three major competitors in the area of word processing and spreadsheet software.

Perhaps you would find some billionaire who stopped trying when faced by the deterrent of a wealth tax, but if there was a billion to be made in the first place, surely there will be plenty o others to fill their place in the market.

I don't think incentives matter in the conventional sense at these levels. They will make the best accounting decisions possible, and most likely this would not involve getting out of the game.

Tom West writes:


Well, if we look at other countries/cultures *or* if we go back in time in North America, CEO salaries were much lower.

Are you saying that every other place *besides* modern day North America is irrational?

Far more likely that non-market factors are more important to CEO compensation than the ideal, but imaginary perfect market negotiations.

Oddly enough, corporations are run by people, and as such, they tend to act like... people. It's not evil, it's just people. The only irrationality would be expecting otherwise.

andy writes:

Andrew, it seems to me that the argument could be different. The CEO wage is set by supply/demand. If the supply is inelastic, you can tax them, lower their wages without any effect on their performance. If you add income effect, they might work even more.

Andrew_FL writes:

@Tom West-No, because there isn't some fixed ideal salary for a CEO. People in the past and in other places are not less rational than Modern Day North Americans. But they aren't more rational, either.

What's far more likely is that what makes sense in different times and places, with different people...differs.

And I do expect people to act like people. Which means I expect them to act rationally, in the sense that I expect them to always act in a way which they believe will alleviate some uneasiness.

@andy-The idea that their labor supply is inelastic strikes me as highly implausible. Why should it be the case, other than this unshakeable need to believe people cannot possibly have anything but indifference between two large numbers in compensation?

You both seem so sure of what motivates CEOs. What grants you these mind reading powers?

andy writes:

The idea that their labor supply is inelastic strikes me as highly implausible. Why should it be the case, other than this unshakeable need to believe people cannot possibly have anything but indifference between two large numbers in compensation?

They surely are not indifferent between two large numbers - I am very sure they will prefer the larger one. The logic behind inelastic supply curve is different.

I certainly do not have mind reading powers; it just seems to me you could construct an argument quite plausibly - and inelastic supply curve doesn't seem to me that implausible.

Roger McKinney writes:

Piketty's admission that an 80% tax would not bring in much revenue for the state but would merely punish the wealthy fits the definition of envy perfectly. Piketty has elevated envy to a virtue.

AS writes:

Your analysis is spot-on, and expresses my own thoughts more eloquently than I ever could. I especially like the point about giving people the freedom to opt-out of social security and invest on their own at higher returns. It's ironic how policies enacted to help the poor actually backfire and make them poorer than they otherwise would be without top-down interference from overconfident and economically-illiterate central planners. It's a vicious cycle as the the poorer the poor get, the more they vote for policies which appear, on the surface, to help them, but in actuality only impoverish them further.

John B writes:

Re Nathan W

A 2% annual tax on wealth is huge and would have big impacts.

Given that current 'safe' returns are on the order of 1%, actual inflation is on that order (in fact, larger) and that nominal returns are taxed, a 2% tax reduces wealth. In other words, you can't pay the wealth tax out of the income produced by the wealth and have anything left over; the wealth is thus essentially valueless as an income source; you have less wealth after paying the tax.

Five million isn't that much when you think of professional-class couples near retirement. It's certainly not in the butlers-and-yachts category.

I don't think the impact would be "not playing anymore". I think it would be political fury at the proposers of the tax and a search for alternative investments that either have higher returns or aren't visible to the wealth tax.

I'm not an egalitarian and inequality doesn't bother me, so I don't feel the moral force others seem to feel behind a wealth tax.

Comments for this entry have been closed
Return to top