Incentives And Income Inequality

by: Shareholders Unite

We wrote an article here on Seeking Alpha about the economic effects of rising inequality that caused a little bit of a stir. In the article, we argued:

  • Since the 1970s, income inequality is increasing, most wealth creation is going to an ever smaller top and, most importantly, wages are not keeping up with increases in labor productivity.
  • In order to share in the wealth creation, many wage earners turned to saving less and borrowing more to fill the gap between rising productivity and relatively stagnant wages. This was stimulated by financial deregulation and rising asset prices, until the latter became a bubble and burst.

We already mentioned the heated discussion that followed this proposition. Some reactions were fairly common, along the lines of @perplexedtex:

And if that happened, how would that make you, or anyone else better off? They have a talent that the market values....what's wrong with that? This idea that there is a finite amount of wealth to be had in the universe and that, somehow, it's being hoarded by "the rich" and unavailable to everyone else perplexes me. Eighty percent of all millionaires are self made and there's a real good chance that they were not in the 1% before they were successful. We used to applaud that in America.

Indeed. There is both a widespread fear that attempts to reduce income inequality-- for instance, via more progressive taxes-- might reduce the incentives to work and invest. There is also a rather long tradition in the economics literature that efficiency and equality are to some degree mutually exclusive goals.

Argentina and Brazil

We came to the idea that rising inequality might have been a factor in producing the financial crisis when we were writing an article about Argentina, based on some quite surprising economic data. In both Brazil and Argentina, strong economic expansion went hand in hand with reductions in inequality.

Argentina's case-- while not nearly as well known as that of its heralded bigger neighbor-- is the more remarkable, as its economy grew twice as fast as Brazil while having a host of market unfriendly policies. Also, the commodities boom only explains a fraction of Argentina's growth. It's much less important than generally assumed.

We were left wondering what else could explain such a booming decade? Could there be a causation which runs something like this: measures to reduce income inequality increase domestic demand because lower income brackets have higher marginal propensity to consume? It fits the facts in Argentina and Brazil. Both countries showed strong expansion of the domestic market.

Since income inequality in the U.S. is rising, and now almost as large as in Argentina or Brazil, could the causation run the other way around in the U.S.?

This led us to Fordism, the link between mass production and mass consumtion via linking wage increases to productivity growth. This was an important underpinning under the golden years from 1950 to 1970. The link was severed in the last couple of decades.

There is already IMF research that shows that the greater the levels of inequality, the less sustainable are upturns in the economy.

The economists found that income distribution contributes more to the sustainability of economic growth than does the quality of a country’s political institutions, its foreign debt and openness to trade, the level of foreign investment in the economy and whether its exchange rate is competitive. It’s not too hard to see why. Extreme inequality blocks opportunity for the poor. It can breed resentment and political instability — discouraging investment — and lead to political polarization and gridlock, splitting the political system into haves and have-nots. And it can make it harder for governments to address economic imbalances and brewing crises. [NYT]

So, one might construct the following argument. Reducing income inequality might have detrimental effects on incentives to work and invest. However, some of the economic consequences of that might be compensated by the higher marginal propensity to consume of lower income groups, especially if one starts from a very skewed income distribution (like Argentina and Brazil).

Money as a proxy for status, power and success

But there is something else. Consider the following statement from top golf player Luke Donald:

It [topping both lists] means the most. Forget the money, forget all the trappings of golf. To do what no-one else has done is what we all practice for. [BBC]

Ask yourself this. How much of the rich, let alone the very rich, are exclusively, or even mainly motivated by money as such? For starters, money, just like almost anything else, suffers from decreasing returns after a certain amount. One can only eat so many meals a day, live in so many houses, etc.

But that's not the main thing. There are good reasons to think that money is not the motivator in itself, but functions as a proxy for status, power and success, at least beyond a certain amount. We could unleash a wealth of evolutionary psychology literature here, but won't.

It is useful to keep in mind that other countries with well functioning economies function with much less income inequality, and so did the U.S. between 1950 and 1970/80. Jobs still got created in those places.

If money-- beyond a certain point-- suffers from decreasing marginal utility and functions as a proxy for status and power, it essentially becomes a positional good. The consequences of this are rather far-reaching.

The money game in the very high income brackets becomes what economist would call a 'rank-order tournament.' What's important in such a rank order tournament (as the name suggest) is not the absolute pay-off, but one's position in the hierarchy (one's 'rank order' in the 'tournament').

This suggests that absolute values of incentives could be significantly reduced without losing much-- if any, power-- as players are much more obsessed with relative (rank order), than with absolute positions (earnings). As Robert Frank has argued (for a long time), this even makes most of the winners not real winners:

They’ve all built bigger mansions and staged more lavish parties, yes, but in so doing, they’ve simply raised the bar that defines what’s considered adequate in these categories.

It's akin to the familiar stadium metaphor:

all stand, hoping for a better view, only to discover that no one sees any better than if all had remained comfortably seated.

So incentives matter, but the relative value matters much more than the absolute value after a certain amount. They're positional goods, proxies for status and power, and they have a habit of creating a race to keep up, or face loss of status.

Here is the impression that Lane Kenworthy got from reading Steve Jobs' biography:

Jobs himself seems to have been driven mainly by a passion for the products, for winning the competitive battle, and perhaps for status among peers. The satisfaction of achieving excellence and of beating one’s opponents appears to have been far more important than monetary compensation. Excellence and victory were their own reward, rather than a means to the end of financial riches.

Apart from money as a proxy for status, power and success, there is also such a thing as intrinsic reward. The reward one gets from a job well executed, or the simple recognition of one's peers. There are those who warn that relying too much on financial incentives might produce some harm to intrinsic motivation.

All this leads to a rather important conclusion, it's not absolute incentives that matter, but relative incentives. In theory, one could half income differentials and incentives would still essentially be in place. As long as people who are more talented put more effort in (or are at the right time at the right place) and get a bigger reward, it doesn't mean all that much how much bigger the reward is, as it's the relative standing that is likely to be much more important.

The rising inequality is basically an arms race to increase one's standing, but it doesn't necessarily depend on the size of the prize, and the arms race itself has destructive side effects.

Reducing inequality as a coordination problem

This essentially means that reducing inequality is a coordination (or collective action) problem. If we could somehow find a way to reduce the absolute values of incentives, everybody could be better off.

But it's very hard to get that going. In fact, market forces are working the other way around as globalization and technology give the best performers in most categories a vastly different proportion of the spoils (the so called 'winner-takes-all' phenomenon). The problem is that we need everybody to curtail at the same moment.

It's like banker's bonuses, if they start somewhere, there is something of a risk some bankers 'flee' elsewhere with less restrictions. The same has been said about steeply progressive income tax scales. Something of an arms race in reducing marginal rates has ensued in the last couple of decades, in order to keep the 'best and the brightest' at home. It's the same kind of coordination problem.

Enter Robert Frank

Now, the same Robert Frank has come up with a rather interesting policy proposal. No progressive income tax, no, a progressive consumption tax. He claims that it could:

reduce the costs of growing income disparities, while at the same time freeing up several trillion dollars of additional resources each year—more than enough to pay down the federal debt and rebuild our crumbling infrastructure—all without requiring painful sacrifices from anyone.

A sort of progressive 9-9-9, perhaps? Here is how it works:

Families would report their taxable income to the IRS (ideally under a tax code that greatly simplifies the calculation of taxable income), and also their annual savings, as many now do for IRAs and other tax-exempt retirement accounts. The difference between those two numbers—income minus savings—is the family’s annual consumption expenditure. That amount, less a large standard deduction—say, $30,000 for a family of four—is the family’s taxable consumption. Rates would start low and would then rise much more steeply than those under the current income tax.

Very high marginal income tax rates discourages some productive activity. However, even if the highest marginal consumption bracket is 100%, it might even encourage productive activities. How's that? Well, a family planning a $2M expansion of its mansion would face another $2M in taxes, the total cost would be $4M.

This is a bit of a disincentive to consume in the extreme luxury segment, and could very well lead to a reduction in the size of these expansions (and other luxury expenditure):

The fiscal magic occurs because other wealthy families who’d also planned additions to their mansions would respond in a similar way. And since no one denies that, beyond some point, it’s relative, not absolute, mansion size that really matters, the smaller additions would serve just as well as if all had built larger ones.

More income would be saved at the top (which is where the progressive consumption tax has its biggest bite). So more could be available for investment, rather than consumption.

Just as interesting, the prospect of an introduction of this kind of tax could act as a considerable spur to the economy:

knowledge that the tax was coming would stimulate a burst of private spending that would help get the economy back on its feet. Anyone who was thinking about buying a bigger yacht or building a bigger mansion would rush to do so before the tax took effect.

And how would it reduce inequality? Well, obviously by the progressiveness of the consumption tax, that is, higher marginal tax on higher consumption. But there is a second effect:

The wealthy family that builds a bigger mansion or stages a more lavish wedding celebration almost surely had no intention of harming others. But its actions nonetheless harm others, by shifting the frames of reference that shape what they must spend in those domains. The progressive consumption tax creates an incentive to take those external costs into account.

It reduces some of the 'arms race' in positional goods, goods for which relative standing matters much more. That race is pretty much comparable to the evolutionary arms race of male deer creating ever more lavish antlers to attract mates. If they could all agree to be a bit less lavish, everybody would benefit.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.