Brad DeLong writes:
The big rise in inequality in the U.S. since 1980 has been
overwhelmingly concentrated among the top 1% of income earners: their
share has risen from 8% in 1980 to 16% in 2004. By contrast, the share
of the next 4% of income earners has only risen from 13% to 15%, and
the share of the next 5% of income earners has stuck at 12%. The top 1%
have gone from 8 to 16 times average income, the next 4% have gone from
3.2 to 3.7 times average income, and the next 5% have been stuck at 3
times average income.
It’s hard to attribute this pattern to a rise in the premium salary
earned by the well-educated by virtue of the skills their formal
education taught them. Such a rise in the education premium would
produce a much smoother rise in relative incomes among the whole top
tenth of the income distribution. The cross-percentile pattern doesn’t
It is especially hard because most theories of the rising education
premium attribute it to skill-biased technological change generated by
the high-tech computer industrial revolution. But the high-tech boom’s
effects on overall productivity became large only in the second half of
the 1990s, well after the biggest increases in inequality. The timing
doesn’t fit either.
Something else is going on.
Leisure time improvements, opportunities like the Internet, and CPI quality mismeasurement have made the lot of the bottom tiers somewhat better than these statistics indicate. (The absorption of would-have-been real wage gains into more expensive health benefits is another relevant factor.) Still the relative pattern is undeniable.
My intuition is that there has been an increase in the ability of very smart and very wealthy people to buy up undervalued assets and turn them into greater value. Improvements in capital markets and market liquidity are behind this trend, as well as the mere demonstration effect that many people have tried this and succeeded. Julio Rotemberg remains an underappreciated economist. American entrepreneurs were building up capabilities which exploded in value once the economy stabilized in the early 1980s. Michael Milken-as-we-knew-him could not have existed in 1973.
Note two features of this hypothesis: First, it is correlated with but not coincident to productivity improvements, which will follow with some lag, thus fitting the data. The capital gains come before the improvements. Second, it might explain the non-linearity of the spike in incomes. A modest multi-millionaire is not wealthy enough to play at T. Boone Pickens or Warren Buffett. He can "go along for the ride" in a hedge fund, but won’t reap most of the value from the major arbitrage opportunities. Those end up concentrated in a relatively small group of people and of course this tendency may be self-reinforcing with the accumulation of wealth and arbitrage expertise.
It might also help explain why the Bush changes in marginal tax rates seem to have produced more revenue than had been expected (yes I know about the weird baselines and projections and the like, but still it was a surprise and a welcome one). It is not a classic supply-side substitution labor supply effect, but rather a wealth effect. The very wealthy can put their assets to especially productive uses, or at least especially high capital-gains-producing uses. That puts high capital gains revenue, high productivity, and growing disparity of incomes all together in the same pot. Sound familiar?
Addendum: Here is Brad DeLong on press coverage of the revenue boost, and also on the small size of the self-financing effect; we also thank him for the endorsement (though note I’ve never been a member of any political party).