By Peter Orszag
(Bloomberg News) Over the past three decades, the highest incomes in the U.S. have risen dramatically, and that has appropriately received lots of attention. At the same time, however, these high incomes have also become much more volatile, and that has gone almost unnoticed.
Conventional wisdom suggests that low-income households experience the greatest changes in response to macroeconomic conditions -- their income falls the most when the economy weakens, and it picks up the most when the economy recovers.
That conventional wisdom is in need of some updating. Today, the impact of macroeconomic events on household incomes forms a U-shaped curve -- it is greatest at the bottom and the top of the income distribution and smallest in the middle.
The strengthening link between high incomes and macroeconomic activity provides some insight into a stunning set of statistics: In 2010, according to research by Emmanuel Saez, an economist at the University of California, Berkeley, households in the top 1 percent of income distribution accounted for an astonishing 93 percent of aggregate income gains. During the slump from 2007 to 2009, according to the same data set, that group also accounted for a very large share of aggregate income losses -- almost half of the total decline.
The tighter connection of the affluent to the macroeconomy isn't limited to income. The Bloomberg same-store retail-sales indexes show a disproportionate decline for high-end stores in 2009, and then particularly rapid growth since 2010.
Thirty-Year Change
These recent data won't be particularly surprising to those who have read an impressive, but little-noticed, study published in 2010 in the Brookings Papers on Economic Activity by Jonathan Parker and Annette Vissing-Jorgensen. The two Northwestern University economists carefully document the increased "cyclicality" of high incomes over the past 30 years.
Since the early 1980s, the income of the top 1 percent has fluctuated more than average over the business cycle, rising five percentage points more per year than the overall average during economic expansions and falling 3.7 percentage points more per year during recessions. Before 1982, the highest-income category fluctuated less than the average.
Parker and Vissing-Jorgensen show that the increased cyclicality holds under a variety of income definitions, can't be explained by capital gains or stock options, and also applies to consumption. And they demonstrate that the connection between incomes and the macroeconomy is even tighter for the top 0.1 percent than for the top 1 percent, and even more so for the top 0.01 percent.
The impact of the business cycle on income thus now has a different shape from what it had 30 years ago. The "beta" (that is, the percentage change in family income in response to a 1 percent change in national income) is now 0.76 for families in the bottom quintile of the income distribution and 0.90 for families in the second quintile. For middle-income families in the third and fourth quintiles, it is less than 0.70. That's consistent with the conventional wisdom: The impact declines as income increases over that range.
What's new is that beta now rises again as we move further up the income distribution, reaching 1.01 for families in the 95th to 99th percentiles and 2.16 for families in the top 1 percent. So the bottom and top are most affected by the macroeconomy, with the impact on the top, if anything, actually larger than the bottom.
In another paper, released this week by the National Bureau of Economic Research, the economists Fatih Guvenen of the University of Minnesota, Serdar Ozkan of the Federal Reserve Board and Jae Song of the Social Security Administration use data from Social Security earnings records to examine how people are affected by the macroeconomy. Their analysis, like the Parker and Vissing-Jorgensen results, shows that the business cycle now affects very top earners much more strongly than in the past. As the authors note, "the fortunes of very high income individuals require a different classification, one that varies over time: more recent recessions have seen substantial income losses for high-income individuals, unlike anything seen in previous ones."
Four points follow from this change in the way incomes fluctuate over the business cycle. First, we shouldn't expect a high-pressure economy, despite its other attractive features, to reduce inequality in the same way it has in the past -- at least if the metric of inequality is how the top 1 percent fares relative to the middle.
High Incomes Rising
Second, a common cause may be triggering both the increased incomes and the increased sensitivity to macroeconomic conditions at the very top. Parker and Vissing-Jorgensen, for example, argue that changes in technology may be causing both the rapid surge in income and the increased cyclicality at the top. Erik Hurst, an economist at the University of Chicago, says higher incomes could even be viewed as compensation for the additional volatility: "Households with very high incomes may have anticipated the increase in risk," he writes, "and if so one would expect them to have demanded compensation for bearing that risk."
Third, one of the best ways to damp income volatility is to make the tax code more progressive, cushioning the blow from declines in income, while limiting some of the upside from gains. The tax system in the U.S. is highly progressive at the bottom of the income distribution -- which is why the betas on after-tax income for families at the bottom are substantially smaller than those on pretax income.
For the very top of the income distribution, however, the tax code is not progressive. In 2009, taxes rose as a share of income up to about $1 million or so. At that point, according to Internal Revenue Service data, the effective income-tax rate stabilized before declining a bit for families with incomes of more than $10 million. Introducing more progressivity into the tax code above $1 million would help to reduce after-tax income volatility at the very top.
Finally, if anything, high-earning households should be the ones most in favor of aggressively boosting the economy in the short run -- and not just out of benevolence. Yet I suspect, without definitive proof, that support for additional stimulus declines as one moves up the income scale.
(Peter Orszag is vice chairman of global banking at Citigroup Inc. and a former director of the Office of Management and Budget in the Obama administration. The opinions expressed are his own.)