Riding the High Income Wave

It may be heretical to suggest it, but those who experienced the greatest shock to their incomes were likely the top 1 percent of earners. That’s right, America’s highest incomes may have been hit hardest in terms of the scale of the contraction, according to research by a pair of economists at the Kellogg School of Management. In fact, incomes of the superearners are vastly more sensitive to economic cycles of booms and bust than previously realized, conclude Annette Vissing-Jørgensen and Jonathan A. Parker, both professors of finance at the Kellogg School of Management.

Previous studies hold that the poor typically bear the worst of recessions. But the past three decades have bucked this trend of past years, Vissing-Jorgensen and Parker found in a recent study published by the Brookings Institution in its Papers on Economic Activity. They say that multiple lines of evidence reveal that since the early 1980s, the incomes of the top 1 percent of earners in our nation have become 2.4 times more sensitive to economic expansions and contractions than those of average earners. This cyclicality, they say, is closely coupled to a well-studied phenomenon where these top earners are also taking home an increasingly greater share of the nation’s income. Further bucking conventional wisdom, the researchers found that incomes of the top earners were even more volatile—rising more in booms and falling more in recessions—than those of folks at the very low end of the income distribution.

To reach these conclusions, Parker and Vissing-Jorgensen used several datasets based on tax returns and population surveys to scrutinize the distribution of annual incomes in each year and track the top 1 percent of earners, or the top 1 percenters. The top 1 percenters are not necessarily the same individuals each year, just those earners who land at the frothy top. They have increased their share of the nation’s aggregate income from 8 percent in the early 1980s to 18 percent in 2008. The top 0.01 percent increased their share of aggregate income from 0.7 percent to 3.3 percent for the same time period. As a whole, the top 1 percenters more than doubled their income share between 1982 and 2008, while the top 0.01 percent more than quadrupled theirs.

To put a number on what this means, consider that in 2008 the top 1 percent of earners had, on average, an income 14 times the size of an average person’s. This equates to an average income of $906,000 for the top 1 percent as a whole and $17.1 million for the top 0.01 percent. In other words, the superearners are gobbling an ever-larger slice of the nation’s income pie, but, like an acquired allergy, that bigger slice is making them ever more sensitive to economic dips and peaks.

“What we found is that the rich ride the waves of the economy much higher and much lower than the average earner,” Parker says. “We’re becoming a winner-take-all society, where the top earners win big and lose big.”

The question is, why?

Rise of Cyclical Incomes

Vissing-Jorgensen and Parker say that before the 1980s, the cyclicality of top earners was slightly less than average. Between 1947 and 1982, their incomes rose on average “1.2 percentage points per year less in each boom and fell by 1.1 percentage points per year less in each recession” compared to average households, the pair report. But after 1982, the top 1 percenters’ incomes rose on average 5 percent more per year during booms and fell 3.7 percent more per year during recessions. The incomes of the top 1 percenters were, on average, 2.4 times more cyclical than those of average earners.

Parker says they were careful to exclude fluctuations in earnings from capital gains or stock options, which would be closely tied to the peaks and valleys of the stock market. Similar findings held for just wage and salary earnings. “This is not happening due to just stock options,” Parker says. “The cyclicality holds true in a more general sense. But we can’t rule out the idea that the increased use of performance-based compensation is part of the causal mechanism.”

The team also found that the higher someone landed in the income distribution, the more cyclical their income tended to be. In the most recent economic expansion, from 2003–2007, the top 1 percenters’ incomes increased annually by an average 7.8 percent. But a different story unfolded when Vissing-Jorgensen and Parker further stratified the top 1 percent. They found the 99.0th to 99.9th percentile of earners expanded their incomes each year by 5.6 percent, the 99.9th to 99.99th percentile expanded theirs by 8.7 percent, and the incomes of the top 0.01 percent shot up a whopping 13.9 percent. In an almost mirror pattern, earners in these same brackets watched their incomes shrink annually during the early part (2007–2008) of the recent recession by 6.7, 8.9, and 12.7 percent, respectively.

When the pair compared the earnings cyclicality across the nation’s income distribution, they noticed a skewed U-shape: the middle income brackets were comparatively unexposed to economic fluctuations, the lowest quintiles were highly exposed, and the top 1 percent were extremely exposed.

Income inequality and cyclicality of the top earners held true across industries and were not driven by any one sector, Parker says. These facts also held true across recent decades and different countries. When Parker and Vissing-Jorgensen analyzed ten countries that have had similar sharp increases in wage inequality, they found that the same countries that had big rises in the income share of top earners also had big increases in the exposure of top incomes to economic fluctuations. This suggests that the phenomena may operate independent of social norms, tax and fiscal policies, and governmental regimes. However, it does not hold true across all countries, Parker says.

The findings that incomes of the top 1 percent have become more cyclical since 1982, and that the incomes of the top earners within the top 1 percent are even more sensitive, poke holes in the previously-held belief that the top earners can ride out downturns with few disruptions. The fact that these findings hold true in some countries with different tax, social, governmental, and financial norms adds to their validity. But at the same time, the findings raise big questions: Why are the super-rich earning so much more, and why has their income grown more sensitive since 1982?

Enter the “Superstar” Worker

In the early 1970s, the income gap between the haves and the have-nots began to grow. Economists explained the phenomenon in terms of technological improvements that favored those with high skills and education. The idea was that computers, for example, raised the productivity of those with high levels of education relative to those with lower levels. But in the early 1980s, the income share of the top 1 percent suddenly accelerated. This happened at exactly the same time that their earnings became more exposed to national income changes. (Wondering why that 1982 benchmark keeps popping up? “It’s not that a switch was flipped in 1982; it’s just that this trend started then,” Parker says. In other words, the date emerged from the data and is not due to changes in tax or fiscal policies.)

Vissing-Jorgensen and Parker posit that skill and education levels may not be driving volatility, but that the “superstar” worker effect is likely more responsible. Such workers can harness information communications technology to, well, simply do more. Vissing-Jorgensen and Parker first floated this explanation in a previously published paper, but expand upon it in their recent one.

“Think of it this way,” Parker offers. “A sports stars 30 years ago might be on TV only occasionally on a nationally televised game. But for the most part, the entertainment they provided was to the people who actually attended the games. Now, nearly every single professional sports game is televised, sometimes across the world. So the ability of one individual to entertain everybody has gone way up.” In this new world, the best entertainers provide a huge amount of prized entertainment for everybody, and so they get more of the income, he says. Wage inequality across professional sports players increases, and the best are paid more while the rest are paid less.

Or you might think of it in terms of an online retailer who hawks their wares at a slightly reduced price. Anyone, with a few clicks of a button, can compare prices in cyberspace, and so these sellers soon outcompete the others and take a larger share of the income.

In other words, the ability of these superstars to scale up their productivity allows them to provide a low markup to a vast market, driving up their earning potential. But when fluctuations or ripples squeeze those small markups, the superstars’ incomes take a huge hit.

“What we offer in our paper is evidence of a new fact, that the cyclicality of the top earners’ income is linked somehow to the increased share of aggregate income that they earn,” Parker says. “We think the information communications technology narrative does the best job of fitting the existing facts in conjunction with this new one.”

Parker says the paper does not make any pronouncements about whether the widening income gap is good or bad for our society. Instead, he says it is “focused on documenting the changing nature of our labor markets, proving the existence of this new fact, and ruling out some candidate explanations for this new fact.”

“The most plausible explanation of our finding is that information and communication technologies have changed and increased the extent to which there is a winner-take-all society,” Parker says. “Such a change can increase the extent to which superstars earn the bulk of the income, and therefore bear the bulk of the fluctuations.”

Related reading on Kellogg Insight

Today’s Rising One-percenters: The growing gap between the very rich and everyone