The rich are different still

They are more subject to the slings and arrows of outrageous fortune than they were prior to 1982, according to the work of two Northwestern economists, Annette Vissing-Jorgensen and Jonathan A. Parker.

We document a large increase in the cyclicality of the incomes of high-income households, coinciding with the rise in their share of aggregate income. In the United States, since top income shares began to rise rapidly in the early 1980s, incomes of those in the top 1 percent of the income distribution have averaged 14 times average income and been 2.4 times more cyclical. Before the early 1980s, incomes of the top 1 percent were slightly less cyclical than average.

And, it isn't because of stock options or capital gains.  Their surmise is that it is because higher income workers use more new technology and thus operate at a different scale than the rest of us;

We argue that these facts are not inconsistent with the hypothesis that the increase in top income shares was caused by rapid technological progress in information and communications technologies (ICT) since the early 1980s. If improvements in ICT have increased the ability of the most talented workers to handle more work or to scale their ideas by working with more production inputs, then the ICT revolution could have caused the incomes of the highest paid both to rise and to become more sensitive to economic fluctuations.

They find the same thing in Canada, which has a different tax code (and culture);

In the Canadian tax data, top income cyclicality is quite similar to that in the United States during the past quarter century. Further, in the Canadian data we are able to follow families across years (that is, we use panel data). Families in the top 1 percent of the income distribution in one year have income changes to the next year that are almost twice as cyclical as for the average. 

The Canadian data helps, especially since the data used for the USA comes from the usual suspects, Emmanuel Saez and Thomas Piketty, with its well known problems.  Also, they find confirmation in the consumption data, as it too shows greater cyclicality for top earners.

The greater volatility comes mostly from lower pay for hours worked (in a recession) for the highly compensated, not (as for low income earners) from reduced hours worked. The logic being that the highly skilled, working with more and better capital must be subject to lower profit margins and more exposed to cyclical fluctuation.

Makes as much sense as anything else, and more than many theories.