[A] new analysis, by Stephen J. Rose of George Washington University, adds an important wrinkle to the story: Income inequality has not actually risen since the financial crisis began.

How could that be? Because the crisis, which ran roughly from 2007 to 2010, reduced the pretax incomes of the wealthiest Americans more than the incomes of any group. The wealthy have indeed received the bulk of the gains since the recovery began, but they still haven't recovered their losses. Meanwhile, the steps that the federal government took in response to the crisis, including tax cuts and benefit increases, have mostly helped the nonwealthy.

Fascinatingly, Mr. Rose's case is not based on a new or previously undiscovered data set. It's based on the same statistics most commentators have been using to discuss inequality. The most up-to-date numbers come from the pathbreaking analysis of tax records by Emmanuel Saez, the University of California, Berkeley, professor who often collaborates with Thomas Piketty. A second set of statistics comes from the Congressional Budget Office.

Since the financial crisis and recession began, the incomes of the highest-earning households have fallen even more than the income of others.

Both point in the same direction: The income of the top 1 percent - both the level and the share of overall income - still hasn't returned to its 2007 peak. Their average income is about 20 percent below that peak. Yet we have all become so accustomed to rising inequality that we seem to have lost the ability to consider the alternative. Maybe it's because many liberals are tempted to believe inequality is always getting worse, while many conservatives are tempted to believe that the Obama economy is always getting worse.