In his Dec. 4 speech on inequality, President Obama set the tone for politics in 2014. One claim he made, in particular, is of central importance in understanding what’s happening to American workers.

“Since 1979, when I graduated from high school,” Obama said, “our productivity is up by more than 90 percent, but the income of the typical family has increased by less than 8 percent. Since 1979 our economy has more than doubled in size, but most of the growth has flowed to a fortunate few. The top 10 percent no longer takes in one-third of our income; it now takes half.”

Obama, apparently, relied on two separate data sources for these claims. The first, regarding family incomes, is likely inflation-adjusted figures from the Census Bureau, which show that the median income for U.S. families grew just 7.8 percent from 1979 to 2012. The Census data, which includes only cash income received by families, is taken from the Current Population Survey, also known as the household survey.

The second statistic, about the top 10 percent's share of U.S. income, is from a dataset constructed by the economists Emmanuel Saez and Thomas Piketty from Internal Revenue Service data.

Their figures show the share of total income, including capital gains, of the top 10 percent of earners growing from just over a third (34.21 percent) in 1979 to just over 50 percent (50.42 percent) in 2012.

Why did the president need to rely on two data sources to illustrate the divergence in outcomes between the typical family and the top 10 percent? One reason is that in the Current Population Survey, top incomes are replaced with lower incomes to protect the confidentiality of respondents. Thus the truly rich won’t be represented in the CPS data, but they are in the IRS data that Piketty and Saez obtained, allowing them to shed light on the rise of the 1 percent in recent years.

What the Census and Piketty-Saez data have in common is that they both measure what might be called market income: Workers’ cash earnings, before benefits, taxes and government transfers. Together, they give a consistent and accessible picture of developments in U.S. labor markets over the past 30-plus years.

The flip-side of the ease of using those sources, though, is that they miss out some of the important and massive changes in U.S. demographics and compensation practices over that time, including the decrease in family sizes, the shift toward payment in the form of benefits — particularly health care instead of cash — and changes in the tax code.

The Congressional Budget Office, in a report that made use of both CPS and IRS data, provided some clarity on how the picture painted by President Obama would change if those demographic and other changes were taken into account.

Adjusting for changes in household size and taking into account benefits and government transfers, the CBO found that median before-tax income grew roughly 36 percent between 1979 and 2010. With taxes taken into consideration, the median household's income was 48 percent higher. (It's not yet as clear what has happened through 2012, in the wake of the recession. But Manhattan Institute scholar Scott Winship has estimated that poor and middle-class incomes had fully recovered their 2007 peaks by 2011 and have grown since then, thanks to safety-net transfers.)

In an apples-to-apples comparison, however, the top 1 percent grew by 201 percent over that period, according to the CBO.


In other words, when household size, benefits, transfers and taxes are taken into consideration, it’s clear that the middle class has not done as poorly as Obama indicated in his speech — in fact, they’ve experienced gains several times larger than he said over the past 30-plus years. But the top 1 percent have done several times better than that.

And it’s a matter of judgment which adjustments, if any, to income figures are appropriate. After all, it’s possible that the changes in household sizes and fringe benefits are, in themselves, a symptom of middle-class struggles in labor markets, and shouldn’t be corrected away.

In a 2012 study of the economic health of the American middle class, economists Richard Burkhauser, Jeff Larrimore and Kosali Simon wrote that for those who are “interested in how middle class Americans are compensated for their time in the labor market, for example, it is appropriate to use pre-tax, pre-transfer [market] income” statistics, such as the ones President Obama referred to in his Dec. 4 speech.

“However, for those interested in the overall economic resources available to individuals,” the economists wrote, “it is more appropriate to consider income defined as broadly as possible,” as it is in the CBO study.

That distinction might help sharpen the debate over inequality: Americans continue to make real gains in economic resources, even if labor market remuneration isn’t what it used to be. Meanwhile, by all measures we have, the 1 percent is taking off.

For the week ahead:

On Tuesday morning, a number of left-of-center experts, including top White House economic adviser Jason Furman, will speak at the Economic Policy Institute about raising the minimum wage, a major domestic effort backed by the president.

Outgoing Federal Reserve Chairman Ben Bernanke will speak at a Thursday event at a new Brookings Institution research center focused on central banking that will include a number of top economists. In recent weeks, Bernanke has tried to explain and justify the unprecedented actions he has taken over the past few years to ease the money supply and support the tepid economic recovery.

Also Thursday, the Bureau of Labor Statistics will provide an inflation update with its release of the Consumer Price Index for December. The CPI is the most closely watched index of inflation, and recently it has been running low, rising to around 1.2 percent in November.

The BLS also will release the Job Openings and Labor Turnover Survey on Friday morning. JOLTS, as its known, provides data on labor market “churn” — openings, hirings, layoffs and quits. Churn remains low in the wake of the recession.