The following post by Ruy Teixeira, author of The Optimistic Leftist and other works of political analysis (cross-posted from his facebook page):

What’s happening with the middle class?

Does the middle class want to get ahead faster or stop falling? It makes a difference which of these is correct, when thinking about what message to promulgate and what programs to emphasize.

Noah Smith rounds up data that suggest a focus on getting ahead faster might be warranted, despite the well-known problems with wage gains since the 1970’s.

“The average American has, in fact, seen modest gains since the early 1970s; the falling wages of production workers don’t tell the whole story. A more comprehensive measure is median real person income. This, it turns out, has risen substantially since 1974 — though at a slower pace than in the past decades. If the consumer price index is used as the inflation measure, real income has gone up by about a third. If personal consumption expenditure inflation — which covers more goods and takes greater account of changes in consumption habits — is used instead, the rise is more than 40 percent:

The median American’s income fell in the late 1970s, then began a steady multidecade rise, interrupted by recessions in the early 1990s and early 2000s. In the 2000s, incomes began to stagnate, then took a disastrous beating during the Great Recession. But the recovery beginning in 2013 was robust, and by 2016 income was at a record high.

Personal income looks at individual adults. But other measures, such as median family income, tell the same story of a slow and bumpy rise.

What explains the difference between wages and income? Two things. First, wages aren’t the only way Americans make money in the market. Income from assets, like retirement accounts and pensions, is increasingly important, as are nonwage compensation like employer contributions to retirement accounts. Second, the income numbers include government transfers, which have shifted more and more income from rich Americans to those who earn less in the market. These factors are all bigger than in the 1970s:

Increased redistribution has been helping the poor as well as the middle class. Recent calculations by the Center on Budget and Policy Priorities show that child poverty in the U.S. has fallen to record lows once government assistance is taken into account.”

I would add to Smith’s account the following:

Consider the basic measure of a society’s affluence, GDP per capita. Per capita GDP in the US rose by 111 percent between 1947 and 1979. Between 1979 and 2007 (the last business cycle peak) growth was slower, but per capita GDP still rose by 67 percent over the time period . Obviously, the US became a much richer society over that time period, despite the slower growth.

Of course, this growth has been very unequally distributed, so the effect of this growth on living standards has been much more modest than that suggested by the substantial increase in GDP per capita. The starkest measure of this are the figures for growth of family income from the Census Current Population Survey (CPS). In the 1947-79 period, median family income went up 113 percent, closely matching the gain in GDP per capita over the time period. But in the 1979-2007 period, median family income grew from around $56,000 to $66,000 (2011 dollars), a gain of only 18 percent . Obviously, this lags far behind the growth of GDP per capita over the same time period. On the other hand, it is a gain of nearly a fifth—modest in comparative terms but not nothing and certainly not backsliding.

Moreover, the CPS data do not take into account the changing size of households, the value of non-cash benefits (food stamps, employer-provided health insurance, etc) and changes in the tax structure. Thus—and there are endless arguments about this among economists —the CPS data may underestimate the gain in living standards over time. Indeed, once all that is taken into account, the Congressional Budget Office (CBO) found that real (inflation-adjusted) after- tax income for the median household grew 50 percent between 1979 and 2007. Again, even this figure lags behind the growth of GDP per capita and is short measure compared to the 314 percent increase for the top 1 percent—but it is far from nothing. Even if one splits the difference between the CPS and CBO figures—in effect, assuming some of the CBO income is not as important as the unadjusted cash income measured by CPS—that would still give median income growth of 34 percent between 1979 and 2007. This is disappointing by historical standards but is far from the miserable picture embraced by many on the left. As the Pew Research Center notes, 84 percent of today’s adults have family incomes above what their parents had at similar ages .

Also lost in the standard tale of middle class decline is the fact that life cycle improvements in living standards have not been repealed by the relatively poor post-1979 environment. That is, it is still the case that as people age, they and their families typically get substantially better off. For example, economist Stephen Rose studied the same individuals as captured by the longitudinal Panel Survey of Income Dynamics and found that 20-31 year olds in 1979 experienced a median growth rate of 56 percent in their income as they aged to 48-59 by 2007.

Speaking of the middle class, this can be another source of definitional dispute between researchers. It is quite possible, for example, for the middle class under some definitions to become smaller even as there is considerable upward mobility from the middle class. This is demonstrated by a 2015 report from the Pew Research Center . According to Pew’s definition of the middle class—those with size-adjusted household incomes between two-thirds to double the median—the middle class shrank from 61 percent of adults to 50 percent in the 1971-2015 period. However, most of that shrinkage was due an increase in the share of adults who were in the upper middle or highest classes (up 7 points) rather than an increase in the share of adults who were in the lower middle or lowest classes (up 4 points). So the middle class, under their definition did shrink, but primarily because of upward, not downward, mobility.

Another excessively gloomy claim about the last several decades is that middle class jobs are disappearing and being replaced by “McJobs”. However, this view equates the decline of low skill, relatively well-paid jobs like those in manufacturing—which has been going on since 1948–to an overall decline in middle class jobs, which is not merited. The middle class jobs of today are in the growth areas of offices and high skill services. These two areas of the economy now provide 64 percent of all jobs and have expanded more as a share of jobs since 1967 than manufacturing and related jobs have declined. Thus, middle class jobs are not disappearing but have rather have moved to different sectors that require higher levels of education and cognitive training.

When thinking about progress in living standards it is also important to keep in mind the ways life has improved for most Americans that are not reflected in income or jobs data . For example, American life expectancy has gone up 5 years since 1979. Homes are far bigger (median new home size has risen from 1600 to 2600 square feet since 1979) and more well-appointed; food and clothing are cheaper and take up a smaller proportion of family budgets; cars are safer and get better gas mileage; access to travel and leisure, including foreign travel, has gone up; and device-enabled connection to the internet has brought the typical American into contact with a universe of information and entertainment that was literally unthinkable 30 or 40 years ago.

That’s progress. Now what we need is more of it–and faster please.