The lost decade

From The Washington Post’s Wonkblog:

U.S. job growth surges in July

By Ana Swanson | August 4 at 8:34 AM

The U.S. economy added 209,000 jobs in July, according to government data released Friday morning, surpassing economists’ expectations and suggesting the economy continues to thrive after an extended streak of job gains in recent years.

The unemployment rate ticked down to 4.3 percent, compared with 4.4 percent in June, and wages rose by 2.5 percent from the year before to $26.36 in July.

“It was pretty solid across the board,” said Michael Feroli, chief U.S. economist at JP Morgan. “It suggests there is really no slowing in the momentum of the labor market.”

That first part was unremarkable, the kind of story one would expect the day the official unemployment numbers are released. The writer continued to note that the job growth numbers are very similar to those in 2016.  But further down was the part I found most significant:

July’s additions also signify a notable turning point for the U.S. economy: After accounting for shifts in population, the level of employment has returned to what it was at in November 2007, before the recession decimated the job market, according to research published Friday morning by the Brooking Institution’s Hamilton project.

“It does not mean there’s no slack in the economy, [or] that we’re at full employment. But it does mean the job losses from the great recession are behind us,” said Diane Whitmore Schanzenbach, one of the report’s authors.

And the Brooking Institution’s article:

The Closing of the Jobs Gap: A Decade of Recession and Recovery

by Diane Whitmore Schanzenbach, Ryan Nunn, Lauren Bauer, and Audrey Breitwieser | August 4, 2017

The Jobs Gap Has Closed

The Great Recession caused labor market devastation on a scale not seen for many decades. Millions of jobs were lost in the United States during 2008 and 2009, leaving the labor market with a hard road to recovery. Indeed, that recovery has required many years of job growth, and it was only in April 2014 that total employment reached its pre-recession level.

However, this milestone did not mark a return to pre-recession labor market conditions. Because the U.S. population is growing, simply reaching the previous number of jobs is not sufficient to return to pre-recession employment rates. At the same time, more baby boomers have entered retirement, somewhat offsetting the effects of population growth and reducing the number of jobs needed for a full economic recovery.

In order to accurately track the progress of the labor market recovery, The Hamilton Project developed a measure of labor market health—the “jobs gap”—that reflects changes in both the level and the demographic composition of the U.S. population (more details regarding the jobs gap methodology are provided in appendix A). Beginning in May of 2010, The Hamilton Project has calculated the number of jobs needed to return to the national employment rate prior to the Great Recession, accounting for population growth and aging.

In figure 2, we apply the jobs gap methodology to three other recent recessions: 2001, 1990, and 1981. Compared to these recessions, the jobs gap during the Great Recession was much larger and took years longer to close. The recessions of 1981 and 1990 involved smaller and briefer jobs gaps, with recovery to the demographically adjusted, pre-recession employment rate after 40 and 48 months, respectively. The 2001 recession saw a more gradual decline in jobs, and a slower recovery; the jobs gap from the 2001 recession did not close before the Great Recession started. The fact that the labor market was not necessarily at full strength at the start of the Great Recession is one reason the closing of the jobs gap does not necessarily signal an end to “slack” in the labor market: the under-used labor that could be profitably employed.

An Uneven Recovery

The labor market recoveries depicted in figures 1 and 2 reflect the overall experience of the entire United States. However, not all regions of the country or demographic groups experienced the same recovery—while some groups have reached and substantially exceeded their pre-recession employment rates, others have lagged behind. Because the payroll employment data (from the Current Employment Statistics survey) do not include information on demographic characteristics, we use individual-level Current Population Survey (CPS) data in these calculations. These data are less current than the payroll data—we use individual-level data through May 2017—and the growth in employment measured in the CPS is somewhat lower. Given these differences and the distinct features of the CPS data, we will now implement the jobs gap concept as an “employment rate gap,” defined as the difference between the demographically adjusted 2007 employment-to-population ratio and the actual employment-to-population ratio at a given point in time.

There’s a lot more at the original, but there is one huge point to be made here: since we never recovered fully from the 2001 recession — worsened in part by the September 11th attacks, the starting zero point for the 2007 recession was lower than it would have been had we fully recovered from that; even though we have, finally, made up the job losses from the 2007 recession, we are still behind where we ought to be, which means that there is still growth potential out there. Given that the 2007 recession began while we were still roughly 4½ million jobs behind full recovery from the 2001 recession, and the Hamilton Projects calculates that we have just now recovered the jobs from the 2007 recession, does than not mean we are still 5 million jobs, adjusting for population growth, behind where we should be, based on jobs prior to the 2001 recession?

But there’s more:

The employment rate gap recovery has been uneven in other respects: notably, women have outperformed men. Two male-dominated occupation groups—production and construction—were particularly hard hit during the Great Recession. Employment in these occupations remains low relative to other occupations, contributing to weaker employment growth for men over the last ten years. In addition, the employment rate of men aged 25 to 54 had been falling for several decades prior to the Great Recession, driven by forces that are still not entirely understood, but possibly contributing to the disparities between the employment trajectories of men and women.

Figure 4 shows the employment rate gap separately for men and women. The immediate employment loss from the recession was somewhat less severe for women, with the gap reaching a trough of -2.9 percentage points in 2011. By contrast, the employment rate gap for men reached a low point of -5.5 percentage points in 2010. Men have considerably more ground to make up than do women to regain their pre-recession employment rate: the gap for men stands at -1.6 percentage points, while it has closed entirely for women. However, it is important to note that men remain employed at a much higher rate than women, even with their relative decline over the past ten years: 65.7 percent of men and only 54.8 percent of women are employed.

One wonders how the gap for women could have reached zero, while for men it remains down 1.6 percentage points, yet the total gap is zero.

The authors claim that “forces that are still not entirely understood” have contributed to a falling employment rate for men for decades, but, the fact is that those forces are completely understood: both automation1 at home and the ever-increasing importation of manufactured goods have depressed the need for manufacturing employees, jobs which have primarily been held by men.  Whether the professional economists understand this, they can address themselves, but it speaks to why the white working class have abandoned the Democrats in recent elections.2

I wrote the title for this article, “The Lost Decade,” before the article itself. But, when we consider the unrecovered jobs from the 2001 recession, it might be more accurately — though less prosaically — The Lost Seventeen Years. With an average job creation rate thus far under President Trump of 179,000 per month, and still being 5,000,000 jobs behind where we should have been considering employment in 2000, it should take roughly 28 more months to close that gap, assuming that the job creation rate remains unchanged.3 Twenty-eight more months takes us until November of 2019, so this article might most accurately been entitled The Lost Two Decades.

We have noted that several recent coal mining operations have been started since Donald Trump became President, but even with those start-ups, automation means that there will be fewer coal mining jobs in those facilities. The President can try to set policies to encourage more American manufacturing, but that will depend entirely upon one thing: will American consumers choose to buy products made here? We are in our current economic situation precisely because consumers have not taken such choices.

  1. See George Jetson’s job: what if there aren’t any jobs in the future?
  2. The Democrats were the ones in power; if things do not get better under President Trump, we can count on a lot of those Trump voters to choose Democrats by the 2020 elections.
  3. We have previously noted that even if the stretch between the end of the last recession and the beginning of the next one equals ten years, the time of the longest gap in recent history, the next recession would begin by July of 2019.