US job creation has been declining since April 2000 and is now in freefall.
Monday, December 6, 2010 at 03:57PM US adult employment as a fraction of all adults reached an all-time high of 66.3 percent in March 2000. Since then, there has been an accelerating downward trend, and the ratio has recently dropped below 61%, a level we have not seen in 25 years.
This ten-plus year decline was not caused by changing demographics. It started when the earliest baby boomers were just turning 55. Moreover, the fraction of those age 65+ who are employed increased from 12.7% in March 2000 to 16.3% in November 2010, according to the same BLS database. Our demographic crisis lies ahead, as seniors are projected to increase from 22% of the population today to 38% by 2030 (slide 6), and of course that will depress the employment to population ratio even more.
The causes were a bursting of the internet bubble initially but more profoundly the offshoring of jobs. The NASDAQ index closed at an all-time high of 5,048.62 on March 10, 2000. A year later, dot-com companies were shrinking and failing right and left, the NASDAQ index was below 2,000, and the US was officially in a recession. The employment-population ratio naturally declined as it has in every recession, but then it did not rebound as it always had in the past. Our trade deficit, only -1% in 1997, exploded to -6% by 2005—a 5% negative shift away from domestic production with domestic labor. To a first approximation, offshoring 5% of GDP eliminated about 5% of the 131 million jobs that existed at the end of 1997, say 6.6 million direct jobs plus perhaps millions more indirectly. (Automation also eliminates jobs, but we have had automation continuously since WWII and have managed until 2000 to have job growth despite that.) As the chart clearly shows, the creation of domestic jobs to replace offshored jobs fell behind population growth, and is accelerating in the wrong direction.
What are the implications?
Except for the very wealthy, employment (including self-employment) is by far everybody's most important source of income. Without employment, most of us would sooner or later become destitute.
Incomes from employment support not only employees but also, typically, family members and, inevitably, government through taxes. Unemployment, therefore, starves families and reduces government revenues at the very time that the demand for government safety net services is increasing.
Our economy is lagging far behind its potential. We know that at least 66% of US adults want to work because that's the fraction that was working a mere decade ago. Yet now only 61% are working. If we could wave a wand and restore the employment-population ratio to the 2000 level, something like 11 million more Americans would have jobs and incomes, and GDP would be 8% greater.
Full employment would eliminate some of our fiscal woes and greatly ameliorate others. For example, during the late 1990s, when employment and real wages were both rising modestly, the Social Security trust fund was getting healthier every year—if that trend had continued, we could be increasing benefits and/or reducing payroll taxes instead of wringing our hands about doing the opposite. State and local governments could have balanced budgets without tax rate increases. And so on. If real median wages were also rising, the American Dream could be resurrected.
A continuing decline in the employment-population ratio, especially if accompanied by continuing decline of average middle-class earnings, would be an unmitigated disaster for America. Every time a family, business, or government tightens its belt a notch, the economy shrinks until all the belts have to be tightened again and more jobs are eliminated—and belts are tightened again. Money becomes unavailable for retirement savings, education, culture, environmental protection, or anything else that is not an urgent necessity. Hope would disappear, and our political system could develop dangerous fissures.
And yet hardly anybody inside the DC Beltway talks about how to achieve economic growth and full employment, or even the desirability of doing so. Nobody proposes any policies that are different from what we've been doing in the previous post-war recessions, or from the normal-times policies of the GOP since the 1970s and both parties since 1992. Why the bipartisan passivity about this? Do they love us less than the Chinese Communist Party leadership loves its citizens?
Skeptic
In response to the comment by PeiperTX44, the post-war growth in the employment-population ratio was For Women Only. For men it declined from 86% to 66% today, while for women it rose from 30% in 1948 to 59% in 2000 when it also began to trend downward. As I and others have pointed out, middle class wages stagnated after 1973 so household and family incomes were able to grow only by adding another wage earner, and perhaps that's PeiperTX44's point--for middle class Americans the problems started long before 2000. Eventually that coping strategy played out and middle class household and family incomes also stagnated. On the post-2000 employment-population ratio trend, middle class household and family incomes will decline at an accelerating pace.
Skeptic
Mike Konczal at Rortybomb has a graph here of the declining employment-population ratios for young people. In 2000, 72% of 20-24 year-olds were employed. Now it's 60%. He calls it a "generational crisis."
Skeptic
From the St. Louis Fed is this graph of an index of total hours worked. If this were adjusted for continuing population growth, it would look like the graph that started this post. Hat tip to Mark Thoma.
Skeptic
According to the St. Louis Fed, employment in all levels of government as a percent of population nearly doubled from 1948 to 1977 and then leveled off. It is now declining.
Skeptic
Brad Delong thinks the employment population ratio important enough that it is one of three graphs at the top of his home page. [UPDATE 2/23/11: Those charts are no longer there.] His plot, from Economagic, appears to be for people 16+ years old, whereas I used 20+.
Skeptic
If you want to remain unperturbed by the employment/population ratio, this is how you do it: Assume a linear trend from 1967, as spencer did at Angry Bear.
ADDED 6/28/2011: You can amuse yourself by extrapolating the straight line and estimating the year in which it will exceed 100%, i.e., more people working than there are people.
Skeptic
Paul Krugman has posted a graph showing that the ratio of Americans age 25-54 in employment has been declining since 2000, while in France that ratio is still increasing.
Reader Comments (2)
Before draw any end-of-the-world conclusions, as you do, I would want to know why the employment-population ration increased so much between 1948 and 1998. I am sure that then, it might have been interpreted as an evil too, along the following lines: The capitalist system is impoverishing everybody, therefore, when families needed only 1 pay check to be able to make it, they now need 2, and people are entering the work force not because they want to but because they have to.
I hope you get my drift.
Trade, immigration, labor-saving technologies are all pushing real hourly wages down. If they refuse to go down unemployment in its various forms is the result. And if they do go down a falling standard of living for working people is the result. Tariffs, immigration restrictions, a shorter standard workweek are all calculated to raise real hourly wages. Alternatively a tax on capital to subsidize wages (or other redistributive schemes) would have the same effect. All of the above would be even better. But if we do nothing we are headed for a racially stratified class society. In fact we are already there. Sad.