(The Wall Street Journal published on November 2nd a synopsis of this post. It was also featured on January 20th in the French weekly L’Express.)
Retraining the employed and the unemployed for higher value-added skills is now more important than simply adding to the number of jobs.
Coal and steel magnate Wilbur Ross, a senior policy advisor to the Trump campaign, has just made in the pages of the Wall Street Journal an economic prediction that looks mathematically unattainable.
Writing with business Professor Peter Navarro of UC – Irvine, Mr. Ross forecast that policies enacted by a President Trump would lead to the creation of 25 million new jobs, ostensibly over an eight year period:
Donald Trump will cut taxes, reduce regulation, unleash our abundant energy and eliminate our trade deficit through muscular trade negotiations that increase exports, reduce imports and eliminate cheating. These policies will double our economic growth rate, create 25 million new jobs, boost labor and capital incomes, generate trillions of additional tax revenues and reduce debt as a percentage of GDP.
To evaluate the 25 million figure, remember that new job creation during the booming Ronald Reagan and Bill Clinton two-term presidencies amounted to 16.1 million and 22.5 million, respectively. Given the growth of the population, you could say that a 25 million target compares reasonably to these figures. It is optimistic but, on the face of it, not outlandish.
That is, until you scrutinize the underlying demographics. A new job requires not only an open paid position but also a person to occupy this position. In order to have 25 million new jobs, we need 25 million people to fill these jobs. So does the US even have 25 million people who would be available in the next eight years? It doesn’t seem like we do.
Now as at any time, there are three main sources of new workers:
Increase in working-age population: During the 1980s decade (which includes the two Reagan terms), the population aged 20 to 64 grew by 18.5 million. During the 1990s (including the two Clinton terms), it grew by 19.1 million. In the coming decade ending in 2025 by contrast, it will only grow by a much smaller 2.6 million.
What explains this decline in growth for this segment? Simply put, there were many more new babies in the 1960s than in the 1910s, resulting in strong growth in the 1980s. But there were only a few more babies in the 2000s than in the 1950s, resulting in lower growth in the 2016-25 decade. Put another way, a rising number of boomers are turning 65 every year and exiting the 20-64 age bracket. This means that, unlike in the 1980s and 1990s, a large percentage of people going into these 25 million jobs will have to come from other sources.
Immigration: Assuming an annual influx of one million immigrants (green card holders), we estimate another 7 new million workers for the decade, and a prorated 5.6 million over eight years.
The idle and unemployed: The current official unemployment rate is hovering around 5%. However, the U6 measure of unemployment now stands at 9.7%, not far from its historical average. If we assume generously that U6 will drop by 3% towards its low of October 2000, there would be an additional 4.8 million people available for work.
Adding all these figures (and making adjustments from 10 to 8 years where needed), we see that there is still a shortfall of 12.5 million people, or half the 25 million needed to meet the new supply of jobs.
Keeping an open mind, we could speculate that an additional 12 or 13 million people, counting for example a large number of women and elderly, could decide to join or remain in the labor force. But this seems unlikely within the big economic boom described by Mr. Ross and Professor Navarro. If the economy will be doing that well, fewer spouses may choose to work and more people will retire early.
Finally, the next President could increase the number of immigrants to address the labor force shortfall. This decision would require doubling or tripling the number of visas and green cards awarded annually, a policy that runs against the grain of Trump campaign pledges.
To sum up, an optimistic level of job creation under the next President, whether Trump or Clinton, would be 12 to 15 million. But even this lower target will prove to be too ambitious if, as is widely anticipated, automation and the internet of things take over more job functions.
Further, because the marginal demand for jobs will be less than in the past, an effort to boost the supply looks not only ill-fated but also misdirected, like that of a general preparing to fight the last war. At this juncture of changing demographics and increased automation, it will be more important to upgrade jobs to higher value-added functions than to simply count the number of net new jobs. Bringing the old jobs back would in theory provide much needed relief to households that are struggling, but retraining these same workers for better jobs will ultimately lead to more favorable outcomes.
In this vein, investments in education and retraining seem more critical now than in the past. Rather than merely adding jobs, a more promising employment strategy for the next President would be to facilitate retraining programs for people who have not kept up with the economy because of outsourcing or other factors.
All this will probably be unconvincing to Mr. Ross and Professor Navarro who downplay the role of demographics in the economy and who believe that a sufficient amount of can-do spirit will overcome the facts of a hard-nosed analysis:
Some falsely assert that the U.S. and other developed countries have settled into a “new normal” of slower economic growth due to greater competition from developing countries and demographic changes beyond our control.
But to quote Mr. Trump’s running mate, Gov. Mike Pence, “People in Scranton know different. People in Fort Wayne know different.”
We shall see.
One clear fact remains however: the golden decades ending in 2005 were in part powered by a fast-growing population and a declining dependency ratio, two conditions that are now fading.