Where have all the workers gone?

In a new research paper, three economists from Washington University in St. Louis ask the question employers have been asking for almost three years now: where are the workers? Looking at the labour market in the United States, Dain Lee, Jinhyeok Park and Yongseok Shin find what others have found: that participation rates are down — slightly — and haven't (as of November at least) recovered to pre-pandemic levels.

But they also find that hours worked are down. People still in the workforce are working a couple of dozen hours a year less than before and that relatively small change is enough to explain fully half the change in total hours worked in the U.S., with departures from the labour force (i.e., reduced participation) explaining the other half.

If you're an employer looking to get work done, whether you can't find bodies or whether the bodies you do find don't want to work quite as much as they used to are just two sides of the same problem. Either way, the labour market gets tighter.

The decline in labour force participation rates during the pandemic — the “Great Resignation” — has been widely noted but professors Lee, Park and Shin argue that in fact what's happened to “part rates,” as economists call them, is pretty much the continuation of a long-term trend. And it's not so much older workers taking early retirement, which has got a lot of attention lately, as it is younger, less-educated workers tending to participate less. Primarily males.

As for the reduction in hours, it's taking place at the other end of the work distribution, among higher-paid workaholics who have been reducing their hours from very high levels. “In 2019,” as the authors put it, “one had to work for at least 2,860 hours to rank in the top 10 per cent of working longest hours. In 2022, he needed to work `only' 2,600 hours to win this honour.” In terms of weekly hours, dividing those numbers by 52, we're talking 55 hours a week down to 50.

Why are these changes taking place? The researchers don't really say, except to stress that it's all likely voluntary: with the labour market so tight over the last few years, people who want to work have been able to work.

They speculate that the U.S. may be falling into line with some of its peers. In 2021, the average U.S. worker worked 1,791 hours, according to the OECD. By contrast the average worker put in 1,685 hours in this country, 1,607 in Japan, 1,497 in the United Kingdom, 1,490 in France and just 1,349 in Germany. (How do the Germans produce all those Leopard tanks working so few hours?)

Research trying to explain this example of American exceptionalism has often focused on how low U.S. marginal tax rates make it worthwhile for American workers to spend so much time on the job. The research question now is whether changes in taxes and also “taxback” rates in various tax credits and social programs have altered workers' calculations.

To try to see what's going on in this country, I checked out what OECD.Stat has to say about Canadian labour force participation rates. In 2021, the latest year for which data are available, they were highest among males 35-39, of whom 93.4 per cent were in the workforce. In that age group, 82.9 per cent of women also participated. The lowest participation is among females 75 years of age and older, of whom only two per cent are in the workforce. In that age group, 6.3 per cent of males are still either working or looking for work.

Ottawa has a consistent policy of wanting labour force participation rates to be higher, presumably so there will be more GDP available to tax. Does it realize that fully 98 per cent of females in the 75-and-over cohort have dropped out?

In fact, as you might put it, the geezers are all right. Between 2010 and 2021, their participation rate went up — by half a percentage point among women and by 1.8 points among men. In most other age groups, participation rates were also up over that period, including among 50-54s, 55-59s, 60-64s and 65-69s. The only age groups where participation fell were at the bottom of the age distribution. The oldest group in which it fell? Men 25-34 and it only declined from 90.7 per cent to 90.4, so it's not as if they're totally slacking off.

But if labour force participation grew in virtually all age categories between 2010 and 2021 — if even people 75 and older are working more than they did a decade ago — how is it that overall labour force participation fell? Time is the answer, as it is to so many questions. Canada is a little older than it was in 2010. More of us have moved into age groups where, though we're working more than people in those age groups worked 11 years ago, we're working less than we did 11 years ago.

No matter how clever policy gets, it can't beat the clock.

PEOPLE WHO WANT TO WORK HAVE BEEN ABLE TO WORK.