Why Donald Trump can’t create 25m jobs

By H.C | WASHINGTON, DC

DONALD TRUMP promises to deliver 25m jobs over the next decade, 18m more than is forecast today. In our print edition, we have consistently argued that this is infeasible, because America is ageing rapidly:

Arithmetic suggests this pledge is fanciful: even if the labour-force participation of 25- to 54-year-olds returns to its record high, only 4.3m new workers will appear by 2024.

But at Free Lunch, a blog at the Financial Times, the excellent Martin Sandbu disagrees:

The US is seriously underperforming relative to its own record on employment. The employment-to-population ratio for 25 to 54-year-olds (which should not be affected by ageing or college attendance) is 78.1 per cent, almost four points below the record set in the last year of the Bill Clinton presidency. The ratio for 16 to 24-year-olds is 10 points down. Restoring those ratios should add about 8m-9m jobs or 6 per cent of the current total.

Say that once-off upward shift of the share of the population work takes place over a decade. If the underlying labour force in addition expands as fast as the adult population growth rate — currently just over 1 per cent per year — that amounts to the labour supply growing at perhaps 1.6 per cent a year, which adds up to Trump’s 25m jobs over a decade.

Who is right? Over the next decade, much of America’s population growth will be in the 65-74 age bracket, which is set to swell by over 35%. That is why Free Lunch's exercise, which assumes that the labour force can grow as fast as the population does, sees Mr Trump hitting his 25m target, whereas ours, which focuses on prime-age workers, does not.

A deeper dig into the demographic projections reveals how outlandish the Free Lunch assumption is. I divided the population aged over 16 into eight age brackets, shown in the chart. The Bureau of Labour Statistics forecasts population growth in each bracket between 2014 and 2024, which is close enough for our purposes. Applying those growth rates to the November 2016 numbers, we can project the population in each age bracket in a decade. Plug in labour force participation rates and you can see how many jobs Mr Trump might create.

The answer is not encouraging. If Mr Trump returns the participation rate in each age-bracket to its level in 1998—when, as Free Lunch notes, the prime-age participation rate peaked—the labour force will grow by only 8.3m workers by 2026. In this scenario, the 16-64 participation rate would hit almost 76%, exceeding its pre-recession level of 75%. It would have grown by 3 percentage points over a decade, only slightly less than the 3.5 point growth Japan, Free Lunch’s favoured benchmark, managed from 2000-2015. Despite all this, Mr Trump would create only perhaps a third of the jobs he promises.

Perhaps, though, this is too pessimistic. In some age brackets—notably, over 65s—participation is higher now than it was in 1998. What if Mr Trump returned participation in every age bracket to its record high? For instance, the record participation for 16- to- 19-year olds is 59.3%, set in August 1978. For 35- to 44-year olds, it is 85.7%, reached in February 1990 and August 1997. And for 65- to 74-year-olds it is 27.7%, set in May 2013.

The results still show Mr Trump missing his target. Even if Mr Trump makes participation not just great, but the greatest it has ever been, and not just overall, but in every specific age bracket, he would unearth only about 18.6m workers. Assuming unemployment also falls to 4%, lower than it is today (and lower than almost all forecasts of the sustainable rate) he would create a little shy of 19m jobs, missing his target by a quarter. Making up the shortfall with workers aged 16-64 would require their participation rate to reach about 81.5%, matching that of Sweden, which has expansive child care subsidies. The 8.5 point growth this would require is well above what America has ever achieved before over a decade (5.5 points in the ten years to 1981, when women were piling into the workforce).

There simply aren’t the workers to fill 25m new jobs, thanks to the ageing population. (There definitely won’t be after an immigration crackdown.) Inflation would take off long before demand-side stimulus created those jobs, and supply-side reform would not be anywhere near fast enough. When people say that America’s growth rate has fallen partly because of demographics, they aren’t just making excuses. The financial crisis put enough people out of work to allow Barack Obama to oversee job creation of 2.5m a year. But that cannot continue for the next decade.