by Dean Baker
Many folks have made a good living for themselves by telling us that the United States, China, or some other country has to worry about running out of people. While I always enjoy ostensibly well-educated people putting their ignorance on display, it is unfortunate when anyone takes this nonsense seriously.
The latest silliness in this vein is a lengthy Washington Post article usefully headlined, “China’s Demographic Crisis Means It is Going to Run Out of Workers.” It gives us the scary projections in the text of the piece as well as almost unreadable graphics. (Jeff Bezos is trying to save money on graphics editors.)
According to the article, in 1990 there were nine people of working age (ages 15 to 59) for every person over age 60. By 2023, that ratio had fallen to 3.4, and the really scary story is that in 2050 it will be down just 1.25 to 1. And for science fiction fans, in 2100 the ratio of working age people to those over 60 will be just 0.8 to 1. That’s really scary stuff.
Before spending too much time freaking out over the story for 2100, it is worth noting that the year 2000 projections for China’s population in 2025 proved not to be very accurate. It’s not likely that our ability to project demographics far into the future, in this case 75 years, has gotten much better. In other words, we should probably take the 2100 projection with more than the usual grain of salt.
But let’s focus on 2050. Let’s say that an average retiree consumes 80 percent as much as an average worker, and we’ll say that all retirees’ spending comes from taxes imposed on workers. In reality, they may have substantial savings to support themselves in retirement; but from the standpoint of the economy as a whole, it really does not matter how retirees finance their consumption, the economy still must be able to produce the goods and services consumed by the elderly.
We can set the output produced by an average worker in 2023 as being equal to 100, as shown in the graph below. If there are 3.4 workers for every retiree, in order to allow retirees to consume 80 percent as much as an average worker we would need a tax rate of roughly 19 percent. This puts the after-tax wage in 2023 at 81 percent of the before-tax wage.
These ratios get much worse if we see the Post’s story where the ratio of workers to retirees falls to 1.25 to 1 by 2050, but it’s not quite the horror story this article wants readers to believe. The reason is productivity growth; the economy can produce more output with each hour of work.
Productivity growth should not be an alien concept. We’ve been seeing it for hundreds of years. This is the story of AI and robots taking all the jobs. As a result of improvements in technology, we will be able to produce more output for each hour of work.
Productivity in China has been increasing rapidly: a rate of well over 5.0 percent annually since the pandemic. It was rising even more rapidly in the decades before. Assuming that productivity growth in China averages 4.5 percent annually between 2023 and 2050, the before-tax level of output in 2050 will be 328 percent of the 2023 level of output.
Since there are projected to be just 1.25 people of working age for each retiree it would take a tax of 39.0 percent to give each retiree 80 percent of the income of the working age population. That may sound like a steep tax, but since before-tax output is 328 percent of 2023 output, this would still leave an average worker with an income that is more than 200 percent of their 2023 before-tax wage and almost 250 percent of their 2023 after-tax wage. That doesn’t sound like much of a crisis.
This simple story does leave many questions unanswered. First, and most importantly, can China expect to see average productivity growth of 4.5 percent? Economists have a horrible track record in projecting productivity, but China has maintained a far more rapid pace of productivity growth for decades. Also, it is investing heavily in advanced technologies, so it doesn’t seem implausible that it will be able to sustain a rapid pace of productivity growth.
There is also the question of distribution. In the United States, ordinary workers have not gotten their share of the gains from growth. We can’t assume that ordinary workers in China will necessarily get their fair share either, but this is an issue of distribution, not demographics. It’s understandable that the beneficiaries of upward redistribution would prefer to focus people’s attention on demographics, sort of like focusing it on immigrants in the United States, but the demographics would not be the real problem in this story.
The simple story above also skips over some obvious points which make the 2050 situation more favorable. Part of the story of the demographic crisis is fewer children. Workers have to support not only retirees by also children. In 2023, there were just under 5 working age people for every child under age 15. This is projected to rise to 6 working age people for every retiree in 2050. While the total ratio of dependents to working age people still rises, the relatively small number of children projected for 2050 mitigates that effect.
The other demographic issue is far more important. These figures assume that people over age 60 do not work. That is the legal retirement age in China at present, but it is likely to rise in the decades ahead. Also, many older people will choose to work, even if they qualify for retirement benefits. In the United States, close to 57 percent of the people between the ages 60-64 are working, as do more than 30 percent of the people between the ages of 65-69.
By 2025, China’s living standards will be roughly the same as they are today in the United States. With people living longer (their life expectancy is already almost the same as in the US), and likely healthier, many older people will opt to work, further mitigating the nightmare the Washington Post is trying to sell to its readers.
It is also worth mentioning that a smaller population means fewer demands on infrastructure, which is not picked up in these calculations, and less strain on the environment. Less crowded parks, beaches, and museums also does not seem like a horror story.
The basic point that people should take away is that productivity growth swamps the impact of demographics. It’s true that we can’t take productivity growth for granted. Countries have gone long periods with little productivity growth, like the United States from 1973 to 1995 and again from 2005 to 2019. But it should be much easier to fix productivity growth than hectoring people into having more kids.
