Japanese Demographics: When Do Things Get Bad?

Yesterday I appeared on the popular Breaking Points podcast to discuss an essay that I wrote a few weeks ago on demographic change and intergenerational conflict. Since then, I have had quite a few people sign up to the Substack. Thank you to those people, and I hope you enjoy it.

Anyway, I thought that it would be interesting — especially to the new sign-ups — to do a post on demographic change. I realised yesterday that I did not really have an answer to the question '“when do we start to feel the pressure of this conflict?” — not because either of the hosts asked the question, but when I reflected on it afterwards.

So, let’s take the case of Japan — which is likely one of the first countries (together with South Korea) that will start to experience the pressure of demographic change. First off, here is the raw demographics data on Japan that I pulled from the data archives and pieced together.

It’s pretty hard to draw any conclusions from that, so let’s break this down into ‘workers’ aged 15-64 and ‘dependents’ aged either 0-14 or 65+. Doing this we can also construct what economists call a ‘dependency ratio’.

What immediately stands out here is that the situation does not look that bad. It looks like Japan has been here before; it’s dependency ratio before WWII was actually higher than it is today. But this is misleading. Why? Because most of those dependents were children.

Recall that my argument is not about dependency per se, but rather about the increase in the elderly population relative to the working age population. This is different from having lots of child-dependents for a number of reasons. Here are just a few:

• Since having children is a choice, parents do not tend to begrudge the children the money that they spend on them.

• Parents still have control over resources when children are the dependents which does not result in the loss of control over these resources when elderly people are the dependents.

• Parents know that children will one day look after them when they are older, so there is a two-way trade — unlike when the main dependents are all older.

• Children may require resources — although they require less resources than elderly people — but they do not tend to own property and assets.

There are many other reasons why I would not expect populations with lots of children to generate intergenerational tensions. But the above should be enough to make the point for now.

Okay, so let’s build an alternative dependency ratio. I call this the ‘elderly dependency ratio’ and it should capture the dynamics that we are interested in.

Now we start to see how unprecedented the present situation is. The elderly typically only made up around 5% of the population. Notice how stable this percentage is. It looks like a natural percentage — something rare in historical social statistics. In 2015 in Japan, however, the elderly make up almost 30% of the population.

What is more, the working age population have also taken a dive. This is leading to our elderly dependency ratio rising from a historically normal reading of around 0.1 to a reading today four-and-a-half times higher at 0.45.

Yet the situation in Japan has not yet completely collapsed. As I discussed in the interview, Japan has seen stagnation in this period. But there is not yet any political and social turmoil. And, if we are honest, the economic situation is not catastrophic.

Before we run some models to see when there might be a tipping point, we should note that Japan is a very different place to the West. As I said in the interview, young Japanese are willing to accept wage stagnation in a way I do not believe Westerners would. With that said, let us continue on with the Japanese statistics.

In this chart I have used simple linear forecasting techniques to project post-2000 demographic trends in Japan into the future.

So, when does the situation become intolerable? By the mid-2030s the elderly will make up nearly 40% of the population, while workers will make up just 50%. That does not seem good. But things get really interesting in the late-2040s when the number of elderly outstrips the number of workers completely. I would consider that a ‘conservative’ estimate of a potential tipping point.

Where are Western countries relative to Japan? Given different birth rates, a simple comprison is unlikely to be totally reliable. But it should be roughly reliable. As of today, 16.5% of the population in the United States are over 65. So, the United States on this measure is around 20 years behind Japan. In Europe, over 20% of the population are over-65. So, that puts Europe around 13 years behind Japan. Although, again, I expect Westerners to be less tolerant of these trends than the Japanese — so, the tipping point in those countries is likely sooner.

In the Breaking Points interview, I said that Gen Xers and Millenials will be the first generations to experience the real effects of these demographic changes. According to this analysis, things will get bad anywhere between 30 and 50 years in the future for the developed economies. I think that this analysis supports that guess on my part.