Demographics

 When I see numbers showing the developing demographic problem, it is often stated as too few workers to support the elderly.  However, there will it seems also be fewer dependents on the other end, children.  Won't that mitigate it and shouldn't charts showing workers to retirees also show workers to total dependents?

Yes. In fact, demographers already have a measure for exactly that: the total dependency ratio, which counts both children and the elderly relative to the working-age population. It is defined as:
Population 15–64Population under 15+Population 65+​

and is usually expressed as dependents per 100 working-age people.

Your intuition is correct that a falling birth rate does offset some of the burden from population aging. Historically, many countries first experienced a declining dependency ratio because the number of children fell faster than the number of elderly rose. Only later, as the large middle-aged cohorts reached retirement, did the total dependency ratio begin to rise again.

The reason you often see charts showing only workers per retiree is that they are usually discussing a very specific issue: pensions, Social Security, Medicare, and similar programs. Children do not draw retirement pensions, so for those programs the retiree-to-worker ratio is the most relevant measure.

However, when looking at the broader economy, your criticism has merit. A society with:

    100 workers,

    40 children,

    10 retirees,

has 50 dependents in total.

If a few decades later it has:

    100 workers,

    20 children,

    30 retirees,

it still has 50 dependents in total. The composition has changed, but the raw dependency burden has not.

That said, there is a wrinkle. Children and retirees consume different kinds of resources. Children require schools, childcare, and eventually universities. Elderly people require pensions, healthcare, nursing care, and long-term care. In practice, the elderly tend to consume much more healthcare than children consume education, especially in wealthy countries. Thus a constant total dependency ratio can still mean rising government expenditures.

I would go even further: the traditional dependency ratio can itself be misleading. It assumes everyone aged 15–64 works and everyone outside that range does not. In reality, many people over 65 continue working, while many working-age adults are students, unemployed, disabled, or retired early. The World Bank and UN both note that dependency ratios are only rough proxies for actual economic dependency.

So your observation is a good one. When someone waves around a scary chart showing "workers per retiree," it is often worth asking, "What happened to the number of children?" Sometimes the answer is that much of the increase in elderly dependency is indeed offset by fewer young dependents. The more complete picture is usually the total dependency ratio, not the retiree ratio alone.

Yes—and that changes the question quite substantially.

If the question is "Will Japan, China, Korea, or Germany decline relative to some other country?", demographics can look rather ominous. But if the question is "Will people around the world become more prosperous?", population aging is a much less straightforward problem.

The crucial variable is output per person, not simply the number of workers. If 100 workers in 2050 can produce what 150 workers produce today, having fewer workers isn't necessarily impoverishing. Automation, AI, robotics, better medicine, energy technology, and accumulated capital can all substitute for some of the labor that demographics take away.

And there is an important benefit on the other side of the demographic transition that gets rather little attention: fewer children mean enormous resources that no longer have to be devoted to raising the next generation. Those resources don't disappear. They can become savings, investment, housing, healthcare, leisure, education per child, or simply higher consumption.

So I would distinguish two things:

The demographic transition can make societies older without necessarily making them poorer.

What is difficult is the distribution of the resources. A society with 30 workers supporting 10 elderly people is not necessarily poorer than one with 40 workers supporting 20 children and 5 elderly people. It depends enormously on productivity and on how the society organizes pensions, healthcare, taxation, capital ownership, and work.

There is also a rather hopeful global feature. Population decline in one country isn't necessarily a global loss. If Korea has fewer workers, Korean factories can become more automated. If Germany has fewer workers, capital can become more productive. If China has fewer workers, its enormous accumulated stock of machinery and infrastructure doesn't suddenly vanish. And people elsewhere can continue becoming more productive.

Indeed, from the standpoint of world prosperity, I think the really important question over the next century may be less "How many workers will there be?" than:

    How much useful output can humanity produce with each worker and each unit of capital?

If productivity rises sufficiently rapidly, humanity could have fewer people, more elderly people, fewer children, and substantially higher living standards simultaneously.

That is a much more interesting possibility than the usual "demographic collapse" headlines suggest.

Another factor is advancing medicine means not just longer lives but longer working lives.  I'm 83 and if I had to I could hold any of a number of jobs.

Absolutely. And that is one of the weaknesses of the conventional "working-age = 15–64" calculation: 65 is an increasingly arbitrary boundary.

If people remain healthy and capable at 70, 75, or even 80, then treating them as economically dependent merely because they crossed an administrative age threshold exaggerates the demographic burden.

Your own example illustrates the point nicely. At 83, you aren't simply an 83-year-old "dependent" in the economic sense. You have accumulated knowledge, experience, judgment, and skills, and you could still contribute economically if circumstances required it. That's a very different thing from assuming that an 83-year-old necessarily requires someone else to support him.

There's an interesting feedback mechanism here. Suppose medicine and healthier lifestyles extend healthy life expectancy by ten years while conventional retirement ages remain unchanged. Society suddenly has a large reservoir of people who are perfectly capable of working but are classified statistically as retirees. If retirement ages eventually adjust, the apparent worker-to-retiree ratio improves substantially.

And it needn't mean forcing everybody to work until 75. Even part-time or intermittent work matters. Someone working 15 hours a week isn't equivalent to a full-time 35-year-old, but they're also not economically equivalent to someone producing nothing.

There's another effect that I think is even more important: the nature of work is changing. Physical deterioration matters enormously less for a desk job, consulting, teaching, writing, advising, operating software, or many other occupations than it did for farming or factory work. A society that moves from physical labor toward knowledge and automated production can make much better use of older people.

So instead of:

workers → retirees → dependents

we increasingly need something more like:

productive people → less productive people → people requiring substantial support.

And those categories don't correspond neatly to chronological age.

That could substantially soften the demographic problem. A country with declining numbers of 25-year-olds isn't necessarily facing a proportional decline in its productive capacity if its 55-, 65-, and 75-year-olds remain healthy, its machines become more productive, and its technology allows each worker to accomplish more.

In fact, there's a slightly amusing demographic paradox here: medicine may be turning "aging" from a period of dependency into a much longer period of potential productivity, while simultaneously extending the period during which people consume healthcare. The latter gets most of the attention because it's expensive; the former is harder to put into a simple chart.

And I suspect the conventional demographic models will increasingly have to grapple with exactly the point you're making: chronological age is becoming a progressively poorer proxy for economic capacity.

Thinking about it, my investment decisions in the States is a form of job via allocation of capital, and of course the enterprises here I have provided capital for, although that is more a transfer than productive work.

Yes. And that's a useful distinction between labor and economic contribution.

Your investment decisions in the United States are not "a job" in the conventional employment-statistics sense, but they are certainly an economic activity. Capital has to be allocated somehow, and deciding which companies, projects, securities, or enterprises receive capital is part of the mechanism by which resources get directed toward productive uses.

There are really several different things going on:

Capital allocation: Your investment choices provide financing to businesses and projects that need capital. At the economy-wide level, this is productive even though you aren't personally operating the machinery or writing the software.

Ownership and risk-bearing: Investors absorb some of the risk of enterprise in return for the possibility of a return. That's economically useful too. Someone has to own the risk.

Transfer: When you put money into an already-existing enterprise here, you're right that the immediate transaction may simply transfer ownership of existing capital or money from one person to another. But that doesn't necessarily mean it's economically unproductive. The seller may then use the money for something productive, while your capital remains invested in the enterprise.

And there's an important point concerning our demographic discussion: capital can substitute for labor. An aging society with fewer workers can potentially compensate by having more capital per worker. A worker equipped with a computer, sophisticated machinery, software, robotics, and AI can produce vastly more than a worker with only hand tools.

That's why I think the phrase "not enough workers to support the elderly" can sometimes conceal the actual economic question. Workers aren't supporting retirees with their bare hands. They're producing things using accumulated capital, technology, infrastructure, and knowledge.

And capital itself doesn't retire at 65.

A factory built 30 years ago can keep producing. A software system doesn't have to retire. A patent doesn't draw Social Security. An investment portfolio can continue allocating capital whether its owner is 40 or 83.

So in your particular example, you're actually illustrating something rather important about an aging economy: an older person can remain economically consequential without being counted as a worker. You may not be standing at a machine producing something, but decisions about where your capital goes can affect what machines get built and what enterprises get financed.

The economic statistics are necessarily crude. "83-year-old = dependent" is a demographic convenience, not a description of an actual human being.