Against "Slightly Against Underpopulation Worries"

Slow population growth is a problem now, not decades from now

Yes, I am having some fun at Scott Alexander’s expense.

This is in response to his post: Slightly Against Underpopulation Worries, in which he pushes back against various people’s concerns re: global population trajectories in which we are likely to peak within the next century globally, and then global human population numbers dwindle…. (if we are so foolish as to extend current trends, but let me not go down that particular path.)

To keep this post focused, I want to address my concern about population trends, which are effects on public finance, particularly on pensions and Social Security-type systems.

Scott Alexander writes:

6. Age Pyramid Concerns Are Real, But Not Compatible With Technological Unemployment Concerns
As birth rates rise, you have many hard-working young people supporting a small number of retirees. As they fall, you have fewer young people and more older people who need support. This either burdens the young, or requires cuts in support for the elderly.

And yeah, to some degree this will happen. I think it will look less like an apocalypse and more like increasing effective retirement ages, but that will suck.

On the other hand, this is basically a complaint about a shortage of labor. And I notice it’s weird to be worried both that the future will be racked by labor shortages, and that we’ll suffer from technological unemployment and need to worry about universal basic income. You really have to choose one or the other. I’m pretty worried about technological unemployment myself.

Another way of saying “labor shortage” is “the value of labor relative to capital goes up”. Workers will be able to expect high salaries and good working conditions. Labor shortages are also periods of intense innovation for labor-saving devices (some historians blame the Industrial Revolution on unusually high wages in the England of the time).

And that’s it.

Who will pay for your retirement?

I’m not going to dig into Scott Alexander’s details, but if I remember correctly, he’s younger than me (I’m 48). He’s an American psychiatrist, I believe, in private practice and probably is expecting to be working for many more decades.

Maybe he’s not thought about how one lives in one’s retirement much, yet.

Maybe he’s expecting Social Security not to be around for him, or, if it is, that it will not be a large portion of his retirement income stream. Fine. Then I assume he has an investment portfolio of assets, which he hopes will accumulate in value, that he can turn into cash once he is no longer getting cash from paying clients. Either those assets will be realized into cash by selling them off to other people, or they will throw off cash flows, also coming from other people.

Note the importance of other people existing and having cash to exchange for those assets which they will value at that future point in time.

What if those people aren’t there?

Some assets require people to exist, like real estate. While some people like collecting houses all over the place, most of us are happy with one place to live in. Maybe some like one house, and one vacation place.

Some assets will lose value as there are fewer people. That’s just how they behave.

It’s okay to own these assets to save for retirement — it’s just that you may have less retirement income than you had hoped for.

Will labor productivity growth really be so high that you will be able to support yourself on what those assets can earn?

Who will pay for Social Security?

For many Americans, Social Security is a large portion of their retirement income, from payroll taxes from current workers. That requires current workers to exist.

And you only get a certain percentage of their labor value to fund your Social Security. Yes, the law can change, but politics is about the possible, and the current FICA cut for Social Security is the 6.2% an employee directly sees and 6.2% the employer also pays (and the self-employed also sees). That amount was last hiked 32 years ago, and was only 1% when it got started. Maybe it can be hiked again.

Before that rate will be raised though, Congress will most likely remove the Social Security tax cap:

If you are not familiar — there’s only so much of one’s wages which are hit with Social Security payroll taxes (there is no such cap for Medicare). That cap is used for payroll taxes… and also for earnings used for determining your Social Security benefit. For obvious reasons, there are limits to how much retirement benefits the federal government is willing to throw in on the behalf of high income folks.

It is less obvious (except for those of us who know why this was done) why the same cap is used for taxation purposes.

It is really unobvious why one would make a “donut hole” so that people like me would still get a few extra hundred dollars per year because there was still a SocSec tax cap… for most of us. I do know people who would go above that attachment point for re-implementing SocSec. I may one day explain this on the blog, but today is not that day!

Today is also not that day.

I will show some of my SocSec taxed wage history, though:

It’s not that interesting, in that all you will see is that I’ve been above the wage cap for most of my actuarial career (so you’re not seeing my full wages past a certain year — just the wage cap). That’s common among actuaries. So you will often hear actuaries being against removing the wage cap.

Well, I’m neutral on it. I am pro two options: just get rid of the payroll tax entirely and fund Social Security from income taxes (and get rid of federal withholding entirely, just to be a real bitch about it) OR get rid of the payroll tax cap and decouple FICA from people’s ultimate benefit.

Both options will annoy people a lot, and the first is a non-starter. I just throw that first one out there for conversation. I really am for it, but I do understand why few people will join me in that stance.

I think the second will actually happen, even if my fellow actuaries don’t like it. People never thought there would be a SALT cap, and here we are.

CBO projects deaths will outnumber births 2043 in U.S.

This projection does not offend my actuarial sensibilities as much as the global ones, mainly because this is over a few decades, and I think it accords with trends fairly well.

Here is Mish’s take:

Aging persons tend to spend less and travel less as the years pass. But they need more health care and it’s expensive.

Yet, there will be fewer people working and paying taxes to support each retiree.

Neither Social Security nor Medicare was created to withstand the cliff event that’s coming when all the boomers retire.

Once Boomers are gone from Congress It will be interesting to see what Millennials and Zoomers do with the mess they inherited.

Nobody cares about us Gen Xers! Boo hoo!

Yeah, well, the Millennials outnumber us (and wait til the Zoomers (my kids) shove you aside, Millennials. BWA HA HA HA)

But let’s look at that projection: CBO Demographic Outlook 2022 to 2052

Here’s the graph Mish pulled:

2043 is not that far away.

Note that the population still grows, as net immigration will increase the population even if there are more deaths than births. That’s how cities like London managed to grow for centuries even though its deaths far outnumbered births before the Industrial Age.

But let’s look at some of the components. Such as fertility:

I should show a more detailed breakout, such as births to those over age 35, or over age 40, because the trend compared to teen births, say, is fascinating. It used to be that peak birth ages were in the 20s, but now it’s in the 30s. I know they push it back only to 1980, but I have some data going back to pre-Baby Boom time, and it’s fascinating to see how it’s changed over the 20th century into the 21st.

For this projection, mortality is fairly boring:

Yes, many actuaries are doing similar projections. I’m not sure this is how it will go.

I’m not breaking out the migration assumption, because I think that’s the iffiest of the assumptions. You can look at the presentation yourself, on the 9th page of the pdf (and the page is numbered 5, because why number the title page and the table of contents, amirite?) The history was very volatile, which is why I think the projection is iffy.

Putting it together, this is how various age groups are projected:

So yes, “working age” (25-54, with 55-64 as an aging adjunct with various levels of disability) is the largest group, but the “retired age” (age 65+) is growing in the projection (thus, the growing number of deaths projected).

And then there’s the public pensions….

Just to plug it again, I recently took the Reason Foundation’s State Pension Plan Projection Tool for a spin:

You can play with the tool yourself here: Reason Foundation’s State Pension Plan Projection Tool

Here is their own post on their results: Unfunded public pension liabilities are forecast to rise to $1.3 trillion in 2022

The reason I bring this up is that there is a connection between these population projections and unfunded public pension liabilities. It’s not specifically their two-year projections I want to point out, but the historical record of funded ratios they’re showing.

Many of these public pensions are underfunded, even though they are putting in full contributions as calculated by the actuaries, because there is an underlying assumption that there is a growing taxpayer base that will be paying more later.

Yes, there are many pension systems that deliberately underpaid, but many made full contributions always and had decreasing funded ratios, even though investments did fine. Calpers had a long, bull market, and they never returned to the full-fundedness of 2001.

Back in 2019, before the pandemic (remember those days?), I wrote a series on whether public pensions were in crisis:

So many old-age programs assume a growing tax base — it’s not only public pensions running into trouble, but other programs such as Social Security and Medicare.

When I say something is in “crisis”, I mean that there is unavoidable disaster unless something changes.

This does not mean it’s panic time.

…..

Getting away from 100% funded is a “crisis” in public pensions because so many do not have good mechanisms to get back to that full funded level. So many plans were at 100% fundedness in 2001… and so many drifted away from that, even before the market drop in 2008.

Since 2008, many plans have failed to improve their situation. They’ve been going more into alternative assets. Some have learned the danger of pension obligation bonds, but others have issued new ones. Many have reduced their discount rates to value the pensions. Some have closed off old plans, and made smaller promises to new entrants.

But many, like Illinois and California, have barely adjusted their behavior.

NOT ACTING NOW WILL LEAD TO PANIC SITUATIONS LATER

If another large market drop occurs in the next decade (forget about next year), many public pension plans are going to be in an awful situation if they cannot adjust pension benefits downward for current retirees and participants. We’ve not had an official recession in over a decade, and not had much of a market “correction” yet.
….

I’m not saying that it’s time to panic (or that panicking will at all help) — but there will be panic if there’s a large market drop. It may make some feel good to blame Trump (or whatever politician is the target at the time), but it will not make pension plan participants whole.

….

There are lots of reasons that it is difficult to adjust public pensions, the biggest element being the political risk. Taxpayers are a diffuse group. The public employees seeing the threat to their benefits are a very interested group.

On top of the political problem, there are legal problems in making these adjustments — many states would require state constitutions be amended to allow for these changes.

I’m not saying making the change from a current system to a new, more stable and sustainable system would be easy. Or even simple.

I am saying that many public pensions will not be able to continue as they are, and the ones that are failing will likely not be able to continue on a pay-as-they-go basis. And there will not be a full bailout (but perhaps a partial one).

You can call that a crisis if you wish.

Again, that was in 2019 that I wrote that.

I could have written that in 2017 or 2014 or whenever. But in 2019, I had seen the end of a very long bull market, and the public pension funded ratios had stalled.

I knew many of these pensions – the pensioners, the politicians, et. al., assumed that future taxpayers would bail them out if the funds ran into trouble.

Not only was I skeptical that the taxpayers would bail them out, but I was also skeptical that the taxpayers would even be there.

I didn’t even have to think about hypotheticals — I saw the taxpayers disappear in Detroit; Prichard, Alabama; and Puerto Rico.

Some nasty things, things got worked out in various ways, but people had promises defaulted on. In Puerto Rico’s case, somebody (the U.S. federal government) came from outside to shower money on them, but for many of the ailing state funds, there will be no outside savior.

There were devastating effects in Prichard, which I keep pointing out — people didn’t get pension payments for over a year, and a few people committed suicide. In Detroit, people who were already retired got payments cut.

These are not small effects to the people who this happened to.

Now imagine that happening in a time of high inflation.

No longer hypothetical nor far away

So, yes, there are problems with slow-growth populations other than just a “shortage in labor”.

Scott Alexander may be dismissive of just that it sucks that the retirement age may rise a bit, but the unfunded liabilities of Social Security and public pensions alone are going to involve some very large adjustments in expectations in a very short time. It doesn’t require a shrinking population, either.

It just requires projections we already see, and it’s even after the two years of extra mortality decreasing the senior population a bit.

(Some people asked me about the impact of COVID on these pension liabilities, and whether it helped re: these unfunded liabilities. Of course, it reduced the liabilities a little, but not a large amount. Sorry, y’all.)

The Social Security Trust Fund will run out in about a decade. The problem is already here as the payroll tax revenue has become insufficient to support cash flows going out.

So yes, fewer people is a problem. Less growth is a problem right now. (Recession is a problem right now.)

To be sure, I don’t think Elon Musk having a few extra offspring is necessarily boosting the growth rate that much, but it surely couldn’t hurt.