Monday, June 14, 2021

60. Things I Can't Predict ⸺ Demographic Doom Podcast Transcript

Below is the transcript for my Demographic Doom Podcast episode #60, released on 15 June 2021. The "home page" for this episode—with annotations, links, corrections and a place for comments—is the YouTube version (15 minutes). The audio version is housed at Podbean and is available on most major podcast platforms, including iTunes and Google Podcasts. The main website for this project is Twitter: @DemographicDoom. Glenn Campbell home page: See bottom for notes on this transcript and how it was generated.

This is the original script. It has not yet been reconciled with the final broadcast.

I'm Glenn Campbell. I call myself a demographic philosopher. I'm looking at life and trying to predict the future through the lens of demography, or the study of human populations.

In this episode, I want to talk about the things I don't know—that is, the things that demographics and macroeconomics can not tell us—as illustrated by all the things that have happened over the past few years that I would have never predicted. 

Of course, I couldn't have predicted the pandemic. It wasn't even on my radar until late January 2020. If you were an epidemiologist, however, you could have predicted that something like this was destined to happen; you just couldn't say when. Demographics and macroeconomics are sort of like that: You can predict that bad things will happen, but you don't know when. 

Epidemiology is a little different, in that you're just waiting for a random mutation or a random jump of a virus between species. The risk has always been there, but it doesn't change much with time. In demographics and macroeconomics, the risks build up over time until things have to start breaking. It's like stretching a rubber band too far: The more you stretch it, the more likely it will break. We know, for example, that there is going to be a worldwide pensions crisis. This is the natural result of too many old people demanding their retirement benefits and not enough young workers paying into the system. I don't know what form this crisis will take when things come to a head, but you know there will be a day of reckoning—or maybe a few decades of reckoning.

There is a saying in economics: "Anything that can't go on forever will eventually end." This is called Stein's Law, from economist Herbert Stein. What he actually said was, "If something cannot go on forever, it will stop." 

Our current demographics and macroeconomics are filled with unsustainable systems that have to stop sooner or later. For example, a government can't keep spending more money than it collects in taxes, printing new money to make up the difference. That's what's happening right now and has been happening for years. Although nothing really bad seems to be happening at this moment, in mid-June 2021, you know it will.

A counterpoint to Stein's law is something I call "Keynes Curse" because it is often attributed to economist John Maynard Keyes, even though there's no evidence he actually said it. Keynes Curse says that "Markets can stay irrational longer than you can stay solvent." In other words, unsustainabilities can go on for a very long time. Anyone who has shorted the market or shorted Tesla stock lately has learned this the hard way. You can know for a fact that a system will fail, but if you don't know when and in what order, this information is pretty much useless to you as an investor.

So I can tell you with great confidence that the US monetary system will fail, but I can't tell you when. There's a fairly limited range of ways the system will break. There can be a massive asset crash that wipes out everyone's investments. There can be hyperinflation, which leaves nominal assets values in place but wipes out the value of the dollar. I can say with great confidence that middle-aged workers who think they have a solid retirement plan will find that it's not there when they arrive, either through inflation or an asset collapse. I can further predict that as some point in the future, the US government will either disintegrate or be forced to live within its means—which itself implies a lot of disintegration.

All of these things are likely to happen sooner or later, because "Anything that can't go on forever will end." I just don't know when or in what sequence or what the secondary damage will be. I just know the ship will sink.

The most serious failing of demographics and macroeconomics is that they can't predict human politics or human behavior. They may be able to predict when social tensions will be high, but they can't predict specifically how those tensions will be expressed.

For example, the hyperinflation of Weimar Germany was a fairly predictable macroeconomic event. If a government keeps printing money without restraint, eventually that money will become worthless. You could have also predicted with great certainty that the hyperinflation would lead to a lot of human misery and popular discontent. What no one could have predicted, specifically, was the rise of Adolf Hitler and where he would lead his country. While you can safely say high economic tensions tend to lead to radical populist movements, you can't say what form they will take.

To me, the great wake-up call to our economic tension was the election of Donald Trump in 2016. This was a pivotal event in my life—and maybe in yours—when I realized that my rational theories of the world didn't work anymore. You can quibble that he didn't win the popular vote, but the fact remains that roughly half the country thought that electing an obvious charlatan was a good idea. The election of 2020 was another harrowing event I never would have predicted. Even after four years of obvious incompetence, roughly half the US population was still willing to reelect him.

The trauma of 2016 is what launched my interest in demography and macroeconomics, but it also forced me to define the difference between what I know and don't know. I know a massive economic collapse is coming, but I can't say how human politics and human psychology will react to that event. I suspect that human politics will eventually trump macroeconomics. From our perspective today, the hyperinflation of Weimer Germany seems almost trivial compared to the World War that followed. That's the sort of thing I can't predict. I can't only live with the fear that another, more efficient Donald Trump could arise at any time.

The pandemic may have been an unpredictable Black Swan, in its specific appearance, but in general, you can predict that Black Swans will happen. They always have. You think you have things all worked out, and then some bolt of lightning hits you out of the blue. The effect of these Black Swans is usually to accelerate destructive processes that were already in play. Government debt, for example, was already a crisis before the pandemic hit, and now it's many times worse. 

A few months or years from now, there is sure to be another Black Swan, and another and another. Although you don't know what each one will be, you know they will happen, and they will accelerate unsustainabilities that were already under way. You just don't know how.

Once I knew that a pandemic was brewing in China, little about how it unfolded was surprising to me. In early February 2020, I remember telling my elderly father that he might not be seeing me for a while because his retirement village would be locked down, and that's exactly what happened. It was easy to predict that the lockdowns would be Draconian and indiscriminate, because government response to a crisis is usually ham-handed.

You can hear my own real-time reaction to the pandemic in the podcast episodes I recorded in the early days, starting on February 3, 2020. The only significant error I'll admit to is that I thought fomites were a source of transmission. In other words: I thought you could get the disease if someone who had it touched a doorknob then you touched it and touched your own face. That turns out not to be true. Most transmission is through the air—by breathing in droplets breathed out by someone with the disease—yet we still have hand sanitizer everywhere. I think I was pretty accurate, however, in predicting the worldwide scope of the epidemic and its devastating effects on the economy.

There were three areas, however, in which I was way off the mark....

My first failure is that I never in a million years would have predicted that the pandemic would result in a rise in stocks and other assets markets. On the contrary, I predicted a devastating crash, which I thought was primed to happen long before the pandemic. It started to happen in March 2020, but then markets reversed course, surged to all-time highs and stayed there. This was a complete disconnect from the real economy, which was on life support.

The second thing I never would have predicted were race riots in the midst of a pandemic. Some cities like Kenosha, Wisconsin were utterly devastated. You can point to certain precipitating events, certain instances of police brutality, for example, but this is nothing new. Racial discrimination is nothing new, so why were these violent protests happening now—especially when gathering in large groups, breathing other people's air, was potentially devastating?

The third thing I never would have predicted were labor shortages in the midst of an unemployment crisis. If you drive down any commercial street in America right now, you see help-wanted signs everywhere, and the wages being offered are rising. The labor shortage is a significant burden to businesses right now, just as things are opening up.

In retrospect, I now have explanations for each of these three anomalies. I can record a whole podcast about any one of them. The point is, I never would have predicted three events before they happened, because neither demographics nor macroeconomics gave me adequate tools to do so. 

No science can predict human behavior or the exact events of history. It's just too complex, and there are too many wildcards. In Weimar Germany, you knew that social tensions were high and that they were destined to be expressed in some way, but you couldn't have predicted Hitler, just like you couldn't have predicted in 2015 that Trump would be President in 2017.

The best that demographics and macroeconomics can do is predict the future of the whole package that were are travelling in—the trajectory of our whole spacecraft. As an analogy, imagine all of the things that could happen on Earth in the next 50 years. It's silly to try to predict world events in 2071, because there are so many factors involved; yet, astronomers can tell you exactly where in space our planet will be in 2071. That much is completely predictable.

Demographics and macroeconomics can place boundaries on our predictions even if they can't predict specific events. In particular, they can place an upper limit on how much future economic growth there can be. The last quarter of the 20th Century saw astounding economic growth. We were building shopping malls, high-rise office towers and new housing developments. Today's demographics can reliably predict that the same growth will never again be repeated, because we just don't have the bodies. 

Macroeconomics can safely predict that tax revenue will never significantly grow, even though government debts have gone through the roof, and this means that one way or another government debt will fail. It is fair to say that most national governments will never again be solvent. They can only keep printing money until something catastrophic happens to make them stop.

The only thing demographics and macroeconomics can do is place a hard ceiling on long-term economic growth, and therefore on the sustainable value of assets and the payability of debt. 

Below that ceiling, anything is possible.


Written, recorded and edited by Glenn Campbell. For annotations, links and corrections, see the description on the video version of this podcast. You can also leave comments there. See here for all my podcast scripts on this blog.

The transcript above is based on the automatically generated YouTube transcript, corrected by me based on my memory of what I said. In general, I make only the minimal changes necessary for clarity. I have not re-checked the transcript below against the actual broadcast. Editing consisted mainly of inserting punctuation and paragraphs and removing repetitive words and phrases. Passages in bold text are ones I consider particularly quotable. Items in [square brackets] are minor grammatical corrections. Items in {curly brackets} are factual corrections or amplifications. —Glenn Campbell

{Transcript backed up to email: 
{Visual version of this script backed up to Twitter on: 
{Video posted on FB album/DD page:
{Video Description backed up by email: 
{Video description backed up on Twitter: 
{Video/audio files copied to permanent backup: