Aging / shrinking populations and inflation; Creditor military action and sovereign credit risk

If a country’s population gets older, would it experience inflation or deflation?

Well, I guess it very much depends on how the old people would choose to spend their money (or not spend at all).

For some of those who believe that aging populations are inflationary, one possible sequence of events looks something like this:

1. The older people retire and stop working, but choose to continue to spend the same amount of money or even more (you know, enjoying the fruits of their younger selve’s labour; or they could be forced to spend their money on necessities such as healthcare);
2. But even if the world still needs to produce the same amount of goods and services, there are fewer workers producing said amounts;
3. Business that want to produce the same amount of products and services would have to compete with other businesses for the shrinking pool of workers;
4. This would drive wages up, which is an inflationary pressure.

For some of those who believe that aging populations are deflationary, one possible alternate sequence of events may look something like this instead:

1. The older people retire and stop working, and consequently choose to spend less money (or it could simply be a function of the being less need to spend, e.g. there is little need to spend on childcare, education, etc. as their children are already grown up);
2. The world needs to produce fewer products and services;
3. Businesses need to hire fewer employees;
4. But the pool of younger workers already exists. This would result in the same quantity of workers competing for fewer jobs, driving wages down, which is a deflationary pressure.

So I think to be able to predict what the future of inflation would look like, you’d have to take a guess (an educated one or otherwise) on how older people of the near future would choose to spend.

But how about the past and the present? How, for example, have the American people spent their money? The chart from the Bureau of Labor Statistics shown below shows that in 2013, peak expenditure happens between the ages of 35 to 54. Spending drops off after that.

Source: US Bureau of Labor Statistics

If this spending pattern persists, this suggests that a US population that continues to age would face a larger deflationary pressure.

So, has it?

Well, it’s not actually really that clear. The graph below plots:

1. People at or above 65 years of age as a proportion (%) of US Census Bureau migration data total population (blue solid line);
2. Year-on-year percentage change of the number of people at or above 65 years of age (yellow solid line); and
3. Year-on-year percentage change of average nominal wages (Bureau of Economic Analysis total national wage data divided by Census population)

from 1948 to 2019.

Sources: US Census Bureau, US Bureau of Economic Analysis

The plot shows that there really isn’t an overall relationship between these datasets. The proportion of people at or above 65 has consistently risen throughout the whole period, but wage growth seems to be all over the place.

If you split it into separate periods, one might be able to force relationships out:

1. In the 50’s and the 60’s, the YoY growth of people >65 is high, but wage growth is all over the place on both top and bottom extremes;
2. In the middle, you see relatively low older population growth, and quite high wage growth;
3. Then starting in the 2000’s to the present, you see an uptick in older population growth again, and wage growth dropping.

(Of course one could ask, if the oil crises of the 70’s hadn’t happened, would we see this higher wage growth in period 2?)

But if one insisted on performing a regression analysis, using data from the 1970 to 2019, you can sort of see a mostly positive relationship between the percentage change of the older population and future wage growth rates. If this relationship is valid, that would instead suggest that an aging population is inflationary (but do note the near zero R-squared values). 

Sources: US Census Bureau, US Bureau of Economic Analysis

If we plotted the same graph above and switching out the horizontal axis for year-on-year change of total Census population, we can see that an increase in population has an immediate effect of decreasing same-year growth of wages (other future years have extremely low R-squared values):

Sources: US Census Bureau, US Bureau of Economic Analysis

So what does this all mean? Well, I really don't know. But I guess on the shorter term at least, and particularly on the labour market front and ignoring supply issues, there are deflationary pressures still given that the current workforce participation rate is still quite low as of February 2021 data:

Source: US Bureau of Labor Statistics via Federal Bank of St Louis

Shrinking populations and inflation

So if we can’t get any clear relationships between an aging society and inflation, what about a shrinking  population? I.e. one that has gone past simply just aging.

For that we would have to look at data from Japan.

Below is a graph showing:

1. Year-on-year total population percentage change (blue solid line, right vertical axis);
2. Year-on-year average monthly household wage (red solid line, left vertical axis); and
3. Year-on-year average monthly household wage less bonuses (yellow solid line, left vertical axis).

Source: Statistics Bureau of Japan

We can see that for the most part until 2017, when the population decreases, wage growth (with or without bonuses) had a habit of going negative sometimes. After 2017, however, wage growth is the highest it’s been since the start of this whole period, despite its population still shrinking.

In short, yeah, if the world’s population does shrink and if things start behaving like Japan, there would be risk that wages would shrink at times.

Creditor military action and sovereign credit risk

If you were lending out money today in the current form of our financial system, it is probably “self-evident” (i.e. “it’s a no brainer”) that if you’re taking on higher risk, you would expect to be compensated more. For example, if you had a really good friend who you know has a history of returning things to you on time and in the same condition as you have lent them (e.g. a lawn mower), if they needed a loan to tide things over, you’d be more inclined to give them one at a low interest rate (or even an interest-free loan if you’re that tight).

On the other hand, if you had to lend to someone who:

• Never returned your favourite crayon in nursery;
• Somehow broke all your 2B pencils during school exams;
• Always promises to buy the next round of drinks but never does

You would be extremely inclined to charge that person some nonsense interest rate.

But from a borrower’s perspective, this “logic” would feel very counter intuitive. Wouldn’t a higher interest rate increase the chances of me not being able t0 pay? If you’re asking more money from me, there’s an increased likelihood that I wouldn’t have enough money!

So, if you had to take out a loan, would you consider this alternative loan were you:

• Get a lower rate of interest
• But at the risk of your creditor taking very physical action on you if you are on the verge of defaulting?

To put it crudely, would you choose to borrow money relatively cheaply, but run the risk of getting shot if you don’t pay up?

The National Bureau of Economic Research published this paper studying historical loans taken out by one country from another. In modern times, if you lent money to another country, there’s quite little you can practically do to recover your money if they chose to not pay up. Historically (1870-1913), however, you could very well just show up at their ports with your fleet of battleships. 

And it turns out, as the paper reports, that this is quite an effective means of getting your money back (who would have thought). Here are the important bits of the paper’s abstract:

We find that “supersanctions”, instances where military pressure or political control were applied in response to default, were an important and commonly used enforcement mechanism from 1870-1913. Following the implementation of supersanctions, on average, ex ante default probabilities on new debt issues fell by more than 60 percent...

The immediate sentence after that surprisingly claims that this outcome is relatively favourable to the debtor country, i.e. interest rates remain relatively low for the country, and that they would easily be able to borrow more money in the near future:

… yield spreads declined approximately 800 basis points, and defaulting countries experienced almost a 100 percent reduction of time spent in default. We also find that debt defaulters that surrendered their fiscal sovereignty for an extended period of time were able to issue large amounts of new debt on international capital markets.

And here are the relevant tables demonstrating the paper’s claims:

Source: Supersanctions and sovereign debt repayment, Kris James Mitchener

Marc D. Weidenmier, 2005.

Originally posted on my Substack at https://aaronleong.substack.com/p/aging-shrinking-populations-and-inflation

Source: somewhere on Reddit