Why do Economists Dread Deflation?

What's So Bad About Everything Getting Cheaper?

Know someone who might like Capital Gains? Use the referral program to gain access to my database of book reviews (1), an invite to the Capital Gains Discord (2), stickers (10), and a mug (25). Scroll to the bottom of the email version of this edition or subscribe to get your referral link!

One of the gaps between the median economist and the public at large is that the public tends to treat inflation as a purely bad thing; the economists largely don't. This is partly because inflation is a very in-your-face cost: the closer you are to needing careful budgeting for groceries, the more obvious it is when the price of eggs suddenly doubles. On the flip side, when the price declines, the natural response is to keep spending within the existing budget (i.e. inflation leads to unpleasant tradeoffs and deflation leads to pleasant ones).

Popular antipathy to inflation makes sense, as it’s the most common way people end up interacting with macroeconomic forces. Since most of us borrow infrequently, and on fixed terms, interest rates have a less direct impact on individual lives than rapidly rising egg prices. And while unemployment obviously matters too, the difference between a serious recession and an unprecedented boom is when the employment rate goes from ~90% to ~96%—even in very difficult times, most people have a job, but regardless of how well the economy is doing, they still end up worrying about inflation.

One fun way to explore this is to compare Gallup polling data on inflation to the actual inflation rate. Gallup asks people to rank their most pressing financial concern, with inflation and the cost of living topping the list. The poll of concerns actually tracks quite well to the inflation rate, with concerns about inflation reliably peaking a month after inflation does. (You can say this lag looks bad, sure, but the average investor takes more than a month to react to some shift in market regime. Back in early February, when UBS declared ChatGPT the fastest-growing consumer product ever, Nvidia was at $217/share. It's at $386 now; markets take longer to discount information the more important it is.)

But one interesting pattern stands out: people were more worried about inflation in the early 2010s, when the CPI was rising around 1% annually, than they were in the mid-2000s when it was running above 3% and sometimes above 4%. Economists were generally more worried about a slow recovery, a credit overhang, a continuation of long-term trends towards lower productivity growth, and the like. It’s not that the average person didn’t feel all of those problems—it’s that they experienced them differently. Prices were too high for them to buy everything they wanted. “Anchoring bias” is another driver here; there’s a natural tendency to treat the price you pay for things when you first start paying for your own meals as a default, and to treat any change since then as bad (of course, salaries also respond to inflation over time, but another set of cognitive biases gives people a habit of attributing their raises to their own hard work). But it's even more accurate to say that real economic growth makes the average person better-off, inflation is just more salient.

So that's one reason it's common to dislike inflation. Why are economists more sanguine?

The economist argument typically hinges on the idea of nominal rigidities: some prices adjust quickly, up and down, while other prices stay flat for behavioral or contractual reasons. A 30-year mortgage will have the same terms for all 30 years, but the real cost of that mortgage is a function of the price level over its lifetime. Banks naturally bake in some assumptions about inflation when they make long-term fixed-rate loans.1 So the real interest rate the bank is expecting to earn already assumes some inflation; if inflation is lower than that, the lender wins their bet, but the borrower is stuck with a loan that's more expensive in real terms.

There's another big category of spending that also displays nominal stickiness: salaries. People hate getting their pay cut, even in cases where the pay decrease is less than the change in their cost of living. So employers are incentivized to preserve morale by laying some workers off rather than cutting everyone's pay. Which is still bad for morale, but the people most upset by the decision aren't around any more. (It's no coincidence that the jobs with the most variable upside are also jobs where either fixed pay is a small component of the total package—think salespeople and investment bankers—or where workers do things one gig at a time, so even when their pay goes down it's not as if a single employer made the decision.)

In that kind of environment, deflation tends to benefit workers with secure jobs, and creditors. But an economy where those are the two biggest winners is an economy that's rewarding risk-averse decisions—if the winners are lenders, people are less likely to invest in the equity of promising young companies, and if the best-paid workers have steady jobs, fewer people want to work for such promising but uncertain companies, too. It can also easily lead to a deflationary spiral: people lose jobs and stop spending as much money, which lowers demand, leading to more layoffs; every round of this means that a larger share of economic output goes to lucky lenders, but it also means that lenders have fewer creditworthy projects to lend against. It's a bad trap for an economy to be in.

Inflation has its own runaway effects, too: it encourages people to front-load their consumption, for example (when inflation is running at 20%, buying a year's worth of something means getting a 10% discount). That means higher consumer spending, meaning even more inflation—and it means that people are denominating their savings in inert physical goods rather than in anything productive. Inflation has historically imposed menu cost: the cost and inconvenience of firms updating prices on everything they sell. Menu costs have an interesting distributional effect: the biggest companies are already changing prices more frequently, and tend to have better IT infrastructure. Menu costs are a trivial concern for a large restaurant chain that uses digital signage for its specials, has in-app ordering, and already runs lots of promotional offers. The cost is higher for a single-location family restaurant that uses printed menus and that doesn't have a team of data scientists calculating complementarities and elasticities across the entire set of their offerings.

In a less financialized economy dominated by small independent businesses, deflation is actually a fairly good thing: sticky wages and cheaper goods mean that everyone is getting richer. And technological progress is in many senses deflationary: the cost of traveling across the Atlantic, or transmitting a message to the other side of the earth, or of producing enough calories to feed someone, have all dropped over time.

But economies get more financialized as the biggest institutions get bigger, so there end up being more fixed costs and more nominal rigidities. Some of that comes back to policy decisions; the US could choose to encourage small business ownership the way that we encourage the channeling of savings into 401(k) plans and IRAs, and we could treat mortgages as a way to borrow for consumption, and set up their tax treatment accordingly, instead of making mortgage interest tax-deductible. But some aspects of financialization are unavoidable: unless you plan to work until the day you die, you will spend some time saving and some time living off of those savings—and for most of us, peak child-rearing years don't coincide with peak earning years, so it makes sense to deploy some of those savings into letting people in their 20s and 30s form families ahead of when they can pay for it out of their savings.

Moderate inflation is actually a small subsidy to economic dynamism. Ongoing inflation in the presence of sticky wages means a continuous pay cut to anyone who isn't able to negotiate a raise. In other words, inflation determines the rate at which people who aren't very valuable in their current job get asymptotically fired. If we assume that the hiring market is imperfect, it's a good thing to slowly reduce the real cost of keeping some workers on staff.

Productivity growth, even when concentrated in just one sector, tends to be deflationary overall. More output for the same input lowers price levels, and that makes sectors that aren't seeing productivity gains look relatively less efficient.

There are many models that work a bit better when you add a bit of noise; overfitting with limited data is easy, and a little entropy can sometimes nudge things to a local maximum. Inflation is one such source of entropy. Total chaos is bad news, but a little extra uncertainty can be a good thing.

Read More in The Diff

The Diff has covered inflation a few times:

  • The 2x2 Inflation Debate ($) looks at the 2021 rise in inflation and the question of whether or not it's transitory and whether or not it's good.

  • Trapped by the Wrong Inflation Stories ($) is a reminder that the causes of inflation in the 70s were very different from the causes of the most recent bout. (An important feature of the inflation debate is that voter turnout is about twice as high among older voters than among younger voters. Which means it's highest among people with bad memories about the 1970s.)

  • Why A "Commodity Supercycle" is Unlikely ($) argues that the demographic and technological contributors to a 1970s- or early 2000s-style sustained rise in commodity prices aren't in place right now.

  • And Has the "Long Deflation" Broken Price Signals for Consumer Goods? ($) looks at the effect of cheap exports from China on companies' pricing decisions, and how this makes it harder than usual to know what the rate of inflation really is. (Another example of this: the "true" rate of inflation was higher than reported in 2020, since many services, ranging from restaurants to education, had a radical decline in quality with no improvement in price. 2021 was partly a normalization year, and also a year when that normalization started showing up in prices.)

A Word From Our Sponsors

There's a reason over 4 million people start their day with Morning Brew - the daily email that delivers the latest news from Wall Street to Silicon Valley. Business news doesn't have to be boring... make your mornings more enjoyable, for free.

1. Or, to be more specific: mortgage rates are set based on other long-term rates, and also include an interest rate premium because of the borrower's option to refinance. That borrower's option is a literal option—yes, the biggest single financial transaction of the average person's life bundles a home and a long-term interest rate derivative—and like other options, that refinancing option's value is partly a function of expected interest rate volatility. This tends to mean that the spread between mortgage rates and bonds of similar maturity goes up as rates rise; it's easier to imagine an eventual return to zero rates than to imagine interest rates doubling from here. And, of course, when rates go down people refinance mortgages. They're a tough asset to own.

Share Capital Gains

Subscribed readers can participate in our referral program! If you're not already subscribed, click the button below and we'll email you your link; if you are already subscribed, you can find your referral link in the email version of this edition.

Help us learn more about you!

We're building a media kit and need some anonymous aggregated subscriber data. If you work in finance and have time for a 3-question survey, that would be very helpful:

Join the discussion!

Join the conversation

or to participate.