(This post originally ran as an opinion piece in Barron’s.)
The U.S. today is experiencing inflation. This is not controversial. But what exactly does it mean?
In the textbook, inflation is a rise in all prices together, caused by an excessive increase in the money supply. But when we measure it, inflation is just a rise in the average price of goods and services. That average might reflect a uniform rise in prices due to excessive money creation. Or, as today, it might instead be the result of big rises in the prices of a few items, for their own reasons.
Over the past year, prices have risen by 7.5%, far above the usual 2% target set by the Federal Reserve. But 70% of that 5.5 points of excess inflation has come from two categories that make up just 15% of the consumption basket: energy (2 points) and new and used cars (1.9 points). Used cars alone make up barely 4% of the consumption basket, but accounted for a third of the excess inflation.
Some commentators have argued that inflation is just a matter of too much money. If that were true, it’s hard to see why so much of it would be flowing to cars. (And before you say cheap financing: Rates on auto loans were lower through most of the 2010s.)
In recent months, vehicle and energy prices have begun to stabilize, while food and housing prices have picked up. These price increases hit family budgets harder. A car purchase can usually be put off, but not rent or groceries. But this is still a story about specific sectors following their own dynamics.
Energy prices are global, and their periodic rise and fall depends mostly on the politics of oil-producing regions (as we are being reminded today). As recently as the summer of 2014, gas prices were higher than they are now, before falling precipitously. No doubt they will fall again, but in the short run there is not much to do about them—though it may be possible to shield people from their impact. In the longer run, decarbonization will leave us less vulnerable to the gyrations of the oil market.
As for vehicles, it’s no mystery why prices soared. Early in the pandemic, automakers expected a long period of depressed demand, and cut back production plans. When the economy bounced back rapidly, automakers found themselves short of key inputs, especially semiconductors. Combine this with a pandemic-induced shift in demand from services to goods, and you have a formula for rapid price increases. The effect was strongest for used cars, whose supply is essentially fixed in the short run.
Housing has made a smaller contribution so far—0.6 of the 5.5 points of excess inflation—but given the way the Bureau of Labor Statistics measures them, housing prices are likely to rise sharply over the coming year. This is a problem. But, it was also a big problem before the pandemic, when rents were rising by nearly 4% annually. Housing affordability is a serious issue in the U.S. But if the question is why inflation is higher today than in 2018 or 2019, housing is not the answer.
Finally, there are food prices, which have contributed about 0.7 points to excess inflation over the past year, and more in recent months. Food prices, like energy prices, are famously volatile; there’s a reason they are both excluded from the Fed’s measure of “core” inflation. They’re also an area where market power may be playing a major role, given the high concentration in food processing. Monopolies may be reluctant to fully exploit their power in normal times; price increases elsewhere in the economy give them a chance to widen their margins.
The great majority of the excess inflation over the past year has come from these four areas. Other sectors—including labor-intensive services where prices have historically risen more quickly—have contributed little or nothing.
The point is not that these price increases don’t matter. Food, housing and energy are necessities of life. People are naturally unhappy when they have to pay more for them. The point is that current price rises are not symptoms of economy-wide overheating.
Some of these prices, like autos, will come back down on their own as supply-chain kinks work themselves out. Others, like housing, will not, and call for a policy response. But that response is not raising interest rates, which would only make the problem worse. The main reason why housing costs are rising is that the U.S. does not build enough of it, especially in the expensive metro areas where employment opportunities are concentrated. Construction is one of the most interest-sensitive sectors of the economy. Rate hikes will cause supply to fall further short of demand.
Some might say that the Fed still controls the overall level of spending in the economy. If people spent less on used cars, wouldn’t they spend more on something else? This ignores the existence of balance sheets. Households hold cash, and finance many purchases—including cars—with debt. Lower used-car prices wouldn’t mean higher prices elsewhere, but higher household savings and less debt.
An inability to build housing where people want to live, dependence on fossil fuels, fragile supply chains and the monopolization of key industries: These are all serious economic problems. But they are not monetary-policy problems. Looking at them through the lens of a textbook story of inflation will not get us any closer to solving them.
Logically speaking the difference is between a supply side problem, that hurts bot capital and workers, and a demand side overheating, where employment reaches some sort of upper limit, the wage share is very high, but the government is forced to continuously pump demand in the economy to increase the profit share so that business don’t close and fire workers, leading to a wage-share spiral.
Supply side problems can certainly happen contemporaneously with demand side overheating, but the proof ov overheating should be low unemployment, an high wage share and a profit crunch, something that I don’t see.
I believe that in normal times what happens in modern capitalist economies is that there is a small supply side deflation due to productivity increases summed up to a bigger demand side inflation due to the fact that governments keep demand higher than would be the natural level, that in turn depends on the fact that the normal tendency of a capitalist economy is that of low employment, not full employment.
So normally governments push employment up through demand creating a bit of overheating but not too much, which is eased by productivity increses; when productivity increases stop or revert for any reason total inflation shoots up and austerians claim that the government is printing too much and overheating the economy, that actually is true but they forget to add that a non-overheated economy is an economy of low employment and low wages.
So the key theoretical problem is the implicit belief that if left alone a capitalist economy would tend to an equilibrium of high wages, low unemployment and low or zero inflation, which is false and in fact a bit weird if you think about it.
Yes this all seems right to me.
I agree substantially with the analysis of aggregate capitalist economy as demand constrained. This has to do with the interests, decisions, and actions of the production sector and the credit issuing sector of the economy. The result is some level of involuntary unemployment except when there is a private credit boom, caused by competition to issue credit to the working class on easy terms, or when there is a government spending stimulus, both of which increase employment, aggregate cash flows, and profits.
I also appreciate the analysis of housing supply constraints and supply chain constraints in the global energy and auto industries. I think there is a glut of vacant commercial real estate that seems to be subsidized by some combination of private credit policy and government programs because the rent is not falling rapidly and there are not high default rates. Instead governments are encouraging workers to go back to the office. So commercial real estate prices are set by some combination of commercial norms and government norms not by competition in so-called “free” markets.
Sorry in advance for length. I think there are two claims here, one about explanation and one about appropriate policy. I’m not sure the connection between the two claims is as close as you are suggesting.
The first claim is that the cause of the current inflation is a set of ‘special factors’ causing an increase in the price of a few items, and not monetary policy. Both sides of this debate confuse me, because I think everyone agrees that monetary policy, market power, automakers’ 2020 production cutbacks, etc., are *all* causing current inflation in the sense that if counterfactually we could vary one of these factors keeping the others fixed, that would result in a different level of inflation. So I assume the disagreement is about explaining the difference between current inflation and some `normal’ benchmark, say 2019, in terms of differences between the level of these causal factors now and in 2019. Heller would argue that if we could set monetary policy back to its 2019 level, this would bring inflation most of the way down to its 2019 level even if the ‘special factors’ were still present. Your first claim (under this interpretation) is instead that if we could set the ‘special factors’ back to their 2019 levels, this would bring inflation most of the way down even if we did not change monetary policy. (Though this can’t entirely be what you mean because fragile supply chains, fossil fuel dependence, etc. obtained in 2019 too.)
Your second claim is that tighter monetary policy is not the appropriate policy response to the current inflation. You seem to suggest that this follows fairly directly from the first claim, but I don’t think it does. Even if higher inflation relative to 2019 was entirely caused (in the contrastive sense) by special factors, tighter monetary policy might bring it down. Equally, even if higher inflation was caused by loose monetary policy, price controls/’supply chain policies’/etc. might bring it down. The question is what policy mix is best.
This question could be *related to* the question of which explanation is correct. If the difference between 2021 and 2019 inflation is mostly due to differences in ‘special factors’, *and* these special factors will mean-revert even absent a change in monetary policy (as you argue in the case of autos), that might imply monetary policy doesn’t have to change much in order for inflation to come down eventually, under a bunch of other assumptions (mostly about whether large transitory increases in inflation move some notion of its ‘underlying trend’). But I think it is better to spell out the link between the two claims more explicitly (or just make the case for your preferred policy mix directly), because it’s not that clear, at least to me. (I appreciate Barron’s has a word constraint!)