Time, memory and inflation

Almost all the available evidence suggests that the U.S. economy is achieving what many economists had thought impossible — a soft landing in which inflation returns to acceptable levels without a recession. Yet polls suggest that the public remains very down on the state of the economy. Why?

I and others have been arguing that there’s a big gap between people’s personal experience and what they say about what’s happening to “the economy,” that is, other people. Surveys aside, consumer spending on discretionary items like travel remains strong, suggesting that people are feeling pretty good about their own situation even as they say that the economy is in the dumps.

Some commentators, however, are strenuously — I might even say weirdly — opposed to considering this possibility, which I find hard to understand. After all, we know that views about the state of the economy are marked by huge partisan differences. This in itself tells us that what people say about the economy is to an important extent driven by what they see on TV and what they think they should say.

However, when it comes to one important issue, inflation, I think there’s a way to look at the situation that makes sense of the divergence between economists — who are almost giddy over the speed with which inflation has come down — and the broader public, which still says that inflation is getting worse.

If I’m right, a lot of it has to do with time — specifically, how far back people look when thinking about inflation.

News reports often lead with inflation over the past year. There are reasons to focus on one-year rates of change: They help smooth out noisy monthly data, and you don’t have to worry about seasonal effects. And in the past, the one-year inflation rate was probably, as my father used to say, good enough for government work.

At this point, however, I’m pretty sure that most working economists consider a year to be too long a lag. The inflation situation is changing rapidly, probably because we are finally working out the supply kinks caused by the pandemic. If we look at core inflation, which excludes volatile food and energy prices, the one-year inflation rate is still unacceptably high, but over the past three months, inflation has run at only 2.4%, and even this number partly reflects delayed effects of a shelter-price surge that ended many months ago.

It might be a few months before we’re willing to declare victory in the war on inflation, but we won’t need a run of improving data — if the next few months look like the past few, we’ll basically be there.

Yet 74% of voters say that inflation is getting worse, not better. Partisanship colors these numbers: Ninety-two percent of Republican primary voters say that inflation is going up. But it probably isn’t all partisanship and misinformation. Instead, I’d suggest, it’s largely about how everyday people think about inflation.

Some commentators have argued that when you say “inflation,” ordinary people think about the level of prices, not the rate of change. That would be very bad news, if true. We haven’t had actual deflation — falling prices — since the 1930s, so if people won’t be happy unless prices go back to where they were a long time ago, it would take a depression to satisfy them.

But I’d argue that it’s not really about levels as opposed to rates of change. How many people are upset that you can’t get McDonald’s hamburgers for $0.15, which is what they cost in the chain’s early days?

Also, consider the last time we saw a big drop in inflation, during the Reagan years. Overall prices continued to rise — in fact, inflation stabilized around 4%, much higher than what we have now. Yet as a 1984 poll suggests, voters nonetheless gave Reagan credit for reducing inflation (credit that actually belonged to Paul Volcker, but that’s another story).

So here’s an alternative hypothesis: Voters assess inflation by comparing current prices with what they remember paying in the past, where the relevant past may be several years ago. Put it this way: When we try to assess macroeconomic trends, we look at inflation over a relatively short period, say three months, but to predict consumer perceptions, we want to look at a much longer period, say three years.

Now, I don’t know whether three years is the right number. But let’s run with it for a minute.

At the time of that 1984 poll, three-year inflation had indeed declined substantially. Right now, by contrast, despite a rapid decline in one-year inflation, three-year inflation is barely off its peak.

Why the difference? The Reagan-era disinflation came after years of high inflation, so when inflation finally began coming down, it did so at all time horizons. This time, however, we had years of low inflation, then a sudden surge in short-term inflation in 2021-22, which is still filtering into longer-term averages even though short-term inflation has come back down.

Or to be less formalistic, arguably right now Americans’ notion of what things “should” cost largely reflects prices before the big 2021-22 run-up, so they don’t perceive any improvement in inflation — yet.

But I’d argue that there are good reasons to expect those perceptions to improve. Over time, memories of prices from, say, 2020 will fade or come to seem less salient, while the relatively low inflation we’re experiencing now will loom larger in public perceptions. Suppose, to be excessively specific, we assume that future inflation proceeds at 2.4% — the core rate over the past three months. What would that mean for the three-year rate of inflation? The answer is that it would gradually decline over the months ahead.

If some people have a shorter time horizon, say two years, perceived inflation will decline even faster.

All of this has obvious political implications. Current polling says that Americans give Donald Trump an edge over Joe Biden in handling the economy, which is unfair but understandable. After all, Biden presided over a big bulge in inflation, while Trump didn’t, and the huge improvement in the inflation picture is too recent to have filtered into public perceptions. If anything, I’m surprised that the gap in perceived economic competence isn’t bigger.

But the election is more than a year away, so there’s still time for memories of lower prices to fade and for the reality of sharply lower inflation to sink in with the public. I doubt that the economy will be a source of political strength for Democrats, but it may well be much less of a drag than many people imagine.

© 2023 The New York Times Company