The real world is nominal

The real world is nominal

Central banks shouldn’t target inflation at all — they should stabilize nominal spending

Central banks shouldn’t target inflation at all — they should stabilize nominal spending

Matthew Yglesias

19 hr ago

David Roberts, one of the best climate policy analysts around, was way ahead of the curve in urging technocrats to abandon the dream of economy-wide carbon pricing as the go-to solution for climate change.

As he emphasizes in one of his posts on the subject, “understanding and planning around political economy is just as important as understanding and planning around physics.” A pricing system is incredibly hard to make sufficiently airtight and rigorous because “it’s a very politically toxic thing to talk about. Any effort to increase the ambition of the overall system means that your opponents can run attack ads saying you’re trying to raise the price of gasoline, which is one of the things politicians hate to be seen doing the most.”

But then in his role as a frustrated progressive guy, he tweeted into the void last week that public concern about inflation is mediagenic.

We also had Sarah Jeong wrongly tweeting that the issue is “rich people flipping their shit because their parasitic assets aren’t doing as well as they’d like,” when in reality stock market gains are pretty clearly outpacing inflation.

But I think inflation anxiety is mostly about gas prices. Political scientists have found that rising gas prices clearly hurt presidential approval. And as Georganas, Healy, and Li find (see also a similar result from Cavallo, Cruces, and Perez-Truglia), impressions of inflation mostly form in response to rises in the price of stuff people buy frequently. So the eye-popping inflation prints that were driven by soaring used car prices got written up by economics reporters and flagged on GOP social media, but they made very little impression on the public because in any given month, very few people will buy a used car.

And beyond that, most car purchases are at least somewhat discretionary. By contrast, you buy groceries all the time and you’re not going to tell your family “sorry, it’s not a good week to eat dinner.”

This kind of inflation gets more media attention because it’s more relevant to more people. At the end of the day, media coverage is fairly demand-driven, and people have a lot of interest in the price of gasoline and bacon.

(Lindsey Nicholson/UCG/Universal Images Group via Getty Images)

That said, a macroeconomist would say that this is actually the kind of inflation that policymakers should pay the least attention to because the price of globally traded bulk commodities is least affected by idiosyncratic aspects of domestic policy. I don’t need to know anything about the fiscal and monetary policy environment in Iceland to tell you that if gas prices have risen here, they’ve risen there, too, because it’s all traded in one global market. So why sweat it? But then you sound like an asshole telling someone whose living standards are plummeting due to rising food and gas prices that what you’re really focused on is the price that a medical practice charges your insurance company.

This is a long lead-up to say that I think “inflation” is a very confusing concept, and it would be better to talk about “nominal income and spending” when doing monetary policy versus the prices of specific things when talking about living standards.

The real world is nominal

People have adopted the annoying convention of referring to inflation-adjusted quantities as “real” when obviously real-world financial transactions are specified in nominal terms.

I have a multi-year contract with Verizon for phone, internet, and television services, and it specifies a price in dollar terms. A mortgage or a car loan comes with a nominal monthly price attached. I’m having some masonry repaired on my house, and it’s taking a while to get the permits, but I agreed a while ago with the company on a price. There’s a house for sale on my block whose owner was overambitious with the pricing, and it’s been listed for a few months now. None of these prices automatically adjust for inflation, and in many cases, there are specific contractual terms that prohibit them from adjusting for inflation.

To say that the inflation-adjusted price of my Verizon bill has fallen over the past year is true. But to say that this falling price is “real” is violence to the English language — the real price is the price that I really pay, and that price is nominal.

It’s obviously possible to write a contract that specifies an inflation adjustment. But to do that you would need to pick an index. The press reports most frequently on the Consumer Price Index, but the Fed tracks the Personal Consumption Expenditures Deflator, which gives a somewhat different result. For dull technical reasons, I think the Fed should probably track the GDP Deflator instead. People sometimes compare U.S. inflation to European inflation without accounting for the fact that the European inflation concept (the Harmonized Index of Consumer Prices) is actually pretty different from the U.S. concept. In the European way, the price of rent is a relatively small share of the index because most people don’t rent their dwellings. In the American way, even homeowners consume housing, so we need to consider the rent that we pay ourselves in our capacity as homeowners. This requires a statistical imputation to guess what owner-occupied housing would rent for it was on the market.

There’s also a question of what rent you should look at. The average rent paid by renters is a slow-moving variable because most rental contracts are somewhat long-term. Landlords like to avoid costly vacancies, and moving is a hassle for tenants. Normally indexes look at this slow-moving average rent, but there’s a theoretical argument that (for some purposes, at least) you should look at spot rents — not “what does the average renter in this city pay?” but “if you had to lease a place today, what would you have to pay?”

This all gets very technical, but the point is that there is no unequivocal answer to the question “what is the inflation rate?” and thus it is odd to refer to inflation-adjusted quantities as “real.” The nominal figures used in actual transactions are real, the statistical adjustments are constructs.

“Real” wages are largely gasoline

The chart below shows average hourly wages over the past five years. You can see that growth happens fairly steadily, with a big discontinuity around pandemic furloughs, which primarily impacted low-wage workers.

And here is the price of a gallon of gasoline during the same period. You can see that gasoline is subject to wild swings in price, not just specific to the pandemic, but all the time. It’s constantly going up and down.

If you combine these series, you can create a chart showing how many gallons of gasoline one hour of work buys you. The “real wage” here exhibits plenty of instability, but as we can see from the two prior charts, the instability is driven by fluctuations in the price of gasoline, not by fluctuations in wages.

When economists are interested in inflation, they tend to focus on “core” inflation — i.e., excluding food and energy prices. The reason they do that isn’t because food and energy are unimportant. It’s that statistically, today’s core inflation is a better predictor of future total inflation than is total current inflation. You can think about why that is in a few different ways, but one is that commodity prices have self-equilibrating properties. If businesses think oil will stay expensive, they invest more in oil drilling so it gets cheaper. If they think oil will stay cheap, they invest less and it gets more expensive. Expensive oil is annoying, but it doesn’t set off an expectations-driven spiral.

At any rate, core inflation — the thing economists and central bankers care way more about — is much more stable than gas prices.

The upshot is that, while in some sense people care about their inflation-adjusted income, in the short term these so-called “real” wages are largely driven by fluctuations in the price of gasoline, not by the state of the labor market. So while as a politician you should try to be seen as worrying about helping people cope with rising gas prices, as a policymaker, it doesn’t make that much sense to worry about the ups and downs of “real” wages. By contrast, nominal income trends have real-world significance.

Inflation isn’t good for debtors — but rising nominal income is

You may remember having heard somewhere that inflation is good for debtors, and possibly associate this view with William Jennings Bryan and old-time populists.

And you may wonder, how is it that gasoline or rent getting more expensive helps me pay off my debts? Obviously, the answer is that it does not. So where did this idea come from?

Well, here’s the story. If your pay rises 10% and prices rise 10%, your “real” wage has stayed flat. But if you have a mortgage, car loan, or any other long-duration nominally-specified financial obligation, the scenario with a higher nominal income is much better for you. This is not a big issue in the economy right now, but debt overhang was a huge issue during Barack Obama’s presidency. If we’d had faster nominal wage growth back then, most people would have been better off even if 100% of the faster growth was clawed back by higher consumers prices. The low inflation climate that prevailed at the time meant that anemic nominal wage gains did concentrate into “real” wage growth. But because the real world is nominal, those anemic nominal wage gains meant real difficulty in climbing out of the financial crisis.

The key thing here is that while inflation isn’t good for debtors, it’s also not so bad as to make it actually true that you just need to adjust everything for inflation and it all comes out in the wash. The nominal quantities matter.

I think the best way to convey this is to forget about inflation and actually just look at the nominal quantities. During the Great Recession, that nominal gross domestic output crashed, and then we never got “catch-up” growth. Instead, we returned to a new level of growth that was somewhat lower than the pre-crisis growth rate.

The pandemic story is very different. The economy crashed harder, but we then got a nominal growth surge and we caught up.

That is a stabilization success story, and now the only big question of stabilization policy is “Can the Fed land the plane now that we need the nominal growth rate to slow back down to a normal-ish 5% or so per year?”

What I like about this perspective is that it gives clear, obvious answers to questions.

Why has nominal growth surged recently? Because Congress and the Fed wanted to catch up after the pandemic.

What do we need to do now? Slow the nominal growth rate back to a sustainable level — starting with the Fed doing less QE and Congress not doing more stimulus — and likely escalate to interest rate hikes.

By contrast, asking about inflation raises a lot of tedious arguments.

Inflation is always or never a monetary phenomenon

Milton Friedman famously said that “inflation is always and everywhere a monetary phenomenon,” which is on its face absurd.

Ryan Dezember at the Wall Street Journal reported on November 4 that wheat prices were rising due to “drought across the northern hemisphere.” And of course throughout human history, bad weather — especially unexpected bad weather — has driven poor harvests and reductions in living standards. How could Friedman claim that monetary policy is responsible for expensive wheat? It’s absurd.

The idea is that a central bank following a strict enough targeting rule would make sure that even though wheat supplies are disrupted, overall prices stay on track.

You could imagine a scenario where expensive gas means people feel poor and eat out less, driving down restaurant prices. Or one where cheap gas means people have extra cash in their pockets so they try to rent nicer places to live and end up driving inflation in the housing sector. That’s Friedman’s point — no matter what happens on the supply side of the economy, it never has to result in a generalized increase in prices.

The problem with people continuing to quote this insight is that contemporary central bankers believe, rightly, that it doesn’t make sense to respond to a drought that increases the price of wheat by trying to raise unemployment that will generate offsetting price decreases elsewhere in the economy. So bankers’ current thinking is you don’t tighten monetary policy in the face of a supply shock. But that has not turned macroeconomic debates into a weird spelunking exercise where people try to dice the data more and more finely to understand exactly why prices are high in this sector or that.

That’s the thing Friedman was trying to warn us away from. There is always some specific micro-level reason that sectors with above-average price spikes are above-average. But there is still a question of how much overall spending there is in the economy. And it’s easier to see central banks’ options if you totally ignore inflation and just focus on total nominal spending, nominal income, or nominal GDP — all of which are different ways of counting the same thing.

Central banking without inflation

Jason Furman, who used to be a chief economist in the Obama administration, has a new paper out titled “What the Federal Reserve Should Do Now” that’s very focused on specific operational questions. He says that instead of saying “rates will probably rise in the second half of 2022 but might go up sooner if inflation is unexpectedly bad,” they should say we should expect at least one rate hike in the first half but they’re open to delaying it depending on what the data says.

I always feel very uncertain that deep in the weeds.

What I think is that central banks should talk less about inflation and more about stabilizing nominal spending.

In the face of a drought, you want to stabilize nominal spending, which means that inflation will probably rise, but it also might not if it happens to be offset by good news on some other front.

In the face of a huge catastrophe that tanks nominal spending, like a pandemic, you want to catch up with a period of very fast nominal spending growth — even if some of that growth takes the form of inflation.

When the catching up is done, you want to slow that nominal spending back down to a sustainable pace.

Right now the economy is obviously not fully healed from the pandemic. But the nominal variables are, in fact, back on track.

The extent of the “real” economy’s recovery depends on a lot of known unknowns (how many early retirees will un-retire if the public health situation improves?) and somewhat imponderable mysteries (is making people use QR codes to order dinner labor-saving productivity or shadow inflation via quality decline?) that the staff economists at the Fed can’t actually figure out the answer to. When they can do is take credit for successful stabilization, promise steady-as-she-goes spending growth going forward, and then hand off all this supply-chain stuff to other branches of the government. And then the White House and Congress can tackle issues like the overregulation of house-building and the throughput of west coast ports as what they are — questions of productivity and growth rather than questions of inflation.