How Goldman’s Inflation Call Went Wrong, According To Goldman

Seemingly everyone wants to claim they “got it right” on inflation in 2021.

Few actually did.

Consider three things. First, it’s easier to make out-of-consensus calls when you’re not accountable. If there are no consequences, (professional or otherwise) for making dour forecasts, you’re free to predict the end of the world every, single day. When those predictions don’t pan out, you can just reschedule the apocalypse a day (or a week, or a month) hence.

Second (and relatedly) hinting at something which turns out to be true using ambiguous language isn’t the same thing as getting it right. For example, repeatedly suggesting that “inflation will prove more persistent than the Fed thinks” is something different than projecting 7% inflation within 12 months. If the forecast calls for generally sunny weather this week and I suggest on Monday that it might be “cloudier than the weatherman thinks,” I’m not a seer if it rains on Saturday.

Third, getting more specific about one’s projections as reality starts to converge with the original pseudo-prediction doesn’t make the original call less ambiguous. To extend the weather analogy, if I suggest the weather will be cloudier than the local meteorologist expects this week, and then subsequently refine that prediction to include explicit calls for rain once dark clouds gather, that doesn’t make me Punxsutawney Phil. It just makes me a pessimist who knows that clouds sometimes presage rain.

With those caveats, it’s fair to say that some people got it more right (or wrong) than others when it comes to inflation in 2021. In a new note (an annual tradition), Goldman took a look back at the 10 questions they asked for 2021 at the end of last year, on the way to juxtaposing what they predicted with “what actually happened.”

One of the questions the bank asked themselves late in 2020 was “Will core PCE inflation exceed 2% at the end of 2021?” Their answer, at the time, was “No.” Needless to say, that was not the correct answer (figure below).

“The inflation surge was the largest surprise to our expectations this year,” the bank’s David Mericle wrote Tuesday.

So, what went wrong? Well, it’s complicated. As Goldman explained,

We expected inflation to bounce above 2% in the spring as we lapped the weakest pandemic base effects, but to come down in the second half of the year as consumption shifted back toward services and production problems diminished, relieving pressure on goods prices. As of mid-year, US automakers also expected these disruptions to end quickly. Instead, demand for durable goods remained strong and spillover effects from the impact of the Delta wave abroad continued to constrain production, resulting in persistent supply-demand imbalances, unusual shortages, and a spike in durable goods’ inflation. On the services side, at the end of last year we underestimated the acceleration in shelter and the jump in food services and accommodations prices driven by pass-through from large wage gains while enhanced unemployment benefits were in place. We corrected these mistakes in the spring by calling for a sharp and persistent increase in shelter inflation and a pick-up in categories that depend on low-paid labor.

Of course, Goldman’s experience wasn’t unique to the bank’s economists. To reiterate: Very few professional forecasters and economists got the inflation call right, where that means quantifying the overshoot a year in advance and detailing the rationale.

I should also note that when it comes to the other nine questions Goldman asked at the end of 2020, the bank’s track record was good. In all, Goldman’s economists and macro forecasters notched five “correct” calls, one “partly correct,” one “almost correct,” and three “incorrects” including the inflation call.

For what it’s worth, the table (below) shows precisely where Goldman’s inflation outlook went wrong.

Some readers have suggested it’s asking too much of pundits and armchair economists to put numbers to their predictions or to otherwise delve into the specifics of their forecasts the way policymakers and Wall Street do. That’s wholly unsatisfying to me. If we expect the Fed and Wall Street to explain their projections in granular detail, why do we not expect the same of critics?

Admittedly, I appeal to common sense on occasion, which is a bit of a cop out. So, I’m just as guilty as anyone else. But let’s face it: Larry Summers (for example) was correct by accident. And that’s the only way you can be correct when the subject is economics.

With that latter point in mind, I’ll leave you with two excerpts. The first, from “Maybe There’s No Such Thing As ‘Economics’,” is an indictment of sorts. In the interest of balance, the second, from my recap of Jerome Powell’s December press conference, is exculpatory.

Just as political science graduate programs sought to replace millennia of rich theory with statistics in a futile quest to make politics a hard science, so too has economics been reduced to increasingly quixotic attempts at quantification.

In both cases, the harder we try, the worse we do. And the sillier we look. Hence wrong-footed pollsters failing to predict the most consequential US election outcome in modern history and PhDs in economics becoming synonymous with dunce caps in the eyes of the public, even as they confer upon recipients the right to indoctrinate new generations of economists with the patently false notion that economics is closer to physics than sociology. The “best” economics PhDs do far better than tenure and book deals. They get to determine the price of money and steer entire economies based on laws that aren’t laws and theories dressed up as facts.


The mitigating factor for the Fed following the Committee’s dramatic policy pivot in the face of the highest inflation in four decades, was simply that everyone is flying just as blind as they are.

Sure, some analysts and a handful of former officials predicted inflation would be higher for longer, but some pundits predicted Donald Trump would win the 2016 election too. And some enterprising individuals predicted the collapse of the US housing market.

In our zeal to celebrate and immortalize out-of-consensus views that turn out to be right, we tend to forget how rare an occurrence that actually is. That’s why contrarian bets pay out so well — because they were extremely unlikely to be winners. Every outcome, no matter how anomalous, seems obvious in hindsight.


Speak your mind

This site uses Akismet to reduce spam. Learn how your comment data is processed.

One thought on “How Goldman’s Inflation Call Went Wrong, According To Goldman

  1. Maybe the idea of doing nothing should apply to forecasts as well as markets. Somewhere back in the mists of (my) time I seem to remember reading that if weather forecasters gave the same forecast every day — tomorrow’s weather will be just like today’s — they would be right two-thirds of the time. Perhaps economists should try this.

NEWSROOM crewneck & prints