Claudia Sahm: it’s clear now who was right

Claudia Sahm: it’s clear now who was right

Robert Armstrong — Read time: 11 minutes

Claudia Sahm: it’s clear now who was right

The economist on how to temper theory with judgment

The eagle representing the US Federal Reserve

© Financial Times

This article is an on-site version of our Unhedged newsletter. Sign up here to get the newsletter sent straight to your inbox every weekday

Good morning. Say what you will about the finance industry, there is one thing it can do well: stitch together barely related, marginally relevant buzzwords when naming a new product. Behold the BTC Private Credit Fund LP, billed as the “first bitcoin private credit fund”. As far as we can tell, the fund — a standard bitcoin lending operation — has nothing to do with private credit as that term is generally understood. But how, you may be asking yourself, can a self-respecting piece of financial marketing in 2023 not include a reference to AI? Well, BPCF has that box checked by having OpenAI CEO Sam Altman as a backer. What a time to be alive. Email us: robert.armstrong@ft.com and ethan.wu@ft.com.

Friday interview: Claudia Sahm 

Claudia Sahm is, perhaps to her chagrin, best known for creating the eponymous Sahm rule, a recession marker developed as part of a discussion of economic stabilisation payments to individuals. She is almost as well known for taking a contrarian view of the policy response to the coronavirus pandemic. She argued early on that the bulk of the post-pandemic inflation was driven by supply constraints that would pass, and urged policymakers not to overreact. Below she offers views on whether the inflation fight is won, productivity, consumer sentiment, the Phillips curve and much else. 

Her comments have been edited for brevity and clarity. 

Unhedged: Have we landed softly?

Claudia Sahm: The soft landing is not here yet. But it is in the bag. Barring an unforeseen massive disruption, we are landing this plane. Now, there’ll be some turbulence as we approach the runway. I fully expect we’re going to have some disappointing reads on inflation. To me, a soft landing is 2 per cent inflation or spitting distance, say under 2.5 per cent, while unemployment stays low, right around 4 per cent or below. We’ll have it next year. You can see the landing strip. 

Unhedged: It’s true, unemployment’s great. The most relevant signals of inflation are within spitting distance of 2 per cent. But no matter how you cut it, wage growth is around 4 per cent. Is that a potential problem?

Sahm: I have not, and do not now, subscribe to the view that the inflation we have been living through since 2021 is primarily demand-driven, like Larry Summers and my friend Jason Furman did. Those folks thought we put too much money into people’s pockets and there was too much pent-up demand. If you were in that camp, you thought we needed to jack up rates and see wage growth come down. 

Wages are rising at a pace that’s better than before the pandemic, which was a very good time for the economy, but we’ve moved out of the very acute labour shortages. And obviously, we want to get workers off the sidelines. To do that, you’re going to have to pay them more! 

I look at inflation and say that’s because of disruptions from Covid and the war in Ukraine. And because those will eventually work out in some way, inflation will come down. That leads to very different policy prescriptions to fight inflation. And it leads to very different views on the things like whether the $1.9tn American Rescue Plan was a good idea; or whether waiting to raise rates was a good idea. If it’s all demand, then you’ve got to destroy demand. But I don’t think it’s all demand.

On wages, too, we have seen some good productivity numbers. If you’re more productive, you get paid more. And that’s coming after the crap productivity growth we had after the Great Recession. If we’re getting better productivity growth, we should not be using pre-pandemic wage growth as the baseline. 

Unhedged: What do you think is driving that bump in productivity? 

Sahm: Productivity is extremely hard to measure and the data is noisy. So we won’t know for sure for a while. But we have people who now work from home and some really crappy jobs are getting automated away. Maybe AI helps make workers more effective. 

The other piece is that business fixed investment has been really solid in this recovery, contrary to what happened after the Great Recession. We’re seeing R&D, we’re seeing capital investments. Doing business investment is the gift that keeps on giving. It shows up in GDP when it happens, and then it gives workers or businesses the tools or knowledge they need to produce in the next round, and the round after that. 

The Holy Grail in macro would be to get another quarter of a percentage point in trend growth. That really adds up. 

Unhedged: You had a recent Substack post called “I was right”, arguing that the US pandemic fiscal response was largely successful. But if the policy was so good, why are Americans so grouchy about it? Why is consumer sentiment so bad?

Sahm: I spent a lot of time thinking about why people are so gloomy. The unemployment rate is low and wage growth is good, but on the other hand inflation has been high. But both were worse in the 1970s, when consumer sentiment was similarly bad. Most Americans are financially better off [than before the pandemic] — whether that’s measured by jobs, wages, wealth or debt. Before the pandemic, a lot of people did not have a financial cushion; now they do. Debt burdens are at record lows.

This is not about economics. When I put my economic adviser hat on, I need to know that, because then I won’t use the sentiment data. It hurts me, because people called the Great Recession before the forecasters did. Sentiment started falling before GDP ever did. Danny Blanchflower had all this research on sentiment, and what a great forecast of recession it was, calling for a recession towards the end of 2021. I didn’t agree because the labour market recovery was gaining steam. Going into 2022, it became clear it wasn’t a recession. And at some point, I realised these [sentiment] data are basically useless to me.

Unhedged: So you’re saying you don’t quite know what it is, but it isn’t the economy.

Sahm: Yeah. I’ve gone through different hypotheses. I think it fundamentally has some relationship back to Covid. Shutting the economy down, sending people home, a deadly virus we didn’t understand — it broke people. Now we have the war in Ukraine. The most optimistic thing I can say is that bad sentiment is related to Covid and Ukraine, and once we get to the other side, that disconnect will close. It just takes time. The questions about the economy or people’s finances are so fundamental to their lives that if you’re angry and scared about one thing, like Covid, you’re going to be angry and scared about your finances, no matter what your bank account says.

There was a recent analysis in the FT, finding that the big gap between US sentiment and the economy wasn’t there for other countries. Those countries had a really hard time with the pandemic, too. Which makes me think it has something to do with processing the pandemic and lockdowns in a highly charged political environment during an election year.

Unhedged: You’ve called for banning the Phillips curve, the economic model positing a trade-off between inflation and unemployment. Now that we have a bit more hindsight, what’s your retrospective on the Phillips curve in this cycle? And if we ban the Phillips curve, what replaces it? 

Sahm: This fundamentally goes back to a view about how much of inflation is demand versus supply. If you think it’s demand-driven inflation, you can fight that with the Fed’s tools. But how do you know how much monetary tightening to do, how much unemployment you need to get inflation down? So then you march off to the Phillips curve. There are more sophisticated versions of the Phillips curve that incorporate supply shocks. No one brought those out. The versions of the Phillips curve that were brought out in policymaking circles went back to the 1950s or 1960s — essentially just inflation versus unemployment. 

The Phillips curve was used by the same people denouncing the American Rescue Plan to make statements like, “We need five years of 6 per cent unemployment.” But it goes back to why did inflation spike, demand or supply? It’s clear now who was right: it was largely supply. It was completely valid to argue in 2021 that when inflation took off, it was demand. The American Rescue Plan was big, it came after two very big fiscal relief packages and the Fed had been adamant about not raising rates. But the fact this year that inflation has notably come down and unemployment has stayed low only happens if it was mostly supply-driven.

In terms of what other model to use, backing off from the Phillips curve would have been a good idea. And then the thing that economists need to think harder about is how we think about supply shocks. Most of the effort in macroeconomic research goes into thinking about demand disruptions. The [industry gold standard] New Keynesian dynamic stochastic general equilibrium model has wedged into it a Phillips curve that can do supply shocks. But we don’t really know how to calibrate [these sorts of models].

A lot of this is art, not science. The academic stuff looks like science, but what actually is useful in the real world is much more judgment-based. But you ought to have tools that at least don’t do damage. The Phillips curve has done damage. 

Unhedged: There’s a lot of worry now about excessive debt and deficits. Olivier Blanchard is saying we need to get r minus g, the real interest rate paid on debt minus the growth rate, on a sustainable trajectory. What’s your perspective on debt sustainability? 

Sahm: First off, Olivier is adorable, what a great way to frame it. My view is that it’s completely misguided to have a discussion about the size of the federal debt. The entire conversation about r minus g, while maybe useful for macroeconomists to think about, ignores that it matters what we spend on. If we are on a path for higher productivity growth after the pandemic, the American Rescue Plan, the infrastructure act, the Chips act, the Inflation Reduction Act — they all get a piece of that pie. 

Unhedged: Folks like Olivier are worried that if long rates are now higher, there’s simply less capacity to spend money the right way, because more of the budget is going to be spent on interest payments.

Sahm: The way I’d frame it is that the stakes are higher and higher for doing spending in a responsible way. If the government is spending in a way that is targeted and effective, say, at raising productivity, that’s money well spent. I would never want to say that spending will “pay for itself”, because that claim has a very dubious history. And yet if you do it right, it does, at least to some extent. 

The other piece is getting into the weeds of what the federal debt is: social security and Medicare. If you do not do something to pull in those expenses, you’re done. That’s a very difficult, delicate conversation. The stakes are higher to spend money well. The stakes are very high to make those entitlement programs cost-effective and well-designed. You can almost cut the whole rest of the budget and you still would have questions about debt sustainability and higher interest rates.

I don’t know that Olivier would disagree with me that entitlement programmes need to be addressed. But as macroeconomists, we’re not setting up a framework that really pushes policymakers to think in those terms. We just say, “It’s too big!”

There is a corollary to [the damage done by macroeconomists relying on the Phillips curve model to analyse inflation] within the debt sustainability debate. The profession got burned, and rightly so, for Greece. The Reinhart-Rogoff argument [that debt-to-GDP ratios over 90 per cent hurt growth] ended up being very bad advice. Macroeconomists need to learn that big numbers aren’t enough. Policymakers need better guidance.

Unhedged: As the author of the Sahm rule, you’ve made the point over and over again that it’s not a rule of nature. Still, financial markets will always want a recession forecasting rule. This time, the inverted yield curve didn’t seem to work. Now the Sahm rule is getting attention. Would you suggest Wall Street use a different rule?

Sahm: No, I won’t give them that. The market is looking for something to predict a recession. The Sahm rule has never been that. It was meant to send out fiscal transfers [as automatic stabilisers]. The typical predictors, like the yield curve and sentiment, have not performed well. You have to go back to 1947 to find a time where we had two consecutive quarters of GDP decline outside of a recession. And we had that last year. Everything is haywire! Rely on the rules of thumb at your peril. Right now, everyone wants some kind of certainty or comfort or guidepost. You ain’t getting it.