The economy keeps surprising us
On Wednesday, the monthly consumer price index report showed that inflation slowed down to 3% during the month of June. That marks a decline of two-thirds from its peak of slightly more than 9% just a year ago. It was the lowest inflation reading in more than two years and marked the first time since early 2021 that hourly wage earnings have increased at a faster rate. A large chunk of inflation has been from rents, which have been going down, but that is just now starting to show up in the data.
Although the inflation problem is not yet solved, it is very encouraging just looking back at where things were one year ago and realizing we can achieve the proverbial soft landing. Since last summer, inflation has come down substantially while jobs have continued to be added each month and the unemployment rate has been steady. Traditional macroeconomics has long believed that there is a tradeoff between inflation and unemployment, known as the Phillips Curve. The idea is that if one goes down, the other goes up.
The Phillips Curve is not perfect (see 1970s stagflation) and there is plenty of debate among economists as to how low unemployment can go and how steep the tradeoff with inflation is. Strict adherence to it though would have meant that the only way the high inflation of the last two years could be brought down is by raising unemployment. Despite the Fed raising interest rates substantially and quickly over the last year, the job market has been amazingly resilient. It is worth asking why that has been the case.
One explanation that I find convincing is from this piece in The Atlantic. It argues that economists and others who thought unemployment would have to go up mistakenly believed that there were no more workers available. The Phillips Curve comes into play here because, according to this mistaken belief, since there were no more workers waiting on the sidelines, unemployment was as low as it could go. Since there was a high demand for workers, but no more workers available, a reduction in demand caused by rising interest rates meant that employers would lower their demand for workers. The result would be layoffs and an increase in unemployment and only then would inflation go down. That is basically what happened in the 1980s.
The problem with that idea, The Atlantic piece says, is that the official unemployment rate does not tell the entire story. It does not count people who could be induced to return to the workforce, but are not currently job searching, as being unemployed. As it turns out, there were millions of people like that. That has been one of many ways the pandemic has thrown traditional economic metrics off.
During and after the pandemic, many people left the workforce. Some retired early, but many were kept from working by everything from long-term health problems to child care to changes in careers and a greater willingness to be selective about which jobs they were willing to do. There have always been people who fell into one or more of those categories, but not on the scale that has happened since 2020.
The unemployment rate is not the only indicator that has been thrown off in the post-pandemic economy. One indicator that has been cited frequently in predicting recessions is the inverted yield curve. What it looks at is the difference between returns on government bonds based on their longevity. Longer term bonds tend to pay more than shorter term bonds. When the opposite happens, the yield curve is said to be inverted. The reason for that is because investors believe interest rates will decline in the future. That is worrisome because interest rates tend to go down when the Fed cuts them and that usually happens when the economy is not doing well.
The inverted yield curve is not scientific by any means. To the extent it has been right it is a correlation and not a causation. The time between when it first inverts and when a recession hits is not uniform either. Still, it has been a decent indicator of future recessions on several occasions going back almost 50 years.
The yield curve has been inverted for over a year now and there is no sign of a recession. Many economists and others who have given it weight before are discounting it now. The biggest problem I have with relying on it to predict the future is the same problem I have with all other economic metrics and indicators. The sample size is very small, way too small to draw any real conclusions from.[i] At best, the inverted yield curve can be said to have been a good indicator five times since around 1980.
That is the problem with trying to predict where this economy is going. Not only is the available sample size very small, but the economy since 2020 is nothing like any of the economies the world has seen. I have been concerned that the raising of interest rates could cause a recession. It would not have surprised me if we had fallen into one by now. At the same time, I am not surprised that we have avoided one either. If there is any rule to follow in this economy, it is to expel the words impossible and inevitable from your vocabulary.
The problem with dealing with a small sample size is not unique to economics. A big interest of mine is electoral politics and trying to predict how that will play out runs into the same issue. I wrote plenty about the midterms last year and how surprised I was at what happened. It was unprecedented for the party in the White House to do well in a midterm when the president had an approval rating below 50%. Prior to last year, it was a law of political physics.
Even though that had been the case every time for as long as polling has existed, the sample size was very small. It was less than twenty, too small to draw any sweeping conclusions from. Drawing conclusions from a small sample size can be done, but it will only work if you’re not out of sample. 2022 was out of sample. To be clear, by “in sample” in this case, I mean the party not in the White House nominates candidates who are not repulsive or extreme. There was no precedent for the party out of the White House nominating the worst possible candidates in almost every key race. The reason Democrats defied history is that their Republican opponents alienated a crucial chunk of people who ordinarily would have voted Republican.
In the case of the current economy, we are not in sample either. To be fair to those trying to prognosticate, economies are vastly more complex than electoral politics and probably most other things, too. Still, many economists and others have adhered to traditional economic models that assume certain things are set in stone and anything else is impossible.
It is a good thing that traditional views about inflation and unemployment have been wrong so far. Had they been right, we would probably be in a recession now and millions would be unemployed. The last year has proven that a recession is not inevitable. At this point, if a recession happens, it will likely be caused by something that is on nobody’s radar or it will be a conscious policy choice. That was not clear a year ago, but it is now.
Looking back at 2021 decisions
Although I make very few predictions in general and almost none about the economy, I did support aggressive fiscal policy in 2021 and favored waiting to raise interest rates. In 2009, stimulus efforts were inadequate and unemployment needlessly stayed high for years. The recovery from the financial crisis was lousy and was not something that should be repeated again. Knowing that, I thought it was better to risk doing too much than to risk doing too little.
Many have blamed the stimulus passed in March 2021, at least in part, for the inflation surge that happened. At the time, there were some warning that it could do that, but there were plenty of others who believed it was either a risk worth taking or didn’t think it would happen. Inflation did wind up surging and the increase in demand from the stimulus measure was relevant to it, but hardly the only cause. Supply chains had a significant effect, too, and so did millions of missing workers who are now working again.[ii]
Politically, it doesn’t matter what the cause of inflation is, at least in the short-term. When people are upset about the economy, the president bears the brunt of it. Long-term, a worry I and many others had was that if the stimulus measures were blamed for inflation, there would be hesitance to do it aggressively during the next downturn and the recovery from it would be slow like after 2008. Hopefully, the next downturn will be a mild one like in 2001 that will not require massive stimulus. Even if the next downturn is similar to 2008 or 2020, I am less worried about not doing enough than I was before.
I hope we have nothing like either of those episodes again in my lifetime, but we might. If we do it will likely be caused by something completely different that may not even exist today. If inflation is brought down without a recession, the downsides of excessive stimulus measures will not be worried about nearly as much should a major downturn happen. The idea that it is better to err on the side of high inflation than high unemployment will have been vindicated (in my view it already has).
With respect to the debate in 2021 about whether inflation was transitory or persistent, I didn’t make any predictions. I wanted it to be transitory and supported policies consistent with that, but arguing over the granular details of what affects inflation, let alone where it might be headed, is far beyond my expertise.[iii] Those who argued it was transitory, including Jay Powell, wound up being wrong. In light of that, was it a mistake to wait until 2022 to raise interest rates?
A year ago it looked that way, but now it’s not so clear. Had interest rates been raised earlier, they could have been raised at a slower pace than they have been since early 2022. Despite the fast raising of interest rates, the job market is still thriving. Hiring has slowed down from its peak months, but it is still happening. Maybe raising interest rates earlier would have had some kind of positive impact on that, but it is not obvious. I am not arguing it was right to wait until 2022 to raise interest rates, but I don’t know that it was wrong either.
As for Powell, I have long been a booster of his and still am. He did more than any individual to prevent another financial crisis in 2020. By guaranteeing that the Fed would do whatever it takes to keep credit flowing, businesses had confidence that the economy was not going to fall off a cliff and were willing to continue lending and borrowing. One thing I like about him is that he is not an economist. He knows plenty about economics and is hardly a bum off the streets, but he has learned a lot on the job. Not having an economics background is what I think allowed him to be bolder and more creative when the going got tough.
For the longest time, the Fed focused more on inflation than unemployment. Powell steered more of the focus towards unemployment and wound up proving that what was thought to be the natural rate of unemployment was not so. Few people remember it, but in 2019 he cut interest rates when unemployment was below 4% and inflation did not increase. Virtually no economist would have done that because they aren’t trained to. Why that episode matters is because it showed that low unemployment can keep going lower without inflation surging and that there isn’t always a need to treat the two as mutually exclusive.
If Powell can successfully tame inflation while avoiding a recession, he will be the greatest central banker of all time. Not one of the greatest, the greatest. He will have done more than anyone else to disprove the notion that there are iron laws of economics. He will also be vindicated in his decision making. The Fed has been criticized by many economists and others for having been “behind the curve” in reacting slowly to inflation. Many of the people making those criticisms are part of the crowd saying that unemployment will have to go up.[iv] It will be Powell who gets the last laugh, along with the millions of workers who don’t needlessly lose their jobs.
Uncertainty can be good
While uncertainty can be maddening, in the case of the economy, I think it is mostly good. What it is telling us is that nothing is set in stone. There are no laws in economics like there are in physics. Things that have happened before might happen again, but aren’t guaranteed to. That might sound like there is nothing we can do about anything, but it’s not. I don’t believe it’s likely that we can prevent financial crises, but we can decide how to respond to them. Downturns may be a fact of life, but slow recoveries are not.
It is very good to know that taking care of one problem doesn’t always have to mean creating another problem. Getting inflation down is important, but so is keeping people employed. Traditional macroeconomics, with a few exceptions, largely says that is impossible. We have seen over the last year that that is wrong. We have seen that jobs can be recovered fast. A sluggish recovery with high unemployment is a choice, not a law of economics that is always true.
For the record, I don’t think traditional macroeconomics should be thrown out. Economic models, like many other statistical models, are great for looking at past data. That is important in its own right, but can also be helpful in dealing with the present and the future, albeit very imperfectly. Even though traditional macroeconomics is limited in its utility now, it was good for explaining much of the pre-2020 economy and may be good again in the future. Just because it wrongly says something is inevitable doesn’t mean it can never happen again, i.e., the next inflation surge may require a recession just like in the 1980s.
As long as it was done in good faith, I don’t fault people for being wrong. Predicting the future is just not something anyone is good at no matter how experienced, smart and credentialed they are. In dealing with an unprecedented problem, everyone is going to get stumped. No subject matter knowledge or prior real world experience can truly prepare someone for something they have never dealt with.
With respect to the current economy, I am not aware of a single person who has not been wrong about something big. In the inflation debate alone, virtually everyone has been wrong about something significant. Those who argued that it would go away quickly were wrong. Those who argued it was going to be persistent usually argued that it would take a recession to bring it down and they, too, were wrong. That is inevitable in a world where things happen in unpredictable and counterintuitive ways. No matter how sophisticated someone’s model is, it can only factor in what has happened before. It’s best to take Jim Morrison’s advice and remember that the future’s uncertain and the end is always near.
[i] Several hundred at least would be needed before any real conclusions could be drawn from a sample. The bigger the sample size, the more representative it is likely to be although it is not foolproof.
[ii] Part of the inflation surge was because there was a surge in wages. With few workers available, employers had to keep paying more to hire or retain workers. The problem was wages increased at a very fast rate, which employers compensated for by raising prices. With more workers joining the workforce, the rate of wage increases is slowing down, which means slower price increases.
[iii] At its core, inflation is too many dollars chasing not enough goods/services. It sounds simple, but a ton of things go into it. Whether there are too many dollars or not enough goods/services are very different problems requiring very different responses. A big part of the debate over what caused the inflation surge in 2021 centered around whether it was too much demand fueled by stimulus measures and low interest rates or too little supply because of supply chain issues. Nobody argued it was 100% one or the other, but some argued it was mostly demand while others argued it was mostly supply.
[iv] If it was up to me, all of those saying unemployment needs to go up would be the first get laid off.