Mistakes, I've made a few

On getting inflation wrong, and on trying to get the future right

Well, shit. 

I think the time has come to concede that I have gotten some big things about the inflation situation wrong. In my defense, we got a huge and unexpected shock in Russia’s invasion of Ukraine. And I have also gotten some big things right. The supply-chain problems that really constrained the economy in 2021 have relaxed considerably, and we have good reason to think that this will weigh on inflation in the months ahead. We don’t appear to have entered a new macroeconomic regime; neither market-based measures of inflation expectations or long-term Treasury yields ever rose by very much, and while wage growth remains high, we haven’t seen steady acceleration, of the sort that might indicate a wage-price spiral. Rate hikes have been really bad for the emerging world, and will continue to be.

But I did not expect inflation to rise as high as it has, I did not expect it to take so long for inflation to peak and decelerate, and I did not think that economic activity in the United States would be this resilient in the face of such aggressive Fed tightening. I got that stuff wrong, and other people got it right. Mea culpa.

Why did I get it wrong? I think I have to allow that it was partly about cognitive error. I can be bloody-minded about things, especially when it feels like people are converging around a particular narrative on the basis of insufficient information; that’s a good trait sometimes but not always, and I try to adjust for this predisposition but I don’t always adjust enough. In addition, I think I screwed up by assuming that we couldn’t get a sustained inflation problem without underlying structural change: like a meaningful shift in labor power. And maybe I also got sloppy, thinking for instance that because some of the people warning about inflation had bad arguments that the case for concern about inflation overall was weak. 

In any case, I wish I had done better, and I will try to do better in the future. And kudos to those who got it right.

In the light of all that, what insight can I offer on where we are now, what’s coming, and what we ought to take away from this whole experience? With humility, here are a few thoughts.

First, it seems important to acknowledge that fiscal policy is really powerful. There will be people who argue that the lesson of the past two years is that fiscal policy is dangerous and can lead to overheating, and it certainly is the case that too much of a fiscal boost in too short a time in the face of constrained supply will generate a sharp rise in inflation. But while the Great Recession and the Covid recession were very different animals, it seems very reasonable to conclude from recent experience that the recovery from the global financial crisis could have been far more rapid, and the broader social costs of the crisis much reduced, had the fiscal response been more aggressive. It will be tempting for some observers to read these two experiences as providing mixed messages about using fiscal policy—that it can be good but it can also be bad, or that we should use it but only reluctantly—and that seems an incorrect reading to me. Both experiences demonstrate that fiscal policy is a good thing to use, and we just need to work on using it more effectively.

Second, central banks still have to do their jobs. The role of the central bank is to coordinate people’s expectations around the future path of demand growth. It might have been reasonable to hope in early 2021 that credibility alone would get this job done, but as inflation rose above central banks’ projections over the course of last year, the Fed needed to adjust its communications and policy to get everyone on the same page, expectations-wise. I don’t think the Fed was wrong in its assessment that transitory factors were pushing up inflation. But powerful transitory forces aren’t a reason to say, in effect, “eventually we will get to where we need to be”; that leaves far too much room for people to form expectations about the future based on other information around them: like soaring asset prices or the fact that everyone they know is buying a ton of stuff or what have you. Rather, transitory forces make it all the more important to be clear about what the Fed wants to see, what it is prepared to do to get there, and then also what it will do if unexpected developments mean that realized inflation misses in one direction or the other. Mea maxima culpa here, as I was definitely *not* encouraging the Fed to do this late last year. 

Third, while we have ended up with more inflation than we wanted, it was absolutely appropriate for inflation to rise well above 2% for a sustained period of time. Too much demand is too much demand, fine. But it is also the case that we have seen wrenching shifts over the past three years in both the composition of demand and in the capacity of the global economy to give consumers what they want. This has entailed enormous shifts in relative prices, and in a few cases those relative-price shifts have made something like a full round trip: way up and then way down or vice versa. The process of managing these shifts has been made much easier by inflation (which allows the real price of stuff to fall farther, faster than would otherwise be the case). But for that, the adjustment process would probably have meant more unemployment and harder times for businesses.

Many people go out of their way to mention how painful inflation is for lots of households, what a bad thing it is—as bad as or worse than recession, etc—and here we need to be clear that this is not quite right. People are hurting, yes, because some things that they need are in very short supply, and as a consequence they can’t buy as much of those things with their available income as they’d like, and the economy as a whole operates much less productively than it otherwise could. The problem we’re facing is real, actual scarcity in the world: of energy, of food, of port capacity, of chips and so on. Making all the excess inflation go away wouldn’t solve that underlying scarcity; it would simply redistribute the pain it causes relative to what we’ve got now. 

It is in some sense understandable that someone with lots of job security would prefer a different macroeconomic policy approach, in which the cost of scarcity was imposed on a relatively small group of people whose disposable income was crushed by the loss of a job, such that those who kept their jobs could buy more things with the income they continue to earn. But it is not necessarily macroeconomically or morally good to give those people what they want, and it is at least a little misleading to say that what’s annoying them is inflation, rather than the underlying problems of scarcity. In the face of big supply-chain problems and a massive food and energy shock, you could be hitting an inflation rate of 2% bang on and people would still be really upset with the state of the economy. 

Fourth, it is probably unreasonable to expect any central-bank target to be able to perfectly accommodate the wild stuff that happened over the past three years, but it seems fair to say that the target we’ve got accommodated the wild stuff less well than alternatives might have. A target for nominal wages or nominal GDP would have allowed the Fed to accept a lot of inflation within its policy framework, and it would have been nice if the Fed had chosen one of those when it updated its policy framework back in 2020, rather than “flexible average inflation targeting”.

Fifth, and relatedly, the decision to stick with an inflation target is going to mean more short-run macroeconomic volatility than is really necessary. I have a feeling that GDP figures for the first two quarters of this year will end up being revised by quite a lot, but for the moment it appears that nominal GDP, not adjusted for inflation, expanded at an annual rate of about 8% in the second quarter of this year. One could decompose this into a rise in inflation of more than 8%, plus a contraction in real output. The Fed wants to get inflation back down to 2%, and it feels that it needs the real economy to be even weaker, such that unemployment rises, in order to achieve this. 

So if we’re unlucky (more on this in a moment), we may end up in a situation in which the Fed continues to choke off real economic activity, and we hit 2% inflation at a level of nominal GDP growth of 2% (implying zero growth across the real economy) or 1% (implying a contraction in real GDP of 1%) or lower still. That’s a huge swing in demand: from 8% growth in nominal GDP to 1%! If the Fed were instead aiming to achieve growth in nominal GDP of 5%, and was agnostic about whether that was mostly real growth or mostly inflation, then it could be happy engineering a decline in nominal GDP growth of just three percentage points. In the context of a growth-busting energy shock, that still wouldn’t be comfortable, exactly; it might well mean we’re living with zero growth and 5% inflation for a little while. But as long as the Fed is serious about hitting 5% nominal GDP growth on a sustained basis, that’s ok; the Fed is inducing a much lower amount of volatility, it’s not trying to guess how transitory a particular inflationary force is or communicate information about its guesses to the public, and it’s not unnecessarily punishing people by throwing them out of work because Russia invaded Ukraine.  

Of course, point six, it is possible that we end up being lucky, because global economic factors reduce inflationary pressures in the US, such that the Fed feels that it needs to induce less domestic unemployment to meet its goals. The wonderful scenario in which this happens is Russia halts its war and deposes Putin and maybe we discover cheap fusion and a bunch of grain we didn’t know we had or something. That’s the less plausible way. The more plausible way is that we “accidentally” use our power as the world’s monetary hegemon to destroy global growth and, voila, inflationary pressures abate. 

There is only so much available oil or wheat or manufacturing capacity to go around, and when lots of economies around the world are doing well there is a lot of competition for those things and the prices of oil and wheat and manufactured goods rise. When lots of economies aren’t doing so well, there is less competition and prices don’t rise so much. For instance, prices for natural resources like oil have gone up by less this year than they could have, because China is very badly managing its property-market problems and its Covid policies and is thus doing very poorly, economically speaking. But there are other things which can hurt global growth, too, and it turns out that tightening by the Federal Reserve is one of them. The dollar is the world’s main reserve currency, Treasury bonds are the world’s most important safe asset, and when the Fed tightens that usually translates into harsher global financial conditions and a rising dollar. 

Now, there’s no iron law that says that the global effects of Fed tightening are more powerful than the effects of tighter policy on America’s domestic economy. But it happens to be the case that America’s recovery from the Covid recession has been particularly robust, and American balance sheets are relatively solid, whereas much of the rest of the world has not recovered as much and is not in the best financial situation and is really struggling with the dual pressures of high food and energy prices and rising interest rates in the US. And so it seems likely that a larger than normal share of disinflation within the US economy may be accounted for by falling world prices for food, energy and other goods and commodities, which is in turn associated with widespread economic hardship across low- and middle-income countries especially. In other words, it is relatively hard to get an American to stop spending right now, but comparatively easier to force residents of struggling middle-income countries to scale back on their consumption by inducing a massive depreciation in their currencies, and so a given amount of the former will come with a larger than normal amount of the latter.

Seventh (sorry, this is starting to get Martin Wolfish), in terms of inducing reductions in spending, Fed policy is also having some other perverse, though totally foreseeable, effects. Inflation is high in part because the cost of housing has gone up a lot, which is partly about increased housing demand but also very much about the fact that we’ve built too little housing in recent years. Now, spending on new housing construction is spending, and if there is too much spending in the economy, then reducing spending on new housing construction is a way to solve your short-run problem of too much spending. But if the longer-run problem is that there is too little housing, then it does seem like kind of a stupid way to tackle the short-run problem. Unfortunately this is what we’re doing; have a look at this chart from Bill McBride at Calculated Risk, which shows the long period of underinvestment in new housing after the global financial crisis, then an upward jump in new housing starts in recent years, followed by a sharp jag downward as the Fed stops the party.

And this is a story which is about more than housing. In the context of persistently weak demand in the 2010s, we got too little investment, which meant that too much of the surge in demand over the past two years could not be met by increased production and instead ended up as inflation. But because investment spending is more cyclical than consumption, an induced downturn will hit investment harder than consumption, “solving” the short-run problem by exacerbating the long-run problem.

Which leads me to my eighth and final point. Big macroeconomic gyrations usually yield important macroeconomic lessons. We don’t know for sure what the lessons of this period will be; that will depend on how exactly this tightening cycle plays out, and also on which intellectuals are most successful at framing the challenges we’ve faced. But one possibility is that the big lesson of this mess is in some ways an extension of the lessons of the global financial crisis and its aftermath. 

After the Great Inflation, there was a push to get most of government out of the business of macroeconomic management, and to make macro stabilization as simple and automatic as possible: thus independent central bankers aiming at public targets, often by following simple policy rules. The global financial crisis and the weak ensuing recovery reduced the appeal of this approach, because it turned out that after leaving everything to central banks you could still end up with massive recession and a decade of pitifully weak growth. But the light-touch approach is an attractive way of structuring things in a world where governments can’t seem to do anything right, and so one might naturally conclude that it is an attractive way of structuring things here and now.

But a world in which we don’t trust the government to do anything is not a world in which we are able to solve our big problems, and we should therefore not be very happy with that world. Instead we should think how best to develop the capacities within government that we need government to have. And part of that is taking a more purposeful and holistic approach to business-cycle management. 

Both supply and demand matter, and because supply matters the composition of demand matters: when we have to cut back, we need to cut back on unnecessary consumption before we cut back on basic spending or on investment. We need to think about the global context; crushing emerging-world economies on the heels of a devastating pandemic and while those economies face rising costs of climate change is really bad. It pushes the world toward horrific humanitarian catastrophes, and it also hurts us over the long run because countries which are rich and friendly to us augment our domestic production much more than countries which are poor and hostile to us. It is fine to want central banks to pursue straightforward targets in transparent ways, but that doesn’t relieve the rest of the government of its macroeconomic responsibilities. 

And it seems just possible that the government could meet them. I know the past few years haven’t gone exactly as one would want them to go. But we should acknowledge that in the face of very difficult circumstances, even the government of the United States of America did some pretty good stuff. It sped along the process of vaccine development. It engineered a massive fiscal response to the massive Covid recession, which helped spark an historically rapid economic recovery. And in the face of supply-chain problems and inflation, it has passed measures to boost investment and may well enact a very sensible, demand-reducing tax measure through the Inflation Reduction Act.

We could clearly improve on this. But we are working our way back from an extremely counterproductive era of intense hostility to the idea that the government might be able to help an economy function better. And we are also struggling through a difficult political moment which is probably best seen as a consequence, at least in part, of that era. Let’s learn from our mistakes and try harder.