Disco inferno

On stagflation, and other risks

I wrote about stagflation last week—specifically, whether we’re in for some—and I didn’t quite get in all the things I wanted to say, so I thought I’d put some additional thoughts here. The question on people’s minds is whether a surge in energy costs is likely to push us into a period of slow growth and high inflation like much of the world experienced in the 1970s. The piece I wrote says no, although that doesn’t mean that high energy costs won’t hurt. The point, rather, is that a lot of things are different now, not least the fact that central banks know that they can bring down inflation and are committed to doing so, so whatever happens it won’t be years of high inflation alongside high unemployment.

But it also seems to me that there’s something a little off about the question itself—whether we’re heading back to the 70s—and the way many people are approaching the strange situation in which the world now finds itself. For starters, there is a disjointedness in time that makes the comparison a strange one. I know it seems like we’ve packed a thousand years into the past twenty months or so, but it is worth remembering that the beginning of the pandemic was not actually all that long ago, that we are still very much in it—there have been roughly 2,000 Americans dying of Covid every day for the past month—and that the inflationary phase of the pandemic is particularly recent. Year-on-year CPI first rose above 2% in March, which is to say seven months ago. 

To just provide a little context, the CPI rose above 2% after about eight years running below it in February of 1966. It didn’t fall back below 2% again until April. Of 1986. The Great Inflation was a very long-lived phenomenon. Even if you start the clock in 1973, it was an era, not an event.

Which is to say two things. First, we are still dealing with the acute effects of the pandemic, and barring some catastrophic mutation in the virus these acute effects will end and then we will see what post-pandemic normal looks like. It has been very difficult throughout this process to understand which wild moves in which variables can be extrapolated and which can’t, and some features of the economic environment—including inflation but also the disruptive circulation and deadliness of the virus—have gone on longer and in more intense fashion than some of us expected. At the same time, when I go back and look at what I wrote earlier this year about how inflation was likely to play out, it holds up reasonably well. Maybe it will turn out that I was wrong and inflation won’t decelerate in 2022, but we don’t know yet because we haven’t gotten there, because it’s only been a few months. It’s just that now we have forty-seven news cycles per day, so every month is a lifetime.

And secondly, in considering whether we are entering a new and potentially stagflationary era, it’s not enough to point to this or that parallel and say because those things are the same the outcomes are the same. The world is more complicated than that. The era of stagflation was challenging for governments to manage not simply because there were oil shocks. Rather, they were negotiating territory they hadn’t encountered before. The two overwhelmingly dominant economic experiences which would have been front of mind for policymakers then were, first, the Depression, and second, the wildly good postwar years. The default setting on policy, for decades, had been to provide adequate demand, and to look out for workers. There were powerful institutional forces geared in this direction.

The world we are in now is not that world. It is different because we know more about what can happen in an economy than they did. We’ve spent half a century trying to understand why things happened the way they did. We spent half a century restructuring the economy, in fundamental ways, as a consequence of what happened. It is because circumstances are wildly different now that we’re seeing a broad move toward policy tightening around the world. Budget deficits are in the process of shrinking a lot, across rich and poor countries alike. Central banks around the world are putting up interest rates, and the big ones that haven’t yet are all but certain to in 2022. 

There’s just very little to be gained from trying to force current circumstances into a 1970s narrative. We should instead be focused on trying to understand the situation we face now on its own terms, to get a better sense of the risks it poses.

If all goes well in the months ahead, and there isn’t another major Covid wave, then high energy costs will weigh on the economy, but the improving public health situation should allow supply-chain problems to slowly unkink themselves. Growth will be slower than we’d like over the winter, and if cold conditions cause lean gas supplies to bite in some countries, then there could be some very uncomfortable months in parts of Europe and Asia, but the outlook for 2022 should be pretty good. But I think many people overestimate how likely it is that this rosy scenario plays out.

In another scenario, policymakers don’t take adequate account of how tightening effects around the world compound, and central bankers, in particular, fail to look through high energy costs to the degree that they should. A soaring inflation driven upward by surges in prices for services could at least potentially be something worth keeping an eye on as a potential source of a wage-price spiral. You can tell a story about firms needing to attract workers with higher wages, and then finding that they’re able to raise prices to cover those wages because the money earned from higher prices is flowing right back into paychecks of people with a high propensity to spend. Surging food, energy and housing costs are different. They squeeze household incomes, leaving people with less money to spend on other things, and the higher returns accruing to owners of resources or rental properties or plain old houses aren’t going to be spent back into the economy nearly as readily as the bigger paychecks brought in by an Amazon delivery person or the waitstaff at a restaurant. 

So while the upward pressure on measures of inflation from the food, energy and shelter categories means that central banks are likely to continue missing their targets for months to come, they are not a reason to begin aggressively tightening monetary policy. Doing that could compound the contractionary effect of expensive food and energy and turn what is currently an uncomfortable overshoot into an uncomfortable undershoot a year from now. Wage growth across the American labor market as a whole is probably good enough now that rising food and energy costs mostly serve to hurry along the process of normalization across supply chains. (Spending on cars, historically, has been very responsive to rising gas prices.) But there is a risk that central bankers blink and overdo it.

But beyond the immediate questions about the trajectory of inflation, I wish The Discourse were less focused on whether we’re going to replay the 1970s and more focused on the potential implications of this ongoing global supply crunch. 

Not all, but most of the inflation pressures we’re experiencing in the US are associated with various supply-chain disruptions: chip shortages are a problem, as are long delays at ports, as are inadequate supplies of fuels. One thing which has been interesting to see is the way in which problems can build on themselves. So, manufacturers and retailers are worried they won’t have critical components that they need, so they over-order, which further stresses supply chains. When they over-order, they need a place to store this inventory, and when warehouse space isn’t available they simply use the containers in which the stuff they ordered arrived as onsite storage. But now that’s another container that isn’t in circulation, which adds to stresses on the system. Then, because it’s no fun to be an overworked trucker or a sailor stuck on a ship anchored off Long Beach for weeks, workers start to quit, or they demand better hours and pay, which further stresses the system.

There’s a tone in some coverage of this stuff, one I’ve adopted myself at times when writing on the whole mess, of “here, now you see how the sausage gets made, and maybe you’ll think about it more the next time you place an Amazon order”. Which is fine to say, you should think about it the next time you place an Amazon order, but perhaps also understates the seriousness of the pressure on the system. Now, it’s hardly a crisis if the board game you wanted to get your nephew for Christmas doesn’t arrive on time. But the value of global trade is nearly 60% of GDP. Lots of cheap plastic toys get shipped over the ocean, but lots of important things do too, and moreover, even the cheap plastic toys are associated with a certain amount of employment and certain financial obligations to bondholders and banks and so forth. There’s a lot of crap that gets shipped, but to a certain extent the global economy runs on the production and sale of crap.

Which is to say, we should be aware of and at least a little concerned by the possibility that stresses might at some point begin to cascade. Maybe the sailors and truckers and longshoremen say to hell with it, or maybe insufficient supplies of fuel and essential pieces of equipment reduce the throughput of nodes in the transport network beyond some threshold past which capacity reductions feed on themselves. Or maybe governments start to intervene in an effort to secure essential supplies—for coal or gas, for instance—thus touching off a round of aggressive and dangerous hoarding.

I don’t think this is particularly likely to happen, although we’ve never stress-tested modern global supply-chain networks to this extent so who knows how close we are to a critical threshold. But the broader point is that global supply chains are incredibly complex systems. Firms scattered all over the world, operating in countries with very different sorts of governments and featuring very different levels of political and institutional health, produce lots of pieces of things which must be brought together in still other countries to be assembled—and then sometimes those things that have been assembled are sent on to still other countries where they meet up with other bits that have gone on long journeys, all before a final good is produced. For shipment to final consumers. 

And we know from experience that complex systems bend in response to a little pressure, but that the application of too much pressure can lead to cascading failures and system collapse. One might point to, say, a banking system in which mortgage loans have been rolled together into pools which are chopped up into securities that are often themselves rolled into pools and chopped up into securities. Which are occasionally rolled together into pools and chopped up into securities. And also the banks have been funding themselves via unsecured short-term debt. Someone who knew nothing of this and was then told about it might think, hm, that seems kind of dangerously complex, but then they’d observe that a modest decline in home prices and a modest rise in defaults didn’t seem to do much damage, and there would be pieces written about how the sausage gets made, up to the point where continued declines in home prices and increases in defaults very nearly blow up the global financial system. 

Again, I am not of the view that a collapse of the global trading system is imminent. But this seems like an understudied sort of risk. And furthermore, there are two things about the present mess which give me pause. One is that it seems like it’s not particularly easy to roll back supply-chain trade just a little bit. A through line in most of the pieces you read about supply-chain woes is that people in boardrooms talk about decoupling from China, but there are no good alternative options, particularly which can be brought online within a year or two. Or they move production from China to Vietnam, which ends up not mattering much because all the components are still made in China. These networks are, again, incredibly complex, and have been built up over decades through accumulation of know-how and capital investment. There may be enormous flexibility in what can be produced along existing chains, even as there is very little flexibility, at least over the near term, in how those chains are structured. And a very large share of them run through an authoritarian state which, we’re told, might invade a neighboring rich democracy within the next few years, which might not bother Tim Cook very much but which would I think put other rich democracies in a position where they could not very well continue trading with China on normal terms.

The other thing which gives me pause is how much of a role extreme weather has played in aggravating these disruptions. Food prices are high, in part, because drought and extreme heat have damaged harvests. Powerful storms have disrupted shipping. China has ordered its mines to churn out a lot more coal, but production has been hit by massive floods in Shanxi province. This is, I think, another reason to be sceptical when economists tell us that climate change is likely to shave just a few points off of global output by 2100. Because we are surrounded by complex systems which operate on the unstated assumption that the climate will be much as it has been for the past 10,000 years, but now, it turns out, it won’t. And it seems unreasonable to me to expect that as climate-related disruptions intensify the operation of our political and economic systems degrades smoothly, rather than smoothly and then all at once.

That’s all very doom and gloomy. But it bears thinking about these things, if only to try to work out how likely they are to happen and what might be done to limit their worst effects. But hey, the good news is: we’re not in for a replay of the 1970s.