At Barron’s: Thank Full Employment, Not AI, for Rising Productivity

(I write a monthly opinion piece for Barron’s. This one was published there in September. My previous pieces are here.)

New data about productivity are some of the best on record in recent years. That’s good news for economic growth. But just as important, it offers support for the unorthodox idea that demand shapes the economy’s productive potential. Taking this idea seriously would require us to rethink much conventional wisdom on macroeconomic policy. 

Real output per hour grew 2.6% in 2023, according to the Bureau of Labor Statistics, exceeding the highest rates seen between 2010 and the eve of the pandemic. That said, productivity is one of the most challenging macroeconomic outcomes to measure. It is constructed from three distinct series—nominal output, prices, and employment. Short-term movements often turn out to be noise. It’s an open question whether that high rate will be sustained. But if it is, that will tell us something important about economic growth. 

Discussions of productivity growth tend to treat it as the result of unpredictable scientific breakthroughs and new technologies, whose appearance has nothing to do with current economic conditions. This view of technological change as “exogenous,” in the jargon, is entrenched in economics textbooks. And it’s reinforced by the self-mythologizing barons of Silicon Valley, who are only too happy to take credit for economic good news. 

The economic conditions that lead companies to actually adopt new technologies get much less attention, as does the fact that much productivity growth comes from people shifting from lower-value to higher-value activities without the need for any new technology at all.

A recent New York Times article is typical. It discusses faster productivity growth almost entirely in terms of the new technologies — AI, Zoom, internet shopping — that might, or might not, be contributing. Not until 40 paragraphs in is there a brief mention of the strong labor market, and the incentives that rising wages create to squeeze more out of each hour of labor.

What if we didn’t treat this as an afterthought? There’s a case to be made that demand is, in fact, a central factor in productivity growth. 

The economic historian Gavin Wright has made this case for both the 1990s — our modern benchmarks for productivity success stories — and the 1920s, an earlier period of rapid productivity growth and technological change. Wright considers the adoption of general-purpose technologies: electricity in the ‘20s and computers in the ‘90s. Both had existed for some time but weren’t widely adopted until rising labor costs provided the right incentives. He observes that in both periods strong wage growth started before productivity accelerated. 

In the retail sector, for instance, it was in the 1990s that IT applications like electronic monitoring of shelf levels, barcode scanning and electronic payments came into general use. None of these technologies were new at the time; what had changed was the tight market for retail employment that made automation worthwhile.

The idea that demand can have lasting effects on the economy’s productive potential – what economists call hysteresis — has gotten attention in recent years. Discussions of hysteresis tend to focus on labor supply — people dropping out of the labor market when jobs are scarce, and re-entering when conditions improve. The effect of demand on productivity is less often discussed. But it may be even more important.

After the 2007-2009 recession, gross domestic product in the U.S. (and most other rich countries) failed to return to its pre-recession trend. By 2017, a decade after the recession began, real GDP was a full 10% below what prerecession forecasters had expected. There is wide agreement that much, if not all, of this shortfall was the result of the collapse of demand in the recession. Former Treasury Secretary Larry Summers at the time called the decisive role of demand in the slow growth of the 2010s a matter of “elementary signal identification.” 

Why did growth fall short? If you look at the CBO’s last economic forecasts before the recession, the agency was predicting 6% growth in employment between 2007 and 2017. And as it turned out, over those ten years, employment grew by exactly 6%. The entire gap between actual GDP and the CBO’s pre-recession forecasts was from slower growth in output per worker. In other words, this shortfall was entirely due to lower productivity. 

If you believe that slow growth in the 2010s was largely due to the lingering effects of the recession — and I agree with Summers that the evidence is overwhelming on this point — then what we saw in that decade was weak demand holding back productivity. And if depressed demand can slow down productivity growth, then, logically, we would expect strong demand to speed it up.

A few economists have consistently made the case for this link. Followers of John Maynard Keynes often emphasize this link under the name “Verdoorn’s law.” The law, as Keynesian economist Matias Vernengo puts it in a new article, holds that “technical change is the result, and not the fundamental cause of economic growth.” Steve Fazzari, another Keynesian economist, has explored this idea in several recent papers. But for the most part, mainstream economists have yet to embrace it. 

This perspective does occasionally make it into the world of policy debates. In a 2017 report, Josh Bivens of the Economic Policy Institute argued that “low rates of unemployment and rapid wage growth would likely induce faster productivity growth.” Skanda Amarnath and his colleagues at Employ America have made similar arguments. In a 2017 report for the Roosevelt Institute, I discussed a long list of mechanisms linking demand to productivity growth, as well as evidence that this was what explained slower growth since the recession.

If you take these sorts of arguments seriously, the recent acceleration in productivity should not be a surprise. And we don’t need to go looking for some tech startup to thank for it. It’s the natural result of a sustained period of tight labor markets and rising wages.

There are many good reasons for productivity growth to be faster in a tight labor market, as I discussed in the Roosevelt report. Businesses have a stronger incentive to adopt less labor-intensive techniques, and they are more likely to invest when they are running at full capacity. Higher-productivity firms can outbid lower-productivity ones for scarce workers. New firms are easier to start in a boom than in a slump.

When you think about it, it’s strange that concepts like Verdoorn’s law are not part of the economics mainstream. Shouldn’t they be common sense?

Nonetheless, the opposite view underlies much of policymaking, particularly at the Federal Reserve. At his most recent press conference, Fed Chair Jay Powell was asked whether he still thought that wage growth was too high for price stability. Powell confirmed that, indeed, he thought that wage gains were still excessively strong. But, he said, they were gradually moving back to levels “associated — given assumptions about productivity growth — with 2% inflation.”

The Fed’s view that price stability requires limiting workers’ bargaining power is a long-standing problem. But focus now on those assumptions. Taking productivity growth as given, unaffected by policy, risks making the Fed’s pessimism self-confirming. (This is something that Fed economists have worried about in the past.) If the Fed succeeds in getting wages down to the level consistent with the relatively slow productivity growth it expects, that itself may be what stops us from getting the faster productivity growth that the economy is capable of.

The good news is that, as I’ve written here before, the Fed is not all-powerful. The current round of rate hikes has not, so far, done much to cool off the labor market. If that continues to be the case, then we may be in for a period of sustained productivity growth and rising income.

New Paper: Rethinking Supply Constraints

I have a new paper on how we conceptualize the supply side of the economy, coauthored with Arjun Jayadev. I presented a version of this at the Political Economy research Institute in December 2022. You can watch video of my presentation here — I come on, after some technical difficulties, around 47:00. (The other presentations from the conference are also very worth watching.) The paper will be published in the upcoming issue of the Review of Keynesian Economics. (The linked version is our draft; when the published version comes out I’ll post that.)

Our fundamental argument is that while macroeconomic supply constraints are normally conceptualized in terms of a level (or level-path) of potential output, in many contexts it would be better to think in terms of a constraint on the rate of change — a speed limit rather than a ceiling.

While this is a general argument, it’s motivate by the experiences of the pandemic and the post-financial crisis recovery of the preceding decade. We think the speed-limit conception of supply constraints makes sense of a number of macroeconomic developments that are hard to make sense of in the conventional view.

First, deviations in output are persistent. We saw this clearly in the wake of the Great Recession, but it seems to be a more general phenomenon. There’s a long-standing empirical finding that there’s no general tendency of output to return to its previous trend. One way we could explain this is the real business cycle way — short-term as well as long-term variation in output growth are driven by changes in the economy’s productive capacity. But of course, there is lots of evidence that business cycles are driven by demand. Alternatively, we could argue that potential grows steadily but actual output may remain far below it indefinitely. I was making arguments like this a few years ago. The problem is that direct evidence on the output gap (unemployment, growth in wages and prices, businesses’ reported capacity utilization rates, etc.) suggest that the output gap did close over the course of the 2010s. So we’re left with the idea that potential output adjusts to actual output — hysteresis. But if we take this idea seriously, it rules out the conventional idea of a level of potential output. In a world where hysteresis is important, a zero output gap is consistent with lots of different level-paths of output; supply constraints only bind the speed of the transitions between them.

Second, there’s no well-defined level of full employment. (Here we have to ding Keynes a bit.) Employment grows steadily over business cycles — there’s no sign of convergence to some long-term trend. Estimates of the NAIRU or natural unemployment rate follow actual unemployment more or less one for one. And if we try to make a bottom-up estimate of full employment — what fraction of the population could plausibly be engaged in paid work — we end up with a value much higher than actual employment even at cyclical peaks.

Third, we observe inflation and other signs of supply constraints in response to changes in the composition of output and employment, and not just in the level. This has been very clear during the pandemic, but there’s good reason to think it’s true in general.

Fourth, increasing returns are pervasive in real economies. This is a bit of a different argument than the first three, since it’s not pointing to a directly observable macro phenomenon. But it’s important here, because it means that we can’t assume that businesses are already using the lowest-cost technique and increasing output will cause unit costs to rise. One way of thinking about this is to imagine a cost landscape that is rugged, not smooth. Moving from one locally low-cost position to another may require traversing a higher cost region, which will appear as supply constraints during the transition. A clear example of this is the transition from carbon to renewable energy sources.

We also argue that this perspective is more consistent with a sociologically realistic view of what “the economy” is. Real economies are not homogeneous “factors” being added to a “production function” which then spits out some quantity of output. They are complex systems of cooperation between human beings, which are embedded in all kinds of other social relationships and the reproduction of households and other social organisms. These relationships cannot be torn up and recreated at any moment — changing them is costly. They evolve only gradually over time. From this point of view, it is wrong to divide the facts about the economic world into a set of long-run, fundamental, exogenous factors and short-run endogenous factors. Who is actually working, and at what, is as much a part of the economic data, no less easily shifted, than the number of people who are potentially available for work.

This way of thinking about the supply side has several implications for policy. First, rising prices and other signs of supply constraints cannot be taken as evidence for the long-run limits on the economy’s productive potential. In general, we should be skeptical of suggestions that recent rises in the prices of energy, food and other essential commodities reflect the “end of abundance”.

On the positive side, our view suggests that the response to positive output gaps should include not only conventional “supply side” measures, but measures to overcome the coordination and information problems and other frictions that limit rapid changes in productive activity. This implies planning of some sort, though not necessarily central planning in the traditional sense. Another implication is that because prices can adjust more quickly than productive activity can (the emphasis on price stickiness is backward in our view), rapid shifts in activity can generate large price spikes that are not informative about long-run production possibilities and produce undesirable shifts in income. This suggests that price regulation has an important role in smoothing the transition fro one pattern of activity to another.

Specific examples and evidence on all these points are in the paper. You should read it! A final point I want to emphasize here is that we are not saying that supply constraints are limits on adjustment speed in an absolute, universal sense. We are saying that insofar as we need a simple, first-cut description of the supply side, we will usually do better to imagine a constraint on adjustment speed rather than on the level of output and employment.

 

Slides on “Rethinking Supply Constraints”

On December 2-3, 2022, the Political Economy Research Institute at the University of Massachusetts-Amherst (where I did my economics PhD) will be hosting a conference on “Global Inflation Today: What Is To Be Done?”1

I will be speaking on “Rethinking Supply Constraints,” a new project I am working on with Arjun Jayadev. Our argument is that we should think of supply constraints as limits on the speed at which production can be reorganized and labor and other resources can be reallocated via markets, as opposed to limits on the level of production determined by “real” resources. The idea is that this makes better sense of recent macroeconomic developments; fits better with a broader conception of the economy in terms of human productive activity rather than the exchange of pre-existing goods; and points toward more promising responses to the current inflation.

I was hoping to have a draft of the paper done for the conference, but that is not to be. But I do have a set of slides, which give at least a partial sketch of the argument. Feedback is most welcome!

Demand, Supply, Both, or Neither?

One way current debates over inflation sometimes get framed is whether it’s driven by supply or demand. Critics of the ARPA and other stimulus measures point to various lines of evidence to suggest that rising prices are coming from the demand side, not the supply side; and of course you can find the opposite arguments among its defenders. This sometimes gets conflated with the question of how persistent inflation is likely to be, with a preference for supply-side explanations putting you on “team transitory.”

In my opinion, the question in this form is not well specified. It makes no sense to ask if price rises are driven by supply or by demand. A mismatch between aggregate demand and aggregate supply is one explanation for inflation. To say “inflation would be lower if aggregate demand were lower” is exactly the same statement as “inflation would be lower  if aggregate supply were higher.” This story is about the difference between the two.

The first step, then, is to think about how we can reframe the question in a meaningful way. The issue of specific indicators is downstream from this, as is the question of how long one might expect higher inflation to last. 

Here are some quick thoughts on how we might clarify this debate. 

1. Insofar as we are explaining price changes in terms of aggregate demand and aggregate supply (or potential output) this is a story about an imbalance between the two of them. If we are using this framework, any change in inflation is fully explained by supply *and* fully explained by demand. To turn this into an either-or question, we have to pose an explicit counterfactual. For example, we might say that current spending levels would have led to no (or less) rise in inflation if it weren’t for pandemic. Or we might say that the disruptions of the pandemic would have led to no (or less) rise in inflation if it weren’t for the stimulus bills. The problem is, these aren’t alternatives — both  might very well be true!

2. If the question is specifically whether current aggregate demand in the US would be inflationary even without the pandemic and Ukraine war, it seems to me that the answer is unequivocally No. The fact that real GDP is no higher than trend is, to me, absolutely decisive here. Suppose we thought that two years from now, the economy operating at normal capacity will be capable of producing a certain quantity of cars, houses, tv shows, haircuts, etc. Now suppose two years pass, and we find that people are in fact buying exactly the quantity of cars, houses, tv shows, haircuts, etc. that we had predicted,. If inflation has nonetheless risen, the only possible explanation, within the supply-vs-demand framework, is that the productive capacity of the economy is less than we expected. If buying a certain quantity of stuff is not inflationary in one year, but is inflationary in a later year, then (within this framework) by definition that means that potential output is lower.

Of course it may be true in this scenario that the nominal value of spending will be higher. But this precisely because prices have risen. Suggesting that higher nominal expenditure explains higher inflation is arguing in a circle — it is using the rise in prices to explain that same rise in prices.

To put it another way, aggregate supply or potential output are describing the quantity of stuff we can produce. It makes no sense to say that the potential output of the US economy is $22 trillion dollars. People who look at nominal expenditure in this context are just confused.

(It’s also worth noting that nominal GDP was below pre-pandemic trend until the last quarter of 2021, at which point inflation had been above target and accelerating for a year already. So this story fails on the basis of timing as well as logic.)

So in this specific sense, I think the supply story is simply right, and the demand story is simply wrong. There is no reason to think that the aggregate quantity of goods and services people are trying to purchase today would be beyond normal capacity limits in the absence of the pandemic.

3. Again, though, it all depends on the counterfactual. It does not contradict the preceding point to say that if the ARPA had been smaller, inflation would be lower. Given a fall in the economy’s productive capacity, you are going to see some combination of lower output and income, and higher inflation, with the mix depending on the extent to which demand also falls. Again, demand and supply are two sides of the same story. It’s perfectly consistent to say that in the absence of the pandemic, today’s level of spending would not have caused any rise in inflation, and that if we had allowed spending to fall in line with the fall in potential, the supply disruptions would not have caused any rise in inflation. 

This means the question of whether ARPA was too big is not really a question about inflation as such. It is not going to resolved by any data on whether inflation is limited to a few sectors or is broader, or whether inflation peaked at the end of 2021 or at some later date. The answer to this question depends on how we weigh the relative costs of rising prices versus lost income and output. The more socially costly you think inflation is, the more you are going to think that ARPA was too big. The more socially costly you think unemployment and the associated loss of income is, the more you are going to think the ARPA was the right size, or too small. It seems to me that this is what a lot of the debate over “supply” versus “demand” stories of inflation are really about.

4. So far, I’ve been using an aggregate supply and aggregate demand framework, which is how people usually talk about inflation. But as readers of this blog will know, I am generally skeptical that rising prices are best thought of at the aggregate level. If we don’t like the aggregate framework, we might tell micro stories. There are several flavors of these, any or all of which which could be true. 

First, there is there’s a story about changes in the composition of demand. It’s easy for a business or industry produce less than it usually does, but hard to produce more — especially in a hurry. So sectors of the economy that face reduced demand are likely to respond with lower output, while sectors that face increased demand are likely to respond in some large part with higher prices, especially if the increase is large and rapid. That means we should expect rapid shifts in demand to be associated with higher inflation, even if the total volume of demand is unchanged. 

That’s one micro story. Another is that when certain sectors of the economy face supply constraints, it may be hard to substitute elsewhere. If demand for the sectors facing bottlenecks is very inelastic, and/or they are important inputs for other sectors, then the fall in capacity in those sectors may have a larger effect on prices than a similar across-the-board fall in productive capacity would.  

Another, simpler micro story is that there is no useful information in the aggregate price level at all. If prices are rising for particular goods and services, that is best understood in terms of production conditions and demand for those particular things.

What these stories suggest is that the aggregate supply/demand framework is less useful when there are large, rapid shifts in the composition of spending or production. That framework may be reasonable when we are talking about an economy undergoing steady growth and want to know if somewhat faster growth (spurred perhaps by across the board easier credit) would lead to higher inflation. But it’s not very useful when the economy is undergoing major qualitative changes. It’s not even clear how the concept of aggregate supply is defined when we are seeing big shifts in the composition of output. 

To be clear, none of these micro stories are necessarily arguments for a “supply side” explanation of inflation. Rather, they are reasons why the supply versus demand framework might not be helpful. 

5. Another story about inflation that often gets conflated with aggregate demand but should not be is the extent to which higher wages are driving inflation. In the standard textbook story, the mechanism by which higher demand raises prices is through higher wages. In the textbook story — this is totally mainstream — there are no constraints on the supply side except the supply of labor. The reason higher demand leads to higher inflation is that lower unemployment leads workers to obtain higher wages, which then get passed on to prices. We know this is the theory the Fed is working with. And people like Krugman are saying that there’s no way to have lower inflation without “getting wage growth down considerably.” 

But even though these two things are linked in the textbook story, they are logically distinct. When we talk about a demand-side story of inflation, we need to be clear whether and to what extent this is a story about wages specifically. And as Krugman emphasizes in the linked piece, the question of whether reducing inflation will require slower wage growth is logically independent of the question of whether higher wages are what has driven rising inflation so far.

6. So what is actually at stake here? Before we start talking about who is on what team, we should be sure we know what game they are playing, and where the goals are.

It seems to me that, operationally, the big question people are arguing about is: if real activity — and in particular the labor market — remains on its current trend, will inflation eventually come down on its own? Or if not, can interventions in specific sectors reduce it? Or is a significant slowing of activity across the board required? This forward-looking question is what the fighting is really about. 

There is some relationship between these questions and the causes of inflation, but they are not the same. For example, it is certainly more plausible that resolving a small number of bottlenecks would bring prices down when the price increases are concentrated in a few sectors. But one might also believe — as I and others have long argued — that we underestimate the flexibility of the supply side of the economy in general. If you believe in hysteresis, then even a general overheating might over time lead to faster growth of potential output, as more people are drawn into the labor force and businesses invest and raise productivity. So unless you start from the premise that potential output is normally fixed, “inflation is widespread, therefore we need less demand” is a non sequitur.

(In this light, it would be worth systematically revisiting the arguments about hysteresis made by Lawrence Summers and others during the 2010s.)

7. The question of whether reducing inflation requires lower demand and weaker labor markets is inextricable from your political program and broader worldview. At this moment, we are on the verge of seeing a major new public investment bill that is being presented as a way of bringing down inflation and creating good jobs. This is a huge vindication for arguments that progressives have been putting forward for years. It would be odd if we now turned around and said, “no no, investing in clean energy won’t help with inflation. And we don’t want more jobs, there’s too much demand in the economy already.” If you think that the clean energy investments in the IRA will in fact bring down prices over time, you need an understanding of inflation that’s consistent with that. And if your view of inflation implies that we would be better off not making those investments at this time, then you need to own that position.

(Some people will say the bill is anti-inflationary because it lowers the deficit. First of all, it seems unlikely that the specific tax increases in the IRA will have much if any dampening effect on demand, while the spending components certainly will boost it. Second, even if you thought the bill as a whole will reduce inflation, you still need to have a view on whether the energy provisions specifically will do so.)

I’m not saying you should just make the economic arguments that support your political program. It’s very important to only say things that are supported by logic and evidence! But presumably, you have the worldview you do because it captures important things you think are true about the world. If your analysis of inflation is not consistent with other things that you strongly believe, that suggests that there might be something wrong with your analysis. 

For example, you might believe that potential employment in the US is much higher than conventional estimates of unemployment or the labor force suggest. You could arrive at this belief on the basis of statistical evidence and also from other beliefs that you are strongly committed to — for example, that women and non-white people are just as capable of useful work as white men are. This argument runs directly against claims that the US is currently facing hard supply constraints, so that the only way the growth in wages and prices will moderate is via lower demand. If the conventional unemployment rate vastly understated the number of people available for work in 2015, presumably it still does today. You can’t just ignore those earlier arguments when talking about current inflation.

Similarly, you might believe that business investment and productivity growth depend on demand. Or you might think they depend on decisions made at the firm level — that corporations face a choice between long-term growth and short-term returns to shareholders, which they will make differently in different institutional and legal environments. These long-standing arguments are relevant to the question of whether it’s plausible that corporate profiteering is contributing to current price rises, and whether changes to taxes and regulation could bring down inflation without any need to reduce demand. They are also relevant to the question of whether given sustained strong demand, supply will eventually catch up, via higher investment and faster productivity growth. This Gavin Wright paper argues that scarce labor and rising wages drove the acceleration of productivity growth in the late 1990s. How convincing you find that argument should be relevant to your assessment of inflation today.

Or again, the question of whether conventional monetary policy should be the go-to response to inflation is not (only) a question about inflation. You could have a fully “demand side” account of rising rents — that they are entirely driven by rising incomes, and not by any change in the housing supply — and yet also believe that higher interest rates, by discouraging housing construction, will only make the problem worse. 

It’s a big mistake, in my opinion, to debate inflation in isolation, or to think that debates over inflation are going to be resolved with statistical tests. We first need to step back and think carefully about what question we are trying to answer, and about what account of inflation is consistent with our broader intellectual commitments. The reason I disagree with someone like Jason Furman about inflation isn’t because I have a different read on this or that data series. (I like his empirical work!) We see inflation differently because we have different ideas about how the world works. 

Alternative Visions of Inflation

Like many people, I’ve been thinking a bit about inflation lately. One source of confusion, it seems to me, is that underlying concept has shifted in a rather fundamental way, but the full implications of this shift haven’t been taken on board.

I was talking with my Roosevelt colleague Lauren Melodia about inflation and alternative policies to manage it, which is a topic I hope Roosevelt will be engaging in more in the later part of this year. In the course of our conversation, it occurred to me that there’s a basic source of confusion about inflation. 

Many of our ideas about inflation originated in the context of a fixed quantity money. The original meaning of the term “inflation” was an increase in the stock of money, not a general increase in the price level. Over there you’ve got a quantity of stuff; over here you’ve got a quantity of money. When the stock of money grows rapidly and outpaces the growth of stuff, that’s inflation.

 In recent decades, even mainstream economists have largely abandoned the idea of the money stock as a meaningful economic quantity, and especially the idea that there is a straightforward relationship between money and inflation.

Here is what a typical mainstream macroeconomics textbook — Olivier Blanchard’s, in this case; but most are similar — says about inflation today. (You can just read the lines in italics.) 

There are three stories about inflation here: one based on expected inflation, one based on markup pricing, and one based on unemployment. We can think of these as corresponding to three kinds of inflation in the real world — inertial, supply-drive, and demand-driven. What there is not, is any mention of money. Money comes into the story only in the way that it did for Keynes: as an influence on the interest rate. 

To be fair, the book does eventually bring up the idea of a direct link between the money supply and inflation, but only to explain why it is obsolete and irrelevant for the modern world:

Until the 1980s, the strategy was to choose a target rate of money growth and to allow for deviations from that target rate as a function of activity. The rationale was simple. A low target rate of money growth implied a low average rate of inflation. … 

That strategy did not work well.

First, the relation between money growth and inflation turned out to be far from tight, even in the medium run. … Second, the relation between the money supply and the interest rate in the short run also turned out also to be unreliable. …

Throughout the 1970s and 1980s, frequent and large shifts in money demand created serious problems for central banks. … Starting in the early 1990s, a dramatic rethinking of monetary policy took place based on targeting inflation rather than money growth, and the use of an interest rate rule.

Obviously, I don’t endorse everything in the textbook.1 (The idea of a tight link between unemployment and inflation is not looking much better than the idea of a tight link between inflation and the money supply.) I bring it up here just to establish that the absence of a link between money growth and inflation is not radical or heterodox, but literally the textbook view.

One way of thinking about the first Blanchard passage above is that the three stories about inflation correspond to three stories about price setting. Prices may be set based on expectations of where prices will be, or prices may be set based on market power (the markup), or prices may be set based on costs of production. 

This seems to me to be the beginning of wisdom with respect to inflation: Inflation is just an increase in prices, so for every theory of price setting there’s a corresponding theory of inflation. There is wide variation in how prices get set across periods, countries and markets, so there must be a corresponding variety of inflations. 

Besides the three mentioned by Blanchard, there’s one other story that inflation is perhaps even more widespread. We could call this too much spending chasing too little production. 

The too-much-spending view of inflation corresponds to a ceiling on output, rather than a floor on unemployment, as the inflationary barrier. As the NAIRU has given way to potential output as the operational form of supply constraints on macroeconomic policy, this understanding of inflation has arguably become the dominant one, even if without formalization in textbooks. It overlaps with the unemployment story in making current demand conditions a key driver of inflation, even if the transmission mechanism is different. 

Superfically “too much spending relative to production” sounds a lot like “too money relative to goods.” (As to a lesser extent does “too much wage growth relative to productivity growth.”) But while these formulations sound similar, they have quite different implications. Intuitions formed by the old quantity-of-money view don’t work for the new stories.

The older understanding of inflation, which runs more or less unchanged from David Hume through Irving Fisher to Milton Friedman and contemporary monetarists, goes like this. There’s a stock of goods, which people can exchange for their mutual benefit. For whatever reasons, goods don’t exchange directly for other goods, but only for money. Money in turn is only used for purchasing goods. When someone receives money in exchange for a good, they turn around and spend it on some good themselves — not instantly, but after some delay determined by the practical requirements of exchange. (Imagine you’ve collected your earnings from your market stall today, and can take them to spend at a different market tomorrow.) The total amount of money, meanwhile, is fixed exogenously — the quantity of gold in circulation, or equivalently the amount of fiat tokens created by the government via its central bank.

Under these assumptions, we can write the familiar equation

MV = PY

If Y, the level of output, is determined by resources, technology and other “real” factors, and V is a function of the technical process of exchange — how long must pass between the receipt of money and it spending — then we’re left with a direct relationship between the change in M and the change in P. “Inflation is always and everywhere a monetary phenomenon.”2

I think something like this underlies most folk wisdom about inflation. And as is often the case, the folk wisdom has outlived whatever basis in reality it may once have had.3

Below, I want to sketch out some ways in which the implications of the excessive-spending-relative-to-production vision of inflation are importantly different from those of the excessive-money-relative-to-goods vision. But first, a couple of caveats.

First, the idea of a given or exogenous quantity of money isn’t wrong a priori, as a matter of logic; it’s an approximation that happens not to fit the economy in which we live. Exactly what range of historical settings it does fit is a tricky question, which I would love to see someone try to answer. But I think it’s safe to say that many important historical inflations, both under metallic and fiat regimes, fit comfortably enough in a monetarist framework. 

Second, the fact that the monetarist understanding of inflation is wrong (at least for contemporary advanced economies) doesn’t mean that the modern mainstream view is right. There is no reason to think there is one general theory of inflation, any more than there is one general etiology of a fever. Lots of conditions can produce the same symptom. In general, inflation is a persistent, widespread rise in prices, so for any theory of price-setting there’s a corresponding theory of inflation. And the expectations-based propagation mechanism of inertial inflation — where prices are raised in the expectation that prices will rise — is compatible with many different initial inflationary impulses. 

That said — here are some important cleavages between the two visions.

1. Money vs spending. More money is just more money, but more spending is always more spending on something in particular. This is probably the most fundamental difference. When we think of inflation in terms of money chasing a given quantity of goods, there is no connection between a change in the quantity of money and a change in individual spending decisions. But when we think of it in terms of spending, that’s no longer true — a decision to spend more is a decision to spend more on some specific thing. People try to carry over intuitions from the former case to the latter, but it doesn’t work. In the modern version, you can’t tell a story about inflation rising that doesn’t say who is trying to buy more of what; and you can’t tell a story about controlling inflation without saying whose spending will be reduced. Spending, unlike money, is not a simple scalar.

The same goes for the wages-markup story of the textbook. In the model, there is a single wage and a single production process. But in reality, a fall in unemployment or any other process that “raises the wage” is raising the wages of somebody in particular.

2. Money vs prices. There is one stock of money, but there are many prices, and many price indices. Which means there are many ways to measure inflation. As I mentioned above, inflation was originally conceived of as definitionally an increase in the quantity of money. Closely related to this is the idea of a decrease in the purchasing power of money, a definition which is still sometimes used. But a decrease in the value of money is not the same as an increase in the prices of goods and services, since money is used for things other than purchasing goods and services.  (Merijn Knibbe is very good on this.4) Even more problematically, there are many different goods and services, whose prices don’t move in unison. 

This wasn’t such a big deal for the old concept of inflation, since one could say that all else equal, a one percent increase in the stock of money would imply an additional point of inflation, without worrying too much about which specific prices that showed up in. But in the new concept, there’s no stock of money, only the price changes themselves. So picking the right index is very important. The problem is, there are many possible price indexes, and they don’t all move in unison. It’s no secret that inflation as measured by the CPI averages about half a point higher than that measured by the PCE. But why stop there? Those are just two of the infinitely many possible baskets of goods one could construct price indexes for. Every individual household, every business, every unit of government has their own price index and corresponding inflation rate. If you’ve bought a used car recently, your personal inflation rate is substantially higher than that of people who haven’t. We can average these individual rates together in various ways, but that doesn’t change the fact that there is no true inflation rate out there, only the many different price changes of different commodities.

3. Inflation and relative prices. In the old conception, money is like water in a pool. Regardless of where you pour it in, you get the same rise in the overall level of the pool.

Inflation conceived of in terms of spending doesn’t have that property. First, for the reason above — more spending is always more spending on something. If, let’s say for sake of argument, over-generous stimulus payments are to blame for rising inflation, then the inflation must show up in the particular goods and services that those payments are being used to purchase — which will not be a cross-section of output in general. Second, in the new concept, we are comparing desired spending not to a fixed stock of commodities, but to the productive capacity of the economy. So it matters how elastic output is — how easily production of different goods can be increased in response to stronger demand. Prices of goods in inelastic supply — rental housing, let’s say — will rise more in response to stronger demand, while prices of goods supplied elastically — online services, say — will rise less. It follows that inflation, as a concrete phenomenon, will involve not an across-the-board increase in prices, but a characteristic shift in relative prices.

This is a different point than the familiar one that motivates the use of “core” inflation — that some prices (traditionally, food and energy) are more volatile or noisy, and thus less informative about sustained trends. It’s that  when spending increases, some goods systematically rise in price faster than others.

This recent paper by Stock and Watson, for example, suggests that housing, consumer durables and food have historically seen prices vary strongly with the degree of macroeconomic slack, while prices for gasoline, health care, financial services, clothing and motor vehicles do not, or even move the opposite way. They suggest that the lack of a cyclical component in health care and finance reflect the distinct ways that prices are set (or imputed) in those sectors, while the lack of a cyclical component in gas, clothing and autos reflects the fact that these are heavily traded goods whose prices are set internationally. This interpretation seems plausible enough, but if you believe these numbers they have a broader implication: We should not think of cyclical inflation as an across the board increase in prices, but rather as an increase in the price of a fairly small set of market-priced, inelastically supplied goods relative to others.

4. Inflation and wages. As I discussed earlier in the post, the main story about inflation in today’s textbooks is the Phillips curve relationship where low unemployment leads to accelerating inflation. Here it’s particularly clear that today’s orthodoxy has abandoned the quantity-of-money view without giving up the policy conclusions that followed from it.

In the old monetarist view, there was no particular reason that lower unemployment or faster wage growth should be associated with higher inflation. Wages were just one relative price among others. A scarcity of labor would lead to higher real wages, while an exogenous increase in wages would lead to lower employment. But absent a change in the money supply, neither should have any effect on the overall price level. 

It’s worth noting here that altho Milton Friedman’s “natural rate of unemployment” is often conflated with the modern NAIRU, the causal logic is completely different. In Friedman’s story, high inflation caused low unemployment, not the reverse. In the modern story, causality runs from lower unemployment to faster wage growth to higher inflation. In the modern story, prices are set as a markup over marginal costs. If the markup is constant, and all wages are part of marginal cost, and all marginal costs are wages, then a change in wages will just be passed through one to one to inflation.

We can ignore the stable markup assumption for now — not because it is necessarily reasonable, but because it’s not obvious in which direction it’s wrong. But if we relax the other assumptions, and allow for non-wage costs of production and fixed wage costs, that unambiguously implies that wage changes are passed through less than one for one to prices. If production inputs include anything other than current labor, then low unemployment should lead to a mix of faster inflation and faster real wage growth. And why on earth should we expect anything else? Why shouldn’t the 101 logic of “reduced supply of X leads to a higher relative price of X” be uniquely inapplicable to labor?5

There’s an obvious political-ideological reason why textbooks should teach that low unemployment can’t actually make workers better off. But I think it gets a critical boost in plausibility — a papering-over of the extreme assumptions it rests on — from intuitions held over from the old monetarist view. If inflation really was just about faster money growth, then the claim that it leaves real incomes unchanged could work as a reasonable first approximation. Whereas in the markup-pricing story it really doesn’t. 

5. Inflation and the central bank.  In the quantity-of-money vision, it’s obvious why inflation is the special responsibility of the central bank. In the textbooks, managing the supply of money is often given as the first defining feature of a central bank. Clearly, if inflation is a function of the quantity of money, then primary responsibility for controlling it needs to be in the hands of whoever is in charge of the money supply, whether directly, or indirectly via bank lending. 

But here again, it seems, to me, the policy conclusion is being asked to bear weight even after the logical scaffolding supporting it has been removed. 

Even if we concede for the sake of argument that the central bank has a special relationship with the quantity of money, it’s still just one of many influences on the level of spending. Indeed, when we think about all the spending decisions made across the economy, “at one interest rate will I borrow the funds for it” is going to be a central consideration in only a few of them. Whether our vision of inflation is too much spending relative to the productive capacity of the economy, or wages increasing faster than productivity, many factors are going to play a role beyond interest rates or central bank actions more broadly. 

One might believe that compared with other macro variables, the policy interest rate has a uniquely strong and reliable link to the level of spending and/or wage growth; but almost no one, I think, does believe this. The distinct responsibility of the central bank for inflation gets justified not on economic grounds but political-institutional ones: the central bank can act more quickly than the legislature, it is free of undue political influence, and so on. These claims may or may not be true, but they have nothing in particular to do with inflation. One could justify authority over almost any area of macroeconomic policy on similar grounds.

Conversely, once we fully take on board the idea that the central bank’s control over inflation runs through to the volume of credit creation to the level of spending (and then perhaps via unemployment to wage growth), there is no basis for the distinction between monetary policy proper and other central bank actions. All kinds of regulation and lender-of-last-resort operations equally change the volume and direction of credit creation, and so influenced aggregate spending just as monetary policy in the narrow sense does.

6. The costs of inflation. If inflation is a specifically monetary phenomenon, the costs of inflation presumably involve the use of money. The convenience of quoting relative prices in money becomes a problem when the value of money is changing.

An obvious example is the fixed denominations of currency — monetarists used to talk with about “shoe leather costs” — the costs of needing to go more frequently to the bank (as one then did) to restock on cash. A more consequential example is public incomes or payments fixed in money terms. As recently as the 1990s, one could find FOMC members talking about bracket creep and eroded Social Security payments as possible costs of higher inflation — albeit with some embarrassment, since the schedules of both were already indexed by then. More broadly, in an economy organized around money payments, changes in what a given flow of money can buy will create problems. Here’s one way to think about these problems:

Social coordination requires a mix of certainty and flexibility. It requires economic units to make all kinds of decisions in anticipation of the choices of other units — we are working together; my plans won’t work out if you can change yours too freely. But at the same time, you need to have enough space to adapt to new developments — as with train cars, there needs to be some slack in the coupling between economic unit for things to run smoothly. One dimension of this slack is the treatment of some extended period as if it were a single instant.

This is such a basic, practical requirements of contracting and management that we hardly think about it. For example, budgets — most organizations budget for periods no shorter than a quarter, which means that as far as internal controls and reporting are concerned, anything that happens within that quarter happens at the same time.6Similarly, invoices normally require payment in 30 or 60 days, thus treating shorter durations as instantaneous. Contracts of all kinds are signed for extended periods on fixed money terms. All these arrangements assume that the changes in prices over a few months or a year are small enough that they can be safely ignored.can be modified when inflation is high enough to make the fiction untenable that 30, 60 or 90 days is an instant. Social coordination strongly benefits from the convention that shorter durations can be ignored for most periods, which means people behave in practice as if they expect inflation over such shorter periods to be zero.

Axel Leijonhufvud’s mid-70s piece on inflation is one of the most compelling accounts of this kinds of cost of inflation — the breakdown of social coordination — that I have seen. For him, the stability of money prices is the sine qua non of decentralized coordination through markets. 

In largely nonmonetary economies, important economic rights and obligations will be inseparable from particularized relationships of social status and political allegiance and will be in some measure permanent, inalienable and irrevocable. … In monetary exchange systems, in contrast, the value to the owner of an asset derives from rights, privileges, powers and immunities against society generally rather than from the obligation of some particular person. …

Neoclassical theories rest on a set of abstractions that separate “economic” transactions from the totality of social and political interactions in the system. For a very large set of problems, this separation “works”… But it assumes that the events that we make the subject of … the neoclassical model of the “economic system” do not affect the “social-political system” so as … to invalidate the institutional ceteris paribus clauses of that model. …

 Double-digit inflation may label a class of events for which this assumption is a bad one. … It may be that … before the “near-neutral” adjustments can all be smoothly achieved, society unlearns to use money confidently and reacts by restrictions on “the circles people shall serve, the prices they shall charge, and the goods they can buy.”

One important point here is that inflation has a much greater impact than in conventional theory because of the price-stability assumption incorporated into any contract that is denominated in money terms and not settled instantly — which is to say, pretty much any contract. So whatever expectations of inflation people actually hold, the whole legal-economic system is constructed in a way that makes it behave as if inflation expectations were biased toward zero:

The price stability fiction — a dollar is a dollar is a dollar — is as ingrained in our laws as if it were a constitutional principle. Indeed, it may be that no real constitutional principle permeates the Law as completely as does this manifest fiction.

The market-prices-or-feudalism tone of this seems more than a little overheated from today’s perspective, and when Arjun and I asked him about this piece a few years ago, he seemed a bit embarrassed by it. But I still think there is something to it. Market coordination, market rationality, the organization of productive activity through money payments and commitments, really does require the fiction of a fixed relationship between quantities of money and real things. There is some level of inflation at which this is no longer tenable.

So I have no problem with the conventional view that really high inflations — triple digits and above — can cause far-reaching breakdowns in social coordination. But this is not relevant to the question of inflation of 1 or 2 or 5 or probably even 10 percent. 

In this sense, I think the mainstream paradoxically both understates and overstates the real costs of inflation. They exaggerate the importances of small differences in inflation. But at the same time, because they completely naturalize the organization of life through markets, they are unable to talk about the possibility that it could break down.

But again, this kind of breakdown of market coordination is not relevant for the sorts of inflation seen in the United States or other rich countries in modern times. 

It’s easier to talk about the costs (and benefits) of inflation when we see it as a change in relative prices, and redistribution of income and wealth. If inflation is typically a change in relative prices, then the costs are experienced by those whose incomes rise more slowly than their payments. Keynes emphasized this point in an early article on “Social Consequences of a Change in the Value of Money.”7

A change in the value of money, that is to say in the level of prices, is important to Society only in so far as its incidence is unequal. Such changes have produced in the past, and are producing now, the vastest social consequences, because, as we all know, when the value of money changes, it does not change equally for all persons or for all purposes. … 

Keynes sees the losers from inflation as passive wealth owners, while the winners are active businesses and farmers; workers may gain or lose depending on the degree to which they are organized. For this reason, he sees moderate inflation as being preferable to moderate deflation, though both as evils to be avoided — until well after World War II, the goal of price stability meant what it said.

Let’s return for a minute to the question of wages. As far as I can tell, the experience in modern inflations is that wage changes typically lag behind prices. If you plot nominal wage growth against inflation, you’ll see a clear positive relationship, but with a slope well below 1. This might seem to contradict what I said under point 4. But my point there was that insofar as inflation is driven by increased worker bargaining power, it should be associated with faster real wage growth. In fact, the textbook is wrong not just on logic but on facts. In principle, a wage-driven inflation would see a rise in real wage. But most real inflations are not wage-driven.

In practice, the political costs of inflation are probably mostly due to a relatively small number of highly salient prices. 

7. Inflation and production. The old monetarist view had a fixed quantity of money confronting a fixed quantity of goods, with the price level ending up at whatever equated them. As I mentioned above, the fixed-quantity-of-money part of this has been largely abandoned by modern mainstream as well as heterodox economists. But what about the other side? Why doesn’t more spending call forth more production?

The contemporary mainstream has, it seems to me, a couple ways of answering the question. One is the approach of a textbook like Blanchard’s. There, higher spending does lead to to higher employment and output and lower unemployment. But unless unemployment is at a single unique level — the NAIRU — inflation will rise or fall without limit. It’s exceedingly hard to find anything that looks like a NAIRU in the data, as critics have been pointing out for a long time. Even Blanchard himself rejects it when he’s writing for central bankers rather than undergraduates. 

There’s a deeper conceptual problem as well. In this story, there is a tradeoff between unemployment and inflation. Unemployment below the NAIRU does mean higher real output and income. The cost of this higher output is an inflation rate that rises steadily from year to year. But even if we believed this, we might ask, how much inflation acceleration is too much? Can we rule out that a permanently higher level of output might be worth a slowly accelerating inflation rate?

Think about it: In the old days, the idea that the price level could increase without limit was considered crazy. After World War II, the British government imposed immense costs on the country not just to stabilize inflation, but to bring the price level back to its prewar level. In the modern view, this was crazy — the level of prices is completely irrelevant. The first derivative of prices — the inflation rate — is also inconsequential, as long as it is stable and predictable. But the second derivative — the change in the rate of inflation — is apparently so consequential that it must be kept at exactly zero at all costs. It’s hard to find a good answer, or indeed any answer, for why this should be so.

The more practical mainstream answer is to say, rather than that there is a tradeoff between unemployment and inflation with one unambiguously best choice, but that there is no tradeoff. In this story, there is a unique level of potential output (not a feature of the textbook model) at which the relationship between demand, unemployment and inflation changes. Below potential, more spending calls forth more production and employment; above potential, more spending only calls forth higher inflation. This looks better as a description of real economies, particular given that the recent experience of long periods of elevated unemployment that have not, contrary to the NAIRU prediction, resulted in ever-accelerating deflation. But it begs the question of why should be such a sharp line.

The alternative view would be that investment, technological change, and other determinants of “potential output” also respond to demand. Supply constraints, in this view, are better thought of in terms of the speed with which supply can respond to demand, rather than an absolute ceiling on output.

Well, this post has gotten too long, and has been sitting in the virtual drawer for quite a while as I keep adding to it. So I am going to break off here. But it seems to me that this is where the most interesting conversations around inflation are going right now — the idea that supply constraints are not absolute but respond to demand with varying lags — that inflation should be seen as often a temporary cost of adjustment to a new higher level of capacity. And the corollary, that anti-inflation policy should aim at identifying supply constraints as much as, or more than, restraining demand. 

At Roosevelt: Reimagining Full Employment

Mike Konczal, Lauren Melodia and I have a new report out from the Roosevelt Institute, on what true full employment might look like in the United States.

This is part of a larger project of imagining what an economic boom would look like. As Mike and I argued in our recent New York Times op-ed, there’s a real possibility that the coming years could see a historic boom, thanks to the exceptionally strong stimulus measures of the past year and, hopefully, the further expansions of public spending on the way. (Interestingly, the term “boom” is now making it into Biden’s speeches on the economy.) If the administration, Congress and the Fed don’t lose their nerve and stay on the path they’re currently on, we could soon be seeing economic growth and rising wages in a way that we haven’t since at least the late 1990s.

This is going to call for a new way of thinking about economic policy. Over the past decade or more, the macroeconomic policy debate has been dominated by a consensus that is more concerned with the supposed dangers of public debt than stagnation, and sees any uptick in growth or wages as worryingly inflationary. Meanwhile, the left knows how to criticize austerity and bailouts for business, and to make the case for specific forms of public spending, but has a harder time articulating the benefits of sustained growth and tight labor markets.

What we’re trying to do is move away from the old, defensive fights about public debt and austerity and make the positive case for a bigger more active public sector. There’s no reason the Right should have a monopoly on promises faster growth and improvements in peoples material living standards. Post-covid, we’re looking at a new “morning in America” moment, and progressives should be prepared to take credit.

One of the great appeals of the Green New Deal framing on climate change is that it turns decarbonization from a question of austerity and sacrifice into a promise to improve people’s material well being, not decades from now but right now, and in ways that go well beyond climate itself. I think this promise is not just politically useful but factually well-founded, and could just as well be made for other expansions of the public sector.

This is an argument that I and others have been making for years. Of course, any promise of faster growth and higher living standards has to confront the argument, enshrined in macroeconomics textbooks, that the economy is already operating close to potential, at least most of the time — that the Federal Reserve has taken care of the demand problem. In that case, the Keynesian promise that more spending can call forth more production would no longer apply.

We’ve tended to respond to this argument negatively — that there is no evidence that the US now was facing any kind of absolute supply constraint or labor shortage before the pandemic, let alone now. This is fine as far as it goes, and I think our side of the debate has won some major victories — Jay Powell and Janet Yellen both now seem to agree that as of 2019 the US was still well short of full employement. Still, I think it’s legitimate for people to ask, “If this isn’t full employment, then what would be?” We need a positive answer of our own, and not just a negative criticism of the textbook view.

This new paper is an attempt to do just that — to construct an estimate of full employment that doesn’t build in the assumption that recent labor market performance was close to it. One way to do this is to compare the US to other advanced countries, many of which have higher employment-population ratios than the US, even after adjusting for age differences. We chose to take a different approach, one that instead looks at differences in employment rates within the US population.

From the executive summary:

This issue brief argues that potential employment in the US is much higher than we have seen in recent years. In addition to those officially counted in the labor force, there is a large latent labor force, consisting of people who are not currently seeking work but who could reasonably be expected to do so given sustained strong labor demand. This implies much more labor market slack than conventional measures of unemployment suggest.

An important but less familiar sign of labor market slack is the difference in employment rates between groups with more- and less-privileged positions in the labor market. Because less-favored groups—Black workers, women, those with less formal education, those just entering the labor market—are generally last hired and first fired, the gaps between more- and less-favored groups vary systematically over the business cycle. When labor markets are weak and employers can pick and choose among potential employees, the gap between employment rates for more- and less-favored groups widens. When labor markets are tight, and workers have more bargaining power, the gap shrinks.

We use this systematic relationship between overall labor market conditions and employment rates across race, gender, education, and age to construct a new measure of potential employment. In effect, since more-favored workers will be hired before less-favored ones, the difference in outcomes between these groups is a measure of how close hiring has gotten to the true back of the line.

We construct our measure in stages. We start with the fact that changes in employment rates within a given age group cannot reflect the effect of population aging. Simply basing potential employment by age groups on employment rates that have been observed historically implies potential employment 1.7 points higher than the CBO estimates.

Next, we close the employment gaps by race and gender, on the assumption that women and Black Americans are no less able or willing to work than white men of a similar age. (When adjusting for gender, we make an allowance for lower employment rates among parents of young children). This raises potential employment by another 6.2 points.

Finally, reducing the employment gap between more- and less-educated workers in line with the lower gaps that have been observed historically adds another 1.8 points to the potential employment rate.

In total, these adjustments yield a potential employment-population ratio 10 points higher than the CBO estimates, equivalent to the addition of about 28 million more jobs over the next decade.

Adding these 28 million additional jobs over the next decade would require an average annual growth in employment of 2.1 percent. The employment growth that would fully mobilize the latent labor force, as estimated here, is in line with the rate of GDP growth required to repair the damage from the Great Recession of 2007–2009 and return GDP to its pre–2007 trend.

You can read the rest here.

The Persistence of Demand

Here’s the very short version of this very long post:

Hysteresis means that a change in GDP today has effects on GDP many years in the future. In principle, this could be because it affects either future aggregate demand or potential output. These two cases aren’t distinguished clearly in the literature, but they have very different implications. The fact that the Great Recession was followed by a period of low inflation, slow wage growth and low interest rates, rather than the opposite, suggests that the persistent-demand form of hysteresis is more important than potential-output hysteresis. The experience of the Great Recession is consistent with perhaps 20 percent of a shock to demand in this period carrying over to demand in future periods. This value in turn lets us estimate how much additional spending would be needed to permanently return GDP to the pre-2007 trend: 50-60 percent of GDP, or $10-12 trillion, spread out over a number of years.

 

Supply Hysteresis and Demand Hysteresis

The last few years have seen renewed interest in hysteresis – the idea that shifts in demand can have persistent effects on GDP, well beyond the period of the “shock” itself. But it seems to me that the discussion of hysteresis doesn’t distinguish clearly between two quite different forms it could take.

On the one hand, demand could have persistent effects on output because demand influences supply – this seems to be what people usually have in mind. But on the other hand, demand itself might be persistent. In time-series terms, in this second story aggregate spending behaves like a random walk with drift. If we just look at the behavior of GDP, the two stories are equivalent. But in other ways they are quite different.

Let’s say we have a period in which total spending in the economy is sharply reduced for whatever reason. Following this, output is lower than we think it otherwise would have been. Is this because (a) the economy’s productive potential was permanently reduced by the period of reduced spending? Or is it (b) because the level of spending in the economy was permanently reduced? I will call the first case supply hysteresis and the second demand hysteresis.

On the left, supply hysteresis. On the right, demand hysteresis.

It might seem like a semantic distinction, but it’s not. The critical thing to remember is that what matters for much of macroeconomic policy is not the absolute level of output but the output gap — the difference between actual and potential output. If current output is above potential, then we expect to see rising inflation. (Depending on how “potential” is understood, this is more or less definitional.) We also expect to see rising wages and asset prices, shrinking inventories, longer delivery times, and other signs of an economy pushing against supply constraints. If current output is below potential, we expect the opposite — lower inflation or deflation, slower wage growth, markets in general that favor buyers over sellers. So while lower aggregate supply and lower aggregate demand may both translate into lower GDP, in other respects their effects are quite different. As you can see in my scribbles above, the two forms of hysteresis imply opposite output gaps in the period following a deep recession. 

Imagine a hypothetical case where there is large fall in public spending for a few years, after which spending returns to its old level. For purposes of this thought experiment, assume there is no change in monetary policy – we’re at the ZLB the whole time, if you like. In the period after the depressed spending ends, will we have (1) lower unemployment and higher inflation than before, as the new income created during the period of high public spending leads to permanently higher demand. Or will we have (2) higher unemployment and lower inflation than if the spending had not occurred, because the period of high spending permanently raised labor force participation and productivity, while demand returns to its old level?

Supply hysteresis implies (1), that a temporary negative demand shock will lead to persistently higher inflation and lower unemployment (because the labor force will be smaller). Demand hysteresis implies (2), that a temporary negative demand shock will lead to permanently lower inflation and higher unemployment. Since the two forms of hysteresis make diametrically opposite predictions in this case, seems important to be clear which one we are imagining. Of course in the real world, could see a combination of both, but they are still logically distinct. 

Most people reading this have probably seen a versions of the picture below. On the eve of the pandemic, real per-capita GDP was about 15 percent below where you’d expect it to be based on the pre-2007 trend. (Or based on pre-2007 forecasts, which largely followed the trend.) Let’s say we agree that the deviation is in some large part due to the financial crisis: Are we imagining that output has persistently fallen short of potential, or that potential has fallen below trend? Or again, it might be a combination of both.

In the first case, we would expect monetary policy to be generally looser in the period after a negative demand shock, in the second case tighter. In the first case we’d expect lower inflation in period after shock, in the second case higher.

It seems to me that most of the literature on hysteresis does not really distinguish these cases. This recent IMF paper by Antonio Fatas and coauthors, for example, defines hysteresis as a persistent effect of demand shocks on GDP. This could be either of the two cases. In the text of the paper,  they generally assume hysteresis means an effect of demand on supply, and not a persistence of demand itself, but they don’t explicitly spell this out or make an argument for why the latter is not important.

It is clear that the original use of the term hysteresis was understood strictly as what I am calling supply hysteresis. (So perhaps it would be better to reserve the word for that, and make ups new name for the other thing.) If you read the early literature on hysteresis, like these widely-cited Laurence Ball papers, the focus was on the European experience of the 1980s and 1990s; hysteresis is described as a change in the NAIRU, not as an effect on employment itself. The mechanism is supposed to be a specific labor-market phenomenon: the long term unemployed are no longer really available for work, even if they are counted in the statistics. In other words, sustained unemployment effectively shrinks the labor force, which means that in the absence of policy actions to reduce demand, the period following a deep recession will see faster wage growth and higher inflation than we would have expected.

(This specific form of supply hysteresis implies a persistent rise in unemployment following a downturn, just as demand hysteresis does. The other distinctions above still apply, and other forms of supply hysteresis would not have this implication.) 

Set aside for now whether supply-hysteresis was a reasonable description of Europe in the 1980s and 1990s. Certainly it was a welcome alternative to the then-dominant view that Europe needed high unemployment because of over-protective labor market institutions. But whether or not thinking of hysteresis in terms of the NAIRU made sense in that context, it does not make sense for either Europe or the US (or Japan) in the past decade. Everything we’ve seen has been consistent with a negative output gap — with actual output below potential — with a depressed level of demand, not of supply. Wage growth has been unexpectedly weak, not strong; inflation has been below target; and central banks have been making extra efforts to boost spending rather than to rein it in.

Assuming we think that all this is at least partly the result of the 2007-2009 financial crisis — and thinking that is pretty much the price of entry to this conversation — that suggests we should be thinking primarily about demand-hysteresis rather than supply-hysteresis. We should be asking not, or not only, how much and how durably the Great Recession reduced the country’s productive potential, but how how durably it reduced the flow of money through the economy. 

It’s weird, once you think about it, how unexplored this possibility is in the literature. It seems to be taken for granted that if demand shocks have a lasting effect on GDP, that must be because they affect aggregate supply. I suspect one reason for this is the assumption — which profoundly shapes modern macroeconomics — that the level of spending in the economy is directly under the control of the central bank. As Peter Dorman observes, it’s a very odd feature of modern macroeconomic modeling that the central bank is inside the model — the reaction of the monetary authorities to, say, rising inflation is treated as a basic fact about the economy, like the degree to which investment responds to changes in the interest rate, rather than as a policy choice. In an intermediate macroeconomics textbook like Carlin and Soskice (a good one as far as they go), students are taught to think about the path of unemployment and inflation as coming out of a “central bank preference function,” which is taken as a fundamental parameter of the economy. Obviously there is no place for demand hysteresis in this framework. To the extent that we think of the actual path of spending in the economy as being chosen by the central bank as part of some kind of optimizing process, past spending in itself will have no effect on current spending. 

Be that as it may, it seems hard to deny that in real economies, the level of spending today is strongly influenced by the level of spending in the recent past. This is the whole reason we see booms and depressions as discrete events rather than just random fluctuations, and why they’re described with metaphors of positive-feedback process like “stall speed” or “pump-priming.”1

How Persistent Is Demand?

Let’s say demand is at least somewhat persistent. That brings us to the next question: How persistent? If we were to get extra spending of 1 percent of GDP in one year, how much higher would we now expect demand to be several years later?

We can formalize this question if we write a simple model like:

Zt = Z*t + Xt

Z*t = (1+g) Z*t-1 + a(Zt-1 – Z*t-1)

Here Z is total spending or demand, Z* is the trend, what we might think of as normal or expected demand, g is the normal growth rate, and X is the influence of transitory influences outside of normal growth.

With a = 0, then, we have the familiar story where demand is a trend plus random fluctuations. If we see periods of above- and below-trend demand, that’s because the X influences are themselves extended over time. If a boom year is followed by another boom year, in this story, that’s because whatever forces generated it in the first year are still operating, not because the initial boom itself was persistent. 

Alternatively, with a = 1, demand shocks are permanent. Anything that increases spending this year, should be expected to lead to just as much additional spending next year, the year after that, and so on.

Or, of course, a can have any intermediate value. 

Think back to 2015, in the debate over the first Sanders’ campaign’s spending plans that was an important starting point for current discussions of hysteresis. The basic mistake Jerry Friedman was accused of making was assuming that changes in demand were persistent — that is, if the multiplier was, say 1.5, that an increase in spending of $500 billion would raise output by $750 billion not only in that year and but in all subsequent years. As his critics correctly pointed out, that is not how conventional multipliers work. In terms of my equations above, he was setting a=1, while the conventional models have a=0. 

He didn’t spell this out, and I didn’t think of it that way at the time. I don’t think anyone did. But once you do, it seems to me that while Friedman was wrong in terms of the standard multiplier, he was not wrong about the economy — or at least, no more wrong than the critics. It seems to me that both sides were using unrealistically extreme values. Demand shocks aren’t entirely permanent, but they also aren’t entirely transitory.  A realistic model should have 0 < a < 1.

Demand Persistence and Fiscal Policy

There’s no point in refighting those old battles now. But the same question is very relevant for the future. Most obviously, if demand shocks are persistent to some significant degree, it becomes much more plausible that the economy has been well below potential for the past decade-plus. Which means there is correspondingly greater space for faster growth before we encounter supply constraints in the form of rising inflation. 

Both forms of hysteresis should make us less worried about inflation. If we are mainly dealing with supply hysteresis, then rapid growth might well lead to inflation, but it would be a transitory phenomenon as supply catches up to the new higher level of demand.  On the other hand, to the extent we are dealing with demand hysteresis, it will take much more growth before we even have to worry about inflation.

Of course, both forms of hysteresis may exist. In which case, both reason for worrying less about inflation would be valid. But we still need to be clear which we are talking about at any given moment.

A slightly trickier point is that the degree of demand persistence is critical for assessing how much spending it will take to get back to the pre-2007 trend. 

If the failure to return to the pre-2007 is the lasting effect of the negative demand shock of the Great Recession, it follows that  sufficient spending should be able to reverse the damage and return GDP to its earlier trend. The obvious next question is, how much? The answer really depends on your preferred value for a. In the extreme (but traditional) case of a=0, each year we need enough spending to fill the entire gap, every year, forever. Given a gap of around 12 percent, if we assume a multiplier of 1.5 or so, that implies additional public spending of $1.6 trillion. In the opposite extreme case, where a=1, we just need enough total spending to fill the gap, spread out over however many years. In general, if we want to get close a permanent (as opposed to transitory) output gap of W, we need W/(a μ) total spending, where μ is the conventional multiplier.2

If you project forward the pre-2007 trend in real per-capita GDP to the end of 2019, you are going to get a number that is about 15% higher than the actual figure, implying an output gap on the order of $3.5 trillion. In the absence of demand persistence, that’s the gap that would need to be filled each year. But with persistent demand, a period of elevated public spending would gradually pull private spending up to the old trend, after which it would remain there without further stimulus.

What Does the Great Recession Tell Us about Demand Persistence?

At this point, it might seem that we need to turn to time-series econometrics and try to estimate a value for a, using whatever methods we prefer for such things. And I think that would be a great exercise! 

But it seems to me we can actually put some fairly tight limits on a without any econometrics, simple by looking back to the Great Recession. Keep in mind, once we pick an output gap for a starting year, then given the actual path of GDP, each possible value of a implies a corresponding sequence of shocks Xt. (“Shock” here just means anything that causes a deviation of demand from its trend, that is not influenced by demand in the previous period.) In other words, whatever belief we may hold about the persistence of demand, that implies a corresponding belief about the size and duration of the initial fall in demand during the recession. And since we know a fair amount about the causes of the recession, some of these sequences are going to be more plausible than others.

The following figures are an attempt to do this. I start by assuming that the output gap was zero in the fourth quarter of 2004. We can debate this, of course,, but there’s nothing heterodox about this assumption — the CBO says the same thing. Then I assume that in the absence of exogenous disturbances, real GDP per capita would have subsequently grown at 1.4 percent per year. This is the growth rate during the expansion between the Great Recession and the pandemic; it’s a bit slower than the pre-recession trend.3 I then take the gap between this trend and actual GDP in each subsequent quarter and divide it into the part predictable from the previous quarter’s gap, given an assumed value for a, and the part that represents a new disturbance in that period. So each possible value of a, implies a corresponding series of disturbances. Those are what are shown in the figures.

If you’re not used to this kind of reasoning, this is probably a bit confusing. So let me put it a different way. The points in the graphs above show where real GDP would have been relative to the long-term trend if there had been no Great Recession. For example, if you think a = 0, then GDP in 2015 would have been just the same in the absence of the recession, so the values there are just the actual deviation from trend. So you can think of the different figures here as showing the exogenous shocks that would be required under different assumptions about persistence, to explain the actual deviation from trend. They are answering this question: Given your beliefs about how persistent demand is, what must you think GDP would have been in subsequent years in a world where the Great Recession did not take place? (Or maybe better, where the fall in demand form the housing bubble was fully offset by stimulus.)

The first graph, with persistence = 0, is easiest to understand. If there is no carryover of demand shocks from one period to the next, then there must be some factor reducing demand in each later period by the full extent of the gap from trend. If we move on to, say, the persistence=0.1 figure, that is saying that, if you think 10 percent of a demand shock is normally carried over into future periods, that means that there was something happening in 2012 that would have depressed demand by 2 percent relative to the earlier trend, even if there had been no Great Recession. 

Because people are used to overcomplicated economics models, I want to stress again. What I am showing you here is what you definitionally believe, if you think that in the absence of the Great Recession, growth in the 2010s would have been at about the same rate it was, just from a higher base, and you think that whatever fraction of a change in spending in one year is carried over to the next year. There are no additional assumptions. I’m just showing what the logical corollary of those beliefs would be for the pattern of demand shocks,

Another important feature of these figures is how large the initial fall in demand is. Logically, if you think demand is very persistent, you must also think the initial shock was smaller. If most of the fall in spending in the first half of 2008, say, was carried over to the second half of 2008, then it takes little additional fall in spending in that period to match the observed path of GDP. Conversely, if you think that very little of a change in demand in one period carries over to the next one then the autonomous fall in demand in 2009 must have been larger.

The question now is, given what we know about the forces impacting demand a decade ago, which of these figures is most plausible? If there had been sufficient stimulus to completely eliminate the fall in demand in 2007-2009, how strong would the headwinds have been a few years late? 

Based on what we know about the Great Recession, I think demand persistence in the 0.15 – 0.25 range most plausible. This suggests that a reasonable baseline guess for total spending required to return to the pre-2007 would be around 50 percent of GDP, spread out over a number of years. With an output gap of 15 percent of GDP, a multiplier of 1.5, and demand persistence of 0.2, we have 15 / (1.5 * 0.2) = 50 percent of GDP. This is, obviously, a very rough guess, but if you put me on the spot and asked how much spending over ten years it would take to get GDP permanently back to the pre-2007 trend, $10-12 trillion would be my best guess.

How do we arrive at persistence in the 0.15 – 0.25 range?

On the lower end, we can ask: What are the factors that would have pushed down demand in the mid 2010s, even in the absence of the Great Recession Remember, if we use demand persistence of 0.1, that implies there were factors operating in 2014 that would have reduced demand by 2 percent of GDP, even if the recession had not taken place. What would those be?

I don’t think it makes sense to say housing — housing prices had basically recovered by then. State and local spending is a better candidate — it remained quite depressed and I think it’s hard to see this as a direct effect of the recession. Relative to trend, state and local investment was down about 1 percent of GDP in 2014, while the federal stimulus was basically over. On the other hand, unless we think that monetary policy is totally ineffective, we have to include the stimulative effect of a zero policy rate and QE in our demand shocks. This makes me think that by 2014, the gap between actual GDP and the earlier trend was probably almost all overhang from the recession. And this implies a persistence of at least 0.15. (If you look back at the figures, you’ll see that with persistence=0.15, the implied shock reaches zero in 2014.)

Meanwhile, on the high end, a persistence of 0.5 would mean that the demand shock maxed out at a bit over 3 percent of GDP, and was essentially over by the second half of 2009. This seems implausibly small and implausibly brief. Residential investment fell from 6.5 percent of GDP in 2004 to less than 2.5 percent by 2010. And that is leaving aside housing wealth-driven consumption. Meanwhile, the ARRA stimulus didn’t really come online until the second half of 2009. I don’t believe monetary policy is totally ineffective, but I do think it operates slowly, especially on loosening side. So I find it hard to believe that the autonomous fall in demand in early 2009 was much less than 5 percent of GDP. That implies a demand persistence of no more than 0.25.

Within the 0.15 to 0.25 range, probably the most important variable is your judgement of the effectiveness of monetary policy and the ARRA stimulus. If you think that one or both was very effective, you might think that by mid-2010, they were fully offsetting the fall in demand from the housing bust. This would be consistent with  persistence around 0.25. Conversely, if you’re doubtful about the effectiveness of monetary policy and the ARRA (too little direct spending), you should prefer a value of 0.2 or 0.15. 

In any case, it seems to me that the implied shocks with persistence in the 0.15 – 0.25 range look much more plausible than for values outside that range. I don’t believe that the underlying forces that reduced demand in the Great Recession had ceased to operate by the second half of 2009. I also don’t think that they were autonomously reducing demand by as much as 2 points still in early 2014. 

You will have your own priors, of course. My fundamental point is that your priors on this stuff have wider implications. I have not seen anyone spell out the question of the persistence of demand in the way I have done here. But the idea is implicit in the way we talk about business cycles. Logically, a demand shortfall in any given period can be described as a mix of forces pulling down spending in that period, and the the ongoing effect of weak demand in earlier periods. And whatever opinion you have about the proportions of each, this can be quantified. What I am doing in this post, in other words, is not proposing a new theory, but trying to make explicit a theory that’s already present in these debates, but not normally spelled out.

Why Is Demand Persistent?

The history of real economies should be enough to convince us that demand can be persistent. Deep downturns — not only in the US after 2007, but in much of Europe, in Japan after 1990, and of course the Great Depression — show clearly that if the level of spending in an economy falls sharply for whatever reason, it is likely to remain low years later, even after the precipitating factor is removed. But why should economies behave this way?

I can think of a couple of reasons.

First, there’s the pure coordination story. Businesses pay wages to workers in order to carry out production. Production is carried out for sale. Sales are generated by spending. And spending depends on incomes, most of which are generated from production. This is the familiar reasoning of the multiplier, where it is used to show how an autonomous change in spending can lead to a larger (or smaller) change in output. The way the multiplier is taught, there is one unique level of output for each level of autonomous demand. But if we formalized the same intuition differently, we could imagine a system with multiple equilibria. Each would have a different level of income, expenditure and production, but in each one people would be making the “right” expenditure choices given their income. 

We can make this more concrete in two ways. First, balance sheets. One reason that there is a link from current income to current expenditure is that most economic units are financially constrained to some degree. Even if you knew your lifetime income with great precision, you wouldn’t be able to make your spending decisions on that basis because, in general, you can’t spend the money you will receive in the distant future today.

Now obviously there is some capacity to shift spending around in time, both through credit and through spending down liquid assets. The degree to which this is possible depends on the state of the balance sheet. To the extent a period of depressed demand leaves households and businesses with weaker balance sheets and tighter financial constraints, it will result in lower spending for an extended period. A version of this idea was put forward by Richard Koo as a “balance sheet recession,” in a rather boldly titled book. 

Finally there is expectations. There is not, after all, a true lifetime income out there for you to know. All you can do is extrapolate from the past, and from the experiences of other people like you. Businesses similarly must make decisions about how much investment to carry out based on extrapolation from the past – on what other basis could they do it?

A short period of unusually high or low demand may not move expectations much, but a sustained one almost certainly will. A business that has seen demand fall short of what they were counting on is going to make more conservative forecasts for the future. Again, how could they not? With the balance sheet channel, one could plausibly agree that demand shocks will be persistent but not permanent. But with expectations, once they have been adjusted, the resulting behavior will in general make them self-confirming, so there is no reason spending should ever return to its old path. 

This, to me, is the critical point. Mainstream economists and policy makers worry a great deal about inflation expectations, and whether they are becoming “unanchored.” But expectations of inflation are not the only ones that can slip their moorings. Households and businesses make decisions based on expectations of future income and sales, and if those expectations turn out to be wrong, they will be adjusted accordingly. And, as with inflation, the outcomes of which people form expectations themselves largely depend on expectations.

This was a point emphasized by Keynes: 

It is an essential characteristic of the boom that investments which will in fact yield, say, 2 per cent in conditions of full employment are made in the expectation of a yield of, say, 6 per cent, and are valued accordingly.

When the disillusion comes, this expectation is replaced by a contrary ‘error of pessimism’, with the result that the investments, which would in fact yield 2 per cent in conditions of full employment, are expected to yield less than nothing; and the resulting collapse of new investment then leads to a state of unemployment in which the investments, which would have yielded 2 per cent in conditions of full employment, in fact yield less than nothing. We reach a condition where there is a shortage of houses, but where nevertheless no one can afford to live in the houses that there are.

He continues the thought in terms that are very relevant today:

Thus the remedy for the boom is not a higher rate of interest but a lower rate of interest! For that may enable the so-called boom to last. The right remedy for the trade cycle is not to be found in abolishing booms and thus keeping us permanently in a semi-slump; but in abolishing slumps and thus keeping us permanently in a quasi-boom.

 

Video: The Macro Case for the Green New Deal

(Earlier this week, I gave a virtual presentation at an event organized by the Roosevelt Institute and the Green New Deal Network. Virtual events are inferior to live ones in many, many ways. But one way they are better, is that they are necessarily on video, and can be shared. Anyway, here is 25 minutes on why the economic situation calls for even more spending than the (surprisingly ambitious) proposals from the Biden administration, and also on why full employment shouldn’t be seen as an alternative to social justice and equity goals but as the best way of advancing them.)

The American Rescue Plan as Economic Theory

So, this happened.

Some people are frustrated about the surrender on the minimum wage, the scaled-back unemployment insurance, the child tax credit that should have been a universal child allowance, the fact that most of the good things phase out over the next year or two.

On the other side are those who see it as a decisive break with neoliberalism. Both the Clinton and Obama administrations entered office with ambitious spending plans, only to abandon or sharply curtail them (respectively), and instead embrace a politics of austerity and deficit reduction. From this point of view, the fact that the Biden administration not only managed to push through an increase in public spending of close to 10 percent of GDP, but did so without any promises of longer-term deficit reduction, suggests a fundamental shift.

Personally, I share this second perspective. I am less surprised by the ways in which the bill was trimmed back, than by the extent that it breaks with the Clinton-Obama model. The fact that people like Lawrence Summers have been ignored in favor of progressives like Heather Boushey and Jared Bernstein, and deficit hawks like the Committee for a Responsible Federal Budget have been left screeching irrelevantly from the sidelines, isn’t just gratifying as spectacle. It suggests a big move in the center of gravity of economic policy debates.

It really does seem that on the big macroeconomic questions, our side is winning. 

To be clear, the bill did not pass because some economists out-argued other economists. It was a political outcome that was driven by political conditions and political work. Most obviously, it’s hard to imagine this Biden administration without the two Sanders campaigns that preceded it. (In the president’s speech after signing the bill, Bernie was the first second person credited.) If it’s true, as reported, that Schumer kept expanded unemployment benefits in the bill only by threatening Manchin that the thing would not pass the House without them, then the Squad also deserves a lot of credit.

Still, from my parochial corner, it’s interesting to think about the economic theory implied by the bill. Implicitly, it seems to me, it represents a big break with prevailing orthodoxy.

Over the past generation, macroeconomic policy discussions have been based on a kind of textbook catechism that goes something like this: Over the long run, potential GDP grows at a rate based on supply-side factors — demographics, technological growth, and whatever institutions we think influence investment and labor force participation. Over the short run, there are random events that can cause actual spending to deviate from potential, which will be reflected in a higher or lower rate of inflation. These fluctuations are more or less symmetrical, both in frequency and in cost. The job of the central bank is to adjust interest rates to minimize the size of these deviations. The best short-term measure of how close the economy is to potential is the unemployment rate; at any given moment, there’s a minimum level of unemployment consistent with price stability. Smoothing out these fluctuations has real short run benefits, but no effects on long-term growth. The government budget balance, meanwhile, should not be used to stabilize demand, but rather should be kept at a level that ensures a stable or falling debt ratio; large fiscal deficits may be very costly. Finally, while it may be necessary to stabilize overall spending in the economy, this should be done in a way that minimizes “distortions” of the pattern of economic activity and, in particular, does not reduce the incentive to work.

Policy debates — though not textbooks — have been moving away from this catechism for a while. Jason Furman’s New View of Fiscal Policy is an example I often point to; you can also see it in many statements from Powell and other Fed officials, as I’ve discussed here and here. But these are, obviously, just statements. The size and design of ARPA is a more consequential rejection of this catechism. Without being described as such, it’s a decisive recognition of half a dozen points that those of us on the left side of the macroeconomic debate have been making for years.

1. The official unemployment rate is an unreliable guide to the true degree of labor market slack, all the time and especially in downturns. Most of the movement into and out of employment is from people who are not officially counted as unemployed. To assess labor market slack, we should also look at the employment-population ratio, and also at more direct measures of workers’ bargaining power like quit rates and wage increases. By these measures, the US pre-pandemic was still well short of the late 1990s.  More broadly, there is not a well defined labor force, but a  smooth gradient of proximity to employment. The short-term unemployed are the closest, followed by the longer-term unemployed, employed people seeking additional work, discouraged workers, workers disfavored by employers due to ethnicity, credentials, etc. Beyond this are people whose claim on the social product is not normally exercised by paid labor – retired people, the disabled, full-time caregivers – but might come to be if labor market conditions were sufficiently favorable. 

2. The balance of macroeconomic risks is not symmetrical. We don’t live in an economy that fluctuates around a long-term growth path, but one that periodically falls into recessions or depressions. These downturns are a distinct category of events, not a random “shock” to production or desired spending. Economic activity is a complex coordination problem; there are many ways it can break down or be interrupted that result in a fall in  spending, but not really any way it can abruptly accelerate. (There are no “positive shocks” for the same reason that there are lots of poisons but no wonder drugs.) It’s easy to imagine real-world developments that could causes businesses to abruptly cut back their investment plans, but not that would cause them to suddenly and unexpectedly scale them up. In real economies, demand shortfalls are much more frequent, persistent and damaging than is overheating. And to the extent the latter is a problem, it is much easier to interrupt the flow of spending than to restart it. 

3. The existence of hysteresis is one important reason that demand shortfalls are much more costly than overshooting. Overheating may have short-term costs in higher inflation, inflated asset prices and a redistribution of income toward relatively scarce factors (e.g. urban land), but it also is associated with a long-term increase in productive capacity — one that may eventually close the inflationary gap on its own. Shortfalls on the other hand lead to a reduction in potential output, and so may become self-perpetuating as potential GDP declines. Hysteresis also means that we cannot count on the economy returning to its long-term trend on its own — big falls in demand may persist indefinitely unless they are offset by some large exogenous boost to demand. Which in turn means that standard estimates of potential output understate the capacity of output to respond to higher spending. 

4. A full employment or high pressure economy has benefits that go well beyond the direct benefits of higher incomes and output. Hysteresis is part of this — full employment is a spur to innovation and faster productivity growth. But there are also major implications for the distribution of income. Those who are most disadvantaged in the labor market, are the ones who benefit most from very low unemployment. The World War II experience, and the subsequent evolution of the racial wage gap, suggests that historically, sustained tight labor markets have been the most powerful force for closing the gap between black and white wages.

I’m not sure how much people in the administration and Congress were actually making arguments like these in framing the bill. But even if they weren’t explicitly argued for, some mix of them logically follows from the willingness to pass something so much larger than the conventional estimates of the output gap would imply. Some mix of them also must underly the repeated statements that we can’t do too much, only too little, and from the recognition that the costs of an inadequate stimulus in 2009 were not just lower output for a year or two, but  an extended period of slow growth and stagnant wages. When Schumer says that in 2009, “we cut back on the stimulus dramatically and we stayed in recession for five years,” he is espousing a model of hysteresis, even if he doesn’t use the word.

On other points, there’s a more direct link between the debate over the bill and the shift in economic vision it implies.

5. Public debt doesn’t matter. Maybe I missed it, but as far as I can tell, in the push for the Rescue Plan neither the administration nor the Congressional leadership made even a gesture toward deficit reduction, not even a pro forma comment that it might be desirable in principle or in the indefinite long run. The word “deficit” does not seem to have occurred in any official statement from the president since early February — and even then it was in the form of “it’s a mistake to worry about the deficit.” Your guide to being a savvy political insider suggests appropriate “yes, buts” to the Rescue Plan — too much demand will cause inflation, or alternatively that demand will collapse once the spending ends. Nothing about the debt. Things may change, of course, but at the moment it’s astonishing how completely we have won on this one.

6. Work incentives don’t matter. For decades, welfare measures in the US have been carefully tailored to ensure that they did not broaden people’s choices other than wage labor. The commitment to maintaining work incentives was strong enough to justify effectively cutting off all cash assistance to families without anyone in paid employment — which of course includes the poorest.  The flat $600 pandemic unemployment insurance was a radical departure from this — reaching everyone who was out of work took priority over ensuring that no one was left better off than they would be with a job. The empirical evidence that this had no effect on employment is informative about income-support programs in general. Obviously $300 is less than $600, but it maintains the priority of broad eligibility. Similarly, by allowing families with no wages to get the full benefit, making the child tax credit full refundable effectively abandons work incentives as a design principle (even if it would be better at that point to just make it a universal child allowance.) As many people have pointed out, this is at least directionally 180 degrees from Clinton-era “welfare reform.” 

7.  Direct, visible spending is better than indirect spending or spending aimed at altering incentives. For anyone who remembers the debates over the ARRA at the start of the Obama administration, it’s striking how much the Rescue Plan leans into direct, visible payments to households. The plan to allow the child tax credit to be paid out in monthly installments may have some issues (and, again, would certainly work better if it were a flat allowance rather than a tax credit) but what’s interesting here is that it reflects a view that making the payments more salient is a good thing, not a bad thing.

In other areas, the conceptual framework hasn’t moved as far as I would have hoped, though we are making progress:

8. Means testing is costly and imprecise. As Claudia Sahm, Matt Bruenig and others have forcefully argued, there’s a big disconnect between the way means testing is discussed and the way it actually operates. When the merits of income-based spending are talked about in the abstract, it’s assumed that we know every household’s income and can assign spending precisely to different income groups. But when we come to implement it, we find that the main measure of income we use is based on tax records from one to two years earlier; there are many cases where the relevant income concept isn’t obvious; and the need to document income creates substantial costs and uncertainties for beneficiaries. Raising the income thresholds for things like the child tax credit is positive, but the other side of that is that once the threshold gets high enough it’s perverse to means-test at all: In order to exclude a relatively small number of high-income families you risk letting many lower-income families fall through the cracks. 

9. Weak demand is an ongoing problem, not just a short-term one. The most serious criticism of the ARPA is, I think, that so many of its provisions are set to phase out at specific dates when they could be permanent (the child tax credit) or linked to economic conditions (the unemployment insurance provisions). This suggests an implicit view that the problems of weak demand and income insecurity are specific to the coronavirus, rather than acute forms of a chronic condition. This isn’t intended as a criticism of those who crafted the bill — it may well be true that a permanent child tax credit couldn’t be passed under current conditions.

Still, the arguments in support of many of the provisions are not specific to the pandemic, and clearly imply that these measures ought to be permanent. If the child tax credit will cut child poverty by half, why would you want to do that for only one year? If a substantial part of the Rescue Plan should on the merits be permanent, that implies a permanently larger flow of public spending. The case needs to be made for this.

10. The public sector has capacities the private sector lacks. While Biden’s ARPA is a big step forward from Obama’s ARRA in a lot of ways, one thing they have in common is a relative lack of direct public provision. The public health measures are an exception, of course, and the aid to state and local governments — a welcome contrast with ARRA — is public spending at one remove, but the great majority of the money is going to boost private spending. That’s not necessarily a bad thing in this specific context, but it does suggest that, unlike the case with public debt, the institutional and ideological obstacles to shifting activities from for-profit to public provision are still formidable. 

My goal in listing these points isn’t, to be clear, to pass judgement on the bill one way or the other. Substantively, I do think it’s a big victory and a clear sign that elections matter. But my interest in this particular post is to think about what it says about how thinking about economic policy is shifting, and how those shifts might be projected back onto economic theory.

What would a macroeconomics look like that assumed that the economy was normally well short of supply constraints rather than at potential on average, or was agnostic about whether there was a meaningful level of potential output at all? What would it look like if we thought that demand-induced shifts in output are persistent, in both directions? Without the assumption of a supply-determined trend which output always converges to, it’s not clear there’s a meaningful long run at all. Can we have a macroeconomic theory that dispenses with that?

One idea that I find appealing is to think of supply as constraining the rate of growth of output, rather than its level. This would fit with some important observable facts about the world — not just that demand-induced changes in output are persistent, but also that employment tends to grow (and unemployment tends to fall) at a steady rate through expansions, rather than a quick recovery and then a return to long-run trend. The idea that there is a demographically fixed long-run employment-population ratio flies in the face of the major shifts of employment rates within demographic groups. A better story, it seems to me, is that there is a ceiling on the rate that employment can grow — say 1.5 or 2 percent a year — without any special adjustment process; faster growth requires drawing new people into the labor force, which typically requires faster wage growth and also involves various short run frictions. But, once strong growth does generate a larger labor force, there’s no reason for it to revert back to its old trend. 

More broadly, thinking of supply constraints in terms of growth rates rather than levels would let us stop thinking about the supply side in terms of an abstract non monetary economy “endowed” with certain productive resources, and start thinking about it in terms of the coordination capabilities of markets. I feel sure this is the right direction to go. But a proper model needs to be worked out before it is ready for the textbooks.

The textbook model of labor markets that we still teach justifies a focus on “flexibility”, where real wages are determined by on productivity and a stronger position for labor can only lead to higher inflation or unemployment. Instead, we need a model where the relative position of labor affects real as well as nominal wages, and  in which faster wage growth can be absorbed by faster productivity growth or a higher wage share as plausibly as by higher prices.

Or again, how do we think about public debt and deficits once we abandon the idea that a constant debt-GDP ratio is a hard constraint? One possibility is that we think the deficit matters, but debt does not, just as we now think think that the rate of inflation matters but the absolute price level does not.  To earlier generations of economists, the idea that prices could just rise forever without limit, would have seemed insane. But today we find it perfectly reasonable, as long as the rise over any given period is not too great. Perhaps we’ll come to the same view of public debt. To the extent that we do care about the debt ratio, we need to foreground the fact that its growth over time depends as much on interest, inflation and growth rates as it does on new borrowing. For the moment, the fact that interest rates are much lower than growth rates is enough to convince people past concerns were overblown. But to regard that as a permanent rather than contingent solution, we need, at least, to get rid of the idea of a natural rate of interest. 

In short, just as a generation of mainstream macroeconomic theory was retconned into an after-the-fact argument for an inflation-targeting central bank, what we need now is textbooks and theories that bring out, systematize and generalize the reasoning that justifies a great expansion of public spending, unconstrained by conventional estimates of potential output, public debt or the need to preserve labor-market incentives. The circumstances of the past year are obviously exceptional, but that doesn’t mean they can’t be made the basis of a general rule. For the past generation, macroeconomic theory has been largely an abstracted parable of the 1970s, when high interest rates (supposedly) saved us from inflation. With luck, perhaps the next generation will learn macroeconomics as a parable of our own time, when big deficits saved us from secular stagnation and the coronavirus.

Talk on the Economic Mobilization of World War II

Two weeks ago – it feels much longer now – I was up at UMass-Amherst to give a talk on the economic mobilizaiton of World War II and its lessons for the Green New Deal.

Here is an audio recording of the talk. Including Q&A, it’s about an hour and a half. Here are the slides that I used.

 

 

The big three lessons I draw are:

1. The more rapid the economic transformation that’s required, the bigger the role the public sector needs to take, in investment especially, and more broadly in bearing risk.

2. Output can be very elastic in response to stronger demand, much more so than is usually believed. There’s a real danger that over-conservative estimates of potential output will lead us to set our sights too low.

3. Demand conditions have major effects on income distribution. Full employment is an extremely powerful tool to shift income toward the lower-paid and to less-privelged groups, even in absence of direct redistribution.

EDIT: The underlying paper is being revised to update the lessons for the present in light of the fact that “the present” is now an acute public health crisis rather than an ongoing climate crisis. The first part of the new version is here. The rest will be forthcoming in the next couple weeks.

You can also listen to an interview with me on Doug Henwood’s Behind the News here.