RI potential paper – J. W. Mason

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.

Some Interviews

One new one, and two older ones I should have posted here a while ago.

The new one is with Seth Ackerman at Jacobin. Its starting point is a new article (co-authored with Arjun Jayadev and Enno Schroeder) I have coming out in Development and Change. But it’s also a continuation of the argument I made in my earlier Jacobin piece on the socialization of finance [*], and in my talk at this year’s Left Forum. (I still hope to get a transcript of that one at some point.)

The older two are both in response to my “What Recovery?” report for the Roosevelt Institute. This one, with David Beckworth at the Mercatus Institute, was a wide-ranging conversation that touched on a lot of topics beside the immediate question of whether we should regard the US economy as having reached full employment or potential output. This one, with Joe Weisenthal and his colleagues at “What Did You Miss” on Bloomberg, was much briefer but still managed to cover a lot of ground.

Supposedly there’s also an interview with me coming out in Der Standard, an Austrian newspaper, but I’m not sure when it will appear.

If you’re reading this blog, you’ll probably find these interviews interesting.

[*] Incidentally, my preferred title was that: The Socialization of Finance. I understand why the editors changed it to the catchier imperative form, but what I liked about my original was that it could refer both to something done to finance, and something done by finance.

Reading Notes: Demand and Productivity

Here are two interesting articles on demand and productivity that people have recently brought to my attention.

The economic historian Gavin Wright — author of the classic account of the economic logic of the plantation — just sent me a piece he wrote a few years ago on the productivity boom of the 1990s. As he said in his email, his account of the ‘90s is very consistent with the suggestions I make in my Roosevelt paper about how strong demand might stimulate productivity growth.

In this article, Wright traces the idea that high wage regions will experience faster productivity growth back to H. J. Habbakuk’s 1962 American and British Technology in the Nineteenth Century. Then he assembles a number of lines of evidence that rapid wage growth drove the late-1990s productivity acceleration, rather than vice versa.

He points out that the widely-noted “productivity explosion” of the 1920s — from 1.5 percent a year to over 5 percent — was immediately preceded by a period of exceptionally strong wage growth: “The real price of labor in the 1920s … was between 50 and 70 percent higher than a decade earlier.” [1] The pressure of high wages, he suggests, encouraged the use of electricity and other general-purpose technologies, which had been available for decades but only widely adopted in manufacturing in the 1920s. Conversely, we can see the productivity slowdown of the 1970s as, at least in part, a result of the deceleration of wage growth, which — Wright argues — was the result of institutional changes including the decline of unions, the erosion of the minimum wage and other labor regulations, and more broadly the shift back toward “‘flexible labor markets,’ reversing fifty years of labor market policy.”

Turning to the 1990s, the starting point is the sharp acceleration of productivity in the second half of the decade. This acceleration was very widely shared, including sectors like retail where historically productivity growth had been limited. The timing of this acceleration has been viewed as a puzzle, with no “smoking gun” for simultaneous productivity boosting innovations across this range of industries over a short period. But “if you look at the labor market, you can find a smoking gun in the mid-1990s. … real hourly wages finally began to rise at precisely that time, after more than two decades of decline. … Unemployment rates fell below 4 percent — levels reached only briefly in the 1960s… Should it be surprising that employers turned to labor-saving technologies at this time?” This acceleration in real wages, Wright argues, was not the result of higher productivity or other supply-side factors; rather “it is most plausibly attributed to macroeconomic conditions, when an accommodating Federal Reserve allowed employment to press against labor supply for the first time in a generation.”

The productivity gains of the 1990s did, of course, involve new use of information technology. But the technology itself was not necessarily new. “James Cortada [2004] lists eleven key IT applications in the retail industry circa 1995-2000, including electronic shelf levels, scanning, electronic fund transfer, sales-based ordering and internet sales … with the exception of e-business, the list could have come from the 1970s and 1980s.”

Wright, who is after all a historian, is careful not to argue that there is a general law linking higher wages to higher productivity in all historical settings. As he notes, “such a claim is refuted by the experience of the 1970s, when upward pressures on wages led mainly to higher inflation…” In his story, both sides are needed — the technological possibilities must exist, and there must be sufficient wage pressure to channel them into productivity-boosting applications. I don’t think anyone would say he’s made a decisive case , but if you’re inclined to a view like this the article certainly gives you more material to support it.

A rather different approach to these questions is this 2012 paper by Servaas Storm and C. W. M. Naastepad. Wright is focusing on a few concrete episodes in the history of a particular country, which he explores using a variety of material — survey and narrative as well as conventional economic data. Storm and Naastepad are proposing a set of general rules that they support with a few stylized facts and then explore via of the properties of a formal model. There are things to be learned from both approaches.

In this case the model is simple: output is demand-determined. Demand is either positive or negative function of the wage share (i.e. the economy is either wage-led or profit-led). And labor productivity is a function of both output and the wage, reflecting two kinds of channels by which demand can influence productivity. And an accounting identity says that employment growth is qual to output growth less labor productivity growth. The productivity equation is the distinctive feature here. Storm and Naastepad adopt as “stylized facts” — derived from econometric studies but not discussed in any detail — that both parameters are on the order of 0.4: An additional one percent growth in output, or in wages, will lead to an 0.4 percent growth in labor productivity.

This is a very simple structure but it allows them to draw some interesting conclusions:

– Low wages may boost employment not through increased growth or competitiveness, but through lower labor productivity. (They suggest that this is the right way to think about the Dutch “employment miracle of the 1990s.)

– Conversely, even where demand is wage-led (i.e. a shift to labor tends to raise total spending) faster wage growth is not an effective strategy for boosting employment, because productivity will rise as well. (Shorter hours or other forms of job-sharing, they suggest, may be more successful.)

– Where demand is strongly wage-led (as in the Scandinavian countries, they suggest), profits will not be affected much by wage growth. The direct effect of higher wages in this case could be mostly or entirely offset by the combination of higher demand and higher productivity. If true, this has obvious implications for the feasibility of the social democratic bargain there.

– Where demand is more weakly wage-led or profit-led (as with most structuralists, they see the US as the main example of the latter), distributional conflicts will be more intense. On the other hand, in this case the demand and productivity effects work together to make wage restraint a more effective strategy for boosting employment.

It’s worth spelling out the implications a bit more. A profit-led economy is one in which investment decisions are very sensitive to profitability. But investment is itself a major influence on profit, as a source of demand and — emphasized here — as a source of productivity gains that are captured by capital. So wage gains are more threatening to profits in a setting in which investment decisions are based largely on profitability. In an environment in which investment decisions are motivated by demand or exogenous animal spirits (“only a little more than an expedition to the South Pole, based on a calculation of benefits to come”), capitalists have less to fear from rising wages. More bluntly: one of the main dangers to capitalists of a rise in wages, is their effects on the investment decisions of other capitalists.

What Recovery: Reading Notes

My Roosevelt Institute paper on potential output came out last week. (Summary here.) The paper has gotten some more press since Neil Irwin’s Times piece, including Ryan Cooper in The Week and Felix Salmon in Slate. My favorite headline is from Boing Boing: American Wages Are So Low, the Robots Don’t Want Your Jobs.

In the paper I tried to give a fairly comprehensive overview of the evidence and arguments that the US economy is not in any meaningful sense at potential output or full employment. But of course it was just one small piece of a larger conversation. Here are a few things I’ve found interesting recently on the same set of issues. .

Perhaps the most important new academic contribution to this debate is this paper by Olivier Coibion, Yuriy Gorodnichenko, and Mauricio Ulate, on estimates of potential output, which came out too late for me to mention in the Roosevelt report. Their paper rigorously demonstrates that, despite their production-function veneer, the construction of potential output estimates ensures that any persistent change in growth rates will appear as a change in potential. It follows that there is “little value added in estimates of potential GDP relative to simple measures of statistical trends.” (Matthew Klein puts it more bluntly in an Alphaville post discussing the paper: “‘Potential’ output forecasts are actually worthless.”) The paper proposes an alternative measure of potential output, which they suggest can distinguish between transitory demand shocks and permanent shifts in the economy’s productive capacity. This alternative measure gives a very similar estimate for the output gap as simply looking at the pre-2008 forecasts or extrapolating from the pre-2008 trend. “Our estimates imply that U.S. output remains almost 10 percentage points below potential output, leaving ample room for policymakers to close the gap through demand-side policies if they so chose to.” Personally, I ‘m a little less convinced by their positive conclusions than by their negative ones. But this paper should definitely put to the rest the idea (as in last year’s notorious CEA-chair letter) that it is obviously wrong — absurd and unserious — that a sufficient stimulus could deliver several years of 4 percent real growth, until GDP returned to its pre-recession trend. It may or may not be true, but it isn’t crazy.

Many of the arguments in my paper were also made in this valuable EPI report by Josh Bivens, reviving the old idea of a “high pressure economy”. Like me, Bivens argues that slow productivity growth is largely attributable to low investment, which in turn is due to weak demand and slow wage growth, which blunts the incentive for business to invest in labor-saving technology. One important point that Bivens makes that I didn’t, is that much past variation in productivity growth has been transitory; forecasts of future productivity growth based on the past couple of years have consistently performed worse than forecasts based on longer previous periods. So historical evidence gives us no reason see the most recent productivity slowdown as permanent. My one quibble is that he only discusses faster productivity growth and higher inflation as possible outcomes of a demand-driven acceleration in wages. This ignores the third possible effect, redistribution from from profits to wages — in fact a rise in the labor share is impossible without a period of “overfull” employment.

Minneapolis Fed president Neel Kashkari wrote a long post last fall on “diagnosing and treating the slow recovery.” Perhaps the most interesting thing here is that he poses the question at all. There’s a widespread view that once you correct for demographics, the exceptional performance of the late 1990s, etc., there’s nothing particularly slow about this recovery — no problem to diagnose or treat.

Another more recent post by Kashkari focuses on the dangers of forcing the Fed to mechanically follow a Taylor rule for setting interest rates. By his estimate, this would have led to an additional 2.5 million unemployed people this year. It’s a good illustration of the dangers of taking the headline measures of economic performance too literally. I also like its frank acknowledgement that the Fed — like all real world forecasters — rejects rational expectations in the models it uses for policymaking.

Kashkari’s predecessor Narayan Kocherlakota — who seems to agree more with the arguments in my paper — has a couple short but useful posts on his personal blog. The first, from a year ago, is probably the best short summary of the economic debate here that I’ve seen. Perhaps the key analytic point is that following a period of depressed investment, the economy may reach full employment given the existing capital stock while it is still well short of potential. So a period of rapid wage growth would not necessarily mean that the limits of expansionary policy have ben reached, even if those wage gains were fully passed through to higher prices. His emphasis:

Because fiscal policy has been too tight, we have too little public capital. … At the same time, physical investment has been too low… Conditional on these state variables, we might well be close to full employment. … But, even though we’re close to full employment, there’s a lot of room for super-normal growth. Both capital and TFP are well below their [long run level]. The full-employment growth rate is going to be well above its long-run level for several years. We can’t conclude the economy is overheating just because it is growing quickly.

His second post focuses on the straightforward but often overlooked point that policy should take into account not just our best estimates but our uncertainty about them, and the relative risks of erring on each side. And if there is even a modest chance that more expansionary policy could permanently raise productivity, then the risks are much greater on the over-contractionary side. [1] In particular, if we are talking about fiscal stimulus, it’s not clear that there are any costs at all. “Crowding out” is normally understood to involve a rise in interest rates and a shift from private investment to public spending. In the current setting, there’s a strong case that higher interest rates at full employment would be a good thing (at least as long as we still rely on as the main tool of countercyclical policy). And it’s not obvious, to say the least, that the marginal dollar of private investment is more socially useful than many plausible forms of public spending. [2] Kashkari has a post making a similar argument in defense of his minority vote not to raise rates at the most recent FOMC meeting. (Incidentally, FOMC members blogging about their decisions is a trend to be encouraged.)

In a post from March which I missed at the time, Ryan Avent tries to square the circle of job-destroying automation and slow productivity growth. One half of the argument seems clearly right to me: Abundant labor and low wages discourage investment in productivity-raising technologies. As Avent notes, early British and even more American industrialization owe a lot to scarce labor and high wages. The second half of the argument is that labor is abundant today precisely because so much has been displaced by technology. His claim is that “robots taking the jobs” is consistent with low measured productivity growth if the people whose jobs are taken end up in a part of the economy with a much lower output per worker. I’m not sure if this works; this seems like the rare case in economics where an eloquent story would benefit from being re-presented with math.

Along somewhat similar lines, Simon Wren-Lewis points out that unemployment may fall because workers “price themselves into jobs” by accepting lower-wage (and presumably lower-productivity) jobs. But this doesn’t mean that the aggregate demand problem has been solved — instead, we’ve simply replaced open unemployment with what Joan Robinson called “disguised unemployment,” as some of people’s capacity for work continues to go to waste even while they are formally employed. “But there is a danger that central bankers would look at unemployment, … and conclude that we no longer have inadequate aggregate demand…. If demand deficiency is still a problem, this would be a huge and very costly mistake.”

Karl Smith at the Niskanen Center links this debate to the older one over the neutrality of money. Central bank interventions — and aggregate demand in general — are understood to be changes in the flow of money spending in the economy. But a lon-standing tradition in economic theory says that money should be neutral in the long run. As we are look at longer periods, changes in output and employment should depend more and more on real resources and technological capacities, and less and less on spending decisions — in the limit not at all. If you want to know why GDP fell in one quarter but rose in the next (this is something I always tell my undergraduates) you need to ask who chose to reduce their spending in the first period and who chose to increase it in the first. But if you want to know why we are materially richer than our grandparents, it would be silly to say it’s because we choose to spend more money. This is the reason why I’m a bit impatient with people who respond to the fact that, relative to the pre-2008 trend, output today has not recovered from the bottom of the recession, by saying “the trend doesn’t matter, deviations in output are always persistent.” This might be true but it’s a radical claim. It means you either take the real business cycle view that there’s no such thing as aggregate demand, even recessions are due to declines in the economy’s productive potential; or you must accept that in some substantial sense we really are richer than our grandparents because we spend more money. You can’t assert that GDP is not trend-stationary to argue against an output gap today unless you’re ready to accept these larger implications.

The invaluable Tom Walker has a fascinating post going back to even older debates, among 19th century anti-union and pro-union pamphleters, about whether there was a fixed quantity of labor to be performed and whether, in that case, machines were replacing human workers. The back and forth (more forth than back: there seem to be a lot more anti-labor voices in the archives) is fun to read, but what’s the payoff for todays’ debates?

The contemporary relevance of this excursion into the archives is that economic policy and economic thought walks on two legs. Conservative economists hypocritically but strategically embrace both the crowding out arguments for austerity and the projected lump-of-labor fallacy claims against pensions and shorter working time. They are for a “fixed amount” assumption when it suits their objectives and against it when it doesn’t. There is ideological method to their methodological madness. That consistency resolves itself into the “self-evidence” that nothing can be done.

That’s exactly right. When we ask why labor’s share has fallen so much over the past generation, we’re told it’s because of supply and demand — an increased supply of labor from China and elsewhere, and a decreased demand thanks to technology. But if it someone says that it might be a good idea then to limit the supply of labor (by lowering the retirement age, let’s say) and to discourage capital-intensive production, the response is “are you crazy? that will only make everyone poorer, including workers.” Somehow distribution is endogenous when it’s a question of shifts in favor of capital, but becomes exogenously fixed when it’s a question of reversing them.

A number of heterdox writers have identified the claim that productivity growth depends on demand as Verdoorn’s law (or the Kaldor-Verdoorn Law). For example, the Post Keynesian blogger Ramanan mentions it here and here. I admit I’m a bit dissatisfied with this “law”. It’s regularly asserted by heterodox people but you’ll scour our literature in vain looking for either a systematic account of how it is supposed to operate or quantitative evidence of how and how much (or whether) it does.

Adam Ozimek argues that the recent rise in employment should be seen as an argument for continued expansionary policy, not a shift away from it. After all, a few years ago many policymakers believed such a rise was impossible, since the decline in employment was supposed to be almost entirely structural.

Finally, Reihan Salam wants to enlist me for the socialist flank of a genuinely populist Trumpism. This is the flipside of criticism I’ve sometimes gotten for making this argument — doesn’t it just provide intellectual ammunition for the Bannon wing of the administration and its calls for vast infrastructure spending, which is also supposed to boost demand and generate much faster growth? Personally I think you need to make the arguments for what you think is true regardless of their political valence. But I might worry about this more if I believed there was even a slight chance that Trump might try to deliver for his working-class supporters.

[1] Kocherlakota talks about total factor productivity. I prefer to focus on labor productivity because it is based on directly observable quantities, whereas TFP depends on estimates not only of the capital stock but of various unobservable parameters. The logic of the argument is the same either way.

[2] I made similar arguments here.

EDIT: My comments on the heterodox literature on the Kaldor-Verdoorn Law were too harsh. I do feel this set of ideas is underdeveloped, but there is more there than my original post implied. I will try to do a proper post on this work at some point.

The Big Question for Macroeconomic Policy: Is This Really Full Employment?

Cross-posted from the Roosevelt Institute’s Next New Deal blog. This is a summary of my new paper What Recovery? The Case for Continued Expansionary Policy, also discussed in Neil Irwin’s July 26 article in the Times.

“Right now,” wrote Senator Chuck Schumer in a New York Times op-ed on Monday, “millions of unemployed or underemployed people, particularly those without a college degree, could be brought back into the labor force” with appropriate government policies. With this seemingly anodyne point, Schumer took sides in a debate that has sharply divided economists and policymakers: Is the US economy today operating at potential, with enough spending to make full use of its productive capacity? Or is there still substantial slack, unused capacity that could be put to work if someone — households, businesses or governments — decided to spend more? Is there an aggregate-demand problem that government should be trying to solve?

It’s difficult to answer this question because the economic signals seem to point in conflicting directions. Despite the recession officially ending in June 2009 and the economy enjoying steady growth for the past eight years, GDP is still far below the pre-2008 trend. If we compare GDP to forecasts made before the recession, the gap that opened up during the recession has not closed at all — in fact, it continues to get wider. Meanwhile, the official unemployment rate — probably the most watched indicator for the state of aggregate demand — is down to 4.4%, well below the level that was considered full employment even a few years ago. But this positive performance only partially reflects an increase in the number of Americans with jobs; mostly it comes from a decline in the size of the labor force — people who have or are seeking jobs. The fraction of the adult population employed is down to 60 percent from 63 percent a decade ago (and nearly 65 percent at the end of the 1990s).

Is this decline in the fraction of people employed the inevitable result of an aging population and similar demographic changes, or is it a sign that, despite the low measured unemployment rate, the economy is still far short of full employment? The Federal Reserve — one of the main sites of macroeconomic policy — has already indicated its belief that full employment has been reached by raising interest rates 3 times since December 2016. Fed Chair and Janet Yellen are evidently convinced that the economy has reached its potential — that, given the real resources available, output and employment are as high as can reasonably be expected.

Other policymakers have been divided on the question, in ways that often cut across partisan lines. Senator Schumer’s statement — that the decline in employment is not an inevitable trend but rather a problem that government can and should solve — is a sign of new clarity coming to this murky debate. Along with his call for $1 trillion in new infrastructure spending, it’s an important acknowledgement that, despite the progress made since 2008, the country remains far from full employment.

In a new paper out this week, we at the Roosevelt Institute offer support for the emerging consensus that the economy needs policies to boost demand. The paper reviews the available data on where the economy is relative to its potential. We find that the balance of evidence suggests there is still a great deal of space for more expansionary policy.

We offer several lines of argument in support of this conclusion.

GDP has not recovered from the recession. GDP remains about 10 percent below both the long-term trend and the level that was predicted by the CBO and other forecasts prior to the 2008–2009 recession. There is no precedent in the postwar period for such a persistent decline in output. During the sixty years between 1947 and 2007, growth lost in recessions was always regained in the subsequent recovery.

The aging population does not explain low labor force participation. It is true that an aging population should contribute to lower employment, since older people are less likely to work than younger people. But this simple demographic story cannot explain the full fall in employment. Starting from the employment peak in 2000, aging trends only explain about half the decrease in employment that has actually occurred. And there are good reasons to think that even this overstates the role of demographics. First, during the same period, education levels have increased. Historically, higher education has been associated with higher employment rates, just as a share of elderly people has been associated with less employment; statistically, these two effects should just about cancel out. Second, the post-recession fall in employment rates is not concentrated in older age groups, but among people in their 20s — something that a demographic story cannot explain.

The weak economy has held back productivity. About half the shortfall in GDP relative to the pre-2008 trend is explained by exceptionally slow productivity growth — that is, slow growth in output per worker. While many people assume that productivity is the result of technological progress outside the reach of macroeconomic policy, there are good reasons to think that the productivity slowdown is at least in part due to weak demand. Among the many possible links: Business investment, which is essential to raising productivity, has been extraordinarily weak over the past decade, and economists have long believed that demand is a central factor driving investment. And slow wage growth — a sign of labor-market weakness — reduces the incentive to adopt productivity-boosting technology.

Only a demand story makes sense. The overall economic picture is hard to understand except in terms of a continued demand shortfall. If employment is falling due to demographics, that should be associated with rising productivity and wages, as firms compete for scarce labor. If productivity growth is slow because there aren’t any more big innovations to make, that should be associated with faster employment growth and low profits, as firms can no longer find new ways to replace labor with capital. But neither of these scenarios match the actual economy. And both stories predict higher inflation, rather than the persistent low inflation we have actually encouraged. So even if supply-side stories explain individual pieces of macroeconomic data, it is almost impossible to make sense of the big picture without a large fall in aggregate demand.

Austerity is riskier than stimulus. Finally, we argue that, if policymakers are uncertain about how much space the economy has for increased demand, they should consider the balance of risks on each side. Too much stimulus would lead to higher inflation — easy to reverse, and perhaps even desirable, given the continued shortfall of inflation relative to the official 2 percent target. An overheated economy would also see real wages rise faster than productivity. While policymakers often see this as something to avoid, the decline in the wage share over the past decade cannot be reversed without a period of such “excess” wage growth. On the other hand, if there is still an output gap, failure to take aggressive steps to close it means foregoing literally trillions of dollars of useful goods and services and condemning millions of people to joblessness.

Fortunately, the solution to a demand shortfall is no mystery. Since Keynes, economists have known that when an economy is operating below its potential, all that is needed is for someone to spend more money. Of course, it’s best if that spending also serves some useful social purpose; exactly what that should look like will surely be the subject of much debate to come. But the first step is to agree on the problem. Today’s economy is still far short of its potential. We can do better.

Posts in Three Lines

I haven’t been blogging much lately. I’ve been doing real work, some of which will be appearing soon. But if I were blogging, here are some of the posts I might write.

Lessons from the 1990s. I have a new paper coming out from the Roosevelt Institute, arguing that we’re not as close to potential as people at the Fed and elsewhere seem to believe, and as I’ve been talking with people about it, it’s become clear that your priors depend a lot on how you think of the rapid growth of the 1990s. If you think it was a technological one-off, with no value as precedent — a kind of macroeconomic Bush v. Gore — then you’re likely to see today’s low unemployment as reflecting an economy working at full capacity, despite the low employment-population ratio and very weak productivity growth. But if you think the mid-90s is a possible analogue to the situation facing policymakers today, then it seems relevant that the last sustained episode of 4 percent unemployment led not to inflation but to employers actively recruiting new entrants to the laborforce among students, poor people, even prisoners.

Inflation nutters. The Fed, of course, doesn’t agree: Undeterred by the complete disappearance of the statistical relationship between unemployment and inflation, they continue to see low unemployment as a threatening sign of incipient inflation (or something) that must be nipped in the bud. Whatever other effects rate increases may have, the historical evidence suggests that one definite consequence will be rising private and public debt ratios. Economists focus disproportionately on the behavioral effects of interest rate changes and ignore their effects on the existing debt stock because “thinking like an economist” means, among other things, thinking in terms of a world in which decisions are made once and for all, in response to “fundamentals” rather than to conditions inherited from the past.

An army with only a signal corps. What are those other effects, though? Arguments for doubting central bankers’ control over macroeconomic outcomes have only gotten stronger than they were in the 2000s, when they were already strong; at the same time, when the ECB says, “let the government of Spain borrow at 2 percent,” it carries only a little less force than the God of genesis. I think we exaggerate power of central banks over real economy, but underestimate their power over financial markets (with the corollary that economists — heterodox as much as mainstream — see finance and real activity as much more tightly linked than they are).

It’s easy to be happy if you’re heterodox. This spring I was at a conference up at the University of Massachusetts, the headwaters of American heterodox economics, where I did my Phd. Seeing all my old friends reminded me what good prospects we in the heterodox world have – literally everyone I know from grad school has a good job. If you are wondering whether your prospects would be better at a nowhere-ranked heterodox economics program like UMass or a top-ranked program in some other social science, let me assure you, it’s the former by a mile — and you’ll probably have better drinking buddies as well.

The euro is not the gold standard. One of the topics I was talking about at the UMass conference was the euro which, I’ve argued, was intended to create something like a new gold standard, a hard financial constraint on governments. But that that was the intention doesn’t mean its the reality — in practice the TARGET2 system means that national central banks don’t face any binding constraint , unlike under the gold standard the central bank is “outside” the national monetary membrane. In this sense the euro is structurally more like Keynes’ proposals at Bretton Woods, it’s just not Keynes running it.

Can jobs be guaranteed? In principle I’m very sympathetic to the widespread (at least among my friends on social media) calls for a job guarantee. It makes sense as a direction of travel, implying a commitment to a much lower unemployment rate, expanded public employment, organizing work to fit people’s capabilities rather than vice versa, and increasing the power of workers vis-a-vis employers. But I have a nagging doubt: A job is contingent by its nature – without the threat of unemployment, can there even be employment as we know it?

The wit and wisdom of Haavelmo. I was talking a while back about Merijn Knibbe’s articles on the disconnect between economic theory and the national accounts with my friend Enno, and he mentioned Trygve Haavelmo’s 1944 article on The Probability Approach in Econometrics, which I’ve finally gotten around to reading. One of the big points of this brilliant article is that economic variables, and the models they enter into, are meaningful only via the concrete practices through which the variables are measured. A bigger point is that we study economics in order to “become master of the happenings of real life”: You can contribute to economics in the course of advancing a political project, or making money in financial markets, or administering a government agency (Keynes did all three), but you will not contribute if you pursue economics as an end in itself.

Coney Island. Laura and I took the boy down to Coney Island a couple days ago, a lovely day, his first roller coaster ride, rambling on the beach, a Cyclones game. One of the wonderful things about Coney Island is how little it’s changed from a century ago — I was rereading Delmore Schwartz’s In Dreams Begin Responsibilities the other day, and the title story’s description of a young immigrant couple walking the boardwalk in 1909 could easily be set today — so it’s disconcerting to think that the boy will never take his grandchildren there. It will all be under water.

What Does Crowding Out Even Mean?

Paul Krugman is taking some guff for this column where he argues that the US economy is now at potential, or full employment, so any shift in the federal budget toward deficit will just crowd out private demand.

Whether higher federal spending (or lower taxes) could, in present conditions, lead to higher output is obviously a factual question, on which people may read the evidence in different ways. As it happens, I don’t agree that current output is close to the limits of current productive capacity. But that’s not what I want to write about right now. Instead I want to ask: What concretely would crowding out even mean right now?

Below, I run through six possible meanings of crowding out, and then ask if any of them gives us a reason, even in principle, to worry about over-expansionary policy today. (Another possibility, suggested by Jared Bernstein, is that while we don’t need to worry about supply constraints for the economy as a whole, tax cuts could crowd out useful spending due to some unspecified financial constraint on the federal government. I don’t address that here.) Needless to say, doubts about the economic case for crowding-out are in no way an argument for the specific deficit-boosting policies favored by the new administration.

The most straightforward crowding-out story starts from a fixed supply of private savings. These savings can either be lent to the government, or to business. The more the former takes, the less is left for the latter. But as Keynes pointed out long ago, this simple loanable-funds story assumes what it sets out to prove. The total quantity of saving is fixed only if total income is fixed. If higher government spending can in fact raise total income, it will raise total saving as well. We can only tell a story about government and business competing for a given pool of saving if we have already decided for some other reason that GDP can’t change.

The more sophisticated version, embodied in the textbook ISLM model, postulates a fixed supply of money, rather than saving. [1] In Hicks’ formulation, money is used both for transactions and as the maximally liquid store of wealth. The higher is output, the more money is needed for transactions, and the less is available to be held as wealth. By the familiar logic of supply and demand, this means that wealthholders must be paid more to part with their remaining stock of money. The price wealthholders receive to give up their money is interest; so as GDP rises, so does the interest rate.

Unlike the loanable funds story with fixed saving, this second story does give a logically coherent account of crowding out. In a world of commodity money, if such ever was, it might even be literally true. But in a world of bank-created credit money, it’s at best a metaphor. Is it a useful metaphor? That would require two things. First, that the interest rate (whichever one we are interested in) is set by the financial system. And second, that the process by which this happens causes rates to systematically rise with demand. The first premise is immediately rejected by the textbooks, which tell us that “the central bank sets the interest rate.” But we needn’t take this at face value. There are many interest rates, not just one, and the spreads between them vary quite a bit; logically it is possible that strong demand could lead to wider spreads, as banks stretch must their liquidity further to make more loans. But in reality, the opposite seems more likely. Government debt is a source of liquidity for private banks, not a use of it; lending more to the government makes it easier, not harder, for them to also lend more to private borrowers. Also, a booming economy is one in which business borrowers are more profitable; marginal borrowers look safer and are likely to get better terms. And rising inflation, obviously, reduces the real value of outstanding debt; however annoying this is to bankers, rationally it makes them more willing to lend more to their now less-indebted clients. Wicksell, the semi-acknowledged father of modern central banking theory, built his big book around the premise that in a credit-money system, inflation would give private banks no reason to raise interest rates.

And in fact this is what we see. Interest rate spreads are narrow in booms; they widen in crises and remain wide in downturns.

So crowding out mark two, the ISLM version, requires us to accept both that central banks cannot control the economically relevant interest rates, and that private banks systematically raise interest rates when times are good. Again, in a strict gold standard world there might something to this — banks have to raise rates, their gold reserves are running low — but if we ever lived in that world it was 150 or 200 years ago or more.

A more natural interpretation of the claim that the economy is at potential, is that any further increase in demand would just lead to inflation. This is the version of crowding out in better textbooks, and also the version used by MMT folks. On a certain level, it’s obviously correct. Suppose the amount of money-spending in an economy increases. Then either the quantity of goods and services increases, or their prices do. There is no third option: The total percent increase in money spending, must equal the sum of the percent increase in “real” output and the percent increase in average prices. But how does the balance between higher output and higher prices play out in real life? One possibility is that potential output is a hard line: each dollar of spending up to there increases real output one for one, and leaves prices unchanged; each dollar of spending above there increases prices one for one and leaves output unchanged. Alternatively, we might imagine a smooth curve where as spending increases, a higher fraction of each marginal dollar translates into higher prices rather than higher output. [2] This is certainly more realistic, but it invites the question of which point exactly on this curve we call “potential”. And it awakens the great bane of postwar macro – an inflation-output tradeoff, where the respective costs and benefits must be assessed politically.

Crowding out mark three, the inflation version, is definitely right in some sense — you can’t produce more concrete use values without limit simply by increasing the quantity of money borrowed by the government (or some other entity). But we have to ask first, positively, when we will see this inflation, and second, normatively, how we value lower inflation vs higher output and income.

In the post-1980s orthodoxy, we as society are never supposed to face these questions. They are settled for us by the central bank. This is the fourth, and probably most politically salient, version of crowding out: higher government spending will cause the central bank to raise interest rates. This is the practical content of the textbook story, and in fact newer textbooks replace the LM curve — where the interest rate is in some sense endogenous — with a straight line at whatever interest rate is chosen by the central bank. In the more sophisticated textbooks, this becomes a central bank reaction function — the central bank’s actions change from being policy choices, to a fundamental law of the economic universe. The master parable for this story is the 1990s, when the Clinton administration came in with big plans for stimulus, only to be slapped down by Alan Greenspan, who warned that any increase in public spending would be offset by a contractionary shift by the federal reserve. But once Clinton made the walk to Canossa and embraced deficit reduction, Greenspan’s fed rewarded him with low rates, substituting private investment in equal measure for the foregone public spending. In the current contest, this means: Any increase in federal borrowing will be offset one for one by a fall in private investment — because the Fed will raise rates enough to make it happen.

This story is crowding out mark four. It depends, first, on what the central bank reaction function actually is — how confident are we that monetary policy will respond in a direct, predictable way to changes in the federal budget balance or to shifts in demand? (The more attention we pay to how the monetary sausage gets made, the less confident we are likely to be.) And second, on whether the central bank really has the power to reliably offset shifts in fiscal policy. In the textbooks this is taken for granted but there are reasons for doubt. It’s also not clear why the actions of the central bank should be described as crowding out by fiscal policy. The central bank’s policy rule is not a law of nature. Unless there is some other reason to think expansionary policy can’t work, it’s not much of an argument to say the Fed won’t allow it. We end up with something like: “Why can’t we have deficit-financed nice things?” “Because the economy is at potential – any more public spending will just crowd out private spending.” “How will it be crowded out exactly?” “Interest rates will rise.” “Why will they rise?” “Because the federal reserve will tighten.” “Why will they tighten?” “Because the economy is at potential.”

Suppose we take the central bank out of the picture. Suppose we allow supply constraints to bind on their own, instead of being anticipated by the central planners at the Fed. What would happen as demand pushed up against the limits of productive capacity? One answer, again, is rising inflation. But we shouldn’t expect prices to all rise in lockstep. Supply constraints don’t mean that production growth halts at once; rather, bottlenecks develop in specific areas. So we should expect inflation to begin with rising prices for inputs in inelastic supply — land, oil, above all labor. Textbook models typically include a Phillips curve, with low unemployment leading to rising wages, which in turn are passed on to higher prices.

But why should they be passed on completely? It’s easy to imagine reasons why prices don’t respond fully or immediately to changes in wages. In which case, as I’ve discussed before, rising wages will result in an increase in the wage share. Some people will object that such effects can only be temporary. I’m not sure this makes sense — why shouldn’t labor, like anything else, be relatively more expensive in a world where it is relatively more scarce? But even if you think that over the long-term the wage share is entirely set on the supply side, the transition from one “fundamental” wage share to another still has to involve a period of wages rising faster or slower than productivity growth — which in a Phillips curve world, means a period above or below full employment.

We don’t hear as much about the labor share as the fundamental supply constraint, compared with savings, inflation or interest rates. But it comes right out of the logic of standard models. To get to crowding out mark five, though, we have to take one more step. We have to also postulate that demand in the economy is profit-led — that a distributional shift from profits toward wages reduces desired investment by more than it increases desired consumption. Whether (or which) real economies display wage-led or profit-led demand is a subject of vigorous debate in heterodox macro. But there’s no need to adjudicate that now. Right now I’m just interested in what crowding out could possibly mean.

Demand can affect distribution only if wage increases are not fully passed on to prices. One reason this might happen is that in an open economy, businesses lack pricing power; if they try to pass on increased costs, they’ll lose market share to imports. Follow that logic to its endpoint and there are no supply constraints — any increase in spending that can’t be satisfied by domestic production is met by imports instead. For an ideal small, open economy potential output is no more relevant than the grocery store’s inventory is for an individual household when we go shopping. Instead, like the household, the small open economy faces a budget constraint or a financing constraint — how much it can buy depends on how much it can pay for.

Needless to say, we needn’t go to that extreme to imagine a binding external constraint. It’s quite reasonable to suppose that, thanks to dependence on imported inputs and/or demand for imported consumption goods, output can’t rise without higher imports. And a country may well run out of foreign exchange before it runs out of domestic savings, finance or productive capacity. This is the idea behind multiple gap models in development economics, or balance of payments constrained growth. It also seems like the direction orthodoxy is heading in the eurozone, where competitiveness is bidding to replace inflation as the overriding concern of macro policy.

Crowding out mark six says that any increase in demand from the government sector will absorb scarce foreign exchange that will no longer be available to private sector. How relevant it is depends on how inelastic import demand is, the extent to which the country as a whole faces a binding budget or credit constraint and, what concrete form that constraint faces — what actually happens if international creditors are stiffed, or worry they might be? But the general logic is that higher spending will lead to a higher trade deficit, which at some point can no longer be financed.

So now we have six forms of crowding out:

1. Government competes with business for fixed saving.

2. Government competes with business for scarce liquidity.

3. Increased spending would lead to higher inflation.

4. Increased spending would cause the central bank to raise interest rates.

5. Overfull employment would lead to overfast wage increases.

6. Increased spending would lead to a higher trade deficit.

The next question is: Is there any reason, even in principle, to worry about any of these outcomes in the US today? We can decisively set aside the first, which is logically incoherent, and confidently set aside the second, which doesn’t fit a credit-money economy in which government liabilities are the most liquid asset. But the other four certainly could, in principle, reflect real limits on expansionary policy. The question is: In the US in 2017, are higher inflation, higher interest rates, higher wages or a weaker balance of payments position problems we need to worry about? Are they even problems at all?

First, higher inflation. This is the most natural place to look for the costs of demand pushing up against capacity limits. In some situations you’d want to ask how much inflation, exactly, would come from erring on the side of overexpansion, and how costly that higher inflation would be against the benefits of lower unemployment. But we don’t have to ask that question right now, because inflation is by conventional measures, too low; so higher inflation isn’t a cost of expansionary policy, but an additional benefit. The problem is even worse for Krugman, who has been calling for years now for a higher inflation target, usually 4 percent. You can’t support higher inflation without supporting the concrete action needed to bring it about, namely, a period of aggregate spending in excess of potential. [2] Now you might say that changing the inflation target is the responsibility of the Fed, not the fiscal authorities. But even leaving aside the question of democratic accountability, it’s hard to take this response seriously when we’ve spent the last eight years watching the Fed miss its existing target; setting a new higher target isn’t going to make a difference unless something else happens to raise demand. I just don’t see how you can write “What do we want? Four percent! When do we want it? Now!” and then turn around and object to expansionary fiscal policy on the grounds that it might be inflationary.

OK, but what if the Fed does raise rates in response to any increase in the federal budget deficit, as many observers expect? Again, if you think that more expansionary policy is otherwise desirable, it would seem that your problem here is with the Fed. But set that aside, and assume our choice is between a baseline 2018-2020, and an alternative with the same GDP but with higher budget deficits and higher interest rates. (This is the worst case for crowding out.) Which do we prefer? In the old days, the low-deficit, low-interest world would have been the only respectable choice: Private investment is obviously preferable to whatever government deficits might finance. (And to be fair, in the actual 2018-2020, they will mostly be financing high-end tax cuts.) But as Brad DeLong points out, the calculation is different today. Higher interest rates are now a blessing, not a curse, because they create more running room for the Fed to respond to a downturn. [3] In the second scenario, there will be some help from conventional monetary policy in the next recession, for whatever it’s worth; in the first scenario there will be no help at all. And one thing we’ve surely learned since 2008 is the costs of cyclical downturns are much larger than previously believed. So here again, what is traditionally considered a costs of pushing past supply constraints turns out on closer examination to be a benefit.

Third, the danger of more expansionary policy is that it will lead to a rise in the wage share. You don’t hear this one as much. I’ve suggested elsewhere that something like this may often motivate actual central bank decisions to tighten. Presumably it’s not what someone like Krugman is thinking about. But regardless of what’s in people’s heads, there’s a serious problem here for the crowding-out position. Let’s say that we believe, as both common sense and the textbooks tells us, that the rate of wage growth depends on the level of unemployment. Suppose we define full employment in the conventional way as the level of unemployment that leads to nominal wage growth just equal to productivity growth plus the central bank’s inflation target. Then by definition, any increase in the wage share requires a period of overfull employment — of unemployment below the full employment level. This holds even if you think the labor share in the long run is entirely technologically determined. A forteori it holds if you think that the wage share is in some sense political, the result of the balance of forces between labor and capital.

Again, I’m simply baffled how someone can believe at the same time that the rising share of capital in national income is a problem, and that there is no space for expansionary policy once full employment is reached. [4] Especially since the unemployment target is missed so often from the other side. If you have periods of excessively high unemployment but no periods of excessively low unemployment, you get a kind of ratchet effect where the labor share can only go down, never up. I think this sort of cognitive dissonance happens because economics training puts aggregate demand in one box and income distribution in another. But this sort of hermetic separation isn’t really sustainable. The wage share can only be higher in the long run if there is some short-run period in which it rises.

Finally, the external constraint. It is probably true that more expansionary fiscal policy will lead to bigger trade deficits. But this only counts as crowding out if those deficits are in some sense unsustainable. Is this the case for the US? There are a lot of complexities here but the key point is that almost all our foreign liabilities (and all of the government’s) are denominated in dollars, and almost all our imports are invoiced in dollars. Personally, I think the world is still more likely to encounter a scarcity of dollar liquidity than a surfeit, so the problem of an external constraint doesn’t even arise. But let’s say I’m wrong and we get the worst-case scenario where the world is no longer willing to hold more dollar liabilities. What happens? Well, the value of the dollar falls. At a stroke, US foreign liabilities decline relative to foreign assets (which are almost all denominated in their home currencies), improving the US net international investment position; and US exports get cheaper for the rest of the world, improving US competitiveness. The problem solves itself.

Imagine a corporation with no liabilities except its stock, and that also paid all its employers and supplies in its own stock and sold its goods for its own stock. How could this business go bankrupt? Any bad news would instantly mean its debts were reduced and its goods became cheaper relative to its competitors’. The US is in a similar position internationally. And if you think that over the medium term the US should be improving its trade balance then, again, this cost of over-expansionary policy looks like a benefit — by driving down the value of the dollar, “irresponsible” policy will set the stage for a more sustainable recovery. The funny thing is that in other contexts Krugman understands this perfectly.

So as far as I can tell, even if we accept that the US economy has reached potential output/full employment, none of the costs for crossing this line are really costs today. Perhaps I’m wrong, perhaps I’m missing something. but it really is incumbent on anyone who argues there’s no space for further expansionary policy to explain what concretely would be the results of overshooting.

In short: When we ask how close the economy is to potential output, full employment or supply constraints, this is not just a factual question. We have to think carefully about what these terms mean, and whether they have the significance we’re used to in today’s conditions. This post has been more about Krugman than I intended, or than he deserves. A very large swathe of established opinion shares the view that the economy is close to potential in some sense, and that this is a serious objection to any policy that raises demand. What I’d like to ask anyone who thinks this is: Do you think higher inflation, a higher “natural” interest rate, a higher wage share or a weaker dollar would be bad things right now? And if not, what exactly is the supply constraint you are worried about?

[1] The LM in ISLM stands for liquidity-money. It’s supposed to be the combination of interest rates and output levels at which the demand for liquidity is satisfied by a given stock of money.

[2] OK, some people might say the Fed could bring about higher inflation just by announcing a different target. But they’re not who I’m arguing with here.

[3] Krugman himself says he’d “be a lot more comfortable … if interest rates were well clear of the ZLB.” How is that supposed to happen unless something else pushes demand above the full employment level at current rates?

[4] It would of course be defensible to say that the downward redistribution from lower unemployment would be outweighed by the upward redistribution from the package of tax cuts and featherbedding that delivered it. But that’s different from saying that a more expansionary stance is wrong in principle.

Demand and Productivity

I’m picking up, after some months, the project I was working on over the summer on potential output. Obviously the political context is different now. But the questions of what potential output actually means, how tightly it binds, and how close the economy is to it at any given moment, are not going away. Previous entries: one, two, three, four, and five.

You’ve probably heard the story about Ed Rensi, the former McDonald’s CEO who claimed the company’s move to replace cashier’s with self-serve kiosks was a response to minimum wage increases.

“I told you so,” he writes. “In 2013, when the Fight for $15 was still in its growth stage, I and others warned that union demands for a much higher minimum wage would force businesses with small profit margins to replace full-service employees with costly investments in self-service alternatives.”

Is this for real? Maybe not: The shift toward kiosks has been happening for a while, so it’s not just a response to the recent minimum wage hikes; and it may not end up reducing labor costs anyway.

But let’s say the move is as as Rensi claims. Then we should call it what it is: an increase in labor productivity. With fewer workers McDonald’s will produce just as many hamburgers; in other words, production per worker will be higher. [1]

As I’ve suggested, this sort of thing is a real problem for a certain strand of minimum wage advocacy. Advocates like to point to productivity gains in response to higher wages as an argument in their favor. (The gains are usually imagined in terms of loyalty, motivation, lower turnover, etc. rather than machines, but functionally it’s the same.) But productivity gains can only reduce the job losses from a minimum wage increase if those losses are large; they are not consistent with a story in which employment stays the same. [2]

But at the macro level, this dynamic has different implications. If the McDonald’s case is typical — if higher labor costs regularly lead to higher productivity — then we need to rethink our idea of supply constraints. There is more space for expansionary policy than we usually think.

Let’s start at the beginning. Suppose there is some policy change, or some random event, that boosts desired spending in the economy. It could be more government spending, it could be lower interest rates, it could be a rise in exports. What happens then?

In the conventional story, higher spending normally leads to greater production of goods and services, which in turn requires higher employment. This leaves fewer people unemployed. Lower unemployment increases the bargaining power of workers, forcing employers to bid up nominal wages. [3] These higher wages are passed on to prices, leading to higher inflation. When inflation reaches whatever level is considered price stability, then we say the economy is at full employment, or at potential output. (In this story the two are equivalent.) If spending continues to rise past this point, the responsible authorities (normally the central bank) will intervene to bring it back down.

This is the story you’ll find in any good undergraduate macroeconomics textbook. It’s a reasonable story, as far as these things go. In the strong form it’s usually given in, it implies a hard limit to how much demand can increase before inflation starts rising unacceptably. Once the pool of unemployed workers falls to the “full employment” level, any further increase in employment will lead to rapid increases in money wages, which will be passed on one for one to inflation.

One place this chain can break is that new workers are not necessarily drawn from the ranks of the currently unemployed — that is, if the size of the laborforce is endogenous. Insofar as people counted as out of the laborforce are in fact available for employment (or net immigration responds to demand), an increase in output doesn’t have to reduce the ranks of the officially unemployed. In other words, the official unemployment rate may underestimate the space available for raising output via increased employment. This motivates the question of how much the the fall in laborforce participation since 2007 is due to demographics, and how much is due to weak demand.

The conventional story can also break down at two other places if productivity growth is endogenous. First, output can increase without a proportionate increase in employment. And second, wages can rise without a proportionate rise in prices.

It’s useful to think about this in terms of a couple of accounting identities, which in my opinion should be part of every macroeconomics textbook. [4] The first is obvious (but worth spelling out), the second a little less so:

(1) growth in demand = percent change in labor productivity + percent change in employment + inflation

(2) percent change in nominal wages = percent change in labor productivity + percent change in labor share + inflation

The standard story is that productivity change on its own due to technology, and the labor shared is fixed and can be ignored in this context. If productivity and labor share can be taken as given, then an increase in demand (money spent on final goods and services) must lead to higher inflation if either employment fails to rise, or if it rises only with higher wages. In this story, if nominal wages rise thanks to a lower unemployment rats, that will pass on one for one to inflation. Pick up an advanced undergraduate textbook like Blanchard or Krugman or Carlin and Soskice, and you will find a Phillips curve of exactly this form, with exactly this story behind it. [5] Policy discussions at central banks conducted in same terms.

This is what underlies idea of hard supply constraints. Output growth is dictated by the fixed, exogenous growth of the laborforce and of productivity. If changes in demand push the economy off that fixed trajectory, all you’ll get is higher or lower inflation. Concretely: To keep inflation at 2 percent, unemployment must be such as to generate nominal wage growth 2 points above the technologically-determined growth of productivity.

But an alternative story is that variation in demand can lead to adjustment in one of the other terms. One possibility is that the laborforce adjusts, as participation rates vary in response to demand conditions. This is what is most often meant by hysteresis: persistent deviations in unemployment from the “natural” level lead to people entering or exiting the laborforce. That implies that even when headline unemployment rates are fairly low, further increases in employment may be possible without a rise in wages. Another possibility is that while higher employment will lead to (or require) higher wages, the wage increase is not passed on to prices but comes at the expense of profits instead. This is Anwar Shaikh’s classical Phillips curve; I’ve written about it here before.

A third possibility is that higher wages are accompanied by higher productivity. Again, this appears as a problem when we are talking about wage increases from legislation, union contracts, or similar developments. But it’s not a problem if the wage increases are thanks to low unemployment. In this case, the joint movement of wages and productivity just means that output can rise higher — that supply constraints are softer. That’s what I want to focus on now.

There are a number of reasons why productivity might rise with wages. Some of them simply amount to mismeasurement of employment — it appears that output per worker is rising but really the effective number of workers is. Others are more fundamental. If productivity responds strongly and persistently to demand, it blurs the distinction between aggregate supply and aggregate demand, to the point that it’s not clear what “potential output” even means.

Suppose we do find a consistent pattern where, if demand is strong, unemployment is low, and wages are rising rapidly, then productivity growth is high. What could be happening?

1. Increased hours. If we measure productivity as output per worker, as we usually do, then an increase in average hours worked will show up as an increase in productivity. There is a cyclical component to this — in recessions, employers reduce hours as well as laying off workers. According to the BLS, seasonally adjusted weekly hours fell from 34.4 prior to the recession to a low of 33.7 in summer 2009. While a 2 percent fall in hours might seem small, it’s a big change in less than two years, especially when you consider that real output per worker normally rises by less than 2 percent a year.

2. Workers moving into real jobs from pseudo-employment or disguised unemployment. In any economy there are activities that are formally classified as jobs but are not employment in any substantive sense — you can take these “jobs” without anyone making a decision to hire you, and they don’t come with a wage or any similar claim on any established production process. Joan Robinson’s examples were someone who gathers firewood in a poor country, or sells pencils on streetcorners in a richer one. You could add work in family businesses and various kinds of self-employment and commission-based work to this category. In countries with traditional rural sectors — not the US — work on a family farm is the big item here. These activities absorb people who are unable to find formal jobs; the marginal product of additional workers here is normally very low. So if higher demand draws people from this kind of disguised unemployment back into regular jobs, measured productivity will rise.

3. Workers may be more fully utilized at their existing jobs. Because hiring and firing is costly, business don’t immediately adjust staffing in response to changes in sales. when demand falls, businesses will initially keep some redundant workers because paying them is cheaper than laying them off and replacing them later; and when demand rises, businesses will first try to get more work out of existing employees rather than paying the costs of hiring more. Some of this takes the form of the hours adjustment above, but some of it simply takes the form of hiring “too little” or “too much” labor for the current level of production. These changes in the utilization of existing labor will show up as changes in labor productivity.

4. Higher wages may lead to more capital-intensive production. This is the McDonald’s story: When labor gets more expensive (or scarcer), businesses use more capital instead. This is presumably what people mean when they say “Econ 101” shows that rising wages lead to less employment (assuming they mean anything at all). This may be seen as a negative when it’s a question of raising wages through legislation or unions, but it shouldn’t be when it’s a question of rising wages due to labor scarcity. Insofar as businesses can substitute machines for labor, rising wages will not be passed on to prices, so there is more space to push unemployment down.

5. Productivity-boosting innovations may be more likely when demand is strong and wages rise. This is a variant of the previous story. Now instead of high wage leading business to adopt more capital-intensive techniques from those already available, they redirect innovation toward developing new labor-saving techniques. Conceptually this is not a big difference, but it implies a different signal in the data. In the previous case we would expect the productivity improvements to be associated with higher investment and to be concentrated at the firms actually experiencing higher wages costs; in this case they might not be.

6. The composition of employment may shift toward higher-productivity sectors. This might happen for either of two reasons. First, higher wages will disproportionately raise costs for more labor-intensive sectors; these higher costs may be absorbed by profits or by prices, but either way they will presumably depress growth in those sectors to the benefit of less labor-intensive, more productive ones. Second, it may so happen that the more income-elastic sectors are also higher-productivity ones. In the short run this is presumably true since durables and investment goods are both capital-intensive and income-elastic. Over the longer run, the opposite is more likely — the composition of demand slowly but steadily shifts toward lower-productivity sectors.

7. The composition of employment may shift toward higher-productivity firms. This sounds similar but it’s a different story. Technical change isn’t an ineffable output-raising essence diffusing across society, it’s embodied in specific new production processes and new businesses — Schumpeter’s new plant, new firms, new men. This means that productivity increases often require new or growing firms to attract workers away from established ones. Given the “frictions” in the labor market, this will require offering a wage significantly above the going rate. And on the other side the fact that the least productive firms can’t afford to pay higher wages will cause them to decline or exit, which also raises average productivity. When wages are flat, on the other hand, low-productivity firms can continue operating. In this sense, higher wages are an integral part of productivity growth. [6]

8. There may be increasing returns in production. It may literally be the case that output per worker rises — at the firm, industry or economy-wide level — when the number of workers rises. Or this may be a more abstract version of some of the stories above. It’s worth noting that increasing returns is an area where the intuitions of people with economics training diverge sharply from people who look at the economy through other lenses. To almost anyone except an economist, it’s obvious that costs normally fall as more of something is produced. [7]

All of these stories imply that higher demand should lead to higher measured labor productivity. But to figure out how strong this relationship is in reality, we’ll look at different data depending on which of these stories we think it works through.

Another important difference between the stories is they imply different domains over which the relationship should operate. The first three suggest a more or less immediate response of productivity to changes in demand, but also one that cannot continue indefinitely. There’s limits to how much hours per worker can rise and how much additional effort can be extracted from the existing workforce, and a limited pool of disguised unemployment to draw from. (The last is not true in developing countries, where the “latent reserve army” in subsistence agriculture may be effectively unlimited.) The other mechanisms are presumably slower, requiring a sustained “high-pressure economy.” With these stories, increased demand may push the economy up against supply constraints, with rising inflation, bottlenecks, and so on; but if it keeps pushing against them, eventually they’ll give. In this case, potential output is a medium-term constraint — over longer periods it can adjust to actual output, rather than the reverse. So in the opposite of conventional story, a temporary increase in inflation can lead to a permanent increase in output. People like Laurence Ball say exactly this about hysteresis, but they are usually thinking of the longer-run adjustment coming on the laborforce side.

If we follow this a step further, we could even say that in the long run, the big problem isn’t that excessively high wages do lead to the substitution of capital for labor but that excessively low wages don’t. People like Arthur Lewis argue that it’s the low wages of poor countries that have led to low productivity there, and not vice versa; there’s a well-known argument that the reason the industrial revolution happened first in Britain rather than in China or India (or Italy or France) is not that that the necessary technical innovations were present only in Britain. They were present many places; it was the uniquely high cost of British labor that made them profitable to adopt for production.

I think that productivity does respond to demand. I think this is a good reason to doubt whether the US economy close to “potential output” today, and to doubt what, if anything, this concept actually means. But I also think we need to be clearer about how they are linked concretely. If we want to tell a story about productivity responding to demand, it makes a difference which of the stories above we have in mind. Heterodox people, it seems to me, are too quick to just invoke Verdoorn’s law (productivity rises with output), and justify it with some vague comments about how labor is used more efficiently when it is scarce. [8] Does this apparent law work via substitution of machines for labor, or through fuller utilization of existing employees’ times, or through reallocation of labor to more productive firms and/or industries, or through a labor-saving-bias in technical change, or pure increasing returns, or what? If you’re just making a formal model it may not matter. But if we want to connect the model to concrete historical developments, it certainly does.

Personally, I am most interested in the reallocation stories. They shift our idea of the fundamental constraint on capitalist economies from biophysical resources, to coordination. The great difficulty for any program of raise or transform production — industrialization, wartime mobilization, decarbonization — isn’t the limited supply of “real” resources, but the speed at which people’s productive activity can be redirected in a coordinated way. This connects with the historical fact that the more rapid and the larger scale is economic development, the more it requires some form of central planning. And it implies that at the most basic level, what the capitalist provides is not money or means of production, but cooperation.

To tell this story, it would be nice if big shifts in productivity growth took the form of changes in the composition of employment, rather than higher output per worker in given jobs. That may or may not be there in the data. For the more immediate question of how much space there is in the US for further expansion, it doesn’t matter as much which of these stories is at work, as long as we can show that at least some of them are. [9]

In the next post or two — which I hope to write in the next week, but we’ll see — I will ask what we can say about the link between demand and productivity based on historical US data. In particular, it’s fairly straightforward to decompose changes in output per worker into three components: within-industry output per hour, within-industry hours per worker, and shifts in the employment between industries. Splitting up productivity growth this way cannot, of course, directly establish a causal link with demand. but it can help clarify which stories are plausible and which are not.

[1] Throughout this discussion, I use “productivity” to mean labor productivity — output per worker or per hour. There is also “total factor productivity,” which purports to be a measure of output for a given input of labor and capital. This concept, which IMF chief economist Paul Romer memorably called “phlogiston,” is measured as the residual from a production function — the output growth the function does not explain. Since construction of the production function requires several unobseravable parameters, total factor productivity cannot be derived even in principle from economic data. It’s a fun toy for economic theory but useless for describing the behavior of actual economies.

Nonetheless it is widely used — for instance by the CBO as discussed here. As Nathan Tankus pointed out to me the other day, under the ARRA Medicare payments to hospitals are reduced each year based on an estimate of TFP growth for the economy as a whole. It’s a great example of the crackpot wonkery of the law’s authors.

[2] Unless productivity improvements all take the form of higher quality, rather than higher output per worker.

[3] This unemployment-money wages relationship was the original Phillips curve, but it’s better now to refer to it as a wage curve.

[4] It’s a topic for another time, but I think it would be very natural to replace the “aggregate supply” framework of the textbooks with these two identities.

[5] Other textbooks, like Mankiw, base the wage-unemployment relationship on a labor-supply curve rather than a bargaining relationship. Graduate textbooks, of course, replace the institutional detail of workers and employers with a single representative agent, in order to make more space for playing with math.

[6] Andrew Glyn and his coauthors have a good discussion of this in the context of the postwar boom in Capitalism Since 1945 (p. 122-123).

[7] For example, here’s Laurie Winkless in Science and the City, which happens to be sitting nearby:

Bessemer’s system rapidly began to change the world of steel manufacturing, and by 1875, costs had dropped to $32 (£23) per tonne. as always, in the supply-and-demand equation, the availability of cheap, high-quality steel made it immensely popular, leading to another huge drop in the price per tonne.

Winkless has made the mistake of studying the actual history of the steel history. If she were an economist, she would know that in the world of supply and demand, immense popularity makes prices rise, not fall!

[8] In Shaikh’s Capitalism, for example, there are a number of models that rely on the claim that productivity rises with output. It’s a big book and I may well have missed a part where he explains more fully why this is true. But as far as I can tell, all he says is that higher unit labor costs “provide a strong incentive for firms to raise productivity.”

[9] The politics of this question under Trump are for another time. But certainly Jeff Spross is right that we don’t want to oppose Trump’s (dubious) plans for a big stimulus by embracing the politics of austerity. We should not respond to Trump by reflexively insisting that the US is already at full employment, and by mocking “vulgar Keynesians” who think there might still be problems for macro policy to solve.

EDIT: Fixed the footnote numbering, which was garbled before.

Can We Blame Low Labor Participation on Past High Unemployment?

Fifth post in a series. Posts one, two, three and four.

We know that US GDP fell sharply in 2008-2009. We know that none of that decline has been made up by faster growth since the recession: GDP today is about 14 percent below the pre-2008 trend, a gap that shows no sign of closing. We also know that one-third of that shortfall is accounted for by slower productivity growth, and the remaining two-thirds by slower employment growth.

To put numbers on it: Over the past decade, US employment rose by a total of 6 percent, or about 0.5 percent per year. This is about half the rate of employment growth over the last ten years before the recession, and less a quarter the average rate for the postwar period as a whole. 2000-2010 was the first decade since the Depression in which US employment actually fell. Since the unemployment rate today is very close to that of ten years ago, this whole slowdown is accounted for by a decline in laborforce participation.

Employment growth, unlike productivity growth, was already slowing prior to the recession, and pre-recession forecasts predicted a further slowdown comparable to what actually occurred. This is consistent with a widely-held view that the slowdown in employment is the result of demographic and other structural factors, not of the recession or demand weakness in general. In the next couple posts, I want to take a critical look at this claim. How confident should we be that employment would be the same today in a counterfactual world where the 2008-2009 didn’t happen? How responsive might employment be to stronger demand going forward? And more broadly, how much do changes in laborforce participation seem to be explained by more or less exogenous factors like demographics, and how much by demand and labor-market conditions?

The rest of this post is about an approach to this question that did not produce the results I was hoping for. So I probably won’t include this material in whatever paper comes out of these posts. But as we feel our way into reality it’s important to note down the dead ends as well as the routes that seem promising. And even though this exercise didn’t help much in answering the big questions posed in the previous paragraph, it’s still interesting in its own right.

Can the fall in laborforce participation be explained as a direct, predictable effect of the rise in unemployment during the recession? It seems like maybe it can. The starting point is the observation that unemployed workers are much more likely to drop out of the laborforce than people with jobs are. You can see this clearly in the BLS tables on employment transitions. As the figure below shows, about 3 percent of employed people exit the laborforce each month, a fraction that has been remarkably stable since the data begins in 1990. Meanwhile, about 20 percent of unemployed people drop out of the laborforce each month.

On the face of it, this 17-point difference suggests an important role for the unemployment rate in changes in labor force participation. All else equal, each year-point of additional unemployment should reduce the labor-force participation rate by two points. (0.17 x 12 = 2.) So you would think that much of the recent fall in laborforce participation could be explained simply by the rise in unemployment during the recession.

When I thought of this it seemed very logical. It would be easy to do a counterfactual exercise, I thought, showing how laborforce participation would have evolved simply based on the historical transition rates between employment, unemployment and out of the laborforce, and the actual evolution of employment and unemployment. If you could show that something like the actual fall in laborforce participation was a predictable result of the rise in unemployment during the recession, that would support the idea that demand rather than “structural” factors are at work. And even if it wasn’t that strong positive evidence, it would suggest skepticism about similar counterfactual exercises using historical participation rates by age and so on.

I mean, it makes sense, right? Unemployed people are much more likely to leave the workforce than employed people, so a rise in unemployment should naturally lead to a decline in laborforce participation. But as the figure below shows, the numbers don’t work.

What I did was start with the populations of employe, unemployed and not-in-the-laborforce people at the end of the recession in December 2009. Then I created a counterfactual scenario for the remaining period using the actual transition rates between employment and unemployment but the pre-recession average rates for transitions between not in the workforce and unemployment and employment. In other words, just knowing the average rates that people move between employment, unemployment and out of the workforce, and the actual shifts between employment and unemployment, what path would you have predicted for laborforce participation over 2010-2016?

The heavy gray line shows the historical fraction of the population aged 16 and over who are not in the laborforce. The black line shows the results of the counterfactual exercise. Not very close.

There turn out to be two reasons why the counterfactual exercise gives such a poor fit. Both are interesting and neither was obvious before doing the exercise. The first reason is that there are surprisingly large flows from out of the labor force back into it. Per the BLS, about 7 percent of people who report being out of the labor force in a given month are either employed or unemployed (i.e. actively seeking work) the following month. This implies that the typical duration of being out of the workforce is less than a year — though of course this is a mix of people who leave the workforce for just a month or two and people who leave for good. For present purposes, the important thing is that exogenous changes to the employment-population ratio decline quickly, with a half-life of only about a year. So while the historical data suggests that a rise in unemployment like we saw in 2008-2009 should have been associated with a large rise in the share of the population not in the laborforce, it also suggests that this effect should have been transitory — a couple years after unemployment rates returned to normal, participation rates should have as well. This is not what we’ve seen.

The large gross movements in and out of the laborforce mean that sustained lower participation rates can’t be straightforwardly understood as the “echo” of high unemployment in the past. But they do also tend to undermine the structural story — if the typical stint outside the laborforce lasts less than a year it can hardly be due to something immutable.

The second reason why the counterfactual doesn’t fit the data was even more surprising, at least to me. I constructed my series using the historical average transition rates into and out of the workforce. But transition rates during the recession and early recovery departed from the historical average in an important way: unemployed workers were significantly less likely to exit the workforce. This turns out to be the normal pattern, at least over the previous two business cycles — if you look back to that first figure, you can see dips in the transition rate from unemployed to out of the workforce in the early 1990s and early 2000s downturns as well. The relationship is clearer in the next figure, a scatter of the unemployment rate and the share of unemployed workers leaving the workforce each month.

As you can see, there is a strong negative relationship — when unemployment was around 4 percent in 1999-2000 and again in 2006-2007, about a quarter of the unemployed exited the laborforce each month. But at the peak of the past recession when unemployment reached 10 percent, only 18 percent of the unemployed left the laborforce each month. That might not seem like a huge difference, but it’s enough to produce quite different dynamics. It’s also a bit surprising, since you would think that people would be more likely to give up searching for work when unemployment is high than when when it is low. The obvious explanation would be that the people who are out of work when the unemployment rate is low are not simply a smaller set of the same people who are out of work when the rate is high, but are different in some way. The same factors that keep them at the back of the hiring queue may make also be likely to push them out of the laborforce altogether. Extended unemployment insurance might also play a role.

It would be possible to explore this further using CPS data, which is the source for the BLS tables I’m working with. No doubt there are papers out there describing the different characteristics of the unemployed in periods of high versus low unemployment. (Not being a labor economist, I don’t know this literature.) But I am going to leave it here.

Summary: The fact that unemployed people are much more likely to leave the laborforce than employed people are, suggests that some part of the fall in laborforce particiaption since 2008 might be explained by the lingering effects of high unemployment in the recession and early recovery. But this story turns out not tow work, for two reasons. First, the rapid turnover of the not in the laborforce population means that this direct effect of high unemployment on participation is fairly shortlived. Second, the rate at which unemployed people exit the laborforce turns out to be lower when unemployment is high. Together, these two factors produce the results shown in the second figure — the fall in participation you would predict based simply on high unemployment is steeper but shorter-lived than what actually occurred. The first factor — the large flows in and out of the laborforce — while it vitiates the simple story I proposed here, is consistent with a broader focus on demand rather than demographics as an explanation for slow employment growth. If people are frequently moving in and out of the laborforce, it’s likely that their decisions are influenced by their employment prospects, and it means they’re not determined by fixed characteristics like age. The second factor — that unemployed people were less likely to give up looking for jobs during 2009-2011, as in previous periods of high unemployment — is, to me, more surprising, and harder to fit into a demand-side story.

Employment, Productivity and the Business Cycle

Fourth post in a series. Posts one, two and three.

Empirically-oriented macroeconomists have recognized since the early 20th century that output, employment and productivity move together over the business cycle. The fact that productivity falls during recessions means that employment varies less over the cycle than output does. This behavior is quite stable over time, giving rise to Okun’s law. In the US, Okun’s law says that the unemployment rate will rise by one point in each 2.5 point shortfall of GDP growth over trend — a ratio that doesn’t seem to have declined much since Arthur Okun first described it in the mid-1960s. [1]

It’s not obvious that potential should show this procyclical behavior. As I noted in the previous post, a naive prediction from a production function framework would be that a negative demand shock should reduce employment more than output, since business can lay off workers immediately but can’t reduce their capital stock right away. In other words, productivity should rise in recessions, since the labor of each still-employed worker is being combined with more capital.

There are various explanations for why labor productivity behaves procyclically instead. The most common focus on the transition costs of changing employment. Since hiring and firing is costly for businesses, they don’t adjust their laborforce to every change in demand. So when sales fall in recessions, they will keep extra workers on payroll — paying them now is cheaper than hiring them back later. Similarly, when sales rise businesses will initially try to get more work out of their existing employees. This shows up as rising labor productivity, and as the repeated phenomenon of “jobless recoveries.”

Understood in these terms, the positive relationship between output, employment and productivity should be a strictly short-term phenomenon. If a change in demand (or in other constraints on output) is sustained, we’d expect labor to fully adjust to the new level of production sooner or later. So over horizons of more than a year or two, we’d expect output and employment to change in proportion. If there are other limits on production (such as non-produced inputs like land) we’d expect output and labor productivity to move inversely, with faster productivity growth associated with slower employment growth or vice versa. (This is the logic of “robots taking the jobs.”) A short-term positive, medium term negative, long-term flat or negative relationship between employment growth and productivity growth is one of the main predictions that comes out of a production function. But it doesn’t require one. You can get there lots of other ways too.

And in fact, it is what we see.

The figure shows the simple correlation of employment growth and productivity growth over various periods, from one quarter out to 50 quarters. (This is based on postwar US data.) As you can see, over periods of a year or less, the correlation is (weakly) positive. Six-month periods in which employment growth was unusually weak are somewhat more likely to have seen weak productivity growth as well. This is the cyclical effect presumably due to transition costs — employers don’t always hire or fire in response to short-run changes in demand, allowing productivity to vary instead. But if sales remain high or low for an extended period, employers will eventually bring their laborforce into line, eliminating this relationship. And over longer periods, autonomous variation in productivity and labor supply are more important. Both of these tend to produce a negative relationship between employment and productivity. And that’s exactly what we see — a ten-year period in which productivity grew unusually quickly is likely to be one in which employment grew slowly. (Admittedly postwar US data doesn’t give you that many ten-year periods to look at.)

Another way of doing this is to plot an “Okun coefficient” for each horizon. Here we are looking at the relationship between changes in employment and output. Okun’s law is usually expressed in terms of the relatiojship between unemployment and output, but here we will look at it in terms of employment instead. We write

(1) %ΔE = a (g – c)

where %ΔE is the percentage change in employment, g is the percentage growth in GDP, c is a constant (the long-run average rate of productivity growth) and a is the Okun coefficient. The value of a says how much additional growth in employment we’d expect from a one percentage-point increase in GDP growth over the given period. When the equation is estimated in terms of unemployment and the period is one, year, a is generally on the order of 0.4 or so, meaning that to reduce unemployment by one point over a year normally requires GDP growth around 2.5 points above trend. We’d expect the coefficient for employment to be greater, but over short periods at least it should still be less than one.

Here is what we see if the estimate the equation for changes in output and employment for various periods, again ranging from one quarter up to ten years. (Again, postwar US data. The circles are the point estimates of the coefficients; the dotted lines are two standard errors above and below, corresponding to a standard 95% confidence interval.)

What’s this show? If we estimate Equation (1) looking at changes over one quarter, we find that one percentage point of additional GDP growth is associated with just half a point of additional employment growth. But if we estimate the same equation looking at changes over two years, we find that one point of additional GDP growth is associated with 0.75 points of additional employment growth.

The fact that the coefficient is smallest for the shorter periods is, again, consistent witht he conventional understanding of Okun’s law. Because hiring and firing is costly, employers don’t fully adjust staffing unless a change in sales is sustained for a while. If you were thinking in terms of a production function, the peak around 2 years represents a “medium-term” position where labor has adjusted to a change in demand but the capital stock has not.

While it’s not really relevant for current purposes, it’s interesting that at every horizon the coefficient is significantly below zero. What this tells us is that there is no actual time interval corresponding to the “long run” of the model– a period long enough for labor and the capital stock to be fully adjusted but short enough that technology is fixed. Over this hypothetical long run, the coefficient would be one. One way to think about the fact that the estimated coefficients are always smaller, is that any period long enough for labor to adjust, is already long enough to see noticeable autonomous changes in productivity. [2]

But what we’re interested in right now is not this normal pattern. We’re interested in how dramatically the post-2008 period has departed from it. The past eight years have seen close to the slowest employment growth of the postwar period and close to the slowest productivity growth. It is normal for employment and productivity to move together for a couple quarters or a year, but very unusual for this joint movement to be sustained over nearly a decade. In the postwar US, at least, periods of slow employment growth are much more often periods of rapid productivity growth, and conversely. Here’s a regression similar to the Okun one, but this time relating productivity growth to employment growth, and using only data through 2008.

prod on emp While the significance lines can’t be taken literally given that these are overlapping periods, the figure makes clear that between 1947 and 2008, there were very few sustained periods in which both employment and productivity growth made large departures from trend in the same direction.

Put it another way: The past decade has seen exceptionally slow growth in employment — about 5 percent over the full period. If you looked at the US postwar data, you would predict with a fair degree of confidence that a period of such slow employment growth would see above-average productivity growth. But in fact, the past decade has also seen very low productivity growth. The relation between the two variables has been much closer to what we would predict by extrapolating their relationships over periods of a year. In that sense, the current slowdown resembles an extended recession more than it does previous periods of slower growth.

As I suggested in an earlier post, I think this is a bigger analytic problem than it might seem at first glance.

In the conventional story, productivity is supposed to be driven by technology, so a slowdown in productivity growth reflects a decline in innovation and so on. Employment is driven by demographics, so slower employment growth reflects aging and small families. Both of these developments are negative shifts in aggregate supply. So they should be inflationary — if the economy’s productive potential declines then the same growth in demand will instead lead to higher prices. To maintain stable prices in the face of these two negative supply shocks, a central bank would have to raise interest rates in order to reduce aggregate spending to the new, lower level of potential output. Is this what we have seen? No, of course not. We have seen declining inflation even as interest rates are at historically low levels. So even if you explain slower productivity growth by technology and explain slower employment growth by demographics, you still need to postulate some large additional negative shift in demand. This is DeLong and Summers’ “elementary signal identification point.”

Given that we are postulating a large, sustained fall in demand in any case, it would be more parsimonious if the demand shortfall also explained the slowdown in employment and productivity growth. I think there are good reasons to believe this is the case. Those will be the subject of the remaining posts in this series.

In the meantime, let’s pull together the historical evidence on output, employment and productivity growth in one last figure. Here, the horizontal axis is the ten-year percentage change in employment, while the vertical axis is the ten-year percentage change in productivity. The years are final year of the comparison. (In order to include the most recent data, we are comparing first quarters to first quarters.) The color of the text shows average inflation over the ten year period, with yellow highest and blue lowest. The diagonal line corresponds to the average real growth rate of GDP over the full period.

What we’re looking at here is the percentage change in productivity, employment and prices over every ten-year period from 1947-1957 through 2006-2016. So for instance, growth between 1990 and 2000 is represented by the point labeled “2000.” During this decade, total employment rose by about 20 percent while productivity rose by a total of 15 percent, implying an annual real growth of 3.3 percent, very close to the long-run average.

One natural way to think about this is that yellow points below and to the right of the line suggest negative supply shocks: If the productive capacity of the economy declines for some reason, output growth will slow, and prices will rise as private actors — abetted by a slow-to-react central bank — attempt to increase spending at the usual rate. Similarly, blue points above the line suggest positive supply shocks. Yellow points above the line suggest positive demand shocks — an increase in spending can increase output growth above trend, at least for a while, but will pull up prices as well. And blue points below the line suggest negative demand shocks. This, again, is Delong and Summers’ “elementary signal identification point.”

We immediately see what an outlier the recent period is. Both employment and productivity growth over the past ten years have been drastically slower than over the preceding decade — about 5 percent each, down from about 20 percent. 2000-2010 and 2001-2011 were the only ten-year periods in postwar US history when total employment actually declined. The abruptness of the deceleration on both dimensions is a challenge for views that slower growth is the result of deep structural forces. And the combination of the slowdown in output growth with falling prices — especially given ultra-low interest rats — strongly suggests that we’ve seen a negative shift in desired spending (demand) rather than in the economy’s productive capacities (supply).

Another way of looking at this is as three different regimes. In the middle is what we might call “the main sequence” — here there is steady growth in demand, met by varying mixes of employment and productivity growth. On the upper right is what gets called a “high-pressure economy,” in which low unemployment and strong demand draw more people into employment and facilitates the reallocation of labor and other resources toward more productive activity, but put upward pressure on prices. On the lower left is stagnation, where weak demand discourages participation in the labor force and reduces productivity growth by holding back investment, new business formation and by leaving a larger number of those with jobs underemployed, and persistent slack leads to downward pressure on prices (though so far not outright deflation). In other words, macroeconomically speaking the past decade has been a sort of anti-1960s.

[1] There are actually two versions of Okun’s law, one relating the change in the unemployment rate to GDP growth and one relating the level of unemployment to the deviation of GDP from potential. The two forms will be equivalent if potential grows at a constant rate.

[2] The assumption that variables can be partitioned into “fast” and “slow” ones, so that we can calculate equilibrium values of the former with the latter treated as exogenous, is a very widespread feature of economic modeling, heterodox as much as mainstream. I think it needs to be looked at more critically. One alternative is dynamic models where we focus on the system’s evolution over time rather than equilibrium conditions. This is, I suppose, the kind of “theory” implied by VAR-type forecasting models, but it’s rare to see it developed explicitly. There are people who talk about a system dynamics approach, which seems promising, but I don’t know much about them.