There is very little humor in the news lately, but every so often an economist comes along who cannot help but evoke peels of laughter because he has no idea that what he is saying is funny.
Here’s a recent example: Commenting on a study that connects manufacturing job losses in the United States due to free trade with increasingly immoderate politics, one of the authors, Gordon Hanson of UC-San Diego, said: “There’s a deeper appreciation for the magnitude of the impact on workers who lose their jobs. But the nature of globalization changed after the end of the Cold War and it took a while for academics to catch up.” Took “a while”…really? Academics, of all people, behind the curve of onrushing reality? It took me several minutes of full-bodied convulsions to catch my breath after reading that.
Everyone understands that the American economic challenge going forward is to generate enough middle-class sustaining jobs to overcome the twin challenges of globalization’s seemingly endless supply of low-wage labor and, probably even more significant at explaining the variance, high-capital-intensity, robotics-infused productivity increases. It’s a real problem, and neither high tariffs nor greater trade union power nor trade restrictions will revive the U.S. manufacturing sector sufficiently to turn the labor market trick that politicians are now promising. So what’s the answer?
Most economists and economic journalists have been taught to recite “service sector,” and so they do. Listen to the NYT’s excellent business journalist Eduardo Porter: “Promises to recapture industrial era greatness ring hollow. The United States, though, does have other options: health care, education and clean energy, just to name a few.” So we’re to become a continental-scale Downton Abbey, or a chapter in Neal Stephenson’s The Diamond Age, with a few very wealthy lords and ladies being served, in essence, by masses of perky personalities with little prospect of their own social mobility in sight? And this arrangement can realistically be compared to the egalitarian-minded (never mind the reality) democratic experience of American history? That’s funny, huh?
The argument that Americans can prosper as a nation by turning to service-related jobs en masse is one that only an economist who lacks a social filter for his data can make. Unfortunately, that seems to be most of them, with TAI’s editorial board member Tyler Cowen being a rare exception—which is why his masterful 2011 essay on the bugaboo of inequality remains the best thing written on the subject to date.
Here’s a recent remark that illustrates the point of flinging about numbers without much clue or care about the real human context. Gary Clyde Hufbauer of the Peterson Institute recently claimed that the benefits of free trade are “ten times the size of the losses” and that free trade helps poorer people because it creates lower prices for products. “The benefits,” he said, “are skewed toward people with lower incomes because they spend a much larger fraction of their income on merchandise.”
In other words, a lot of Americans may lack work that provides financial stability, self-esteem, and dignity for their families, and in all too many cases they have become psychologically déclassé based on previous experience and extrapolated expectations. So they’re angry, but according to Hufbauer—and he’s hardly alone—they shouldn’t be because they can buy heaps of inexpensive plastic junk they either don’t need or can’t keep working from China and other low-wage manufacturing countries. I would love to see Hufbauer, unarmed, say that before a seething mass of Donald Trump supporters—there could be some real Kickboxing Federation-style entertainment value in that.
I stand second to none in my loathing of Trump, but many of his supporters have a real beef, and it’s academics and policy wonks like Hufbauer (but, even more, former policymakers like Robert Rubin and Larry Summers, back in the day) who raised the steer. Sure, the macroeconomic numbers look good, but the benefits on the U.S. side have been highly uneven in their distribution. Those who have been waiting for the trickle down are mostly still waiting, because wealthy corporations and individuals today have vastly more opportunities than before to move their money around in ways that do not align with historical definitions of re-investment in the home economy.
This just inheres in the nature of the subject matter. When you look at defense budget numbers, for example, it’s not outrageous to think that every American benefits from a roughly even share of that investment (assuming for the moment that most of that money really does go for genuine defense and security). But when it comes to macroeconomic and trade data that is not at all the case—never has been, never will be. How the benefits and liabilities play out depend on where someone is located in the import/export nexus that partly defines the economy, and hence on whose ox gets gored in the shuffling about of capital investment and consequent labor market perturbations.
So much, then, for all the earnest and very unfunny Republican China hands out there who claim, without even thinking through the mantra, that the U.S.-China economic relationship over these past twenty or so years has benefitted both parties. It’s certainly been a major cause of Chinese economic expansion and, with it, an indirect source of a burgeoning Chinese military expansion; and of course a lot of American corporations (and the banks that finance their activities) have done very well. But more to the point, it’s also obviously been a major cause of the derangement of the American labor profile that has now yielded political consequences in the form of the populist explosion of the current campaign season. Angry American populism tends to be either highly nationalist or isolationist in foreign policy orientation, oscillating Jacksonian-like between the two. So if in due course the main historical consequence of the Sino-American economic relationship ends up being a major war, which is not as far-fetched a prospect as I wish it were, would these people still want to be making this macro-data-based argument? It would be funny if they did.
Now, as I said already, the labor market shifts caused by allowing capital to flow as freely as goods and services has probably had a smaller impact than the shift to more capital-intensive technologies and techniques. (It’s also the case that a lot of capital flows into the United States as well as out, so it’s not a simple matter to judge how freeing up capital flows plays through the U.S. economy long term.) That’s not my independently formed opinion, because I’m not competent enough in the subject to have one, but pretty much every serious analyst believes this to be the case. But if that’s true, then why has recent economic data seemed to suggest the reverse? Two stories appeared on the same set of data in the April 27 New York Times, one beginning on the front page and one on the front of the business section. So what did they say? Employment and wages were up pretty strongly, but productivity increases were miniscule to negligible. This, both articles and the economists quoted therein said, was a “mystery” or a “puzzle.”
Neither story discussed a third dataset: corporate profits. These, on balance, have not been unhealthy. But why, the economists reason, would companies be hiring and paying high wages to attract the talent they need in the absence of strong labor productivity increases? Maybe the companies are making enough money to justify the hiring without fully knowing why they’re doing so well. The economists apparently believe Robert Gordon’s thesis of a technology plateau, which, if true, means that the employment good news can’t persist unless the productivity index gets healthier.
There is, however, an alternative explanation that supports the CEOs against the wonks. Back in 1991 Paul A. David of Stanford University studied the lag time between the introduction of commercial electricity and productivity increases based on it. He concluded two things: first, that it took an entire generation for the legacy concepts and investments of industry to adapt the new technology—or, in other words, the conservative character of human institutions retarded adaptation for social, not technical, reasons; and second that, because of concomitant shifts in what productivity meant to people as consumption patterns and expectations changed, the measurement of productivity took almost as long to adjust as the communities that were adopting the new methods. As it happened, economists back in the 1920s also denigrated the economic implications of electricity for industrial production. By the 1950s they looked in retrospect like hidebound morons. Oh dear reader, are you beginning to feel the breeze I am wafting your way?
How exactly do economists measure productivity increases today? You don’t know, do you? Neither do I. This is an esoteric subject that only a small number of people really understand. That small number does not include the majority of economists who use these numbers without ever really interrogating the methods that produce them. What if it turned out that we are still measuring productivity changes with mid-industrial age metrics, even though, again, what people today understand as economic value differs significantly from what it was forty or fifty years ago?
This stuff changes, as Balzac realized 175 years ago when he observed that discretionary income had migrated during his lifetime from people buying capital goods (tools, looms, and so on) to people buying status items (better furniture, fancy clocks, and designer door knockers) to people buying experiences (theater tickets and vacations). And it’s no unappreciated coincidence, for my purposes, that Balzac made this observation in his justly famous La Comédie humaine.
This stuff not only changes; it is very hard to measure, and it’s getting harder every day as concepts of economic value relatively dematerialize. Example: As everyone knows, and as we’ve discussed in the inner sancta of TAI’s extremely swank downtown Washington digs, a person with contemporary information-digital technology can watch a huge number of movies on demand at almost zero cost compared to the time when the Blockbuster video rental franchise was a cool new idea, much less the days before television. Are these “productivity”-relevant changes, even though they’re about consuming experiences rather than consuming stuff?
How about the use of the little hand-held computers we still call cell phones, although that’s the least of what they do nowadays? Nothing that we do regularly with these devices—get accurate information in a very short time (if you know how to parse the divide between accurate and the addled); coordinate business and social activity in a far more efficient way (assuming that time has value); check on the health and safety of the elderly, the ill, the handicapped, and assorted risk-prone teenagers, and the peace of mind that ability can bring; and many dozens of other non-trivial uses—actually produces or depends on anything physical. So are these benefits and efficiencies consonant with a post-industrial concept of productivity or not?
No one knows. No one understands how to measure these things. Hardly anyone even thinks about what is actually required to measure them, which suggests the need for economists (and others) to refresh their memories of Simon Kuznets and his work. And what also bears mention is how relatively fast, by any reasonable historical standard, these changes are occurring. The question we ought to be asking is not whether our metrics are still relevant but, in light of the historical experience we know, how could they possibly still be relevant in the face of such changes.
Everyone savvies the old Maslow witticism that if all you have is a hammer every problem begins to look like a nail, and everyone remembers the old saw about looking for a lost item under a lamppost at night because that’s where the light is. We chuckle at these observations. But isn’t it just as amusing that we still count what we already know how to count, and what is relatively easy to count, and imagine that it’s still relevant to a changed reality? It rarely occurs to economists and those who try to pay attention to them that we might have no very good idea how to measure productivity in current technological-social circumstances.
The real mystery, then, is not about any mismatch between our productivity metrics and the employment and wage data; it’s about how seemingly intelligent people can be so dense, again, looking for knowledge in only the familiar places where it likely no longer resides.
You don’t think this is funny? Come on, lighten up a little.