On Tyler Cowen’s “Great Stagnation”
I’m late to this party, but I’m late to every party. Tyler Cowen’s recent microepic, The Great Stagnation, has been pretty throughly chewed over by the blogosphere. The essay is a quick read, thought-provoking, and getting to give Cowen a couple of bucks for the privilege only adds to the pleasure. The quick summary, for those who’ve been living in a cave, is that since 1973, we’ve been living in a “great stagnation”, during which the pace of growth experienced by the median American household slowed relative to expectations set in the period preceding. Cowen suggests that the explanation for this is a slowdown in the rate of technological change combined with an exhaustion of “low hanging fruit” afforded us by earlier advantages and innovations. Cowen simultaneously disputes both conventional measures of economic performance (we do not observe a “great stagnation” in headline GDP) and left-ish arguments that economic unhappiness stems primarily from maldistribution. We suffer instead from an absence of anticipated wealth. As Cowen puts it, we’re simply “poorer than we thought”.
I think the deepest issue Cowen brings out has less to do with technology than with problems in what economists measure (and people perceive) as “revenue” or “production”. In particular Cowen makes two related points:
Many activities that generate apparent revenue are detached from reliable judgments of value. Using revenue as a measure of production requires, at a minimum, that discriminating, budget-constrained actors determine that whatever is “paid-for” offers real-economic value superior or at least comparable to activities that could be inspired by alternative expenditures of the funds.
Cowen is appropriately general with this critique. Expenditures by government may not meet this “market test” because political actors may direct expenditures for reasons other than inspiring high-value economic activity, or because, for informational or organizational reasons, government may be unable to discern relative value. But the private sector is not immune. In spheres such as health care and education, the benefits of private sector as well as public sector expenditures are difficult to evaluate relative to alternative uses of resources. Cowen reminds us that health care and education are widely viewed as “growth” sectors, but to the degree we collectively overpay for them, “revenue” overstates economic value. A substantial portion of these expenditures should probably be accounted for as transfers and excluded from measures of aggregate production. But of course, we have no means of estimating the size of the appropriate haircut.
I’d add another important industry to government, health care, and education: financial services. Like with health care and education, we simply are unable to evaluate the degree to which payments to financial service providers represent wise use of resources and to what degree they represent transfers to financial industry stakeholders. Inherent informational problems associated with investment quality, combined with the temptation by service providers to exploit these difficulties to extract transfers, render financial sector revenue highly suspect as a marker of value. Also, financial services are intimately involved in the other problematic sectors: One thing that binds government, health-care, and education is that all are financed in roundabout and sometimes opaque ways that soften near-term budget constraints and that shift costs and risks, both across time and onto people other than the purchasing decisionmaker. The means by which government, health-care, and education are financed help keep them vulnerable to agency and information problems.
I think of government, education, health care, and finance collectively as the “information asymmetry industry”, and I find it terrifying that many people presume that they are the future growth industries for the United States. Dani Rodrik has pointed out that tradable goods are special, in terms of engendering development in often corrupt emerging markets. Cowen offers an astute explantion: tradables that compete in international markets are usually low-information-asymmetry goods. Apparent value (revenue from trade) and real value are likely to be closely aligned and hard to fake. I worry that specialization in the information asymmetry industry could be an antidevelopment strategy for developed countries.
Conversely, Cowen points out that many new technologies generate value without generating commensurate revenue. It is clear that we would collectively pay a lot more for recorded music or news, for example, because we did in the past (and it’s likely that our reduced payments have more to do with technology and industry changes than with changes in our preferences). Cowen suggests that many of the current era’s technologies are like this, which is nice from a certain perspective (yay! free stuff!) but can cause a kind of sclerosis in an economy that nourishes itself via flows of monetary exchange.
An interesting corollary of Cowen’s argument is that a substantial fraction of notional savings are in fact empty claims — no meaningful real economic activity accompanied the generation of that income, so no real investment could have attended its non-consumption. This suggests that any attempt to mobilize savings in aggregate — any net dissaving — is likely to result in either inflation or displacement of consumption by current earners. I think that this is true, that it is now and increasingly will be a source of social and political problems.
Despite the evidence that Cowen is able to muster (and lively internet debates about kitchen appliances and time travel), I remain agnostic on the question of whether a slowdown in the pace of technological change is largely to blame for whatever it is that ails us. I wouldn’t rule it out, but like Kevin Drum and others, I see a lot of very real “low hanging fruit” arising enhanced capabilities to coordinate and collaborate via internet and information technologies. These benefits go well beyond “cognitive surplus” and bemused infovoria.  Coordination technologies, ranging from assembly lines to the limited-liability joint-stock corporation, contributed mightily to past golden ages of advancement and growth.
Cowen is clearly right that, with the exception of the internet, recent years have offered few technologies that radically and visibly altered how people live. But I think we need to think carefully about an “extensive” and “intensive” margin for technological change. The early industrial revolution largely streamlined manufacture of existing categories of goods (think textile factories rather than spinning wheels and molded products replacing hand-smithed metal). Producers saw their world turned upside-down, but consumers mostly found themselves with recognizable stuff, just more and cheaper. Still, one wouldn’t characterize it as a low-growth era.
Cowen seems really focused on the “extensive margin” — the development of qualitatively new goods. I don’t think “growth”, in aggregate or as experienced by the median family, really captures what Cowen thinks we are missing. Instead, I think that Cowen is lamenting a scarcity of breathtaking resets. Developments like electrification and the widespread adoption of automobiles didn’t make people richer as much as they completely changed the circumstances of everyday life. Indirectly, they also made the production and marketing of previously extant goods much more efficient: Electricity helped make bread cheaper. But I don’t think cheap bread impresses Cowen as much as the fact that, post-electricity, humans colonized the night and Presidents colonized living rooms. Income statistics do end up capturing these sorts of changes, but in a manner that is arbitrary and formal. Ultimately, different technological regimes are incommensurable in welfare terms. While most of us would choose more food over inadequate nutrition and better health over worse, adoption of new technology is less a matter of individual choice than social evolution. One must adopt an automobile-centric lifestyle if workplaces move far from available housing. One must become internet proficient if current modes of employment are contingent upon it. The fact that the gas we are forced to purchase is factored into computations of real income and GDP doesn’t really imply that our preferences have been satisfied more completely than they would have been in a different-technology alternative.  It might be the case that the big changes whose absence troubles Cowen are welfare-destructive in and of themselves, but occurred only because they came bundled with large improvements in the productivity of prior goods, or simply because new possibilities rendered unstable earlier equilibria that were superior outright. If this is the case, we should celebrate rather than fret if recent modes of innovation have succeeded at increasing productivity without reseting the technological terms of our existence.  Ultimately, society is a game and big technological changes rescramble both the available strategies and payouts, leading to changes that are unpredictable both in form and in terms of welfare (under almost any welfare criterion that you might choose).
Suppose that it is true that we’re poorer than we’d anticipated, that past growth trends have eased. It’s not at all clear that what people conventionally think of as technology is the growth-limiting factor. Cowen looks to emerging markets for examples of how the availability of low-hanging fruit — technology and institutions already prevalent in developed economies — can hypercharge growth. But the less-developed world offers a different set of examples as well. There are many many economies where despite free and full availability of scientific information and plenty of institutions to emulate, low-hanging fruit is left to wither on the vine. We have (usually cartoonish and patronizing) explanations for other economies’ failures — “they” are corrupt and their cronies keep them down; they were scarred because of colonization by Belgians rather than Brits; (in whispers) they are culturally or even genetically inadequate to the task of development; they simply fail to make good choices. Whatever your just-so story, it’s pretty clear that from the inside, intelligent people struggle unsuccessfully to find means to overcome barriers that prevent them from picking delicacies that are hanging in front of their noses. We might be in a similar situation, with plenty of technological fruit ripe for the picking, but invisible barriers — political, cultural, whatever — that prevent us from doing so.
A few nights ago, a gentleman accosted me in a dream and declared himself to be “Tyrone”, Tyler Cowen’s evil twin. Tyrone told me that his brother had “as usual” got it all backwards. In fact, he told me, we’ve been in the Great Stagnation for a century as a result of, rather than for the lack of, technological progress. The median household is experiencing wealth stagnation caused by technological change. Households are feeling the pain now more than in the past, even despite a relatively modest pace of change, because over the past few decades we have managed to avoid employing the sort of durable and effective countermeasures to stagnation that have succeeded in the past.
For the most part, Tyrone pointed out, technological progress is labor displacing. It simultaneously creates valuable new techniques for reconfiguring real resources while diminishing the number of people who are required to participate in those transformations, and who can therefore trade their participation for spending power. There is a myth among neoliberal economists that labor markets have always “adjusted” sua sponte: that when laborers were displaced from farms, “higher value” factories arose to employ them; that when the factories were downsized and offshored, a more pleasant, higher-value service economy came to be; etc. That narrative is wrong, he told me. At best it is criminally incomplete. With each technological change, new social institutions had to arise to sustain dispersed purchasing power despite a reduction of numbers and bargaining power of workers in old industries. Displaced workers ultimately did find new work, but only because the new social institutions “artificially” created buyers for all the things displaced workers reinvented themselves to sell. Without this institutional innovation, Tyrone tells me, something like the Great Depression would have been the new normal. Historically, institutions that have arisen to sustain purchasing power despite increasingly labor-efficient core production include direct government transfers and expenditures, labor unions, monetary policy interventions, financial bubbles and financial fraud.
Cowen (Tyler, that is) argues that technological development creates opportunities for “bigger”, more extensive and intrusive government than existed during earlier periods. As Tyler tells the story, there is a progressive expansionary impulse to government, for which technological change creates opportunities, so government expands until those opportunities are fully exploited. Tyrone says his brother has the story backwards. Why, asks Tyrone, does government not only expand in absolute terms as a response to technological change, but also in relative terms? After all, as Tyler points out, private enterprise also has a natural expansionary impulse. With technological change, Tyler writes, “Everything was growing larger.” Yet, to the degree that we can measure it, government has grown dramatically in its share of the overall economy. Why does government win? Tyrone says government is a reluctant adopter of new technology (“Have you been to a government office?”), but that government outgrows the private sector despite this, because the concentration of economic power that attends technological changes demands countervailing state action if any semblance of broad-based affluence and democratic government is to be sustained. Tyrone (who is much more arrogant and less pleasant than his brother) proclaims this to be his “
ironsilicon law”: In (non-terminal) democratic societies, technological change must always and everywhere be accompanied by the growth of institutions that engender economic transfers from the relatively few who remain attached to older productive enterprises to the many who require purchasing power not only to live as they did before, but also to employ one another in novel or more marginal activities that were not pursued before. Inevitably those institutions develop in state or quasi-state sectors (which include the state-guaranteed financial sector and labor unions whose “collective bargaining” rights are enforced by the power of the state). Tyrone tells me that the only thing the post-Reagan “small government” schtick has accomplished is to push this process underground, so that covert transfers have been engineered by a “private” financial sector in ways that are inefficient, nontransparent, and often fraudulent according to traditional laws and norms. Some of these weak institutions upon which we relied to conduct transfers broke in 2008, so now we’re really feeling the pain. We’ll continue to feel the pain until we restore the ability of the financial system to hide widespread transfers, or until we employ some other sort of institution to provide a sustainable dispersion of purchasing power.
Having met both brothers Cowen, I can state with some confidence that Tyler is smarter, better-looking, and much more engaging company over lunch. Tyrone is kind of icky, hygienically speaking, and he strikes me as gratuitously mean. But I think he has a point.
- 14-Apr-2012, 10:00 p.m. EDT: More than a year later, Tyler Cowen and Scott Sumner have said nice things and relinked this piece. I gave it a reread, and though I remain happy with the substance, I found some textual and grammatical embarrassments. So no substantive changes, but I’ve made some small edits:
rendersrender financial sector revenue highly suspect as a marker of value.”
morelikely that our reduced payments have more to do with technology”
- “any attempt to mobilize savings in aggregate — any net dissaving — is likely to
be attended byresult in either inflation or displacement of consumption” (eliminated repetitive use of “attended by”)
 Cowen does acknowledge the possibility the internet is generating low-hanging fruit, but that there are lags in the indirect process by which a communication and coordination technology translates into consumer visible improvements.
 Even a thought experiment that holds health and nutrition constant but contemplates switching technologicolifestyles doesn’t cut it, because preferences are history-dependent. Emigration is a painful option even when better health and more food await on the other side of the border. We love the world that makes us. We’d really need to ask people in 1900 whether they’d prefer to move to our strange, televised world or to stay where they are but with 21st century health and wealth in terms of then-extant goods. If we allowed cross-migration between the 1900 and 2011 on those terms, giving people of both eras ample time to sample both eras, it’s not clear to me in which direction net migration would flow. To run this experiment, we’d have to correct for population differences and exclude groups that were plainly oppressed in 1900. (We’d want to exclude those groups from the sample, if what we are interested in is the welfare associated with technological, rather than social, change. It’s not obvious that the civil rights movement could not have happened independently of electricity or automobiles, although one could make a case that relaxation for the need for exploitative labor was a precondition for that movement.) Also, there’s the difficulty of simultaneously allowing both ample information for comparison and the sublime pleasure associated with not knowing what you are missing with respect to life in the other era.
 I hasten to add that I, personally, am a technophile, and curious enthusiasm for technological resets is much of what I live for. But I don’t think that’s a very general or even common preference.