How much is the Internet worth? How bad is income inequality in advanced economies? What is causing the productivity slowdown?
These three questions are related, but I doubt few people know why. The reason is that all three problems are misinterpreted based on the standard microeconomic measure of income and value. I’m going to make this brief, but keep in mind this is just an introduction to a series of work I’m planning to blow up the policy discussion economists have been having.
To begin, I live in Silicon Valley, literally in a Stanford apartment on Sand Hill Road. I’m not one of the tech geniuses building the future, but I do watch them from a front row seat, and I really believe the value creation here is vastly under appreciated in Washington, not to mention European capitals and most editorial pages. Even my friend Tyler Cowen seems to be missing the big picture:
The productivity slowdown is too big in scale, relative to the size of the tech sector, to be plausibly compensated for by tech progress. (from his March 4, 2016 NYT column, “Silicon Valley Has Not Saved Us From a Productivity Slowdown”)
The mistake economists have been making is to ignore value above the demand curve. Classic economics dating back to Adam Smith recognized a dual paradox of value, with one type of value high in utility (e.g., water, food, tools, clothing) and a more economic type of value signaled by prices in exchange (e.g., diamonds, gold, and other conspicuous goods). My theory is that the paradox is actually a trinity, and the classic dichotomy ignored value in consumption (or maybe a better word is desire). A forthcoming essay in COMMENTARY will make a more detailed case, but it will not include this figure which contrasts a Demand curve (Willingness to Pay, a.k.a. WTP) with something you won’t see in textbooks, yet:
The first objection I’ve heard to this theory is that the WTA-WTP gap has already been explained by the endowment effect, and not only that, but the pioneers of behavioral economics were already awarded a Nobel Prize. Sorry, but no, the endowment effect has been overinterpreted, and it is NOT fully fleshed out. For starters, the effect often disappears for commodities and appears elsewhere when people are not even endowed. It presumes that there is an objective value to a good (the exchange price) which is distorted by irrational subjective attachment. I think this “irrational primates” explanation goes too far. It makes more sense to appreciate a desire for something above the demand curve, to wit: a penniless bum is budget constrained and cannot afford an icy-cold can of Coca-cola (so his WTP is zero), but if he were given a can of soda, his consumption would far exceed zero. Indeed, there is a rich literature on contingent valuation that recognizes the WTA of many things approaches infinity (e.g., the value of an extra decade of life).
Why does this matter to Tyler and the techno-skeptics? Simply put, the WTA value of modern things is vastly higher than older, more tangible, more commoditized goods. I have conducted some preliminary, not-ready-for-peer-review research and discovered a huge gap differential:
I also tested two theories, again just preliminary. My first theory is that consumption valuation increases with time. The endowment effect should be linear — if you take a teddy bear away from a kid for seven days instead of one, he should be seven times as displeased. But I expect displeasure to be nonlinear, namely increasing exponentially. It is. Here’s my question of nearly 50 colleagues at the Hoover Institution about the WTA of pizza over three time periods.
My second theory is that valuation increases with rarity. Commodities that are easy to exchange should yield a WTP-WTA gap that is no larger than the friction of exchange. (And for the endowment nerds out there, I suspect this is most of what the canonical “college mug” experiments of Kahneman were picking up). But what if you were asked to sell a ticket to the concert of your favorite entertainer? What if it was their last concert ever? The way I framed the question to colleagues was by using three different types of tickets to a favorite sports team: a regular season ticket, a playoff ticket, and lastly a ticket to the championship game. The nonlinear valuations were confirmed. To be fair, these were not per-review quality studies, but I have no doubt those will show the same result.
Now let’s connect the dots back to Tyler. Is it true demand for Internet access on airline flights is low? Yes, but that’s explained by the short duration. People are willing to accept loss of something at a low value when the time is brief. I bet the value of a year of Internet is worth millions of dollars to Tyler Cowen, even though he can easily do without it on every flight. And it’s worth more than access to a library because there is no substitute for the Internet. It is unique, not a commodity.
If we did a better job as economists of measuring consumption values, we could chart the exponential increase in progress that has been missing from traditional output & productivity measures. The productivity slowdown masks a consumptivity explosion. And further to the point, even the poorest people in modern economies are enjoying the boom. Lower mortality, cleaner air, better parks, and free wifi are all real trends and really invaluable.