This is the once-a-week free edition of The Diff. Paywalled newsletters this week include: a useful economic law that’s always wrong, a review of a timely Edward Luttwak book (from 1968), and an overview of the game theory of crises.
Amazon Sees Like A State
Thirty years after the end of the Cold War, America’s government enjoys Brezhnevian sclerosis—which septuagenarian will be in charge next year!? Our biggest hyper-capitalist companies run internal cradle to grave, with free, high-quality public transportation and socialized food. They’ve delivered the “what” of utopian socialism (albeit with monster doses of Marxist labor alienation—sure, at one level you’re Organizing the World’s Information, but in practice that means a tiny optimization that makes pages render 1ms faster). But they’ve also delivered the “how”: big tech companies get big when they achieve a monopoly on resolving an information gap. Economic profits ultimately rely on information gaps. There are a few short descriptions for how you aggregate messy user-level information into economically valuable data. Depending on whether you want to be valued at 5x revenue, 10x, 20x, or 50x, you can call it statistics, data science, machine learning, or AI. All of these basically involve converting a question into a giant matrix, and solving it. This is basically what the USSR tried to do.
There are many practical critiques of socialism—body counts, corruption, darkness everywhere. But even before that, there was a damning theoretical critique: the calculation problem. A price system sets an exchange rate between any one thing and every other thing, at any point in time. Stalin didn’t know the exchange rate between a tractor and a sock and a poem. He just had to guess.
(This is one reason Russia embarked on such a breakneck industrialization program. The sooner you achieve post-scarcity, the sooner you stop worrying about tradeoffs. But to get there, you still need to know the exchange rate between an arc furnace, a truck chassis, and a ton of cement…)
In economic terms, the trouble with command economies is that supply and demand are curves, and a price is an intersection between those curves. The price is the most informative point on the curve, but price still projects multidimensional curves onto a one-dimensional number line. The Soviets tried mightily to solve the problem, building “IBM-compatible” machines and ever-more-complex models to meet demand, control prices, and avoid shortages.
It failed: insufficient data and bad incentives. But now Google, Facebook, and especially Amazon do the same thing: they analyze reams of data to implicitly estimate supply and demand curves. Amazon has to explicitly estimate them, since it provides the logistical backend for actually getting products to customers, and treats reliably fast shipping as a key differentiator. When Amazon makes a pricing decision, it’s making a decision about warehouse space, packing efficiency, last-mile shipping capacity, and working capital. And since so many goods it sells are complements and substitutes, it’s making that decision, incrementally, for countless other products besides. This is the sort of complex tradeoff that, in theory, should be intractable for a single institution. In theory, the way to handle complex tradeoffs between products in a catalogue and the entire supply chain required to deliver them is to let prices and incentives do the calculation. In practice, Amazon seems to just do it.
We have reached a weird point in history where we can say that true communism has never been tried, because they didn’t have enough RAM.
When I look at decisions like Amazon only selling masks to hospitals and governments, but using them in warehouses, I don’t see hypocrisy. I see the world’s first functioning command economy elegantly handling an edge case.
The fact that tech companies are very meta makes them a peculiar kind of altruistic: they own a percentage of economic growth, and given their model it’s usually high-margin (and valued at a high multiple). Once a tech company’s share of economic growth multiplied by its price/sales ratio equals one, the company becomes selfishly altruistic, at least for all forms of altruism that can be measured in GDP.
Last year, the US economy grew $850bn in nominal terms, and Amazon’s North America sales growth was $30bn of that. AMZN trades at 3.45x sales, so its altruism quotient is ($30bn/$850bn * 3.45), or 12%. i.e. any time Amazon can spend $1 to make the country $8 better-off, it’s a wash. Since Amazon is growing faster than the US economy, and its price/sales ratio has drifted up over time, this will only rise.
This gives Amazon one very good incentive, and one very bad one. The good incentive is to make the world a better place, as measured by GDP. More money means more purchases of electronics, clothes, food, books, on-demand videos, etc. But the bad incentive is to make the economy more measurable. One way to increase GDP is to invent a fabulously useful gadget. Another way to increase GDP is to substitute a home-cooked meal using $5 worth of ingredients for a food delivery order that costs $30 for something not quite as good.
This tractability argument is straight out of James C. Scott’s Seeing Like A State, which is all about governments' efforts to give people full names, fixed addresses, centrally-registered property rights, and other traits that make them more easily taxed and conscripted. Scott was writing from a historical perspective, and historically the most ambitious people in the world went into government. I don’t know who the most ambitious industrialist in Napoleonic France was, but that industrialist was less of a go-getter than the median French Army captain at the time. Today, Scott’s argument applies to tech, because tech is where ambitious people go. The institutional and legal framework are different—Jeff Bezos is not going to make you march on Moscow midwinter any time soon—but the incentives are the same.
There’s an old Isaac Asimov story, The Evitable Conflict. The story is set in the year 2052, when the world’s economy is run by four supercomputers. In the book, the world’s economy has moved beyond communism and capitalism. When computers make the decisions, it doesn’t really make sense to ask the question. This struck me as perfectly sensible when I first read it (I was 12), totally crazy when I reread it (some time in high school, after a megadose of Thomas Sowell), and intriguingly reasonable now (because it’s 2020).
Which is not to say that the commies were right. They were wrong all along, with disastrous results. It’s a victory lap of sorts that VCs looking for a big score and wall street financiers demanding ever-better quarters have caused big tech companies to casually implement the most effective form of economic central planning and socialist nanny-statism ever devised.
The killer critique of socialist economics was not that it was strictly impossible, just that it was hard enough to be effectively impossible. “Technology” is a shorthand for the process of making impossible things possible, so in a sense it was inevitable that the socialist calculation debate would eventually be resolved in socialism’s favor, just implemented by capitalists and judged by an earnings-per-share metric.
Lenin probably never said that when the revolution comes, capitalists will sell socialists the rope that’s used to hang them. But Huey Long did say “Sure, we’ll have fascism in this country, and we’ll call it anti-fascism.” So it’s all too appropriate to see Soviet-style central planning coming to America, sporting a $955bn market cap.
 I’m back on my footnote game. To flesh out this argument: in the short term, you can earn an economic profit from economies of scale or other cost advantages, but I tend to view those outcomes as distorted. Getting to the ideal scale is a tournament, but “ideal scale” is not a stable quantity. While there are rewards to timing chunky capex well—semiconductor foundries are probably the best modern example—there are big costs to being early. Regional monopolies are also a tournament; if you own the only gas station in a town that can support 1.5 gas stations, you have a good business, but either a) you bought it at a fair market price, or b) your earnings include a risk premium from correctly betting on the size of the market and correctly betting that you wouldn’t split the market with someone else. So those economic profits are somewhat illusory: they’re real to current owners, but that’s survivorship bias. Economic profit from regional and scale monopolies tends towards zero when you count all the people who went out of business trying to earn it.
Information asymmetry persists much longer, and it’s a good description of many other kinds of monopolies. What is a successful biotech company other than a monopoly on the knowledge required to make a particular drug? Of course, tournament dynamics are at play here, too, but it’s an iterated game: the real value of the firm is its ability to repeat the magic trick of inventing a new product. Consumer Internet companies are a sort of economic Maxwell’s Demon, creating a barrier between consumers who are hyper-efficient at expressing demand and companies that are hyper-efficient at satisfying that demand. The big Internet companies charge to make the introduction.
 A biography of Bill Gates written in the early 90s refers to the early 80s as “a period where ‘AI’ was becoming a marketing tool.”
I wrote a piece in Medium’s Marker about pensions. One way to think about pensions, from the recipient’s perspective, is that they’re basically equivalent to an annuity. If interest rates drop, that annuity is worth more today. Which means that to the pension fund itself, lower interest rates imply that their liabilities are higher. Normally, the way to hedge that risk is to buy long-term treasuries. Pension funds have, increasingly, not been doing that; they’ve been investing more and more in equities, including private equity. When the economy is growing, that pays off nicely. During a downturn, they’re hit in two directions: the value of their equities drops, but their liabilities rise.
And in Coindesk, a piece on why Bitcoin is not a safe haven for a dollar-shortage, but is in other cases.
- My friends at Glimpse have a new Covid-19 Trends Tracker, showcasing which products and services are getting the most incremental search interest during the pandemic. Some of these are clear first-order impacts, like “online church” and “bulk ammo” (there’s always somebody taking it more seriously than you). But some will be the start of a longer-term shift.
- One of many exciting real-time economic indicators: hotel occupancy in the last week of March was 22.6%, down 67.5 percentage points YoY. Occupancy is a good metric to track because it measures corporate sentiment, often less visible than consumer sentiment.
- App Annie says time spent on apps rose 20% YoY in Q1, but that appears to be roughly in line with the prior trend. Apparently the growth in Internet usage due to Covid-19 skews towards desktop.
- Bryan Caplan bites the bullet: if the CPI overestimated inflation due to outlet bias and variety before, it’s underestimating it now. Q2 2020 is going to be a very odd quarter for macroeconomic data: the most important ever, in the sense that it will measure a step-function decrease in standards of living around the world, but the least-important ever in the sense that every output will be subject to immense bias. A very approximate measure for the actual GDP hit in Q2: how much would you pay for all of this to go away?
- An unexpected impact of Covid-19: culture drift, because new employees will be onboarded remotely.
- eBay is expediting the one-time shift to e-commerce by waiving seller fees and discounting some services for newly-online physical retailers. The timing here is interesting. Did it take weeks for the opportunity to emerge—or weeks to make a decision?
- This is an incredibly informative piece on toilet paper shortages. Worth reading in full, just to get a sense of how complex modern logistics are, but the short version is: it is in fact reasonable to buy more toilet paper during a pandemic, and stores will indeed run low. The hoarders were right. Costco’s February comp was exactly what the efficient market hypothesis would imply.
New York Legalizes E-Bikes
No idea how they found the time to do this, but New York has un-banned motorized scooters. NYC was an early hotspot for car-sharing, the last major disruption in transportation technology, so it will be be interesting to see which of the surviving scooter rental companies grab share after this.
Greg Mankiw and Davie Weil predicted in the early 90s that housing prices, driven by Baby Boomer demand, would peak some time in the mid-90s. That is not exactly what happened. Part of why their thesis went wrong: as housing prices went up and consumers levered up, mortgages became the major channel through which monetary policy stimulated the economy. Meanwhile, high turnout for older voters meant that the Levered Boomer Demo was a key political constituency. Bubbles lead to bifurcation, especially when you can’t short: if the world consists of housing optimists and housing pessimists, as prices rise the optimists are increasingly the price-setters.
But you can’t outrun demographics forever. Eventually, people downsize or die. That’s starting to happen, according to a new paper. A striking stat: controlling for zip code, prices for 1- and 2-bedroom houses have been rising since 2012, but prices for 4- and 5-bedroom homes have been falling. And there’s a negative correlation between housing price changes and average age by zip code.
(I previously wrote about the economics of Baby Boomers here.)
Following their negative press (covered in yesterday’s newsletter, in which I argued that this happens to every growth company and has never killed any company), Zoom has prudently paused new features for 90 days to lock down security. They also casually mention that Zoom DAUs went from 10m at peak in December to over 200m today. 2,000% growth off a 10m base in three months is a record that may never be surpassed.