Intellectual Property

One major policy domain of which I have a limited understanding is intellectual property law and policy. So I wanted to write a post to talk through my understanding of intellectual property and invite you, readers, to correct and improve me in the comments. In the first part of the post, I’ll try to lay out broad ideas about intellectual property; in the second, I’ll try to apply those ideas to some current controversies.

***

Property: In the modern world, at least since the fall of the Berlin wall, most of us believe that private property is important and deserving of protection. There’s both a deontological, ethical argument and a consequentialist, economic argument for why. The deontological argument hearkens back to John Locke, who famously argued in his Second Treatise that people initially assumed property rights to the land by ‘mixing their labor’ with it; the original landlords, he imagined, were those who put their selves into their land by founding, taming, and tilling it, making that land a kind of extension of their selves; the right to private property was therefore an extension of self-ownership. Modern intellectual heirs to Locke (including those who identify as Nozickians) would argue by extension that when I freely contract with others, offering them my talents and hard work in return for cash compensation which I use to purchase assets or commodities, then I have earned and deserved those assets, and so taking those things from me would be a violation of my self and a denial of my desert. The economic argument for property rights it that people will not build and invest in valuable assets unless they feel assured that they will continue to control those assets and, hence, be able to use them to their profit. Countries that don’t credibly guarantee to protect private property discourage investment and scare their own citizens into investing all their assets abroad, hurting their growth and prosperity. (See Argentina.)

Property, Intellectual: Intellectual property  – typically defined as property that is a work or creation ‘of the mind’ or the ‘result of creativity’ — is similar but different from regular physical property. It’s arguably similar in that (1) the things I create with my mind are a kind of extension of my self, and so it would be a violation for someone to claim my work as their own or appropriate my work for profit without my consent and (2) people and companies will not invest in new ideas, research, creations, and brands unless they can be assured that they will gain compensating benefits for those investments. Since in a competitive market you cannot gain any profit from a thing that everybody else has access to, recognizing exclusive intellectual property rights is thought to be an ideal way to incentivize research and innovation. But intellectual property is different, crucially, from physical property in that it is abstract and hence non-rivalrous — that is, someone can copy my algorithm or song or blog post without taking it from me. If someone takes my physical asset, like my land, I can no longer enjoy and use it; if they copy my song or algorithm or blog post, I still can.

What principles should we use in granting intellectual property rights? I would argue that it’s better to think about property rights primarily through the consequentialist, economic lens, rather than the deontological, natural-rights-based lens. Philosophically, it’s hard to parse the boundaries of individual desert: I largely ‘owe’ my ability to produce creative work to the parents who fed me and read to me as a child, to the public institutions that educated me, and to the political system and culture that were the basis for it all. More practically, even the most Lockean stalwarts would caveat their understanding of property-as-natural right when the consequences are great enough: Suppose a brilliant scientist had discovered and patented a cure for all cancers, but refused to sell or license the patent out of a Kaczynski-esque hatred of technological modernity; in the face of the potential to save millions of lives, would we really have an obligation to ’respect’ this scientist’s ‘natural right’ to his discovery? For another thought experiment: If our ideas are extensions of ourselves, and thus our inviolable natural rights, then shouldn’t, e.g., a policy wonk or mayor who comes up with an innovative policy solution for managing mass-transit or Medicaid logistics be able to patent that method and prevent other municipalities from adopting it? If the creative inventions of our minds are our natural rights, why would we grant patents for, e.g., efficient computer algorithms for managing data, but not for efficient ‘social algorithms’ like those imagined policies? Finally, if intellectual property is a natural right, then how would we justify ever letting patents expire? In short, I think the property-as-natural-right argument doesn’t withstand philosophical scrutiny; we should instead think of intellectual property rights as constructed social tools, artificial legal rights that we as a society assign in the interest of promoting our shared prosperity and felicity.

The basic economics of intellectual property: If we accept the argument above, then we should think about intellectual property rights as economists do, as a tool to maximize social utility. To that end, we want to both (1) give people incentives to produce innovative and creative works in the first place and (2) maximize ordinary consumers’ ability to access and enjoy those goods. These two goals are obviously in tension: If you increase patent protections from 10 year to 15 years, then you give firms even stronger incentives to invest aggressively in research and development (because they’ll be able to command monopoly prices on the innovation for longer), but you’ll also increase by 5 years the length of time that consumers have to wait to enjoy the good at cheap, competitive prices. If you decrease patent protections from 10 years to 5 years, you’ll halve the time consumers have to wait for competitive prices on goods, but you might decrease risky and innovative R&D, as companies fear that they’ll find it hard to make a killing in that time. The economic debate centers on finding the social optimum, given this tradeoff.

As I’ve read up on intellectual property I’ve found that there’s basically a consensus among intellectuals and experts I respect about three very broad things: (1) The basic economic theory — that we generally need some IP protections to give incentives to creators, and the debatable question is how to optimize the tradeoff between giving creators these incentives and giving consumers earlier and more access — is sound; (2) But in its actual legal implementation, there are a lot of problems and abuses in our current IP-law system — our IP system is subject to abuse by extortionate ‘patent trolls,’ we grant patents for small, incremental changes to technologies that may not constitute truly creative breakthroughs, etc.; and (3) Our intellectual property legal protections are probably too strong overall. Item number three here is actually what one would predict given the theory of public choice. Intellectual property law lobbying is a classic example of “distributed costs and concentrated benefits.” Individual firms and patent owners get very big, very obvious benefits from legislative extensions of the protections on their intellectual properties and they lobby accordingly; we individual consumers get hurt by these in delayed access, higher product costs and health-insurance premiums, etc., but because these costs are diffuse and sometimes invisible, we don’t put appropriate pressure on our legislators to stop them. Thus our democracy produces laws whose aggregate costs outweigh their benefits.

There’s also a compelling heterodox viewpoint that our IP laws are radically too strong, and that we should radically weaken all of our IP protections and completely eliminate many of them. But in the rest of this post, I want to first touch on a potpourri of IP issues, using the consensus ideas, and then finish up by touching on the heterodox idea a little more.

***

Types of IP: First, let’s distinguish among types of intellectual property. The least controversial type is trademarks. Protecting trademarks simply amounts to preventing vendors from lying about who they are to consumers; this makes it easier for us find what we want from sources we trust. Very few people are against the indefinite extension and protection of trademarks. Industrial design rights basically amount to the same thing. Copyrights protect creative, artistic works, allowing their authors to control their use, replication, and distribution. Individual copyrights extend to 70 years after their authors’ deaths; corporate copyrights last until 120 years after their creation. It seems very hard to justify these extremely long copyright lengths. Personally, I find it hard to motivate myself with the prospect of the financial returns my blog posts will generate 1 year after my death, much less 70 years thereafter. Patents protect inventions and discoveries and allow their holders to control their use and sale for 20 years after the initial filing date (in the U.S.). While 20 years of patent protection doesn’t seem outrageous when compared to the length of copyright protection, it certainly seems like a long time for those hoping to access life-saving drugs at a competitive cost, or an energy company hoping to use some patented chemical in prototyping a new super-efficient battery, or non-Amazon e-commerce sites hoping to implement one-click shopping. In other words, it’s still an urgent question, are we offering too-strong patent protections?

Industry variance: In thinking about patents, we need to distinguish among industries. There’s no reason in principle to think that the ideal patent regime would be the same across all lines of business. Judge Richard A. Posner has argued that the pharmaceutical industry is a classic example of one that does require patent protection, due to its high, up-front R&D costs and uncertain payoffs, but that most other industries “would do fine” without patents. The I.T. industry in particular seems to be characterized by lots of lawsuits over patents held on ergonomic quick-fixes that seem more like part of the companies’ marketing than their R&D. Do we really think that Apple would not have developed the swipe function on the iPhone without the promise of patent protection? Or Amazon and its one-click-shopping option, for that matter? According to this table, all industries outside of pharmaceuticals and chemicals think the overwhelming majority of their patents would have been developed and implemented absent patent protection.

Patent trolls: These companies, which allegedly buy up dubious and less-than-innovative patents in order to shake down unsuspecting businesses with legal threats (usually coming formally from shell companies) are back in the news. In recent years companies using certain scanners and producing podcasts, for example, have received demands for cash from companies holding patents on ideas that only vaguely prefigured podcasts and contemporary scanners. People in the tech industry overwhelmingly say that they feel that innovation is being stalled by tech firms’ constant legal anxiousness that they’ll be found in violation of some esoteric, vague patent. Part of the problem is that the overstretched and understaffed U.S. patent office has granted a lot of vague patents that it probably should not have. The President is currently proposing new rules that would require that patent-holders be disclosed in patent arbitration cases; this would, at least, expose and hopefully shame the most blatant patent trolls. A more general idea for mitigating patent trolling is that we should be able to patent only implementations, not purely abstract ideas.

Music today (back to copyrights): The music industry today is an instructive case. As we all know, it’s very easy to download and torrent music for free online and so young people generally do not pay for the copyrighted music they listen to. And yet it’s commonly observed that musicians are doing better than ever before today. How’s that? The radically expanded access to music that we consumers are all now enjoying, and the ease with which we can share and recommend our friends, has whetted our collective interest in musicians and we now pay more to see more live shows than ever before. What musicians have lost in CD sales they’ve largely made up in ticket revenues. I suspect that in the future, the authorities will largely grow to accept a world characterized by (1) not-for-profit illegal downloading of media; (2) for-profit ventures like Spotify that stream music for users for small fees or advertisements and pay relatively small per-play fees to creators; and (3) pop music and movie producers that know they have to be extra spectacular to draw people into concert venues and theaters, and indie bands that learn how to cultivate voluntarily supportive cult followings. More generally, the fact that musicians have flourished despite the effective erosion of their copyrights, thanks to second-order effects of music’s increasing availability, strengthens the case for reconsidering intellectual property rights in other domains as well.

Financing medical innovation through public prizes: Pharmaceutical patents are possibly the most controversial domain of patents, first for the obvious reason that denying or charging prohibitively high prices for necessary medicine horrifies us, and second because many people see pharmaceutical companies as patenting a lot of not-so-innovative incremental changes to extant drugs, and then pushing these patented drugs, which they can sell at monopoly prices, on insurers, doctors, and consumers, driving up costs for all of us, without providing true innovation or benefits. (Now, notably, I think it’s silly to blame patent rights per se  or corporate greed here — the root problem (if I may pause to grind my ax) is the total lack of individual incentives in our insane health care system. If the medical market were more cost sensitive at every level, and we consumers were rewarded for our choices to reduce our costs, then we would simply choose to use less expensive, unpatented drugs, unless the more expensive, patented ones offered compensating benefits.) But given that my preferred healthcare policies are not likely to be implemented, how else could we mitigate this problem of wasteful pharmaceutical investment and innovation? One clever idea would be to stop granting new pharmaceutical patents and instead begin offering public prizes. I.e., the government would offer $5 billion for whichever company could first produce a drug that met some well-defined criteria in improving our treatment of AIDS/Alzheimer’s in XYZ ways. Theoretically, this would stop pharmaceutical companies from overinvesting in small, incremental pharmaceutical innovations and encourage them to focus on our most pressing health needs, as defined by smart public authorities. Once the government had awarded the prize to the victorious pharmaceutical company, any company in the world would have the right to vend it, and so its price would quickly be driven down to its marginal cost of production, immediately widening its availability. This is a clever idea, but it certainly has some problems: As a public-sector entity, such a prize-granting agency would face political pressures to focus on politically popular cures, and underinvest in less salient ones; with no ‘bottom-line,’ its revolving-door bureaucrats might overpay pharmaceutical companies generally, just as, today, government contractors are seen as overpaid; there would be huge liability issues and public outrages when prize-winning drugs turned out to have mild side effects or tradeoffs or cause 3 people 5 sleepless nights, which might also drive such a public entity to be way too conservative in awarding prizes and publicly offering drugs. I’m not sure whether our current system or this proposal is more imperfect.

Intellectual property abroad: Enforcing U.S. patents and trademarks abroad, particularly in China, India, and Africa, is a legally and morally tricky issue. Legally, sovereign nations are sovereign within their boundaries, and so U.S. patents qua U.S. patents simply don’t apply outside of the U.S. — we can only persuade and, sometimes, pressure these government through trade retaliation to adopt and enforce their own laws protecting U.S.-based IP. Morally, there’s an argument to be made that that developed-world creators’ primary markets will always be in rich, developed nations. If developed-world consumers produce strong enough incentives for innovation, then there’s a strong humanitarian argument for letting it slide when poor countries violate IP laws and use our innovations very cheaply to save lives and develop their lagging domestic economies. At the same time, we can also understand the distress American consumers feel when they find that drugs that were developed in U.S. labs and are prohibitively expensive in the U.S. are cheap and over the counter in India. And since China, India, and Africa contain most of the people in the world, as they develop economically, they’ll become increasingly important factors in firms’ incentives to innovate, and so at some point it’ll be key for them to get more serious about intellectual property. 

***

The radical argument: Some smart, legit folks argue that we should go a lot further than I’ve advocated here in radically weakening all, and completely abolishing much, intellectual property protection. Proponents of this viewpoint point to the fashion industry: There, new clothing designs enjoy no patent protections (because clothing, due to its utilitarian function, does not constitute art that can be copyrighted) and yet we still see plenty of innovation. Perhaps we would see the same in other industries if we abolished their IP protections: firms would arguably continue to invest in R&D and innovation, even absent patent protections, seeking the financial rewards of ‘brand value’ in being the first and the best to implement their innovation. In addition, absent protections, there would be more technologies and ideas in the public domain, which would give us all more resources to draw upon in producing new innovations. A software engineer would have a somewhat smaller financial incentive to make any new software innovation, but she would have a lot more ideas and software engineers to draw on in producing new ideas. So it’s plausible that abolishing patents here could produce more innovation in the aggregate. 

On the whole, then, these heterodox thinkers argue, we’d be better off with much, much weaker IP protection. I’m not sure this argument is right, particularly for firms in industries with very large up-front R&D costs, like pharmaceuticals and alternative energy. I also worry that once firms couldn’t enjoy monopoly rights to their patents, they would respond by aggressively and permanently guarding the secrecy of their innovations, which wouldn’t be great for human progress in the long run. But I think the example of the still-innovative fashion industry, and the surprisingly still-successful music industry should both push us to think very boldly about more narrowly circumscribing intellectual-property rights. 

Some thoughts on higher education

After healthcare, the biggest, growing expense that is dragging on every middle-class American’s well-being right now is probably the cost of higher education. Full tuition and expenses at top colleges in the U.S. is famously surpassing $60,000 a year. Newly-minted college graduates are taking five to six figures of debt into an economy with extremely high youth unemployment, in which a college degree is no longer a guarantee of a stable middle-class existence. New J.D.s are famously graduating from law school with six-figure debt loads and declining job prospects. American medicine is facing a shortage of general practitioners, at least partly because a lot of young M.D.s can’t bear the expense and work of medical school and internships if they’ll be condemned to a life making only (!) $300,000 a year, as non-specialists. These have a lot of serious, second-order, distributed costs that we don’t always think about: economically indebted young people are more risk-averse, less confident, more prone to depression and anxiety; the ever-growing costs of labor in services that employ credentialed professionals get passed on to all of us when we use their services; less savings ends up invested in other useful places in the economy; a kid whose parents grew up working class, but who are now middle-class and ineligible for financial aid, might choose to go to attend a state school instead of a prestigious Ivy League university, meaning that high college costs drag on social mobility even given generous financial-aid packages. But one cost that sticks out to me is that I think parents should be allowed to have a little fun and live large once their kids have graduated form high school. And many parents who fund their children’s educations are spending all of their savings — money which they could have put to a lot of other fun and worthy uses.

So it’s a big deal. What’s driving these rising costs? Economists who research this talk about a bunch of different things. First, since the 1970s, the “skills premium” in American wages has increased — that is, the differential between college-graduates’ and high-school graduates’ has grown. This, in turn, is explained by the fact that the U.S. has continued transitioning from a manufacturing-based economy to a services-based economy driven by information and knowledge. So as the financial returns to college education have increased, the purchase price that colleges can demand has naturally increased as well (particularly given that available spots at elite colleges have not kept pace with population growth). But colleges are non-profit — so where has all this extra money gone? One major rising cost is faculty salaries, and this has to do with a nifty economic concept called Baumol’s cost disease — since technology and globalization have increased the productivity of highly-educated professionals in other fields, such as law and finance, academe has had to raise its faculty salaries in order to compete with those industries for the highly educated, even though faculty productivity has not increased. Then, there are a lot of other assorted sources of growing costs: increases in administrative and non-faculty university staff (including yours truly!); all the indoor rock-climbing gyms and exorbitant athletic facilities and other frivolities designed to lure high-school seniors who do not know what money is.

These high costs and frivolities may be tolerable in a time of affluence. But since the recent recession, people have become increasingly upset. A spate of books have been written questioning whether college is still worth it. (For the record, in terms of financial returns, strictly, there’s no question that college is still ‘worth it’, in that the college wage-premium easily repays the cost of college, though there is a legitimate debate about the source of this advantage, i.e., whether it comes from real ‘value-add’ to graduates or mere signaling). In the startup community, it’s increasingly fashionable to advocate “hacking” your education, outside of prestigious brick-and-mortar universities.

Normatively, it’s very important to look for policies and innovations that can decrease the cost of providing higher education. Descriptively, colleges may face a much less compliant clientele unless they lower their prices (already, law-school applications have fallen of sharply). How could this happen? As in any other industry, decreases in costs will have to come from competition and technological advance. The major technological change that could impact higher education is the internet in general, especially Massive Online Open Courses (MOOCs), such as those being offered by edX and Coursera. The best argument for universities experimenting with employing MOOCs is that college costs are currently so unacceptably high that we should be open to almost any experiments to help control higher-education costs. But in the rest of this post, I want to consider MOOCs, and argue that they’re not just a valid experiment, but they’re likely to be a part of the right answer as well.

***

What is the value of higher education? It might be most helpful to partition higher education into two parts. Part 1 is higher education’s instrumental value — i.e., it’s practical, it’s skill-acquisition, it’s relevant to jobs, it’s giving people abilities that will match them up with what the market is demanding. Part 2 is about education’s intrinsic value — i.e., finding yourself, inhabiting unusual and novel perspectives on life, learning to better understand and empathize with others, asking question that are just worth asking for their own sake, etc.. We probably get more of Part 1 in STEM classes and lab work. We probably get more of Part 2 in English and philosophy classes and in the conversations we have with our fellow bright young collegians. Now, this taxonomy is imperfect. It’s likely that things like “communication skills” and “teamwork” and “leadership” — all skills that employers look for — are things that we develop in late-night conversations and philosophy papers and extracurriculars. It’s also the case that computer science, cognitive science, and physics all are intrinsically meaningful and beautiful as well, and can even expand our curiosity and empathy. But this imperfect schema might help our thinking a bit as we move forward.

In particular, I think there’s little controversy that MOOCs could be extremely useful in at least contributing to the provision of Part 1 of higher education. Indeed, MOOCs might be able to take over the majority of the work for many classes in this category. This past year, I took an introductory computer science course in my free time and never once attended the lecture in person. I sometimes watched a live feed of the lectures from my office — I usually watched them after the fact. But it wasn’t clear to me why the professor was still lecturing in person — few people attended class in person anyways, and  he’s been giving the same intro course for many years now. Tellingly, when a Monday class was cancelled due to the Boston marathon bombing, the professor simply had us watch his lecture from the previous year. In these courses, I also didn’t get much individual attention from my overworked, grad-student Teaching Assistants. I benefited more from online fora where I could exchange questions and tips with other students. And my problem sets probably could have been graded by a computer instead of these TAs — professors can easily write programs that, in a few seconds, throw thousands of different potential inputs into a program to make sure that the programs output the correct answer.

So I think it’s a no-brainer that universities should broadcast and offer credit for MOOC-based intro CS courses and other similar introductory STEM courses. For intro chem and bio classes, universities would likely employ mixed model, where students would watch lectures online, but attend lab in person. This would free professors up from intro teaching duties that they generally don’t enjoy. And by allowing students to choose from a variety of MOOC courses to use toward their college credit, students can be matched up with professors whose teaching styles fit them best, whatever university they’re situated at. This choice could also (once professors receive compensation for the MOOC use of their courses) bring competitive pressures to bear on professors’ teaching efforts.

For more advanced classes across the STEM category, I imagine mixed models would prevail. For a higher level course on, e.g., the theory of efficient algorithms, professors might want students to watch some lectures, write some programs, and master some content via MOOC-style recorded content, and automatically-graded problem sets, but the professor might want the students to then attend some seminar discussions on the much trickier theoretical stuff. Or a university might offer calculus, linear algebra, differential equations, real analysis, and mathematical logic, as MOOCs, while expecting math majors to attend seminars on the later, pure math theory courses in the curriculum, in person.

But the coolest thing about MOOCs is how they might provide more freedom and flexibility to people in seeking necessary job credentials. If you were, say, a successful engineer in Pakistan, but you moved your family to the U.S. (out of fear of religious persecution, or a desire to provide a brighter future for your grandkids), your lack of U.S.-based academic credentials might prevent you from landing a job in the U.S. that could fully employ your talents. Or if you’re a 25-year old mother stuck in a mid-level job, you might feel that your kids’ existence will preclude you from going back to school for a law degree or a CS masters degree. If these people could attend classes online at night, and get credit for their actual knowledge, however they attained it, and then get matched to jobs that are appropriate for their talent levels, that would our economy a lot more fair and efficient.

***

Can MOOCs also change the provision of Part 2 of higher education? I have a couple of thoughts about this. First,  ”asking deep, meaningful, philosophical questions for their own sake” sounds really nice — and it’s something I sincerely believe in in the abstract — but it’s probably not of much interest to the vast majority of people and is probably, in fact, a luxury that disproportionately appeals to that class of people who write about ideas and run universities for a living. The idea that a good education should not concern itself with utility is a luxury of those who will never need to really worry about unemployment.

Second, it’s hard to say how MOOCs will contribute to Part 2 of higher education, because it’s really hard to define what exactly Part 2 is, and how we measure it in the first place. We say that the liberal arts should make us more empathetic people and open our minds. So what do we make of the value of the liberal arts education of a recent cultural studies BA who gives no money to charity, spends no time interacting with people who lack his cultural markers and affiliations, and is completely intellectually incurious about non-Marxist veins of economic thought and aggressive towards those who are? Did this person fail at his liberal-arts education in the same way that, say, a computer science major who couldn’t build an app did?

Third, every other Yale College graduate I talk to says the same thing, that the most meaningful aspect of their time at Yale was their constant conversations with each other — i.e., the philosophy and political theory that happened outside of the classroom. Right now, people tend to have these excellent transformative experiences at college. But in principle, it’s not clear why that has to be the case — and it’s also not clear how much of Yale’s $200,000 tuition expenses are necessary to facilitate those experiences.

So how will MOOCs transform Part 2 of education? The conventional wisdom is that they’ll only slightly change it, as part of a blended model — i.e., that  students may watch recorded lectures from great teachers, will still attending seminars in person and having their essays graded by people. But I think it would be interesting to see how far we could pushing using digital technologies for Part 2. Professor Gregory Nagy, a professor of Greek at Harvard, has made a compelling case that automated multiple-choice grading in Humanities courses can be useful, when well-designed:

A little later, Nagy read me some questions that the team had devised for CB22x’s first multiple-choice test: “ ‘What is the will of Zeus?’ It says, ‘a) To send the souls of heroes to Hades’ ”—Nagy rippled into laughter—“ ‘b) To cause the Iliad,’ and ‘c) To cause the Trojan War.’ I love this. The best answer is ‘b) To cause the Iliad’—Zeus’ will encompasses the whole of the poem through to its end, or telos.”

He went on, “And then—this is where people really read into the text!—‘Why will Achilles sit the war out in his shelter?’ Because ‘a) He has hurt feelings,’ ‘b) He is angry at Agamemnon,’ and ‘c) A goddess advised him to do so.’ No one will get this.”

The answer is c). In Nagy’s “brick-and-mortar” class, students write essays. But multiple-choice questions are almost as good as essays, Nagy said, because they spot-check participants’ deeper comprehension of the text. The online testing mechanism explains the right response when students miss an answer. And it lets them see the reasoning behind the correct choice when they’re right. “Even in a multiple-choice or a yes-and-no situation, you can actually induce learners to read out of the text, not into the text,” Nagy explained

But there’s another possibility. Everything I’ve discussed so far has centered on simply complementing or replacing some of the features of current universities, within the structure of universities as they exist today. But the most truly “disruptive” proposal for online education is currently coming from the Minerva Project. The Minerva Project intends to have a highly-selective admissions process (it aims to get ‘Ivy-League quality students’) and then house them at different dormitories, on a rotating basis, over the four years of their education. Meanwhile, they’ll watch recorded lectures from top scholars online (meaning–the top scholars only need to be involved in the production of course material once), while they’ll interact with, and be graded by, newly-minted PhDs who are currently out of jobs. Minerva claims that by cutting out the expenses of university infrastructure, athletic fields, etc., it will be able to charge half the tuition of most top-tier universities today. And by housing elite students together, they’ll maintain the benefits of late-night dorm-room conversations, etc.. By moving them around the world, from Paris to Sao Paulo, etc., every few months, they’ll make them more cosmopolitan citizens of the world.

Will it work? It’s not clear. But we need to try.

Inflation basics [Econ for poets]

I recently had a conversation with a smart acquaintance about monetary policy, and we discussed the new Bank of Japan’s governors’ promises to push for higher inflation in the country. I tried to argue that we had good reasons to believe that such an inflationary policy could boost the real economy, while my friend argued against me. But eventually, I realized that the friend and I were doing a bad job articulating what, exactly, drives inflation, and this was a drag on our conversation. I suspect that there are a lot of us who know how to use all the words we see associated with inflation in magazines (“money supply,” “loose monetary policy,” “inflation expectations,” etc. etc.), who may even remember a mathematical formula from Intro Macro (MV = PQ), but who, when we dig a little deeper, have to admit we don’t have a clear grasp on what’s going on. So I thought I could do the blog world a favor by writing a very back-to-basic post (in English words) on what inflation is exactly and how it happens.

***

What is inflation? It is a rise in the prices of goods and services. What causes inflation? Most people would say that  inflation is driven by an increase in the amount of currency or money in the economy — the “money supply.” The intuition here is that if an economy produces the exact same amount of goods in year 1 as in year 2, but there is twice as much money in circulation in year 2, then prices will have to double in order to sort of “soak up” the extra money. I think that’s the implicit metaphor most of us have for how it works: The monetary price of real goods is determined by the amount of money in circulation relative to the amount of real goods; and inflation (and deflation) is driven by increases (and decreases) in the money supply. Now, the interesting thing about this is that it is mostly true in practice but not entirely true in theory. To get a much better grasp  on this, we need to go back to very basic theory, to make sure we’re clear on things, and then we need to clarify exactly what we mean by the “money supply.”

Who sets prices? Theory: In a market economy, everybody sets prices. That is, the price of anything in a market economy is the price at which sellers choose to sell their goods, provided that they can find buyers. So any full explanation of inflation has to answer the question: Why, exactly, did sellers choose to raise their prices and why did buyers go along with it? So let’s start with an incredibly simple model: Adam and Barbara are stranded on a desert island and they have their own economy. Adam grows peaches on his peach tree; every day, he harvests a bushel, eats a peach for himself, and sells the rest to Barbara; Barbara then eats a peach, turns the rest into peach juice, drinks some of it, and sells the rest back to Adam; Adam drinks some of the peach juice and uses the rest to water/fertilize the soil of his peach tree. One day, a $10 bill falls from the sky. Adam and Barbara decide to use this for their transactions: First, Barbara gives Adam the $10 bill in exchange for his peaches; then Adam gives Barbara the $10 back for her peach juice.

Now, suppose that two more $10 bill falls from the sky, one into Adam’s hand and another into Barbara’s. What will happen? Will prices triple? Well, that’s up to Adam and Barbara. They might just decide to save their new $10 bills and continue trading one day’s worth of peaches and one day’s worth of juice for $10, every single day — the only thing that would have changed from before would be their “savings.” But it also is possible that prices could increase. Maybe one day Adam gets greedy for dollar bills, and decides to demand $20 from Barbara for his peaches — he knows she has the money, and since he’s her only supplier, she has to consent. At that point, since Barbara now expects she’ll have to pay $20 for future supplies of peaches, she’ll start charging $20 for a day’s worth of peach juice in order to maintain her living standard. So suddenly prices double, just like that. And it’s also possible — this is the really interesting part — that prices could more than triple. Perhaps Adam gets really greedy and starts to charge $40 for his peaches — more than all the currency in the economy — and Barbara responds by charging $40 for her peach juice as well. One way this could work is that, first Barbara buys half a day’s supply of peaches for $20, makes half a day’s supply of peach juice and sells it for $20, and then uses that $20 to buy the next half-day’s supply, etc. Another way they could do this would be to use the magic of credit –  Adam or Barbara hands over $20 for the full amount of peaches/peach juice and also a promise to pay another $20 that night. At the end of the day, after their two transactions, each is $20 in debt to the other, but each earned $20 in cash from that day’s transaction, so they simply swap $20 to settle up.

Now, notably, this simple model is not a good a good one, because it leaves out (1) the reason money is useful and influences our behavior in the first place, namely that is completely fungible and usable across a broad array of transactions that would otherwise be complicated by barter and (2) competition, which is the major thing that stabilizes prices in the first place. But the point of this model has been to get us beyond our implicit metaphor that prices have to “soak up” the supply of money. Adam and Barbara — the market — are in charge of the prices they set, and they do so according to their own purposes. They could randomly double or halve their prices at their whims. And what’s true for Adam and Barbara is also theoretically true for all of us. If every single person in the world were to wake up in the morning and decide to double the prices they pay and charge for absolutely everything (including doubling, e.g. the amount of credit they demand from and extend from others), then this could work without a hitch — every numerical representation of the value of every good would change, and nothing else would.

The above is just a verbal expression of the familiar “Equation of Exchange” that we see in Econ 101, MV = PQ. In this equation, P represents the price level and Q represents the total quantity of real goods sold — multiplied together, PQ thus simply represents the nominal value of all real goods sold in a given time period. So in the second iteration of our fictional desert-island economy above (where Adam and Barbara were each charging $20), PQ = $40 per day. What about the other side of the equation? M represents the supply of money (a total of $20 in that part of the thought experiment). And V is stands for velocity of money, or the number of times any given unit of that money changes hands in a transaction, per time period; in our thought experiment, since $40 worth of goods changed hands a day, and the amount of money was only $30, then the velocity of money was 1.333 transactions per day (($40 of transactions/day) / $30). If you think carefully about this, you can see that MV = PQ is an axiomatic mathematical identity: The total monetary value of all transactions taking place in a given period of time must necessarily be equal to the amount of money there is times the number of times the average unit of money changed hands in a transaction. If prices suddenly double, while everything else stays the same, it must necessarily be the case that money is changing hands twice as fast, doubling V.

So let’s now think about some of the things that happened in our thought experiment, in terms of this identity, PQ = MV. At first, there was $10 in the economy, and $20 worth of purchases, because the $10 bill changed hands twice a day. So PQ = $20 and MV = 2 * $10. It balances! Then $20 fell from the sky. In one scenario, Adam and Barbara didn’t change their prices, so PQ still was equal to $20. Since M was was now equal to $30, V must have fallen to 2/3rd. In other words, since they were still just doing the same transactions, at the same dollar value, even though there were two new $10 bills hanging around, the ‘velocity’ of any given $10 bill was now 1/3rd of what it had previously been — only 2 $10 bills changed hands per day, even though there were 3 of them in the economy. In the scenario after that, both Adam and Barbara raised prices to $40, meaning that PQ was now equal to $80. Because M was equal to $30, V was necessarily 8/3 transactions per day — that is, the average $10 bill changed hands more than twice, because of how Adam and Barbara transacted four times per day.

So going forward, let’s keep in mind this main theoretical takeaway: The only fundamental constraint on prices is the mathematical identity that PQ = MV. So, if the money supply, M, say doubles, that could cause prices to double, but it’s also possible that the extra money could get “soaked up” by a lower velocity of money, i.e., people choosing, for whatever reason, to hold on to any given dollar in their hands for longer before spending it (and it’s also possible that we could see a little bit of each, or that velocity could surprisingly increase, leading to more than double inflation, etc., etc., etc.)

What influences prices? Practice: In theory, the only certainty about the price level is the identity that MV = PQ — the velocity of money could double one day, and halve the next, making prices double and halve in turn. But in practice, things are much different. First, we don’t, in practice, all just wake up in the morning and all collectively decide to double or halve the velocity of money. If I own a shop and I double my prices one day, my competitors probably won’t, and so all my customers will leave me and buy from them. If I suddenly halve my prices, I’ll run out of goods real quick and won’t make a profit. So, because most firms (hopefully!) face real and prospective competitors and don’t like selling things at a loss, the velocity of money, V, doesn’t just randomly, wildly oscillate on its own. This means that if both the quantity of real goods an economy is producing, Q, and the money supply, M, are held relatively constant, then we won’t usually see wild fluctuations in the price level, P.

And second, in practice, changes in the supply of money do not usually get entirely absorbed/cancelled out by changes in the velocity of money. Just think about it: If you suddenly had an extra $100,000 would you hide it all under your mattress? Maybe you would hide some of it (you would probably save much — but these savings would be someone else’s credit, which we’ll get to later), but probably you would increase your spending at least somewhat. And if all of us suddenly got an extra $100,000 we would all probably start to spend a bit more. Since our increased spending would amount to an increase in nominal demand for goods, we would expect prices to rise. So the Econ 101 explanation here is that increases in money lead to an increase in nominal demand, which causes nominal prices to rise. If you prefer narrative to graphical style thinking, think of it this way: if we helicopter-dropped an extra $100,000 into everyone’s bedroom, workers would demand higher pay to work overtime (since they already have such great savings), people would take vacations and bid up the price of spots at restaurants and on airplanes, everyone would be willing to pay more for houses, bidding up prices, etc., etc. But people also would hold onto or save much of that $100,000, meaning that velocity of any given dollar would slow down at first, and so the extra money supply wouldn’t be immediately ploughed into higher prices. So usually the price level should correlate and moves with the money supply, but not immediately in a perfect, linear 1-to-1 relationship.

What is money? In the first few iterations of the desert-island thought experiment, “money” basically means “paper currency.” But in the modern world, most of what we call “money” is actually just debits and credits in bank accounts. For example, if you have accumulated $10,000 in cash at work, and you put that into a checking account, you still have $10,000 in “money” (because you can withdraw at any time) even though your bank is not keeping those $10,000 locked away in a vault. Your bank likely lent most of those $10,000 in cash out to somebody else, and so now there is $19,000+ in “money” resulting from your deposit, even though there was only $10,000 in cash. Indeed, if the person who got that loan from the bank spends her $9,000 to hire somebody a job, and that hiree then saves his $9,000, and the bank then loans out those $9,000 in cash to somebody else, then there is now $28,000 in money. As we can see, in the modern world, “money” is very different from “currency,” and so economists have very categories for measuring the money supply. “M0″ refers to actual physical currency in circulation; “MB” (the Monetary Base) refers to currency in circulation, currency stored in bank vaults, and Federal Reserve credits to banks (see below); “M1″ refers to currency, bank deposits, and traveler’s checks; “M2″ includes savings accounts and money-market accounts as well; “M3″ includes all those and a few other savings/investment vehicles. As you can see, M0 through M3 are ordered according to their relative liquidity — M0 is just actual cash, which is completely liquid, and M3 includes things that might take a bit more time for you to withdraw — savings accounts and money-market funds. Money, in the modern world, exists on a spectrum of liquidity. Indeed, it’s arguable that ‘money’ in these traditional categories is too conservatively defined. If you have $10,000 invested in an index ETF, and you can exit the ETF at any moment, you might think of those $10,000 as your money, but the Federal Reserve, at least when it pays attention only to M0-M3, would not.

So how does the Federal Reserve control the money supply? It doesn’t do so by “printing money,” as Fed-skeptics often put it — it’s even more aerie than that! The Fed actually mostly influences the money supply just by entering credits and debits into its and other banks’ digital balance sheets.  Suppose a bank has $100 in deposits from savers like you and me, and it has loaned those $100 to General Electric. At this point, there are $200 ($100 in deposits, and $100 in cash on hand for GE). But now, the Federal Reserve can buy GE’s debt obligation from the bank; the bank thus gets $100 (or whatever the market purchase price of the loan was) in cash credit from the Federal Reserve, which it can then loan out to another company, like Ford. So now there’s $300 of money in the economy ($100 for GE and Ford each and $100 for the banks’ original depositors), with the extra $100 having been created simply by the Fed crediting another bank’s account.

In reality, due to ‘fractional reserve banking,’ each purchase of X that the Federal Reserve makes creates much more than X new money, because banks often lend to other banks, or banks’ loanees deposit some of their loans in other banks, etc. So the Federal Reserve can have a large impact on the money supply simply by purchasing banks’ assets — by giving these banks fresh money, it allows them to lend more money to other people/banks who will lend to other people/banks who will lend again, creating new money at each iteration.

***

I hope this is all the basic background one needs to understand the talk about inflation that we see in the business press. But I want to quickly touch on some implications:

1. This reason all this theory is important is that it explains why Federal Reserve policy is controversial and debatable. If there were a simple, linear relationship between the money supply and the price level, there would be no controversy — we could easily and uncontroversially predict inflation by quantifying the money supply. But Fed policy right now is controversial, for some, because we can’t actually be sure how changes in the money supply will affect inflation over the long run. It’s theoretically conceivable that a central bank could increase the money supply while observing very little inflation, because people largely hide their new money under their mattresses, only to see that 5 years later, everyone suddenly starts spending their mattress-savings, sending prices skyrocketing. The complex psychological factors that influence the velocity of money, including self-fulfilling expectations about inflation (see below), mean that there is always some uncertainty about what the consequences of the Fed’s actions will be. For the record, I’m not very worried about the prospect of very high inflation. The market’s expectations for future inflation are priced into price difference between TIPS (Treasury Inflation Protected Securities) and regular, non-inflation protected Treasuries. And TIPS continue to show low inflation expectations. If I were smarter than the market, I should probably be a billionaire right now. People who are very certain that high inflation is coming should put their money where their mouths are, by putting most of their savings in inflation-protected securities.

2. Expectations for inflation are largely self-fulfilling: If you expect wage rates to rise 10% next year, you might try to lure a new hire with a contract at a 8% premium (relative to current wages), to lock her in at a price that will be a 2% savings relative to what you expect for the future. If you expect prices for your supplies to rise next year, you might raise prices on your merchandise right now, in order to earn enough cash to afford those higher-priced supplies. If you think your competitors are raising their prices right now, then you know you can raise your prices without losing customers. Etc., etc., etc.. The fact that inflation is a sort of self-creating phenomenon, ultimately based on everyone’s best guess about what everyone else thinks about what everyone else thinks about how much prices will rise in the future, is one thing that sometimes makes it hard to control. Most episodes of hyperinflation ultimately originate from governments printing massive amounts of new money — but from there, inflation radically outpaces the printing presses, as everyone keeps raising prices in response to everyone else’s price hikes in a downward spiral. More, one of the most effective ways for the Fed to control inflation is for the Fed chairman to literally make statements — in words — about future inflation. If the Fed says, “we are committed to ensuring that inflation is 3% next year,” the average company will have a good reason to raise prices by 3%.

3. Most mainstream economists believe that moderately higher-than-usual inflation can help boost an economy out of a recession. There are at least four mechanisms through which inflation can benefit a recessionary economy:

          (i) If you own a company and you expect prices to be 8% higher next year, all else equal that fact will make you more inclined to purchase more merchandise now, while prices are still lower. You also might ramp up your production and investment right now, so you’ll be well-position to meet that high nominal demand.  This boost can help an economy get out of the recessionary downward spiral in which low demand and low production begets more low demand and low production.

          (ii)  Most of us think about our salaries in nominal terms. Most of us do not like to take paycuts. However, during a recession, individual workers’ productivity decreases (i.e., if I’m a car salesman, I’m worth more to my company during a time when lots of people want to buy cars). The problem is that if workers’ contribution to companies’ bottom lines decreases, but workers’ salaries stay the same, then firms will hire less and fire more, and/or become less competitive. Inflation allows firms to lower their employees’ real wages, without needing to lower their nominal wages. Economists think this is a good thing — the alternative to lower real wages during a recession is mass unemployment and bankruptcy.

          (iii) Inflating a currency typically devalues it relative to other world currencies. If we make the dollar worth less relative to the Brazilian real, then Brazilians will be able to more easily afford to buy American goods. This should help America’s exporters, which is another thing that can help drag a country out of a recessionary downward spiral. (The flip side of this, of course, is that it will be more expensive for Americans to import things from Brazil — so policymakers have to think carefully through the full industrial implications of a devalued currency).

          (iv) Inflating a currency benefits debtors (at the expense of creditors). If I owe my very wealthy landlord $1 million next year, but prices rise 15% in the interim, then the “real” value of my obligation to my landlord will only be some $850,000. If I as a middle-class consumer am more likely to spend extra money than my ultra-wealthy landlord, then this inflation-driven decrease in my debt/increase in my wealth (and decrease in my landlord’s wealth) will mean greater net demand in the economy. Again, this short-term boost to demand can help jolt an economy out of a downward spiral. You often hear that the problem we’re facing in the U.S. is that, after the financial crisis, everybody tried to “de-leverage” (that is, reduce their debt obligations) at the same time, which led to a “demand shortfall.” (This is often called the “paradox of thrift” — saving more money is good for any individual, but when everybody does it at the same time, it can cause a recession). Inflation can make it easier to reduce our debt obligations, thus weakening the demand shortfall problem that comes with deleveraging.

On the flip side, most mainstream economists believe that in non-recession times, relatively low, stable inflation is good. This is because it’s easier for people to enter into short-term and long-term economic contracts when they can have relatively certain expectation about what things will cost and be worth in the future.

The weird and awful/wonderful economics of taste and contemporary artisanship

This is a post about a weird and interesting space in economic theory, but it starts with a short anecdote.

Today, I went to my local barbershop and sat for an extra half hour browsing terrible magazines so that I could get my hair cut, specifically, by the owner of the place, an older man with blazing white hair and a thick Greek accent that he still retains from his boyhood in Samos. I feel subjectively that I look better when I get my haircut by the owner, as compared to the other barbers. But as a good junior social scientist, I always try to be skeptical of subjective impressions. Objective social science has been very good at obliterating a lot of our pious impressions about the superior quality of goods produced by lofty artisans and craftsmen — in blind taste tests connoisseurs can’t distinguish a fine wine/cheese from an ordinary one, etc. So what about my haircut? Is there any objective basis for my belief that the owner gives me a better one? What could explain my impression?

I have a couple of hypotheses:

#1: My first was that it’s is that it’s totally an illusion and I’ve just been primed by the owner’s foreign accent, old age, etc., to trust him as a craftsman. I.e., perhaps when the other barbers, with thick south Boston accents, cut my hair, prejudice leads me to watch their work with an overly-critical eye. I look in the mirror afterwards seeking to identify their mistakes and misjudgments and find them for this very reason. The owner’s wise-old-man aesthetic primes me to discover the evidence of his excellent good taste when I look in the mirror, and I see it for this very reason.

Is hypothesis #1 correct? Maybe — perhaps even probably. But my girlfriend, who is a fairly unbiased intellect and never present when I get haircuts, has agreed that my haircuts with the old man have been better. So I want to investigate the possibility that his haircuts really do look better. What in turn could explain this?

#2: The main objective difference I can observe in the barbershop is that the old owner of the place uses only scissors, while the other barbers use the modern electric tools. Could this explain the difference? It’d be really easy to say something like this: “Modern electric scissors save time but sacrifice quality. They impose uniform lengths and increments on men’s hair, while a truly good look depends on the layered textures, and smooth, non-discrete cuts that come only from scissors and the experienced judgments of a craftsman.”

This story could be true, but I’m skeptical. The reason I’m skeptical is that in an alternative universe people might be telling the exact opposite story just as plausibly. Supposed we lived in a world in which fine electric-mechanical devices were prohibitively expensive and rare. Scissors were abundant, but electric scissors were a luxury that only elites could afford. In this world, I’d bet that the electric look would be vaunted as desirable and superior. People in this world would probably say things like: “The electric haircut is a huge improvement over its pre-industrial equivalents. It allows the highly trained electric-scissor-certified barber to cut the hair in fine and exact geometries, as opposed to the rough, shabby, hastily layered looks of the past. A buzz cut is chic, crisp art deco on your head. Such a pity that only a few can afford it…”

See the problem? Our story about how the truly authentic scissored haircuts are better sounds nice; but there’s no way to objectively confirm it, so a person who is a critical outsider to our culture would argue that we’re just reverse-engineering a rationalization for our prejudices. If this is true, my impression makes a lot of sense: I don’t like the look my head gets when electric-scissored because of the cultural/affiliational/class-based reactions that have been ingrained into all of us. In my city, the buzzed, electric-scissored look is associated with the military, chains, Budget Cuts, etc. The look of hair cut by scissors, by contract, is associated with people and places that are willing to pay and wait extra to achieve a more fashionable appearance. And so the old-fashioned-scissored look seems more attractive not because of anything inhering in its geometry, but because of associations inhering in our culture and affiliations.

***

So the theory here is that there’s a kind of circular process going on: (1) Aesthetics and taste are not objective. (2) Electric scissors take less time and training to operate properly, so haircuts done with them are cheaper. (3) Therefore, aesthetics aside, income-constrained people will be more disposed to get electric-scissor haircuts; the hairstyles of elite people and elite urban areas will disproportionately be drafted by real scissors. (4) Therefore, the culture will come to associate electric-scissor haircuts with low social standing and regular-scissor haircuts with high social standing. (5) Therefore, the old-fashioned scissor haircuts will be upheld as “objective good taste” and self-conscious elites will be willing to pay more and wait longer for them, which will reinforce the distinction.
It is the superior price efficiency of the electric scissors that causes the look they produce to be associated with low social-standing, which causes it to be devalued. A generalization of this insight is that in matters of ‘taste’ (which is to say: in markers of social distinction) democratizing, price-lowering innovations are at least partly self-defeating.

***

This basic idea is key to understanding a lot of markets based around taste, cultural affiliations, etc., and is also troubling to the general optimistic picture of how markets work. Normally, we hope markets work something like this: When we all really want and/or need something, we bid up the price of it; the high price attracts entrepreneurs who want to make a lot of money meeting this demand; entrepreneurs uncover new technologies and production processes to make the thing more cheaply; the entrepreneurs compete with each other to market the good, driving their prices down; and so now everyone can get the thing they want on the cheap. See, e.g., automobiles, computers, etc. But for goods whose value comes at least partially from social distinction (i.e., “positional goods”), entrepreneurs can’t do quite so much good for us, because the technology and production processes that broaden access to the good will, ipso facto, reduce the value of the good (and be panned by cultural arbiters as ‘bad taste’). The value that electric scissors could provide to the world has been partially limited by the fact that their efficiency created a new distinction.

I find this interesting purely as a theoretical contrast to classical economic theory: In these domains, technology improves the objective features of a good, but in doing so detracts from its value as a token in human social hierarchies. In the supply-and-demand curves we saw in Econ 101, the demand for a good increases as its price declines; for these positional goods, the relationship is more ambiguous. But beyond theory, there are a couple interesting implications:

(1) Right now, Apple enjoys famously high margins on and earnings from its products. As Apple faces increasing competition and loses market share, it might be tempted to lower its prices, the natural response for any company fighting off competitors. As an economist, I should love this decision — more individuals could buy more great Apple products more easily. But if I were a consultant to the company, I might be hesitant: It seems to me that a large part of Apple’s brand value comes from the price distinction itself. Today, buying a non-iPhone smartphone labels you as someone who’s too eager to save a couple hundred bucks, a gaffe among yuppies. So Apple lowering its prices might not unambiguously raise its sales. What can Apple do? Personally, I think there’s just realistically no way Apple can keep up its current earnings and margins and so the company warrants its very low PE ratio. But this is not what consultants are hired to say.

(2) This theory provides some hope for an “artisanal economy” in the future. The basic idea, which I first heard proposed by Adam Davidson, is this: Throughout human history, improvements in technology have improved human welfare overall, even though technological disruptions caused short-term harm to the workers whom they made obsolete. But now some really smart people are starting to worry that this time is different. Once artificial intelligence advances sufficiently that robots can do literally anything that humans can do, there will be no way that we humans can complement technology and we’ll all start to be replaced by it instead. So who will have jobs in the future? Well, people who are part of protected licensing cartels might: As long as the government says you need to see a human doctor to get XYZ prescription, doctors will still have jobs. The people who own the capital and intellectual property used to make the robots will also still have plenty of income. But what about the rest of us?
Davidson has proposed that the future looks like Brooklyn, NY, in whose hip neighborhoods you can find artisinal offerings of just about anything. How is this economy supported? Mostly by people across the river, in Manhattan, whose incomes are either directly or indirectly tied to financial services. Are artisanal versions of goods better than their mass-produced industrial counterparts? A lot of artisanal foods probably wouldn’t come out ahead in a blind taste test, but artisanal goods in general are useful for us for expressing cultural affiliations and in-the-know-ness, or adding a unique quality to a dinner party or a unique aesthetic to an interior design. Artisanal goods are mostly useful as social tokens. And that’s a good thing. As such, they’re largely protected from competition from technology, because getting them cheap and efficiently is not the point — the point is having the experience of visiting the artisan’s boutique shop in a hip neighborhood, and telling the story of the good when you bring it home. I wonder if the economy of the future will look a bit like the economy that currently crosses the East River: technology does all the real work in satisfying our objective basic needs; the owners of capital and intellectual property earn huge profits as a result; and the rest of us are employed in vaguely creative professions, doing things that objectively robots could do, but which some rich capitalists want a unique human fingerprint on. I will let the reader decide whether that is utopia or dystopia.

Final post on health insurance: Back to basic theory and forward to controversial implications

Dear readers, many of you found my last post, a satire of American health insurance, amusing. But others of didn’t like that I departed from my usual style of careful and judicious exposition of economic theory. And still others complained that while I cruelly satirized others, I didn’t put forth any controversial views of my own, to expose myself for criticism. So in this post I’ll do both: I’ll explain the basic theory of health insurance and make a shocking claim about some drugs that I think should simply be offered over the counter, where competition would drive their costs so low that there would be no point in increasing insurance premiums in order to cover them.

What good is insurance? Insurance vs. prepayment

Suppose you live in a world designed by economic theorists, and you’ve just bought a house for $100,000 (lucky you!). Now suppose also that there’s a ferocious monster that comes from the sea, and there’s a 1 in 10 chance that in one of the next ten years, the monster will come and destroy your house. Indeed, every $100,000 house in the area has a 1 in 10 chance of being destroyed by the monster in any given year. What do you do? The answer should be pretty obvious. You should get together with, say, 1,000 other households with houses costing $100,000 in the area; you should all agree to pay in $1,000 a year to a single fund over the next 10 years; and the fund should pay out a full $100,000 to the victimized household whenever the monster destroys their house. Given the law of large numbers, you have good reason to believe that this fund should usually stay solvent; but there’s always the risk that more than 10 houses might get destroyed in any particular year. So to insure against that risk, you could raise the premiums to, say, $1,100 a year. And also you need some people to administrate the fund — treasurers, accountants, etc. — so you can raise the premiums to $1,150 to pay their salaries. Seems like a pretty good deal, right?

What’s the basic logic behind this? A freshman finance student might actually say this is a bad deal. After all, your ‘expected return’ in going uninsured is -$10,000 (.1 x $100,000) while your expected return while being insured is now -$11,500 (10 yrs x $1,150/yr). So why do we humans think this is a good deal? Well, to go back to deep economic theory — utility and all that — the idea here is that losing something worth $100,000 is more than ten times as sucky as losing $10,000. Paying an extra $1,150 a year for insurance is no fun; but losing $100,000 in assets could be absolutely devastating for a middle-class household — disrupting to all their life plans, etc. Rare big losses are much worse than small predictable losses. And this fact about the relationship between human happiness and money — the fact that huge losses are more than x times as sucky as premiums of 1/x — actually has a cool and important implication: Given that fact, we can improve human welfare and make everybody better off without producing any new goods or physical technologies, just by entering into contracts to socialize our risks through the ‘social technology’ of insurance.

So that’s what insurance is for. Insurance should be distinguished from prepayment. The insane way in which we do health insurance has confused the two. If I know with certainty that every five years my house will need $5,000 in repairs, then the hypothetical house insurance above shouldn’t pay for it, and it wouldn’t make any sense for me to try to make some new house-care plan to pay for it. Why? Well, if I wanted insurance to cover my house repairs, then the insurer would have to incorporate the certainty of those extra $10,000 in costs into my and everyone else’s premiums. And since there would be legal red tape and administrative expenses involved with taking in and paying out those $10,000, the insurer would need to add extra costs on top of that. I would end up paying in $11,000+ to get paid out $10,000. That’s not a good deal for me or anybody else. If we imagine drawing a spectrum from ‘pure prepayment’ to ‘pure insurance,’ then insurance is less and less useful as we get closer to the prepayment spectrum. If there’s a 25% chance each that my house repair expenses will be either $4,000, $5,000, $6,000 or $7,000, then in practice there’s not really much to be gained from buying insurance, even if theoretically there’s uncertainty about costs involved. 

So, insurance only adds value and makes us all better off by socializing the risks of unpredictable and incompletely controllable but very costly events. Prepayment doesn’t really do us any good, because it just pushes costs around and through extra levels of bureaucracy which add to its costs. The more that things which are effectively prepayments get incorporated into all of our insurance plans, the less well off we all are overall. Most of us understand this in other contexts: we don’t expect our car insurance to pay to refill our gas tanks; such a policy would involve transaction costs that would only increase the effective cost of driving, making us all worse off. But we lose track of this in healthcare, mostly because we’ve so completely confused prepayment and insurance that most people don’t even think to make the distinction anymore, and so lots of activists and special-interest groups have been very successful in getting their pet treatments covered by insurers, even when it doesn’t many any sense or do anyone any good when one thinks about the underlying economics.

Now, I’m trying to keep this, for now, on the abstract level of insurance per se. But here I need to make a theoretical point that only applies to health insurance: None of the above is fundamentally changed when your employer pays for your insurance. When you try to convince someone to hire you, you’re basically saying something along the lines of, “I can produce an extra $70,000 of value for you, and I’ll only cost you $65,000 in a salary, so you can add $5,000 to your bottom line by hiring me.” Now, when your required health coverage costs your employer an extra $5,000 a year, then that’s part of your employer’s effective cost of hiring you. So you’ll need to change your spiel to, “I can produce an extra $70,000 of value for you, and I’ll only cost you $60,000 in a salary and $5,000 in health coverage. So you can add $5,000 to your bottom line by hiring me.” What our employers pay for us in insurance therefore effectively comes out of our own salaries. And so we still really don’t want coverage for predictable, small-scale expense (that is, prepayments), even in our “employer-provided” health insurance.

Why health insurance is difficult:

Now, I proposed an ideal thought experiments above in which a bunch of homeowners got together and realized it was in their interest to socialize their risks through insurance contracts — every household made itself better off, with no government mandates or complex regulation required. Every household probably feels like it got a pretty good deal (whereas many of us don’t feel anything of the sort about our health plans). So one way to approach to the question of “What do we need public policy to do in the health insurance space?” is to ask the question, “How is real-world health-care different from that thought experiment?” I see a few main dimensions:

1. Individuals are short-sighted and emergency rooms are a classic public (non-excludable) good: The fundamental dilemma of health insurance is that none of us wants to turn dying uninsured patients away from emergency rooms. But some individuals will always think, “It will never happen to me.” And as long as we give emergency-room care to uninsured patients, then individuals have less reason to buy insurance up front. And as long as some individuals thus refuse to buy insurance up front, but we still all need to foot the bill for those who end up in emergency rooms, then health-care expenses go up for the rest of us, and more people feel that it is not worth it to buy insurance…and so forth. So this feedback loop screws up free-market health insurance. This is why I think the mandate is actually the very best thing from Obamacare.

2. Pre-existing conditions exist: In the thought experiment, every house had equal risk of being destroyed by a monster, and so they could all insure themselves at low and equal rates while still being profitable “bets” for their insurers. In reality, some people are born with, or acquire at early ages, conditions that entail so many risks and are so expensive to treat that no insurer could profitably insure them at a reasonable rate. This is another well-known problem with a competitive free-market for health insurance. I’m not sure what the ideal policy for providing for people with pre-existing conditions is, but I’ve seen a number of attractive proposals.

3. Assorted others: Information asymmetry, moral hazard, etc.: Information asymmetry means that you usually know more about your health risks than your prospective insurers would, so in a perfectly free market you might just wait until you’ve already gotten sick to buy insurance, and buy that insurance while hiding your condition from your insurer; this would make insurers suspicious of what their applicants aren’t telling them, thus driving up costs for everybody. Moral hazard means that as long as your insurer is footing the bill for your health costs, you  might not make as strong efforts to control those costs. I.e., you have less of an incentive to stop smoking if your insurer will not incorporate the cost of your eventual lung-cancer treatment into your premiums. As long as you’ll only need to pay a $5 copay, you might try to get your doctor to write you a phony script for Adderall so you can resell it on your college campus during finals week, etc., etc.

So those are just a couple of quick caveats and reasons why health care is unlike the ideal property-insurance scheme we laid out above. But outside of these classic and well-understood market failures, I want to make the way we provide healthcare to look more like competitive markets in other domains. For really big, uncontrollable, costly health risks, we should all socialize our risks through insurance. For regular, predictable, and medium-low costs, most of us should pay out of pocket and buy over-the-counter as much as possible, to reduce the added costs of extra layers of bureaucracy and force sellers to meet us at a low price — like grocery markets, etc. Folding the two together within healthcare only adds opacity, confusion, and extra cost. So now I’m going to expose myself to criticism by drawing out the implications.

Controversial implications: 

I think the major, under-discussed potential mechanism through which to reduce health costs in the U.S. is for our society to stop being so fearful of drugs and medicine, and begin to offer a lot more treatments over the counter (or at least delegate to pharmacists — who have a lot of training — the right to dispense these drugs). This would free up doctors’ time and lower costs for individuals — two unmitigated goods. Would lots of deregulation lead to some abuse, too? Yep. Will some people take drugs they don’t really need? Certainly (as they already do right now, by faking or exaggerating symptoms in their costly appointments with their doctors). But there are tradeoffs in life. And right now, health expense are unacceptably high. We should be willing to accept some relatively small risks to keep them down.

The only real prescription I have personal experience with is an Abuterol inhaler, which I used for exercise-induced asthma as a high-schooler. I see no reason why inhalers shouldn’t be available over the counter. As a child, I used to have scary fits of wheezing on the soccer field, and I knew that inhalers were an easy treatment, but I never wanted to see a doctor about it because I didn’t want the stigma of being ‘sick’ or needing ‘curing.’ Even today, I occasionally wheeze when I work out with the Cambridge Sports Union, but I don’t want to go through the trouble of seeing my doctor about it, because my time is very valuable and I don’t want my doctor to see me as an anxious neurotic. In high school, when I was more concerned about exercising my full abilities as a track athlete, I had at some point consented to see a doctor, and he prescribed me the basic mainline inhaler on the basis of nothing other than my subjective self-report. So why did I have to go through him then and why would I have to go through him now? You can’t get high off an asthma inhaler; I’ve never heard of anyone being addicted to them; the worst side-effect is the potential for slight light-headedness, or insomnia if taken before bed — but people are smart enough to deal with these themselves. I expect some people to object that if albuterol inhalers were available over the counter, we would expect body-builders and weekend warriors and high-school athletes to seek them out as minor performance enhancers. But so what? All these groups already down red bulls and b-vitamins and protein shakes with dopamine-precursors in them as performance enhancers. These are all at least as risky (which is to say, not very risky at all). The risks of albuterol-inhaler abuse are very low, so I don’t see why we as a society have a strong interest in adding to our collective health costs in order to stop amateur athletes from seeking a little extra broncho-dilation.

Probably the most obvious and salient example of a thing that should be available over the counter is contraceptives. To go back to our theoretical scheme: most people use contraceptives for most of their sexually-active lives, whenever they are not eager to conceive children. More, oral contraceptives are very chemically simple and the pill is very old, so we would expect it to be trivially inexpensive in a competitive and transparent market. There is (obviously) no real potential for abuse of contraceptives; they are trivially easy to use correctly. More, society as a whole has a strong interest in reducing the number of unwanted conceptions, and so we want to proactively make it as easy as possible for people to access contraception. Teenagers might feel embarrassed about having to talk to an adult doctor about sexual activity in order to obtain a prescription for oral contraceptives; others might just never get around to dealing with the hassle of arranging a doctor’s appointment. That has very serious, very bad consequences. Society has no interest in forcing healthy young people to further stress the time and resources of our doctors through trivial appointments to get their prescriptions approved and refilled periodically. In a sane world and a competitive market, $5 packs of a month’s supply of the pill would be available in Wal-Greens and CVSs right next to the men’s condoms. This is an absolute no-brainer.

My next example is one that I think people will instinctively find controversial but which I think our logic inevitably leads us to: I think basic SSRI (selective serotonin reuptake inhibitor) anti-depressants should be available over the counter. According to what I’ve read perhaps a majority of people will go through depressive episodes in their lives, whether prompted by seasonal change, disruption in social/kinship groups, grief, the existential ‘who am I?’ angst of college years, minor head injuries in car accidents, etc.. The fact that perhaps a majority of us will use them over the course of our lives means that we’re largely ‘prepaying’ for them; and because they’re pretty basic and many classes have been around for so long, we would expect many solid SSRIs to be very inexpensive over the counter. While sadness is a part of the human condition, paralyzing anxiety and depression that prevent one from doing anything about that sadness should not be. I’m sure there are many people who are experiencing depressive episodes, but are unwilling to see doctors due to the stigma of ‘mental illness.’ But they should be able to see, on their own, if SSRIs could help them out. SSRIs are old, safe, very well-tested, and really quite moderate medicines. They operate by inhibiting the reuptake of serotonin in the body, thereby increasing the supply of neurotransmitters available throughout the system. This can reduce anxiety in the short-term; over the medium-term, the increased supply of serotonin, in my understanding, causes the body to produce more receptors for serotonin and other neurotransmitters, elevating the body’s production of BDNF (brain-derived neurotropic factor), which is involved in the construction of new synapses and the elevation of mood. SSRIs have a lot of other potential uses; doctors commonly prescribe them to aid recovery from minor concussions for high-school football players, by preventing insomnia and aiding the repair of synapses, etc. There are even broad cognitive benefits associated with SSRIs’ elevation of BDNF — I can even imagine a reasonable person arguing that relatively healthy people should consider using them to enhance their lives.

Over the long term there are some withdrawal symptoms from SSRIs, just like there are withdrawal symptoms from caffeine, nicotine, intimate relationships, and sunshine… I think if we lived in a world with less stigma about ‘mental illness’ and drugs per se we would all have to concede that, rationally, SSRIs aren’t that scary. They don’t give their users a high; there’s little potential for abuse (or at least abuses that can’t be easily achieved with other available drugs). The major, well-known side effects of SSRIs are weight gain and sexual dysfunction — so, thanks to our universal vice of vanity, it’s unlikely that healthy young people would abuse SSRIs unless they were gaining some offsetting benefits. Would people use SSRIs recreationally? Maybe. So what? For all of human history we’ve been messing around with our neurotransmitters. The enlightenment was fueled by caffeine, which allows the brain to flood with serotonin and dopamine by blocking the action of adenosine, the sleepiness neurotransmitter. Caffeine has negative side effects, but most of us have learned how to consume just enough caffeine to make ourselves alert without making ourselves jittery or anxious. Nicotine operates as an MAOI (a monoamine oxidase inhibitor) which prevents the breakdown of neurotransmitters, and is thus not very different, conceptually, from a reuptake inhibitor. Another primary way in which we screw with our neurotransmitters now is through alcohol, which has its effect largely by acting as a GABA agonist (you can think of GABA as the neurotransmitter that turns parts of your brain off, causing disinhibition and a reduction in anxiety) and also through effecting dopamine and serotonin. But alcohol is also happens to be a poison, which kills brain cells, is frequently addictive and deadly, and not very good for our brains even in moderate doses. It seems likely to me that many alcoholics grow reliant on alcohol for anxiolysis; if such people could access the anxiolytic benefits of SRRIs without the stigma of seeing a psychiatrist, we might be able to prevent some very sad cases of alcoholism.

Because I’m a healthy young person, I have very little personal experience with medical treatment. This is why I’ve focused so far on treatments that I know about because they’re in the public eye. But I’m confident there are many many others — indeed, the more commonly used and known a drug is, the stronger is the case for offering it over the counter, because its very commonness (a.) puts it on the ‘prepayment spectrum’ and (b.) means that we’re all likely to be well-informed about it and (c.) is preliminary evidence that it’s not very dangerous. I’ll add to this list a few others: (1) Viagra — I doubt very much that one gets an addictive high from it (outside of the activity it enables); it is a great modern breakthrough cure for one of the great ailments of aging men everywhere, an ailment which has lamented since the time of the ancient poets; the symptoms of the dysfunction that it cures are, erm, unmistakable. There’s no reason to put the majority of older men, and a minority of younger men, through embarrassing appointments with their doctors, appointments which would consist solely of an unpleasant self-report and an expensive doctor’s bill. Just make it available over the damn counter. (2) Lots of basic antibiotics: As a kid I got ear infections (like a lot of kids) and I recall that it was burdensome to get its very basic but very effective treatment; urinary tract infections, etc., are a regular part of  the lives of a large portion of female population. People shouldn’t have to delay until they can get a doctor’s appointment to get the basic anti-biotics needed to treat these simple, regularly-occurring ailments.

There are a lot of other such things out there. As I conceded, my personal experience with medicine is limited. But whenever I do get interested in a medicine, I almost invariably come away feeling astonished by how much — and how expensively — we have all been babied by the FDA and the medical profession. My latest ailment is ruptured peroneal tendon. I’ve become interested in a very interesting and innovative therapy to help speed my recovery from the injury — LED Phototherapy, a.k.a Low-Level Laser Therapy. You can look up the research on it — it’s fascinating, effective, and completely free of negative side effects by every account I’ve seen. But only recently has a simple laser for home laser tendon therapy been approved for online sale. It’s still around $200. My mechanical engineer friends tell me they could make one these things in their garages in 30 minutes. I’ve read that until a few years ago, the FDA had approved the laser therapy for use on the wrist but not for use on the fingers. What is the possible theoretical basis of the idea that laser therapy could work on the joints in the wrist, but would be prohibitively dangerous for use on the fingers? It just makes no sense. These things should be available for $5 a piece at CVS by now…

***

So short executive summary: A major, insufficiently explored, politically orthogonal way to decrease medical costs in the U.S. would be to increase over-the-counter availability of a large number and variety of medicines. Making more treatments available over the counter would (1) decrease the demand for doctors’ time, lowering costs thusly, (2) force medical providers to compete in a real, free-market, lowering their prices, and (3) just generally make life easier and more convenient for all of us. There could be some minor downsides. But I think if we thought seriously about tradeoffs, we’d come to the conclusion that there are a lot of medicines for which the tradeoffs would easily be worth it.

Universal Access to Food [A bitter satire]

In May 2008, James Dapper graduated from Yale College and moved to New York City with a mission: To guarantee the people of the city universal access to food. While writing his senior thesis, James had discovered that there was not a single agency in the whole city devoted to providing food to all of the city’s residents. Nobody was in control of the provision of food to New York City residents! Why, for all we knew, people could be starving in the streets right now, James realized.

James took a job at the mayoral office, and, after several weeks of pushing paper, managed to sneak in a private meeting with the mayor. “Do you know,” James asked the mayor who sat across from him, “that this city has no agency or policy to guarantee universal access to food?” The mayor was shocked. The city government had 14 separate agencies devoted education; 35 devoted to celebrating New Yorkers’ various heritages; 11 devoted to the arts; and 49 devoted to medicine. But not even one devoted to food! What an oversight! How could this have happened?! “Food is essential,” the mayor said. “Sometimes I feel as if I couldn’t live without it. And here, without even thinking about it, I’ve been allowing the people of my city to go without food all of this time.” Dropping his head, he asked in despair, “How could I have been so stupid?” The mayor gave James an immediate raise, made him his leftenant, put him in charge of New York City’s new Department of Food and empowered him to run New York City’s new Edicare system, which was to provide edibles and care to the people of New York.

***

James was not, at first, sure how he would go about using his new agency to guarantee the people of New York food. But one thing was certain: letting people run wild and purchase their own food just could not possibly work. He had learned in his social psychology and cognitive science courses at Yale that individuals — especially individuals who did not go to Yale — were not rational. Left to their own devices they might eat too much of the wrong foods; or they might eat no food at all; or they might spend all of their paychecks right away, wasting all of their money on vacations or video games, and then find, a week later, that they had no money for food; or they might get ripped off in a market in which grocers could charge whatever they wanted. Because individuals are imperfectly rational, James realized, there was an obvious solution — have the government take control of the food supply.

But how?

James first idea was simple enough. He would compile a complete registry of every single individual in New York City, create a new agency of food-wonks who would use Big Data to determine everyone’s BMI, age, sex, musical tastes, exercise habits, etc., and consult with a panel of experts to determine the ideal diet for each individual. New York City would erect an enormous Food Warehouse on Randall’s Island, and, every morning, send out thousands of trucks to deliver to the people of New York their daily tailored quotas of food. For-profit Big Food grocers were no longer necessary.

But problems soon emerged in the project. The food warehouse construction project went way overbudget and way over-time; the food-wonks had trouble keeping track of all the data surrounding who had moved where, who was new to the city, who had moved out, who had died, whose BMI had changed, etc.; the food truckers were usually not able to reach all of their thousands of destinations per day, and they went on strike when they were asked to work overtime. It just wasn’t working; and the people of New York City began to protest the long waits they had to endure for the food trucks to arrive and the onerous taxes that had been levied to support the Warehouse and the pensions of the public truckers’ union.

James knew that this single provider model was the ideal policy in theory, but he soon realized — embittered, the idealism of his college days shattered — that it was simply implausible. The logistics were too complicated in a large overpopulated city like New York. The selfish, for-profit food vendors of the city were too powerful a lobby. The public was too unwilling to bear the tax hikes which were self-evidently in its interest. So after more consideration, he came up with a sophisticated but cynical compromise, which he thought would buy off the grocery lobby, reduce the governments’ logistical and administrative burdens, and guarantee near-universal access to food.

***

The idea was this: All New York City employers would be given massive tax incentives to offer their employees comprehensive food insurance. The employers would pay in, say, $1,500 a month per employee into food plans run by the new regulated food insurers, such as Eatna and the KaiserRoll Foundation. Eatna and KaiserRoll would themselves cut deals with the major grocers to allow their food-plan subscribers to use their food-plans to buy select, approved items from the grocers. Indeed, once New York City residents’ employers had paid the $1,500 a month premiums, the NYC residents would only be required to pay grocery $1 per visit in co-pays (Dapper’s Department of Food banned co-pays higher than that). There would be some tradeoffs here, of course. Because the major food insurers had cut deals with the major grocers, smaller grocers and potential competitors were not covered by the insurers’ plans and they went out of business. And a lot of employers went out of business because they couldn’t afford the premiums on the plans. But still. Citizens could have food for only a $1 copay per grocery visit!  Imagine that! A triumph!

***

However, as James became more involved in food policy, he began to discover dangers in the food supply, of which he had never previously been aware. A number of foods, he found, were dangerously untested; some had negative side effects; some had long-term health consequences. For example, after scientific review it was found that consumption of chocolate and coffee ice cream could both raise heart rates and blood pressure and cause insomnia. Left unregulated, Dapper concluded, people would cause themselves serious sleep and hypertension problems. Coffee derivatives and chocolate were immediately banned. Further research suggested that bananas had negative interactions with drugs like MAOIs — bananas were thus banned, too. Soup was too high in sodium — it could cause dehydration or high blood pressure — and thus was banned. Eggs were too high in cholesterol. Milk and cheese could cause gastrointestinal distress in adults. Nutella and Chinese food were found to be highly addictive, particularly for vulnerable populations like young university students. Why, for a certain portion of the population, peanuts and other nuts could even be literally deadly. James couldn’t believe, reading all of this research, that the poor people of New York had been exposed to all of these dangers without any authority taking control of the process.

James saw the clear conclusion: Edicare needed to create a new professional class of experts to control people’s food choices and steer them away from dangerous foods. Since food could be deadly, this class had to be extremely well-educated, the proverbial cream of the proverbial crop. They would be required to obtain a Food Doctorate (FD) degree from a prestigious university, by spending 7 years in food research, and obtaining certification from the board of the American Food Association (AFA). Then they would be put in charge of authorizing people to buy approved foods. In order to get peanut products at a grocer now, you would have to present a card and an I.D. proving that you were not allergic to peanuts. That was the simple and obvious solution for saving people from themselves. Think of all the otherwise inevitable peanut-butter deaths the Dapper administration had thus prevented!

Of course, with this protected certification cartel of highly-compensated F.D.s and their armies of administrators and secretaries now involved in everyone’s purchase of basic foods, premiums had to increase quite a bit. And then, of course, there was the fact that James Dapper’s Edicare required by law that food insurers and grocers be held liable for any accidents people suffered eating their food. So torts pushed costs up a bit more… One Olympic long-distance runner sued Eatna because the insurer had allowed him to access corned beef, which, he claimed, caused him to gain weight, miss out on an Olympic medal, and thus lose his shoe contract and source of livelihood. The family of one Long Island resident sued KaiserRoll after the father died of a heart attack (the father, they claimed, had been allotted too many hamburgers for his own good). One woman sued Eatna, because it had given her too much coffee, which made her jittery and caused her to write several tweets that she later regretted; another woman in turn sued Eatna for restricting her access to coffee, without which, she claimed, she was completely unemployable.

So while these legal liabilities and legal ass-covering expenses increased costs a bit (average food premiums now ranged around $3,000 a month), the process for getting food was now very simple and fair, in contrast to the dog-eat-dog chaos and disorder of the prior, completely unregulated market. You simply had to get full-time employment with an employer who would pay $3,000 a month on your behalf to a food insurance plan; then you had to schedule to get an appointment with an F.D., to receive written notes of permission to access the food you desired; then, you could go to any of the two or three grocery stores that were approved by your plan and pay a $1 copay, and present your I.D. and your food permission forms to the cashier, in order to get foods ranging from carrots, to wheat bread, to orange juice, low-fat turkey, to salmon and broccoli and nectarines and beyond! The grocer would pass on the true cost of your foods, which you never saw, to your food insurance plan, which would use that to set your employers’ premiums. Many New Yorkers remembered with horror the days in which grocers could change the prices of the foods they were offering at their profit-maximizing whim and would aggressively push dangerous foods such as coffee ice cream. Now, New Yorkers (or at least the dwindling minority of them who were still employed) had the comfort of knowing they would never pay more than a dollar and that all of their foods were safe and approved by the authorities.

***

To Dapper’s consternation, though, a few more problems emerged. First, because people only bore the cost of their $1 co-pays, a black market developed in which people began to resell some of their food to unemployed, uninsured people without food plans. This was clearly dangerous, Dapper and the AFA concluded, as the people buying in these black markets were surely naive to the potential dangers of food consumption. So the Dapper administration passed onerous new Rockefeller Food Laws, which would land New York City residents up to 20 years in jail for reselling food to those who lacked proper food prescriptions.

Then, food prices continued to skyrocket. Because individuals neither knew nor felt the costs of their foods, grocers began participating in arms races to patent new foods — most of which were rather trivial and small variations on long-extant foods, such as Churkey, a literal blend of chicken and turkey meat — in order to secure monopoly prices. Food insurers were not allowed to limit access to, or charge higher copays for, these novel patented foods under Dappercare. So, unbeknownst to the consumers themselves, their insurers were paying  thousands of dollars for Churkey when they (the consumers themselves) would have been happy to have bought turkey and chicken separately at much lower cost. The grocers naturally loved this, though. Un-patented foods were taken off of the shelves; grocers funded studies that concluded that the old, unpatented foods were now dangerous and untrustworthy; patented foods alone were vended; costs and grocer profits skyrocketed. While some right-wing think-tanks claimed that the root cause of the rising high costs of food (in contrast to the decreasing costs of goods in almost every unregulated market) was the lack of competition and transparency in the market and the fact that individuals did not know or feel any of their own costs, all Serious food experts at Respectable media outlets agreed that the real problem was Big Foods’ offerings of an unnecessary and proliferating panoply of superfluous and decadent new food patents. So James Dapper created a new Independent Panel to determine which foods New Yorkers really needed. After a substantial and costly review process, it was determined that risotto was unnecessary and costly; muffins were completely decadent; blueberry juice was debauched. The range of foods that grocers were allowed to supply was thus severely restricted. Some consumers grumbled, but the growers and suppliers of foods that passed the Independent Panel’s scrutiny were very happy indeed — now that grocers had fewer legal substitutes for their products, they could raise their prices still further.

***

New York City under James Dapper’s Edicare was vaunted as one of the great triumphs of modern history. By 2013, everyone in the city, provided they had an employer who was willing to pay just $3,000 a month for their Edicare premiums, now had guaranteed access to 20 carrots, three gallons of soy milk, and 15 loafs of low-glycemic gluten-free bread, and two chickens every single month without exception, provided they attained and maintained approval for these purchases from a certified F.D. at their monthly required F.D. checkups. Sure, the unemployed had almost no way to access basic food; sure, some people noted that New York City residents’ incomes and standard of living were now disappearing into the skyrocketing costs of food coverage; sure, many employers began completely freezing new hiring or left New York entirely, given how expensive food-insurance premiums had become. Sure, haute cuisine and the emerging creative experimental foodie culture were crushed (but these were just for privileged yuppies anyways). But the important thing is that people were not left alone, vulnerable to the predatory whims and dangers of the for-profit grocers, as they once had been. Full-time employed New Yorkers finally were able to go to sleep at night knowing that they would wake up the next morning with guaranteed access to food.

***

Of course, there was a right-wing reaction against this march of progress, as James had expected. At Yale, he had often been warned that, beyond the gates of his residential college, there were lots of people out there who just couldn’t stand the sight of progress. And now these people were emerging, as it were, from the compost bins of America. As The New York Times’ main municipal policy reporter wrote,

While scholars and progressives have vaunted Dappercare as the first municipal program in history to guarantee access to food, several controversial industry-funded think-tanks and astro-turf protesters are trying to roll back Edicare’s universal food coverage. For example, the Competitive Food Institute (CFI), a controversial right-wing think-tank backed by a number of ultra-wealthy hedge-fund managers, released a report which noted that the “effective cost” of food on New York City residents was now approximately 10 times what it had been prior to Dappercare, and that the “consolidation of the industry and destruction of smaller competitors” under the “maze of laws and regulations” of Dappercare could drive prices even higher. The CFI has advocated a complete elimination of the Department of Food, a policy that most citizens polled regarded as extreme, and argued that individuals should be forced to “pay for their own food”; the CFI has advanced an untested and abstract theory that competition would drive down costs of food to the point where ordinary people could afford it on their own. But progressive scholars have noted the clear regressive implications of such a policy. For example, because men have higher average incomes than women, if people were forced to pay for their own food,  women would have less access to food than men would. Similarly, young unemployed people (unemployment has notably recently skyrocketed, though the causes of this are controversial), in particular, would have trouble affording as much food as employed 40-somethings.

The New York Times story soon went viral among young web-savvy activists. It traded on Twitter with hashtags like #EqualFoodAccess and mottos like “Don’t roll back food progress.” On Facebook, people shared photos in which they held banners with mottos like, “I am a citizen; I am a father; I am on Dappercare,” and “I am a fireman: You rely on me; I rely on Edicare,” and “I am for progress; I am for security; I am for the Department of Food.” The viral phenomenon silenced the Competitive Food Institute and its donors were stigmatized and uninvited from all the cool parties and the few and dwindling opponents of Dappercare were shamed into the silence.

Progress was secured. Fin.

***

***

***

Dear readers, as you can no doubt sense, this is an extremely bitter, market-leaning, satire of U.S. healthcare. But I do not want you to take away from this satire that I think the market for health insurance is just like the market for food or that American healthcare would benefit from complete deregulation. Rather the purpose of this satire is to apply the dumb arguments people make about healthcare in another, more familiar, context — the context of the purchases we make at grocery markets every day — in order to force us off our cliched arguments and make us think more imaginatively about healthcare. While the market for food seems to work very well as a largely deregulated market with some basic subsides for the very poor (food stamps), there are well-known market failures that make health insurance very different. Notably, problems of information asymmetry, unpredictable but extremely costly risks, the existence of people who tragically have pre-existing conditions, etc., mean that the market alone will not deliver ideal healthcare. The purpose of this satire is not to deny those facts, or to deny the measures we need to take to solve those problems, but to make the point that “the public needs access to x” is not a good argument for “the government must manage the market for x” — in most situations in life, markets are very good at delivering to us the things we need access to. Usually, individual people meet their needs by making their own choices among competing suppliers  in an open market — rarely can central authorities manage the informational and political barriers to improving on this process.

We do, in my view, require policies that will provide subsidies/vouchers to the very poor; we do need ways to socialize the risks of unpredictable accidents; we do need to collectively cover the exorbitant costs of those who tragically suffer from true pre-existing conditions. Otherwise, however, most of our healthcare system would benefit from reforms oriented around transparency, individual choice, and competition, which would make it more like the markets for food and computers and air travel, deregulated industries which have massively improved over the past decades, unlike healthcare. More drugs should be available over the counter even if they have the (horrors!) side effects of gastrointestinal distress or drowsiness. Individuals should be trusted more to take their own risks and bear the consequences of their own free choices (i.e.: we need tort reform). Credentialed, certified doctors do not need to control access to basic antibiotics or contraceptives. We should all pay out of pocket for drugs that are helpful but not medically necessary, like Ambien, just as we do for other ordinary goods. We shouldn’t let outraged and unserious culture warriors who are focused on frankly trivial components of healthcare to distract what is actually the most pressing policy issue for middle-income people in America today — controlling rising costs. My last post on Singapore’s healthcare system outlines some general principals for how we can socialize major risks while still incentivizing individuals to control costs by making them feel their costs.

Additionally, this is not a satire against Obamacare. Most of the insane policies satirized in this post predated Obamacare; Obamacare even attempts to solve one of the problems highlighted in this essay, the fact that health-coverage is perversely linked to employment. Obamacare also wisely solves the basic problem of information asymmetry in the market for health insurance via its individual mandate. But Obamacare did fail to address many of the root ills that are driving the exorbitant rising costs of U.S. healthcare, including lack of transparency, weak incentives on individuals to control costs, too-limited over-the-counter access, restrictions on the agency of nurse practitioners and pharmacists, legal and regulatory complexity that snuffs out potential upstart competitors, the FDA’s over-conservatism toward drugs with mild side effects, companies’ inefficient patent-law arbitrage, the power of the doctors’ cartel, etc.

Singapore’s Healthcare System

I’m going to caveat this whole post by saying that health policy is not my expertise. (I spend a lot of time reading about economics and policy, and still would have trouble fully explaining the structure of health-care provision in the U.S. — but maybe this is part of the problem with U.S. healthcare?) But I’ve read a number of attractive things about Singapore’s healthcare system, and so I wanted to share my understanding of, and takeaways from, its Platonic ideal.

The basics, as I undertand them, are this:

First, everyone in Singapore has health-savings mandatorily and automatically deducted from their paychecks and placed into high-interest accounts. Since most people’s health expenses are low when they’re young, most people quickly accumulate a substantial buffer of health savings, which continue to compound over time.

Second, when it comes time to go to the doctor, you can pay for many, but not all, things out of this ‘Medisave’ account. Most medically necessary interventions and prescriptions qualify. Checkups for minor and non-life-threatening ailments or prescriptions for drugs that are helpful, but not actually cures for dangers-to-be-insured-against (e.g., an Ambien to help with jet-lag on international travel), might not be. This ensures that people don’t burden their health savings too much with their neuroses and sniffles, but also ensures that, when medical interventions *are* necessary, the money is there. It also requires medical providers to lower their costs to a point where they can actually attract demand in a free market — e.g., if people have to pay the full cost of Ambien, rather than a meaningless copay, you have to lower the price to a point where it’s worth it from an individual’s perspective.

Third, very interestingly, you can ‘donate’ some of your accumulated medi-savings to your family members. This increases your incentive to keep saving more and more and not overspend even if you are precociously and congenitally healthy, and provides an extra line of support to those who are congenitally and precociously unhealthy, provided that they have loving families with some healthy members. (It’s also interesting and heart-warming to me, because in economics we usually think of incentives as working on individuals, but this is an example of incentives working on the ‘extended self’ of a family. It also provides an extra level of ‘socialization of risk’ at the extended family level.)

Fourth, the government offers very low-cost and subsidized catastrophe insurance. This catastrophe insurance is ‘means-tested,’ meaning that if you have a million dollars of wealth lying around, the catastrophe insurance might not pay out even if you get in a car accident that runs up to $40,000 of medical expenses — because while your accident was tragic, you can plainly pay for it yourself. But if you’re middle class and that same accident would bankrupt you and your lifetime Medisavings, the catastrophe insurance would cover it. Catastrophe insurance represents the most basic, important function of insurance — to socialize the risks of unpredictable, rare, and extremely costly events, so that people don’t have their lives ruined by events over which they have no control.

Fifth, there are basic subsidies for the very poor. For some people, the regular required Medisave and catastrophe-insurance contributions are quite costly, and they, and they alone, receive subsidies. This means that the most vulnerable members of society are supported in procuring healthcare, but the median consumer of medical services has no incentive to consume more than is rational from his own cost/benefit analysis. By targeting subsidies at the very poor, Singapore’s health-care system provides universal access without (as we do here in the U.S.) incentivizing the over-consumption of medical resources.

Sixth, the government makes the market for medical services more competitive by enforcing radical transparency. Healthcare providers are required by law to publish their prices for services, in order to enable and encourage people to shop around for bargains. The U.S. system is radically untransparent. If your child has an ear infection in the middle of the night, and you go to an overnight emergency room to pick up a basic antibiotic (which must be a highly dangerous and addictive drug, given that only AMA-certified mandarins with seven years of education are allowed to dispense it!), the doctor who scribbles her signature on the prescription may charge $500. But you never see that cost — it is absorbed by your insurer who incorporates it into the annual costs paid by your employer, which employer has its medical costs subsidized by the government. We are five or six levels removed from the actual costs of our medical decisions, and so it’s no wonder at all our expenses are so irrationally high.

Seventh, at a certain age, Singaporean citizens can pass on what they have remaining in their Medisave accounts into their savings or retirement accounts. That is, once they’ve made it through most of their lives, they are rewarded for their efforts to control costs and allowed to spend the cash on other needs and wants. This simply closes the circle of giving people incentives to keep their costs low and allowing them to make their own tradeoffs about medical vs. other goods.

***

This system seems pretty theoretically ideal. It guarantees universal access via subsidies for the very poor and a mandate to ‘Medisave’ on everyone else. It achieves the most basic, fundamental function of insurance via cheap catastrophe insurance. And it keeps the costs on the public very low by relying on strong incentives at the individual and family levels, price transparency, competition, means-testing, and the general principal that individuals ought to bear their own costs for most things. (Ideal theory suggests that it might also be optimal to provide extra incentives for preventive steps — e.g., subsidizing gym memberships to nudge us to be healthier, and less costly, later on. But given that real-world governments are imperfect and subject to corruption and capture, Singapore’s more basic, keep-it-simple-and-stick-to-the-fundamentals approach is probably a better template for real governments.)

Singapore’s system is based around recognizing realities and trade-offs which are unfortunately a “third rail” for politicians to speak of in the U.S. Namely, medical resources are scarce, and health is one good among many that we want to enjoy in life. So, yes, sometimes it is rational to not get this checkup and not to get that prescription. If people knew and felt the costs of their medical services, they would be able to make these trade-offs more rationally. More, insurance adds value when it actually insures — socializing the risks of the irregular, the unpredictable, and the unavoidable. (Auto insurance does not cover the cost of refilling our gas tanks, because that is not what insurance is for.) And the Singaporean healthcare system exemplifies this. I would like an Ambien the next time I travel to Asia, but ‘I would like x’ is not tantamount to ‘it is rational for x to be fully covered by insurance.’ It would be better for society as a whole if I would bear the full cost of my non-clinical sleep aid and if the company that makes the drug were forced to meet me at a price which I myself would be willing to pay.

***

One thing that struck me about Singapore’s healthcare system is that in popular political cosmogonies, we posit the ideals of ‘strong active government’ and ‘individual choice and competition’ in opposition to each other. But Singapore’s system could be seen as both more government-driven and more market-driven than its Western equivalents. It begins with a universal government mandate in order to provide a well-defined public good — but then relies on intense competition, individual choice, transparency, simple and understandable rules, and strong incentives, to keep costs low.

This is my way of saying that I think the popular political cosmogony is misleading, and we should have fewer conversations about ‘big government’ in the abstract versus ‘free-markets’ in the abstract, and keep our eyes on the goal of ‘efficient provision of public services’ while being open to intelligent combinations of government mandates and market incentives/competition in achieving that goal. It’s not useful to say that Singapore’s system is characterized by ‘bigger’ or ‘smaller’ government than the U.S.’s — it’s just smarter government.

Anat Admati’s simple proposal for stopping financial crises: Target debt, not assets

Obviously, the financial crisis of 2008, and the subsequent recession and anemic recovery, was a really big deal. Even if we bounce back to 3% GDP growth rates in this year and the next, the second-order and aggregate effects of the financial crisis will continue to drag on American and global economic growth for literally decades to come. Probably the biggest cost of the recession has been high unemployment among the young, which has prevented members of my generation from accumulating skills, experiences, and savings that they otherwise could have — skills, experiences, and savings that could have done much to contribute to our economic future.

So how can we stop a financial crisis like our last one from happening again? Well, to massively oversimplify, the last financial crisis happened because banks had taken out huge amounts of debt to buy assets whose values were tied to the housing market, and the housing market faltered, causing the value of those assets to decline, which left some financial institutions insolvent, fundamentally unable to meet their obligations to others, and all the panic and uncertainty meant that even fundamentally sound banks lost access to credit they needed to hold them over through the crisis. So how do we stop this from happening again? Well, most of the discussion has centered around regulating banks’ assets. Most people want more regulations and stronger regulators on banks asset purchases–passing regulations to require banks to take on less risk and giving regulators more authority to look at their balance sheets and make them change their asset allocations if they’re being too risky.

But there’s a theoretical problem with this line of thinking: Financial institutions really don’t like going bankrupt (though, notably, the policy of Too Big to Fail can cause a problem of “moral hazard” here). They really do their best to find assets that will increase in value over time. Plus, banks these days — for better or worse — employ a lot of the smartest people in the world — economists, physics and math PhDs, etc. — to model what’s happening in the economy, figure out the probable impacts on their assets, and use that to figure out how to help their bank prosper. And this means that it’s not realistic to expect that the next financial crisis will be averted because a few government regulators getting paid $120,000 a year go up to a few Goldman Sachs economists making $5 million a year, and say, “Hey, look, your assets are going to decline in value, and you’re going to go bankrupt,” and the Goldman Sachs economists will say, “Oh, crap, we hadn’t thought of that.”

If that sounds snarky, let me put it more formally: The value of an asset represents the market’s best assessment of the total discounted future expected returns of that asset. To say that “the value of these assets will decline in the future” is an inherently counter-cultural, quixotic, non-consensus prediction, because the market incorporates its predictions for the future into the current market value of assets. If regulators are smarter than the market and can predict the future better than the market can, then they all should have already made billions and billions of dollars doing their own trading by now. (They generally have not.) In other words, declines in the value of assets are by definition unpredictable — so giving regulators power to stop banks from buying assets that they (the regulators) think are unwise purchases will almost certainly not work. To illustrate this basic theory with actual history: In the mid 2000s through 2007, the Fed assured us over and over again that the housing market was no cause for concern — in late 2007, most economists did not think that the U.S. would enter a recession in 2008 (we were already in one at the time). Regulators will not predict the next financial crisis in advance, because financial crises are by their nature unpredictable and unpredicted.

So what else can we do? Instead of giving more power to regulators, could we give more power to formal, unbiased, conservative regulations about the kinds of assets banks can hold, i.e., requiring that they buy relatively higher amounts of very safe assets, like U.S. Treasuries? This is, in my view, a better line of thinking, but not the ideal primary policy approach. Indeed, one could argue that one contributor to the last financial crisis was, e.g., the requirement that banks hold a certain portion of AAA-rated assets, and the ratings’ agencies stupidly giving Mortgage-Backed Securities AAA ratings. Ironically, the fact that banks could formally meet some of their requirements for AAA assets by buying these MBS actually helped drive up the demand for, hence the price of, MBS, which could have occluded and distorted price signals about their riskiness. In other words, ultimately the “more regulation of asset purchases” idea falls to the same argument as the “stronger regulator power over asset purchases” argument — if we knew which assets were risky in advance, they wouldn’t be so risky. Another objection is that we as a society actually do want banks to do plenty of risky investing, in, e.g., innovative but young companies with uncertain but potentially awesome futures. The tech bubble of the late 90s eventually got overheated, but it’s basically a pretty great thing that American capitalism could hook up a lot of brilliant entrepreneurs in California with the money they needed to implement their crazy ideas to change the world. It’s not clear that we’d be better off as a society if more of that money had gone into pushing yields on U.S. Treasuries even lower.

So what do we do instead? The big idea that’s catching on in the econ blogosphere, and which I’ve been persuaded by, is that we ought to stop focusing on banks’ assets per se, and instead focus on how they finance those assets. One way to think about this is that, as I wrote above, we’ll never see the next big decline in asset values in advance — it will always, by its nature, be unpredictable — but we can increase the chances that the financial system will be robust through such a period. How could we do this? It’s simple: If banks financed more of their assets with equity, and less with debt, they would be able to suffer a greater decrease in the value of their assets without becoming insolvent. So we simply force banks to have more equity relative to their debts: we could do this by simply making them reinvest all their earnings (i.e., not pay out any dividends) until they met the desired ratio. This idea is being advocated most powerfully and vociferously by Professor Anat Admati, as in her new book, The Bankers New Clothes.

Let’s step back to make sure we’re all absolutely clear on the terminology here: If I’m a business, every purchase I make is formally financed by either equity or debt. When I first start my business, I invest $10,000 — that’s equity; when I get a $10,000 loan from a bank, that’s debt. When I spend that money to buy up office space and inventory, then I have $20,000 of assets, financed equally by debt and equity (meaning I have a ‘capital structure’ of 1 to 1). If I make $5,000 right away, then those profits count as new equity immediately, and so I have $15,000 of equity for $10,000 of debt. If I pay those $5,000 out to the owner (myself) as dividends, then those $5,000 are in my personal bank account, and longer on the company’s balance sheet, so the company is back to the 1 to 1 capital structure ($10,000 of debt and $10,000 of equity). If my office catches on fire and now my assets are worth only $10,000, then I now have 0 in equity, because I still owe $10,000 to my creditors. If I invite a partner to come share ownership of the company with me, his/her investment is new equity.

In the run-up to the financial crisis (and still today), banks were famously highly ‘levered’; Lehman Brothers’ assets were financed by some 30 times as much debt as equity. This is sort of like buying a house for $300,000, while making only a $10,000 down payment. What’s so bad about taking out all this debt? The problem is that, the more debt/less equity you have, the greater are your chances of bankruptcy. You legally have to pay off your debts regardless of circumstances (your debt does not decrease because you had a bad year) but your equity just goes with the flow of your assets. If my company has $100,000 in assets, with a capital structure of 1 to 1, and our assets then decline in value to $80,000, then that sucks for me and my fellow owners — our equity just fell from $50,000 to $30,000 — but we can still pay off all our debts and remain a going concern. But if we had financed our $100,000 in assets with a leverage ratio of 9 to 1 ($90,000 in debt and $10,000 in equity), then the same decline in the value of our assets would leave us completely insolvent.

When banks are levered up 30 to 1, just a 3% decline in the value of their assets can leave them insolvent, unable to meet their obligations. When lots of banks are levered up this much, even smaller declines in the value of their assets can put them at risk of insolvency, which can, in turn, force them all to sell off assets in fire-sales, pushing down the value of financial assets even further, or cause them to lose access to credit, leading to a self-fulfilling prophecy, financial contagion, and a credit crisis necessitating bailouts, etc. In other words, each bank’s leverage has negative “externalities” on society as a whole.

***

Why do banks take out all of this debt? There’s one fact everyone agrees on: One major contributor is the debt bias in the U.S. tax code. Corporations can deduct the interest they pay on their debt for tax purposes, while they cannot deduct the dividends they pay out to shareholders — indeed, dividends get taxed twice, first as corporate profits and then as income for the owners who get them. This debt bias gives banks a relatively greater incentive to take out more debt. It also means, unfortunately, that if we did undertake Admati’s proposed reform without getting rid of the biased tax incentives against equity, banks would see their costs of funding rise, which could increase the cost of credit throughout the economy. (N.B.: She does want us to get rid of the debt bias as a part of her proposed package of reforms.)

But what if we could get rid of the debt bias? Then could we all agree to increasing banks equity-ratio requirements? This is where the discussion gets tricky and contentious. A lot of bankers are arguing that even if we could get rid of the debt bias, higher equity-ratio requirements would be a bad idea, because they would decrease banks’ Return on Investment (ROI), and hence their value. Think of it this way: Suppose I invest $50 million in a bank, and the bank gets another $50 million in loans, and buys $100 million in assets, which appreciate, over the year, to become worth $120 million. The bank needs to pay back $55 million to its creditors ($50 million plus 10% interest), but the other $65 million is all mine. I make a 30% ROI, even though the bank made only a 20% return on its investments, because the bank was levered up. If it weren’t so levered up, I wouldn’t make as much. If the bank had funded all of its assets with a $100 million investment from me, then I would only get a 20% ROI.

And this is definitely, obviously true — when a company is doing well, leverage multiplies the amount it can return to its shareholders, particularly when interest rates are low. The problem is, when the company is not doing well, leverage multiplies how much the shareholders get hurt. There’s a formal mathematical expression of this idea which proves that (in the absence of tax biases), the capital structure  of a company is irrelevant to its value. The math is hard to express, but here’s an easy way to think about it: Suppose a company has a very reliable business model, and so it’s thinking about levering itself up an extra two times, in order to increase the take-home ROI of its owners.  This isn’t a horrible idea, but it’s also not necessary, for a simple reason: If the investors have faith in the company’s reliability, then they could just lever their own investments in the company up, taking out debt to increase their equity stakes, which would have the exact same effect on their take-home ROIs. So the debt-equity capital structure/ratio is irrelevant to the company’s value to its shareholders — it just shifts around the risk.

***

One last quick note: A bedeviling misconception is the language that suggests that higher equity-ratio requirements mean that banks will have to ‘hold’ more equity, which will decrease their ability to lend, hence the supply of credit in the economy. This is totally insipid and false. Banks’ loans are assets — equity vs. debt are the way of financing those assets. Banks do not ‘hold’ equity. As soon as I invest in a bank, it can lend that money out. Banks ‘hold’ reserves as the Federal Reserve — but this is not at all affected by, and has nothing to do with, their equity. Admati’s proposals have nothing to do with how much cash banks have to keep in the bank.

***

So here’s a three-step process to make our financial system ten times as safe as it is right now:

(1) Get rid of the debt bias in the U.S. tax code.

(2) Require banks to have equity ratio requirements of 20%. An easy and orderly process for getting banks to reach this level would be to forbid them all from paying out dividends (i.e., requiring them to reinvest all of their earnings) until they reach that level.

(3) Let banks make all the risky investments and chase all the profits they want — and next time their bets don’t work out, let their shareholders, and not the U.S. taxpayers of the financial system as a whole, bear the cost.

Here’s all the interesting stuff in Nate Silver’s The Signal and the Noise

I’ve been immersing myself in statistics textbooks and software recently, as a part of a class and my general career interests. So over a weekend ski trip, I took on a lighter version of the work I’m doing by reading Nate Silver’s The Signal and the Noise: Why So Many Predictions Fail—but Some Don’t. Silver has been thoroughly and well-reviewed since his book was published shortly after the presidential election. So I won’t need to introduce him or the basics of what he does. My post will just highlight some of the more interesting, surprising, and difficult-to-articulate stuff in the book, particularly those that are related to topics in economics we’ve already discussed.

***

At the heart of the book is a powerful and important idea: The truth about the world, as best as we can understand it, is probabilistic and statistical, but we humans are unsuited to statistical and probabilistic thinking. What does this mean? Let me give a couple of examples. People say things like, ‘Uncle Lenny died of cancer because he was a smoker.’ The unjustified certainty with which we use the word ‘because’ here reveals a lot about how we think. After all, a large fraction of us—smokers and non-smokers alike—will die of cancer. We know with certainty that smoking statistically increases one’s risk of developing cancer, but we can’t say for sure that Uncle Lenny in particular wouldn’t have developed cancer if he weren’t a smoker. A more rational thing to say would be ‘Uncle Lenny died after being long weakened by cancer. As a smoker, he was more likely than the general population to contract cancer, and so it’s likely that his smoking was a significant contributor among other risk factors to his development of a cancer that was sufficient to contribute to his death.’ But that lacks a certain pith. See the problem? We all know that the underlying reality is that a wide variety of different risk factors contribute to and explain cancer, but we humans like to trade certain and definitive statements about linear causation, rather than thinking about a complex system of inputs that take on different values with different probabilities and interact with each other dynamically to produce distributed outputs with certain probabilities. In other words, we humans like to reason with narratives and essences, but the truth of the world has more to do with statistical distributions, probabilities, and complex systems.

Other examples of essentialist thinking are: When we have a hot summer, we often say that it was caused by global warming; on the other hand, global-warming deniers will say that we cannot make any such attribution because we had hot summers from time to time even before the industrial revolution. The most realistic thing to say would be, “global warming is increasing our chances of experiencing such a hot summer, and thus the frequency of them.” Another example: people will say that, “Kiplagat is an excellent long-distance runner because he is a member of the Kalenjin tribe of Kenya.” This ‘because’ is not entirely justified, but neither is the offense that sensitive people take to this claim, when they say things like, “Not every Kalenjin is a fast runner! And some Americans from Oregon are great runners, too!” The most precise way of putting the underlying truth would be, “Kiplagat is an excellent long-distance runner. He is a member of the Kalenjin tribe, which is well-known to produce a hugely disproportionate share of the world’s best long-distance runners, so this is one major factor that explains his ability.”

Why are we bad at probabilistic thinking, and locked into definite, essentialist, narrative styles of thinking? The axiomatic part of the explanation is that our brains have evolved to reason the way they do because these styles of reasoning were advantageous throughout our evolutionary history. We humans have been built as delivery mechanisms for our masters and tyrants—our genes. They encode instructions to make us and or brains work in the ways that helped them (the genes) get passed down through the generations, rather than working in ways that lead us to the strict truth. Probabilistic thinking takes a lot of information gathering and computational power—things that either weren’t around in our evolutionary history, or were costly luxuries. So our brains have evolved mental shortcuts or ‘heuristics’—ways of thinking that give us the major advantages of surviving and reproducing in the probabilistic world, without all of the costs. Our ancestors did not think, and we do not think, ‘Three out of the last five of our encounters with members of this other tribe have ended badly; so we can conclude with X% certainty that the members of this other tribe that we see here are between Y% and Z% likely to have a hostile attitude toward us.’ Rather, our brains tell us, ‘This enemy is evil and dangerous; either run away or fight—look, I’ve already elevated your heart-rate and tightened up your forearms for you!’ I.e., it gives us an essential claim and a definitive takeaway. In the modern age, public authorities say, ‘Smoking will give you cancer,’ which gets across the main takeaway point and influences behavior in important ways, more powerfully than ‘Lots of smoking generally contributes to a higher probability of developing cancer.’

Our brains are also wired to see a lot of patterns, causation, and agency when they aren’t there. As Silver notes, the popular evo-psych explanation for this is that it is more costly to erroneously attribute a rustling in the woods to a dangerous predator, and to take occasional unnecessary precautions against it, than it is to erroneously always assume that all rustling just comes from the wind, and get eaten alive when a predator actually does appear. Since missing a real pattern is more costly than falsely seeing an unreal one, we tend to see more patterns than there really are, and believe that we can discern a predictable pattern in the movement of stock prices, or get impressed by models that, e.g, predict presidential elections using only two small metrics, finding them more impressive than predictions that rely on aggregates of on-the-ground polling. Our basic innate failures at thinking statistically are reinforced by the culture around us, which accomodates/manipulates us (in good ways in bad) by appealing to our need for narrative approaches to understanding the world.

But now we live in the modern age. Our needs are different than they were in our evolutionary history, and our evolved psychology should not be destiny. We need to learn to reason more truly—which means probabilistically and statistically. Silver explores how we have succeeded and failed in doing this with examples drawn from baseball, online poker, politics, meteorology, climate science, finance, etc.

***

Why is it so important that we learn to reason probabilistically and statistically?  There are two main reasons. The first is very practical, and the second is more theoretical but ultimately very important. First, we obviously base our plans for and investments in the future around our predictions of what the future will be like. But the future cannot be known with absolute certainty, so we need to make rational decisions around a probable distribution of outcomes. For example, in chapter 5, Silver recounts an example of a flood which the public authorities predicted would rise to 48 feet—since the levees protecting a neighboring area were 51 feet tall, locals assumed they were being told that they were safe. But the 48 foot prediction obviously had a margin of error and, this time, it was off by more than 3 feet, and the levees were overrun. Given how dangerous it is to be in a flooded area, the local residents, had they understood the margin of error in the prediction and the probability of the levees being overrun, would have decided it was worth evacuating as a precaution—but they weren’t made to understand that the authorities’ prediction in fact entailed a range of possibilities. This is a very concrete example of the ubiquitous problem of reasoning, planning, and acting around a single true expectation, rather than weighting a range of possible outcomes.

Another example of this is how climate scientists don’t feel like they can give probabilistic statements to the public, like, ‘The most likely outcome is that, on our current path, global temperatures will rise on average 2 degrees Celsius over the next 100 years, and we have 90% certainty that this increase will range between .5 and 3 degrees. Additionally, we fear the possibility that there could be as-yet-imperfectly-understood feedback loops in the climate which could, with 5% probability, raise temperatures by as much as 8 degrees over the next century–while the chance of this is low, the potential costs are so high that we must consider it in our public-policy responses. Additionally, the coming decade is expected to be hotter than any in the last 100 years, but there is a 10% possibility that it will be a cool decade, from random chance.’ The public—you and I—are not good at dealing with these kinds of probabilistic statements. We demand stronger-sounding, definitive predictions—they resonate with us and persuade us, because they’re what our brains are comfortable dealing with. And a lot of the confusion in public debates surrounding scientific matters comes from our demand for definitive answers, where science can only offer a range of probabilities and influences. Climate scientist Michael Mann was quoted in the book as saying, “Where you have to draw the line is to be very clear about where the uncertainties are, but to not have our statements be so laden in uncertainty that no one even listens to what we’re saying.”

But the second, more fundamental reason for why we need to get better at probabilistic prediction is that offering and then testing predictions is the basis of scientific progress. Good models, particularly those that model dynamic systems, should offer a range of probable predictions—if we can’t deal with those ranges, we can’t test which models are the best. That is, we as a society would be ill-advised to say to climate scientists, ‘You predicted that temperatures would rise this decade, but they didn’t—neener neener.’ Rather, we should be savvy enough to understand that there’s a margin of error in every prediction, and that the impact of some trends can be obscured by random noise in the short run, and so the climate scientists’ claim that temperatures are rising is true even if it did not appear in this particular decade.

The rest of this post consists of some of the more interesting of the book’s ideas about statistical reasoning, and some of the barriers thereto, after a brief discursion on economics.

***

I’ve written in the past about the Efficient Market Hypothesis and about the value of short-selling, so it piqued my interest when Nate made some interesting points that related the two. One challenge Nate presents to the EMH is the two ‘obvious’-seeming bubbles that we have experienced in recent memory—the late 90s tech bubble, and the mid-2000s housing bubble. Now, it’s obviously very easy to call bubbles in retrospect, with the benefit of hindsight. But let’s accept for the sake of argument that we really could have seen these bubbles coming and popping—in the 90s, P/E ratios were hugely out of whack, and in the 2000s, housing prices had accelerated at rates that no underlying factor seemed to explain. The question is, why didn’t people short these markets and correct their exorbitant prices earlier?

Well, part of the problem is that in certain markets it can be difficult to accumulate a large short position without huge transaction costs, sufficient to move prices to a more rational level. But Silver’s more interesting argument is that institutional traders are too rational and too risk-averse relative to their own incentives. Counterintuitive, right? What does Silver mean? Let’s imagine that we’re in a market that looks a little overheated. Suppose there’s a market index that currently stands at 200, and you’re an analyst at a mutual fund and you think that there’s a 1/3^rd chance that the market will crash to an index of 50 this year. That’s a big deal. But there’s still a 2/3^rd chance that the party won’t stop just this year, and the market index will rise to 220 (a 10% return—not bad). In this scenario, the bet with the highest expected return is to short the market, a bet with an expected return of about $.18 on the dollar ( (1/3 * 150 – 2/3 *20) / 200). Going long in the market has an expected loss of the same. So if your goal is to maximize your expected return you go short, obviously.

The problem is, institutional traders don’t have an incentive to maximize their expected return, because the money they trade is not their own. Their first incentive is to cover their asses, so they don’t get fired. And if, in this scenario, you prophecy doom and a market crash, and short the whole market two years in a row, while the market is still rising, you’ll have a lot of outraged clients and you will get fired. And that’s the most likely outcome–the 2/3rds probability that the bull market will continue, and return another 10% this year. If you go along with the crowd, and continue to buy into a bull market that becomes overpriced then, well, when the music stops and the bubble pops, you’ll lose your clients’ money, but you won’t look any worse than any of your competitors. So this may be why a lot of bubbles don’t get popped in good fashion. It’s not that institutional investors are irrational—it’s that they are being rational relative to their career incentives, which are not well-aligned with market efficiency as a whole.

What’s the solution to this problem? Well, part of it is to get more really good short-sellers. One interesting tradeoff here is that the market is most efficient when people are (1) smart and (2) putting their own money on the line. Right now, we’re seeing a transition in which mutual funds and such are becoming more and more common, and so a larger portion of trading that is done in financial markets comes from institutions rather than from individual retail investors. These institutional traders may be smarter than independent retail investors, but they’re not betting their own money, which means their incentives are not well-aligned with market efficiency—the mutual fund’s first incentive is to avoid losing clients, who will bail out if the fund misses out on a bull market in the short term. So institutional investors will face a lot of pressure to keep buying into bull markets even when they know better. In short: don’t expect bubbles to go away anytime soon.

***

Silver discusses some of implications of the fact that predictions themselves can change the behavior they aim to predict. This is particularly pertinent in epidemiology and economics. For example, if the public authorities successfully inform the public that, this year, the flu is expected to be especially virulent and widespread in Boston, Bostonians will be especially inclined to get vaccinated, which will then, in turn, cancel the prediction. So was the prediction wrong? Maybe, but thank God it was! In economics, if the economics establishment sees that some developing country is implementing all of the ‘right’ policies, it will predict lots of economic growth from that country—this will cause a lot of investments and optimism and talent to flow into that country which could ‘fulfill’ the prediction. On the most practical level, this means that in these scenarios it’s very difficult to issue and then assess the accuracy of predictions. On a philosophical level it may mean that a perfect prediction that involves human social behavior may be impossible, because it would require a recursive model in which the prediction itself was one of the inputs.

A lot of this reasoning here raises a moral quandary. Should forecasters issue their predictions strategically? We know that public-health authorities’ predictions about how bad a flu outbreak will be will influence how many people get immunizations. The Fed’s predictions about the future of the economy influence companies’ plans for the future, which plans can then fulfill the Fed’s predictions (i.e., if a company is persuaded by the Fed that there will be an economic recovery, then it will ramp up its production and hiring right now, in order to meet that future demand, which will help fulfill that prophecy). Should these and similarly situated agencies therefore issue their predictions not descriptively, but strategically, i.e., with an eye to influencing our behavior in positive ways? In practice, I assume the agencies definitely do. The Fed has consistently optimistically over-predicted the path of the U.S. economy since the financial crisis. This is embarrassing for it, but any cavalier expression of pessimism from the Fed very well could have tilted the U.S. into a double-dip recession. The obvious problem is that when public agencies make their predictions strategically rather than descriptively, they could, over the long run, dilute the power of their predictions in the eyes of the public—i.e., people might start to automatically discount the authorities’ claims, thinking “this year’s outbreak of avian flu, much like last year’s, will affect 10^3 fewer people than the authorities suggest, so I don’t actually need to get a vaccination.”

***

Silver offers a lot of helpful reminders that rationality requires us to go beyond ‘results-oriented thinking.’ On televised poker, for example, commentators praise the wisdom and perspicacity of players who bet big when their hands weren’t actually all that strong, statistically speaking, and who win either because (1) they caught a break on the last cards dealt (in Texas hold’em style) or (2) they were  lucky enough that everyone else had even weaker hands. But while commentators may praise these players’ prescience, we should call these bets what they are—dumb luck. We shouldn’t evaluate people’s decisions after the fact using perfect information, what we know now. We should evaluate how rationally they acted given the information they had access to at the time. And betting big with a weak hand, without any information that other players’ hands are even weaker, is never the smart or rational thing to do—even though it will luckily pay off in some chance occasions.

***

‘Big Data’ is a modish term right now. An essayist in Wired claimed a few years ago that as we gain more and more data, the need for theory will disappear. Silver argues that this is just the opposite of truth. As the amount of information we have multiplies over and over again, the amount of statistical ‘noise’ we’ll get will multiply exponentially. With all this data, there will be more spurious correlations, which data alone will not be able to tease out. In the world of Big Data we’re going to need a lot of really sound theory to figure out what are the causal mechanisms (or lack thereof) in the data we have, and which impressive-seeming correlations are spurious, explained by random chance. So theory will become more important, not less.

***

One big takeaway for me, as I read Silver’s accounts for how statistical methods have been applied to improve a variety of fields, is that we are very easily impressed, sometimes intimidated, by mathematical renderings of ideas, but statistics really is not rocket science. The computations that statistical software can do at first seem complex, but they’re all ultimately built on relatively easy, intuitive, concrete logical steps. Same with models: the assumptions on which we build models, and the principles we use to tease out causation and such from within the wealth of data, are ultimately pretty intuitive and straightforward and based in basic logical inference. In reading Silver’s account of the how the ratings agencies got mortgage-backed securities wrong in the run-up to the financial crisis, I was astonished by just how simple the models the agencies were using were. That is, even those of us who like to bash the financial sector still tend to assume there’s some sophisticated stuff beyond our ken going on in there. But Silver reports, for example, that the ratings agencies had considered the possibility that the housing market might decline in the U.S., but continued to assume that defaults on mortgages would remain uncorrelated through such a period. The idea that mortgage-defaults would always exhibit independence—and that the rate of default as a whole could not be changed by global economic conditions—is flatly ridiculous to anybody who takes a moment to think imaginatively about how a recession could affect a housing market. But because the ratings agencies’ ratings were dressed up in Models based around Numbers on Spreadsheets, Serious People concluded that they were Serious Ratings. A lesson for the future: Don’t let yourself be bullied into denying the obvious truths or accepting obvious falsehoods just because they have been formulated in mathematical notation. A seemingly sophisticated mathematical model is in truth a very bad one if its basic assumptions are incorrect.

The lesson here is not that we should eschew statistical methods—it’s that we should get in on the game and improve the models, instead of being cowered by the people who wield them. Indeed, another striking part of the book was Silver’s admission that his own famous political-prediction model on his Five Thirty-Eight blog is not terribly sophisticated—it’s only been so successful because everyone else’s standards in the political world have been so low. And the statistical methods that revolutionized baseball drafting and trading, as recalled in Moneyball, weren’t that sophisticated either—they were just low-hanging fruit that hadn’t been eaten yet.

***

The more polemical parts of the book center on Silver’s righteous claim that pundits be held to account for their predictions. Silver points out that political pundits, like those who appear on the McLaughlin group, regularly get their forecasts wrong in very predictable ways, and never get called out on them or punished. As one who, like Silver, gets angry when people make plainly descriptively untrue statements about the world, I did enjoy his righteous outrage. But I think that in this, he (and I) get something basically wrong—namely, being a political pundit and appearing on the McLaughlin Group are fundamentally not truth-seeking activities, and so their failure to deliver truth should be completely unsurprising and probably doesn’t even qualify as a real indictment in the pundits’ minds. The goal of the people engaged in these activities is not to uncover the truth, but to root for their team. So of course the Republican pundits on McLaughlin group always predict Republican electoral victories, as the Democrats predict Democratic victories. That’s what they’re there for.

More fundamentally, I think Silver under-estimates how uncommon it is for people to think about the world in a descriptive truth-seeking manner. Most of us most of the time are not engaged in truth-seeking activity. Most of us typically choose the utterances we issue about the world on the basis of loyalties, emotional moral commitments, etc.. Thinking about the world descriptively is just not the natural mode for most people. When a Red Sox fan, in the middle of a bad season, says something like, “The Red Sox are going to win this game against the Yankees,” we shouldn’t actually take him to mean, “The Red Sox are certain to win this game” or even necessarily “The Red Sox have a better than even chance of winning this game.” Rather, the real content of his statement is better translated as, “Rah, rah, goooo Red Sox!”  For most people, statements that they phrase as predictions are not a matter of descriptive analysis of the world—they’re statements of affiliation, hope, moral self-expression, etc. The social scientific and descriptive mindsets are very rare and unnatural for humans, and if we’re going to get angry about people’s failures in this respect, we’re going to be angry pretty much all the time.

But I do agree with the basic takeaway from this polemic: Silver wants to make betting markets a more common, acceptable, and widely-expected thing. If we were forced to publicly put our money where our mouths are, we might be more serious and humble about the predictions we make about the future, which should improve their quality. I’ve long relied on Intrade to give me good, serious predictive insights into areas where I have no expertise, and do wish liquid betting markets like it, where I can gain credible insights into all kinds of areas, were more common and entirely legal.

***

A lot of expert reasoning goes into building a good model with which to make a prediction. But what about us general members of the public who don’t have the time to acquire expertise and build our own models? How should we figure out what to believe about the future? Silver provides some evidence that aggregations of respectable forecasters (i.e., those who have historically done very well) are almost always better than any individual’s forecasts. E.g., an index that averages the predictions of 70 economists consulted on their expectations for GDP growth over the next year does much better than the predictions of any one of those economists. So in general, when we’re outside of our expertise, our best bet is to rely on weighted averages of expert estimates.

But there’s an interesting catch here: While aggregates of expert predictions generally do better than any individual experts, this fact depends upon the experts doing their work independently. For example, Intrade has done even better than Nate Silver in predicting the most recent election cycles, according to Justin Wolfers’ metrics. So does that mean that Nate Silver should throw away his blog, and just retweet Intrade’s numbers? No. And the reason is that Intrade’s is strongly affected by Silver’s predictions. So if Silver were, in turn, to base his model around Intrade, we would get a circular process that would amplify a lot of statistical noise. An aggregation ideally draws on the wisdom of crowds, law of large numbers, and the cancelling-out of biases.  This doesn’t work if the forecasts you’re aggregating are based on each other.

Aggregations of predictions are also usually better than achieving consensus. Locking experts together and forcing them to all agree may give outsized influence to the opinions of charismatic, forceful personalities, which undermines the advantages of aggregation.

***

Nate argues, persuasively, that we actually are getting much better at predicting the future in a variety of fields, a notable example of which is meteorology. But one interesting and telling Fun Fact is that while meteorologists’ actual predictions are getting very good, the predictions that they are compelled to present to the public are not so strong. For example, the weather forecasts we see on T.V. have a ‘wet bias.’ When there is only a 5-10% chance of rain, the T.V. forecasters will say that there is a 30% chance, because when people hear 5-10% chance they think of it as an essential impossibility, and become outraged if they plan a picnic that subsequently gets rained on, etc. So to abate righteous outrage, weather forecasters have found it necessary to over-report the probability of rain.

Meteorologists’ models are getting better. We humans just aren’t keeping pace, in terms of learning to think in probabilities.

***

But outside of the physical sciences, whose systems are regulated by well-known laws, we tend to suck at forecasting. Few political scientists forecast the downfall of the Soviet Union. Nate attributes this failure to political biases—right-leaning people were unwilling to see that Gorbachev actually was a sincere reformer, while left-leaning people were unwilling to see how deeply flawed the USSR’s fundamental economic model was. Few economists ‘predicted’ the most recent recession even at points in time when, as later statistics would reveal, we were already in the midst of it. Etc., etc.

***

Silver points out that predictions based on models of phenomena with exponential or power-law properties seem hugely unreliable to us humans who evaluate these models’ predictions in linear terms. A slight change in the coefficients in the parameter can have huge implications for the prediction a model makes if it is exponential. This can cause a funny dissonance: a researcher might think her model is pretty good, if its predictions come within an order of magnitude of observations, because this indicates that her basic parameters are in the right ballpark. But to a person who thinks in linear terms, an order-of-magnitude error looks like a huge mistake.

***

Silver briefly gestures at a thing that the economist Deirdre McCloskey has often pointed out—that our use of ‘statistical significance’ in social science is arbitrary and philosophically unjustified. What is statistical significance? Let me back up and explain the basics: Suppose we are interested in establishing whether there is a relationship, among grown adults, between age and weight—i.e., are 50-year olds generally heavier than 40 and 35-year olds? Suppose we sampled, say, 200 people between 50 and 35, and wrote down their ages and weights, and then constructed a dataset. Suppose we did a linear regression analysis on the data, which revealed a positive ‘slope,’ representing the average impact that an extra year of life had on weight in the sample. Could we be confident that in general, for the population of people between 35 and 50 as a whole, this relationship holds? Not necessarily. Theoretically, there’s always a chance that our sample set is different—by pure chance—than the general population, and so the relationship in our sample cannot be generalized. There’s a possibility that the relationship we observed between age and weight is not a true relationship at all, but was just a matter of chance. And (as long as our sample was truly randomly selected from the population) we can actually calculate the probability of this possibility, using the data’s standard deviation and the size of our sample. In statistics, we call it the p-value, and a p-value of .05 means that there’s a 5% chance that a relationship observed in a sample is just an illusion, produced by chance. In contemporary academe, social scientists by convention will generally publish results with a ‘statistical significance’ of 95%–i.e., where the p-value is lower than .05. But applying this rule mechanically actually doesn’t make much sense. It means that today, a statistical analysis that produces a result with a p-value of .050 will get published, while one with a p-value of .051 will not, even though the underlying realities are almost indistinguishable. There’s no fundamental philosophical reason for setting our general p-value cutoff at .05—indeed, the basic reason we do this is that we have 5 fingers. In practice, this contributes to the rejection of some true results and the acceptance of some false results. If we accept all findings that establish ‘statistical significance,’ then we’ll accept a lot of false results. For example, if a journal publishes 100 research findings, all of which have a p-value of .03, passing statistical significance, we would expect that, on average, 3 of these findings would actually be incorrect, illusions of the samples from which they were built. (This is, by the way, after controlling for the possibility of the data being incorrectly obtained.)

***

On page 379, Silver has what is possibly the greatest typo in history: “ ‘At NASA, I finally realized that the definition of rocket science is using relatively simple psychics to solve complex problems,’ Rood told me.” (I am envisioning NASA scientists carefully scribbling down the pronouncements of glazy-eyed, slow-spoken Tarot-card readers.)

***

The final chapter in the book, on terrorism, was fascinating to me, because with terrorism, as with other phenomena, we can find statistical regularities in the data, with no obvious causal mechanism to explain those regularities. In the case of terrorism, there is a power-law distribution relating the frequencies and death tolls of terrorist attacks. One horrible feature of the power-law distribution of terrorist attacks is that we should predict that most deaths from terrorism will come from the very highly improbable, high-impact attacks. So over the long-term, we’d be justified in putting more effort into preventing e.g., a nuclear attack on a major city that may never happen, as opposed to a larger number of small grade terrorist attacks. Silver even argues that Israel has effectively adopted a policy of ‘accepting’ a number of smaller-scale attacks, freeing the country to put substantial effort into stopping the very large-scale attacks—he shows data suggesting that Israel has been able to thus ‘bend the curve,’ reducing the total number of deaths from terrorism in the country that we would otherwise expect.

***

But the big thing I was hoping to get from this book was a better understanding the vaunted revolution in statistics in which Bayesian interpretations and ideas are supplanting the previously-dominant ‘frequentist’ approach. But I didn’t come away with a sound understanding of Bayesian statistics beyond the triviality that it involves revising predictions as new information presents itself. Silver told us that the idea can be formulated as a simple mathematical identity: It requires us to give weights to the ‘prior’ probability of a thing being true; the probability that the new information would present itself if the thing were true; and the probability of the information presenting itself but the thing still being false. With these three we can supposedly calculate a ‘postperior probability,’ or our new assessment of the phenomenon being true. While I will learn more about the Bayesian approach on my own, Silver really did not convey this identity on a mathematical level, or help the reader understand its force on a conceptual level.

Overall, then, I found the book disappointing in its substantive, conceptual, and theoretical content. A lot of the big takeaways of the books are moral-virtue lessons, like, “Always keep an open mind and be open to revising your theories as new information presents itself”; “Consult a wide array of sources of information and expertise in forming your theories and predictions”; “We can never be perfect in our predictions—but we can get less wrong.” All great advice—but not what I wanted to get from the time I put into the book. The sections on chess and poker are interesting and good journalism, too, but they will do little to advance the reader’s understanding of statistics, model-building, or the oft-heralded “Bayesian revolution” in statistics, etc. But maybe I’m being a snob and wanting more of a challenge than a book could pose if it expected to sell.

–Matthew Shaffer

What good is short-selling? (Econ for poets)

If you follow the business press, you’ve probably seen the raucous unfolding story about Herbalife, a company whose share-price has tumbled then oscillated ever since Bill Ackman, the hedge-fund manager, took a short position in the stock a couple months ago. Ackman has alleged that Herbalife is actually a pyramid scheme — i.e., that its revenue primarily comes not from its sale of actual goods, but from its ‘multi-level marketing’ strategy in which its distributors recruit new individuals to sign up as distributors, and take a portion of that sign-up fee in return. That is, Ackman alleges that Herbalife distributors are only making money by taking one-off payments from new distributors, which is obviously not a sustainable business strategy over the long run (how will Herbalife’s distributors make money once all 7 billion people on earth have been recruited, if it can’t make money by actually selling its goods?). Ackman wants the authorities to investigate Herbalife’s business model. Others, like Carl Icahn, have come to Herbalife’s defense, saying that Ackman’s allegations are misplaced and, more, since these false allegations have unjustly driven the company’s stock-price downward, the company is now a very good buy.

This story will, no doubt (as is every story’s wont), continue to unfold. But I wanted to use this opportunity to explain and explore the basic theory of short-selling in financial markets. Short-sellers don’t have a good reputation with the companies they target for short-selling, or with members of the public who think that short-sellers hurt the companies they target or profit from others’ losses or hurt the market. But I want to argue that short-sellers play a very valuable role. This post will have four basic parts: (1) I will explain what short-sellers do, emphasizing that they do not directly ‘take capital away’ from companies, and therefore do not directly hurt them. (2) I’ll argue that we as a society do not want the stock market just to go up and up as high as possible, but, rather, we want it to be correctly priced. In part (3) I’ll combine points (1) & (2) to argue that short-sellers play a valuable role. And (4) I’ll caveat my roseate view, and acknowledge and address some criticisms of short-sellers.

***

(1) What is short-selling? The basics: Suppose you have a very good reason to think that a stock is underpriced — that, all things considered, it will return more than the market rate in the future. What should you do? Obviously, you should buy it, which amounts to placing a bet on the stock’s rise. Colloquially, we call this ‘taking a long position’ in the stock. Now suppose that, after a few years, other investors fall in love with the stock, and, now, you think it’s overpriced. What do you? Obviously, you sell. But what if you see a stock that is overpriced, but you don’t own the stock in the first place? What do you do? It would obviously make no sense to buy it in order to then sell it — that would just incur two fees from your broker. So what can you do? Is there any way that you can bet on the decline of a stock’s price if you don’t own the stock in the first place (or bet on its decline beyond just selling off all of your shares)?

As you’ve probably guessed…yes, you can short-sell or ‘short’ the stock. How do I do that? Technically, when I short a stock is that I borrow it from someone else for a contracted time and at a contracted price (colloquially, we call this ‘taking a short position’ in the stock). This allows me to profit from the stock’s decline over the period of the contract. Here’s how it works: Say that stock in QWERT is trading for $100 a share. I could pay somebody else $10 to ‘borrow’ their stock for 1 year — if they expect the stock to rise, stay flat, or even only fall a little bit, then this is a great deal from their perspective. Then, I could immediately sell the stock at the market rate of $100. Then, at the end of the year, if the price of QWERT’s stock has declined to, say, $70 a share, I could repurchase the stock at this new, lower price, before returning it to the party I borrowed it from. So I paid $10 to borrow it for the year, sold it for $100, and then bought it back for $70 — I made a cool $20 while effectively investing only $10 of my own money for the year. (Modern markets are sufficiently sophisticated that I don’t actually write up individual contracts to borrow every stock I short — I can do it with a click of a button. But this transaction is legally happening somewhere underneath my click.)

If I’ve belabored this explanation a bit, it’s because I want to make clear a couple of key points: First, a ‘short position’ (just like a ‘long position’) is simply a transaction in secondary financial markets between consenting adult investors that doesn’t directly impact the capital that the company itself can access. A short-sale is the flip-side to any long position in a stock — long investors are gambling that the stock is underpriced, while short investors are gambling that it’s overpriced. What does this mean? Well, first it means that the common financial metaphors that compare short-sellers to sharks and predators are misleading. Short-sellers aren’t hurting other investors without their consent — when you borrow a stock to short it, your counterparty knows exactly what you’re doing, and makes a deal with you anyways, because s/he disagrees with your assessment. And short-sellers don’t directly harm the businesses they target (N.B. I’ll caveat this later). A company gets the equity capital that it needs in order to grow and function from its Initial Public Offerings and other direct share offerings. But as soon as a company sells shares to the public, the money it received on the sale belongs to it. So increases and decreases in the price that those shares trade for in secondary financial markets (i.e., fluctuations in the stock price) have no direct effect on the company’s store of and access to capital. To repeat: The equity that gets traded in secondary financial markets is completely distinct from the equity that is on the company’s Balance Sheet.

So what is short selling? Here’s another way to define it: It’s a legal transaction in which I’m a nice guy who takes the opposite side of two trades, paying a fee to a guy who wants to loan out his share for a year, and selling to a gal who really wants to take a long position in the stock. If I’m lucky, I make a profit on the trades. If I’m not, he and she do. The company’s day-to-day operations are usually completely unaffected by my trade.

***

(2) What do we want the stock market to do? Some theory: When I was a wee lad, before I understood basic financial theory, I thought of the stock market — as represented by the S&P500 index charted on TV screens and newspaper front pages — as a sort of agentic and determined creature, struggling admirably and valiantly and against adversity to move uphill. The S&P 500 chart was, I thought, a measure of the economy as a whole, and the higher it climbed, the better the world was. And wee-me was not the only one to think this way. Indeed, there’s interesting research at the intersection of cognitive science and financial theory that shows how even sophisticated financial commentators imbue their descriptions of stock prices with normative and agentic metaphors — an increase in stock prices is described as “the market vaulted to new heights,” while a decrease is inevitably written up as “another slip in a faltering market.” The basic metaphor that this language embeds and subconsciously conveys to us is: “The market is a self-willed agent, and it is an excellent thing when it ‘rises,’ and a sad thing when it ‘falls.’”

But this is not actually a rational way to think about the stock market. We don’t necessarily want stock prices to ‘climb’ higher and higher. Rather, we just want stocks to be priced correctly. Why? Well, there’s one very obvious and practical reason, and another less-obvious but more fundamental reason. The obvious reason is that when asset valuations just climb and climb, that causes a bubble, and bubbles usually pop, and cause a lot of instability and hell when they do. Bubbles are bad on the way up and on the way down — on the way up, I look at my stock portfolio and think I’m wealthier than I truly am and spend way too much; on the way down, I get upset about how much wealth I’ve lost and become risk averse and don’t buy enough. But is ‘popping’ the only problem with over-high asset valuations? What if we had a magic-wizard policy that could stop bubbles from popping by banning short-selling, etc., to keep bubbles permanently inflated? Would super-high asset valuations be a good thing in this magical world? Even here, economists would say ‘no,’ because even in this magic world, over-high valuations lead to a ‘misallocation of resources.’

What does this mean? Let’s explore a very simple model. Suppose that I have $100 in savings, which I’m considering investing in the IPO of PetApps.com, a new startup website that sells Apps that help your furry friend keep tabs on what other pets in the neighborhood are up to. (Yes, I’m being derisive.) It’s a ‘roaring’ ‘bull’ market that everyone wants a piece of, driving up equity valuations ‘through the roof.’ I realize that the PetApps.com IPO is overpriced relative to its true, fundamental value, but the market has so much ‘momentum’ that I can cash out of my investment while it’s still moving upwards. Should I invest in PetApps.com? Well, the answer depends on who’s asking the question. From my own selfish perspective, I should — I can make a profit by buying and selling quickly during this frenzy. But from society’s perspective, this investment would be a bad thing. Why? Well, when we say that shares of PetApps.com are ‘overpriced,’ we’re saying that, for their cost, these shares will not earn good returns in the future. In other words, capital invested in PetApps.com will not generate as much value as it would elsewhere; more colloquially, the management of PetApp.com is too incompetent to handle all that money wisely. That means that we would all be better off I invested my cash in some company that was undervalued, or in municipal bonds, or even if I just spent it on a vacation now, which would generate income for airlines, etc.

To generalize this thought experiment: At any given time, we have a lot of good options for what we can do with our money. Given that, we don’t just want to just put more and more value into any particular asset, because that would detract from the money that we could use for the other goods. Rather, we want to price each good correctly; this is what economists mean by ‘allocating capital efficiently.’ We as a society don’t just want company shares to sell for high prices per se during their IPOs — we want them to sell for correct prices, providing the company with exactly as much capital as it can use efficiently.

What about the secondary market (i.e, the buying and selling of stocks that you and I can do through E-Trade after the IPO)? As we noted above, trading in the secondary market does not directly effect the capital available to a company. So does it matter to the real economy? I think so. One way to think of secondary markets is as one big ecosystem that supports the ‘primary’ equity markets of IPOs. That is, primary investors in IPOs only invest in the first place because they are counting on the fact that they’ll be able to cash out by selling their shares, whenever they want, into a liquid market. If they made a wise investment decision during the IPO, investing capital in a company that went on to use it to do something transformative (like Apple), they’ll cash out into rising secondary markets, and make a killing. If they invested unwisely, wasting society’s scarce resources on Pets.com, they’ll lose a lot in secondary markets. Trading in secondary markets thus rewards and punishes investors for making efficient and inefficient investment decisions. It’s the ecosystem that is essentially supporting the basic business of getting good investments into good companies.

Also, companies often sell new share issues, well after their IPOs. Those shares will be sold at a price that reflects the total ‘market capitalization’ of company, calculated as the share price times shares outstanding (i.e., if your total market cap is $100,000, on 100 initial shares, and you issue 100 new shares, your 200 shares should now all sell and trade for $500 a piece, since the total value of the company should remain more or less unchanged). And so the same basic principles apply: We want secondary markets to price shares correctly, not highly, in order to prevent destabilizing bubbles, and to properly reward and punish good and bad allocation of capital. A rising S&P 500 is thus only a good thing if the S&P 500 had formerly been undervalued; a declining S&P 500 is a good thing if it had been overvalued. This is why the normative and agentic metaphors that our financial commentators use for the ‘climbing’ and ‘slipping’ market are problematic and misleading.

***

(3) Does short-selling help the market do what we want it to do? An argument: So let”s put the pieces of the puzzle together. How do short-sellers help support the economy? Well, the most obvious and commonly cited way is that they help provide ‘liquidity.’ If you want to buy a stock in financial markets, you’ll need to buy it from a seller — often a short-seller. But the more fundamental good they provide is that they help correct over-valued stocks. Recall that a short-sale is only profitable if the stock price actually does end up dropping, as the short-seller predicts. Short-sellers, by entering the market and becoming sellers, increase the for-sale supply of a stock and decrease the demand for it, bidding its price down. Short-sellers also have an incentive to publicly reveal their short position, to persuade other investors to drive down its stock price. Thusly do short sellers correct the prices of stocks that they believe are overpriced. This provides the indirect good of supporting the efficient allocation of capital, as discussed above. And often it has more direct, tangible benefits. Short-sellers have historically been very good — often better than the official regulators — at sniffing out and exposing accounting fraud and large companies. The threat of drawing the attention of short-sellers can help scare management teams (whose compensation is typically tied to the stock price) into behaving, being honest and transparent with analysts, not paying out lots of company cash for ‘consulting’ from shell companies they themselves (the managers) own, etc., etc. Short-sellers may be the best and most effective regulators the market has.

***

(4) Can there be abusive, harmful short-selling? Some caveats: I’ll admit that my perspective here is generally pretty positive about the value short-sellers provide, and I hope I’ve persuaded my readers to share this general feeling. But I think there are special and marginal cases in which short-selling can be harmful. What are these? The first is, most obviously, when a short-seller is wrong and deceptive about his evaluation of a stock’s true value. Suppose Bill Ackman is completely deluded about Herbalife, and the business is truly sound. If this turns out to be the case, then the stock price will eventually rise again, and Ackman’s hedge fund will suffer greatly, punishing him for his false assessment, and Herbalife will be fine. But what if Bill Ackman, having realized that Herbalife was sound, quietly exited his short position, without telling anybody? He might profit from doing this, since Herbalife’s stock has already dropped a great deal just on his allegations. If this happened, Ackman would hypothetically profit from, essentially, tricking the markets; Herbalife would have wasted a lot of its valuable capital on its legal threats to Ackman, its PR campaign, etc.. The markets would have, in this special case, rewarded the wicked and punished the good. Could this happen? Theoretically, yes, but in practice it’s unlikely. If he did this, Ackman would ruin his reputation and credibility on Wall Street for the rest of his life, which would be very costly to him in the end — so it’s very improbable that he or any other investor of significance would. More, delusion and deception are native to the human condition, and hardly unique to short-sellers. And why should we say that erroneously shorting a stock is so much worse than erroneously boosting a stock?

In theory, as we’ve noted, a short-sale can only profit you if the company’s stock price actually does drop, as your short-sale predicts. So a short-sale does not reward you for just attacking fundamentally sound companies. Are there exceptions to this? One possibility is that there could, in some marginal cases, be powerful self-fulling prophecies. Again, as I’ve emphasized, trading equity and balance-sheet equity are distinct; so short-sellers don’t deprive companies of the capital they need to do business. But my understanding is that for some companies, lending terms and other obligations are tied to their shares’ trading prices.  E.g., a company might be required to pay a higher interest rate on debt if its shares fall below a certain price. Or a bank that has entered into lots of derivatives contracts might be required to post lots of extra collateral immediately if its share price falls; posting this collateral could, in turn, force the bank to sell off other assets in a fire-sale, which would hurt its core business, initiating a downward spiral. These sorts of effects can be particularly harmful in major economic or financial crises — and so sound regulation should guard against these sorts of systemic and spiraling risks.

But we also hear this concern voiced outside of crisis situations. Some people worry about more mundane ways in which this self-fulfilling prophecy can work its evil. I.e., there’s a fear that short-sellers put heavy pressure on management teams to pay too much attention to short-term stock prices, which could cause them to lose track of sound management for the long run. Is there truth in this idea? Honestly, I’m pretty skeptical. On the most basic theoretical level, the value of a share consists in a slice of all the future profits of a company — so placing ‘the long-term value’ of a company in opposition to ‘its short-term stock price’ is a false dichotomy. More practically, it seems that it would require heroic skill to take down a fundamentally sound business just by psyching out the management. I suspect that this ‘concern’ about ‘harmful short-term pressure’ from short-sellers is largely mongered by management teams who aren’t very good at what they do, and want some pre-fabbed catch-phrases to take to the press, so that the big bad mean short-sellers will leave their company alone!

***

(5) Things I didn’t just write: This post has not argued that everything in the financial sector,  and our public-policy approaches to it, is a-okay. It has not argued that equity-trading is necessarily the most morally worthy of professions (I will also not say it is particularly morally unworthy). It has not argued that there is no excess of high-frequency trading in the markets. It has not argued that it is no problem that so much intellectual talent in the U.S. is pulled into the financial sector as against, say, engineering or teaching. It has not argued against circuit-breakers to prevent the massive crashes that can come from panic psychology or algorithm-driven trading. It has argued that short-sellers provide a valuable service that is essential to a modern economy, and the language and metaphors we use to describe them are misleading.

–Matthew Shaffer