I eagerly awaited and quickly devoured SuperFreakonomics when it appeared a few weeks ago. And while many reviewers are focusing on the substance of the book, I’m struck by two shifts in the Levitt/Dubner method.

First, SuperFreakonomics is more of an effort at problem solving. The original Freakonomics book showed how creative econometrics applied to historic data could be used to uncover the “hidden” causes of observed behavior. To be sure, SuperFreakonomics retains many examples of the hidden-side-of-everything data mining. But the new book is much more of a solutions book. It uses economic thinking to generate new ideas to solve really big problems. Levitt and Dubner are admirably leveraging the success of the first book to try to make the world a better place. They are on the lookout for concrete suggestions to reduce the lives lost from hurricanes, hospital infections, global warming, automobile accidents and even walking drunk.

In the original book, number crunching itself was the solution. Forensic number crunching could help identify whether Sumo wrestlers had thrown a match or whether Chicago teachers were cheating on test scores. In the new book, number crunching is instead used to verify that a particular solution (such as hand-washing or ocean cooling) is likely to work.

The Randomization Lens

The second methodological shift is subtler. The first book focused on historical data. For example, a core story of the original book looked at data on crime and abortion. In a truly inspired moment, Levitt (and his coauthor John Donohue) were able to show that legalizing abortion reduced the amount of crime — 18 years later. Mining historic data can produce truly startling results.

But a higher proportion of the new book is devoted to studies that use randomized field experiments to find out what causes what. If you want to know whether offering donors a two-for-one matching grant produces more charitable donations than a one-for-one grant, you randomly assign potential donors to receive one of these two solicitations and then look to see whether the two groups give different amounts.

One sign of the shift toward randomization is the prominence of John List and his rise to fame in the economics profession. John is one of the great field experimenters in economics today. He’s the kind of guy who goes to baseball card shows and at random treats one set of card dealers differently from another and then sees whether they offer different prices. (You can read an excerpt of the book’s discussion of List here).

SuperFreakonomics not only relates the results of more randomized experiments than Freakonomics did, it also explains how the idea of randomized experiments is leading statisticians to think more clearly about how to use regression analysis to test for causal effects with historic data. There is a new zeitgeist in the way economists think about running regressions. Today, statistical economists explicitly think of their regressions in terms of randomized experiments. They think of the variable of interest as the “treatment” and ask themselves what kind of assumptions they need to make or what kind of statistical procedures they need to run on the historic data to emulate a randomized study. This new way of thinking is very much on display in the truly excellent (but technically demanding) book, Mostly Harmless Econometrics: An Empiricist’s Companion, by Joshua Angrist and Jorn-Steffen Pischke. (I praised the book in a previous post because it “captures the feeling of how to go about trying to attack an empirical question….”). For example, Angrist and Pischke show that the regression-discontinuity design (which I’ll say more about in a later post) provides causal inference from historic correlation because it emulates randomized assignment of a treatment to otherwise similar subjects.

What Economists Would Really Like To Do

SuperFreakonomics very much reflects this new randomization lens as a way of thinking about data-mining. Without off-putting jargon, Levitt and Dubner explain how regressions can give you quasi-experimental results. Indeed, with help from my Kindle, I found three parallel descriptions that turn on making the randomization analogy. For example, listen to how they describe testing for sex discrimination on the job:

Economists do the best they can by assembling data and using complex statistical techniques to tease out the reasons why women earn less than men. The fundamental difficulty, however, is that men and women differ in so many ways. What an economist would really like to do is perform an experiment, something like this: take a bunch of women and clone male versions of them; do the reverse for a bunch of men; now sit back and watch. By measuring the labor outcomes of each gender group against their clones, you could likely gain some real insights. Or, if cloning weren’t an option, you could take a bunch of women, randomly select half of them, and magically switch their gender to male, leaving everything else about them the same, and do the opposite with a bunch of men. Unfortunately, economists aren’t allowed to conduct such experiments. (Yet.)

They go on to describe how, in the absence of randomized data, some (limited) progress might be gleaned by looking at the historic experience of transgendered people — before and after sex reassignment surgery. They take a similar approach when tackling the question of testing physician quality:

What you’d really like to do is run a randomized, controlled trial so that when patients arrive they are randomly assigned to a doctor, even if that doctor is overwhelmed with other patients not well equipped to handle a particular ailment. But we are dealing with one set of real, live human beings who are trying to keep another set of real, live human beings from dying, so this kind of experiment isn’t going to happen, and for good reason.

Since we can’t do a true randomization, and if simply looking at patient outcomes in the raw data will be misleading, what’s the best way to measure doctor skill? Thanks to the nature of the emergency room, there is another sort of de facto, accidental randomization that can lead us to the truth.

The “next in line” queue at some emergency rooms provides quasi-random assignments and allows researchers to emulate the results on a randomized test. The magic “really like to do” words appear a third time when Levitt and Dubner talk about testing whether more incarceration would really lower the crime rate:

To answer this question with some kind of scientific certainty, what you’d really like to do is conduct an experiment. Pretend you could randomly select a group of states and command each of them to release 10,000 prisoners. At the same time, you could randomly select a different group of states and have them lock up 10,000 people, misdemeanor offenders perhaps, who otherwise wouldn’t have gone to prison. Now sit back, wait a few years, and measure the crime rate in those two sets of states. Voilà! You’ve just run the kind of randomized, controlled experiment that lets you determine the relationship between variables.

Unfortunately, the governors of those random states probably wouldn’t take too kindly to your experiment. Nor would the people you sent to prison in some states or the next-door neighbors of the prisoners you freed in others. So your chances of actually conducting this experiment are zero.

That’s why researchers often rely on what is known as a natural experiment, a set of conditions that mimic the experiment you want to conduct but, for whatever reason, cannot. In this instance, what you want is a radical change in the prison population of various states for reasons that have nothing to do with the amount of crime in those states. Happily, the American Civil Liberties Union was good enough to create just such an experiment.

The methodological repetition across these examples is one of the book’s strengths. This is really the way that many empirical economists talk to themselves about testing. Regardless of the problem, we often now start with the same basic question.

One of the great early stories from SuperFreakonomics is the finding that “even after factoring in the deaths [innocent bystanders from drunk driving], walking drunk leads to five times as many deaths per mile as driving drunk.” The substantive fact is not only surprising, but the story also metaphorically foreshadows the book’s new emphasis on experimental approaches. After all, what makes a drunkard’s walk so dangerous is that the drunkard lurches from side to side randomly.

When I’m upset about the minor annoyances of life, I sometimes find it helpful to think of the price I’d charge for enduring the annoyance. For example, when my wallet was stolen, I wondered how many dollars would someone have had to pay me to consent to the taking. This thought experiment is comforting because the amount is usually much less than the daily fluctuation in my stock portfolio. Why should I be so bent out of shape for something that “all in” has a frustration cost of X, when I routinely endure a portfolio loss of 2X or 3X without wrinkling a brow?

But Orlando Magic coach Stan Van Gundy has an even better “Why Not?” idea about monetizing frustration. After one of his players, Matt Barnes, was fined $20,000 for throwing a basketball into the stands, Van Gundy joked:

Barnes should consider throwing cash into the stands instead of a ball next time.

“That’s basically what he did,” the coach said. “At least if you did that, it’d be the same amount of money, and you’d be very popular. If he threw $20,000 in cash, he’d be very popular.”

Actually throwing money into the stands might cause a riot. But you could imagine a team keeping some cash on hand at courtside to let players, who were about to commit a finable offense, bypass the NBA middleman and give the fine directly to some designated recipient. Instead of throwing a ball into the stands, Barnes could have ceremonially and publicly deposited cash into a courtside forfeiture drawer — with the money going to charity or to rebate part of the ticket prices.

Publicly forfeiting money is a pretty credible way to signal that you are upset about a blown call. It is not cheap talk. Forfeiting money to fans or a charity that players like makes the talk a bit cheaper because players might get some value from making such a donation. But then again, the current NBA fines suffer from a similar problem:

The NBA and National Basketball Players Association (NBPA), per the league’s collective bargaining agreement, equally split fines paid by players, then donate the respective shares to the charities of their choice.

Rasheed Wallace or Mark Cuban may be less deterred in criticizing the refs if they think the ensuing fine will ultimately go to a worthy charity. If Wallace plans to give a bunch to charity anyway, he can just reduce his non-fine giving so that the fine won’t have much of an effect. The NBA wisely chose to “not specify which charities benefit from the league’s donations.” It’s harder to rationalize that your fine is going to a good charity that you support if you don’t know the beneficiary of your fine.

The commitment website I cofounded, stickk.com, takes a similar approach. Users who put money at risk to keep a commitment have the option of designating a specific charity or anti-charity. But we also give them the option of forfeiting money (if they fail to stickK to their goal) to a more ambiguous “charity” option — where we, like the NBA, choose not to specify the recipients clearly. When you’re trying to deter particular behaviors, sometimes it is better to make it harder to put a price on the frustration.

Students at University of California schools have been protesting the decision of the Board of Regents “to raise undergraduate fees — the equivalent of tuition — 32 percent next fall.” But higher tuition, if it is accompanied with higher financial aid for lower- and middle-income students, improves equity. As Aaron Edlin and I wrote back in 2003:

It might seem … that raising state college tuition is plainly a bad thing. High tuitions mean students will find it harder to finance college — and may not even attend, or may drop out due to costs. And for the students who attend state colleges, many of whom are of modest means, the tuition crunch may be especially painful.

In fact, that is absolutely not the case. The truth is that increasing public college tuitions are not a problem at all. Indeed, the biggest problem in pricing tuition at public universities is not that the poor pay too much, but that the rich pay too little.

Tuition increases are actually a good idea — as long as they are matched with financial aid, including scholarships, for poor students.

The Huge Gap Between Average Public and Private University Tuitions

Consider a comparison: U.C. Berkeley offers more courses taught by more Nobel laureates than Yale. Yet Yale charges $28,400 per year in tuition and fees, while Berkeley charges $5,858.

And this is no anomaly: tuitions at public universities average $4,694 compared with $19,710 at private colleges. In short, public university tuition, on average, costs less than one-quarter of private university tuition. (And that is even in light of this year’s public university tuition increase of 14 percent — the largest in at least a quarter of a century.)

Who benefits from the low public-school tuitions? A disproportionate amount of the benefits go to rich students who attend schools like Berkeley because of the way financial aid operates.

The Problem for Poor Students Is Low Financial Aid, Not High Tuition

What happens, then, when public university tuitions rise, as has occurred recently? Perhaps surprisingly, the situation becomes fairer.

The rich Berkeley student now must pay a tuition much more commensurate with what he or she can afford. And the poorest Berkeley student are typically not much worse off: as tuitions have risen this past year, those from the poorest families saw their financial aid packages rise almost dollar for dollar.

For poor students, then, the important issue isn’t tuition so much as financial aid. [If] students can’t afford the fee increases at UC Berkeley, … the answer isn’t a tuition decrease; it’s a financial aid hike.

Why Public Universities Should Continue to Raise Tuitions Even More

Thus, the member schools of the California system, for example, would be wise to radically increase both their tuitions and their financial aid.

For instance, suppose UC Berkeley raised its tuition by $20,000 per year and gave all but its richest students an extra $20,000 scholarship. With the extra money it got from its richest students, it could balance its budget. And, having done so, it would not need to burden students even from middle-class families.

A side benefit of raising tuition and financial aid is that it would increase UC’s position in U.S. News rankings which turn in part on the amount of financial aid granted:

Would Berkeley deserve this position increase? Absolutely. The bump up in ranking might seem to be the result of sleight of hand or subterfuge; after all, Berkeley’s increased financial aid would be required only because of its own decision to raise tuition.

But in fact, the change would only equalize Berkeley with schools like Yale, which currently get a ranking advantage using the very same “sleight of hand.” That is, Yale chooses to charge a very high tuition, but then effectively waives a great deal of it through financial aid.

Moreover, Berkeley is already, in effect, giving lots of financial aid out — but it goes to the wrong people, and it isn’t counted in U.S. News ranking. Every affluent student who attends Berkeley, not Yale, in effect gets a $20,000 scholarship to do so. The current aid is just given in the hidden form of low tuition.

Financial aid is, at its core, a price-discrimination scheme. Consumers pay different prices (net of financial aid) for the same service. Higher education is the very rare market where the seller says “Tell me in detail about your ability to pay, and I’ll tell you what your (net) price will be.” But instead of maximizing firm revenue, the goal is to enhance equity. By increasing the effective tuition for some of our wealthier students, we might be able to reduce the price for some of the less wealthy.