An Interesting Market Failure

Will the market mechanism evacuate efficiently? George Stigler says yes; Tyler Cowen is not so sure.

If you are evacuating a city suddenly, along a constrained road or path, ideally (at least by economic standards, which may or may not be your final moral theory) you wish to favor the people who are young, productive to others, and people who value their own lives highly and are risk-averse. A market auction tends to favor the wealthy and in this context many of the first leavers in line will be inefficiently old

TC also notes that the old may also have a higher discount factor for spendig their current wealth, adding to the dilemma.

Rare Terrorists and Bayes' Rule

Matt Yglesias says "Very Rare Terrorists are Very Hard to Find":

monitoring the UK’s 1.5 million Muslims is a lost cause. If you have a 99.9 percent accurate method of telling whether or not a given British Muslim is a dangerous terrorist, then apply it to all 1.5 million British Muslims, you’re going to find 1,500 dangerous terrorists in the UK. But nobody thinks there are anything like 1,500 dangerous terrorists in the UK. I’d be very surprised if there were as many as 15. And if there are 15, that means you’re 99.9 percent accurate method is going to get you a suspect pool that’s overwhelmingly composed of innocent people.

This is basically Bayes' Rule at work. A hypothetical 1.5 million-person pool of suspects in Great Britain, only 1/10,000 of whom are terrorists. An identification method that is 99.9% effective means that with probability 0.999 a Muslim will be identified as a terrorist when he is one, and a 0.001 probability that he will be identified as a terrorist when s/he is not. Thus, you will have about 1,500 identified as terrorists by the intelligence technique. Thus, the probability of someone is a terrorist given that he
has been identified as such by the method is just 0.01 (although this
is a marked improvement over 0.0001 obtained by profiling all Muslims)!
The good news is that you have almost certainly netted all 15 terrorists in that pool. The bad news is that you've arrested, and harassed 1,485 innocent British Muslims (about 99% of those identified as terrorists).

I've done a similar calculations for using torture and the so-called 1% doctrine. Here is a paper by Hugo Mialon, Sue Mialon and Maxwell Stinchcombe on the subject. They find that legalizing torture creates a disincentive for using other means of investigation, even in cases where there is low evidence of terrorist involvement.
HT: TC at MR

More Superfreaky Controversy

Is Levitt really surprised when he attracts controversy anymore or is he just a glutton for it. He insists in Superfreakonomics, and on his blog, that for a given distance traveled, it's safer to drive drunk than to walk drunk, and that it might be a tossup between driving drunk and taking a cab. Couple things: First, the option of driving home drunk, may lead to a longer distance to travel, thus eliminating the option of walking. In other words, as Andrew Gelman points out, it's stupid to assume that "all else is equal."
Here's another thing. Drunk walking has few, if any, negative externalities. Drunk walkers are unlikely to careen off the road and do harm to an innocent bystander. So, even if there is a tossup between the two in terms of private benefits, the public cost is nowhere close.

The Pelzman Effect and Moral Hazard in Mountaineering

I caught the update on NPR on the search for three missing mountain climbers in Oregon this morning at about 6:50am while driving to work. I don't have any real comment on that search, as it is terrifying and sad to think of their fate.

However, the story had an add-on about a proposed policy: Republican John Lim has rallied around a group of Oregonians supporting mandatory radio beacons. The policy effects of such beacons are unclear, however. First, they might not be very effective. The current leader of the search effort says in the clip that no one is saying, "if only they had a radio beacon." Second, as the piece suggests, amateur climbers will feel safer about their odds of surviving the difficult climb, and be more likely to get into an accident. This is a specific type of moral hazard known as the Pelzman Effect. Essentially, more beacons may mean more total accidents, and, even if they rescue success rate improves marginally, more total accident deaths.

Question about the housing bubble/financial crisis.

If the housing bubble and financial crisis can be attributed to a small incentive to give a small percentage of home loans to poor borrowers in the US, why did UK and Germany have bubbles and crashes, too, and why did UBS and RBS have their own subprime crises? I mean it seems like an obvious way to debunk the "deregulation and 'innovation' were great it was the liberals' fault for trying to help poor people."

Kenneth Arrow

Another in a set of interviews with great economists by Conor Clarke at Atlantic Monthly:
Part one: Economics and business cycles
Part two: Health Care
I guess this is one of my favorite quotations from it:

One point was that health is a random event. It's not like buying automobiles. Whether you're sick or not is hard to predict. Some get sick and some don't. That uncertainty makes it an ideal scenario for insurance. Some houses burn and some don't, but you know whose. So you have fire insurance. (And by the way, financial problems have the same characteristics, and I was always interested in the subject -- the uncertainty.)

Later,

But in the case of health care there are three players: the insurance company with the health plan, the physician, and the patient. The physician presumably has a better knowledge of what the patient needs -- at least better than the insurance company does. So the insurance company could never put together a bill. There is also a Physician and patient relationship, but the physician knows more than the patient.

There are information asymmetries in this story. Health insurance is limping along. It's limited in scope, and then you other consequences. Insurance companies have high premiums to protect themselves. The ones who come to the insurance company are sicker and the people have to pay more. You have adverse selection. You have moral hazard. And the doctor does what's on the safe side -- defensive medicine -- without regard to cost. These are fundamental conditions that make health insurance difficult.

In other words, even if a 2 quart bottle of ketchup costing twice the price of a 1 quart bottle shows market efficiency for ketchup, insurance (and financial) markets work much differently. Of course, I've made these points about adverse selection and moral hazard casually myself, too, but getting them from a Nobel (who isn't Krugman) probably adds more weight.

On the policy side, Arrow had a nice result in his 1951 Ph.D. thesis known as the impossibility theorem.

Terrorism and Oil

This is an interesting post on the World Bank PSD blog, with nice links to support.

Driver's Licences

With all due deference to the esteemed Professor Borjas, I have to take some issue with him over his recent postings about driver's licenses for undocumented immigrants. On the surface the idea that "illegals" should not be enabled with the privilege of driving is seductive, but if we dig deeper, the idea that they should be denied may be worse. If we "punt" the debate over whether illegal immigration should be accepted, it still doesn't necessarily make sense to deny themthe right to apply for and receive a driver's license. If illegals aren't licensed, the likely alternative is not that they won't drive, or even that they won't cross the border, rather the alternative is 12 million unlicensed, unmonitored, unaccountable, and uninsured motorists on the road. In other words, we bear the burden of their added risk. A similar argument was applied to the "bad driver fees" several states added to certain types of traffice violations-- states like Michigan noticed that the fees led to a substantial increase in uninsured motorists driving on licenses that had been suspended because of a single arbitrary "bad driver" incident. If illegal immigration continues to grow, this risk can only seem likely to increase.

Negative Loss = Positive Gain?

Check out this article from Popular Science's Brilliant 10 for '07. Basically it's about some of the cognitive similarities between monkeys and humans, and whether we seem to "miscalcuate" expectation in the same way. First, let me say that my only experience with the brains of monkeys is in the first person-- i.e. my own relatively unevolved mind. But, the article is interesting because the monkeys are put in 2 scenarios: In scenario 1 they "pay" 1 token for 1 slices of apple, but are given 2 with probability 0.5; In scenario 2 they "pay" 1 token for 2 slices of apple, but are only given 1 with probability 0.5. In both scenarios they can expect 1.5 slices. It is unclear to what extent the monkeys understand the "rules" before hand, or what group is used as a control, but scenario 1 seemed to be much preferred by the monkeys, and it seems it is more preferred by the more evolved primates that have opposable thumbs. I would guess that playing the game many many times with the monkeys would converge to a more uniform reaction in terms of "monkey mood."

But, people are the same way, at times, especially when uncertainty is involved. I'm not sure what the appropriate economic explanation, but there seems to be some argument here for a more integrated role of psychology in explaining economic behavior, and may explain some of the puzzles in economics. Small deviations from "full rationality" at the micro level can easily perpetuate persistent deviations from the rational equilibria that are used in neoclassecal macro/finance mondels. Mostly, I'm looking for comments and/or discussion.

Baby Crib Recall

China and TRADE HAVE NOTHING TO DO WITH THIS! It's a design flaw by US designers and managers!

Equal Time for Stupid Democrats

I just want to take a moment to call out the Democrats on Iraq, because they're not any smarter when it comes to making logical arguments about their positions on Iraq than those other guys. Every time a democrat argues that we should get out because we should have never gone in to begin with s/he is making the same fallacy in judgement as the Republicans who say we should stay there because we did decide to go. The decision to have gone in the first place is a foregone conclusion, and all of these twits need to learn from the past without re-arguing it. No decision about how best to proceed should be based on anything other than weighing the present and future gains to our National and global security against the marginal consequences and costs of those proposals.

Iraq and the Sunk-Cost Trap

Just a quickie today. I'm calling out Mike Huckabee and half of the Republican field for falling into what anyone with an undergraduate business degree knows to be a "sunk cost trap." Regardless of how you weigh the costs and benefits of various alternatives to how we should move forward in Iraq, it is a disservice to our troops to stay the course simply because we decided to go in the first place. To me, continuing a policy (whether that policy is good, bad or ugly) on the basis of events already behind us is a sad mistake in how to move forward in an optimal fashion.

The argument I'm calling out is the "honor" argument, which basically stakes the claim that changing course, redeploying, etc. is not an option because so many service men and women have already made the ultimate sacrifice. What bigger disservice could we do to their memory than to not learn from their sacrifice? Would it not be appropriate to allocate the lives of the men and women who carry the flag behind the fallen in a way that best serves our National Interest instead of squandering their service to a failed policy? In otherwords, the lives lost, are sadly and tragically already lost-- they are a "sunk cost" that can never be recovered. Let's ensure that any additional loss in life maximizes the benefit of security obtained from those future casualties.

The Dirty Secret Behind "The Dirty Secret Behind America's Energy Future"

Jeff Godell's book, "Big Coal: The Dirty Secret Behind America's Energy Future," has gotten a resurgence of media attention in recent days because of the events that have transpired in the Utah mine tragedy and the more recent tragedy of a similar nature in China. One of the things Godell blasts the coal industry for is safety (hence the newfound media attention). Now, if you've read my other posts on energy, you'll know that I'm no blind defender of Big Coal. I am, however, a big fan of making sure that if I'm going to pour my outrage, and that I target it where it will do the most good.

Around the time of a BIG EVENT with LOTS OF MEDIA, nearly every politician has to make sure they are on the public record "saying the right thing," and showing the appropriate amounts of outrage over the incident, resolve to do more from now on, and contrition that nothing was done sooner that would have saved the lives of the noble citizens who met their ultimate demise. But do we mete out our attention, sympathy, outrage and action equally? Well, OF COURSE NOT. It's a "law of small numbers" thing (the principle comes from the properties of the Poisson Distribution of statistics and referenced in Thomas Pynchon's fictional novel, "Gravity's Rainbow"). Basically, the way I'm going to apply it is to say that because certain events are so unexpected generally, that when they do occur they tend to grab our attention and affect behavior disproportionately to their risk of occuring again. My point is, there are hundreds of on-the-job fatalities every year (over 1400 in manufacturing) and almost none of them make Headline News. Why? They're just too damn common to be newsworthy! Hence, there is almost no public outcry for the government to take action. It's kind of like flying versus driving-- people are terrified of planes even though it's widely known that driving is statistically much more risky. (Again, there are so many damn car wrecks there's no way the news is going to start sending crews to them...) Does the fact that a certain type of fatality is more "common" or "mundane" make it less tragic?

Back to mining... Have I mentioned that economists like to back stuff up with data? So, the Coal mining industry actually has a much LOWER incidence of on-the-job fatality than the general manufacturing sector (according to the Bureau of Labor Statistics, manufacturing had about 3 fatal incidents per 1,000 workers in 2006 whereas coal had about 0.8 per 1000 workers). So, even if, as Mr. Godell asserts, coal accidents increase drastically over the next several years, it is unclear that our regulatory efforts are best-placed in that sector. It is conceivable that many more lives would be saved by fortifying safety regulations in a more general way rather than targeting coal. Not only that, it seems to me that the dangers of mining coal underground are not unforseen, so miners who choose that line of work are making an informed choice-- and are compensated for the perceived risk. In fact, despite a lower incidence of fatalities, COAL miners out-earn their colleagues in the manufacturing sector by about $7 per hour.

So, with all of this in mind, where should we spend our regulatory dollars? In coal, where there were just 78 fatalities in 2006, and incidents were relatively rare, or in the manufacturing sector or in a more general way altogether, where there are far more fatalities (and lives that could potentially be saved), but far less media attention? Should the media's crocodile tears and misinformation dictate where our tax dollars are spent? Maybe I'm a jerk, but as my students know, I want to make sure that I get the most "Bang" for my buck.

All New Meaning to "Cost of Living"

Several days ago, I was on a rant about Lou Dobbs calling economists "idiots" and "jackasses." But, while we tend to know what we are talking about a good deal of the time (usually with data and ituition to back up our stories), we are also really big jerks. Case in point: most "normal" people would hear the phrase, "you can't put a price on a human life" and agree without a second thought. Eonomists hear this, and ask, "can you back that claim up with data?" In fact, human life does have a price put on it, and economists aren't to blame.

Economists have actually done calculations to try to infer the cost of dying using data that are readily available. Frank Ackerman has "priced" a number of things, including death, in his pamphlet, "Pricing the Priceless." So how much is a human life worth? On average, between 1 and 6 million, depending on certain characteristics. But here's the catch: Computations aren't entirely made on the basis of earnings, life expectance at birth, etc.-- They're made by observing and evaluating our own behavior. If you feel this cheapens your worth as a person, you have yourself to blame. Here's one approach that's been used. Take construction workers, and survey their hourly earnings. Now survey the compensating differential between working on the first floor and working on the thirtieth floor. Next, calculate the increase in the risk of dying on the job, and there you have it. The price these construction workers have voluntarily put on their own lives using this back-of-the-napkin approach is about 2 million dollars.

So, the next time you run a red light,. think about the 30 seconds it saves you, the risk of dying you assume by doing it, and the fact that somewhere an economists might be watching.

Tomorrow, I will focus on the information distortion we get from the media-- are we evaluating the risk of death or disaster appropriately when we weigh whether or not to fly vs. drive, or in measuring our level of outrage for government action for things like mining regulations, bird flu precautions, etc.