« Will European social democracy survive? | Main | Gerald Dworkin writes to me »
Katrina remedies?
What are the options? I don't buy the Leeson-Sobel notion that cutting FEMA aid will improve long-term performance in disaster-stricken areas. The real question is what we should do ex ante.
Howard Kunreuther and Mark Pauly promote a traditional idea:
This paper explores options for programs to be put in place prior to a disaster to avoid large and often poorly-managed expenditures following a catastrophe and to provide appropriate protection against the risk of those large losses which do occur. The lack of interest in insurance protection and mitigation by property owners and by public sector agencies prior to a disaster often creates major problems following a catastrophic event for victims and the government. Property owners who suffer severe damage may not have the financial resources easily at hand to rebuild their property and hence will demand relief. The government is then likely to respond with costly but poorly targeted disaster assistance. To avoid these large and often uneven ex post expenditures, we consider the option of mandatory comprehensive private disaster insurance with risk based rates. It may be more efficient to have an ex ante public program to ensure coverage of catastrophic losses and to subsidize low income residents who cannot afford coverage rather than the current largely ex post public disaster relief program.
The goal is to make people internalize the social costs of placing their assets in a vulnerable position. If you own a home by a questionable levee, you have to buy insurance. Maybe the price of that insurance will tell you not to keep the home. I have two problems with this idea. First, in distributional terms we will essentially end up confiscating the homes of many poor people. Second, insurers can be notoriously reluctant to write policies for high-risk areas, or they will write policies with exorbitant non-expected-utility-based rates. (Is any of this regulatory? Why don't the markets pool out the risk? Is there a principal-agent issue within the insurance company?) It might lead to far more confiscation or abandonment than is efficient.
The correct ex ante policies toward disasters remain an underexplored are of microeconomics. And why private insurance doesn't do a better job of insuring against long-run risks..well...that is perhaps the leading question of applied microeconomics today.
Here is my previous post on libertarian policy recommendations and Katrina.
Posted by Tyler Cowen on October 31, 2006 at 03:52 AM in Economics | Permalink
Comments
do i sense a business opportunity in lon-term insurance here? to be fait, berkshire hathway does do long
term insurance.
Posted by: sa at Oct 31, 2006 7:35:36 AM
Second, insurers can be notoriously reluctant to write policies for high-risk areas
Yes, and? High insurance costs are the market's way of saying "Don't build here, stupid."
If we're going to subsidize poor decisionmakers, let's do so in a way that we don't have to do it again at some point in the future. Give current home-owners a one-time housing voucher. If they then decide to build there anyway, make them pay high mandatory insurance rates sufficient to cover the risk.
Posted by: Christopher Rasch at Oct 31, 2006 9:07:53 AM
Insurance companies are also
notoriously reluctant to pay claims
following disasters (their advertising
not withstanding). Absent some provisions
requiring relatively rapid settlement
of claims, even appropriately-priced
insurance provides no guarantee that
people suffering catastrophic damage
will be compensated.
And, by the way, I thought Tyler's
argument was that insurance companies
are offering insurance with higher-than-
"correct" rates.
Posted by: Donald A. Coffin at Oct 31, 2006 10:31:40 AM
Donald Coffin is surely correct that many companies pay claims slowly and reluctantly. Just talk to some Katrina victims about their experiences. I think economists who want to explore the problem Tyler mentions would do well to stop building models for a while and go out and see how things really work.
The insurers have a number of major advantages over their policy-holders in these cases.
First, they are holdng the money, and the only way to get them to pay before they want to is litigation. They have all sorts of delaying mechanisms available - claims processing, etc.
Second, the policyholder will often be in a difficult financial position, reducing bargaining power. In other words, the policyholder will be willing to accept a substantial discount to get the money quickly.
Third, litigation is not only costly and time-consuming, but, tales of avaricious trial lawyers notwithstanding, is a process that likely favors the company.
These sorts of considerations - more or less the grossly unequal bargaining relationship - must be taken into account for economic theory to say much useful about this problem.
(Please, no nonsense about how competition among insurers wipes all these issues away.)
Posted by: bernard Yomtov at Oct 31, 2006 11:22:43 AM
First of all from a libertarian stand point...If you build your house by a levee and you don't have any kind of insurance you are basically self insuring. This principle is similar to increasing a deductible from $500 to $1000. If you take on the responsibility of the first thousand dollars you are self insuring and if a disaster strikes you owe the first $1000. Also, if you build your house next to a levee no one should have to be forced to buy insurance. This similar to being forced to buy health insurance, whether I want to or not. The other problem is that the government continues to subsidies people who choose to live next to a levee. They never learn a lesson. "I don't have to worry, the government is there to take care of me, even though I know the dangers."
There was a program that talked about the New Orleans levee and there was a "levee expert" on the show. His statement caught me off guard. It's not a matter of how much a levee can hold til it breaks its a matter of when it will break. He said point blank that no levee system in the world is perfect, the Dutch haven't even come up with a levee that will hold back every break. If the government continues to subsidize people living next to a levee because the poor schmucks lost their house, why not keep living there.
Secondly, in response to Bernard's last line. Competition does keep this from happening. But the law must be willing to enforce contracts. The dispute in the Katrina area is over language, "wind blown vs flood". This makes a big difference in claims, because it is contractual. Regulation on insurance limits the number of insurers within the markert. Let's say an insurance company gets a poor rap for bad claims service, it has several options, either improve service or get out of that line altogether. Auto insurance may be a little easier to understand. If someone runs into my car and I don't believe my insurance company is handling the claim well, what am I likely to do? Go to another company that better claims. It behooves the insurance company to provide great claims service. In a catastrophe there are limited resources, even for big bad insurance companies. (Don't come back with I am insurnace company apologist, I just live in the real world.) The issues never get wiped away, but it can certainly hurt an insurance company's bottom line. There are good companies and bad companies. Eventually the companies that provide great service will be better off.
Auto industry is another good example of how good and reliable service benefits companies. If I can limit who has the ability of competition to get into the market, then I can continue to provide crappy service.
Sure state insurance comissioners can do what they want, but I don't want to go too much into the Iron Triangles.
Posted by: Matt at Oct 31, 2006 12:02:53 PM
--First, in distributional terms we will essentially end up confiscating the homes of many poor people. ---
Or rich Californians who keep insisting on building
their homes on the sides of mountains.....
How about 3 strikes and you're out?
Posted by: Sandy P at Oct 31, 2006 12:55:48 PM
Bernard makes excellent points. To assume the mostly poor and undereducated residents of the lower ninth worth exhibit the rationalty of homo economicus is ridiculous. Hence this point by Matt, "First of all from a libertarian stand point...If you build your house by a levee and you don't have any kind of insurance you are basically self insuring." is true but ultimately unproductive.
Matt also says: "The other problem is that the government continues to subsidies people who choose to live next to a levee. They never learn a lesson."
Christopher says:
"If we're going to subsidize poor decisionmakers, let's do so in a way that we don't have to do it again at some point in the future. Give current home-owners a one-time housing voucher. If they then decide to build there anyway, make them pay high mandatory insurance rates sufficient to cover the risk. "
I direct them to this article by John Stossel raises questions about the rationality (at least of the rich) to build in vulnerable areas.
Finally, the primary insurer in the NOLA area is State Farm which had inadequate reinsurance to respond to a major disaster like Katrina. Much like FEMA, insureres were not prepared to respond to a major disaster.
Posted by: Irving Washington at Oct 31, 2006 12:59:32 PM
Matt, I think you're being a bit naive here. Changing carriers for bad service won't correct this problem: the events are sufficiently rare that you don't have the opportunity to test them out until it's too late; how a carrier handles small scale payments is not necessarily indicative of how they will handle your large scale loss (you may have a better chance of getting paid if they were good on your auto policy, but the scale of losses is different since the magnitude to you is great and the magnitude of other policies that they must pay is likewise great); and, if everybody were dissatisfied after Katrina and everybody switched, but the companies saw no net change in their number of customers, they wouldn't learn the lesson that my switching was intended to teach them (they found some other sucker to replace me).
Posted by: RSaunders at Oct 31, 2006 1:07:32 PM
the events are sufficiently rare that you don't have the opportunity to test them out until it's too late
Right, but you don't learn just from your own case, you learn from others. Insurance companies develop a reputation. And the more successful an insurance company is, the more data there is on them, and the more you know how they deal with ever rarer cases.
There is also a possible setup: let an intermediary decide whether the payment is merited, rather than letting them be a judge in their own case. Private courts. Much cheaper and quicker than public courts. Such market innovation is curtailed by government interference, and government does interfere in insurance. Lesson: more laissez-faire, please.
if everybody were dissatisfied after Katrina and everybody switched, but the companies saw no net change in their number of customers
The same argument would apply equally well to any competitive market. If all providers of X (anything - maybe pizza) were equally bad, and if all customers switched leaving the pizza sellers on net no worse off than before, then pizza sellers would (so goes the argument) not learn anything. But obviously, that's not how it works at all. Your scenario is at best a statistical unlikelihood on the order of all the molecules of a gas in a room spontaneously moving to one side of the room, which technically they can do.
Posted by: Constant at Oct 31, 2006 1:30:02 PM
I think RSaunders makes my point about competiton more clearly than I did. The difficulty is that policy buyers really have little information as to how the insurers will perform in the case of a catastrophe. If company A charges a higher premium than B with the argument that they will provide better claim service there is seldom a good way to determine whether that is true or not. This is unlike auto insurance because auto claims are fairly common and generally relatively small, so you can get information from friends or maybe other sources like Consumer Reports.
As to Matt's argument about "enforcing contracts," this ignores two important issues. First, where language is ambiguous the insurance company will have much better legal resources available. More important, the very time that it takes to litigate is a huge advantage to the insurer in any pretrial settlement negotiations. When your house has been destroyed, and the company you worked for is out of business, you have a very high discount rate for future payments. In other words, you will typically be willing to accept much less now than you might get after a year or two of fighting the insurer.
Posted by: bernard Yomtov at Oct 31, 2006 1:39:14 PM
Bernard makes excellent points.
Bernard's points appear to be made in ignorance of economics. He seems to think that economists fail to take into account what he is saying, but the reality is that they do take it into account and he fails to take into account what they are saying. Specifically, he treats the points that he raises as if they were intrinsic properties of the insurance business, where any economist can easily see how these aspects of the insurance business can be expected to arise out of unhealthy economic processes - possibly market failure, but more likely government interference. Matt explains some of this. Bernard is describing outcomes as if they were inputs to an economic model.
To assume the mostly poor and undereducated residents of the lower ninth worth exhibit the rationalty of homo economicus is ridiculous.
Expressions of contempt for the poor, such as the above, are no substitute for argumentation.
Posted by: Constant at Oct 31, 2006 2:00:30 PM
Rsaunders Point: "Changing carriers for bad service won't correct this problem: the events are sufficiently rare that you don't have the opportunity to test them out until it's too late." May be true. Honestly, but after the fact that insurance companies don't do well is exactly what Consumer Reports and other rating companies take into consideration. GM has a big hole to dig out in regards to reputation and they had just a bunch of little more frequent mistakes. I would even argue that the events aren't sufficiently rare. That would be like people who live along coastal Florida to say, hey, it happens every other year. I didn't think this would be the year. What exactly is sufficiently rare? That argument obviously doesn't fly if all the reports kept saying that an event like Katrina was bound to happen.
To assume that people in the 9th ward in NOLA aren't smart enough to understand economics is ridiculous in it self. That's probably why they live there. The land was probably cheaper in the first place, it was the lowest level in the whole city. Because I can't afford to send my children to private school doesn't mean that I don't know that they would get a better education in a private school. James Tooley points this out in "Backing the Wrong Horse: How Private Schools are Good for the Poor," http://www.fee.org/pdf/the-freeman/0605Tooley.pdf Just because some one is poor doesn't mean they don't know how to make decisions in a rational or values generated manner. That's why the rich continue to build on the beach at the expense of the tax payer and I believe that's what John Stossel was talking about.
Posted by: Matt at Oct 31, 2006 2:39:10 PM
Bernard's points appear to be made in ignorance of economics.
Constant,
You have already established in previous comments that you are an arrogant asshole. No need to provide further evidence.
Your points seem to made out of ignorance of how the world works, and unthinking devotion to libertarian dogma.
Insurance companies develop a reputation.
Reputation effects. The last refuge of libertarian theology.
he treats the points that he raises as if they were intrinsic properties of the insurance business, where any economist can easily see how these aspects of the insurance business can be expected to arise out of unhealthy economic processes - possibly market failure
As a matter of fact insurance companies by and large prefer not to pay. That seems like an intrinsic feature to me. If that doesn't fit into your wonderfully clean and abstract and utterly sophomoric view of economic processes, too bad. Though I do give you credit for conceding that there is some slim possibility of market failure.
Bernard is describing outcomes as if they were inputs to an economic model.
I am describing incentives and bargaining relationships that ought to considered when trying to understand the process, because they have important effects on the outcomes. Those lovely graphs in chapter one don't exactly tell the whole tale.
Private courts - right. That'll equalize things. Faster? Why? Do you think there will just be all these private judges, etc., just waiting around for the disaster? Cheaper? Why? And what about fairness? Who will pick these private judges?
Expressions of contempt for the poor, such as the above, are no substitute for argumentation.
Neither is misrepresenting others' meanings, something you are quite fond of.
Posted by: bernard Yomtov at Oct 31, 2006 3:03:24 PM
That's it...if you feel like you can't prove a point, be flippant and try to insult the person you are trying to debate. Because that's how you win an argument.
That only proves one thing...who the real arogant a**hole is.
I guess belief that government can cure all isn't theology either. Because by faith alone will government save everyone.
Where did I go wrong?
Posted by: Matt at Oct 31, 2006 3:10:06 PM
Bernard writes: "As a matter of fact insurance companies by and large prefer not to pay. That seems like an intrinsic feature to me. If that doesn't fit into your wonderfully clean and abstract and utterly sophomoric view of economic processes, too bad."
But in fact I am perfectly aware that if they can at all get away with it, insurance companies invariably will choose not to pay. Ask Tyler Cowen, ask Matt, ask anyone you're targeting with your remarks whether they are aware that businessmen would screw the customer if they could get away with it, and I am sure that they would all respond that they are perfectly aware of this and take it into account in their thinking. And for you to believe otherwise, for you to believe the nonsense that you are stating, only shows that you truly do not understand what they are saying and why. You think you are telling them something they do not already know, but in fact you're only revealing what you do not know.
Posted by: Constant at Oct 31, 2006 3:44:32 PM
Interesting datapoint I did not find out until last month, the ninth ward actually has the one of the highest elevations in the city.
Second data point. Due to natural processes the land closest to the river and the levees will tend to have the highest elevation.
Posted by: TJIT at Oct 31, 2006 3:45:03 PM
There are are many issues, but one of the most basic ones seems not to even have been mentioned -- if you live next to a levy, your concern is that the levy might break and your house might flood. But private insurance companies do not write flood insurance coverage. The federal government does.
Moreover, at least as it constituted currently, the National Flood Insurance Program's existing mandates that those who live in 100-year flood plains and have federally regulated loans must purchase flood cover generally does not apply to those communities that are protected by levies. In point of fact, one of the greatest forces that drives communities to BUILD levies in the first place is to be freed of the mandatory purchase requirements.
Posted by: R.J. Lehmann at Oct 31, 2006 3:58:35 PM
By the way, on this point:
"Finally, the primary insurer in the NOLA area is State Farm which had inadequate reinsurance to respond to a major disaster like Katrina."
This was more of a concern to Allstate, which is a publicly traded company with shareholders to please. State Farm is a mutual, owned by its policyholders. And since a) the company hasn't taken any sort of significant hit to its solvency and b) its own surplus dwarfs ALL of the reinsurance written for ALL homeowners risks in the United States, it's kind of hard to make the case that not having more reinsurance really "hurt" them at all.
Posted by: R.J. Lehmann at Oct 31, 2006 4:06:30 PM
"And why private insurance doesn't do a better job of insuring against long-run risks..well...that is perhaps the leading question of applied microeconomics today.
"
Really? It seems like one that has a pretty simple answer -- because current tax and accounting laws do not allow insurers to reserve for potential catastrophic risks, and current state rate regulation regimes do not allow insurers to price for potential catastrophic risks.
Posted by: R.J. Lehmann at Oct 31, 2006 4:16:51 PM
perhaps the leading question
That can be read two ways - e.g., "leading question" in the courtroom sense. It supplies part of the answer by suggesting, through its wording, that the problem is with private insurance rather than with government regulation.
Posted by: Constant at Oct 31, 2006 4:27:16 PM
People living in the ninth ward (not to mention NOLA and LA in general) lag behind national measures of formal education. These are facts.
I did some quick googling and found this site:
http://www.dollarsandsense.org/archives/2006/0306wagneredwards.html
According to this site, 40% of the people in the ninth ward lack a high school diploma, 29% only have a HS diploma/GED, and 7% have at least a Bachelor's degree. If nearly 2/5 of the population did not even finish HS, I seriously doubt they have the level of education (though they are probably capable of it) to act as economically rational as the people in Stossel's article. Furthermore, even highly educated people often do not make decisons that maximize utility. I think I remember reading this in Thaler's "The Winner's Curse". Isn't the concept of "bounded rationality" a tenant of behavioral economics? I freely admit I lack the background in economics of Prof. Cowen and certainly Constant. However, I fail to understand how merely mentioning the the undereducation of the 9th ward populace implies I have contempt for the poor. I am merely pointing out they lack the economic education of Constant Economicus. And this lack of education has to play some role in their economic decision making.
Posted by: Irving Washington at Oct 31, 2006 6:10:18 PM
That's it...if you feel like you can't prove a point, be flippant and try to insult the person you are trying to debate. Because that's how you win an argument.
I think you are describing Constant's comment as well as mine. Calling someone ignorant is no sounder an argument than calling someone an a**hole. Yes, I was insulting, and I concede that it was inappropriate, but no more so than Constant.
I guess belief that government can cure all isn't theology either. Because by faith alone will government save everyone.
Where did I go wrong?
In assuming that I claimed government was a cure-all.
Despite all the accusations and claims that I am just repeating what everyone knows and takes into account, I see no explanation from either Matt or Constant as to how market forces would overcome the difficulties I mention. I see no discussion of how to make private insurance function better. Nor do I see any answer to my questions about why private courts would be superior.
So if you want to talk about who is not addressing the points raised maybe you should look elsewhere.
Posted by: bernard Yomtov at Oct 31, 2006 6:51:53 PM
If nearly 2/5 of the population did not even finish HS, I seriously doubt they have the level of education (though they are probably capable of it) to act as economically rational as the people in Stossel's article.
So let's subsidize economic and personal-finance education. Or are you suggesting that the government can second-guess everyone's economic decisions?
Posted by: guest at Oct 31, 2006 6:52:32 PM
Where is it written that people must rebuild in New Orleans? If you weren't 100% insured, then your sunk costs were underwater as soon as the levee broke. Puns intended.
Just because some people had sunk costs doesn't mean there is any point in throwing more money after those losses. I believe there is a Fallacy named after this.
If the former residents of New Orleans have no place to live anywhere in the US right now, then that is a problem. But I don't hear anyone saying that. They are saying it is a problem that although they have a place to live outside of New Orleans, that taxpayers or someone have an obligation to subsidize a move to somewhere that they can't afford. Maybe someone can subsidize a move for me to the Hamptons. I promise to say Thank You for your misplaced generousity.
Posted by: happyjuggler0 at Oct 31, 2006 7:09:05 PM
Irving - Behaving rationally given the incentives does not require a high school diploma, nor does it require even average intelligence. As Matt points out, the people we are discussing do in fact appear to be acting rationally given the incentives they are presented with. Therefore, there seems to be no call to consider them to be irrational.
Moreover I would argue that the conclusions from the assumption of rationality are robust. Given a large population, only some of them need to be rational in order to produce outcomes very similar to the outcomes that a perfectly rational population would produce. consider the length of the the lines at two checkout counters. Suppose that the two lines are equally fast. Then if only half of the population chooses the fastest (i.e. shortest) line and the other half chooses a line randomly, then the lines will still stay about the same length most of the time. For simplicity's sake, suppose that, first, six people choose their line randomly, and then six more people choose the shortest line. Even in the worst case scenario, if all the random choosers choose one line so that one line has six people and the other line has zero, then the rational choosers will all choose the other line, so that in the end the lines will be of equal length. This is a simple demonstration of why it is not necessary for more than a fraction of the population to be rational.
We can can construct similar examples to illustrate that perfect rationality is also not required. Suppose that each person has a split personality and chooses rationally only half the time. Again, the outcome will be two lines of about equal length - close to what we would expect if everyone were perfectly rational all the time.
Posted by: Constant at Oct 31, 2006 8:26:16 PM
