A short history of FEMA:

Courtesy of Kevin Drum, read the whole thing.  Hindsight is easy, but what should FEMA be doing?

My view is the following.  Many levees are genuine public goods, and should receive government support, from the federal government (e.g., Army Corp; here is a brief history of their involvement) if need be although perhaps not ideally.  FEMA should not be in the business of flood insurance, nor should FEMA reimburse local governments for snow plowing.  Here is a Cato critique of FEMA.  Here is a libertarian article on why a limited governmental response to the Chicago fire was best.  Here is another libertarian critique.  Here is an AEI article that FEMA invests too much in earthquake safety.  Here is an argument that FEMA should not have been made part of the Department of Homeland Security.

Here is a recent piece on cuts to levee subsidies; the news will hurt the Republicans.  Here is a short piece on how revenue from airport privatization could have been used to shore up New Orleans levees.

Libertarian readers, do you care to argue the levee should not have been subsidized?  Do you favor real privatization, not as a Port Authority or Federal Reserve may be private, but in the true market sense?  (Here is a short history of the Louisiana levee authorities; their status has evolved over time.)  If you take that position, you have a few alternatives:

1. We rely too much on unreliable levees, and privatization/non-subsidization would reveal their true social costs and induce people to move elsewhere.

2. A privatized, non-subsidized levee would engage in a successful long-term contract with city residents; see the Demsetz-Williamson debate.  The government still would have to force residents to make the relevant tax payments, for free rider reasons.

3. A levee contract could be written without use of coercive taxation; see this piece on assurance contracts.

4. A private levee authority would invest in water safety out of fear of being sued.  Furthermore these ex post legal incentives would be reliable and would not involve more government intervention than ex ante regulatory incentives.

5. A private levee authority would be forced by its insurance company to build good protection and also hold huge capital reserves.  Their cost of capital and costs of production would remain lower than the government's.  You can hold this position in conjunction with #3, or believe that coercive taxation would remain necessary.  But in any case it probably requires reliance on #4.

I am not willing to defend any of these five positions, but what do you say readers?  The current government system, obviously, does not have a sterling record.  Comments are open, and note I have placed a similar post at (with separate comments), Eugene has suggested I do some selective cross-posting.

Andy Freeman (mail):
Why does "public good" imply "federal government"?

Shouldn't the good folks in New Orleans and/or LA decide whether keeping that particular piece of ground dry is worth the expense? After all, they're the primary beneficiaries. And, to the extent that that piece of dry land has extra value that the rest of us benefit from, they can tax our access to it.
9.2.2005 11:55am
DNL (mail):
My take is similar to Mr. Freeman's, above.

Levees are not a public good if you describe the market as "the United States of America." They are certainly excludable in that their placement dictates who gets their benefit. (That is, a levee used in New Orleans cannot be used in Biloxi as well.)

But if one rearticulates the market as "New Orleans" or even "Southern Louisiana," then it becomes non-excludable. (They are non-rival in any case.) So, the governmental body that should tax to fund them is the state/local one, not the federal, and even many ardent libertarians would find no problem with it.
9.2.2005 12:09pm
"A short history of FEMA"? So, Tyler, FEMA's history begins in 2001? Give me a break. That "history" was exactly the same in 2000 and before.

"Here is a recent piece on cuts to levee subsidies; the news will hurt the Republicans." Cuts to levee subsidies in fiscal 2006. What year is it now, Tyler?

Sheesh, pure uninformed Bush Bashing. I'd certainly have expected more intelligent criticism here.
9.2.2005 12:27pm
AppSocRes (mail):
The problem with the two earlier posts is they ignore the larger picture that federal/state/local government and private interventions on the Mississippi upriver of Louisiana and New Orleans, e.g., erosion and flood control measures in the great plains along the Mississippi and its tributaries, create problems downriver in the delta and environs. I believe there is a complex of common and statute law governing upstream use of riverine sources that derives from issues surrounding dams, watermills, and other diversions. Pure libertarian philosophy needs to be very carefully thought out in its application to this situation.
9.2.2005 12:30pm
Andy Freeman (mail):
Upstream activities do not affect sea-level or the level of the near-by lake, the two sources of the flood water. At most, they affect the variations in river-flow.

In any event, any connection is attenuated by distance and it disappears entirely once you're outside the area that flows into the Mississippi. For example, activities in CA have very little effect.

In short, it's not enough to say "other people's activities have an effect", you have to identify which other people and look at the magnitude of said effect. At some point, it's not "proximate cause" (or whatever the relevant term in this area of the law is).

BTW - some of the upstream activities reduce the cost of NO's levees by reducing the reasonably forseeable maximum flow. Should those folks get to charge NO?
9.2.2005 1:01pm
DC Lawyer:
The only intelligent comment yet is the one by AppSocRes. Mr. Freeman clearly knows nothing about coastal ecology and would be advised not to opine on it. Army Corps of Engineers projects, mostly to subsidize the region's oil and gas industry, prevent the natural siltration that should occur in the Mississippi Delta. The resulting loss of wetlands places New Orleans and the bayou communities at significantly higher risk.

Also do people really believe that the protection of New Orleans is a problem only for the people of New Orleans? If that were so, why are YOUR gas prices rising? What about the income New Orleans derives for the State of Louisiana and the regional if not national economy? You really think those of us far away derive no benefit from and have no responsibility for these places?

Its unquestionable that the levees are a public good and that there are some things governments must do. When you insist on "starving the government until you can drown it in a bathtub" (eh Grover Norquist?) this is the result. When politicians of all stripes prefer to fund pet projects for their district (A $200 million bridge to an uninhabited island in Alaska, Sen. Murkowski?) instead of the $100 million the Corps requested for levee improvements (they got only $40 million), this is the result.

It's not government that's the problem. Its the lack of foresight of our so-called leaders. And both parties are to blame.
9.2.2005 1:15pm
Shelby (mail):
Tyler: Here's another relevant data point. The Army Corps of Engineers says that even if they'd gotten full funding under the Bush Administration, it would have made no difference in the current situation.

Andy: It's not strictly true that upstream activities don't play a significant role. The degree of sedimentation, and the actual place where the river flows, are both dramatically affected by upstream river-control projects. Some experts apparently think the river would have completely changed its lower course (upstream of N.O.) absent extensive engineering.

That doesn't invalidate any of the political or economic principles here, though. We should also take into account the role the levees play in facilitating (or inhibiting?) commerce along the Mississippi. That does play a huge role in the national economy. My uneducated sense is that the levees have tended to aid commerce, in part by enabling New Orleans to play a consistent role in managing shipping and warehousing.
9.2.2005 1:18pm
AppSocRes is absolutely correct that the risk to any individual spot anywhere on the river is affected by the actions of those on other places on the river.

(An example: Davenport, IA does not have a levee protecting its downtown. The topology on the IA side at that part of the river is that the land rises quickly to bluffs, and so only the riverfront park and the adjacent 5-6 blocks flood, the rest is high enough up. If they were to build a levee, they would not be allowed to build it right on the edge of downtown with the park a devoted flood plain. Or even to build it right at the current river's edge. No, they would have to build it out into the channel, making the channel narrower. This is because the federal rules require some direct economic benefit in addition to the benefits of flood control, and so they need to reclaim land from the river. The problem is that narrowing the channel and not letting downtown Davenport flood is a double-whammy against both downstream and upstream communities, especially on the IL side where there is a broad plain of bottom-land. If Davenport builds a levee, it will make flooding much worse elsewhere. In this case the government screw-your-neighbor regulations are an attempt to create a market-like mechanism. The city politicians in Davenport are acting to protect the neighbors' best interests because they don't want the political flak. If there is a real market, then a private company will have a fiduciary responsibility to its stockholders to place the levee out in the channel and capture the economic value of the reclaimed river bottom because they will have no liability for the way that they increased flooding in neighboring communities.)

The other, even more intractable, problem is that from an engineering point of view you can argue that the levees around New Orleans did exactly what they were supposed to do and did not "fail" at all. For 200 years the levees have been engineered to decrease river flooding and increase hurricane damage. If the river had been allowed to flood New Orleans every year, then that plot of land would be covered 10-20 feet deeper with river silt, and would not be below sea level at all, and it would also be miles further inland. From a technology point of view, the only way to substantially decrease the risk from the river has been to substantially increase the risk from the gulf.

I don't see why a levee company should bear any liability for hurricane damage. If I hire a company to dig a hole next to my building to act as a flood retention pond, and then a flood comes and fills up my pond and spares my building, where is the failure? The nation hired the Army Corps of Engineers to do the equivalent of digging a hole to act as a flood retention pond for the Mississippi. Is it the Corps fault that we built a city in the middle of it?

cathy :-)
9.2.2005 1:27pm
Dallas (mail):
DC Lawyer's argument has a problem: it assumes that those refineries and port facilities could not be effective anywhere else and the assumption that the leevees are required to derive the benefit received by the rest of the nation.

New Orleans' port and refinery facilities do give benefit to the rest of the nation; that is not in question in my opinion. However, could those facilities have been built elsewhere and still provided the same benefit? I would argue that you could build/rebuild those facilities elsewhere. Whether it be on the Texas coast (likely Houston/Galveston) or farther east (could you not build port/refinery capabilities in Alabama/FL areas?), there are other places where refinery and port operations could be build at or above sea level? Opposing viewpoints?

For that matter, many of the refining operations are not even in the core New Orleans area protected by the leeves. That 'investment' to reclaim the wetlands areas is not even relevant to these refinery operations.

I am fairly libertarian and generally opposed to large scale public works programs. There is, in my opinion, no inherent right to build residential or commercial facilities in an area subject to natural dangers. Nor is there a right to federal protection of those areas in order to allow commercial activity. If someone wants to take those risks, fine. If someone wants to insure those activities, fine. If not, caveat emptor.

That said, I am not opposed to public works projects in a local area that are favored bay a majority of residents in that area. If the residents in southern LA favor leevees and vote favorably to be taxed in order to build and support them...then have at it.

I suppose I support the same thing at a federal level, though I imagine it would be a much more difficult referendum to pass. Explaining that only by building these leevees can the delivery of gasoline be assured seems a tough (and probably false) argument.

I have to support The Speaker of the House on this matter; I cannot imagine feeling good about spending tax dollars to rebuild the leevees and the accompanying commercial activity in those below sea-level areas. We probably will wind up footing the bill (as taxpayers), but it strikes me as a fools' errand.
9.2.2005 1:34pm
DC Lawyer:
They say a gaffe is when a politician in Washington tells the truth. I hate to say it but Hastert is right. We should think seriously about whether New Orleans, Biloxi, Gulfport and a lot of other little coastal towns should be rebuilt or returned to nature.
9.2.2005 1:44pm
Anon7 (mail):
Pesky little thing about ports--by definition, they pretty much have to be on or near the water.

As for the person who said to relocate the New Orleans port traffic to Houston or Mobile, I would like to see his or her plan for shifting the flow of the Mississippi River several hundred miles. It is its position on the Mississippi that makes (made) New Orleans the largest port in the United States.
9.2.2005 1:59pm
Shelby (mail):
Some quick-and-dirty research suggests there are many ways of measuring port size, and New Orleans/Southern Louisiana doesn't lead them all. In any event, though, an assumption of approximate economic rationality suggests the Port of New Orleans is there, and plays the role it does, because that made a lot of economic sense.

That said, we should consider whether to let the Mississippi adopt whatever its current "natural" course is, and rebuild N.O. according to that result.
9.2.2005 2:19pm
Cheburashka (mail):
Levees are plainly a public good that the government should provide, and plainly the governments primarily responsible are those of New Orleans and Louisiana.

How many billions does those governments spend each year on make-work jobs and "racial balancing" programs?

I don't think its appropriate for localities to hold themselves hostage by refusing to provide necessary services unless the Federal government pays for it.
9.2.2005 2:30pm
The end of the Mississippi is naturally a delta because that what happens when a big river meets an ocean at a place where the terrain is relatively flat. In other words, the deep navigable channel fractures into a gazillion tiny unnavigable creek- and stream-sized little trickles. Naturally, the river moves a gazillion tons of silt down the river and spreads it out over the delta.

When they started the levee system, they picked the largest of those delta trickles and said, "ok, this one is the river." Then they built levees up where the delta starts fanning out to prevent the water from spreading into the delta, so that they could use the water volume to make their chosen channel navigable. That decision there was where they made the trade -- allow the gulf to remove the delta so that we have a navigable river all the way out to the sea. The slow-and-steady forces of nature have been removing the delta slowly and steadily over a century, and now one of nature's fast-and-furious forces has removed a huge chunk of what remained.

The fundamental nature of this problem is that you can't have a river delta above sea level and not have the river frequently flooding over the delta land mass to deposit more silt and have a navigable channel all the way out to sea. Before trying to construct market mechanisms that will give us all three at once, how about stepping back and remembering that the Fundamental Theorem of the Market is that there is no such thing as a free lunch.

cathy :-)
9.2.2005 2:43pm
Abe Delnore (mail):
Galveston was one of the leading ports on the Gulf until it was destroyed by hurricane in 1900. This was on the scale of what you are seeing today in Gulfport and Biloxi. As the mayor of Houston remarked, but for the grace of God any of the proposed alternate locations could have been devasted.

New Orleans is in a peculiarly bad position from the perspective of hurricane survivability, but great benefit is derived from fixing the course of the Mississippi and locating a large port there. I suppose we could let Old Man River roll where he pleases, but this would mean major disruption of one of the nation's commercial disruption every few decades. The impulse to control the Mississippi and populate its banks was not entirely frivolous.

--Abe Delnore

9.2.2005 2:52pm
cirby (mail):
Any criticism of FEMA should, as a matter of course, start with a note that their primary function is to write checks, not to physically respond to disasters.
9.2.2005 3:06pm
Andy Freeman (mail):
> Mr. Freeman clearly knows nothing about coastal ecology and would be advised not to opine on it.

Let's review.

> the larger picture that federal/state/local government and private interventions on the Mississippi upriver

Mr. Freeman knows enough about the geography to know that "upriver" from NO is not coastal, and Mr. Freeman was responding to the upriver discussion.

I'll be happy to discuss coastal activities with someone who understands (geography again) that they are fairly local to NO.

Yes, ACOE has been involved. The question is whether federal money should have been or whether the costs of NO should be borne primarily by NO, much as the benefits are.
9.2.2005 4:30pm
Andy Freeman (mail):
> You really think those of us far away derive no benefit from and have no responsibility for these places?

The observant reader would have noticed that I pointed out that the locals receive the benefits, so the question is whether they should pay the costs. I'm pretty sure that one can build refineries elsewhere. Why shouldn't we build them where they can pay for themselves?
9.2.2005 4:33pm
Andy Freeman (mail):
> When politicians of all stripes prefer to fund pet projects for their district (A $200 million bridge to an uninhabited island in Alaska, Sen. Murkowski?) instead of the $100 million the Corps requested for levee improvements (they got only $40 million), this is the result.

The money for that bridge could not have come from AK. The money for LA's levees could have come from NO if it wasn't being diverted elsewhere, such as AK.

Pork is an unavoidable consequence of massive federal funding for local projects.
9.2.2005 4:37pm
tz (mail) (www):
I think I would be in the assurance contract camp OR, I would use a bond sale and place a lein on all the affected property since the values of all the property would rise because of the levee. That could be paid back immediately, or at the time the property is sold. Thus no one would have to pay a tax, but there would be no "free riders". I don't know what supermajority would be needed to impose such leins, but I would think 25% should not have a heckler's veto if it can be done without immediate taxation.

The larger problem is the lack of incentives for Government to really do a good job. For whatever reason, the levees weren't built or maintained properly. But that only shows up if there is a disaster. However a (typical for-profit) corporation isn't necessarily a good alternative. Consider Enron who hid debt "off the books". Would some other corporation hide the safety problems "off the books" and hope there would be no hurricane until after the CEO was in Paraguay? A hurrican may not hit anyway. Or in Enron's case, the tech bubble might not burst and energy might rise monotonically.

Corporations tend to try to maximize profits over the short term. Government tends to target voting blocks but in the same sense maximize profits (they might do a shoddy levee which would employ more people or benefit a contributor). You can have an occasional Marcus Aurelius, but they are statistically rare enough.

There is the Red Cross, or consider the opensource movement that engineered Linux, Apache, and Open Office. For some reason people think the alternatives are FEMA v.s. ENE. But there are other societal structures. Neither governmental nor corporate, but voluntary. Some of the Rural electric coops might be a better model.

I will also note that it is perfectly reasonable NOT to build houses where floods occur (or build them cheaply or on stilts, etc.). What happens seems to be Private Developer gives a little $ to politician who then spends X on a levee, which allows developer to make several times X on the now more valuable land. But the developer would not or could not pay for the levee.

What about the grandfathered buildings in CA?

What about the land in the path of any lahar from Mt. Ranier?

What about the New Madrid fault where there are no earthquake building codes?

What about a subduction of Hawaii or LaPalma that would create a large Tsunami in regions where that rarely occurs?

What about a Tambora or Toba eruption that gives us another "year without a winter".

The big problem with the libertarians saying "insurance will take care of it" is that you can't know risks. There are many 1/10K risks, and we usually see several in one lifetime, but we don't know what they are beforehand.

Another problem is that the oceans are cyclic. The Atlantic cycle is 60-70 years - does anyone remember the 1920s FL land boom and the Marx Bros. "Cocoanuts"? That ended with a series of bad hurricanes. But then the "calm" part of the cycle occured so people wondered why no one bought all this wonderful costal property. They are finding out. 20-30 years from now when no one wants it because of hurricanes, it will be safe to go back - but only for 30-40 years.

It was not certain that Katrina would hit New Orleans until it did - and it could have had a storm surge that was immediately over the levees, but didn't. It could have been stronger or weaker. We know things now, but I've already noted a lot of other potential but far from certain disasters we won't address.
9.2.2005 6:29pm
Thom (mail):
What is the conventional wisdom in welfare economics concerning the geographical component of public goods analysis?

Engineering issues aside (since I'm no engineer), levees that maintain a shipping channel (and do little else) would seem to be a public good to more folks than levees that maintain a shipping channel and protect low lying areas from even occassional flooding of brackish water. However, my welfare economics stury was so long ago and so superficial that I cannot be certain of this.
9.2.2005 7:10pm
GBCNH (mail):
All, I would be interested in comments on the related effects of the the federal flood insurance program, which similarly artificially encourages development activity in hurricane and flood prone areas (and, of course, encourages redevelopment in the exact same locations).

Regarding the impact of a "reduced" New Orleans, I think that we are going to have a fairly limited period of disequilibrium, after which various "key" business activities will shift. I read that already various import and export shipping activities are moving to other ports - import shipping to gulf ports, and export shipping might go out via the St Lawrence or other modal transportation. The longer the disruption in New Orleans, the more difficult it will be to re-establish the activity in NO or nearby locales.
9.4.2005 9:35am