Volcano Monitoring and the Stimulus: Cost Effective and a Clear Public Good

I've already talked about the basic dishonesty Bobby Jindal exhibited when he took a swipe at the mention of "volcano monitoring" in the stimulus - Jindal claimed that there was $140 million in there for "volcano monitoring", when it's actually only one of a number of projects listed under that line - but there's something more important that I didn't discuss. I took a swipe at the messenger, but what about the message? Jindal may be a liar, but that doesn't make him wrong.

He is wrong, of course. He delivered the argument dishonestly, but the argument still fails on the merits. Volcano monitoring is a legitimate governmental function, and it would still be a good investment even if we were spending the entire $140 million on nothing but monitoring.

Cost:

Before I get into the public policy questions, let's take a quick look at the costs. Volcano monitoring is (as many others have already pointed out) something that needs to be done if you want to avoid losses of life and property in a volcanic eruption. Unless you're near a Hawaiian-type volcano, with it's picturesque slow-flowing basaltic lavas, you really need to get out of the way before the mountain goes boom and falls down. If you want to be able to get out of the way before the insanely hot wall of burning rock, mud, and ash hits you, you probably want to have someone monitoring the thing.

The best example we've got of a case where volcano monitoring has worked really well is the 1991 eruption of Mt. Pinatubo in the Philippines. The Philippines Institute of Volcanology and Seismology had been monitoring the volcano for a while, and when it started to show signs of life USGS geologists came to assist. The US Air Force, acting on the advice of the USGS scientists, evacuated Clark Air Force base.

(Source: Wikipedia)

That's a 1990 picture of part of Clark's flight line. Take a look at the row of aircraft. The three or so visible at the far right are F-4 Phantoms, which were being phased out of service at the time, but the 12 planes to their left are F-16 fighters. The early model F-16s cost $14.6 million each (in 1998 dollars). Twelve aircraft times a $14.6 million unit cost works out to $175 million. That, you'll note, is well under the $140 million that Jindal (falsely) claimed was going to be spent on volcano monitoring, and that's just those 12 aircraft.

I think we can reasonably say that volcano monitoring is cost effective.

Public Policy:

OK. Volcano monitoring is cheap. But is it something that the government should be spending money on?

This is a subject I'm reasonably comfortable addressing, particularly in the aftermath of my recent debate with Timothy Sandefur. During that debate, Tim gave us a good explanation of what a "public good" is in the technical sense of the term:

"Public goods" doesn't mean "good for the public," which science obviously is. It means something that is non-rivalrous in consumption and non-excludable: something you can't keep people from stealing and something that everyone can enjoy at the same time.

Volcano monitoring seems to fit that definition reasonably well, at least for a given definition of "everyone". Anyone who is near a potentially active, monitored volcano benefits from the monitoring - they've got a greatly improved chance that they'll get sufficient warning to clear the danger zone. That's true regardless of the number of people in the danger zone, so it's clearly non-rivalrous in consumption. It's also non-excludable. Even if we ignore the moral difficulties that would ensue if only people who paid for a monitoring service were warned, it's still impossible to keep people from stealing the service (if half your neighbors all suddenly decide to go on a long vacation in overloaded cars, you're probably going to be smart enough to clear out).

Volcano monitoring is clearly a public good. The next question we have to address is whether this is a public good that can or should be provided by private enterprise. It might be possible to make a case that it could be, but it's going to be a real stretch.

If you had to come up with a group that would have an economic incentive in doing volcano monitoring, the insurance industry would probably be the first one that comes to mind. If you take a minute or two to think about it, though, they don't really have all that much of an incentive to do full-time monitoring.

Remember, most of the insurance losses in an eruption are going to be property losses. Most of these losses are not going to be preventable. You can move yourself and your car out of the hazard area, but (Hawaii again occasionally excepted) you're not going to be able to move your house. Industries won't be able to take their heavy equipment, and commercial outlets aren't going to be able to move most of their stock out. That creates an incentive to do periodic hazard assessments in order to set appropriate premiums, but it doesn't provide an incentive to do continuous monitoring.

There's a bit of a difference when you look at life insurance, but there's not much more incentive there. Volcanoes are typically going to represent a low risk compared with day-to-day life most of the time. The number of insurance companies that are typically writing policies in an area is going to cut the risk for each individual firm, and it's also going to provide an incentive for some of the firms to try to get the free ride by refusing to participate in a monitoring program. Their ability to buy reinsurance is also going to reduce their incentives. The reinsurers, in turn, have the same free rider issues, and most of their risk is also going to come from the (non-preventable) property losses.

If there's an industry other than insurance that would have a motive, I'm not sure what it would be. Volcano monitoring isn't just a low-cost public good, it's a low cost public good that is unlikely to be delivered by private firms. We've met the classic conditions for a service that's best provided by the government.

The only question we've got left is whether this is best done by the federal government, or whether it's better left to states and/or localities.

Each of the volcanoes at risk of erupting is found in a single state, which would seem to be an argument in favor of making this a state function. However, the effects of an eruption are not necessarily limited to a single state. The Mt. St. Helens eruption, for example, dropped measurable amounts of ash across a large chunk of the country. A Yellowstone eruption is unlikely, but potentially has global effects.

The active volcanoes are not restricted to a single state. They're spread out over a number of states. A single federal monitoring program also has quite a few efficiency advantages over a many separate state programs.

Finally, the volcano monitoring also fits into the existing federal role in interstate commerce - in particular, it fits into aviation safety. In 1989, a KLM 747 with 245 people on board lost power in all four engines as a result of flying through a volcanic ash cloud. The plane landed safely, but that's obviously not an event that you want to repeat (particularly since the plane suffered $80 million dollars in damage).The NTSB noted that one of the causes of the incident was, "the lack of available information about the ash cloud to all personnel involved."

Stimulus:

The last question we need to address in order to see if Jindal was making any sense at all is whether spending money on volcano monitoring represents a good stimulus. The stimulus, after all, is supposed to be a short-term funding boost. Volcano monitoring is a long-term project. If we fund the volcano monitoring with stimulus monitoring, what's going to happen when the short term funds dry up?

In this case, I don't think we need to worry. If we take a moment to look at the language in the bill, most of these concerns should dry up. The law specifies that the money is to be spent on:

...repair, construction and restoration of facilities; equipment replacement and upgrades including stream gages, and seismic and volcano monitoring systems...

This isn't money for the program as a whole, or for things like salaries that are going to be an ongoing expense. This is money that's going to be used to improve the infrastructure that the volcano monitoring system uses. It's money to replace older equipment with newer, and to do things like repair the buildings that are used. It's going toward infrequent expenses, and it will go back into the economy in the form of purchases from equipment vendors and payments to contractors. In the short term, it's the government acting as a consumer when citizens are reluctant to spend. In the long term, it's buying us a better volcano monitoring system - basically, we're spending the money on something that we have a use for.

Jindal:

Like Palin with her "fruit fly research" complaints, or McCain's moaning about bear DNA and "overhead projectors," this is clearly another case where a Republican in the national spotlight is saying things about science that just don't make sense. I'm not sure if this is because they think that bashing science is a good way to score cheap political opponent, or if they just don't understand how any of this science ties into public policy. Both are bad; I don't know which is worse.

But it really has to stop.

Update

I just had a chance to read through some of the backlog in my google reader. I was relieved to note that Paul Krugman cited Volcano Monitoring as an example of a classic nonrival, nonexcludable public good.