In the Wake of the Gulf Oil Disaster, We Must Conserve Energy, Develop Alternatives, and Revisit the Choice Among Oil, Coal, and Nuclear Energy
	By NEIL H. BUCHANAN
	Thursday, June 3, 2010

The widening environmental catastrophe in the Gulf of Mexico, with a still-flowing gusher of oil destroying the ecology of an entire region and its waters, rightly continues to make headlines. By now, this crisis has reached the stage where politicians are finally feeling pressured to respond. Gone are the days of dismissive comments to the effect of "We need the oil, so we shouldn't overreact."

Although it might turn out to be simply impossible to fix the damage that has already been wrought, the jockeying for political advantage that we are now witnessing indicates that both parties have concluded that there is public outrage to be harnessed. Drilling and licensing moratoria are in place, and the betting odds favor more policy and regulatory reactions in the coming months.

As I argued in a FindLaw column posted shortly after the disaster's onset on April 20, the most fundamental lesson of the Gulf disaster is that America's consumption of energy is expensive — more expensive, in fact, than we have ever been forced to acknowledge. Our energy consumption has always been much more expensive than the price of energy would suggest, because we have failed to include all of the costs of energy in the prices of gasoline, electricity, and so on. When people die from respiratory diseases, or when entire forests die from acid rain, the hidden costs of energy consumption become overt — but the costs still are not borne by those who are ultimately responsible for those harms: the users of energy. That is, by us.

In this column, I will discuss the new realities of energy policy, in light of the changed calculus of costs and benefits after the Deepwater Horizon explosion. By far, the "cheapest" path is to increase conservation efforts. Beyond that, however, the new reality requires a re-examination of the choices for the immediate future among oil, nuclear, and coal to fulfill the bulk of our energy needs.

None of the choices is even close to being cost-free, but there is now a fairly clear least-bad choice among the three: Very reluctantly, I conclude that it is coal. Even so, it is essential to remember that this reliance should last only for the shortest term possible, until non-fossil-fuel alternatives become more widely available, and as the effects of stepped-up conservation efforts come to fruition.

Small Costs and Large Unknowns: Taking Low-Probability Disasters Seriously

In my recent FindLaw column, I discussed the notion of "externalities" in the context of energy usage. Energy production and consumption, as currently practiced, allow some of the most serious costs of energy use to be ignored by private decision- makers. Therefore, the producers and users of dirty and dangerous forms of energy have clear incentives to expand their operations well beyond the level of energy production that they would choose if they were forced to face — and internalize — all of the consequences of their actions.

What makes the Gulf disaster unique, however, is the role that low-probability disasters with extremely high costs play in the determination of the "true costs" of a particular choice of energy source. When one source of energy carries with it the risk of extremely high costs that would result from highly unlikely events, what is the best way to compute the costs and benefits?

The simplest analogy is to a game of chance. Suppose that you were given the choice of betting one dollar on the possibility that an unbiased random-number generator would choose your number from among one billion numbers. How much would the payoff have to be for you to play? Setting aside cognitive preferences such as risk- aversion, your answer should be $1 billion. Conversely, how much benefit must we gain from an activity that has a one-in-one-billion chance of resulting in a $1 billion disaster? The answer — which is notably more troubling, from an intuitive standpoint — is one dollar. That is, one dollar of benefit would exactly balance $1 billion worth of disaster costs, if that disaster has only a one-in-a-billion chance of happening.

In the context of energy production, one could imagine even a non-corrupt energy regulator assessing the option of deepwater oil drilling—pre-April 20—by calculating the probability of a major ecological disaster — which at that point may have seemed low — and its likely costs, and concluding that the risk-adjusted costs were not as great as the benefits of increased oil production. Of course, for the cost-benefit analysis to have been, at that time, used correctly, it would have been necessary for the regulator to require that those risk-adjusted costs be included in the price of oil, which was not done; but conceptually, it is at least imaginable that one could have looked at the potentially high costs of a major disaster and still have concluded that drilling should proceed.

The reactions to the Gulf disaster can thus be seen in one of two ways. Either people are now over-emphasizing the still-tiny likelihoods of environmental disasters, just as people statistically overreact to the possibility of earthquakes in the immediate aftermath of a major quake, or we are now newly evaluating our choices based on new evidence. In my judgment, even if many people are guilty of the former reaction — which would be an over-reaction — it is entirely appropriate to engage in clear reflection that takes into account what we know now that we did not know previously.

How Can We Even Measure the Likelihood of Very Rare Events?

Being able to compute risk-adjusted costs requires, of course, that we accurately measure the costs, as well as the risks, before they occur. Both of those measurements, however, are problematic when there is no basis of experience upon which to base one's estimates.

For example, how could anyone have made even a ballpark guess, ahead of time, of what the costs of the Deepwater Horizon explosion and its aftermath would be? (After all, we still, more than a month after the fact, have no idea how high these costs will ultimately turn out to be.) We have never seen anything remotely like this, and the analogies from smaller-scale disasters such as hurricanes and tanker spills are imperfect at best. We do not know in advance, for example, whether the costs of a uniquely large disaster like this one will be proportional to the costs of smaller disasters or will, instead, rise exponentially as the disaster's magnitude expands.

Similarly, there is no statistically-reliable way to measure the likelihood of a low-probability event of the sort that the energy sector so often presents. Familiar statistical forecasts such as those predicting life expectancies, touchdowns per game, and educational costs per pupil are based on large caches of data; and the forecasts derived from such data are regularly tested against reality and updated appropriately. If one predicts that an event will happen once every billion years, however, how can that prediction be tested? If, ten years after the prediction, the event still has not occurred, there is no way to know whether the forecast is accurate or not.

Of course, it is not appropriate to respond to the limits of such forecasts by simply ruling out all high-cost, low-probability activities. Still, outcomes like the Gulf disaster remind us that our collective sense that the risks of certain high-cost, low-probability events are acceptable may simply mean that we are living in denial. Sometimes, a newly-licensed driver learns the hard way that reckless high-speed driving has worse consequences than he thought it would. If he lives through the experience, he will (one hopes) change his behavior.

Regulatory Failure and Environmental Disasters: What Should Regulatory Solutions Look Like?

The federal and state governments, of course, already have in place various legal regimes that are designed to alter the choices between production now and disaster later. As we have learned from the Gulf disaster, however, there has been a bipartisan willingness (more pronounced among Republicans, to be sure, but hardly alien to Democrats) to rig the regulatory system to ignore the consequences of, say, environmental waivers for deepwater drilling.

Under the current klieg lights of scrutiny, it is easy to imagine that the regulators can be brought into line to do their duty in the future. Such an outcome is certainly what everyone should desire. When the news cycle turns, however, we know that there will once again be major failures of the regulatory agencies, due to corruption, myopia, and simple neglect. When that happens, we can be sure that activities with high-cost outcomes will once again be inappropriately approved.

In the face of that reality, what should energy policy look like? The most sensible change is to intensify our efforts to reduce energy consumption. Conservation policies can begin to pay off nearly immediately. Even though they seem expensive up front, such policies are in fact very inexpensive compared to the true, long-term costs of most current energy sources. They also have no "big disaster" downside. At worst, if we "conserve too much," we will simply have caused the economy to be smaller than it otherwise might have been. That is unfortunate, but it hardly compares to what we are seeing in the Gulf of Mexico.

Most non-fossil fuels at least promise a similar outcome. Solar energy, for example, offers the prospect of high returns on its costs. No one has suggested, as far as I am aware, that a major disaster could result from expanded solar power. (Granted, even "clean" energy sources can be based on hardware that has been produced from environmentally-damaging components, such as mercury — meaning that the composition of such hardware must be monitored closely. But simple and relatively easily-enforced regulations can simply ban the use of such components.)

Until conservation and alternative energy sources can truly change the energy landscape, however, the question is this: What should we do about our "base" energy sources: oil, nuclear power, and coal? We now have direct evidence, in the form of the Gulf disaster, of how much more costly oil production is than we used to believe. What about nuclear and coal?

The advocates of nuclear power, including President Obama, suggest that energy can now be produced in nuclear plants in an environmentally-responsible way. Such an assessment, however, assumes — as it did in the case of offshore oil drilling — that there is some way to determine, with some degree of accuracy, the costs and probabilities of low-likelihood events such as nuclear plant meltdowns, contamination from radioactive wastes, and terrorist attacks on nuclear reactors.

Given our new and hard-won knowledge of just how badly regulation can go awry, it is simply not very reassuring to be told that new nuclear technologies pass the cost-benefit test. That is, after all, exactly what the Energy Department (under both Presidents Bush and Obama) said about deepwater drilling. How much would it cost if a major American city became uninhabitable, with numerous Americans left sick, dying, or at risk, due to a nuclear accident or terrorist attack?

Coal Has Important Costs — But No Massive Downside Comparable to the Gulf Disaster

What about coal? Tragically, we know that coal production costs lives — most vividly when coal mines collapse and trap miners inside, but also as part of the daily dangers of drilling caverns underground. Just this April, a West Virginia mine explosion left twenty-nine dead. We also know that coal mining is environmentally damaging to the areas being mined, especially when entire mountaintops are blown off and the wastes are allowed to fill the surrounding valleys. All of these activities must also be regulated — and, again, regulated at least as well as fallible humans can regulate them.

What seems not to be in the cards with coal, however, is any Deepwater Horizon-like (or Chernobyl-like) high-cost, low-probability event. Coal is, of course, a major contributor to global warming. The probabilities and costs of global warming's worst consequences are also difficult to measure, for exactly the same reasons described above. If the choice for the next few decades, however, is between increased reliance on coal, oil (which also contributes to global warming), or nuclear power — while we aggressively promote conservation and alternative sources to change the energy mix as quickly as possible — then it seems that coal is the least bad choice among a number of unattractive alternatives.

Again, none of these choices are good ones; and all of them have been under-priced for generations. What we now know for certain is that the regulatory system is much worse than we thought at providing ongoing oversight to try to align social costs with benefits in the energy sector. That realization now calls for a change in the structure of the energy sector, as opposed to tinkering with the regulatory apparatus.

Until the energy mix can be fundamentally altered, however, we should reconsider our confidence that major disasters are unlikely and that they can be managed effectively when they occur. That was a nice dream, but it is long past time to wake up.

Neil H. Buchanan, J.D. Ph. D. (economics), is a Visiting Scholar at Cornell Law School, an Associate Professor at The George Washington University Law School, and a former economics professor.