On Risk Policy

            ‘The greater the risk, the greater the reward’ is a saying that holds true in many circumstances. The question is: is it better to accept a greater risk and the potential for a greater reward, or is it better to accept a lesser risk for a lesser reward? There is no objectively right or wrong answer to this question. It will depend on the nature and magnitude of the relevant risks and rewards being considered, and on the subjective risk tolerance levels and subjective valuations of the individual decision-maker.

            

           What does this mean from a policy point of view? It means that, ideally, if we want as many people as possible to derive as much utility from their actions as possible, individual people should decide, based on their own risk tolerance levels and subjective value judgments, what risks they are willing to take on for what rewards.

            

           In order to approach this ideal, two pre-requisites must be met, to a greater or lesser degree. They are: 1. Sufficient Knowledge: It would be best if the decision-making actor knew the nature and approximate magnitudes of the risks and rewards in question. 2. Linking Risk to Benefit: If an actor takes on a certain amount of risk, it would be best if he were recompensed via the prospects of greater benefits, proportional to the amount of risk taken on. Thus, if an actor takes on all of the risk of a course of action, he should also get most of the reward. If the actor takes on half of the risk of a course of action, he should get almost half the reward, etc… 

            

           A real world decision that probably approaches this risk evaluation ideal is the decision of whether to wear a bicycle helmet while biking or not. Most people know how much they prefer (in terms of comfort and style) not wearing a helmet to wearing a helmet. Most people also know that not wearing a helmet puts them at greater risk of head injury if they get into an accident. If we assume that there is no socialized medicine (which is a perfectly proper assumption for a libertarian-leaning policy analysis), the main bearer of the risk of not wearing a bike helmet is the actor himself; it is his own head’s safety that is on the line. The benefit of not wearing a helmet in terms of comfort and style obviously accrues to the biker himself. Thus, the two prerequisites are met to a high degree: the actor has sufficient knowledge of the nature of the risks and rewards, and risk is linked to benefit for the actor.

            

           In cases like this, I think it is best to just let the individual decide for themselves. The policy of trying to ‘protect individuals from themselves’ is unwise. It causes harm to the individuals in question, who aren’t going to be grateful for this ‘protection’ and will resent it. It sets dangerous political precedents, as any tyrannical rule could be justified based on this excuse. And the only real beneficiaries of these policies are nosy busy bodies and politicians/bureaucrats who love meddling in other people’s affairs just for the sake of it, attitudes that it would be best not to encourage. 

            

           Unfortunately, not every case where risk is involved approaches the risk evaluation ideal as fully as cases such as the bicycle helmet decision. The ideal and its two prerequisites are still good things to keep in mind though, even when analyzing these more difficult cases. I submit that the goal of policy in the field of risk should be to advance these two prerequisites as much as possible, and thus approach the risk assessment ideal as much as possible, in as many circumstances as possible. When faced with harder cases, the temptation for many people is just to call on coercive governments to do the risk assessment and to make risk-related decisions for people. This is not a solution to the risk problem though: it’s just a call for the problem to be ‘solved’ in a largely arbitrary fashion and for the ‘solution’ to be uniformly imposed on masses of people, without properly taking into account the subjective preferences of individuals.  

            

           Let us consider one of these harder cases now, and see what could be done to improve the situation. Take the decision of whether to build a nuclear power plant in a particular region or not, given our present institutional arrangements. Let us check our prerequisites: 1. Most people have only sketchy notions about how safe nuclear power plants are. In addition, the particular design and location of the plant can significantly change how safe the plant is. 2. The main benefits of a new nuclear power plant take on the form of lower electricity rates for residents of a very large swath of territory around the plant. In the event of a nuclear meltdown though, those most at risk would be those living in a much smaller swath of territory around the plant. The result of this disjoint is the prevalence of the ‘not in my backyard’ mentality, coupled with complaints about high electricity prices when no power plants can be built, due to the local opposition.   

            

           Thus, the two prerequisites are not met sufficiently, but in addition, it is pretty much impossible for individual decision-making to prevail in this case. If two neighbors disagree on whether they want the plant in their area or not, they can’t both have their way, as they could if they disagreed about whether or not to wear a helmet while biking.

            

           Before throwing our hands up in despair and calling for the government to nationalize the nuclear power industry and to force people to accept their choices of power plant location, let us see if we can’t find a way to approach the risk assessment ideal more closely for this case. So that we have a clear idea of our goal, let us begin by imagining a fantasy scenario that approaches the ideal. Imagine that everyone knows the percent chance per year of a nuclear meltdown for each plant design and geographical location. Imagine that the nuclear plant is somehow equally divided throughout the whole large region that is benefitted by the lower electricity prices, such that the total risk of radiation in case of meltdown from the one plant is spread around this large region evenly. Imagine that individual people can decide whether to opt in or out of taking on the risks associated with the plant. If they opt in, they get cheaper electricity prices. If they opt out, in the event of a meltdown they are magically protected from the radiation, but must pay higher electricity prices. In this fantasy scenario, individuals are empowered to make the decisions themselves, based on their own subjective valuations, and the prerequisites for the ideal are met perfectly.

            

           Now, this scenario could obviously never become a reality. Nevertheless, there are things that can be done in the real world to approach this ideal. One thing that could be done would be for the power company to offer to pay the residents of the at-risk area in exchange for allowing the plant to be built in this area. Since contracts with individual residents would be unfeasible, bargaining with nuclear power companies could fall into the purview of the regional air protection associations (an association of all of the residents of the region, in their capacity as collective owners of the region’s air. See http://thinkingabouthumansociety.blogspot.ca/2013/03/thoughts-on-environmentalism-pollution.html for a full discussion of this.)

            

           As laid out in the associations idea, members of the association (all the residents of the region) would receive equal cash dividends, based on the compensation amount agreed upon by the association and the power company, if the power plant were allowed. This way, though the economic benefits of the power plant are still spread out throughout a large geographical area, buying the consent of the more immediate region’s inhabitants would now constitute a cost of production of building and operating the power plant. With a higher cost of production, energy consumers would have to pay a higher price for the power than if consent were not needed. Hence, the large pool of beneficiaries ends up funding the special monetary benefit that goes to those whose regional association agreed to bear the brunt of the risks.

            

           As for the knowledge prerequisite, in a free-market, it is likely that there would be respected third-party, private nuclear safety certification associations who would put their reputations on the line by rating the safety of proposed nuclear power plants. A reputation for integrity and a high level of expertise would be the main business assets of such associations, assets which they would seek to improve and protect. These associations could be commissioned by air protection associations, by power companies seeking a third-party stamp of approval, or could just receive their revenue as a pure information provider, selling information to consumers or selling ‘eye time’ with these consumers to advertisers. If these associations did their job well, which I think the forces of free-market competition would push them to, then these expert assessments could give those living in the region to be affected, a better idea of the magnitude of the risk they are being asked to take on. That way, they can make their decision with the benefit of more relevant knowledge.

            

           Thus, while still not the ideal fantasy scenario imagined above, these institutional tweaks would greatly improve the prospects for rational, individual-based risk assessment procedures in this hard case. While these institutional tweaks might vary from case to case, something similar could probably be thought of for many cases where risk assessment needs to be improved.