A policy conversation about gun violence

Here is a very real question: having agreed that one school shooting is too many, how many are acceptable in a free society that allows gun ownership?

WARNING: This article contains a discussion of statistics, which I know very little about, and probably everything statistics-related in this article is erroneous. But I provide links to the information I am discussing. So feel free to draw your own conclusions.

Let us consider what the actual rate of school shootings is. Your child is likelier to die while traveling in a motor vehicle to school than to be killed in a school shooting. Your child is likelier to die in a bicycle accident than to be killed in a school shooting. Your child is likelier to drown in a pool than to be killed in a school shooting. Schools are generally quite safe from being shot up.

So what we are talking about is making marginal gains through policy. I don’t mean marginal in the sense of utility; I mean marginal in the rate of gain. The disconnect is between the urge to do something (or nothing) and the utility of those decisions. We are looking at situations where quite literally a change we make today may take 20 years to be measurable in figures that people can appreciate.

If we can, through legislation, achieve an annual rate of decline in school shooting fatalities of 2%, what is the economic and opportunity cost we are prepared to pay for that decrease? Is a decrease in fatalities the goal, or a decrease in incidents? What measures will accomplish both? These are the hard questions that people are not asking, will not ask, and are not interested in addressing because it is an admission that people will still die. That children will still die.

There will never be a time when there are no school shootings. Even with a total gun ban in America, there will still be school shootings. So the questions are, in my mind, the following:

1. What is the short term goal?
2. What is the long term goal?
3. How do we measure progress toward each of the goals?
4. What is considered successful? When is the mission accomplished?
5. What is the price we are putting on a given marginal return of improvement in individual safety?
6. What infringement on liberty are we willing to pursue to gain any improvement in individual safety?
7. What cost is too high, in money or liberty, to save a single life, if at the end of the day we have one person die per year in school shootings? Put another way, how many dead children do we assume are a natural consequence of life?

Before we say that we cannot put a price on a child’s life, consider that the legal system does this all the time in personal injury and wrongful death cases. In an imperfect world, we need to be able to examine these questions. Alabama spent years on a pilot project to study the effectiveness of seat belts in school buses. They concluded that to save a single life in Alabama through the use of school bus safety belts would cost $8 million dollars and take over two years. In the end, the state decided that that price was too high, particularly since there were less expensive, more effective strategies.

How did Alabama figure this out? First, they had a goal: to evaluate the effectiveness of seat belts in school buses. Second, they had a school district to operate as a test case. Third, they had a time frame for effectiveness. Fourth, they had a heuristic for evaluating the data collected. Fifth, they had the legislative will to evaluate the results of the data seriously. When the data concluded that life one would be saved after a cost of $32 million dollars over a period of ten years (at a cost of $7,000 to $10,000 per school bus in the state), the state concluded that there were less costly ways to save student lives than putting seat belts on school buses, which are statistically the safest way to get to school.

There are several types of gun control measures generally permissible under the United States Constitution and the Second Amendment. With regard to the trade in guns, only Congress can directly affect interstate commerce; broadly speaking, Congress can use the interstate commerce clause to prohibit in a substantial degree the exchange of guns, ammunition, and accessories over state lines. This is what the so-called Assault Weapons Ban of 1994 addressed.

Other, more meaningful gun control policies must be enacted at the state level. In some states, such as Mississippi, this is effectively impossible, because of the right to bear arms enshrined in the state’s constitution, which would require a constitutional amendment to address. Additionally, in most states, there are laws on the books that prevent local passage of gun control, enshrining state preemption over political subdivisions (counties, municipalities, etc.). Serious gun control regimes must therefore arise at the state level. Serious does not mean severe, here. It means rigorous, i.e., developed with a systematic approach towards a measurable goal.

The thing that prevents us from taking this approach is that no one wants to do a true cost-benefit analysis of gun control. Democrats are afraid that the cost of effective gun control per life saved is too high. Republicans are afraid that the broadest gun control measures are the least expensive and most effective. In this sense, the ban on federally funded public health research into guns is useful for both parties, because they can continue to make their cases to their bases without regard for the evidence.

Semi-automatic high-capacity small-calibre rifles are highly lethal. So are handguns. Most mass shootings occur at relatively close ranges, because it is harder to hit a target at a longer range. Most shootings in the United States involve handguns. Very few involve semi-automatic high-capacity small-calibre rifles, but those that do have a generally higher casualty rate than those that do not. All of these facts, however, are somewhat inconsequential because we do not know what our goal is.

I recommend that people who are serious about determining what is the best, most effective policy for preventing child deaths in school shootings examine their assumptions, determine what their actual goals are, how to measure them, and get to work studying the problem. It is painfully apparent that they cannot answer some very basic questions that are necessary for us to have an effective policy discussion. For the sake of argument, I am accepting the idea that we can achieve measurable gains in student safety through expanded gun control. I am also assuming that any such gun control will comply with the Constitution (which presupposes the idea that the 2nd Amendment may be repealed as part of it). I want to know what would be considered success. I want to know the cost of the successful measures.

Using the figures from the Alabama school bus safety belt study: If it costs $32 million dollars to save 4.27 children per 10 years per state, for example, are we willing to spend $1.6 billion dollars nationally to save less than 22 children per year ($7.5 million per child, which is a $2 million less than the NHTSA value of a statistical life for 2016)? If so, what is the attendant liberty cost? How do we measure the cost of being unable to purchase a handgun for self-defense, which the Supreme Court currently considers the measure of the limits of the 2nd Amendment? NOTE: These numbers are unlikely to be comparable, since we average about 22 student deaths per year from school shootings, and there are relatively few school shootings annually, whereas many children ride buses to school every day. The costs may be lower, or they may be higher.

It cannot be wrong to ask these questions if we want to address the very real problems of our violent society. It is by most measures far less violent than it has ever been. But it is still violent. If we want answers to these questions, we need to ask them.