Safety, at what price?
A few days ago I was sent an interesting article by Mad Rocket Scientist, Regressive Effects of Regulation, authored by Diana Thomas at George Mason University’s Mercatus Centre.
The paper’s central focus is on a trade-off that we all deal with every day, and is a major focus of government policy – the trade-off between prosperity and safety. Every time a government mandates a new safety precaution (or not), requires more inspections of industrial sites (or not), or decides to beef up airport security (or not) it is making a trade-off that many people are uncomfortable discussing, but is essential to the policy-making process: trading off human life against money (or equivalent value that can be easily converted into money).
It seems repugnant to compare something as important as life to something as routine as money, but it must be done, otherwise how do you decide whether making cars 1% more expensive to save 100 lives per year is a good idea? The only alternative to making a trade-off is to treat life as infinitely valuable, which sounds good until you realise that this would most likely lead to speed limits being reduced to 10 km/h, with violators being sentenced to life in prison. If we want to be able to weigh lives against anything but other lives we’re going to end up with a trade-off of money for life, directly or indirectly. It’s not just governments that face this choice either, every time you get in a car you are making a trade-off between the convenience of getting to your destination, and the probability you will be killed or severely injured during the trip. Equally, anyone who works in a dangerous occupation is implicitly trading some of their expected lifespan for cash.
So what does this have to do with the article? Well, Thomas’s argument is that the current level of safety regulation in the US leads to sub-optimal trade-offs for low-income people.
Her argument runs as follows:
- Safety regulation is set to please the wealthy and middle class, who hold most of the political influence.
- Given their constraints, the poor demand less safety than the rich.
- Therefore the poor could be made better off by loosening safety regulation.
I won’t bother discussing step 1, I think we can take it as read. The more interesting one is premise 2. What Thomas demonstrates, by looking at how much people of different incomes are prepared to pay for a reduced risk of death, that safety is what economists call a Normal Good, which means that richer people consume more of it than poor people (this is true of most goods, which is why they are called “normal goods”). This means that poor people are more likely to choose a cheaper, more dangerous option than richer people. This should come as no surprise, poor people have to stretch their budgets further so they will compromise on a range of non-monetary values. But this has policy implications – greater safety regulation, even where it makes rich or middle-class people better off, may make poorer people worse off, as they value the safety less and find the increased product costs as more problematic that richer people do.
So what are the policy implications of this work? The simplest option would simply be to focus primarily on the desired safety-income trade-offs of poor people These will produce too little safety for the rich, but they can always buy products that have a higher-than-mandated safety level. However, for our less-libertarian readers, there are some other policy options to consider:
- Subsidies over Mandates – instead of the government mandating a safety level, it could subsidise safety features. This would lead to safer products, but the taxpayer would bear most of the cost of the safety instead of consumers. This would promote safety while diminishing the adverse effects on the poor.
- Direct Transfers – If your preferred policy adversely affects the poor, but you don’t want to change it – compensate! Of course the trick is getting the rich and middle class to go along with it.
- Multiple Safety Ratings. Instead of having a single safety level, you could have multiple levels, and products could then be labelled with which safety standard it passes (this happens with some products already, though I think most examples occur through the private sector).
Tracing the distributional effects of a policy can be difficult and time-consuming. Often the real winners and losers can be a group of people you don’t expect. So if you want your policy to disproportionately help the poor (or at least avoid disproportionately harming them), it’s important to take some time to work out who the winners and losers of your policy will be and, once you know, what you plan to do about it.