What priors do you assume when you hear about a regulation?
“Priors,” or “prior probabilities” are a Bayesian concept that says, “we don’t know the attributes of a thing for certain, and perhaps if we investigate more we will find out things with more certainty, but from the start we know (uncertainly) SOMETHING about most topics.” So, for example, suppose that you hear your friend Bill has a new girlfriend. You probably will assume that his new girlfriend has, for example, two arms. You know that not every person in the world has two arms, but most do. So your prior that his girlfriend has two arms is probably something north of 95%. If you know Bill well, you might have some additional priors about his new girlfriend. For example, you might assume that she is likely to be black (60%) or white (20%), but less likely to be asian (10%), hispanic (5%), or other (5%), based on your previous experience with Bill.
It is apparent to me, when engaging people, particularly liberal people in the Bay Area or online, that many people have priors for regulation that look something like “this regulation addresses a real problem (80%),” and more broadly “this regulation is a good idea (70%).”
My priors look very different, particularly when dealing with local business regulation. I generally assume something like:
- This regulation was created to address a real problem: 50%
- This regulation actually addresses the problem it was designed to address: 95%
- This regulation does not create new and different problems as bad or worse than the problem it was designed to address: 90%
- This regulation is not out-of-date: 70%
- This regulation is enforced fairly: 50%
- This regulations costs to enforce are in line with its benefits: 50%
And I think that you can reasonably multiply those priors together to get an overall prior that “this is a good regulation”: 7.5%
Why is my view of local business regulations so dismal?
Local regulations are for the most part not created by a public outswell of opinion (think about it: when was the last time you participated in demanding a new local law, and were successful increating one?). They generally are the result of the passionate lobbying of small numbers of local individuals and businesses. People believe all kinds of crazy shit, and people are also not above passing laws narrowly in what they believe to be their best interests. People get freaked out about all kinds of ideas that are not real problems. For example: stranger child-abduction, cell phones causing cancer, and the necessity of flame-retardant chemicals in furniture. It’s entirely possible that whatever the regulation was intended to address is not something that actually even happens, at least with any frequency. Or the law is covertly intended to squash competition — larger businesses will sometimes push for safety or other compliance laws that are expensive to comply with (but which scale well), making another hurdle for smaller competitors. This toy safety law is a possible example. Another might be an ostensible environmental law in Dallas: taxis that were hybrids got priority to pick up passengers in the airport line. But of course only Yellow Taxi (the large fleet of the area) could afford the capital investment to immediately create a fleet of hybrids, and so it got a powerful economic advantage over its competitors (all of this is according to a cab driver in Dallas I talked to — I have no primary reference).
And then whoever crafted the law was, frankly, probably not the best and the brightest. The people who draft really good law do not work for cities or local politicians. Sorry. So even if the need is real, it might just straight up fail to actually do what it’s trying to do (not very likely, but possible). More likely, it will create perverse incentives and, while it does do what it’s trying to do, it creates whole new classes of problems, such as rent control laws leading to un-upgraded apartments, creating economic winners of people who sublet forever, etc.
Even if the law was at some point good law, it might not be any more. Cities evolve, the culture around them evolves, and technology evolves. Wiretapping laws created in the 1960’s were used in the 21st Century to prevent recording the police during arrests. These laws accrete; they are not taken out once a decade to see if they still make sense.
Local police are not equipped to understand and seek compliance issues with the surprisingly large number of local laws. Local laws are typically not actively enforced: they instead are unenforced until someone brings a complaint against someone else (or until someone is looking for an excuse to cause trouble for someone). Almost definitionally, local business regulation will be unevenly enforced.
And, finally, people aren’t good at just eye-balling cost/benefit analysis. And with local law, there was probably not extensive analysis of a law by anyone more qualified than an overworked local lawyer. Even if there are real benefits to the law, and it doesn’t create brand new problems, if it just straight up costs a lot to comply with and to enforce, it may simply not be worth it.
The more sunlight a law sees, the less likely I regard all of these problems. I don’t think that this analysis applies to laws against, say, murder. But local law just inherently sees a lot less attention than state or national law. And business law is boring. Most people never concern themselves with business law, or only do in very narrow areas that affect them personally. Local business law is doubly unlikely to have been exposed to a lot of attention that will bring these kinds of problems to light.
What are your priors?