Monday, October 22, 2012

I am very skeptical of how useful this new regulation data will be

Which is a shame, because I'm guessing a lot of work went into it.

The Mercatus Center at GMU has released REGDATA, which allows you to compare the number of binding regulationsby industry from 1997 to 2010. I've scanne the working paper, the code book, and some of the files, and it looks like they've searched for standard regulatory language "shall"/"must", etc. in the CFR and alotted them by industry.

Quantifying regulations is obviously a worthwhile goal, but I would have thought one of the primary facts of regulatory economics is that different sorts of regulations have different results. I don't understand why adding up how many times an industry "shall" do something tells you anything.

"Bus companies shall put a yellow line near the front of the bus and a sign that tells everyone that federal law prohibits people from stepping across the yellow line" [for non-Americans, this is a weirdly emphatic, micromanaging, and highly visible regulation that's often puzzling to American bus riders]

is massively different from:

"All buses shall only use biofuels derived from hemp and recycled tie-dye t-shirts"

It's a little nanny-statish but you could see how the first one would have negligible or maybe even small positive value (there are some plausible negative externalities on you from a fellow bus-rider running up past that yellow line and yelling at the driver). The second regulation would have a large negative welfare impact.

These are goofy examples, but the point is regulations range from highly damaging to completely benign to highly beneficial.

I've seen this sort of thing done less formally in blog comment threads a lot, when people list the number of pages of federal regulations to prove what an awful country we live in. I've never understood why people think that means anything. I don't think formalizing it in a database improves the argument for this sort of data at all.

I think research on industrial regulation should probably take a page from labor economists working on labor market policy. Look at the policies and regulations one at a time. Don't think you can add up how many policies there are and that that tells you anything at all.

9 comments:

  1. it's the mercatus center. allowing for the idea that regulation is something multifaceted that can have good or bad effects detracts from their goal of complete deregulation of everything

    ReplyDelete
    Replies
    1. But as far as I can tell the bias is as likely to work against them as for them. Actually what's most likely is that you'll just get mush out of it.

      Granted it will afford the opportunity to write a steady stream of ideological pieces that simply say "look regulation is getting worse and worse cause the number is going up!". But some people will try to do actual analysis with this and I can't see anything really intelligible coming out of it one way or the other.

      Delete
  2. "I've seen this sort of thing done less formally in blog comment threads a lot, when people list the number of pages of federal regulations to prove what an awful country we live in. I've never understood why people think that means anything."

    Well, if you have to comply with the regs, the number of pages of difficult to comprehend crapola that you have to wade through means something to you. Not that you are wrong about the effect of regs. :)

    In my youth I was all for deregulation, but now I have seen the effects. We need good regs. And the more detailed the regs are, the more chance of getting bad regs, IMO.

    ReplyDelete
    Replies
    1. Min is quite right. The number of regulations as well as their meaning is important because the information must be digested and conveyed to everyone relevant. I've spent many long hours reading through regulatory requirements documents.

      Delete
    2. I totally agree with you guys on the work of reviewing it all, but this is in a lot of ways transactions cost type stuff that we're talking about. I'm not sure that's what a lot of people are thinking of.

      The other thing I wonder about (not knowing these issues as well) is what the enforcement is like.

      Delete
  3. "there are some plausible negative externalities on you from a fellow bus-rider running up past that yellow line and yelling at the driver"

    But, the bus company is responsible for passenger safety. So it must ensure that bus riders don't interrupt the driver. Buses in Ireland and the UK have signs about this even though they don't have yellow lines on the floor. I have no idea if there's a regulation about that here, but it's quite possible that there isn't.

    ReplyDelete
  4. Daniel, you are truly uninformed.

    This is the kind of blah, blah, blah about regulation that comes from people who have no legal training or understanding.

    The reason for the yellow line regulation is that is the only kind of regulation our super conservative courts will enforce.

    If you write a regulation that reads business owners shall provide reasonable notice where to stand in when the bus is in operation you will get two responses.

    First, bus owners will complain that the regulation is not definite enough, that they don't know what is required.

    Second, these same business owners will say that it costs to much to exercise judgment.

    All this, even though for 125 years the common law has recognized that business owners have a duty to adopt reasonable regulations to assure the safety of customers.

    ReplyDelete
    Replies
    1. If common law already recognized that why the regulation?

      Delete

All anonymous comments will be deleted. Consistent pseudonyms are fine.