Beyond Cost-Benefit Analysis of Regulation

According to the Wall Street Journal, Barack Obama has just set the "all-time regulation record," in terms of the number of new "major regulatory rules." That last phrase is Beltway jargon for any regulation costing over $100 million. I don't doubt that government regulations wreak economic havoc, but I am intrigued at how vague a criterion this is, and it was the following paragraph that helped me realize it:

Sam Batkins of the American Action Forum, who did the study, calculates that the economic cost of all this adds up to $743 billion, based on data provided by federal agencies. Mr. Batkins doesn't say this, but that estimate is almost surely an understatement because agencies routinely low-ball the costs and overestimate the benefits of the rules they propose. [bold added]

That's an obvious problem, but it's worse than that in two differing ways. First, the whole premise of most (if not all) of these analyses is that there is some "public" that collectively owns your money and time and for whom there is a "benefit" (conveniently defined by the same bureaucrats who will spend your money). Second, and as a direct consequence, even the most scrupulous attempts to estimate the benefits of some regulations will result in an overblown "benefit" for certain activities of the government. Consider a hypothetical regulation that would save on administrative costs for an agency tasked with something the government has absolutely no business doing. A "savings" of $200 million from a billion-dollar budget that should be wholly off the books means the remaining $800 million is more likely to keep on being wasted. So, despite attempts to gauge the economic impact of regulations, there is every reason to suspect they are even worse for the economy than this story indicates.

But such analyses also present the following interesting difficulty for advocates of limited government: There are regulations out there that might effectively exist (as, say, industry standards) in a truly free society, due to such factors as potential loss of business, the threat of legitimate litigation, or watchdog groups similar to the Consumers Union. The government shouldn't be doing these, either, but any cost-benefit analysis of such regulations would need to include their benefits.

But that difficulty suggests that we shouldn't be haggling over the costs of regulations, anyway. Such an exercise can be useful, but it is no substitute for a debate about whether the government ought to be attempting to run the economy at all.

-- CAV