We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Regulation and data

This article from the Washington Post, on the application of the little known Data Quality Act to hobble the regulatory leviathan, is full of unintentional insights. The Data Quality Act is, well, let the Post tell it, and let the insights begin!

The Data Quality Act — written by an industry lobbyist and slipped into a giant appropriations bill in 2000 without congressional discussion or debate — is just two sentences directing the OMB to ensure that all information disseminated by the federal government is reliable.

The first insight is, of course, the clonking great pro-government, pro-regulation bias that the Post brings to this story. Note the disparaging terms applied to this piece of legislation, which has a genesis and a pedigree that is totally ordinary – most legislation is the product of interested parties, and most finds its way onto the books via massive omnibus bills that no one reads. However, these routine facts of Washington life are given ominous prominence only when the media outlet is opposed to whatever was done. The rest of the story is riddled with similar bias – in the Post’s world, regulation is always good, always to protect the people, never fails a cost-benefit test, always supported by the preponderance of the scientific evidence, etc.

The next set of unintentional insights comes to us when the relatively innocuous purpose of the Act collides with the prerogatives of the regulatory state.

But many consumers, conservationists and worker advocates say the act is inherently biased in favor of industry. By demanding that government use only data that have achieved a rare level of certainty, these critics maintain, the act dismisses scientific information that in the past would have triggered tighter regulation.

First, of course, note who the Post asks for their opinion. Of equal interest is the rather revealing admission that, in the past, regulation was apparently handed down on the basis of information that was, how to put this, of less than adequate quality. Declining to regulate because the data isn’t there is, of course, a Bad Thing.

These final comments surely need no elaboration.

“It’s a tool to clobber every effort to regulate,” said Rena Steinzor, a professor of law and director of the Environmental Law Clinic at the University of Maryland. “In my view, it amounts to censorship and harassment.”
. . . .

Yet Steinzor, the Maryland environmental lawyer, and other critics complain that the OMB’s involvement politicizes the process. The expertise of the handful of scientists hired by Graham, they say, cannot match that of the thousands of experts on agency staffs.

Tweet about this on TwitterShare on FacebookShare on LinkedInShare on TumblrShare on RedditShare on Google+Share on VK

7 comments to Regulation and data