Late last year, the U.S. Department of Labor proposed an important and controversial federal regulation without a serious analysis of its costs and benefits, and of its likely effects on low-income American workers. That’s a big mistake, a disservice to the public and a bad precedent. It might cause legal trouble as well.
The absence of such an analysis, including numbers, is inconsistent with decades of practice supported by both Republican and Democratic presidents. Under a series of executive orders starting with President Ronald Reagan and most recently reaffirmed by President Donald Trump, presidents have required all executive agencies to accompany “major” rules with a “regulatory impact analysis.” That analysis must quantify the costs and benefits of both proposed and final regulations.
At least as much as their Democratic counterparts, Republican presidents have insisted that the Office of Information and Regulatory Affairs should oversee the regulatory process to ensure that potential costs and benefits are revealed to the public.
That’s important. Regulations usually shouldn’t go forward unless the benefits justify the costs. Cost-benefit analysis isn’t perfect, but it’s the best method we have for assessing the consequences of what the government is proposing to do. Without a catalog of anticipated costs and benefits, we can’t know whether a regulation is likely to help people or hurt them.
Clear presentation of anticipated consequences, for public scrutiny and review, can also help correct mistakes — and show an agency that it ought to consider different solutions to a problem (and possibly no regulation at all).
Finally, analysis of costs and benefits can reduce the influence of political dogmas, unreliable intuitions and interest-group power. If an agency is required to assess the actual effects of its proposals, and show that on balance they are good, everything else starts to look less relevant. The government’s attention is focused on the right questions.
I was privileged to serve as administrator of the Office of Information and Regulatory Affairs from 2009 to 2012, and I saw, close up, the immense importance of cost-benefit analysis to democratic accountability. The obligation to produce some kind of cost-benefit analysis is a crucial safeguard.
Which brings us to the Department of Labor, which has proposed to rescind a 2011 regulation that requires tips to be treated as the property of employees who receive them. Under the 2011 rule, employers may not direct employees to turn over their tips. The new proposal would eliminate that prohibition. Among other things, it would allow employers to require “tip pooling arrangements” in which, for example, waiters and waitresses would have to share their tips with kitchen workers.
No one should deny that the Trump administration may reconsider previous regulations. It deserves great credit for its effort to cut regulatory costs. But the tipping proposal came without sufficient analysis of its effects. Would workers end up losing a lot of money on balance? Would employers end up taking significant amounts of the tips?
According to a report from Bloomberg Law, the Department actually produced such an analysis, and it suggested that workers could lose billions of dollars annually. Labor Secretary Alexander Acosta didn’t like that conclusion. At first, he is said to have worked to alter the analysis to show that it would have a smaller impact on workers than it originally suggested. According to the report, he insisted that the proposal should come out with hardly any analysis of the central questions.
The public explanation accompanying the proposed rule admits that some of the key figures “have not been quantified” on the ground that the department was not able to come up with reliable numbers. Instead of offering some projections, the department asks for public comments about the effects of its proposal, noting that it may add an analysis when it issues a final rule.
That’s not good enough. In many cases, agencies lack the information that would allow them to specify costs and benefits, and they do the best they can. They might be able to project a range. A safety regulation might prevent somewhere between 50 and 250 deaths; an environmental regulation might cost between $300 million and $600 million.
If agencies can’t specify a range, they might identify lower or upper bounds (while candidly acknowledging uncertainties). They have sophisticated, time-honored techniques for dealing with gaps in available knowledge.
The Labor Department’s silence on the potential impact of its proposal might also create legal trouble. When agencies propose their regulations to the public, they usually invite public comment on the supporting analysis — which allows careful scrutiny and correction of errors.
A failure to allow public scrutiny of an agency’s analysis creates a risk that a rule will be overturned in court.
At some point before it finalizes the proposal (if indeed it chooses to do so), the Labor Department should re-propose the rule to the public, along with a fuller supporting analysis — or at the very least, issue such an analysis and allow ample time for public comment.
There’s a broader point here: Put the politics to one side. Whether the government is imposing or eliminating regulatory burdens, that’s the right thing to do.
— For more columns from Bloomberg View, visit http://www.bloomberg.com/view.