Reinventing Cost-Benefit Analysis

If the goal is to give decision makers the tools to make better decision, a single-dimensional metric isn’t the way to go.

One key issue facing Biden on January 20 will be the role of the the White House regulatory czar. The Office of Information and Regulatory Affairs (OIRA) is a tiny White House agency that is virtually unknown to the public. Yet it exercises outsized influence. OIRA is charged with screening all proposed government regulations using a strict cost-benefit analysis.  Biden may want to rethink that role a little.

Obviously, when making a decision, you generally want to know the positives and negatives. For forty years, U.S. regulators have used a quantitative method of doing so, which involves converting everything into cash equivalents. A lot of the existing process is really useful. Trying to stuff everything into dollar terms is less so.

Today, important regulatory decisions are accompanied by a regulatory impact analysis. The impact analysis contains a great deal of really valuable information. For instance, the RIA for regulation under the Clean Air Act will assemble and assess the scientific information relating to the risk posed by a pollutant. This may involve both the use of existing scientific studies and of EPA models to determine how the pollutant would spread.  EPA then models how a new regulation would affect pollution levels and the resulting risks.  On the other side, EPA attempts to determine how industry would comply and to estimate the costs of compliance.  This discussion is the bulk of the RIA, and it’s information that we’d really like to know.

Where things go off track is the effort to stuff everything into the metric of dollars and  cents.  There are several problems here.  First, we can only do this with the effects we’ve been able to quantify, and even there, large uncertainties might exist.  If we’re not sure of the size of the risk, it’s hard to put a dollar value on its seriousness.

Second, it’s pretty unsatisfactory to assign a cash value to an endangered species or a child’s asthma attack. If anything, these problems are worse outside the environmental area.  Consider the Justice Department’s effort to put a cash value on avoided prison rapes.

Third, the process is opaque, involving mysterious parameters such as the “VSL” and the “discount rate.” It’s not easy to explain these key concepts to non-economists – a category that contains virtually all politicians, journalists, and ordinary citizens.   The intellectual world of the technocrats who apply cost-benefit analysis is far removed from the thinking of policymakers.

And fourth, there are a lot of things that we just don’t fit into the analysis, such as social inequality, human rights, and personal dignity.  In theory, agencies are free to consider those. In practice, these intangibles are often marginalized in the analysis. If Biden wants these intangibles to be given serious weight, use of cost-benefit analysis is a detriment.

In short, a monetized cost-benefit analysis is actually not a particularly useful tool for realworld decision makers. The information in regulatory impact analyzes is very useful, but the one-dimensional cost-benefit metric is much less so.  Policymakers would be better served by a multi-dimensional assessment of regulatory impacts.  In particular, it seems to me, a policymaker needs answers to the following five questions:

  1. What problem is the regulation trying to solve, and how bad is the problem?
  2. To what extent would the regulation fix the problem?
  3. How much would it cost, in industry compliance costs and  in jobs?  Would there be economic benefits, such as sparking growth in other industries?
  4. How would the regulation impact values such as inequality, human rights, personal dignity, the rights of future generations? Where relevant, I would include the intrinsic value of nature in this analysis.
  5. In weighing the final decision, are there any benchmarks we can use, such as the tradeoffs made in existing regulations or statutes, by individuals in private life, or implicitly made by Congress in statutes?

The first three questions track current regulatory impact analysis.  The fifth one takes what is useful about cost-benefit analysis and reframes it in better terms.  The fourth is marginalized in the current approach.

Economics is going to play an important role in this analysis. What will be different is that we won’t try to turn a complicated decision into a single number.  Rather, the technocrats will give the relevant information to agency heads in usable form and let them decide how to weigh it.

It’s unrealistic to expect the White House to stay completely out of regulatory issues.  They’re too important to a President’s agenda. They’re also too important politically to be ignored.  I’d suggest that White House review should primarily be aimed at quality control.  Many agencies do careful jobs, but some don’t, and even normally careful agencies can slip up.  The White House should review the RIA to make sure the agency has done its homework. It can also take a hard look at issues that might come up in litigation.  Except in unusual cases, the White House should set the Administration’s agenda but should generally leave the ultimate policy judgments to the agency heads appointed by the President.

Cost-benefit analysis is an overly simplistic way of making tough decisions. It’s time we moved beyond it, so we can focus on all of the factors that really matter to us in decisions.  Monetized costs and quantified benefits are among those factors, but they’re far from being the only ones.   If the goal is to make decisions that reflect society’s values, we can do better than a single-dimension dollar metric.

 

 

, , , ,

Reader Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

About Dan

Dan Farber

Dan Farber has written and taught on environmental and constitutional law as well as about contracts, jurisprudence and legislation. Currently at Berkeley Law, he has al…

READ more

POSTS BY Dan