EPA's Regulatory Blinders

A Washington Times article by Art Fraas and Randall Lutter exposes a systemic flaw in the way EPA attempts to justify its regulations, including the new Tier 3 rules. In March, EPA prosed new regulations on gasoline to reduce the sulfur content of gasoline. In this forum, we have written plenty of critiques of the Tier 3 regulations, showing the flaws in the case for it. However, Fraas and Lutter point out a new problem, which affects not just Tier 3 but EPA’s approach in general: When assessing the costs and benefits of a proposed rule, EPA only considers the rule versus the status quo, rather than other alternatives that might be even superior. In a cast like Tier 3, this is an expensive omission because the rule would cost Americans $3.4 billion annually. First I’ll quote from Fraas and Lutter, and then I’ll give an illustrative example to see how powerful their objection is.

Here Fraas and Lutter make their case against the EPA’s failure to truly consider alternatives:

…the [EPA] did not make reasoned determinations that new rules are the best or most cost-effective way to protect public health.

One of the rules in question — the EPA’s recent Tier 3 proposal to reduce emissions from cars and light trucks — illustrates the problem. This rule would cost Americans about $3.4 billion annually, according to the agency, placing it among the administration’s most costly rules. The EPA, however, proposed the rule while estimating net benefits of only the proposed option. It did not identify, let alone evaluate, any alternatives.

Failure to analyze alternatives is first among the cardinal sins that tempt regulators. Without analysis of alternatives, they may claim that their preferred alternative is good — meaning better than doing nothing. They have no basis, however, for saying that their rule is cost-effective or best among plausible alternative approaches.

The EPA’s failure to analyze alternatives violates White House directives. President Clinton’s Executive Order 12866, which President Obama endorsed, requires “an assessment, including the underlying analysis of costs and benefits of potentially effective and reasonably feasible alternatives.” [Bold added.]

Let me illustrate Fraas and Lutter’s important point with a simplistic example. Suppose UPS is trying to decide on the appropriate rule for how often each 18-wheeler in its fleet should have its oil changed. If there’s no policy—meaning that the oil is never changed—then eventually each truck will break down and require major repairs to the engine, costing many thousands of dollars per truck, per year.

To avoid this horrible outcome, a dashing executive gives a PowerPoint presentation to his fellow UPS decisionmakers, and shows them the cost/benefit analysis of requiring daily oil changes. True, it would be expensive, not only in terms of paying for the oil change itself, but also (more important) UPS is losing out on the shipping that can’t be achieved while each truck is being serviced. The policy of insisting on daily oil changes effectively reduces the capacity of UPS’ fleet of trucks, for a given number of vehicles.

Even so, the hotshot executive makes a convincing case that UPS can’t afford not to implement this new rule. Without daily oil changes, the engines eventually break down, often leaving drivers stranded on the interstate, with trucks loaded with time-sensitive packages. Thus, although the costs of the new policy are admittedly high, the benefits (in terms of avoided expenses, but also extra revenue from happier customers) are much greater. The new rule thus easily passes the cost/benefit test, and UPS implements the policy of requiring all of its trucks to receive daily oil changes.

A silly story, to be sure. But what specifically is wrong with our tale? The answer is obvious, because I’ve picked such an exaggerated example: Just because getting a daily oil change is more profitable than never getting oil changes at all, doesn’t mean that it’s a sensible policy. A policy of weekly oil changes would be better still, since it would lower the costs while retaining just about all of the benefits of the daily policy. And switching to a more flexible rule, where the oil is changed based on a two-pronged trigger of either miles driven or time elapsed, would be even better still.

As the silly UPS example illustrates, it’s not enough to evaluate a proposed policy in a vacuum, comparing it only to the status quo. Even if it passes a cost/benefit test in isolation, it still might be a ridiculous policy. Rather, the appropriate approach is to compare a proposal to a range of policies, to see if any outperforms it. Furthermore, when it comes to government regulations, it would be nice to see considerations of how introducing pro-market reforms could help matters—rather than piling on ever more mandates and restrictions.

Beyond the specific problems with the new Tier 3 rules, Fraas and Lutter have penned a general critique of EPA. When performing cost/benefit analyses of proposed rules, EPA should consider several alternatives, to test the robustness of its results. After all, these costs cost Americans billions of dollars. As Fraas and Lutter put it: Just because a policy is better than nothing, doesn’t make it a good policy.

IER Senior Economist Robert P. Murphy authored this post.