Eye on OIRA: Sunstein Brings Behavioral Economics to NHTSA Tire Fuel Efficiency Program

On March 19, OIRA Administrator Cass Sunstein issued the office's first Review Letter of the Obama Administration, telling the National Highway Traffic Safety Administration (NHTSA) to redo their studies on how to design the labels for the agency’s new “Tire Fuel Efficiency Consumer Information Program.” (For background on Review Letters and the other types of OIRA letters, see here.) Those new studies will delay implementation of the tire efficiency regulation by at least half a year, and likely longer.

Under the tire efficiency program, NHTSA must develop a system for setting tire fuel efficiency ratings and design a label for tire manufacturers to affix to their products that convey these ratings to consumers, so that consumers can consider the fuel-efficiency effects of different tires when deciding which to buy.

When NHTSA sent its draft final rule to OIRA in December, it appeared that the rule was nearly ready to go. The agency had devised a system for rating tire fuel efficiency, and a label for conveying that information to consumers. The agency used extensive survey and focus group research as it designed the label, and it found through that research that it had created labels that were comprehensible and effective in conveying information about a tire’s fuel efficiency rating, and the significance of this rating.

By the time the draft final rule emerged from OIRA three months later, it had been blown apart. The rule was now divided into two components: one focusing on the system for setting tire fuel efficiency ratings and one on the labels. On March 30, NHTSA published a final rule putting the tire fuel efficiency ratings system into effect, but the agency postponed finalizing any provisions that established the label design criteria until it has completed a comprehensive new set of studies on how best to design the label. According to NHTSA’s research plan, this study will not be completed until September 30, 2010—a full six months later. After that, NHTSA must propose a new label design requirement and go through the whole notice-and-comment process again. It could be well over a year before NHTSA is able to issue a final rule for its label. Until then, the agency’s recently adopted tire fuel efficiency ratings system must languish without providing any benefit to consumers.

So, why the new tests? With the Toyota unintended acceleration scandal laying bare NHTSA’s severe resource constraints, one might expect the agency to avoid such unnecessary and duplicative undertakings.

It’s not entirely clear what Sunstein’s problem with the original studies is. The documents and emails exchanged between OIRA and NHTSA while the rule was under regulatory review would normally provide the best evidence. These documents and emails have not been made public yet, even though Executive Order 12866—the presidential order that governs the regulatory process—requires their disclosure (see Section 6(b)(4)(D)). OIRA’s failure to publish such documents—in clear violation of Executive Order 12866—has been commonplace throughout the Obama Administration, just as it was during the Bush Administration.

The vague language in Sunstein’s ReturnReview Letter provides some insight, though. Sunstein states that the tests “should aim to measure consumers’ understanding of the label, and their likely behavior given that understanding, instead of merely identifying their preferences among labels. . . . NHTSA should give greater weight to scientifically valid experiments than to focus testing.” This suggests that NHTSA’s original tests may have fallen short because they did not attempt to put a precise quantitative value on how the different labels under consideration affected hypothetical consumer choices. For example, these tests were not designed to identify whether the comprehensibility of Label Design Scheme A is 10 “units” greater than that of Label Design Scheme B, and 20 “units” greater than that of Label Design Scheme C—or to determine a quantified difference in consumer's decisionmaking between such choices.

In the context of a classroom or a research laboratory, it might be interesting to try to put a numerical value on concepts like “comprehensibility” or “ability to influence” so that they can be precisely measured and compared. To those of us who have to live, and work, and consume in the real world, the whole enterprise is quixotic. This kind of behavioral economics experiment has only one certain effect: Delaying an important regulation. Executive Order 12866 requires no such experiments, but since the White House has not (yet) issued a new order on regulatory review, OIRA is going ahead with these initiatives even under the old (but standing) order.

NHTSA has serious work to do, including reducing traffic fatalities and injuries and curbing greenhouse gas emissions from cars. The public interest is not being served if this crucial protector agency is forced to dedicate its scarce resources toward this experiment.