Empirical evidence suggests that expert agencies perform no better than courts.
A fundamental premise underlying the U.S. administrative and regulatory system is that the subject matter expertise of administrative agencies vests them with the ability to enforce their organic statutes more effectively than any of the three branches of the federal government acting independently. At least in part for this reason, Congress grants agencies the authority to guide policy through statutory interpretation, to engage in rulemaking, and to render adjudicatory decisions. Similarly, perceived expertise guides not only the decision of when it is appropriate for Congress to delegate authority but also whether to delegate to more specialized expert agencies or to courts of general jurisdiction.
This “expertise hypothesis” implies administrative agency decision-making adds greater value to law enforcement, all else equal, than does delegation to Article III courts. This prediction implicates a number of critical issues concerning institutional design and the optimal roles of administrative agencies and courts; indeed, substantial scholarly attention has been paid to these questions even though relatively little direct empirical evidence exists to confirm the expertise hypothesis itself.
One way to test the hypothesis is to compare agency performance in adjudicative decision-making to federal judicial performance in deciding cases involving similar legal issues. We take this approach, comparing antitrust decisions before the Federal Trade Commission (FTC) with those issued by Article III courts. While adjudicatory decision-making is just one of the many functions delegated to administrative agencies, and is only one portion of the FTC’s competition policy portfolio, it provides a useful laboratory for comparative institutional analysis of antitrust enforcement agencies and courts.
The FTC’s competition policy decisions have particular policy relevance in light of recent calls for expansion of its enforcement authority under section 5 of the FTC Act, which authorizes the agency to prohibit “unfair methods of competition.” It is widely understood the FTC has rarely been successful in § 5 enforcement efforts targeting conduct that does not violate the Sherman or Clayton Acts despite repeated attempts and occasionally strong policy preferences of the Commission to avoid the constraints imposed by the judicial interpretations of the antitrust laws. Consistent with those policy preferences, and armed with evidence that some antitrust cases may be too complicated for generalist courts to assess, these commentators invoke the expertise hypothesis as justification for their proposals to grant the FTC greater enforcement powers.
The comparison we conducted is one between institutions – the FTC and the federal judicial system – and not individual judges and commissioners. There is no doubt the antitrust and economic experts at the FTC are well equipped to analyze all modes of business dealings; in this sense, agencies certainly have greater economic expertise than courts as a general rule. However, like the FTC, courts incorporate economic expertise into their decision-making. While the FTC seeks to incorporate its staff’s expertise, courts seek to incorporate the expertise of expert witnesses. Both institutions are organizations with complex internal structures with different modes of transmitting economic expertise into decision-making outputs and different constraints and incentives.
We sought to determine whether the FTC’s institutional structure facilitates the transmission of economic inputs into its adjudicatory decisions more effectively than Article III judges in similar cases. The implicit assumption underlying the expertise hypothesis, and many of the calls for increased delegation to the agencies, is that they are better equipped to convert that expertise to litigation outcomes. If agencies do not hold such an advantage in converting expertise to outputs, interesting and important questions are raised concerning the optimal institutional design of the FTC and of administrative agencies generally. Our preliminary evidence suggests precisely this outcome as we find no advantage for administrative agencies.
We began by comparing the appeal rates of federal district court judges and the FTC’s five-member Commission as the primary performance measure to compare these institutions. Following a paper I published with Michael Baye, we used appeal rates here because they are a signal generated by actual costs incurred by the parties who, informed by their own economic experts, have determined the initial court committed a reversible error. All other things being equal, an appeal signals that at least one party believes it can convince a higher court an error has occurred. For these reasons, we use appeal as our primary performance measure though results are unchanged if we consider reversal rates.
Our comparison suggests the Commission’s decisions are more likely to be appealed (and reversed) than those of Article III judges. Taken at face value, the comparison implies that when it comes to adjudicatory decision-making, the Commission does not perform as well as district courts. We acknowledge that these differences in appeal rates may be the result of factors that influence the decision to appeal from a Commission judgment differently than from a district court judgment. For example, the selection of cases by the FTC may differ systematically from the cases brought by plaintiffs to federal court, though it is not necessarily the case that those differences imply a higher appeal rate for agencies. Further, differences in procedure between litigation in federal court and at the agencies provide different incentives to settle. However, when we conducted further analysis to control for such factors to the extent possible with available data, our basic results remained unchanged. Further, we found that – looking exclusively at FTC cases – appeal rates are not reduced when the Commission modifies the initial decision of the FTC administrative law judge.
The expertise hypothesis posits that the institution with greater expertise will consistently outperform any institution without endogenous expertise. That assumption should be grounded in a comparative analysis of the institutions and the processes translating expertise to decisions. The inability of an agency like the Federal Trade Commission to translate its expertise into higher-quality decision-making than found in the courts renders it at best ineffective and at worst costly to society. We find evidence suggesting the Commission does not perform as well as generalist judges in executing its adjudicatory function. Given the specialized nature of modern antitrust analysis and its strong reliance upon economic expertise, our findings raise important questions about the validity of the expertise hypothesis. Our study highlights the need for more research on the relationships among institutional design, agency expertise, and the quality of agency outputs.
This essay draws upon the authors’ recent paper, “Do Expert Agencies Outperform Generalist Judges? Some Preliminary Evidence from the Federal Trade Commission.”