Behavioral economics and Fed policymaking.

• Jurisdiction: United States
• Date: 22 September 2016
• Author: Calabria, Mark A.

For every bias identified for individuals, there is an accompanying bias in the public sphere.

--Cass Sunstein (2014:102)

Behavioral economics has continued to gain momentum in challenging the standard rational actor model in economics. With a few exceptions, the emphasis has been on the cognitive failure of individuals outside of government. Niclas Berggren (2013: 200) estimates that 95.5 percent of behavioral economics articles in the leading economics journals do not contain an analysis of the cognitive ability of policymakers. In this article, I offer a preliminary analysis of potential cognitive failures in the Federal Reserve's conduct of monetary policy. Proposals to "debias" monetary policymaking are offered, along with a discussion of how the Fed's existing institutional structure ameliorates or exasperates potential biases.

Behavioral Economics in Monetary Policy

Initial attempts to incorporate behavioral economics into monetary policy focused on how policymakers might better account for the behavioral biases of market participants, especially in the area of price and wage setting (Rotemberg 2008: 34-36). Behavioral economics has been used to provide a new rationale for reliance on a Phillips curve framework, as well as to place greater weight on unemployment. As Janet Yellen (2007: 15) has stated, "Behavioral macroeconomic models also provide theoretical underpinnings for the view held by most policymakers that, in the short run, monetaiy policy can and should strive to stabilize the real economy." More recent scholars (e.g., Orphanides 2015) have attempted to explain the Federal Reserve's own actions via the lens of behavioral economics. Those efforts are extended in this article.

Are policymakers at the Fed likely to experience cognitive failures? DellaVigna (2009: 364) has argued that politicians "are experienced agents facing high-stake incentives and significant competition" suggesting that at least some elements of the political realm are less susceptible to cognitive failure than households. Setting aside whether DellaVigna's characterization of politicians is accurate, the question of competition and experience has been demonstrated to reduce cognitive failure (List 2003, 2004). However, while there may be some modest choice among currencies, the Fed has few, if any, real equals. The level of competition facing the Fed does not even approach that likely necessary to serve as an effective feedback mechanism.

If competition is unlikely to exist in sufficient degree to eliminate cognitive failures at the Fed, will experience suffice? Members of the Federal Open Market Committee (FOMC) are often professional economists and usually quite advanced in their careers. They serve long terms (in principle, at least). On the other hand, economic crises are, thankfully, not a frequent occurrence. This means that policymakers rarely have the opportunity to learn from them. Federal Reserve Chair G. William Miller stated as much during an FOMC meeting in December 1978 (quoted in Abolafia 2012: 99):

Boston Fed President Frank Morris: Mr. Chairman, I don't think we understand what is really going on in the economy.

Chairman Miller: I'll go along with that.

President Morris: I think it's because we haven't had enough experience judging the reaction of both the consumer and the investor to an economy with a high rate of inflation.

Chairman Miller: That's right, we haven't had any experience.

While Morris and Miller found themselves in an environment of unusually high inflation, other periods have found policymakers wondering how to maneuver in an environment of unusually low inflation (Akerlof, Dickens, and Perry 1996: 50-52). I shall return to this point in more detail below, but both statements from policymakers and the broad economic trends suggest experience alone will be insufficient to eliminate any behavioral biases among policymakers.

Potential Biases in Monetary Policy

Cognitive science has identified a number of potential decisionmaking biases. Some are more robust than others. Below, I discuss the following biases: availability bias, representativeness bias, status quo bias, loss aversion, and overconfidence. I also offer examples of how each of these may be present in monetary policymaking. The discussion is in no way exhaustive, but rather focuses on the most likely biases, as well as those for which we have some evidence. There remains considerable room for additional research into these and other cognitive biases and their related impact on monetary policy decisions.

Availability Bias

When asked to consider the frequency of possible outcomes, individuals often quickly go to what they can remember about similar situations. Ask any New Yorker about terrorist attacks and 9/11 is unlikely to be far from mind. Ask an economist about inflation and you may well get a lecture on the Great Inflation of the 1970s. Ask about a stock market bust, and you might get the answer "1929." Ask about when allowing a large hedge fund to fail resulted in a collective financial yawn and you might get a puzzled look (think "Amaranth Advisors LLC"). Where an answer comes easily ("1929") there is a tendency to overestimate the frequency of such events, whereas when an answer requires some research or consideration ("Amaranth") the frequencies tend to be more accurately estimated or even underestimated. This is the heart of availability bias (Tversky and Kahneman 1973: 16.3-65.).

Monetaiy authorities must often make quick decisions. Should institution X be saved before the markets open in Asia on Monday morning? Will the smallest sign of deflation (or inflation) spiral out-of-control? Will the large scale purchase of assets stimulate lending and the economy? While there will be, of course, some analysis conducted on these questions, it would not be too ungenerous to describe some actions by the Fed as "spur of the moment." Witness the ever-changing rationales given for denying assistance to Lehman Brothers.

Given the handful of almost mythical events in the history of monetary policy and financial crises, it is perhaps not surprising that these same events dominate discussions of policy. Federal Reserve Chair Ben Bernanke has occasionally been called the "perfect man" for the 2008 crisis, because of his long study of the Great Depression. But this assumes we were facing another Great Depression. As it's impossible to know whether that was actually the case in 2008, it is not a stretch to conclude that, given the actions ultimately taken, the implied probabilities of another Great Depression were magnitudes higher than the actual probability. There may be no bigger availability heuristic in macroeconomics than the Great Depression. In all likelihood...