Chair the Fed: Insights from game usage data

Abstract

Chair the Fed is an award-winning online educational game developed by the Federal Reserve Bank of San Francisco to help players learn about monetary policy. Players assume the role of Fed Chair and adjust the federal funds rate to try to achieve low inflation and low unemployment. If successful, they are reappointed to another term. By investigating anonymous user data from a three-month period in 2019, we find that about 20 percent of games completed result in reappointment. Chances of reappointment improve with game experience; however, players exhibit more skill in addressing some situations than others. Given the widespread interest in the game, reflected by over 80,000 games initiated per month, the implications of the game for improving economic literacy are important.

Keywords:

• College economics
• economic education
• games and simulations
• monetary policy
• pre-college economics
• serious games
• teaching economics

JEL codes:

• A20
• E50

Acknowledgments

Many thanks to Humberto Barreto and to two anonymous reviewers for helpful comments, and to Kevin Cook for data assistance and insights. Opinions expressed in this paper are those of the authors and do not necessarily reflect the views of the Federal Reserve Bank of San Francisco or the Board of Governors of the Federal Reserve System.

Notes

1 The underlying algorithm in the game computes a score, which players do not see, that provides a threshold for the reappointment decision.

2 See “Chair the Fed Videos” on the SF Fed’s Web site: What Economic Model is Used in the Chair the Fed Game? (FRBSF n.d.[a], [b]). https://www.frbsf.org/education/teacher-resources/chair-federal-reserve-economy-simulation-game/chair-the-fed-game-videos/what-economic-model-is-used-in-the-chair-the-fed-game/

3 Switching from the money supply target to the interest rate target in the game mirrors the Fed’s abandoning of the money supply as a policy tool in 1993 in favor of interest rate targets. See Federal Reserve Bank of New York (FRBNY 2008), “Money Supply,” and Steven Greenhouse “Fed Abandons Policy Tied to Money Supply” in the New York Times, July 23, 1993.

4 The 12 FAQs include information about the Fed’s role and goals, the federal funds rate, and game design. Not surprisingly, earlier questions are accessed more frequently than later questions.

5 Page views are much higher than games started in part because each game has multiple page views. Page view data were collected through Google Analytics.

6 Note that the percent of games completed in table 2, 65.8 percent, refers to the percent of games completed per session (as a proxy for individual players) rather than the overall percentage of games completed shown in table 1.

7 We also investigated whether accessing the game resources Your Job and the FAQs affected reappointment and found that it did not. This is likely because those playing in a classroom setting had already been introduced to the game by their instructors. These players also may be more knowledgeable about monetary policy, which would explain why they get reappointed without reviewing the links.

8 So, for example, inflation gaps of 1 and −1 would equal 3 percent and 1 percent inflation respectively. Unemployment gaps of 1 and −1 would equal 6 percent and 4 percent unemployment respectively.

9 Also, our experience is that players may get reappointed with gaps of 1 or −1. We provide the detail in table 3 to invite readers to draw their own conclusions about what choices may be deemed correct and incorrect.

10 Given the massive size of our dataset, traditional scatterplots were not subject to clear interpretation. Binned scatterplots split the changes to the federal funds rate into equal-sized bins and compute means for the respective inflation or unemployment gaps. The resulting scatterplots are basically the plot of the conditional expectations of these means.

11 As well, a quick perusal of principles of economics textbooks on our shelves shows substantially more coverage of inflation as compared to deflation. This imbalance in textbook coverage may influence content coverage in economics classrooms.

12 We are unable to report systematically whether players change interest rates in the right direction in response to shocks (as we did with inflation and unemployment gaps in table 3) because the immediate response to a shock depends on the player’s current position relative to the dual mandate goals and also due to policy lags built into the game.

13 We use reappointment rather than percent of correct choices as the measure of success in part because, as discussed earlier, it is not always possible to discern definitively what is a correct versus an incorrect response. However, we also find that, as with reappointment, our measures of correct responses increase with game experience. This is to be expected because getting reappointed (or not) is an aggregated measure of all responses in completed games, so game experience should affect correct responses and reappointment in the same direction.