A student asked to grade her own work might give herself all As, or at least higher grades than a teacher would. A similar thing is happening with central bankers, research suggests—and with potentially significant economic consequences.
When researchers at central banks evaluated quantitative easing, one of the banks’ key responses to the 2008–09 financial crisis, they found a bigger impact than did independent economists from academia, according to the National Bank of Slovakia’s Brian Fabo, the European Central Bank’s Martina Jančoková, and Chicago Booth’s Elisabeth Kempf and Lubos Pastor.
Conflicts of interest may explain the rosier conclusions of the bank researchers. “Central bankers evaluating their own policies is not unlike pharmaceutical ﬁrms evaluating their own drugs,” the researchers write. “Both have skin in the game.”
Fabo, Jančoková, Kempf, and Pastor examined 54 research articles, released between 2000 and 2018, about the effects of quantitative easing, or QE, in which central banks bought securities in the open market to help bolster the economy during and after the financial crisis. Research studies that included central-bank officials among their authors found QE to have a higher impact on economic output and inflation than articles written solely by academicians, the researchers find.
Articles with central-bank authors found the peak impact of QE on economic output to be about 0.7 percentage points higher than articles without central-bank authors, the researchers find. The cumulative impact of QE at the end of 2018 was found in central-bank research to be half of a percentage point higher, Fabo, Jančoková, Kempf, and Pastor report. Those are significant differences, they write. All of the research with central-bank authors reported a statistically significant effect of QE on output; only half of the academic papers did so.
Central-bank papers tended to find QE to be more effective at boosting prices, the researchers find. On inflation, the central-bank articles found that QE had a peak effect 1 percentage point higher than all-academic articles reported, and a cumulative effect that was around the same.
And articles with central-bank researchers used more favorable language in the abstracts of the papers than those by academics, with more positive adjectives and fewer negative adjectives, according to the study.
Managerial influence at central banks could be the reason. Bank management both influences the research process and decides who gets promoted. Central-bank researchers who found larger effects of QE on output were more likely to get promotions, the study finds, and there’s some evidence that the connection was strongest for senior researchers—those who may be especially reliant on the support of the bank’s top management to be promoted.
Fabo, Jančoková, Kempf, and Pastor report one counterexample: researchers at Germany’s Bundesbank suggested that QE had less of an impact on output than did articles by academics. However, the Bundesbank’s top brass has been critical of QE, suggesting its researchers may have been operating under a different line of scrutiny.
When a central bank evaluates its own policies, that can lead to overly optimistic research outcomes, which in turn can shape the public’s perception of how effective the central bank’s policy was, the study indicates. This is particularly relevant as central-bank efforts play a key role in bolstering the economy amid the COVID-19 recession.
However, the researchers stop short of concluding that central-bank studies are tainted by bias, calling it “an important but messy question.” Central banks may have an incentive to view their own policies favorably, but they also want to preserve their credibility. The researchers write that they hope their study will help central banks “think through the implications of this conflict for their research processes.”