Line of Inquiry: Michael Weber on How Diversity Can Help the Fed

So one way in which central banks try to actually stimulate aggregate demand in the current low interest-rate environment, meaning when the conventional monetary toolbox is shut off, is through direct communication with ordinary households. For example, if the Fed were successful in actually convincing households that inflation will be lower going forward, households might now actually postpone purchases into the future. And through that mechanism, the Fed would be able to manage aggregate demand intertemporally over time. So that’s one reason why, in fact, actually the Fed is so keen, as many other central banks, in coming up with ways of communication that reaches the broad population overall.

Now, of course, another reason why we might be concerned about people not actually being aware of what the Fed is doing, and households having a low level of trust, is the fact that if that’s the case, populist governments could actually have an easy way of undermining the credibility of the Fed and ultimately politicizing monetary policy-making. And that’s why it’s of paramount importance for central banks to ensure that the overall broad population has a high level of trust in the central bank and what it’s trying to achieve.

We surveyed 10,000 people in the United States and then randomly provided or made salient the representation of a woman, Mary Daly, the president of the San Francisco Fed, and a Black man, Raphael Bostic, the president of the Atlanta Fed, and wanted to understand whether making salient their representation on this policy body was actually increasing the level of trust and the extent to which women and African American survey participants would actually incorporate the identical information compared to women and African American survey participants that instead actually were made salient the representation of a white man, Thomas Barkin, the president of the Richmond Fed. We actually show that women and African American survey participants that see Raphael Bostic or Mary Daly, the female and African American representation on this Federal Open Market Committee, are way more likely to incorporate the identical information that we provide in the survey into their own forecast for macroeconomic aggregates—think about inflation or the unemployment rate over the next 12 months—relative to women and African American survey participants that saw Thomas Barkin, a white man, who is also part of this Federal Open Market Committee. And of course, you know, all groups were more likely to incorporate information that they received relative to a control group that didn’t receive any information.

Of course, an important question is: What are the mechanisms and channels through which making salient the representation of women and African American participants on the FOMC resulted in this higher likelihood that women and African American survey participants were incorporating the provided information into their own forecast for the macroeconomy? In the paper, we actually showed that two complementary channels are mainly responsible for this updating of individuals’ expectations. In particular, we show that, on the one hand, there is a trust mechanism channel, but there’s also an information acquisition channel.

So what we show ex ante, meaning before we provide any information and make salient the representation of women and Black participants on the FOMC, we do observe in the data that female and African American survey participants actually have a very low level of trust in the FOMC and the Federal Reserve overall, compared to white male survey participants. However, we also then find that women and African American survey participants that do see that in fact, actually, Mary Daley and Raphael Bostic are part of this important committee, they substantially increase the level of trust in the Fed—and, actually, trust in the Fed along two dimensions, that they actually achieve the inflation and unemployment target, but also that they care about the well-being of all Americans, including people like themselves in the survey.

The second channel we want to study in the paper, we find evidence that it’s responsible for our results, is what we call an information acquisition channel. Specifically what we mean by that is, you know, maybe if I’m a woman and I don’t see any representation of a woman on this policy body, I might just not be interested to begin with of gathering and learning about what this institution is about. And indeed, for example, if you ask about basic pillars of policy-making ex ante, before our experimental intervention, we do see that women and African American survey participants, for example, are less aware that the FOMC is in charge of setting policy rates, or for example, what the inflation target of the Federal Reserve is. Instead, actually, then what we do in the follow-up experiment, we actually try to understand if us making salient the representation of a female policy maker on the FOMC increases the willingness of women to gather information about the FOMC.

And the way we do that is we, again, randomize our survey population into three groups, and every group has a choice in reading two separate articles, an article about the Congressional Budget Office versus the Fed. The first group is not told who the article is about. The second group has a choice about CBO, but this time Phil Swagel, a white man, and Richard Clarida, also a white man on the FOMC. The third group instead has the choice between CBO and Phil Swagel, and this time a female policy maker on the FOMC. And we do see that, actually, once we make salient that actually the FOMC has female representation, women have about a 10 percentage point higher likelihood to choose the article about the Fed relative to women that actually have the choice only between two male policy makers, Clarida versus Swagel, or actually unconditionally just Fed versus CBO. So making salient the fact that this institution, actually, in fact, also has female policy makers increased substantially the willingness of women in our experiment to gather information about the Federal Reserve.

So another mechanism we actually then wanted to shed light on is whether our results would be more consistent with what we call homophily, the fact that people tend to maybe react more strongly once actually someone that looks like themselves is actually making policy decisions, over that what we call a taste for diversity, someone that is not part of the majority population on that body is also represented. And so what we find in the paper is that women are way more responsive, not only to seeing Mary Daly being on the FOMC, but also to seeing Raphael Bostic. And now if you think about it, you know, especially white women, they actually, if we would think about Raphael Bostic versus Thomas Barkin, and, you know, Thomas Barkin would at least share their race, if not gender. So we would expect if it was a pure homophily type of argument, that white women might actually respond more strongly to Thomas Barkin than Raphael Bostic. But we do see that actually women, especially white women, are way more responsive to Raphael Bostic relative to Thomas Barkin, which for us, in addition to other results we discussed in the paper, is more consistent with a heterogeneous taste for diversity, relative to a pure homophily-type argument.

Now, of course, there are good reasons why we might want to increase the diversity of important policy committees like the Federal Open Market Committee, like equity, whether, like, legitimacy reasons, some people have highlighted that, you know, a more diverse committee might also be less subject to groupthink because different backgrounds and experiences are represented. But if you now want to think about maybe like a pure efficiency argument in the sense, like, we want to make sure that if the central bank communicates, as many people as possible are responsive. If we now increase the diversity, some critics have argued potentially the majority population as currently representative on the FOMC, white men, might in fact be put off, let’s say. So now, if they see that in fact there are female or African American representation on the FOMC, in fact, they might become less responsive to the provided message. Instead, actually, what we find is that white men, they respond equally to our provided information independent of whether it is associated with a picture by Thomas Barkin, Raphael Bostic, or Mary Daly, which shows to us, and we show that in the aggregate, that in fact, actually, monetary-policy communication is more effective because it doesn’t put off white men, instead it actually increases the responsiveness of female and African American survey participants, and so overall, in fact, enhances the communication of the Federal Reserve.