Closing remarks

In line with the convention theme, I’ve organized my remarks across the previous, current, and way forward for R&S. My reflection on the previous, whereas idiosyncratic, can be, I imagine, indicative of a by way of line that characterizes the division traditionally. For me, that previous begins in 1983, after I got here to the Board as a newly minted faculty graduate to be a analysis assistant (RA) within the Authorities Finance Part, now often known as Fiscal Evaluation. Certainly one of our major duties was to maintain observe of the unstable, and considerably unpredictable, fluctuations in tax receipts. The each day numbers had been transmitted not by laptop, however by phone. Different knowledge retrieval required RAs to go to the Treasury instantly within the luxurious conveyance of a D.C. taxicab, reimbursed by the Board—just a little perk that, because it occurred, by no means got here my approach. In our part, although, it was plenty of time on the cellphone, and the 2 economists I labored with, Al Teplin and Wolf Ramm, would take down the numbers and cross them to me.

Clearly, many issues have modified in R&S over the previous 40 years. However one factor that hasn’t modified a lot is the position of RAs in analyzing knowledge and collaborating in different vital features carried out by the division. Again then, armed with these tax receipt numbers, I might be anticipated to place collectively a projection. My bosses had their very own projections, in fact, and so they subjected my work to very cautious overview. However I used to be on condition that task as a result of there was a widespread view then that RAs had been right here to study and to contribute meaningfully to analysis and coverage work, and I felt a substantial duty to do the perfect I may with the vital work I had been given. A lot has modified at R&S, however these values stay. As a younger faculty graduate, I used to be impressed with how severely folks in R&S took their obligations, no matter their position. I’m not the one particular person whose expertise as a youngster in R&S was formative. In my case, it may appear outstanding {that a} former R&S RA has returned as Vice Chair, however for many years, former RAs have been returning and serving in management roles on the Board and all through the System.

With respect to the long run, my expectation is that immediately’s challenges will persist. I’ve already talked about one problem, which is to ensure we’re buying the precise “various” knowledge, making the precise inferences, and appropriately utilizing instruments corresponding to synthetic intelligence, together with machine studying, to research this wealth of information. These instruments could also be superb at estimating correlations, however we all know that correlation doesn’t indicate causation; thus, their usefulness could also be restricted. Figuring out a correlation within the knowledge with out figuring out causation could lead policymakers to incorrect conclusions and due to this fact lead us to implement inappropriate insurance policies. As well as, these instruments are superb at knowledge mining, which may be harmful if it offers us the false sense {that a} good in-sample match results in a great out-of-sample forecasting mannequin. Economists in R&S and different divisions, in addition to lecturers, have made progress on this entrance, however there may be room for additional enchancment. Along with the problem of how one can greatest use these instruments and knowledge on the Fed, R&S faces the problem of understanding what these new applied sciences imply for labor productiveness and the financial system.

The opposite problem I wish to spotlight is that of constructing choices beneath uncertainty. This isn’t a brand new problem, in fact. John Maynard Keynes and Frank Knight supplied book-length remedies of the topic a century in the past (Keynes, 1921; Knight, 1921). As well as, in 2003, Alan Greenspan noticed, “Uncertainty is not only an vital characteristic of the financial coverage panorama; it’s the defining attribute of that panorama” (Greenspan, 2003). The truth that uncertainty has been a problem for therefore lengthy makes me suppose that it’s going to proceed be a problem going ahead, and within the wake of the pandemic, it’s extremely salient.

In 1967, William Brainard argued that uncertainty in regards to the energy of financial coverage implied that coverage ought to reply extra cautiously to shocks than could be the case if this uncertainty didn’t exist (Brainard, 1967). Brainard’s attenuation precept is a traditional instance of what has come to be often known as the Bayesian strategy to uncertainty and is usually cited as the inspiration for the gradualism within the adjustment of financial coverage. That stated, strategies based mostly on theories of ambiguity aversion led to anti-attenuation. In different phrases, to guard in opposition to uncertainty, the suitable response could also be a fast and powerful financial coverage response. In fact, the precise response to uncertainty might be context particular and thus varies over time.

How to answer uncertainty is a matter that the Federal Open Market Committee, with the assistance of R&S, has been wrestling with for a very long time. Whereas I may very well be improper, as of immediately, I don’t foresee that some omnipotent synthetic intelligence will be capable to completely steadiness the dangers of doing too little or an excessive amount of. Due to this fact, I count on that lengthy into the long run, we’ll be counting on the data and common sense of the folks of R&S to weigh the dangers and to assist in offering choices to financial policymakers in order that they’ll make knowledgeable choices.

In closing, congratulations on a century of service to the Fed and the American folks. I’ve each expectation that the following 100 years of the Division of Analysis and Statistics can be simply as integral to the prosperity and well-being of the American folks because it has been to this point.

Thanks!

References

Brainard, William C. (1967). “Uncertainty and the Effectiveness of Coverage,” American Financial Evaluation, vol. 57 (Might), pp. 411–25.

Dessaint, Olivier, Thierry Foucault, and Laurent Fresard (2023). “Does Various Information Enhance Monetary Forecasting? The Horizon Impact,” Journal of Finance, forthcoming.