The Hutchins Roundup brings the latest thinking in fiscal and monetary policy to your inbox. Have something you'd like us to include in the next Roundup? Email us and we'll take a look.
This edition was written by Chesapeake Dowdy, Chase Parry, Andrew Rosin, and David Wessel
Many white-collar professions develop expertise through learning by doing, raising the question of whether AI helps or hinders skill development. In a randomized controlled trial with 133 patent lawyers across 11 U.S. intellectual property law firms, David Autor of MIT and co-authors find that lawyers using AI to help draft patents outperformed those who didn’t on measures including enforceability, completeness, and clarity, with the largest benefits accruing to junior lawyers. After 90 days, lawyers were asked to review and redline a patent draft without any AI. Those who had been using AI outperformed those who had not, but the performance gains accrued only to senior lawyers. Juniors showed no average gain, and their scores became more dispersed, with more poor and more good scores and fewer mediocre ones. The authors interpret this pattern as consistent with AI helping some junior lawyers acquire skills while hindering skill development for others. Among seniors, AI use produced more good scores and fewer mediocre and poor scores, rather than increasing performance at the very top.
Four times a year, the Federal Reserve publishes the Summary of Economic Projections (SEP), which reports Federal Open Market Committee participants’ projections for the federal funds rate and other macroeconomic variables. Stefania D’Amico of the Federal Reserve Bank of New York and co-authors ask how markets react to the information provided in these projections. Using overnight index swaps to measure market expectations for the fed funds rate, the authors find that about 20% of a surprise in the median SEP fed funds rate projection immediately passes through to market expectations. The authors argue that this partial adjustment is consistent with markets viewing the SEP as a conditional projection subject to considerable uncertainty rather than a commitment. If markets viewed the SEP as a commitment to a particular policy path, the pass-through would be one for one. Surprises in the SEP’s GDP and inflation projections also move market expectations for the fed funds rate, suggesting that markets learn from the FOMC’s economic outlook. But controlling for those surprises leaves the pass-through from fed funds rate surprises essentially unchanged, suggesting that the rate projections provide information about the FOMC’s reaction function. In a model simulation, a 25-basis-point surprise in the median SEP rate projection boosts market expectations by only 5 basis points initially but by about 18 basis points after two quarters. This gradual reaction suggests that the dot plot contains information about future monetary policy that markets incorporate only gradually.
How does the design of government transfers—in-kind versus cash, one-time versus monthly—affect households’ spending decisions? Standard economic theory predicts that cash and food benefits of the same value should have similar effects on spending patterns if the food benefit is smaller than what the household would normally spend on groceries, because households could simply reduce their spending on food and free up funds for other purposes. But in an analysis of four large pandemic-era programs, Diane Schanzenbach of Georgetown University and co-authors find that food benefits raise spending at food stores more than cash transfers of similar value. Households increased food-store spending by 20 cents for every dollar received in one-time in-kind transfers, for example, but increased spending by only 7 cents for every dollar received as a one-time cash transfer. Households also increased food-store spending more in response to monthly transfers than to one-time transfers. For example, for in-kind food benefits, households increased food-store spending by 43 cents for every dollar received in monthly payments but only by 20 cents for every dollar received in one-time payments. The authors suggest that households spend more out of monthly transfers because they expect future payments, whereas one-time payments are more likely to be smoothed over time. Households spend more out of in-kind transfers because they “treat a benefit labeled for ‘food’ as belonging in the mental account for food purchases, even when the restriction is not economically binding,” the authors add.
“I believe that I can use the balance sheet of the U.S. for foreign policy. So we have a foreign policy goal, and that is to create allies in the Western Hemisphere. Argentina was the front-runner for that, and it was my belief that the Milei government had sound policies. The opposition was using their capital markets to try to create a panic and cause a currency crisis. I could step in, bridge them through the election, which makes money for the U.S. government. Then they got to the other side of it, capital markets were good, and President Milei won, beat expectations and they’re off to the races,” says Scott Bessent, U.S. Treasury secretary.
“But more importantly, with the Argentine election, that was followed by Chile who went from like a crazy left-wing government back to a center-right government. Colombia has now gone from a hardcore Marxist government back to a market-based government. Ecuador, Bolivia, all these countries are following, same thing.
“And whenever people say, ‘Oh, well, the Treasury Secretary is taking a risk,’ it’s my dream. I have asymmetric information, I am the house now. So when we intervene with the Japanese yen, I have pretty good insight into what the Bank of Japan’s going to do, what Japanese policymakers are going to do, but you can bet against me if you want.”
About the Hutchins Center on Fiscal and Monetary Policy at Brookings
