
Full guide here, along with every discussion I've given since 2007: drric.web.unc.edu/research/d… They've gotten better over time. Yours will too, if you're willing to put in the work.

Professor of Finance at UNC Chapel Hill; Faculty Director Investment Mgmt Center of the Kenan-Flagled Business School; Research Associate NBER; Editor at RAPS

Full guide here, along with every discussion I've given since 2007: drric.web.unc.edu/research/d… They've gotten better over time. Yours will too, if you're willing to put in the work.

10/10 A discussion is not an ambush. Slides to the authors the night before. If you think you've found a fatal flaw, tell them privately first. You might be wrong, and finding that out in front of an audience backfires spectacularly.

9/10 Use AI as the muscle, not as the brain. You read the paper. You come up with the comments. Then AI can tighten the writing and fix your Beamer. The other order is shirking, and people can tell.

8/10 Time constraints are binding. Dry run, time it, cut. If it runs 15 and your slot is 10, cut five minutes' worth. No exceptions. Don't tell yourself you'll go faster.

7/10 Starting early is always a good idea. Two weeks out: read it, let it sit. One week out: jot down comments, you'll have too many, that's the point. Three days out: write the script.

6/10 You are not Jerry Seinfeld. Few people are actually funny. Some senior people get away with it, and that's where the damage starts: juniors watch, conclude it's fair game, and repeat the joke. If you're a young scholar, you'll never regret a joke you did not say.

5/10 An out-of-the-box comment beats a check that ends in appendix 42. The narrow comment is where you're most obviously competent. It's also where only three people in the room can follow you.

4/10 Go the extra mile. Do some work: a small regression using related data, a manageable extension of the model, a magnitude check on the back of an envelope. Engage with the paper as a researcher, not as an auditor.

3/10 The right comments need the right delivery to land. Write a script. It's where you plan the pause, and the one "I want to emphasize this" that goes on the thing that matters most.

2/10 Take your audience's limited attention span into account. Two slides to summarize the paper. One is ideal. Don't fill the slides: the audience should be listening to you, not reading.

1/10 If criticism is all you have, you haven't thought hard enough. You're not there to outsmart the authors or catch them in a gotcha moment. The job is to help them make the paper better.

Remember how you fell in love with this job. The intuition, the idea, the "wouldn't it be cool to do XYZ," the night you couldn't sleep because you were scribbling equations. Approach your next discussion with that spirit. It can be one of the most rewarding things you do.

This summer a few junior colleagues asked me for feedback on their discussions. I realized I had a lot to say and nowhere to point them. Discussions are the one skill nobody teaches you in grad school. So I wrote it down. Ten suggestions for young scholars. 🧵

A day after the inflation awakening, implied rates keep on climbing. By early 23, the Fed will have hiked rates by an additional 200 points. Rates are expected to stay at that level for at least 6 months. Markets expectations and Fed much needed correction are finally converging.

Remember those days? We are now on track for 4 consecutive 75bps hikes. The Fed should have seen the writing on the wall and be forthcoming about its policy. reuters.com/world/us/feds-po…

Implied rate changes from current level for each of the next 11 FOMC meetings. Two more 75bps hikes are now a lock. Hard to believe that a little over a month ago only half of that was priced in.