Charles Martineau

John Cochrane of the Uni­ver­sity of Chicago wrote a bril­liant advice paper to PhD stu­dent few years ago. The paper should be read reli­giously by all PhD research stu­dents. The sec­tion on sem­i­nar pre­sen­ta­tions (sec­tion 4) cov­ers short and sweet guide­lines to increase the prob­a­bil­ity of suc­cess when it comes to pre­sen­ta­tions. It may be incom­plete on var­i­ous pre­sen­ta­tion aspects but nonethe­less one of the best place to start think­ing and struc­tur­ing your pre­sen­ta­tions. Below I copy/paste the sec­tion 4 of the paper (page 11–12). Read it care­fully… it is not long!

4 Sem­i­nar pre­sen­ta­tions
You will not believe how fast the time will go by.

Since time is lim­ited, it’s espe­cially impor­tant to get to the point. We can’t skim to the
impor­tant stuff in a seminar!

You don’t need any lit­er­a­ture review or moti­va­tion in a sem­i­nar. Just get to the point.
Gene Fama usu­ally starts his sem­i­nars with “Look at table 1.” That’s a good model to
emulate.

Don’t “pre­view” results. It wastes time; why say it twice rather than say it once, right?

Don’t make slides with a bul­let point for every word you intend to say. This forces you
into a pre­planned order, and then you can’t change on the fly when you figure out how fast
time is going by. Slides are fine that only con­tain equa­tions, tables and graphs — things
we really need to see. At most use words for the one or two really impor­tant things you
want peo­ple to know, e.g. “Identification: inter­est rates do not respond to fiscal shocks in the Ricar­dian model.” Also, you want peo­ple to remem­ber the struc­ture of the model, definitions of vari­ables, etc. If you have too much junk on the slides, peo­ple can’t see the util­ity func­tion while you’re talk­ing about the pro­duc­tion func­tion, so they get lost. Peo­ple don’t remem­ber equa­tions from one slide to the next.

You have to leave slides up for a decent amount of time in order for peo­ple to digest
them. That means you will not be able to put up 1 slide per minute!

As in writ­ing the paper, your main objec­tive is to get to the #1 impor­tant con­tri­bu­tion
as fast as possible.

Most sem­i­nars are a dis­as­ter. They start with point­less moti­va­tion and pol­icy impli­ca­tions, which the audi­ence can’t fol­low since we don’t know the result. Then we get a long
lit­er­a­ture review, which is even more bor­ing since we don’t know the point of this paper
much less what every­one else did. Then we get a results pre­view. Usu­ally, the pre­sen­ter says “I’ll pre­view the results now because I may not have time to get to them all,” a strangely self-fulfilling prophecy. Since show­ing the main results is the only rea­son you came, why not just start right now! Worse, the rea­son we run out of time is because we wasted half an hour on the stu­pid pre­view! The sem­i­nar then bogs down as peo­ple start ask­ing ques­tions about the pre­viewed results; most of the ques­tions are dumb (“I mea­sure the demand elas­tic­ity at 0.3.” “But how did you iden­tify sup­ply shifts?”) since they will be explained in a proper pre­sen­ta­tion of the results. But the ques­tions are totally rea­son­able since the claim with no doc­u­men­ta­tion is mean­ing­less. Next, we get (in empir­i­cal papers) some “theory” that is really beside the point and only serves to pro­voke more need­less argu­ment (no, there really is no way to dis­tin­guish the “behav­ioral” and “ratio­nal” expla­na­tion. Clever audi­ence mem­bers will come up with sto­ries that reverse all the signs.) Then we get some dis­tract­ing pre­lim­i­nary results and tables and graphs of unre­lated obser­va­tions. More point­less dis­cus­sion erupts; peo­ple don’t know what point the speaker is try­ing to make and the dis­cus­sion goes off in to tan­gents. Finally the speaker sees there is only 10 min­utes to go, tells peo­ple to be quiet, and the main results go by in a big rush. Every­one is tired and con­fused and doesn’t fol­low any­thing. I timed the finance work­shop last win­ter quar­ter and not one paper got to the main results in under an hour!

Lis­ten to the ques­tions, all the way to the end, then count to three before answer­ing.
Yes, you’re in a rush, and yes, you think you can guess what the ques­tion will be and you
know the answer. This isn’t a game show, and much of the time you actu­ally don’t know
what the ques­tion will be.

Keep a sheet of paper handy. You may not have a quick answer to every ques­tion, and
some ques­tions may point to good things to change in the paper.

You can­not make it too sim­ple. Most pre­sen­ters, espe­cially Ph. D. stu­dents over­es­ti­mate
dra­mat­i­cally how much the­ory peo­ple can digest in one sit­ting, and how quickly they can
mem­o­rize and digest mod­els and results.

Speak loudly, slowly and clearly.

There’s noth­ing wrong with end­ing early!*

—

* I bolded the last line… and not John Cochrane. Fin­ish­ing ear­lier does not kill anyone.