Nick HK

Sort of tangent to this, but I think it would be interesting to have a journal where the publication/r&r decision is the author's, but the final round of reviews are published alongside, and maybe the paper is given a quality grade or something.https://twitter.com/JakeVigdor/status/1075463641451786240 …

I wonder what is the most-spoken language in which nobody has ever rapped

But don't assume that the behavioral change fixes the real world problem, or even that removing the math fixes that problem. Those are both just guesses based on simplistic models (sounds familiar!) /fin

I think that Economics undergrad education could stand a more real world bent (doesn't everyone?). And additional behavioral where relevant would also be good. For example, I'd never teach anything about personal savings without talking about Thaler.

"teaching a bunch of heuristics and biases" and "teaching economics in a way that it can be applied in the real world" are not the same thing, and in fact in some cases the first can get in the way of the latter

Or take the common critique about the reliance on simplistic models like supply and demand. The problem with S+D isn't rationality, which it barely cares about. It's the competitive markets assumption. And it still provides super valuable real world intuition.

(imagine the delight if we did all put in a bunch of behavioral and suddenly we're teaching everyone Choquet integrals)

Further, these critiques treat "behavioral" and "real world applications" as equivalent, as well as "behavioral" and "less focus on math" which are both obviously wrong. Behavioral and rational models can both be real or abstract, math or intuition.

(can't tell you how many things I've had to change in my behavioral class in the few years I've been teaching it because something on the syllabus turned out not to replicate. I lead off the class with a warning about this)

Which ignores, of course, that (1) there are many, many reasons an econ model can be wrong, adding in behavioral will only address one slice (and could well make it more wrong), and (2) uh, behavioral is pretty new, a lot of it is still wrong too

The usual implication is that the rational model is proven wrong by behavioral, and so every wrong prediction and simplified model would become accurate if only you did it over again with heuristics and biases.

IMO any citation format requiring more information than author, year, title, publication, and link for any entry is obsoletehttps://twitter.com/JakeAnbinder/status/1074136801625952256 …

"The effect could be positive because of mechanism X, or negative because of mechanism Y. We look at the data and find that it's positive overall, implying X is stronger than Y." "Sure, you find that it's positive, but I bet once you take Y into account it's negative."

i wonder how much of a pay cut i would accept in exchange for never having to grade anything ever again

for the record: all else equal, it's much better for a dollar of debt to be carried by a grad student than an undergrad; grad students are much more likely to be able to afford to pay it back

i have a news article in Principles to explain supply elasticity about how people are buying more machinery to make bitcoin as price rises. today i'm replacing it with a rather opposite article.