David Powell

I try not to think too much about what I eat since I’m supposed to avoid processed foods.

Stop saying correlation does not equal causation. Is correlation > causation? Less than?

A lot of parenting is trying to find healthy foods that your kids will eat and when they actually do like something deciding that means that it must not be healthy then and restarting the whole cycle.

If you’re overidentified in one paper, you should be allowed to be underidentified in your next one.

When do they announce the Nobel Prize in Literature Reviews?

A conference where you can discuss whatever working paper out there is bothering you.

Summary below with results from 3-year study in Kenya assessing whether savings interventions can strengthen economic prospects for the future and also reduce engagement in behaviors that increase HIV/STI risk. Our results are extremely promising!

Partial identification is for quitters.

Elaborate gender reveal parties…but for journal decisions.

There’s always a chance that you won’t make it to your destination. So when someone asks for your ETA, you should tell them that it’s undefined.

This paper’s main contribution is that it does not claim to make any contributions, opening the door to all types of new publications which also don’t make contributions.

I am a one issue voter — stop letting indoor playgrounds require you to purchase their “special” 4-dollar socks.

A Jeopardy spinoff in which everyone must answer in the form of an exclamation.

People are worried about the water usage required for AI data centers, but economists see an opportunity to use a rainfall instrument.

Ugh, every journal issue is special.

A true Labor Day would be randomized across people so we could identify the causal effects.

People use abbreviations without even knowing what they stand for. Few people can tell you what i.e. is short for. The “i” is short for “i didn’t” and the “e” is short for “explain that well enough the first time so here’s my second attempt.”

Everyone knows that if you have young kids, you end up going to a lot of birthday parties. But they don’t tell you that you’ll end up talking to parents there that don’t even know their ORCID.

My advice would be to write a paper about sample selection bias. Looking at papers about sample selection bias published in top-5 journals, they get a lot of cites.

Academic hack: Be such a notoriously awful reviewer that everyone adds you as a coauthor.

Imagine how much money we’d save on signage if everyone’s default assumption was that surfaces are slippery when wet.

Is anyone currently studying the biggest public health concern of our times? (How my kids like to touch all my food before washing their hands)

Can raising the minimum wage help curb “deaths of despair?” LDI Fellow @davidpowell.bsky.social found that minimum wage increases had, at most, modest effects on overdose deaths and suicides—suggesting more targeted health policies may be needed. Read more: https://bit.ly/4g79GVk

I’m glad that the defibrillator at this playground is labeled an “Emergency Defibrillator” so I know not to use it recreationally.

This paper has several limitations. First, not all researchers will appreciate it. Only the smartest and best looking will believe it should be immediately accepted at a top journal.

It’s so cool that we have all these AI tools now to free up time to talk about AI tools.

They’re called standard errors because everyone gets them wrong.

Probabilities can be unintuitive. Take this example: when I drive one kid, there’s a 70% chance that they will need to go to the bathroom within the first 10 minutes of the drive. But if i drive both of them, there’s a 180% chance we will need to stop in the first 10 minutes.

Just read a paper that said “Data is…” when it clearly should say “Data am…”

I wanted to share some notes from spending the day not at NBER: 1) I was a comfortable temperature. 2) I had decent chicken for lunch. 3) I made a casual remark about something and didn’t receive a response using the phrase “diminishing marginal utility” or “reverse causation”

Good reminder of how hard it is to access health insurance for people who are disabled by their conditions but in the waiting period for coverage for SSDI/Medicare. The mortality during the wait period for people with cancer (neoplasms) is nearly 40%. www.healthaffairs.org/doi/epdf/10....

YOU are producing the evidence. End every paper title with “: Evidence from Me”

“It’s like trying to put sunscreen on a three year old” seems like it would be a useful expression.

If you need beta-carotene, just regress on carotene.

I invented a shift-non-share instrument but I’m not letting anyone else use it.

To give you a sense of just how long 250 years is…if you had started working on three economics papers on July 4, 1776, you would probably already have two published.

Stop assuming your popcorn is independent.

What was the meat version of Arrow’s Impossibility Theorem?

I found it

If you review a paper and the authors thank you for the comments and say that they improved the paper, then you can add “influencer” to your bio.

The most underappreciated people in our profession are the ones that select the “best referee” winners and I hope we can find a way to recognize them.

I do worry that economists might use AI to begin writing papers outside of their field.

I have decided to accept a recent compliment of my artwork as “pretty good” though it was given under the impression that our six year old drew them.

Trying to work outside more to improve the external validity of my results.

Your paper shows a clear positive effect. Please add a model which shows that the effect is ambiguous.

This paper tries to do much. Please split it into three papers so we can reject each one individually.

Change the default. Make everyone not thinking about submitting a proposal send a Letter of No Intent.

This playground says that my kids play at their own risk but little does it know that I have mitigated some of that risk through something called a high deductible plan.

If you’re facing a word limit, one way to cut down on words is to delete the limitations section of your paper and then just not have any limitations in your paper.

Do people also hallucinate fake papers? Is that the point of psychedelics?