As lots of people are liking a post I wrote that has the word "advice" in it, I need to issue my periodic reminder that my advice may in fact be terrible. I know very little, in general.
Story. Name changed to protect the (mostly) innocent.
My friend Todd is over 65 and lives in CA. The other day, Governor Newsom goes on national TV and says those over 65 are now eligible for the vaccine. Todd goes online, and makes an appointment at a mass vaccination site.
Professional news: I got tenure
@ChicagoBooth
! As of July 1, I'll be Professor of Marketing.
I'm very grateful for the time I've had here to start my career. I look forward to continuing to learn from my amazing colleagues.
Check-in worker loses her mind. Says if he gets a vaccine, she's calling the police. She won't be silent while someone tries to "jump the line"
Knowing if the police come, vaccines will be disrupted for the rest of the day, the director relents. Todd leaves unvaccinated.
A couple of thoughts to reviewers if you want to be helpful to the AE (particularly if it's me).
1. Be less negative (even if you say reject). If we took the average reviewer's advice, 99% of papers would go directly into an incinerator against a backdrop of evil laughter.
1/4
Director of the vaccination site eventually agrees with him. They have the dose ready. The governor told them >65 was eligible! They gave him an appointment. None of this is his fault.
To enter Canada you need a negative PCR test in the 72 hours prior to your original departure flight time.
Does this effectively do anything useful? As far as I can tell, all it really achieves is limiting admission to people who can afford to pay $375 for a rapid PCR test...
Funny enough, there's no place to turn around the car, so Todd has to go through the vaccination line, but is not allowed a vaccine. As he goes through the line, he keeps arguing. Shows the article on his phone. Re-iterates that he was allowed to make an appointment.
Upon arrival, Todd checks in and is asked for his health care worker ID. He's not a health care worker. He says "The Governor got on TV and said >65 was eligible. And the website allowed me to make an appointment"
Check-in worker says "NO" and puts a large X on his windshield
Warms my heart to hear from former students with life updates.
If you're wondering whether or not to send that life update email to a professor whose class you enjoyed, I say you should send it 🙂
This.
Soooooo many health econ talks start with some version of "we spend more on healthcare than any other developed nation but have lower life expectancy."
And soooooo few of those talks are talking about deaths among young people.
Important part of understanding the difference in life expectancy between US and peer nations is that it's driven by higher mortality for the *young*. Elderly Americans have some of the lowest death rates, while in the US ages "25–29 experience death rates nearly 3 times higher"
Really proud of this paper. I think the paper has a lot to say about the observed state of the world when it comes to overall TV advertising effectiveness and profitability.
I want to be clear about what the paper doesn't say.
1/
New estimates of TV advertising effectiveness across 288 CPG brands show that advertising generally has little effect on sales and most brands earn a negative ROI. The results cast doubt on the sustainability of financing TV content markets with ad revenue
When I first came to Chicago, ride share was consistently 50% cheaper than taxi.
Tonight to get home from dinner, a taxi was $15 to Uber/Lyft $37. And the taxi was 4 minutes (calling on Curb) wait to ride share's 8 minute wait.
Amazing how things have changed.
I really miss going to the office and seeing my colleagues on a daily basis. Getting coffee. Talking about research. Talking about not research. It's one of the things I like best about my job. And I really miss it.
Working Paper Thread! “Generalizable and Robust TV Advertising Effects” joint with my talented co-authors,
@Guenter_Hitsch
& Anna Tuchman.
In this paper, we independently estimate advertising elasticities across 288 pre-selected brands. 1/40
I have two students who just accepted faculty jobs, an RA who is joining our PhD program, an RA I'm trying to renew for another year.
All immigrants.
What are the consequences for all of these wonderful people? What can I do to help them? How can I channel my anger right now?
This is my number one piece of advice on "null results".
Before doing any analysis at all, think hard about what you would learn under any contingency of results. If the paper is only interesting if you get one specific result, maybe rethink your research question...
6) Be thoughtful when you start a project. I like thinking of a new project this way: “If I DO get a null, is it still somewhat interesting?” If I’m not sure the answer is “yes,” I try to think about the idea some more. Slow science, and letting an idea sit for a bit, is good!
About a month ago, U of I and Notre Dame had upticks in cases that were seemingly on every news site & all over twitter. Hadn't heard any follow-up, so I just looked it up.
Both in person and <1% positivity.
"We show that increased immigration significantly raises the staffing levels of nursing homes in the U.S., particularly in full time positions. We then show that this has an associated very positive effect on patient outcomes"
I always just lived the life I wanted to live pre-tenure. Figured if I didn't make tenure, I'd just find another job.
It's an underappreciated point that if you write good papers but just don't quite write enough for the place you start, you'll very likely get another good job.
I have refereed a lot of papers recently. A lot. When reading the other referee reports, I often see: (1) Reviewers suggest a result is "obvious" (2) Reviewers suggest a result is "not surprising." They are equally annoying to me but not the same thing.
My daughter was so excited for her first Thanksgiving that she came 10 days early. We were home in time for Thanksgiving dinner.
You'll be happy to hear that Mona has been an exceptional big sister to baby and protector of her mama.
Love to all of you.
Setting the fraud question aside. I think it's likely that we just generally write too many papers. Saying no to jumping on a marginal project opportunity is such an important (and difficult) skill for making sure you have enough time for your best work.
Plant based meat was never supposed to be about being healthy.
It is about allowing people to eat the junk food they like without the environmental and ethical costs of raising and killing animals.
Fake meat in a nutshell:
- Not better for you
- Safe to eat
- Highly processed but so what
- Better for the planet when it's instead of beef but not pork
- Better for animals
People think it's healthy so it will be a health minus but a planet plus.
@ashdgandhi
Yup. This was all figured out ex post. Todd figured he was eligible not just because of the governor, but also because he was able to make an appointment.
Oh other thing. Reviewers are quite often wrong about things. It's OK. We're all wrong sometimes, especially when we only spend a few hours on an airplane while jetlagged.
But given that there's a reasonable chance you could be wrong, maybe work in some humility/uncertainty.
New working paper!
"Estimating the Value of Offsite Data to Advertisers on Meta," with Nils Wernerfelt,
@tuchmanna
& Robert Moakler
We ask, "how much would it hurt advertising efficiency if Meta couldn't use 3rd party data for ad targeting?"
I go back and forth about whether we academics:
(1) genuinely don't realize we're left of the population (bubble effect)
OR
(2) do realize we're left of the population but just think our way should be the way despite its relative unpopularity ("we're so wise" effect).
I wish academics, both as individuals but especially as professional associations and other institutions, would think harder about how to maintain and even gain credibility in a world where their political opinions are well to the left of most people.
Not just grad school. Elite levels of anything do this, basically by design.
If you achieve high enough, you'll eventually be selected into narrower and narrower pools until you feel like you aren't good enough.
Drinking shit wine straight out of a bottle in complete darkness and complete silence with my roommates at our dining table right after we took our first micro exam.
No words, just shaking our heads and passing the bottle around.
What is your most unhinged story from grad school?
Mine was probably the time I picked a dead flounder up off the lab floor and then ran to a public bathroom with it in my hands and sobbed over its body in a stall.
Happy to say I'm starting on as a Co-Editor at
@QME_Journal
.
I think it is a great journal that has published many high quality influential papers at the intersection of quantitative marketing and economics, so I'm really excited to be a part of it.
A few thoughts:
The COVID-19 infection rate in countries under
@JoeBiden
's travel ban is LOWER than the infection rate in countries from which entry is allowed. (And of course, MUCH lower than US).
From
@ECDC_EU
and
@OurWorldInData
.
If you're submitting an empirical paper to an academic journal, please don't write the intro like you are doing policy advocacy.
Maybe it's just me, but it instinctively makes me believe your results less if your introduction reads like it is part of a political campaign rally.
Very interesting work on how economists perceive null results TL;DR.
1) Economists do penalize null effects
2) How to communicate the null matters.
3) Surprisingly, they penalize them even if that null effect is "surprising," inconsistent with bias in favor of surprising results.
1/ People often say that null results are penalized in the publication process. Is that true? In a new paper with
@cp_roth
,
@FelixChopra
, and Andreas Stegmann, we examine whether and why that's the case! Read on to learn more!
Working Paper Thread! “Preferences for Firearms and their Implications for Regulation”, joint with Sarah Moshary and Sara Drango.
We estimate consumer preferences for firearms and run counterfactual simulations predicting the effects of candidate polices.
Really quite excited to say that "TV Advertising Effectiveness and Profitability: Generalizable Results from 288 Brands" with
@Guenter_Hitsch
and
@tuchmanna
is now forthcoming at Econometrica
@ecmaEditors
I'm heading to Cambridge for the NBER Digitization, IO and Health Care meetings this week.
The summer institute is the time I have always felt the most heightened sense of professional & social anxiety. Even now that I'm tenured and am NBER affiliated. I suspect I'm not alone.
There is no bigger piece of *luck* that I did *nothing to deserve* than being born to amazing parents in a country where people aren't trying to kill me for no good reason.
I wish every child were so lucky.
Would love to share a new working paper "Depression and Shopping Behavior", joint with
@KatherineMeckel
.
It's a descriptive paper linking shopper panel data from
@iriworldwide
to ind level health survey data. We're particularly interested in depression.
Late in the game PREDOC opportunity!
I'm looking for someone to join my team
@ChicagoBooth
for the next 1-2 academic years.
Looking for:
-Excited/motivated to learn about quant marketing/econ/policy
-Coding skills
-Attention to detail
Apply here:
Lukewarm Take: The biggest difference between a talk that goes well and a talk that goes poorly is whether or not the research question is clearly and precisely stated from the very beginning.
Went to a restaurant last night that automatically included a "suggested" 25% tip. 25%! And not a large group, just the two of us.
The only way to give a more customary 20% tip is to cross out the pre printed tip and write in a lower number.
This is weird, right?
Important snarky thread. Politically, we hold Warren (and other Ds) to a *much* higher standard of how to pay for their pet projects than any GOP candidate in my adult life.
FWIW I prefer holding candidates to high standards, but would rather we did so symmetrically.
2. Evaluate the paper for the question it *is* asking, not for what you *wish* it were asking. Do your best to identify what you think the paper intends to accomplish. If all the reviewers disagree about what they think it intends to accomplish, that's diagnostic!
2/4
Referees are advice-givers, not deciders. Editors are deciders.
I also wish that editors mostly didn't send papers back to reviewers after revision. If the reviewer did a good job explicating the concern, editor should be able to tell if the concern was addressed.
A modest proposal: editors should state which subset of referee comments should be responded to.
If it's not in the list, it's not grounds for rejection at the next stage.
Some editors do this to some extent already 🙏🙏🙏
But it can be even more explicit. 1/5
Today is my grandmother's 91st birthday. I'm so thankful to still have my grandmother. I know not everyone is so lucky. I'm also thankful I'm able to come and see her for her birthday ❤️
"Experts" lying to people about their area of expertise in an attempt to get them to behave in ways consistent with the experts' values and often quite different from the target's values is bad, actually.
This seems to be a lesson we really don't want to learn.
It's not that climate change is fake or that it doesn't pose any serious challenges, but there is widespread misinformation about the current trajectory scenarios.
It's supposed to motivate more mitigation, but I think it mostly doesn't.
Excited to say that this paper is now accepted and forthcoming at AEJ: Micro! I’ve largely explained what it does already, but I want to pitch two reasons why you should cite the paper. One is theoretical about advertising, the other applied (Health Econ, specifically). 1/9
And now, as promised, a thread on my new *working* paper (feedback welcome): “Promoting Wellness or Waste? Evidence from Antidepressant Advertising” (link) 1/22
It remains a mystery to me why we aren't talking about nursing homes 24/7. Massive share of deaths. Huge scope for improvement, actionable *now*.
Seems not enough people follow
@DavidCGrabowski
. Please do.
The fun part of
@JDVance1
being the VP nominee is we can tell everyone about the time that
@smilleralert
,
@asacarny
,
@agoodmanbacon
and I argued with him on twitter dot com over the meaning of "lack of statistical significance"
Khoa has subtly made an very deep point here.
While the causal effect you estimate might be enough to convince you of a policy's merit, it might not convince someone else, *even if they believe the causal effect is correct.*
I feel like the pro-gun crowd won't read through the 70-page-long AER papers w/ rigorous analysis on the causal effect of gun policies that you are sharing and reconsider their political stance on the issue though.
Our celebration of
#historicpreservationmonth
continues as we highlight some of our dedicated historic preservation specialists. Elaine Shapiro is a
@ttinc
MVP in our work on the restoration and modernization of Chicago's iconic Wrigley Field ballpark.
Really annoys me to hear/read "I controlled for everything." If you controlled for everything, why is there still variation in X to estimate beta?
Theoretical justification of why residual variation is as good as random > "I controlled for everything"
Jason explains nicely why.
There are two facts which make correlations controlling for confounds uninformative about many (but not all) causal effects: 1) the R^2 of the mechanisms we understand is low, 2) our uncertainty about not well-understood mechanisms should be high. (1 / about 13-15)
Hey
#EconTwitter
! As some of you know, my family is in the wine importing and sales business. (Quigley Fine Wines -
@Quigleywines
).
They are opening a beautiful Tasting Room in San Diego! The *very same* San Diego where
#ASSA2020
will be held! Soft opening will be DURING ASSA!
As far as I can tell from my end, nothing has changed on Twitter other than people on my timeline talking about how all the changes are horrible.
Did I miss something?
I've often believed the lie that if I only got this, or did that, or was included with them, or achieved this milestone that life would finally be good.
Helpful for my soul to see Dax Shepard shatter that illusion.
"I encourage you to think about the other referee comments and incorporate them if/as useful, but if I did not specifically mention them above you do not need to give me a specific “response” to them."
Gah I love this editorial attitude...
I'm a die hard Bills fan.
But this is a terrible deal for the people of New York.
Giant transfer from the taxpayers to one of the richest families in the state, for almost entirely private gain.
I can't tell you just how disappointed I am in this administration's disposition towards asylum seekers.
Biden campaigned on a more humane immigration policy and he has failed spectacularly.
@JohnHolbein1
"If R2's priors (and lack of surprise) are based upon a paper in the literature we failed to cite, we would be happy to cite it and are happy to provide additional supporting evidence on this important question. If not, we're thrilled to bring rigorous evidence to R2's priors."
3. Unstructured lists of ~18 "concerns" aren't helpful. Would be great if you could highlight the 1-3 biggest issues the paper faces. If suggesting reject, it helps us to know what you think is *the* pivotal reasons for reject. If revise, need to know what's really *needed*.
3/4
This is my argument for why health economists should study opiods, car accidents and guns.
Health econ has clearly studied the first a lot. But could definitely use more health economics thinking on the second two.
Contrarian thoughts for
#EconTwitter
discussion du jour:
1. Getting PhD/doing research is not a good fit for all.
2. Constantly consider your outside options. They are almost always better than you realize.
3. Exercising outside option does not make you a failure at life.
Congratulations to Rachael Meager who was presented the AEJ Best Paper Award for Applied Economics at
#ASSA2023
for her paper “Understanding the Average Impact of Microcredit Expansions: A Bayesian Hierarchical Analysis of Seven Randomized Experiments.”
Immigration is America's superpower, so this should be raising alarm bells for policymakers:
The United States is no longer the top choice of foreign talent when asked where they would move for work.
4. If you're suggesting R&R, give a proactive reason why the paper *should* be published. What do you see as the important contribution? How can the authors better highlight and support that contribution?
Much more helpful than a list of demands or "concerns".
4/4
Kitkat and Reese’s Peanut Butter Cup customers won’t tolerate price hikes. For every 1% increase in pricing, volume drops 2.1% and 3.9%, respectively. - UBS
Reminder PARALLEL PRE-TRENDS IS **NOT SUFFICIENT** for diff-in-diff to work! You need parallel *counterfactual* trends post treatment! No amount of matching can make that assumption empirically testable. It's an assumption that has to be justified with theory.
Do you use diff-in-diff? Then this thread is for you.
You’re no dummy. You already know diverging trends in the pre-period can bias your results.
But I’m here to tell you about a TOTALLY DIFFERENT, SUPER SNEAKY kind of bias.
Friends, let’s talk regression to the mean. (1/N)
Found and made an appointment for my 79 year old neighbor to get vaccinated (in IL). Drove her in the snow. Helped her get checked in.
Afterward she said she never would have been able to get an appointment on her own. She had tried for weeks before by calling sites directly.
Every day when I walk Mona, I pass by a neighbor's house, and say hi to their aging golden retriever. I always told both he and Mona to sit, and I would break a treat in half, giving each half. Almost every day for the last 6.5 years.
Yesterday I learned that he passed.
🐾💔
2020 refereeing observation: Every time I said "Accept as is" and another referee still wanted more revisions, other referee became de facto dictator, and my "Accept" largely seemed ignored.
Lesson: Don't say "Accept " without an accompanying strong affirmative case for paper.
I find life to be more enjoyable when I assume that the people I talk to are decent people, acting in good faith and aren't out to get me or offend. And I think it's an accurate assumption basically all the time. And even when it isn't, I'm still happier when I assume it is. 🤷♂️
I never was a bath guy. Then I moved to a house with an excellent bathtub and Sonos in my bathroom. That and I realized that hot bath water was essentially a weighted heated anxiety blanket.
Now incense is burning, Norah Jones playing and I'm shit posting.
Life is good.
Hi folks! Happy to say I'll be taking on the roll of Public Editor
@QME_Journal
, together with
@SeilerStephan
.
Follow the account for news from QME and summaries of new papers coming out in the QME!
Hello Twitter! This is the official account of Quantitative Marketing and Economics ().
The account will be run by the new QME Public Editors
@btshapir
&
@SeilerStephan
.
Please follow for summaries of new papers coming out in QME!
I've had the same faculty support person my entire time at Booth, Stephen. He's amazing. Over the years, he accumulated other responsibilities and has little by little supported fewer faculty. He just got promoted high enough that he won't be doing any faculty support anymore.
My grandmother's nursing home is badly understaffed. This issue hits very close to home for me.
It's quite clear that having insufficient hands harms her quality of life. Upsetting but not surprising to learn that it also harms health outcomes.
"We show that increased immigration significantly raises the staffing levels of nursing homes in the U.S., particularly in full time positions. We then show that this has an associated very positive effect on patient outcomes"
I spent the last few days in Buffalo visiting my Grandma Fran. It's always a good day when I can play cribbage with my Gram.
Visiting her nursing home it makes me smile every time she interacts with a nurse or aid. She treats every last one of them as if they were her grandkid.
Periodic announcement:
I don't hire from cold emails. If I'm hiring a research assistant or postdoc, I list it formally, advertise it broadly and do a proper search.
We also don't admit PhD hopefuls from cold emails- there's an official application process and timeline.
@economeager
@hamandcheese
@ne0liberal
I've always been a big believer that further explaining a joke when everyone already understands the references and the punchline makes it much funnier. :-)
At the time it seemed like a coordinated effort prevent economist influence on policy.
Many economists had good ideas that were suggested in good faith and were attacked or mocked.
I wish we had listened more to
@ATabarrok
,
@IvanWerning
,
@ProfEmilyOster
, Eric Budish & others.
Fauci interview “we should have listened to economists”
Painfully takes me back to early 2020…
My coauthors Acemoglu Chernozhukov Whinston and I put out a paper studying costs vs benefits and plus one overlooked essential economic idea: Targeting.
Reaction: Not entirely satisfied with the paper after revision, but it's not wrong and it has a clear contribution.
Recommendation: Accept
I feel like this is a place that typically leads to reject or another long list of revision demands. But I think it should be an accept.
I've crossed 4,000 followers without noticing, it seems. Why 4,000 of you follow me is beyond me. 😂
Sending much love to all of you. Peace, love, dogs and soup.
Sarah Moshary & I are hiring a pre-doc
@ChicagoBooth
.
Looking for someone who is:
- Organized & pays close attention to detail
- Excited to learn about research in IO, marketing & policy.
- Interested in going on to a PhD in related field.
@econ_ra
Smaller cities have also become better to live in in the last decade IMO. Renewed downtowns, better restaurants, museums, etc.
I think the superstar cities managed fine w/ terrible housing policies when the outside option wasn't great. But now there are lots of good options!
It is truly remarkable when lawyers, consultants, educators + advertising folks feel they can’t afford to live comfortably in your city.
That is where NY, SF, DC, LA and Seattle are now.