dep director @devpolicy + econ assoc prof @ouranu. priors @dartmouthecon @stanford @ouranu @pmc_gov_au @uq_news. very personal account. 🧑‍🧒‍🧒+🌻+🦋
Canberra
- Tweets 53,040
- Following 3,081
- Followers 6,473
- Likes 87,567
Pinned Tweet
Some "overdue" news: my first paper on palm oil, started over a decade ago, was accepted for Christmas and is in print open access at the @JIntlEcon today. It's quite different from earlier versions, and the (excellent) review process improved it a lot.
sciencedirect.com/science/ar…
1
4
26
3,010
Ryan retweeted
Econ is greatly positioned for the AI wave because the obsession with publishing a few killer papers in top journals leads economists to use AI for quality enhancement rather than to increase quantity. This is exactly what you want!
1
11
73
3,639
Ryan retweeted
Immigration experts have never known more, but mattered less. In my new article, "Immigration for Realists," I argue that we have ourselves contributed to this gap and suggest ways to fix it. We misread the public, misrepresent the evidence, and lose our credibility as a result.
4
17
67
5,581
Ryan retweeted
He lives
Want to reduce crime on the subway? Enforce the fares.
My latest for @WSJFreeEx
12
171
3,839
145,498
Ryan retweeted
PSA to economist/EA parents: Esther Duflo has a children's book that's so well done that my son (3) asks for it every night. and now says for Christmas he wants to buy mosquito nets so the "friends in Africa" do not get malaria. highly recommend! amazon.com/dp/9353455863?lv=…
4
4
49
1,703
Ryan retweeted
Also the fact that the AEA allows you to “register” an RCT without posting a plan for analyzing your data was a huge mistake. The rate of “pre-registered” studies that are just a vague abstract and some poorly described outcomes is like 100%.
2
2
29
2,289
Ryan retweeted
The fact that historical practice with pre-registration has this incompleteness problem is important to recognize. But beyond the obvious need to discount whatever bonus you might have assigned to pre-registered studies, there is a path forward.
1/4
Also the fact that the AEA allows you to “register” an RCT without posting a plan for analyzing your data was a huge mistake. The rate of “pre-registered” studies that are just a vague abstract and some poorly described outcomes is like 100%.
1
3
6
1,335
NBER WPs signal value reduce to the reality that yes its just a WP for now, maybe a first version
If you are spamming weak genAI papers into the NBER Working Paper series, what is your world model?
What if everyone did this?
Maybe the perpetrators don't know they are making the world worse, they think their ideas are so so good and now humanity can benefit from them.
1
10
2,817
this is such a good paper
We think of trade costs as being about transportation and tariffs. But really, the cost of trade is above all the cost of information. It's harder to sell things far away because you know less about it. Cheap intelligence -- AI -- really matters here, as did mobile phones. 1/
10
1,023
Ryan retweeted
We think of trade costs as being about transportation and tariffs. But really, the cost of trade is above all the cost of information. It's harder to sell things far away because you know less about it. Cheap intelligence -- AI -- really matters here, as did mobile phones. 1/
1
20
93
5,879
I feel like this is truly a generational thing: it would be lovely if my kids liked me and cared for me when old, or had kids of their own, but i have no expectation of any of it and will be most happy with them exercising their own agency and doing what makes them happy
1
4
591
Ryan retweeted
PhD students do this sometimes — and since they're new to research and writing, often the AI *can* express the ideas better than them.
But if you outsource this step, when are you going to get good at it?
I worry this practice may be cognitively disabling.
3
5
41
9,893
Ryan retweeted
The International Rescue Committee is searching for a new research director--one of the best research jobs in the world for social scientists and designers. Oversee work on:
- humanitarian aid
- health
- climate
- education
- children
- sexual violence
lnkd.in/p/giU-3CCY
1
20
40
6,528
Ryan retweeted
Chris put together a really cool online version of the paper with interactive figures. It gives versions of our empirical figures under different assumptions etc. Which, we hope, is useful for readers! I'll write a thread on the empirics next week.
chrisedmond.net/unbundling-l…
In a new theory of bundling frictions, increasing productivity of cognitive skills explains changing patterns in within occupation wage inequality in the US, from @chrisedmond and @Simon_Mongey nber.org/papers/w35783
20
86
9,462
Ryan retweeted
Does LaTeX enhance perceptions of research quality in economics? sciencedirect.com/science/ar…
7
19
229
44,096
Collier at the sandwich bar energy @JustinSandefur
As I walk the streets of Berkeley, I see many dogs and few children. As more young people make this choice, what are the consequences for their older relatives? What are the social consequences of private choices? Let the debate and discussion begin.
1
541
Ryan retweeted
I read the paper, the Times article back when it came out, and I understand the reference perfectly well. That doesn’t make it above criticism. I think it’s gratuitous here. There are real problems with choosing to describe family relationships and women’s reproductive decisions in those terms, and pointing me back to Becker doesn't address it. Especially if, as you claim, you hope your research will lead to better and more conversations between families then yes you of course shoulder some responsibility for the rhetoric you choose to deploy as you push these ideas forward.
I would offer that you do not respond to your (predominately but not exclusively) female critics with suggestions that they must just be ignorant about economics or whatever it is
Have you read our paper? The Nobel Laureate Gary Becker introduced "The Rotten Kid Theorem" 30 years ago. Our paper is written for academic economists and our paper extends his ideas. Would you like to interview us for your journal? (1/3)
13
8
210
65,162