When Israeli supermarkets were forced to adjust prices under a reform, firms learned new strategies and improvedโbut still priced inefficiently.
Research by Avner Strulov-Shlain ๐๏ธ shows how firms learn from experience. https://ow.ly/IPYm50XjOT1
Just published in @jpube.bsky.social:
"Impacts of home-care subsidies: Evidence from quasi-random assignment"
By Yuval Ofek-Shanny, @avnershlain.bsky.social, & @dzeltzer.bsky.social
www.sciencedirect.com/science/arti...
#econsky #publiceconomics
โฝ For an academic study, we (@akraft.bsky.social and I) wish to interview US Gas Station owners to learn more about their business practices. Please share or send anyone you know to Andreas or me. Thanks!
Whimsy Planning Commission announcing their improv show every last Tuesday of the month at Woodlawn Tap at 7:30pm
We'll be back at Jimmy's on Tuesday 3/25 and EVERY LAST TUESDAY OF THE MONTH! Come once and you'll come again.
@abionsky.bsky.social @williamhjhubbard.bsky.social @avnershlain.bsky.social @dfordon.bsky.social @elderserpiente.bsky.social
Thanks Neale!
Thanks, Colin!
I am very happy to share that I am being promoted to Associate Professor of Marketing at The University of Chicago Booth School of Business.
I am incredibly grateful for my family, and for my friends, coauthors, mentors, and colleagues who made it all possible. Excited for the years ahead!
A math question for 7th graders in Chicago Public Schools showing a linear relationship between cars odometers and price.
Decrease in average car price with discontinuities at 10k miles. From Lacetera Pope and Sydnor (2012)
No wonder retailers ignore left digit bias. Indoctrination against it starts in school!
Cc:
@justinsydnor.bsky.social @nicolacetera.bsky.social
My reaction to @cfcamerer.bsky.social as well ::blushing::
@annastansbury.bsky.social, your question is a core interest of mine and I have a couple of projects in the works on the topic. I would love to talk more if youโre interested.
And it is also crucial to note here that if the price list range or steps were different, the confused results would have been different. It just so happens that random answers => the four fold pattern.
An instant hit paper claimed that "Decisions under Risk Are Decisions under Complexity".
Re-analysis shows no evidence in support of the claim, and perhaps even the opposite:
- The experimental design causes noise to generate the results.
- For non-confused participants risk != complexity.
๐จ Oprea (2024 AER) argued that prospect theory choice anomalies were not due to risk, but due to complexity-driven mistakes.
But this new analysis convinces me that Oprea (2024) is substantially wrong. In my opinion, the paper should be retracted.
This figure shows the relationships between expectations, goals, and performance that we test in this paper. Solid lines indicate those that are supported by data, while dashed lines are not.
In summary:
- Higher expectations increase goals (A) and performance (C).
- Goals reflect expectations but do not affect them (!B).
- Having a goal improves performance (D) but the level does not matter (!E).
Check the paper for more details and findings papers.ssrn.com/sol3/papers....
The figure plots regression-adjusted average performance across treatment groups. Coefficients and 95% confidence intervals from Table 3 with the mean in the omitted treatment (Medium practice, Baseline Goal) added. On the left panel, we plot the effects of Practice Difficulty assignment, and on the right the effects of Goal Elicitation treatments.
The figure plots the histogram of performance (number of tables solved) relative to the goal in the Baseline goal treatments. Data are winsorized below at -25 and above at +25.
How do goals & expectations affect what participants do?
Easy practice & simply setting a goal both improve performance.
Goal level itself matters less once expectations are set.
Participants care about the goal (see bunching), but only locally.
The figure shows the regression-adjusted average expected probability of reaching the goal for the different treatments. Coefficients and 95% confidence intervals from an OLS regression of the expected likelihood of reaching a goal on Practice Difficulty and Goal Elicitation treatments, controlling for skill. Mean in the omitted treatment (Medium practice, Baseline Goal) added.
The figure plots the regression-adjusted average goals (dark circles) and median expected performance (light triangles) across treatment groups. Coefficients and 95% confidence intervals from Table 2 with the mean in the omitted treatment (Medium practice, Baseline Goal) added. On the left panel, we plot the effects of Practice Difficulty assignment, and on the right the effects of Goal Elicitation treatments. For both outcomes, sample is limited to those who reported monotonic beliefs.
Do goals shape expectations? We find that expectations drive goal-setting.
Manipulating expectations changes goals, but tweaking goal difficulty doesn't significantly affect expectations.
The figure plots a histogram of the self-reported believed likelihood of reaching their goal across participants who set a goal in the Baseline goal treatment.
What goals do people choose?
Participants typically set goals they expect to hit ~78% of the time. Few choose goals they think are less than 50% likely to be achieved.
In this new working paper we use a real-effort experiment to study the relationship between self-set goals and expectations, and how they affect performance.
Do goals affect or reflect expectations? Do goals affect performance, conditional on expectations?
Reassessment?
(Not a Netanyahu fan) but getting the four soldiers home and then standing grounds against violations was smart and fruitful.
www.haaretz.com/israel-news/...
An alternative is that they think it will not collapse and they are trying to show strength of sorts and exert as much psychological torture on Israelis.
Of course itโs not good. But if you were Israel and Hamas started violating the agreement ever so slightly but blatantly, what would you do?
As @joshtdean.bsky.social figured out โ Nighshade is of no help here.
At least the answer doesn't change
@aleximas.bsky.social referred me to Nightshade (nightshade.cs.uchicago.edu/whatis.html). Basically, if we find a good-for-its-time way to hide the instructions from AI, such as Nightshade, then can proceed.
I have so many responses but instead ::sigh::
Itโs better to combine losses ๐ฌ
The end of online experiments?
in Hebrew we say you are in a state of โemotions stormโ or even just โstormedโ when having lots of emotions. I am stormed with hope, relief, anxiety, dread, regret, gratitude, anger (to name a few).
Welcome back Romi, Emily, and Doron โค๏ธโ๐ฉน
twitter.com/haaretznewsv...
A road-trip inspired research idea, free disposal:
Estimate the effect of speed limits on accidents by using interstate roads and difference in states limits. E.g, the I-55 is up to 75mph in AK, 70 in MO and 65 crossing to IL. With good enough data can be quite credible IMO.
Apparently.
Also, mine wonโt be the 100th streak to start on Jan 1st. I have a pretty long no-running streak though.
Cool. Whatโs the data here? Initially I thought itโs your streaks but even you canโt have 99 January firsts in eight years.