Open ehuppert opened 3 years ago
Thank you very much for sharing your research with us! I have question regarding your recent project with Lyft. I think we are assuming that other features other than the price of a trip are equal for trips below and above the round price. However, I suspect that Lyft might have a sophisticated algorithm that assigns different waiting time based on the cent value of a trip. Have you found any feature of a trip changing below and above the round price?
Hi Dr. Pope, thank you for sharing your work. As someone who just went through the car buying process, I was surprised and intrigued by the results of your paper. It seems that for all the research, car facts, and consumer reports, our purchase may have been influenced by the smallest of heuristics. One of the questions that I had was how can consumers identify the most salient features in a buying process, given the overwhelming amount of information?
You also point out that there is some awareness on the supplier side in this phenomenon, whereas the same was not seen in the consumer. This may be slightly of out bounds, but whether humans can accurately describe their own decision-making and whether machines should also be expected to do the same, are questions we often debate in CSS. I wondered if the undergrad students in your survey would have identified the mileage as a major factor in their decision making or if they would have pointed to other factors.
Thanks for bringing the talk here! The conclusion is interesting, and I happen to know friends who run the used car business, hopefully they can apply this conclusion to increase their revenue!
Thank you for sharing us such an interesting paper!
Thank you so much for sharing the interesting research!
Thank you for sharing this interesting research on the left-digit bias!
First, I have a question about part V, which aims to understand the incidence of inattention on the different actors in the used-car market. Before part V, it seems that the wholesale operator conducts offline auctions, while Car.com offers customer choices online. So I wonder whether there could be biases caused by such a difference? For example, when people filter their choices online, they directly choose a range below 10,000 and search for car information, which might be different from the offline case when people could observe cars with miles just below 10,000.
Secondly, I wonder if there some differences between cars with miles above and below 10,000 miles. I feel that a crucial assumption here is there no much difference between cars with miles just above and just below 10,000 miles. But, at least in China, most cars will take motor vehicle maintenance at 10,000 kilometers, so I am thinking about whether there is such cost in the research.
Thanks
Thanks so much for the presentation, Professor Pope! I'm slightly curious about how such bias will change under gain frames, as mileage seems to be under a loss frame where individuals should be more sensitive to the digits, while it is also easy to imagine someone caring about every penny for his/her salaries or bonus under the gain frame - would that somehow potentially conflict with loss aversion to some extent? Thank you!
Thank you for presenting your interesting research. I was wondering if Lyft has an algorithm that optimized the price of operations. I am curious about how you have put that into your considerations. Thank you!
Dr. Pope,
Thank you for sharing your work with us! I have a more psychological question regarding the awareness and potential exploitation of consumers by firms and other market agents.
It has long been discussed that firms occasionally load too much information onto consumers in order to give an aura of transparency (thinking of hundred-page EULAs or terms and conditions). This informational overload potentially resulting in "forced inattention" has been utilized by the auto sales market among others for years. Since consumers are averse to too little information (deeming a transaction as "sketchy" or "dangerous') and too much information (making poor choices because they didn't read the fine print), how do we hold firms accountable for not preying on our natural cognitive biases?
This question may range too far into ethics and philosophy than behavioral economics, so I completely understand if it is not easily answerable in the time allotted.
Hi and thanks for presenting at our workshop. I was wondering - what types of branding or marketing strategies your research supports as ways to more effectively communicate with consumers?
Thanks for coming to our workshop. I'm curious if you think that these problems and biases can be more easily solved with a cleverer algorithm?
Thank you so much for the presentation! This research is very insightful and it is surprising that heuristic thinking can play such a big role in this market with relatively large stakes. The identification is quite clean the remarkable and the structural estimation for the inattention model is very convincing in accounting for the phenomenon.
Given the empirical pattern that the results are driven by final customers rather than professional agents, do you think we can design some clever mechanism to nudge these final customers to make rational decisions instead of suffering welfare loss in the used car dealership market?
Thanks!
Hi Professor Pope, I'm looking forward to your presentation tomorrow. It's interesting that this effect shows up when the buyers are all used car dealers, since my intuition tells me that they should be less susceptible to these kinds of heuristics in the domain of cars than the average person would be. So your idea that used car dealers are playing to their end customers makes intuitively more sense to me. The left-digit heuristic doesn't seem that concerning to me in the car context, but in other contexts you've suggested, like hiring decisions, seem concerning. Do you have any ideas on how it can be worked around?
Thank you for this interesting paper! Looking forward to your presentation.
Hi Professor, I wonder if the left digit effect is largely due to our every day experience, the left most digit has the greatest impact on our activities as it determines the range we are operating in. So for instance in an exam 90 and 97 are probably the same to a student but 83 and 90 are very different. I guess this is due, in part, to the way in which our number system is set up and the arbitrary cut offs imposed on it (90% is good, 87% not so good, xyz will happen 50% of the time). There is computational biology literature suggesting humans can pick up large changes in probability distributions but not smaller ones. It seems like thats what is happening here, you choose 40k miles and then you dont care if its 40,100 or 40,999 because in your every day experience as long as you focus on that leading 40 or 4 youre good. You try to reduce the cognitive load of every decision. I love this paper and I see it references the Basu paper that I am a big fan of.
Thank you for this interesting paper! Looking forward to your presentation.
Thanks so much for the presentation, Professor Pope! Looking forward to your presentation.
Thank you so much for coming Professor Pope! I am curious would it be feasible to test the heterogeneity among sellers given different sets of dealers by experimental design? Thank you!
Thank you for your presentation. I wonder how behavioral economics can interact with marketing science.
Thank you for this interesting paper! Looking forward to your presentation.
Thank you for the presentation. Very interesting topic indeed. As someone from a more psych and neuro background, I am curious if there are some biophysical constraints/optimisation objectives that results in this bias. Do you think the inattention parameter could correspond to some biophysical phenomenon? Totally understandable if this is outside of the scope of economist's intersts.
Thank you for your presentation. I am interested in the solution for this phenomenon. We have observed this effect and its influence, and what would be your advise to avoid or circumvent this issue? Thank you.
Thanks for coming! I'm wondering if the algorithm can be further improved.
Thank you in advance for your presentation! Could you talk more about the solution to such biases?
Hi Professor, I'm quite curious about what kind of interventions we can do to reduce the left-digit bias (eg. whether adding some energy labels would promote the choice of durables). And can we design some experiments to further explore this pattern?
Hi Prof. Pope, thanks for coming to our workshop! I think this topic is very interesting because we can find that commercials that utilize this left-digit bias all the time--899.99 seems much cheaper than 900.01! The used car market is also a great idea because we can estimate the value of the used car by its mileage. I notice you used a lot of different visualization in the body of your study, can you explain their intuitition? Thank you!
Hi Dr. Pope,
Thank you for sharing your research with us! It was interesting to learn about the left-digit bias, which is something I've never heard of. It's also very intriguing to think about the psychology behind why we think this way, and why we focus on these particular numbers.
Thanks again, and I look forward to your presentation!
Thank you for sharing your research! Looking forward to your presentation.
Thank you for sharing your work with us.
Thanks so much for the presentation, Professor Pope! It’s interesting to know how people’s “irrationality” in reality fits into the pricing model. Looking forward to your presentation!
Thanks for sharing your inspiring work! I believe the left-digits heuristics is underlying many markets, for what purpose you choose to investigate the car market? Could you please explain the potential psychological mechanism underlying the left-digits heuristics? Why would the customer exhibit such heuristics? Thanks so much for your patience.
Thanks so much for sharing this paper and look forward to your presentation tomorrow!
Thank you for sharing with us this presentation! I wonder whether increasing the number of permutations would increase the accuracy of the model.
Thank you so much for your interesting work. I am wondering if the sellers can customers both realize left digital bias, does that mean the problems you mentioned may largely cancel out?
Thank you very much for sharing your work with us! This paper sheds light on the left-digit bias when people make decisions. The structural model and the RDD identification is clear and easy to follow, and the conclusion about the inattention parameter is insightful. I have the following two questions:
Thanks again.
Thank you for presenting in advance. In terms of the left-digit bias, why do you think it is convincing to use the information provided by undergraduate students. As the survey shows, students are able to recall the first digit of milage, do you think it has a general meaning for other experiments.
thanks for speaking at our workshop! I'm looking forward to hearing your talk tomorrow.
Thank you very much professor Pope for coming to our workshop! The revealed point that even in markets with large stakes and easily observed information, information-processing heuristics matter is truly a surprise. I look forward to more intuitions behind this point.
Thank you so much for your presentation! I wonder if you have noticed some interesting heuristics on the nowadays social media that worth being discussed. I hope to get your view on the subject of social media.
Thank you for sharing your research, Dr. Pope! I was wondering if you are able to share some of the applications for your findings in other areas of interest.
Thank you for your presentation ! And look forward to your presentation tomorrow.
Thanks a lot for sharing! Following @MkramerPsych and @hesongrun , I wonder how the insights could inspire market regularization, as a potential effort to improve social welfare and enhance a strategy-proof mechanism design.
Thank you for the presentation! I also would like to hear more about the solution to such biases?
Thank you for the presentation! I have read papers about the application of deep learning models in behavior economics. I wonder whether you have ever considered comparing the traditional (behavior) economics models with new "AI" algorithms? Thanks!
Thanks for sharing your work Dr. Pope. I would like to hear more about the potential mechanisms at play while we unintentionally apply this heuristic in our everyday lives. Why do you think we developed this heuristic? Does it make comprehending written text easier for us? Would you say that this heuristic still applies for cultures where language is written from right to left like Arabic? Looking forward to the talk.
Thank you so much! I was wondering if there were any differences in these biases based on country context or other cultural factors or if this was a pretty universal bias?
Thank you for this interesting paper! Looking forward to your presentation.
Thank you so much for your paper! Really looking forward to your presentation tomorrow!! How do you think AI can improve the traditional economics research?
Thank you for sharing this interesting research with us. It is inspiring to see how deep learning framework can be successfully applied in economic problems. My question is how is this DNN model different from traditional methods in the field of economy?
Thank you very much, Professor Pope! I am very interested in the influence of heuristic processing on economic behaviors. I was wondering that who will be influenced more by the left-digit bias, consumers or sellers? If it's the consumers who should be more responsible for the discontinuous drop, then how can we try to eliminate the influence? For example, how can the sellers or the platform remind the consumers about the real difference between 7990 and 8100 is much smaller than 7600 and 7990?
Comment below with questions or thoughts about the reading for this week's workshop.
Please make your comments by Wednesday 11:59 PM, and upvote at least five of your peers' comments on Thursday prior to the workshop. You need to use 'thumbs-up' for your reactions to count towards 'top comments,' but you can use other emojis on top of the thumbs up.