"Mr. Hansen's models could lead a casual observer to conclude that race plays a
significantly larger role in Harvard s admissions process than it actually does... The models
incorporate far fewer variables than those prepared by the parties economic experts for this
litigation and omit many variables that are important to the admissions process. Compare [PX12
at 33 ] , with [ PD38 at 26 ]. Even Mr. Hansen's most complete model almost certainly suffers
from considerable omitted variable bias in light of the likely correlation between race and
important variables that Mr. Hansen did not include. Most notably, his models contain no
controls for socioeconomic and family circumstances that correlate with race and also affect
admissions decisions. See [PX12 at 33] . Given these deficiencies in the models, they are
entitled to little weight for the purpose of determining whether Harvard discriminates against
Asian American applicants, particularly given the availability of the experts far more
comprehensive models and the testimony offered by fact witnesses in this case. See Oct 19 Tr.
19: 19– 20:8... Hansen's models do suggest, consistent with other evidence, that Asian
Americans applicants excel in academic metrics; that tips for legacies and recruited athletes
result in more white students being admitted; that a projection of Harvard' s class based only on
the profile ratings, academic metrics, and athlete and legacy statuses is incomplete and results in
a projected class that is vastly less racially diverse than the one Harvard achieves; and that ,
absent any consideration of race,Harvard s classes would have drastically fewer African
American and Hispanic students."

A group of colleages and I have been thinking about the sunk cost fallacy a lot lately, contemplating at its prevalence and the possibility that it may be a features rather than a flaw.

So I wrote a cute story about it.
 
Assumption -> in caveman times, life expectancy for infant born with n guardians (parents/elders etc) is higher than life expectancy for infant born with n-1 guardians.
 
(written in my finest caveman dialect)
___________________________________________________________________________________________________________
Grog and Urg are male and female cave-persons, respectively, who share cave. They are part of local tribe, where each member serves distinct role. Grog is hunter for tribe, Urg is gatherer. Grog and Urg are familiar with birds and the bees, and are expecting baby Grug in many moons.
 
Suppose, in waiting for Grug to arrive, things between Grog and Urg deteriorate. Grog begins long think about Grog, Urg, and baby Grug.
 
First, picture a version of Grog that “thinks at the margin”. His decision to leave the cave (and Urg) would be made by weighing the benefits of leaving Urg against the costs of leaving Urg. Among the benefits may be more disposable income, and possibly more leisure time. Among the costs would be the possible disutility of living alone, and the uncertainty regarding baby Grug’s survival. If Grog doesn’t care much about baby Grug’s life expectancy, and therefore doesn’t care about passing his genes to offspring, then his assessment may be that the benefits of leaving Urg exceed the costs of leaving Urg, and they separate. Grug is raised with one less guardian. If baby Grug really does have a lower probability of survival with one less guardian, then the probability of Grog’s genes surviving in the long-run is low if he thinks this rationally.
 
Let’s now imagine a Grog that falls for the sunk cost fallacy. This would be Grog looking back on all the time invested in the relationship. The time spent fending off wolves from cave to protect sleeping Urg, the times Grog went without food so that Urg could eat, etc. These sentiments make him overestimate the cost of leaving, at the margin. So Grog continues to invest in the relationship: continues to fend off wolves, give Urg his gruel, etc. The result of his continued investment in the relationship is that Grog and Urg do not separate and raise baby Grug together. So, it’s possible that Grog falling for this sunk cost is really his lizard brain trying to commit Grog to propagating his genes- the child has a better shot at making it past infancy with both Grog and Urg together. So he falls for the sunk cost fallacy and baby Grug has a better shot at survival. Furthermore, Grug has better odds at surviving long enough to have a child and pass the genes responsible for this behavior to the next generation. Probability of Grog’s genes surviving in the long-run is higher than the case above.
___________________________________________________________________________________________________________

So what is this story really saying? Alexander McQuoid paraphrases my story in the following way:
Sunk cost fallacy helps overcome inefficiencies associated with externalities? If I only think about my private forward looking utility, I will systematically undervalue actions that impose externalities on the community, which leads to long-term inefficiencies that can lower my ability to pass on my genes? So systematic selection of genes based on those that make decisions thinking about sunk costs, not just on the margin. 

Related to the following post on reddit.com/r/academiceconomics

Q: I'm a freshman at a not so great school for economics (didn't get into the big name schools and this was cheapest option), but I really like economics. In high school I won a bunch of awards in economic competitions. Not trying to brag, just to prove that I'm interested and have some aptitude.I've reached out to my current professor and other professors in the department but none of them are really doing any research. The few that are just want upperclassmen to do data entry work for a long term project. Most people at my school just take economics to work in business. I was recommended to do the MA in four years but that seems like a long term thing.

A: 

​My perspective as a tenure-track faculty member doing empirical research at a small public 4 year degree granting school:
 
You are a freshmen, so unless you are a significant outlier, you've not been exposed to econometrics.
 
Many faculty may hesitate to have you do anything other than data entry if you haven't taken econometrics. Why? Econometrics is a language that you need to learn to speak fluently before you can really add value beyond simple tasks like data entry and literature reviews. Don't forget, having you involved isn't costless, many of us are quicker at these tasks on our own or with colleagues, rather than having a student involved.
 
So given I have to invest time in the relationship, I need to invest that time in an optimal way. What is not optimal is for me to attempt to teach you econometrics just for the sake of having you as an RA. You'll take the class, it's a requirement, so this just wouldn't make sense. This is absolutely nothing personal, it's just prudent time management.
 
I poach the top 2-3 econometrics students as research assistants and co-authors on wild ideas we have a common interest in, but wouldn't really consider students early in their career because they need to develop their intuition more through exposure to the coursework. If you think you can add value beyond just data entry tasks, then signal it, show them you can add value. It would definitely shut up my concerns if you walked in with some code and preliminary attempts at addressing an interesting question.
 
You shouldn't discount the experience of doing simple tasks like data entry at the very start of your career. You signal your interest in the field and that you are serious about the profession, you may form great bonds with professors or other RAs, you will do a lot of reading that will help you write better, you will learn nuances of cleaning data, etc etc etc.
 
My guess is the regional Fed would send you info about one of their public outreach programs, the Fed challenge, etc, as ways of being exposed to economics and research. It's not reckless to reach out, but I wouldn't hold your breath.



​

I'm super pleased to announce that my article with a dear friend and colleague Alex McQuoid has been published in Applied Economics!

We have a few free copies that can be accessed if you click here. At some point this link will no longer provide free copies, a pre-print version can be found here on ResearchGate and embedded in the post below. 

Of all the various podcasts I listen to, The Behavioral Grooves podcast (also available through iTunes) has consistently been the most enjoyable entrant to the list. Since it's summer, I've been doing some leisure reading in the area and joining a few social media groups.

On the behavioral economics group on LinkedIn, I recently stumbled on an annual guide in applied work, The 2018 Behavioral Economics Guide. Seems to be great reading for anyone trying to learn more about the field. The guide has an introduction by Robert Cialdini , author and Professor Emeritus, and editorial content organized by Alain Samson. Doing a little more research, I see that many of the previous guides are available on Alain Samson's ResearchGate page.  The 2015 Behavioral Economics Guide has an introduction by Dan Ariely.

Super cool, excited to read more. 

Skimming the content, I notice a section on the organization of behavioral insight units. I've always tossed around the idea of trying to informally start a "Nudge Unit" at FSC: something where faculty and students can collaborate to evaluate various nudges that could improve life at Farmingdale. Organizing such an activity would not be trivial, so I'm glad for the guidance that this guide provides.