What preferences are revealed by market designs? by Oğuzhan Çelebi (who is on the job market)

Oğuzhan Çelebi is on the job market this year (listed on the job market websites of both MIT and Stanford where he's finishing a two year postdoc). He has a new paper that takes a really novel approach to market design.  It looks at the implicit preference relation (if one exists) that a mechanism reveals when a revealed preference analysis is done of the choices it makes from different possible menus of alternatives. The specific focus is on preferences for diversity revealed by mechanisms used in connection with affirmative action.

Diversity Preferences and Affirmative Action, by Oğuzhan Çelebi 

Abstract: "In various contexts, institutions allocate resources using rules that determine selections given the set of candidates. Many of these rules feature affirmative action, accounting for both identity and (match) quality of individuals. This paper studies the relationship between these rules and the preferences underlying them. I map the standard setting of market design to the revealed preference framework, interpreting choice rules as observed choices made across different situations. I provide a condition that characterizes when a rule can be rationalized by preferences based on identities and qualities. I apply tests based on this condition to evaluate real-world mechanisms, including India’s main affirmative action policy for allocating government jobs, and find that it cannot be rationalized. When identities are multidimensional, I show that non-intersectional views of diversity can be exploited by dominant groups to increase their representation and cause the choice rules to violate the substitutes condition, a key requirement for the use of stable matching mechanisms. I also characterize rules that can be rationalized by preferences separable in diversity and quality, demonstrating that they lead to a unique selection within the broader set of policies that reserve places based on individuals’ identities."

########

He's a target of opportunity for any department interested in market design.

Should you get a Ph.D.?

 One of the emails I answered yesterday asked 

"As I get closer to graduation, I’m really curious about how people balance academic research with practical applications.  ... how do you see the future of combining theory with real-world impact? And what advice would you give someone like me who’s trying to decide between diving deeper into academic research or jumping into industry?

Here is how I replied:

Hi: academia versus industry isn’t a once and for all decision, but going for a PhD is, and it’s definitely worth thinking about.

Getting a PhD is pretty much essential for entry into an academic career, but is at least somewhat optional for most other careers (although sometimes particular specialties in computer science or biotech require a PhD even for industrial jobs).

So…the downsides of getting a PhD are pretty clear.  It’s hard, not always fun, it takes a long time, and doesn’t come with any guarantees.

The upsides are more subtle: it can be fun, and it opens up new ways of thinking, and things to think about.  But the big reward, if it is a reward, is the entry into academia.  And the big reward of academia is the ability to schedule your own mind—to be able to think about what you want, as long as you want.

If that doesn’t sound like a huge prize, then there are lots of rewarding things to do outside of academia (and many of them pay better).  The world is full of interesting, important, hard problems, and many of those are most actionable outside of universities, in industry or non-profits, or governments and NGOs. 

But the world has its own schedule and priorities, you can’t always choose what to think about, and you often have to come to a conclusion and move on before you’re ready…

So, for me universities have been great.  I’ve been able to engage with the world too.  And I couldn’t have gotten the job without a PhD (although I’m an economist whose PhD wasn’t in Economics).

I can’t guess what’s right for you, but those are some things to think about.

############

If the question had touched more on teaching, I could have added that a Ph.D (or other terminal degree) is  largely needed for teaching in research universities and liberal arts colleges, where you can teach the most advanced students. But there are obviously lots of teaching opportunities that don't require a Ph.D., and ways to be a mentor that aren't formal "teaching," so some of the above comments apply there too.

Pediatric Heart Transplants: rethinking the waitlist, by Power, Sweat...Almond et al.

 Here's a paper on the design of the waitlist for pediatric heart transplants.  It's accompanied by an editorial in the journal, and a discussion at Stanford Medical School.

Here's the article

Alyssa Power, MD,a,* Kurt R. Sweat, PHD,b,* Alvin Roth, PHD,b John C. Dykes, MD,a Beth Kaufman, MD,a Michael Ma, MD,c Sharon Chen, MD, MPH,a Seth A. Hollander, MD,a Elizabeth Profita, MD,a David N. Rosenthal, MD,aLynsey Barkoff, NP,a Chiu-Yu Chen, MD, PHD,a Ryan R. Davies, MD,d Christopher S. Almond, MD, MPH, Contemporary Pediatric Heart Transplant Waitlist Mortality  Journal of the American College of Cardiology, Volume 84, Issue 7, 13 August 2024, Pages 620-632

ABSTRACT

BACKGROUND In 2016, the United Network for Organ Sharing revised its pediatric heart transplant (HT) allocation policy.

OBJECTIVES This study sought to determine whether the 2016 revisions are associated with reduced waitlist mortality and capture patient-specific risks.

METHODS Children listed for HT from 1999 to 2023 were identified using Organ Procurement and Transplantation Network data and grouped into 3 eras (era 1: 1999-2006; era 2: 2006-2016; era 3: 2016-2023) based on when the United Network for Organ Sharing implemented allocation changes. Fine-Gray competing risks modeling was used to identify factors associated with death or delisting for deterioration. Fixed-effects analysis was used to determine whether allocation changes were associated with mortality.

RESULTS Waitlist mortality declined 8 percentage points (PP) across eras (21%, 17%, and 13%, respectively; P < 0.01). At listing, era 3 children were less sick than era 1 children, with 6 PP less ECMO use (P < 0.01), 11 PP less ventilator use (P < 0.01), and 1 PP less dialysis use (P < 0.01). Ventricular assist device (VAD) use was 13 PP higher, and VAD mortality decreased 9 PP (P < 0.01). Non-White mortality declined 10 PP (P < 0.01). ABO-incompatible listings increased 27 PP, and blood group O infant mortality decreased 13 PP (P < 0.01). In multivariable analyses, the 2016 revisions were not associated with lower waitlist mortality, whereas VAD use (in era 3), ABO-incompatible transplant, improved patient selection, and narrowing racial disparities were. Match-run analyses demonstrated poor correlation between individual waitlist mortality risk and the match-run order.

CONCLUSIONS The 2016 allocation revisions were not independently associated with the decline in pediatric HT waitlist mortality. The 3-tier classification system fails to adequately capture patient-specific risks. A more flexible allocation system that accurately reflects patient-specific risks and considers transplant benefit is urgently needed. 

 ###########

Here's the accompanying editorial in JACC

Getting to Transplant Should Not Be the Goal, by David L.S. Morales MD and Benjamin S. Mantell MD, PhD

And here's the Stanford article:

Heart transplant list doesn’t rank kids by medical need, Stanford Medicine-led study finds. More babies and children survive the wait for a heart transplant than in the past, but improvements are due to better medical care, not changes to wait-list rules, a new study finds. August 5, 2024 - By Erin Digitale

“The current system is not doing a good job of capturing medical urgency, which is one of its explicit goals,” said the study’s co-lead author, economist Kurt Sweat, PhD, who conducted the research as a graduate student in economics at Stanford University. "

#########

Kurt's job market paper was on heart transplants for adult patients.

Friday, April 26, 2024

Kurt Sweat defends his dissertation

Update: here's another commentary on the article:

Fewer Kids Now Die While Awaiting Heart Transplant, but There’s Room for Improvement Twenty years of data show mortality has dropped. Still, with one in eight children dying on the wait list, more needs to be done.  By Yael L. Maxwell

Redesigning academia

 Here's an article that presents for discussion possible market design changes in academic earth sciences.  Some of them don't apply immediately to Economics (e.g. we already admit grad students to departments without assigning them to specific advisors and grants), but others refer to much broader practices.

Kemeny, P.C., Phillips, A.A. and Johnson, D.L., 2024. Replaying the tape of academia: Fourteen alternative practices for the physical sciences. Perspectives of Earth and Space Scientists, 5(1), p.e2024CN000240. 

Kurt Sweat and Vincent Jappah graduate

 Kurt Sweat is a new PhD economist (and if you look closely you can barely make out how expertly his doctoral hood was given and received).  Vincent Jappah is a new Masters in Econ, heading towards a PhD in Health Policy.

Three Hats

Earlier:

Friday, April 26, 2024

Kurt Sweat defends his dissertation

Gonzalo Arrieta defends his dissertation

Gonzalo Arrieta defended his dissertation last week.

Here's his job market paper:

Procedural Decision-Making In The Face OfComplexity by Gonzalo Arrieta and Kirby Nielsen

Abstract: A large body of work documents that complexity affects individuals’ choices, but the literature has remained mostly agnostic about why. We provide direct evidence that individuals use different choice processes for complex and simple decisions. We hypothesize that individuals resort to “procedures”—cognitively simpler choice processes that we characterize as being easier to describe to another person—as the complexity of the decision environment increases. We test our hypothesis using two experiments, one with choices over lotteries and one with choices over charities. We exogenously vary the complexity of the decision environment and measure the describability of choice processes by how well another individual can replicate the decision-maker’s choices given the decision-maker’s description of how they chose. We find strong support for our hypothesis: Both of our experiments show that individuals’ choice processes are more describable in complex choice environments, which we interpret as evidence that decision-making becomes more procedural as complexity increases. We show that procedural decision-makers choose more consistently and exhibit fewer dominance violations, though we remain agnostic about the causal effect of procedures on decision quality. Additional secondary evidence suggests that procedural decision-making is a choice simplification that reduces the cognitive costs of decision-making."

##########

Another of his papers is a really creative investigation of human welfare:

What You Don't Know May Hurt You: A Revealed Preferences Approach by Gonzalo Arrieta and Lukas Bolte

Abstract: The dominant approach to welfare is based on revealed preferences and thus is restricted to settings where the individual knows their preferences have been fulfilled. We use a choosing-for-others framework to experimentally study welfare when what the individual believes to be true differs from what is actually true. About 42% of participants see welfare as independent of beliefs; 22% see welfare as only depending on beliefs; and 29% see a lower, but still positive, welfare effect when beliefs are fixed. Furthermore, the average participant values accurate beliefs. Our results suggest most people support the idea that welfare goes beyond awareness, which can inform media regulation, informational policies, and government communication.

 

Here's a figure from the instructions about the creation of "real" and "fake" inscribed copies of books. A third party judged the welfare to the recipient

"Our altruistically revealed preference paradigm consists of asking surrogate participants to trade off a monetary bonus given to the Receiver, and the Receiver getting the books with the original notes over those with the fake notes. The bonus amount is a surprise to the Receiver to minimize concerns that they use it to deduce which books they got (i.e., to maintain obliviousness). Our three requirements allow us to interpret the bonus amount that leaves participants indifferent between giving the original and fake notes as a measure of the change in the Receiver’s welfare. As a benchmark, we also elicit the welfare effect when the Receiver does learn which notes they get."

********

Welcome to the club, Gonzalo.

Talks to doctors and students in Italy

 I just returned from a busy trip to Italy, that began with a talk to physicians in Rome about kidney exchange, after which I moved on to talk to students in Padua.

Both were fun in different ways.  Here are some of the students gathered for a spritz after one of my talks in Padua.

Kurt Sweat defends his dissertation

Kurt Sweat defended his dissertation this week. As you can see in the picture below, we celebrated with the traditional toast, but instead of champagne we're drinking  In-N-Out milk shakes.

 

He and I are flanked by the other members of his committee: Itai Ashlagi, Paolo Somaini, Frank Wolak and Han Hong.  He was also advised by Dr. Chris Almond.

  His job market paper, and the other papers in his dissertation, all concern heart transplants.

Endogenous Priority in Centralized Matching Markets: The Design of the Heart Transplant Waitlist (Job Market Paper) 

[Link to current draft]

"Centralized matching markets that prioritize specific participants to achieve certain policy goals are common in practice, but priority is often assigned using endogenous characteristics of participants. In the heart transplant waitlist in the United States, the treatment that a patient receives is used to assign waitlist priority. Policymakers recently changed the prioritization in an attempt to reduce waitlist mortality by assigning higher priority to patients receiving specific treatments previously associated with high waitlist mortality. First, I document a significant response to waitlist incentives in treatments given and transplants that take place. Then, I develop and estimate a structural model of treatment and transplant choices to evaluate the effect of the policy change on patients' outcomes and doctors' decisions. I find three main results from my model. First, there is little change in aggregate survival, and the effect of the change has been mainly redistributive. Second, the change has effectively targeted patients with lower untransplanted survival, with these patients receiving higher expected survival under the current design. Third, the effect on survival is largely driven by changes in the decision to accept/decline offers for transplants rather than directly due to a change in treatment decisions. The policy implications suggest that future designs of the waitlist should disincentivize declining offers for transplants."

Welcome to the club, Kurt.

Ariel Pakes celebrated at the Nemmers Prize conference

 Ariel Pakes could certainly also win the prize for being loved by his students and colleagues: 

CONFERENCE IN HONOR OF ARIEL PAKES

Computer science award

A computer scientist whose work I follow has won an award that reflects on both his teachers and students.

Aaron Roth receives 2023 CyLab Distinguished Alumni Award 

"Aaron Roth, the Henry Salvatori Professor of Computer Science and Cognitive Science at the University of Pennsylvania, has been named CyLab's 2023 Distinguished Alumni Award winner.

...

"Roth earned his Ph.D. in Computer Science from Carnegie Mellon University in 2010, where he was advised by former CMU Professor Avrim Blum. His dissertation, 'New Algorithms for Preserving Differential Privacy,' gave new methods for performing computations on private data.

"Nominated by his former advisee, now Assistant Professor in CMU's School of Computer Science, Steven Wu, the award recognizes Roth's excellence in algorithms and machine learning, leadership in the field, and commitment to his students.

"As my advisor, Aaron is nothing less than a beacon of inspiration, marked by his relentless curiosity, exceptional instinct for identifying the most exciting questions, creative problem-solving acumen, and impeccable eloquence in communication," said Wu.

"Advising is one of the best parts of my job," said Roth. "Being recognized by one of my former students at the University where I earned my Ph.D. is really special."

Paying it forward

 Scott Cunningham, an economist who devotes a lot of his efforts to providing public goods, recently had a post on the phrase "paying it forward." He writes that he connected it with a movie with a similar name, but has recently come to view it differently (for reasons I find too embarrassing to quote, but related to the fact that I use the phrase now and then.)

Wikipedia says "Pay it forward is an expression for describing the beneficiary of a good deed repaying the kindness to others instead of to the original benefactor."  It goes on to say "Robert Heinlein's 1951 novel Between Planets helped popularize the phrase."  I could have first seen it there, as I read much of Heinlein's science fiction when I was a boy.

My associations with the phrase now mostly come from the motivations and actions of some living kidney donors, particularly in kidney exchange chains.

The phrase is certainly is evocative of what we do so much of in academia (when we're doing academia well): it describes the relationship between studying and teaching, and between teachers and students.

********

Scott's post announced that, as part of paying things forward, he's funding a prize for young economists.

Stanford graduation--Alex Chan, Ph.D.

 Congratulations Dr. Chan.

Welcome to the club, Alex.

Alex Chan makes his 'final presentation' at Stanford.

 

Note Alex's shoes

Dissertation advisors and job market outcomes by Rose and Shekhar

 Here's a paper on the economics job market, and the influence of dissertation advisers.

Adviser Connectedness and Placement Outcomes in the Economics Job Market, by Michael E. Rose and Suraj Shekhar, forthcoming in Labour Economics

Abstract: We study the role of social networks in the academic job market for graduate students of Economics. We find that the connectedness of a student’s advisor in the coauthor network significantly improves her job market outcome. We use two identification strategies and find that a) higher Eigenvector centrality of an adviser leads to her student getting placed at a better ranked institution, and b) larger distance between an adviser and an institution decreases the probability that her students are placed there. Our study sheds light on the importance of social connections in a labour market where information frictions regarding job openings are virtually absent.

...

"Our setting, the academic job market for Economists, is special in that information frictions regarding job openings are (almost) absent due to Job Openings for Economists. Thus, our finding that social networks play a role in this market is likely because they help decrease the uncertainty about an applicant’s quality."

********

I'm reminded of the timeless joke about how rabbits eat wolves: if you don't know it, there are many versions on the internet, here  (and this one comes with a bonus joke: Rabbit's Ph.D. Thesis and Lion's Watch Repair Business).

Alex Chan

 Congratulations, Alex.

Alex Chan

@alex8chan

·

Mar 24

I will join

@HarvardHBS

as an Assistant Professor next academic year! to the sacrifices my family made for me + their support… #HBS #FirstGen + my advisors who made this dream possible #AlRoth

@MurielNiederle

@LaurenceCBaker

@MichelleM_Mello

@I_am_NickBloom

@SIEPR

@Stanford

And earlier (in October)

Welcome to the club, Alex.

Erling Skancke defends his dissertation

 Erling Skancke defended his dissertation last week:

Here's his job market paper:

Welfare and Strategic Externalities in Matching Markets with Interviews (Job Market Paper) Recent debate in the medical literature has raised concerns about the pre-match interview process for residency and fellowship positions at hospitals. However, little is known about the economics of this decentralized process. In this paper, I build a game-theoretic model in which hospitals conduct costly interviews in order to learn their preferences over doctors. I show that increased interview activity by any hospital imposes an unambiguous negative welfare externality on all other hospitals. In equilibrium, both hospitals and doctors may be better off by a coordinated reduction in interview activity. The strategic externality is more subtle, and conditions are derived under which the game exhibits either strategic complementarities or substitutes. Moreover, an increase in market size may exacerbate the inefficiencies of the interview process, preventing agents from reaping the thick market benefits that would arise in the absence of the costly interviews. This effect increases participants' incentives to match outside of the centralized clearinghouse as markets become thicker, jeopardizing the long-term viability of the clearinghouse. The model also provides new insights into several market design interventions that have recently been proposed.

Congratulations, Erling! 

Welcome to the club.

David Zuckerman defends his dissertation

 David Zuckerman defended his dissertation yesterday.

Here are the papers he defended:

Preferences for Compensatory and Retributive Justice

I experimentally investigate third-party preferences for victim compensation and offender punishment when one party has harmed another. I find that if the harm is intentional, third parties not only display an increased demand for punishment, but also an increased demand for compensation. I refer to these additional demands for punishment and compensation as the demand for retributive justice and compensatory justice, respectively. Demand for retributive justice is positively correlated with demand for compensatory justice. However, third parties do not treat the two types of justice as substitutes or complements. Moreover, I generally find that these types of justices do not take victim knowledge of offender punishment nor offender knowledge of victim compensation into account. I then extend my investigation to a risky setting, where the offender's action may or may not end up harming the victim. I find that for both compensation and punishment, not only does the intent behind the action matter, but the consequences of the action as well. However, this is primarily driven by third-party distributional preferences; the outcome does not seem to affect the demand for retributive or compensatory justice.

A Theory of Chosen Preferences

(with B. Douglas Bernheim, Luca Braghieri, and Alejandro Martínez) [Online Appendix] [2019 Working Paper Version]

American Economic Review, Vol. 111, No. 2 (February 2021), pages 720-754

We propose and develop a theory of preference formation based on the idea that people evaluate their lives according to worldviews that provide accounts of success and failure, and that they choose those worldviews subject to feasibility constraints. Our framework highlights the role of mindset ï¬,exibility, a trait that determines the relative weights the decision maker places on her current and anticipated worldviews when evaluating future outcomes. We show that our theory generates rich behavioral dynamics, thereby illuminating a wide range of applications and providing potential accounts for a variety of observed phenomena.

Unseen Preferences: Homophily in Friendship Networks

Revise and Resubmit, Journal of Economic Behavior and Organization

Homophily is typically measured using a single dimension to define groups. However, people generally display preferences over multiple dimensions. I develop a simple model that characterizes agents both by a (discrete) "type" and a (continuous) "personality" value. Agents have preferences-for-similarity over both dimensions, but homophily is only measured with respect to type. I identify sufficient conditions on preferences such that a strongly stable matching exists, and show via simulations that the friendship patterns generated by these matchings qualitatively line up with several patterns noted in the data. The matchings can be calculated via an algorithm where agents "search" through utility space for friends. Increasing costs to friendship cause both within-group and systematic across-group heterogeneity in the extent of students' search through utility space. Majority-group "outliers" - those who must search an extensive amount in utility space for friends - form a disproportionate number of cross-group friendships. These outliers are the key reason why mid-sized groups exhibit the highest levels of homophily bias, a distinctive feature noted in prior literature.

Mazel tov, David.  Welcome to the club.

Süleyman Kerimov defends his dissertation

 Süleyman Kerimov defended his dissertation yesterday, in Stanford's MS&E department. He studies matching, and will teach at Rice next year.

His main advisors are both named Itai.

These are the papers he spoke about:

• Dynamic Matching: Characterizing and Achieving Constant Regret, with Itai Ashlagi and Itai Gurvich, . [pdf] [SSRN]


• Scrip Systems with Minimal Availability, with Itai Ashlagi, working paper.
  • Appeared as an extended abstract in the 15th Conference on Web and Internet Economics (WINE 2019).

Mazel tov and tebrikler, Suleyman.

Welcome to the club.

70: three score years and ten

 If you persist, you find that 70 isn't nearly as old as you once imagined it to be (and  as it used to be, maybe not even that long ago). Certainly three score years and ten is no longer an inspirationally long life.

(The Hebrew verse 10 of Psalm 90 simply says "70 years." I think that "three score and ten" is a flourish originally introduced in the King James translation.

יְמֵי-שְׁנוֹתֵינוּ בָהֶם שִׁבְעִים שָׁנָה

"The days of our years are threescore years and ten, or even by reason of strength fourscore years...")

Earlier this month I was surprised at what I thought would be only a Hanukkah party, by a pre-birthday celebration in which a number of my students located a pretty big group of my former students and postdocs to produce this video of birthday greetings and reminisces. 

I've been lucky in my students and young colleagues, long ago and still today.  Thanks to Alex and the local gang, to all those on the video, and to all those who weren't found by the surprise team, who are all remembered fondly and gratefully...

Alejandro Martínez-Marquina defends his dissertation

 Alejandro Martínez-Marquina defended his dissertation this week.

 The three papers he chose for his dissertation are these:

When a Town Wins the Lottery: Evidence from Spain

(with Christina Kent) [Slides] [Draft]

"How do local wealth shocks impact economic activity? For over two centuries, Spain has conducted a national lottery which often results in the random allocation of up to $800 million in cash to the citizens of one town. This is the only case in the world where individuals living in the same location randomly receive pure wealth shocks of this scale. Leveraging data on town-level lottery ticket expenditures, we compare winning towns to non-winning towns that had the same probability of winning. We find that although consumption increases, the lottery causes a slowdown in economic activity and deters new migration to towns that won in recent decades. However, an analysis of a century of lottery winners reveals large and persistent increases in population for towns that won in earlier periods."

The Burden of Household Debt

(with Mike Shi)

"We propose that holding debt causes worse financial decisions using a novel experimental design where we randomly assign debt. Our design isolates the consequences of holding debt while controlling for potential confounding factors such as initial wealth levels, selection, risk, and time preferences. Our findings show that debt causes behavioral biases detrimental to subjects' financial payoffs. However, subjects' strategies are not random but instead debt-biased, consistent with an additional penalty for holding negative balances. We refer to the financial losses caused by debt as the Burden of Debt and provide evidence that, under certain circumstances, these behavioral biases can compound and lead to substantial losses. Furthermore, we show in additional treatments how these debt-biased behaviors can also deter subjects from borrowing and forego profitable opportunities."

Ingraining Traditional Gender Roles in the Classroom: Evidence from the Spanish Social Service

[Slides]

"This study uses a regression discontinuity framework to examine the long- run effects of conservative education on women's' family and labor decisions. In 1939, the Spanish dictatorship created the Social service, a compulsory 6- month training program aimed at relegating women to the roles of mothers and housewives. We exploit the discontinuity induced by the sudden abolition of the Social Service, in addition to variation in the age of enrollment, to examine the consequences of attending the program. Using historical enrollment records and the universe of birth certificates, we find the Social Service was successful in instilling the regime's ideology. Women exposed to the class get married and have kids at younger ages, consistent with the desire to form a family sooner. In addition, they are more likely to declare being housewives when their first child is born. Future work will explore the underlying mechanisms and the effects on children by surveying women who enrolled around abolition."

Welcome to the club, Alejandro.

Aleandro ( top center) with Chenzi Xu, Muriel Niederle, Doug Bernheim, Al Roth, Ran Abramitzky