Wednesday, November 30, 2011

School Choice and Education Reform at Brookings

The Brookings Institute hosts a session on School Choice and Education Reform, featuring Joel Klein, who was chancellor of New York City schools when the high school choice plan there was revamped.

"Large numbers of parents choose where their children are educated by moving to a school district or neighborhood that gives them access to good public schools, but school selection through residential choice is not an option for parents who are poor or unable to relocate. These parents are forced to take whatever is available to them through their local school district, and the schools that serve them do not have to worry about competition. While some districts are satisfied with this status quo, others have embraced policies that make school choice widely available and expose schools to the consequences of their popularity.

"On November 30, the Brown Center on Education Policy at Brookings will host a discussion exploring the critical role of school choice in the future of education reform. Senior Fellow and Brown Center Director Russ Whitehurst will preview the Education Choice and Competition Index – an interactive web application that will score large school districts based on thirteen categories of policy and practice – and announce the Index’s initial rankings of the 25 largest school districts in America. 

"Following his remarks, Joel Klein, the executive vice president of News Corporation and the former New York City Schools chancellor, will deliver a keynote address offering his reflections on the successes and challenges surrounding the expansion of public school choice in New York City"

The Brookings Institute has also published a companion website and report: The Education Choice and Competition Index: Background and Results 2011. The report contains a ranking of school choice plans around the country.

One of the criteria is the "Assignment Mechanism
Our framework places considerable emphasis on the processes by which students are assigned to schools, treating it as a major category for evaluating choice and competition.  The antithesis of choice is an assignment mechanism based on residence, with little or no chance of parents being able to enroll their child in a school other than the one in their neighborhood. In contrast, the paragon of assignment systems is one in which students are assigned to schools through an application process in which parents express their preferences and those preferences are maximized.  We score districts based on where they stand with respect to these two poles. "
The report is here: The Education Choice and Competition Index: Background and Results 2011.

And it's conclusion (drumroll....) is

"The high score overall goes to New York City, with Chicago in second place.
Both received letter grades of B. The low score goes to Orange County, Florida,
which received a grade of  D. New York  performed  particularly  well in  its
assignment mechanism,  its provision of relevant performance data,  and  its
policies and practices for restructuring or closing unpopular schools.  Chicago, in
contrast to New York, has more alternative schools, a greater proportion of
school funding that is student-based, and superior web-based information and
displays to support school choice.  If the best characteristics of Chicago were
transferred to New York and vice versa, both would receive letter grades of A."

Tuesday, November 29, 2011

Early admissions and early decision

A number of stories follow this year's trends in early admissions (with Harvard back in the early game).

Early applications on rise at colleges
"High school seniors hoping for an advance nod from Harvard University have swamped it with an unusually large group of early applications that represent the most economically and ethnically diverse set of students in the school’s history.

"Harvard canceled its early action program in 2006 because of worries that privileged applicants were getting an edge and holding back attempts to recruit a more diverse student population, a fear backed up by studies showing that early deadlines tend to draw a whiter, richer applicant pool than conventional winter and spring cutoff dates do.

"But the school reinstated the program this year and announced yesterday that it drew 4,245 applicants, an increase from the 4,010 who applied in 2006. Almost 72 percent of this year’s applicants need financial aid, and numbers of African-American, Latino, and Native American students are all up."
see also

The Flock of Early Birds Keeps Growing
"In the end, the story of early-action in 2011 is not really about what happens at the most-selective colleges (they will be just fine, by any measure). The story’s really about how, in the wild ecosystem of enrollment, one institution’s actions and policies will affect another—and what that will mean for individual students down the line.

"At Georgetown, Mr. Deacon suspects that Harvard and Princeton siphoned many early applicants out of his institution’s pool this year. So he expects that Georgetown’s yield for early-action applications, which dropped from 60 percent to 46 percent back in 2007, will go back up to where it used to be, which would be good news for the university."

Harvard College Receives 4,245 Early Applications
"“Early admission programs tend to advantage the advantaged,” then-Interim University President Derek C. Bok said in a statement in 2006. “Students from more sophisticated backgrounds and affluent high schools often apply early to increase their chances of admission, while minority students and students from rural areas, other countries, and high schools with fewer resources miss out.”
Harvard and Princeton retreated from that stance earlier this year, with each announcing the return of early admissions within hours of each other. Though Harvard administrators had hoped other colleges and universities would follow suit in eliminating early admission, that trend never materialized.
In the February announcement of early admission's return, Fitzsimmons argued that the circumstances had changed and that a broader group of students sought to apply early."

This story in the NY Times Choice blog starts with a good table listing schools with binding early decision (including some universities with two rounds of binding early decision), non-binding early action, and restrictive/single-choice early action. (They even have a printer-friendly version of their chart.)

Monday, November 28, 2011

Quick links to economics job market candidates, from the NBER

If you are hiring new Ph.D. economists this year, it's time to begin inviting them to interviews at the ASSA meetings in January. If you haven't started doing your homework yet, this big list of candidates from the NBER might help. It lists Ph.D. Candidates in Economics, in alphabetical order from Alabama (1) to Yale (18) (where the numbers in parenthesis are the number of candidates, by my count).

(Some of the NBER's links seem to be from last year, here's Penn State's list for this year (7).)

Sampling from the NBER's big list yields some big producers of new Ph.D.s (and I have surely missed some). But see the ones below, with ten or more new Ph.D.s on the market.
I wouldn't have guessed this order.

Harvard (31), of whom I'm particularly fond of these 5);
NYU (25);
Northwestern (25);
Cornell (24)
MIT (23);
UCLA (21);
UC Davis (20)
Maryland (20):
OSU (20)
Chicago (19);
Columbia (19);
Wisconsin (19);
LSE (18);
Michigan (18);
Yale (18);
Princeton (16);
Stanford (16);
UCSD (15):
BU (14);
Berkeley (13);
Brown (13)
Duke (12);
Penn (12);
Illinois (10);
Minnesota (10);
Rochester (10):
Cal Tech (9) [I know, 9 isn't double digits...]

Sunday, November 27, 2011

Matching in Practice conference: 28 November, Budapest

Peter Biro writes:
Hi Al,
the European research network on Matching in Practice will have its third workshop on 28 November, this time in Budapest. You can find the program and other details at
Third Workshop on Matching in Practice

Third Workshop on Matching in Practice

Following the first two workshops in Berlin and Brussels, the Matching in Practice network will meet in our institute on 28 November 2011.

Organisers Péter Biró (IEHAS), Estelle Cantillon (ULB, Brussels) and Dorothea Kübler (WZB, Berlin)

10:00-10:15 welcome and presentation
10:15-11:00 Tamás Fleiner and Naoyuki Kamiyama: A Matroid Approach to Stable Matchings with Lower Quotas
11:00-11:30 coffee break
11:30-12:15 Sebastian Braun, Nadja Dwenger, Dorothea Kübler and Alexander Westkamp : Implementing quotas in university admissions: An experimental analysis
12:15-13:00 lunch
13:00-13:30 discussion on the network
13:30-13:50 László Kóczy and Zsolt Szelíd: Resident matching in Hungary
13:50-14:35 Caterina Calsamiglia and Antonio Miralles: All about priorities: No school choice under the presence of bad schools
14:35-15:00 coffee break
15:00-15:45 John Kennes, Daniel Monte and Norovsambuu Tumennasan: The Daycare Assignment Problem
15:45-16:15 coffee break
16:15-17:00 Heinrich Nax and Peyton Young: The Evolution of Core Stability in Decentralised Matching Markets

Saturday, November 26, 2011

Gift registries for everything

Are people starting to think what gifts to buy you?

US Airways sends me the following email:

Don’t just wish for the miles you want…Ask for them!
Create your Dividend Miles Gift Registry today.

Dreaming of a trip but don't have enough miles? Create your Dividend Miles Gift Registry so your friends and family can help you. Whether it’s a special event like your honeymoon or birthday, or just the vacation you’ve wanted to take – ask and you shall receive.

Here’s how:
  • Create your Gift Registry
  • Invite friends and family to gift or share miles
  • Track your progress
  • Book your award trip!
  • Miles Gift Registry

Friday, November 25, 2011

Liver transplants for alcoholics?

Here's hoping you didn't drink too much yesterday: Study stirs debate over transplants for alcoholics

"Some gravely ill alcoholics who need a liver transplant shouldn't have to prove they can stay sober for six months to get one, doctors say in a study that could intensify the debate over whether those who destroy their organs by drinking deserve new ones.

"In the small French study, the vast majority of the patients who got a liver without the wait stopped drinking after their surgery and were sober years later. The study involved patients who were suffering from alcohol-related hepatitis so severe that they were unlikely to survive a six-month delay.

"The findings, reported in Thursday's New England Journal of Medicine, could boost demand for livers, already in scarce supply, and reopen a bitter dispute over whether alcoholics should even get transplants.

"The controversy peaked in the 1990s when celebrities with drinking problems Larry Hagman, David Crosby and Mickey Mantle got liver transplants. More recently, British soccer star George Best received a new liver in 2002, started drinking again and died three years later.

"Alcohol can cause lethal, liver-destroying diseases such as cirrhosis and hepatitis. Nearly one in five liver transplants in the U.S. go to current or former heavy drinkers. Transplant hospitals commonly require patients waiting for a new liver to give up drinking for six months as a way of assuring doctors they are serious about staying sober after the operation.

"Drinkers severely ill with hepatitis account for a very small share of patients needing transplants. The French study suggests that dropping the six-month rule for these patients would increase demand for livers by only about 3 percent.

"The study's lead author, Dr. Philippe Mathurin of Huriez Hospital in Lille, France, said a strict application of the six-month rule may be unfair to such patients. He said they are just as deserving as other liver patients, many of whom have diseases caused by poor lifestyle choices such as drug use or obesity.

"Mathurin said he favors keeping the rule for other alcoholics with liver disease, noting that some can recover liver function simply by staying sober.

"Dr. Robert S. Brown Jr., transplant director of New York-Presbyterian Hospital/Columbia University Medical Center, agreed it is time to rethink the six-month rule. "The challenge of this paper is to come up with better ways, both to treat alcoholism as a disease and to predict who will succeed with transplantation," he said.

"Mathurin acknowledged that such a change could put more patients on the waiting list for organs, and said: "It means we have to increase the number of donors."

Thursday, November 24, 2011

Raising turkeys for Thanksgiving

Happy Thanksgiving y'all:)

Here's timely advice from North Carolina Cooperative Extension if you're thinking of starting a small turkey operation for next year: Raising Standard Turkeys for the Holiday Market.

The first two points sound like the voice of experience:

"a. Identify processing plant before obtaining poults
"b. Identify feed source before obtaining poults"

And here's a report about a somewhat larger operation: In This Town, Turkey Picks Up Bill for Dinner, concluding with this:
"“You look at it every day, and you get to where you don’t really care for turkey,” Mrs. Farmer said. “That’s why I get a ham.”"

Wednesday, November 23, 2011

College admissions and searchable data

A slightly breathless but interesting story in The Washington Monthly about college admissions, and how it may be transformed by data mining (and focused on the company Connectedu): "The End of College Admissions As We Know It: Everything you’ve heard about getting in is about to go out the window."

"Because the information that might help them is entombed in file folders, colleges resort to an expensive, inefficient, scattershot strategy. A typical midrange private college looking to enroll 1,200 freshmen might buy a list of 350,000 names from the College Board. An expensive but poorly targeted direct-mail campaign leads to 11,000 applications. They accept 5,000, of whom only 1,200 choose to enroll. Of those, more than half drop out or transfer, leaving the college struggling to bring in enough tuition revenue to pay their bills and left with no option other than buying another 350,000 names.

"Students have a similar problem. It’s hard to choose the right college. Higher education is what economists call an “experiential good,” something you can’t fully understand until after you purchase and experience it. As parents of college age children know, students often assemble a list of prospective schools through a frighteningly arbitrary process of hearsay, peer misinformation, and fleeting impressions gained during slickly produced college tours. Or, worse, they don’t assemble a prospective list at all and default to inexpensive, nearby institutions. Some of those local colleges are terrible places to go to school. (See “College Dropout Factories,” September/October 2010.) Too many students don’t find out until it’s too late.
"There are other players in this market, including the Common Application and a company called Naviance, which offers electronic college planning tools for high school students. The virtue of ConnectEDU, though, is that it spans the entire process, from late middle school into college and beyond. The company’s first foray into the market came in 2006, when it signed up three colleges and fifteen high schools. In 2007, it was up to thirty-five high schools and 300 colleges. It began signing up school districts instead of individual schools, then moved to contracts with entire states, starting with Michigan. The number of high schools increased to 700 in 2008, 1,700 in 2009, and 2,500 in 2010. That amounts to about 2.5 million students. The Miami-Dade County school system joined the network last year. The state of Hawaii signed up in May 2011.

"The number of colleges using the service has also increased, to 450, representing a decent—though not quite commanding—subset of the schools that receive large numbers of applications. Yale signed up in 2008."

HT: Stephanie Hurder

Tuesday, November 22, 2011

Surrogacy as a business

Excerpts from The Red Market: Cash on Delivery

"India legalized surrogacy in 2002 as part of a larger effort to promote medical tourism. Since 1991, when the country’s new procapitalist policies took effect, private money has flowed in and fueled construction of world-class hospitals that cater to foreigners. Surrogacy tourism has grown steadily here as word has gotten out that babies can be incubated at a low price and without government red tape. Patel’s clinic charges between $15,000 and $20,000 for the entire process, from in vitro fertilization to delivery, whereas in the handful of American states that allow paid surrogacy, bringing a child to term can cost between $50,000 and $100,000, and is rarely covered by insurance.
"Dependable numbers are hard to come by, but at minimum, Indian surrogacy services now attract hundreds of Western clients each year. Since 2004 Akanksha alone has ushered at least 232 babies into the world through surrogates. By 2008 it had forty-five surrogates on the payroll, and Patel reports that at least three women approach her clinic every day hoping to become one. There are at least another 350 fertility clinics around India, although it’s difficult to say how many offer surrogacy services, since the government doesn’t track the industry. Mumbai’s Hiranandani Hospi- tal, which boasts a sizable surrogacy program of its own, trains outside fertility doctors on how to identify and recruit promising candidates. A page on its website advertises franchising opportunities to entrepreneurial fertility specialists around India who might want to set up surrogacy operations with an endorsement from Mumbai. India’s Council on Medical Research (which plays an FDA-like role—except that it has far less power to actually enforce its edicts) predicts that medical tourism, including surrogacy, could generate $2.3 billion in annual revenue by 2012. “Surrogacy is the new adoption,” says Delhi fertility doctor Anoop Gupta.

"Despite the growth projections, surrogacy is not officially regulated in India. There are no binding legal standards for treatment of surrogates, nor does state or national authority have the power to police the industry. While clinics like Akan- ksha have a financial incentive to ensure the health of the fetus, there is nothing to prevent them from cutting costs by scrimping on surrogate pay and follow-up care, or to ensure they behave responsibly when something goes wrong.

Monday, November 21, 2011

Matching and market design session at SAET in Australia in July

Fuhito Kojima writes:
Call for submissions: I am organizing a session on matching and market design in SAET 2012:
which will be held in Queensland, Australia, from June 30th to July 3rd.
Al is scheduled to give a plenary talk (Sir John Hicks Lecture) and Tayfun already offered to come to the session, and it will be an interesting conference for us!
So I would like to encourage your submission:
If you are interested, you please send a message directly to me by December 25th this year with the paper draft (if not yet available, please send me the abstract and/or slide if available).
I will get back to you about the decision by January 15th.
Hope that I can pique your interest!

Thanks for reading this. A happy thanksgiving.


Fuhito Kojima

Sunday, November 20, 2011

Schools' choices and school choice in England

In the U.S., charter schools are typically not allowed to be choosy: they often have to offer admission strictly by lottery. In England, there's controversy over religious schools, and the following story illustrates some of the issues.

Catholic school in new row over school admissions

"Coloma Convent Girls' School in Croydon was reported to the official admissions watchdog by the local diocese amid claims its entry rules are “discriminatory”.
The over-subscribed school – which is rated outstanding by Ofsted and regularly appears towards the top of league tables – gives more “points” to families who take part in parish activities.
But the Catholic Archdiocese of Southwark complained that the system discriminated against single parents who were unable to find the time to take part in parish work.
It also said it was unfair on immigrants who did not share the same “tradition of community service” and struggled to provide written evidence of volunteering because English was not their first language.
The move comes after Diocese of Westminster shopped the hugely popular Cardinal Vaughan Memorial School to the admissions watchdog, complaining that its entry rules were too elitist and effectively penalised the less devout.

Saturday, November 19, 2011

School choice in New Orleans

New Orleans has some ambitious plans for a unified school choice plan in a city whose schools will include many independent charter schools. Here's a ten minute video of an interview with Recovery School District Superintendent John White. Starting at minute 3 it is about school choice.

This is work in which IIPSC, the Institute for Innovation in Public School Choice, is lending support in market design and technology.

Here's an earlier story: Recovery School District chief plans central enrollment system, technical training, more

"The head of the state-run Recovery School District, which governs most of the city's public schools, issued a wide-ranging strategic plan Tuesday aimed at tackling the district's most chronic shortcomings.
"He is promising a central enrollment system by next year, so parents and guardians -- especially those looking to place children midyear -- will be able to ask the district to find them an open spot, rather than having to contact one school after another.
"Most, if not all, of the district's schools will eventually be charters...

"On the central enrollment system ... White provided new details, explaining that parents will be asked to list their top school choices on a common application. The district will be able to weigh factors like a school's distance from a child's home, and assign each student to a building by looking at available seats in both charters and direct-run campuses."

Friday, November 18, 2011

School Choice in Chicago

CPS Application Plan Moves Ahead

"The Chicago Board of Education approved a resolution Wednesday that would require all incoming 9th grade students to apply for entrance into the public school system next year and opens the door to an application process for all grades.
The resolution authorizes Chicago Public Schools officials to create a single application for the high school admissions process that could be introduced as early as the 2012-2013 school year and to develop a similar application for elementary school admissions.
Currently, students can enroll in their neighborhood school without applying. If they choose to go to a magnet, selective, charter or other school outside the neighborhood they live in, they must fill out separate applications.
It is not known if charter schools would be included on the application. District spokeswoman Marielle Sainvilus said the resolution is the first step in creating a common application and none of the specifics have been decided.
The resolution also calls for the district to contract with the Institute for Innovation in Public School Choice, a non-profit school choice consultancy, to develop a “school choice matching system.” District officials did not elaborate on what the system would entail.
The current school application process can be dizzying and has spurred the creation of small consulting businesses to help parents navigate the system.
Jackson Potter, staff coordinator for the Chicago Teachers Union, said Tuesday that the resolution raises questions about how the district will ensure that all students, especially those that are homeless or have unstable situations, are able to enroll in school.
The Institute for Innovation in Public School Choice has worked with other urban districts, such as New York and Boston, to overhaul their admissions processes. In New York, all 8th grade students are required to fill out an application and rank up to 12 schools they would like to attend. The district then sorts them based on their preferences. In Boston, all students submit an application, listing the schools they would like to attend. The district then assigns students based on their choice and a number of other considerations, such as how close they live to a school and if a sibling is already enrolled."

Thursday, November 17, 2011

School Choice in Denver

Denver is implementing a new school choice system. IIPSC, the Institute for Innovation in Public School Choice, is helping. Denver Public Schools will be the first large district to unite charter schools and district managed schools under one umbrella of centralized choice.

New, Unified DPS Enrollment Process Backed by Coalition of Education Advocates
DENVER – A coalition of 10 local education advocacy groups announced their support today for Denver Public Schools’ (DPS) new, unified enrollment process, SchoolChoice. 
The coalition worked closely with DPS and national enrollment experts to develop SchoolChoice, a unified enrollment process for the 2012-13 school year that is designed to simplify how families enroll students in schools. The group will continue to assist the district with the campaign to notify parents and community leaders as well as the implementation and review of the new process.
“Years ago, families identified significant drawbacks and hurdles to the way DPS enrolled students in schools across the district.  Many families were spending days or weeks trying to navigate multiple deadlines and driving across the city to find different forms. And many other families didn’t participate at all.  We all came together to find ways that the current process could be simplified and streamlined,” said Amy Slothower, Executive Director, Get Smart Schools.  
“We believe the new process will increase access to the district’s highest-performing schools and make enrollment much easier for students, their families and their schools,” Amy continued.
Prior to the introduction of SchoolChoice, charter and magnet schools each conducted their own enrollment.  When combined with the district’s own choice process for its traditional neighborhood schools, there were more than 60 different enrollment processes with different forms and due dates for public schools across Denver. The range of due dates, methods and applications made enrollment overwhelming for many parents and made planning for the school year challenging for schools.
SchoolChoice is designed to simplify school enrollment because every transitioning student (incoming K, 6th and 9th grades and any student who wants to choice into a different school) will prioritize their schools of interest on a single form with a single deadline. All transitioning students will then receive a school assignment, based on their preferences, during a single week in March.
Putting into place one process for all district schools and one form per grade will simplify what has been a confusing and often overwhelming system for students and families to navigate.
 “We’re excited because SchoolChoice will ensure a fair and equitable enrollment process for all DPS families,” Slothower continued. “With the help of our partners, the district and the Walton Foundation, we’re making it easy to connect students with the schools that best fit their needs.”
SchoolChoice is being funded by a combination of public and private funding in the form of grants and bonds, including a substantial grant from the Walton Foundation.
The full list of coalition members includes:
·        Get Smart Schools
·        The Walton Family Foundation
·        Denver Public Schools
·        A Plus Denver
·        Colorado League of Charter Schools
·        Colorado Succeeds
·        Donnell-Kay Foundation
·        Metro Organizations for People
·        Stand for Children Colorado
·        University of Colorado Denver

Update: see also

Wednesday, November 16, 2011

Congratulations to Muriel Niederle

If that's white smoke coming out of the faculty meeting at the College of Cardinal, it must mean that Muriel Niederle has just been promoted to full professor. An excellent day for market design, experimental economics, and Stanford. (Congratulations to L.E. and N.B. too:)

Tuesday, November 15, 2011

Boston school choice: waiting lists

The Boston Globe looks at waiting lists for Boston Public Schools. Some wait lists don't move until after the first week of classes, when the fact that some assigned students have left the school district can be verified.

CLASSES IN SESSION; THOUSANDS IN LIMBO: Parents frustrated as children languish on waiting lists

"School opened with almost 10,000 students - nearly 18 percent of the student body - still on waiting lists, trying to get into different schools than they were assigned. Some, like Mayes’s son, were held up because they came late to the process. Others applied on time but were disappointed by their assignments and hoping for better placements.

"Most would never get called. Those who did might wait days or weeks for an opening. Some might not be notified of vacancies until November, forcing families to make agonizing decisions about pulling children out of classrooms they have grown used to.

"Boston’s school lottery is a balancing act. Designed to give every family a chance at getting into a high-achieving school, the lottery lets parents request seats in schools outside their neighborhoods. The intent is to spread opportunity in a city with uneven schools and keep options open for parents, but the unintended consequence, too often, is disruption. Since school started in September, about 750 students moved off waiting lists and into different schools, leaving altered class lists and new vacancies to be filled behind them.

"Last-minute changes are inevitable in a city with a highly mobile population, where hundreds of students move during the summer, but Boston’s assignment system adds - and indeed fosters - additional layers of delay and uncertainty.

"Families who were asked to choose a school last winter or spring were never forced to commit to one. Students could show up - or not - in September. If they didn’t, the district left their seats open for eight school days before releasing the spots to wait-listed students, tying up thousands of seats for the first two weeks of school. The number of no-shows, eight days into this school year, was 2,810.
"Why does a school district that starts the assignment process so early not finish it before school starts? Administrators say they can’t start assignments until late in the summer because they have to wait for 3,000 students to finish summer school in August to find out who will actually be promoted and who will have to repeat a grade. The school district’s hotline, which fielded 15,000 calls in five weeks, only has a temporary staff of a dozen and only opened in late August.

When folks get back from their vacations and hotline staff starts making the phone calls, parents are home, they’ve made decisions, it works,’’ said Jerry Burrell, director of enrollment and planning and support. “Any earlier, it just doesn’t work.’’

Here is the Boston Public Schools student assignment policy for waitlists:

  • BPS will create wait lists for all schools where there are more applicants than available seats for a particular grade.
  • A student’s place on the wait list is based on the registration period when the student applied, sibling priority, the school choices selected on the application, and a random number.
  • No student will have a lower place on a wait list than any student who applied in a later registration round, regardless of priorities. However, within each period, a student’s place on the wait list can change if his/her priorities change, which may affect the placement of other students on the wait list. 
  • Families registering for any grade, K0 through 12, may be placed on up to three wait lists. Families who are assigned to their second choice school may be on the wait list for their first choice school. Families assigned to their third choice school may be placed on the wait lists for their first and second choice schools.  And families assigned to their fourth choice school or higher, or who are unassigned (kindergarten only) or administratively assigned (see below), may be on wait lists for their top three choices.
  • Families may request that a student be added to any wait list (to a school for which they are eligible to apply). However, students may not occupy more than the number of wait lists prescribed above. Students already on the maximum number of wait lists must go off one list in order to be added to another.
  • Any student who remains a Boston resident may remain on a wait list after the beginning of the school year, regardless of whether or not the student attends the Boston Public Schools.
  • All wait lists expire at the end of the second marking period (January of the following year).
Coming off a wait list
When seats become available, students will be assigned from wait lists in the following order, beginning with students who applied in the earliest rounds:
If the school has not reached its 50% walk zone target, students are assigned from wait lists in this order:
1. Students with sibling + walk zone priority
2. Students with sibling priority
3. Students with walk zone priority
4. Students with no priorities
If the school has reached its 50% walk zone target, students are assigned in this order:
1. Students with sibling priority (no additional priority for walk zone)
2. All other students (no walk zone)
The random numbers assigned to families during registration will be used as "tiebreakers" among students with the same priorities.
From mid-March through mid-August, as seats become available, children are automatically moved off the wait list into their chosen school. Families receive notification about their new school assignment with a letter sent in a mail.
For kindergarteners after mid-August, if a space becomes available at a school with a wait list, families on the wait lists are contacted in order. Families have 24 hours to decide if they want to attend the school. Families are contacted only at the phone numbers they listed on their registration form. This process continues into the school year through January as seats become available.
For students in grades 1 through 12, if a space becomes available at a school with a wait list, families on the wait list are contacted from mid-August through the end of September. After September, families are only contacted about transferring schools where they are wait-listed after marking periods (mid to late November and late January). Families have 24 hours to decide if they want to attend the school and families are contacted only at the phone numbers they have listed on their registration form.
All lists, regardless of grade, expire at the end of the second marking period (January of the following year)."

Monday, November 14, 2011

A supply and demand model for stable matchings, by Eduardo Azevedo and Jacob Leshno

Much of two-sided matching theory has been developed using discrete mathematics, which has led to beautiful theorems about lattices and stable matchings that are optimal for one side of the market or the other. Since the agents on each side of the market can be quite heterogeneous (which is why who is matched to whom is important), there is a lot of information in each stable matching, which can therefore be hard to summarize succinctly.

Now Eduardo Azevedo and Jacob Leshno have a very nice model of two sided matching in which one side of the market is a continuum, or can be approximated as such via a converging sequence of finite markets, while the other side is finite. Think college admissions, with lots of students and a finite set of colleges. In their model, there is a (generically) unique stable matching, and it can be characterized by a low dimensional set of price-like scalar thresholds, one for each college.

Modeling a continuum of students means that students have to be characterized in some sense, and they find a simple, general way to do this, that works well in the finite case also. A student is characterized by an arbitrary preference over colleges (there are only finitely many of these, since there's a fixed finite set of colleges) and a vector of scores, one for each college, which characterizes each college's preferences over students (each college ranks students in their order on the college's own score). They show that a stable matching becomes a very simple object--it is the solution to a set of supply and demand equations, in which each college is characterized by a threshold, and students are admitted to their most preferred college among those that score them above that college's threshold.

This simple characterization of stability considerably simplifies the investigation of comparative statics.They illustrate this with a price-theoretic analysis of how competition induces schools to improve quality, and why this may not benefit some types of student.

It looks like this will be quite a useful model.

Here's the paper and abstract:

A Supply and Demand Framework for Two-Sided Matching Markets by EDUARDO M. AZEVEDO AND JACOB D. LESHNO

Abstract: "We propose a new model of two-sided matching markets, which allows for complex heterogeneous preferences, but is more tractable than the standard model, yielding rich comparative statics and new results on large matching markets.

"We simplify the standard Gale and Shapley (1962) model in two ways. First, following Aumann (1962) we consider a setting where a finite number of agents on one side (colleges or firms) are matched to a continuum mass of agents on the other side (students or workers). Second, we show that, in both the discrete and continuum model, stable matchings have a very simple structure, with colleges accepting students ranked above a threshold, and students demanding their favorite college that will accept them. Moreover, stable matchings may be found by solving for thresholds that balance supply and demand for colleges. We give general conditions under which the continuum model admits a unique stable matching, in contrast to the standard discrete model. This stable matching varies continuously with the parameters of the model, and comparative statics may be derived as in competitive equilibrium theory, through the market clearing equations. Moreover, given a sequence of large discrete economies converging to a limit economy, the set of stable matchings of the discrete economies converges to the stable matching of the limit economy.

"We bound the rate of convergence of the set of stable matchings of large discrete economies to the continuum approximation, and show that comparative statics regarding the unique stable matching of the continuum model extend to strong set ordering of the sets of stable matchings of approximating discrete economies. We model the transferrable utility case, as in Becker (1973). We characterize the limit of school choice mechanisms used in practice, generalizing previous results of Che and Kojima (2010). Finally, we illustrate the model's applicability by quantifying how competition induced by school choice gives schools incentives to invest in quality. Specifically, we show that schools have muted, and possibly even negative incentives to invest in quality dimensions that benefit lower ranked students."

As I've mentioned in previous blog posts about some of their other papers (here about Eduardo and here about Jacob), they are both on the job market this year. Here are their other papers (Eduardo's; Jacob's). You could hire them both:)

Sunday, November 13, 2011

For Love or Money

Kim Krawiec writes
"The conference volume For Love Or Money: Defining Relationships in Law and Life has finally been published, yeah!  The introduction is here.  My three posts, done just before and after the conference and discussing the event are herehere, and here.  The table of contents is [here]."

Here are the posts on my blog that come up when I search for "Krawiec".

Saturday, November 12, 2011

Market design at INFORMS, Nov 13-15 in Charlotte NC

When I got my Ph.D. in Operations Research in 1974, it looked like game theory might find a natural home in OR and Management Science. Shortly after, the new journal MOR was formed, with Bob Aumann as the area editor for game theory.  But...things didn't turn out that way, and game theory caught on in Economics much more than in OR. And game theory eventually gave birth to the emerging field of market design.

It looks like market design may be a theme that brings game theory back into OR/MS. Here are some sessions to be presented at the forthcoming INFORMS meetings...

Cluster Information

Title: AuctionsChair: Bob Day,Assistant Professor, University of Connecticut, 2100 Hillside Rd, Unit 1041, Storrs CT 06269-1041, United States of America,
Co-Chair: Martin Bichler,TU Munich, Boltzmannstr. 3, Garching 85748, Germany,

Sunday Nov 13, 08:00 - 09:30 : Combinatorial Auction Design and Applications IChair: Dries Goossens,K.U.Leuven, Naamsestraat 69, Leuven 3000, Belgium,
Sunday Nov 13, 11:00 - 12:30 : Core-Selecting AuctionsChair: Bob Day,Assistant Professor, University of Connecticut, 2100 Hillside Rd, Unit 1041, Storrs CT 06269-1041, United States of America,
Sunday Nov 13, 13:30 - 15:00 : Spectrum AuctionsChair: Bob Day,Assistant Professor, University of Connecticut, 2100 Hillside Rd, Unit 1041, Storrs CT 06269-1041, United States of America,
Sunday Nov 13, 16:30 - 18:00 : Panel on Research Challenges in Market DesignChair: Martin Bichler,TU Munich, Boltzmannstr. 3, Garching 85748, Germany,
Monday Nov 14, 08:00 - 09:30 : Combinatorial Auction Design and Applications IIChair: Benjamin Lubin,Assistant Professor, Boston University, 595 Commonwealth Ave, Information Systems Department, Boston MA 02215, United States of America,
Monday Nov 14, 11:00 - 12:30 : Procurement Auctions and Combinatorial ExchangesChair: Damian Beil,Associate Professor, University of Michigan, 701 Tappan St, Ann Arbor MI 48109, United States of America,
Monday Nov 14, 13:30 - 15:00 : Algorithmic Game TheoryChair: Kevin Leyton-Brown,University of British Columbia, 201-2366 Main Mall, Vancouver BC, Canada,
Monday Nov 14, 16:30 - 18:00 : Behavioral Market DesignChair: Kemal Guler,HP Labs, 1501 Page Mill Rd MS 1040, Palo Alto CA, United States of America,
Tuesday Nov 15, 08:00 - 09:30 : Auctions and Trading AgentsChair: Wolf Ketter,Assistant Professor, Rotterdam School of Management, Burgemeester Oudlaan 50, T9-07, Rotterdam 3062PA, Netherlands,
Tuesday Nov 15, 11:00 - 12:30 : Algorithmic and Implementation Issues in AuctionsChair: Sasa Pekec,Duke University, Fuqua School of Business, Durham NC, United States of America,
Tuesday Nov 15, 13:30 - 15:00 : Auctions and their ApplicationsChair: Ravi Bapna,Professor, University of Minnesota, 4600 Washburn Avenue South, Minneapolis MN 55410, United States of America,
Tuesday Nov 15, 16:30 - 18:00 : Auction Design with Multi-dimensional TypesChair: Sasa Pekec,Duke University, Fuqua School of Business, Durham NC, United States of America,

Friday, November 11, 2011

How to allocate goods when the waiting list is essentially infinite. New queues for overloaded systems, by Jacob Leshno

Matching is about who gets what when allocation isn't entirely by price. And a common means of allocating scarce goods is by waiting lists. But there are special problems to consider when the goods being allocated are so scarce that most of those waiting will never receive an allocation. How should applicants be matched to goods when they become available, and how can applicants be given an incentive to pass on goods that might more efficiently be allocated to someone else, if match quality is private information? In considering these things, Jacob Leshno opens up a new front for market design, and introduces a novel kind of queue and some new ideas to the venerable study of queues.

For example, the city of Chicago caps the waiting list for public housing at 60,000 people. The total supply of public housing that they administer is around 20,000 units, and units only become vacant after being occupied for several years, so it's clear that most people who are eligible for public housing will never get it; the system is simply overloaded.  How should the places be allocated? The whole point of public housing is that we don't want to use price to determine who gets what by pricing the poorest out of the market.

Dynamic Matching in Overloaded Systems by Jacob Leshno
Abstract: "In many assignment problems items arrive stochastically over time. When items are scarce agents form an overloaded waiting list and items are dynamically allocated as they arrive; two examples are public housing and organs for transplant. Even when all the scarce items are allocated, there is the efficiency question of how to assign the right items to the right agents. I develop a model in which impatient agents with heterogeneous preferences wait to be assigned scarce heterogeneous items that arrive stochastically over time. Social welfare is maximized by appropriately matching agents to items, but an individual impatient agent may misreport her preferences to receive an earlier mismatched item. To incentivize an agent to avoid mismatch, the policy needs to provide the agent with a (stochastic) guarantee of future assignment, which I model as putting the agents in a priority buffer-queue. I first consider a standard queue-based allocation policy and derive its welfare properties. To determine the optimal policy, I formulate the dynamic assignment problem as a dynamic mechanism design problem without transfers. The resulting optimal incentive compatible policy uses a buffer-queue of a new queueing policy, the uniform wait queue, to minimize the probability of mismatching agents. Finally, I derive a robustly optimal policy which uses a simple rule: giving equal priority to every agent who declines a mismatched item (a SIRO buffer-queue). This robustly optimal policy has several good properties that make it a compelling market design policy recommendation."

That is, to reduce mismatches, the allocation process has to give some people the incentive to wait when they are offered a space of the “wrong” type for them, e.g. in a wrong location.  It is easy to incentivize the first such person, e.g. you could promise him that if he declines the space currently being offered, he will be the first to get a space in his preferred location when it becomes available. If the anticipated wait is short enough, this could be an attractive proposition. But suppose the next person on line likes the same location as the first person? You couldn’t offer him as good a deal, since he would now have to wait for the second space to become available in his preferred location. So, it could be that, before someone who actually prefers the current location comes to the head of the line (which they have to reveal, since it’s private information), a mismatch would occur when a person decides it’s better to take the wrong location than to wait further for the kth place at the location he prefers. In a simple model with just two locations, Jacob shows (the far from obvious conclusion) that maximizing the number k of people willing to wait for their preferred location is the same as minimizing the steady state probability of mismatches.

But conventional queue disciplines (e.g. FIFO, LIFO, etc.) don’t maximize the number of people willing to wait in this buffer queue of folks who would have been mismatched if they had taken what was first offered to them. In solving the optimization problem, Jacob discovered/invented a new queue discipline that he calls uniform wait. People who choose to wait in the buffer queue for a given location are given different probabilities of being called to the head of the queue as different housing spaces arrive to be allocated, based on how many people have joined the buffer queue so far, so that the expected wait for any of the 1,…k people who might decide to wait is equal at the moment that they have to decide whether to join the buffer queue. That is, if you are the first to join, you will get the location you are waiting for if it arrives before someone else has joined, and after that you will have some probability of getting each apartment-at-the-right-location as one arrives, and that probability will depend on how many other people are also waiting. In this way, the 2nd through kth people who join can also be given good incentives to wait for a perfect match; they might not have to wait for all the people ahead of them to be served. Everyone faces the same choice; to take the offered apartment (at the wrong location) or to have a fixed (uniform) waiting time for one at the location they prefer, whether they would be the first person in the buffer queue or the kth. Mismatches will now occur only when this buffer queue has so many people that new people prefer to accept a mismatch than to wait.

Jacob also proposes an almost optimal, more robust solution (that doesn't have to be tuned to the particular parameters of the problem) which is to place mismatched applicants into an unordered buffer, from which they are selected at random when an object of the kind they are all waiting for becomes available.

Incidentally, Jacob is a local hero:
"The Martin Award is awarded to HBS doctoral students enrolled in the PhD in Business Economics program who have excelled at conducting outstanding academic research. This year, the faculty have selected one student to receive the Martin Award.
Jacob Leshno – Jacob studies market design, specifically the dynamic allocation of scarce resources through the use of waiting lists."

Jacob is on the job market this year, you could hire him. Here's a link to his papers.

Thursday, November 10, 2011

Should you guess on the SAT? And do you? Katie Baldiga finds men and women are different.

What accounts for the gender gap in (high) test scores, for the fact that SAT scores predict college success less well for women then for men, and perhaps for other gaps that persist between men and women?  An innovative line of inquiry is carried out in a paper whose title telegraphs what may be part of the answer:
Gender Differences in Willingness to Guess and the Implications for Test Scores by Katherine Baldiga

Here's the Abstract:  "Multiple-choice tests play a large role in determining academic and professional outcomes. Performance on these tests hinges not only on a test-taker's knowledge of the material but also on his willingness to guess when unsure about the answer. In this paper, we present the results of an experiment that explores whether women skip more questions than men. The experimental test consists of practice questions from the World History and U.S. History SAT II subject tests; we vary the size of the penalty imposed for a wrong answer and the salience of the evaluative nature of the task. We find that when no penalty is assessed for a wrong answer, all test-takers answer every question. But, when there is a small penalty for wrong answers and the task is explicitly framed as an SAT, women answer signifi cantly fewer questions than men. We see no differences in knowledge of the material or confidence in these test-takers, and differences in risk preferences fail to explain all of the observed gap. Because the gender gap exists only when the task is framed as an SAT, we argue that differences in competitive attitudes may drive the gender differences we observe. Finally, we show that, conditional on their knowledge of the material, test-takers who skip questions do significantly worse on our experimental test, putting women and more risk averse test-takers at a disadvantage."

Katie's experiment is designed as follows. Each subject participates in only one of four experimental conditions, determined by whether there is a penalty for answering incorrectly or not (i.e. whether points are subtracted for wrong answers or not), and whether or not the test is framed as an SAT, by reminding participants that the questions are drawn from SATs and will be scored like SATs. That's the "between subject" part of her design, and the abstract makes clear how those comparisons played out.

The "within subject" part of the design is that every subject participated in three tests. The first consisted of the SAT questions, and subjects were free to skip questions they did not feel confident they could answer. The second was a test of risk aversion. The third test consisted of the same SAT questions as the first, but subjects were asked to answer every question even if they were not confident that they knew the answer. Having the data from the three tests for each subject allows Katie to compare, on the first test, subjects who did equally well when they answered all questions, and to determine how much of the skipping of questions can be accounted for by differences in their risk aversion. This is what lets her see that the women skipped more questions, and got lower scores than the men, even when they could answer the same number of questions correctly. Which could be one of the reasons why the women's scores would predict future performance less well than the men's.

Katie is a theorist as well as an experimenter: here are her other papers, on social choice theory. She is on the job market this year; you could hire her.

If you are at the ESA meetings in Tucson tomorrow you can also listen to her in what looks to be a great session (I've heard all the speakers before):  Friday, November 11, 1:20 pm – 2:40pm,
Session 4, Ocotillo: GENDER 2

Katherine Baldiga, “Gender Differences in Willingness to Guess”
Johanna Mollerstrom, “Framing and Gender: It's all about the Women”
Muriel Niederle, “Do Single-Sex Schools Make Boys and Girls More Competitive?”

Wednesday, November 9, 2011

Matching Japanese Doctors: problems with the current mechanisms, and suggestions for improvement by Yuichiro Kamada and Fuhito Kojima

How to get more doctors into rural hospitals? That's a problem that confronts Japanese medical administrators, and American ones too. Apparently the Japanese Ministry of Health, Labor and Welfare has somewhat more control over the medical labor market than is available in the U.S., since they have instituted regional caps on how many new doctors--residents--can be assigned to urban regions.

But the way that they have implemented these caps, and integrated them with the job market for Japanese medical residents, isn't efficient. This is pointed out in a new paper which also proposes and analyzes an alternative design with more appealing properties. (A related paper also considers a simpler fix for the problem, and discusses why this wouldn't quite work...). Here's the new paper:
Improving Efficiency in Matching Markets with Regional Caps: The Case of the Japan Residency Matching Program by Yuichiro Kamada and Fuhito Kojima), December 2010. (Revise and Resubmit, American Economic Review.

(A non-technical introduction to this paper (in Japanese) is here (written by Yuichiro Kamada, Fuhito Kojima and Jun Wako), and here's a short summary in English, which also discusses why a simple fix--an "iterated deferred acceptance algorithm"-- wouldn't be strategy proof for doctors): Stability and Strategy-Proofness for Matching with Constraints: A Problem in the Japanese Medical Matching and Its Solution by Yuichiro Kamada and Fuhito Kojima, September 2011, Forthcoming, American Economic Review Papers and Proceedings.)

Here's the Abstract: "In an attempt to increase the placement of medical residents in rural hospitals, the Japanese government recently introduced "regional caps" which restrict the total number of residents matched within each region of the country. To accommodate regional caps, the government modified the deferred acceptance mechanism in a particular manner. Motivated by this policy change, we study the design of matching markets under constraints on doctor distribution. This paper shows that the Japanese mechanism may result in avoidable inefficiency and instability and proposes a better mechanism that improves upon it in terms of efficiency and stability while respecting the regional caps."

The inefficiency that they observe occurs arises because of the way the regional cap is translated into caps on the number of residents that can be hired by individual hospitals in the region. If the regional cap is to be, say, 75% of the total of the regional hospitals' original capacity to receive new residents in the Japanese Medical Resident Matching Program (JRMP), then each hospital's individual capacity is set to be .75 of its original capacity. The inefficiency arises when some hospitals fail to fill all of their capacity defined in this way, while other hospitals, which have filled their new, artificially low capacity, are prevented from hiring extra residents even though they could do so without violating the cap on the number of residents allowed in the region. These hospitals could even be part of blocking pairs that would not violate the regional caps, i.e.the outcome would be unstable even under an appropriately defined notion of stability with regional caps.

The JRMP uses a doctor-proposing deferred acceptance algorithm (like the U.S. NRMP, although apparently without the many match variations involved in the American match.) One way proposed to fix the problem described above would be to run the deferred acceptance algorithm once, and if some hospitals in a region had empty positions, allow these to revert back to other hospitals in the region and run the JRMP again. Yuichiro and Fuhito observe that this iterated deferred acceptance algorithm wouldn't be strategy-proof for doctors: it might give doctors incentives to truncate the preference lists they submit.

Instead, they observe that each proposed matching determines not only an assignment of residents to hospitals, but also, for each region, a vector of how many residents have been assigned to each of the region's hospitals. If this vector is evaluated according to some "regional preference relation" over vectors that obeys the substitutes property, then a notion of "stability under regional preferences" can be defined that allows many of the recent results from the literature on matching with contracts to be applied. (An example of a regional preference with the substitutes property would be to prefer vectors in which hospitals were closer to having proportional caps to those in which some hospitals received disproportionately more residents than others.)

They propose what they call a (doctor-proposing) Flexible Deferred Acceptance Algorithm, which allows hospitals to more flexibly determine how many residents to hire, while respecting the regional caps. The way it works is that, after a conventional doctor-proposing step of the deferred acceptance algorithm, each region selects its most preferred feasible vector of capacities, and hospitals in each region choose their choice sets with respect to this capacity, and reject only those applicants who aren't chosen under this capacity. It produces an outcome that is both stable under regional preferences and strategy proof for doctors.

It looks like it could be put to use in Japan in the years to come.

Yuichiro is an already-widely published game theorist with broad interests, mostly in more classical kinds of game theory. His papers are here. You could hire him this year.

Tuesday, November 8, 2011

Market design in a future of trusted smart markets: paper by Eduardo Azevedo and Eric Budish

The recent NBER conference on market design had a number of remarkable papers. One of them, by Eduardo Azevedo and Eric Budish, seems to me to offer a tantalizing glimpse at what market design might look at some time in the not too distant future when there is a high level of trust in computerized "smart" markets in which a proxy agent reliably acts on your behalf.

At least that's one way to interpret their paper "Strategyproofness in the Large as a Desideratum for Market Design." Among other things, it considers a kind of extension of the revelation principle that would allow a non-strategy-proof mechanism with an attractive Bayesian Nash equilibrium to be converted into a direct mechanism (e.g. one in which agents were asked to reveal their preferences) that would be "strategy proof in the large," (SP-L) i.e. approximately strategy proof in large markets, and strategy proof in the limit. (Another big contribution of their paper is making precise the idea of strategy proofness in the large, which, they argue, may be a desirable criterion when no strategy proof mechanism exists, or when markets are large...the idea is that mechanisms are SP-L but not strategy proof when they allow players' reports to influence prices in ways that vanish in the limit, but mechanisms that aren't even SP-L allow more fundamental manipulations, e.g. they don't give you what you want even when you're a price taker.)

About the revelation principle type mechanisms hey say:
"The construction works as follows. Agents report their types to our mechanism. Our mechanism then calculates the empirical distribution of these types, and then “activates” the Bayes-Nash equilibrium strategy of the original mechanism associated with this empirical. If agents all report their preferences truthfully, this construction will yield the same outcome as the original mechanism in the large-market limit, because the empirical distribution of reported types converges to the underlying true distribution. The subtle part of our construction is what happens if some agents systematically misreport their preferences, e.g., they make mistakes. Suppose the true prior is u , but for some reason the agents other than agent i systematically misreport their preferences, according to distribution m. In a finite market, with sampling error, the empirical distribution of the other agents’ reports is say m^ . As the market grows large, m^ is converging to m, and also i’s influence on the empirical distribution is vanishing. Thus in the limit, our construction will activate the Bayes-Nash equilibrium strategy associated with m. This is the “wrong” prior – but agent i does not care. From his perspective, the other agents are reporting according to m, and then playing the Bayes-Nash equilibrium strategy associated with m, so i too wishes to play the Bayes-Nash equilibrium strategy associated with m. This is exactly what our constructed mechanism does on i’s behalf in the limit. Hence, no matter how the other agents play, i wishes to report his own type truthfully in the limit, i.e., the constructed mechanism is SP-L."

The attraction of such a mechanism of course is that it doesn't depend on the agents reaching a Bayes-Nash equilibrium, which is the problem with mechanisms whose desirability is based on the attractiveness of their equilibrium behavior. Equilibrium may be hard to reach, and such mechanisms may perform badly in practice as a result. But coordination on an equilibrium is much easier when truth telling is a dominant strategy.

The reason this seems like a future mechanism rather than one that is promising for practical application right now is that it is pretty opaque, the opposite of transparent. I can't yet imagine going to e.g. a school district and proposing such a mechanism, which you'd have to sell to parents by saying "tell us your true preferences, and we'll act on your behalf to get you your highest ranked school choice by playing the equilibrium that will arise when we see the choices of all families." The problem is not just that the equilibrium might be hard to describe in the abstract, but that this difficulty is compounded by the fact that assignments will depend in this hard to describe way on an unknown distribution of preferences.

But what might be a tough sell today will be a much easier sell when everyone is accustomed to having their data automatically backed up in the cloud by software that optimizes performance based on things only it observes, and to having their electricity consumption mediated by smart meters that run the air-conditioner in a way that reduces costs based on spot prices, etc. is like that. Just as bridges have gotten longer and stronger over time, there's no reason to think that the market designs of today will be the ones we build in the future. The prospect of confidently putting yourself in the hands of a non-transparent automated expert that you may not understand, a "Martian system" so to speak, may be agreeable to the general public of the future.

(The phrase "Martian system" is one I recall from the early days of expert systems and decision aids. The idea was that you were likely to trust an automated adviser more if you could understand its reasoning, and so judge when its advice was likely to be correct. If you got a non-intuitive answer from an opaque oracle, a "martian system" instead of an expert system, you might worry that the answer was wrong because of wrong inputs or bad construction, and so ignore it. But a transparent system might convince you that a non-intuitive answer was correct, if you were more confident that when it wasn't correct you could tell. But if the martian adviser became so reliable that you could be sure he would not produce an incorrect answer, his opacity might become less of a drawback, since you could rely on him anyway.)

By the way, did I mention that Eduardo is on the job market this year? He's a talented theorist with broad interests who has already made important contributions to matching theory, among other things. Here are his papers. You could hire him.