Harvard's financial report

Here's a link to the Harvard University Financial Report for fiscal year 2009.

Simultaneous offers in a computer science search at Harvard

Michael Mitzenmacher writes about some job searches in computer science, including one that he chaired that was able to make 6 simultaneous offers, and ended up hiring 3 people. He argues that making simultaneous offers sends a signal. (It is certainly a signal about the support the university is offering to the department, and the direction that it is taking...)

"in our last search (which I was leading), where we ended up making 6 offers (and got 3 acceptances). We (the hiring committee) recognized that we were making a rather significant request to have 6 simultaneous outstanding offers. We also recognized the dangers in trying to sequentialize these offers. First, there was the internal danger -- the complex discussions (we had such a great committee, we wouldn't have argued) we would have had to undertake to rank-order everyone we wanted to make an offer to. And second, there's the external danger that the candidate -- who will, of course, find out they were the "second choice" -- takes the ordering as a negative signal and becomes more inclined to take another offer. One can argue whether or not a candidate should take such an ordering as a signal, or whether such a reaction is a purely emotional response. (See Mihai's post, for example, and judge for yourself in that case.) But it was clear to us that, even if no such signal was intended, there was a high risk that would be the interpretation from the standpoint of the candidate.Mihai's post provides a solid empirical data point that we were right to have this concern; it's something I will keep in mind (and if necessary point to) in future hiring discussions. I'm glad we were able to make 6 simultaneous offers, and give all of the candidates we made offers to the right signal."

HT: Itai Ashlagi, who is on the job market this year

Interviewing for the medical match

A recent article* in the Journal of the American Medical Association talks about the strategic behavior that precedes the submission of rank order preference lists to the medical match. The author was a student member of the NRMP board of directors who then participated in the match, and he describes the ways in which residency directors skirted the Match rules against soliciting information or commitments about how students would rank them. He also describes the kinds of signaling and information exchange that might not be unusual in other job markets, but which might have less consequence than in a centralized clearinghouse.

"I was startled when my first interview with an assistant program director abruptly turned from an easygoing chat to an unfriendly challenge:“Why would you ever come here?” Throughout the rest of the season, other interviewers often pushed the MPA’s boundaries, asking me,“How seriously are you considering our program?” or similar questions. Such inquiries are not violations, strictly speaking, but they still suggested that I had to make a commitment to be competitive. Worse, several interviewers did commit unambiguous violations: “If you want to match here, you have to let us know,” or “If we had a position for you, would you come here?”
...

" The atmosphere of gamesmanship extended beyond interviews. Some programs offered me formal “revisits” while others left it to me to request them. I wondered if it was necessary to travel to programs to improve my chances there. Soon after, program directors, faculty, or even residents I had met in passing started making recruitment calls, sometimes weekly.
No matter how friendly these callers were, their overtures always seemed to end with awkward pauses inviting me to make a commitment. After each one finished, I gave my standard reply, which soon became rote: “I loved visiting your program and would be honored to train there.” "

..."(Another program director inveighed against gamesmanship on the morning of a visit, but later that day, an interviewer stated outright that in order to match there I had to make a commitment to them.)... As the ranking deadline approached, I felt compelled to tell my top-ranked program that it was first, and painstakingly crafted enthusiastic e-mails to others. Many of the other applicants I know did the same.
Most of us applicants were rewarded for participating in these courtship rituals. I was told I was “ranked to match” by a number of programs, and it was public knowledge that other students were receiving similar commitments. That said, whatever relief these assurances gave us was tempered by horror stories from years past.One student had been sent real estate clippings from his top-choice program, covered with breathless Post-it notes: “Looking forward to seeing you here!” Weeks later, he was shocked to match at a different program."

*(Fisher, Carl Erik, 2009, "Manipulation and the Match," JAMA, 302,12 (Oct 1), 1266-7.)

Inside a philosophy search (from the labor demand side)

Lou Marinoff describes the search process that led to the hiring of two philosophers this year at City College of the City University of New York. They received 637 applications. They could have used an electronic application process...

"This posed a major filing challenge since less well-organized applicants sent in their materials in dribs and drabs. Hundreds of letters of recommendation not bundled with applications had to be opened, tallied and filed. At times, six boxes had to be displaced to file one piece of paper. It became a common sight to behold members of our search committee trundling half a dozen boxes to their offices on handcarts so they could keep up with the incoming deluge."

"The committee’s next task was to compile a “long list” of applicants to interview at the upcoming American Philosophical Association (APA) Eastern Division annual meeting. We had less than four weeks to identify and notify the candidates we wanted to meet with in time for the event, held December 27–30 in Philadelphia."

"How did we prune our field from 637 to 27? An important selection criterion was holding a Ph.D. from a good university. ...
A second criterion was research and publication. ...
Third, we needed evidence of undergraduate teaching ability as well as versatility. ...We looked for evidence of outstanding teaching ability, variety, and potential for curriculum development.
Finally, we wanted evidence of administrative service. "

After the interviews came the flyouts:
"We attained consensus on six finalists whom we invited to campus in February 2009. They ran a gauntlet of meetings, and each one had to give a “guest lecture” for an undergraduate course in progress – as opposed to reading a paper in a departmental colloquium. We knew they could all read and write well enough, or they wouldn’t have been finalists."

And finally came the negotiations with the administration
"Not long after handing our final rankings to Dean Reynolds, he called me into his office. He informed me that the abundant fruits of our search, and the energy we had invested in it, had not escaped the notice of our senior administration. Prudentially, I recalled the second of three infamous Chinese curses: “May the government be aware of you.” But this time it was a blessing."

Academic hiring wikis and jobmarket rumors

The internet offers a number of free-form information sharing sites for academic job seekers.

Had a bad experience on the job market? (Think things are hard in economics?) Check out the Universities to fear wiki, on which anonymous accounts of bad interviews in (mostly) humanities departments are recounted (e.g. travel expenses not reimbursed after flyouts, etc.)

See also Universities to love, which is also part of the Academic Jobs Wiki collection (you could start one in your field...).

Some specific fields that seem to have active rumor/wiki participation (by no means all accurate in content or agreeable in tone):

Economics Job Market Rumors

Political Science Job Rumors

IR RUMOR MILL and Jobs Board (if you have to ask, "IR" = International Relations)

Sociology Job Market 2009-10 and Wiki

Theoretical Particle Physics Jobs Rumor Mill
Theoretical Particle Physics Postdocs Jobs Rumor Mill

Astrophysics Job Rumor Mill (where you can find an ad for an Intergalactic Medium Fellowship)

Nobel to Elinor Ostrom and Olliver Williamson

I like it. I recall thinking when Ronald Coase won a Nobel that Williamson should have shared it. And Lin Ostrom is one of the few political scientists whose work I know reasonably well, very much oriented towards design (including the use of laboratory experiments).

Salary databases

This week, as economics graduate students start preparing their job market information, I'm planning to have several posts about job markets. I'll concentrate mostly on academic job markets, mostly in other disciplines. Today I'll focus on information about salaries.
Markets do many things, and one of them is to set wages. Americans are reticent about income, so there's always some interest in stories like this one in the Washington Post: What Washingtonians Make , from the President to an Abraham Lincoln portrayor (not the same guy)
One domain in which salaries are more or less public is government, which includes the pay of professors at state universities. At universities where salaries are public, this changes the culture a bit. When I was a professor at the University of Illinois, the fact that salaries were public (in those days you could check out the state budget book from the library, and find your name in it) meant that my young colleagues and I talked about our raises, and what they might mean, much more freely than at universities where salaries were private, not to say secret.
Non profit corporations of all sorts also have to reveal limited salary information on IRS form 990, stating the pay of officers and the top 5 salaries to non officers. This involves some game playing. When that requirement went into effect, I was teaching at the University of Puttsburgh. The first year, if memory serves, the top 5 salaries to non officers all went to active surgeons in the medical school. The second year, none of those surgeons was listed, the top 5 non-officer salaries were all much lower, and went to medical school department chairs. The surgeons weren't paid less, they had just shifted their compensation from reportable salary to non-reportable practice plan payments.
Below is a miscellaneous collection of searchable databases, including State salaries and Form 990's, thanks to the Freedom of Information Act.
State salary databases
http://wikifoia.pbworks.com/State-Salary-Database from Wikifoia, "The wiki for helping people understand and use the Freedom of Information Act at the state and local level."
Sunshine Review, "Establishing the Standard for Government Transparency" Public Employee Salaries

http://umich.highedsalaries.com/ Higher Education Salaries, seems to have salaries for Michigan, George Mason U, and Purdue... (Michigan was a pioneer in making salaries public on the web Faculty and staff salary record.Ann Arbor, Mich. : University of Michigan (excel spreadsheets, by year)) {Update from the comments: The student newspaper at Michigan has put up a better, searchable database of the Michigan salary information: http://data.michigandaily.com/tmdsal }
Ontario public salary disclosure for 2009 Ontario universities http://www.fin.gov.on.ca/english/publications/salarydisclosure/2009/univer09.html
British Columbia public salaries, with a searchable database.

Non profit companies: ERI Nonprofit Organization Information (IRS Form 990 data, including "Part V List of Officers, Directors, Trustees and Key Employees," and "Compensation of Five Highest Paid Employees Other Than Officers, Directors and Trustees (Schedule A)". (Some organizations fill out these forms more transparently than others.)
Private sector: I don't know of any comparable resources for private sector salaries (where no sunshine or freedom of information acts apply), although there are places where you can get samples rather than full databases, e.g. Glassdoor.com has http://www.glassdoor.com/Salaries/index.htm, and PayScale has a variety of surveys reporting average salaries, e.g. http://www.payscale.com/research/US/Country=United_States/Salary , with an index here http://www.payscale.com/index/US
Misc. updates:
Here is the summary of the Association to Advance Collegiate Schools of Business 2008-09 US SALARY SURVEY REPORT.
Here is a collection of state university salary databases.
and here: http://www.collegiatetimes.com/databases/salaries
university and college president salaries
The Dec. 2010 Forward publishes a list of salaries of Jewish community organization leaders.

Last minute tickets at sold out Fenway

The Red Sox, whose home games are always sold out, reserve some tickets for last minute purchase: At Sold-Out Fenway, a Way in for Patient Fans .
"For every home contest, whether a weekday game in April or a Game 7 in October, the Red Sox set aside some tickets for fans who did not plan ahead. There may be a few dozen seats available, there may be a few hundred. The point is, they are available, every game, for those who waited too long but are willing to wait a bit longer.
Starting five hours before the scheduled first pitch, fans can line up on the sidewalk near Gate E, around the corner from hectic Yawkey Way. There is a red-and-white sign reading “Game Day Ticket Sales,” and there are a pair of green roll-up doors. Two hours before the game, the doors are raised and tickets are sold at face value. "
...
"“To us, it’s part of a long-term strategy,” said Ron Bumgarner, the team’s vice president for ticketing. “We do not want every fan at Fenway Park to be a season-ticket holder with 81 games. We want as many different people in Red Sox Nation as possible to be able to come to the games.”
Such a system comes with strict rules, and several copies of them are pasted near Gate E. First, five hours is deemed the perfect amount of wait time — too short to spend the night (which used to be allowed, but created the predictable unruliness) but long enough to make it a bit of a chore.
Second, fans must stay in line and cannot save spaces for other fans. In other words, no sitting across the street at the Cask ’n Flagon, or under the center-field stands at the Bleacher Bar, draining beverages while a buddy reserves a spot until the last minute.
Third, once the tickets are purchased — only one per person — fans must enter the ballpark. That means no scalping the ticket to the highest bidder, a temptation for big games."

Sentenced to prison? Hire a consultant.

The NY Times reports (in the Sports section) that
Consultants Are Providing High-Profile Inmates a Game Plan for Coping

"The former Giants wide receiver Plaxico Burress, who is serving a two-year sentence for a weapons charge, recently joined a growing list of high-profile inmates who have hired prison consultants to help them navigate their entry to a confined life. Others have included Bernard L. Madoff, Michael Vick, Mike Tyson, Martha Stewart and Leona Helmsley."

Book publishers, and book citations

It is customary when citing a book in the bibliography of a scholarly article to include the publisher, and the name of a city associated with that publisher. This seems to be a holdover from the days when publishers were relatively small, and perhaps when the British and American versions of a book with the same publisher might have some differences. And maybe knowing the city would help you find the book.

But book publishing has changed, and when I look at the front material of a scholarly book today, as often as not I see something like
"Morgan Kaufmann Publishers, An Imprint of Academic Press, A Division of Harcourt, Inc., San Francisco, San Diego, New York, Boston, London, Sydney, Tokyo."

Maybe it's time to change our citation conventions. (Or maybe, next time you're in Sydney, you'll have a special opportunity to pick up that book...)

Complementarities in markets and computers

One thing that makes it difficult for markets to clear efficiently is if goods are complements, so that you only want good A if you can also have good B (e.g. you need two licences of adjacent-frequency radio spectrum to carry out your broadband business plan, or you are willing to take job A only if your spouse can get an appropriate job in the same city). If the market isn't designed to take this into account, there might be coordination failures, in which some buyer ends up having paid good money but having gotten only part of what he needs, or in which you and your spouse end up with jobs in different cities. Combinatorial auctions, and labor clearinghouses that accomodate couples, are market designs that seek to avoid these kinds of coordination failures.

The same kind of thing can cause your computer to freeze. If program A needs parts X and Y of your computer memory, and so does program B, and they each lock up one of those components while they try to call the other, you may have to send the children out of the room and reboot. The programming solutions for avoiding this can involve some sort of clearinghouse. Here's an article on the subject from Dr. Dobb's Journal: Lock Options, A compile-time deadlock prevention scheme.

The introduction begins:
"The two major problems in concurrent programs are data races and deadlocks. Races occur when two or more threads are accessing shared memory without proper synchronization. Deadlocks occur when synchronization is based on locking and multiple threads block each other's progress. In a typical deadlock scenario, thread A locks the mutex X, and thread B locks the mutex Y. Now, in lockstep, thread A tries to lock Y and B tries to lock X. Neither can make progress, waiting forever for the other thread to release its lock. The program freezes.
In a sense, data races are the opposite of deadlocks. The former result from not enough synchronization, the latter usually happen when there's too much synchronization and it gets out of hand.
There are sophisticated schemes to prevent races and/or deadlocks, but they either require special language support or strict discipline on the part of programmers. Here I present a solution based on a well-known deadlock-avoidance protocol and show how it can be enforced by the compiler. It can be applied to programs in which the number of locks is fixed and known up front."
HT: Ted Roth

Real estate; disaggregating a bundle of services

There are cracks beginning to show in the way real estate is sold. For a long time, the only realistic option for selling a home was to employ a full service realtor, who sells a whole package of services bundled together, and charges five or six percent of the selling price of the house.

But as the internet has eroded the Multiple Listing Service monopoly on home listings, it is beginning to be possible for sellers to put their own homes on the market, with the assistance of a la carte services from realtors.

One such service in the Boston area, http://eplacehomes.com/ offers just such an a la carte "menu of services". It will be interesting to see where this leads, in a market whose standard, full service contract, with a 6% realtor's commission, became strained as house prices rose into seven figures in many communities. (See this earlier post on broker rebates to buyers.)

Skydeck360: Signaling in the MBA Job Market

Harvard students Abhinav Agrawal, Patrick Chun and Yasser Salem have launched a platform for students to signal interest to employers looking to hire MBA students. The premise of the platform is quite similar to the signaling mechanism used by the AEA, in that the scarcity of the signals is what creates credibility:

"Skydeck360 will go live in October, and will initially be available for only [1st year MBA] students. The tool is designed for simplicity: all students have to do is sign up on the Skydeck360 website and indicate their top choice companies. This data will then be automatically aggregated and shared with recruiters by Skydeck360 in early November."

One interesting facet of the platform is that it is run as an entrepreneurial venture, rather than by the school itself, and so is an interesting marriage of signaling and entrepreneurial market design...

Article: http://www.harbus.org/home/index.cfm?event=displayArticle&uStory_id=677fb978-c83d-4c25-bff3-0ee233de1a5d

Matching with contracts

When Marilda Sotomayor and I wrote our 1990 book on matching, we divided the book into models without money (like the marriage model) and models with money (like auction models), and showed how similar results obtained for both kinds of models. There's been lots of recent progress in unifying those models and understanding why the results were so similar. The most recent paper in this literature is by John Hatfield and Scott Kominers, Many-to-Many Matching with Contracts, and I asked them to write a blog post about it, which appears below.

"A few years ago, Hatfield and Milgrom (2005) introduced "(many-to-one) matching with contracts," a generalization of classical two-sided matching in which contracts between agents specify both (1) a matching and (2) conditions of the match (such as wages, hours, or specific responsibilities). The results of Hatfield and Milgrom (2005) are surprisingly general--although presented using matching-theoretic language, the Hatfield and Milgrom (2005) model encompasses not only two-sided matching but also several package auction models.Hatfield and Milgrom (2005) assumed a many-to-one matching market, that is, a market in which agents on one side of the market (the "doctors") were never allowed to sign more than one contract. Although reasonable for some applications of matching (such as school choice), some matching applications (such as the United Kingdom Medical Intern match) allow "many-to-many matching," in which every agent can be assigned to multiple agents on the other side of the market.In "Many-to-Many Matching with Contracts," we develop a theory of "many-to-many” version of matching with contracts which extends the model of Hatfield and Milgrom (2005) to allow all agents to accept multiple contracts.This framework extends contract matching to a host of applications not previously covered by generalized matching theory, including the United Kingdom Medical Intern match (discussed above), the United States National Resident Matching Program (which allows agents to pair together and match as "couples" who can receive two jobs), and the matching of consultants to firms. Additionally, many-to-many matching with contracts generalizes the theory of bilateral "buyer-seller markets".
We prove that substitutability of preferences (for agents on both sides of the market) is both sufficient and necessary to guarantee the existence of stable many-to-many contract allocations; in many-to-one applications, in contrast, weaker conditions than substitutability guarantee existence. This result shows that, in general, a stable match is not guaranteed to exist in the matching with couples problem, since couples' preferences are generally assumed to be non-substitutable by nature. These results also provide insight into why the necessity result does not hold in the many-to-one matching case, and allows us to identify a new class of non-substitutable many-to-one preferences which are in some sense projections of substitutable many-to-many preferences, and for which the existence of a stable match is guaranteed."

Market prediction and the Challenger disaster

Michael Trick has a post on Models, Information, and Market Rationality which shows a graph of the stock prices of Morton Thiokol, Lockheed, Martin Marietta, and Rockwell, in the hours after the Challenger disaster.
"The stock price for all of the companies immediately dropped 7-8% after the disaster. Within an hour, three companies went back up to being just 2-3% down, while one company further decreased: Morton Thiokol. The company responsible for the O-ring (of Richard Feynman and ice water fame): Morton Thiokol. It is certainly provocative that the market seemed to know something immediately that took an investigation months to determine. ...
But, as Bryan reminds me, this was not exactly a mystery to everyone at the time: the engineers involved strongly suspected early what the issue was and later fed that information to Feynman. So the information was out there and perhaps that information leaked out to the market in the immediate aftermath of the explosion. So perhaps it is not so mysterious after all. And there may well be other explanations for the larger drop off by Motton Thiokol."

Prediction markets and Olympic cities

Over at MidasOracle.org the word is that Chicago won’t have the Olympics in 2016, despite the predictions of a number of well known prediction markets.

"The Chicago candidacy, which was favored by the prediction markets ...is the one that fared the worst."

"The prediction markets are not able to forecast which country will get the Olympics. The IOC is a close aristocratic group that does not leak information. Hence, it is not possible to aggregate information." (emphasis in original)

I guess President Obama also thought that Chicago had a good chance.

Incentives and food safety

In a disturbing story on food safety, the NY Times focuses on ground beef and reports E. Coli Path Shows Flaws in Ground Beef Inspection .

The problem has to do with both incentives and regulation. While companies have incentives to try to avoid selling contaminated food, it turns out there are incentives for not knowing where the contamination originated, and this makes it hard to eliminate. The problem is that most ground meat is a mix of meat scraps purchased from many providers.

"Meat companies and grocers have been barred from selling ground beef tainted by the virulent strain of E. coli known as O157:H7 since 1994, after an outbreak at Jack in the Box restaurants left four children dead. Yet tens of thousands of people are still sickened annually by this pathogen, federal health officials estimate, with hamburger being the biggest culprit. Ground beef has been blamed for 16 outbreaks in the last three years alone, including the one that left Ms. Smith paralyzed from the waist down. This summer, contamination led to the recall of beef from nearly 3,000 grocers in 41 states. "
...
"Ground beef is usually not simply a chunk of meat run through a grinder. Instead, records and interviews show, a single portion of hamburger meat is often an amalgam of various grades of meat from different parts of cows and even from different slaughterhouses. These cuts of meat are particularly vulnerable to E. coli contamination, food experts and officials say. Despite this, there is no federal requirement for grinders to test their ingredients for the pathogen. "

That is, food processors are required to test their final product, but not their ingredients.

"Those low-grade ingredients are cut from areas of the cow that are more likely to have had contact with feces, which carries E. coli, industry research shows. Yet Cargill, like most meat companies, relies on its suppliers to check for the bacteria and does its own testing only after the ingredients are ground together. The United States Department of Agriculture, which allows grinders to devise their own safety plans, has encouraged them to test ingredients first as a way of increasing the chance of finding contamination.
Unwritten agreements between some companies appear to stand in the way of ingredient testing. Many big slaughterhouses will sell only to grinders who agree not to test their shipments for E. coli, according to officials at two large grinding companies. Slaughterhouses fear that one grinder’s discovery of E. coli will set off a recall of ingredients they sold to others." (emphasis added)
...
"The retail giant Costco is one of the few big producers that tests trimmings for E. coli before grinding, a practice it adopted after a New York woman was sickened in 1998 by its hamburger meat, prompting a recall."
...
"But even Costco, with its huge buying power, said it had met resistance from some big slaughterhouses. “Tyson will not supply us,” Mr. Wilson said. “They don’t want us to test.” "

Kidney sales

Here's a very interesting 9 minute YouTube video on kidney sales, including in the U.S.:CNN Reports On The Buying & Selling Of Human Organs - 10/03/09

It argues that (illegal) sales in the U.S. may still be rare, but not rare like unicorns.

HT to Chris. F. Masse, http://www.midasoracle.org/

Course allocation by Budish, updated

Eric Budish's updated (and still remarkable) paper is here:
The Combinatorial Assignment Problem: Approximate Competetive Equilibrium From Equal Incomes,
and Eric himself is now at Chicago's Booth School of Business.

Here is my previous post on the first version of that paper (which was Eric's jobmarket paper).

Hotbeds of innovation: food in Portland

Why are Silicon Valley and Cambridge MA hotbeds of innovation? A nice NY Times story about the culinary scene in Portland Maine, by Julia Moskin, might give some insights: In Portland’s Restaurants, a Down East Banquet .
"Portland’s many chefs and bakers, its turnip farmers and cookbook sellers and assorted mad food geniuses are gearing up for another lively winter.
“I wouldn’t call it a competition, I’d call it a collective,” Josh Potocki, the chef and owner of 158 Pickett St. Café in South Portland, said of the city’s food scene. “We are all trying to raise the level of food in Portland to insanely high.” ...
"Chefs here feed off one another’s work in a way that’s impossible in larger cities (Portland’s population is about 65,000, and it has a compact urban center), constantly eating in and commenting on one another’s restaurants. “I’ve made enemies, for sure,” said Joe Ricchio, a bartender who makes Vietnamese pho on his days off, has a weakness for flaming scorpion bowls, and writes a blog titled Portland Food Coma.
In 2007, Mr. Ricchio started a festively debauched event now known as Deathmatch, a kind of extended “Iron Chef” dinner, with each invited chef contributing a course. “Each one takes five years off your life,” Mr. Ricchio said."...
"Most of Portland’s young chefs ...have worked in — and walked out of — one another’s kitchens. ...Many have cycled through the twin temples of Sam Hayward’s Fore Street or Hugo’s on Middle Street, where Rob Evans is the chef. These are the kitchens that first defined Portland as a destination for rigorously local and regularly delicious food. "...

“Ninety percent of the best restaurants are chef-owned, small, single-operator places,” said Samantha Hoyt Lindgren, an owner of Rabelais, an all-food bookstore on Middle Street ...“That makes a huge difference in the quality of the food and the relationships with purveyors,” she said."..."And for chefs to become owners, the price of entry — key money, kitchen renovation, etc. — is relatively low. "

For related earlier posts, see Market for hand crafted food and Market for hand crafted food, continued