My mentor Paul Milgrom

 My mentor Paul Milgrom

Author Luo Hantang

Im an assistant professor of finance at the University of Chicago Booth School of business. I am engaged in market design, including financial derivatives, real estate market, bilateral bargaining and allocation of natural resources use rights.

I graduated from Stanford GSB economics in 2019, and Paul Milgrom was one of my two mentors. Paul is an economist and thinker who covers a wide range of fields. As an economist, his 10 most cited papers are distributed in more than five different fields.

In recent years, he has been working with computational scientists to develop an incentive auction method in the signal band, which combines the complex satfc algorithm. Paul is still learning new things, building connections in various fields, and trying to break through the application boundaries of auction. As a student, I am particularly encouraged.

Among auction theorists, Paul is undoubtedly the most important one in terms of contribution theory model. Paul has contributed to almost every field of auction economic theory.

Although his paper has been cited a lot, his achievements in practice may be more important: he has been involved in the design or consultation of almost all band auctions since the first auction by the Federal Communications Commission (FCC) in 1994.

Looking at the written records of the Nobel Committee, it is more his auction design work in practice that makes Paul win the Nobel Prize in economics. Many of the contents of this work have not been well cited by the academic community, because it is specific to the situation and can not be generalized. But this work is far-reaching, and Ill try to describe some of them.

Auction practice vs auction theory

Economist Ronald Coase has an interesting idea: lets auction bands! Whoever is willing to pay the highest price for the band should get it.

Through auctions, the government will not only save time in processing applications, but also raise a lot of revenue through spectrum sales. Coase attended a hearing before the FCC in 1959 to discuss the feasibility of the auction. At that time, the biggest question was: is the auction a fools Day joke? The FCC believes that it is impossible to auction directly to ensure that the spectrum ends up on the right people!

Interestingly, at some point before the auction, the FCC has no bandwidth (pun) to process all applications, so it can only be allocated by lottery. Coase Theorem says that the market is the most efficient way, but it is not.

I dont know how long after that, when the FCC found out that auctions really worked, many countries around the world tried auctions, and there was always debate about the best design. Nowadays, most bands are sold by auction.

Nowadays, the band auction uses a very professional process, which is completely different from the graduate theory class, including simultaneous up auction, combined clock auction, agent clock auction and so on

Why is practice different from that taught in graduate school? This has the particularity of band auction. The model setting of one unit is more concise, which can be used to explain the design of auction mechanism under single parameter to students.

Incentive auctions involve the work of many computer scientists, such as Kevin Leyton brown, who have studied how to combine auctions with basic knapsack problem solving. Its a difficult thing, and theres a theoretical basis behind it - Paul Milgroms work on total substitutes and up auctions, matroids, etc.

Therefore, auction in reality can not be summarized as concisely as one-dimensional mechanism design, and it is even more difficult to teach students. The practical application is usually not as direct and clear as what is taught in school.

In some cases, academic papers will be divided. Some widely concerned theoretical articles are rarely applied in reality. However, many important problems in specific real environment have not been paid attention to by theoretical research.

One example is online advertising auctions. Companies like Google and Facebook have come up with clever tricks like bid leveling.. I worked on Facebooks advertising team in 2013 and found it interesting and practical, but it looked completely different from what was in the classroom.

Maybe its the psychological subjective feeling: the actual work and the theoretical work of the auction have some differences. There are a lot of cool apps that affect billions of dollars in the market, but economists pay little attention to it.

Economists have a lot to do in this field. Market design is chaotic, complex and detail sensitive. But isnt it the same in the real world? If you want to do a job that has a real impact on the world, then you need to face it. And Paul is willing to roll up his sleeves and deal with the real problems of chaos and make a difference. His research inspires peoples minds, and it is also my goal to follow.

he near future

Paul still does a lot of work in academia and policy. On the policy front, he recently participated in the design of the 2017 incentive auction, a successful collaboration between economists and computer scientists. This is a very difficult problem for two reasons. First, its a two-way auction: it buys spectrum from the broadcaster at the same time, repackages it, and sells it to new buyers. Secondly, there are many complicated restrictions on the means used in the process of repackaging. These are basically high complexity problems, which need complex computer methods to deal with. A lot of computer experts, like Kevin Layton brown, are involved.

Paul has also written a number of policy documents related to his work, such as redesign spectrum licensing (2017) and 5g fuel: flexible use and valid licensing (2019) with Glen Weyl and I. He is also engaged in basic research. He recently published a game theory paper with Josh mollner: the appropriate equilibrium of expansion. In addition, he and his students wrote a paper on investment incentive mechanism: investment incentive in near optimal mechanism. All in all, Paul is still active in many areas.

each students

Pauls interdisciplinary knowledge really helped his students, and I started to study the JMP topic because Paul invited Darrell Duffie to his study group to speak on financial auctions, which led me to engage in financial related research.

I reported to Paul about ten different versions of my employment thesis, and he showed up every time he gave a speech. He might be really bored later. But it shows how much time he spends on students. Without Pauls advice and support, my senior brothers and sisters, and some luck, it would be difficult for me to write my employment thesis.

Paul was always generous with his students, not only in writing papers, but also in learning how to speak. After each student in the group has spoken, we will circle around the conference room: each student in the audience must comment on the speakers academic content and presentation style. Sometimes, the feedback process takes up half of the whole speech time.

I think thats a big reason that Pauls students are doing very well, because he urges us to work hard on interesting and important issues, and at the same time learn to show and communicate our results clearly.

Paul taught his students very successfully. In 2017, Piotr dworczak worked at Northwestern University. Yu Ning worked at Emory University in 2015, and Shengwu Li got a job at Harvard. Muhammad akbarpour was a postdoctoral fellow in Chicago and returned to Stanford Business School in 2016. Shota Ichihashi published his paper on the job market in the top issue of the American economic review (AER) the second year after graduation.

For such a good result, I think part of the reason is the choice (only the best student can make him a mentor), but a large part of it is his teaching. He often gathered a group of smart people in a room, had a lot of discussion about papers, and other more general issues: important things that happened in academia and practice. Suowus employment paper was drawn from the divergent discussion in the group, which is a very interesting model.

My inspiration is: first, spend more time listening to other peoples feedback on their work and skills; second, let smart and interesting people talk in the same environment, and cool things will happen.

Anthony Zhang Lee is a visiting scholar at rohantang and an assistant professor of finance at the Booth School of business at the University of Chicago


ThePrizeinEconomicSciences2020. .Thu.15Oct2020.



Milgrom, PaulR.andMollner ,Joshua,ExtendedProperEquilibrium(2020).Availa bleatSSRN:https / / / / .2139/ssrn.3035565.


Produced by Netease Research Bureau (wechat official number: wyyjj163)

Netease Research Bureau is a financial and professional think tank created by Netease News. It integrates the original multimedia matrix of NetEase Finance and economics, relies on the wisdom of hundreds of top economists at home and abroad, conducts rational and objective analysis and interpretation on hot topics of economics, and creates a leading financial think tank with attitude. Welcome to contribute (contribution email: [email protected] uff09u3002

Take a look at the contents not seen here

[wonderful recommendation] Click to enter Netease Research Bureau, China Version

[wonderful recommendation] Click to enter Netease Research Bureau u00b7 International Edition > >

Wei endText.article_ bottom{ width:660px;margin :50pxauto0;}# endText.bottom_ title{padding- bottom:15px;border-bottom :3pxsolid#ddd;}# endText.bottom_ titleh3a{color:#333;font- weight:normal;font-size :20px;font-family:MicrosoftYahei;text-d ecoration:none; I am sorry. endText.part {border- bottom:1pxsolid#e2e2e2 ;margin- bottom:5px;padding-bottom :10px;}# -border{border- bottom:0; I am sorry. endText.partul {margin- top:30px;float :left; width:330px; I am sorry. endText.partulli {font- size:14px;color :#333; B ackground:url ( )leftcenterno-repeat;margin- bottom:20px;padding-left : 10px; font family: Song typeface; line- height:15px; I am sorry. endText.partullia {color:#333;text-d ecoration:none; I am sorry. endText.partullia :hover{text-d ecoration:underline; I am sorry. endText.part.img_ news{ float:left;margin-left :50px;margin- top:25px; I am sorry. endText.part.img_ news.img_ box{ width:280px;height :150px; display:block; I am sorry. endText.part.img_ news.img_ boximg{ width:280px;height :150px;}# endText.part.img_ newsp{font- size:12px;color :#666;text- indent:0;margin :0;}# endText.part.img_ newspa{color:#666;text-d ecoration:none; I am sorry. endText.part.img_ newspa:hover {text-d ecoration:underline; }.w9# endText.article_ bottom{ width:600px; }.w9# endText.partul { width:320px; }.w9# endText.part.img_ news{margin- left:30px; }.w9# endText.part.img_ news.img_ box{ width:250px; }.w9# endText.part.img_ news.img_ boximg{ width:250px;height Netease Research Bureau economists recommendation list: 25 books to enhance your horizon! Liu Yuanchun: faster than expected growth may make China take the lead in the game of big powers Wang Yongjun: promoting development by poverty alleviation: theoretical interpretation and theoretical progress of Phoenix model Zhang Yansheng: China still has two economic indicators to add fire in the fourth quarter: Comments on economic data and policy outlook in the third quarter; the Australian market with the best performance in the world in the past 120 years has not been able to withstand Dong Keyong On the long term reform of Chinas pension system? Science Popularization: masks are high-tech products

Australia, the worlds best performing market in the past 120 years, cant afford it

Why cant I buy a mask? Science Popularization: masks are high-tech products