Back from Barcelona GSE "Trobada" VIII

I have no idea with what "trobada" means, and apparently it is not a spanish word. Yes, it is from Catalan and it means "conference"..... So Barcelona GSE "Trobada" VIII is the annual conference for GSE afflicted professors.

This year it is held in a very beautiful and historical building, Casa Convalescència. Oh, I admit I should have brought my camera there. What a pity! I can only download some pictures from the Internet and paste them here.

Casa Convalescència


Today's topics were very interesting. I listened to three presentations.

Prof. Giacomo Ponzetto (CREI-UPF and Barcelona GSE)
"Politics, Asymmetric Information, and Protectionism"

Prof. Johannes Gierlinger (UAB and Barcelona GSE)
"Hedging Priors: On Novel Instruments to Insure against Ambiguity"

Prof. Davide Cantoni (UPF and Barcelona GSE)
"Natural Experiments in Economic History"

I was there because  the last one, natural experiments in economic history, attracted me. It was a very brilliant idea to "design" natural experiment in economic history. Well, one interesting thing is that nearly all people admit that we need to learn economic history, but this course is still not available here (sorry, it is available. I made a mistake), and not many students will actually registered for it, perhaps due to the way it is taught and the form of exam.

Another thing I was considering for the whole morning was that what's the difference between the questions European economists care about and those from U.S. or China. In China we don't need to think about "lobby", since the political system is so different. If institution plays an important role in determining the economic growth speed and pattern, what can we conclude from China's example?  Is the potential lobby power (like the pressure from big companies) growing and influencing policy makers? Will they still consider the global situation and make a wise choice?

A big problem for China now may be the exchange rate. G20 Conference is holding, and everyone is predicting the trend of future's world economy. I have heard that whenever the Chinese department of commerce is going to make a new economic policy, they will face different pressure from all related industries. Therefore, can they still stay balanced and choose the best strategy for the whole economy? It is definitely hard, and we cannot forecast the result. Economic models cannot tell everything in the real world. We can learn from economic theories, or the real history.

[Update 2010.10.24]

1. Thanks for Jodi's email, and now I update the new pictures. For more, see

2. I emailed Prof. Davide Cantoni and got the list of papers he mentioned on the conference (He is really a nice guy, and I do want to take his economic history course now). I'd like to paste the list here, and so, enjoy~~~

Acemoglu, Johnson, Robinson, "Colonial Origins of Comparative Development" (

Jeremiah Dittmar, "Information Technology and Economic Change: The Impact of the Printing Press" (

James Feyrer and Bruce Sacerdote, "Colonialism and Modern Income: Islands as Natural Experiments" (


Elise Huillery, "History Matters: The Long Term Impact of Colonial Public Investments in French West Africa" (

Melissa Dell "The Persistent Effects of Peru's Mining Mita" (

A paper that gives a good review on this strand of research more generally is Nathan Nunn's contribution to the "Annual Review":

mother breastfeed and children health

This is an interesting topic: the behavior of mother breastfeed and children health. Yes, it is the topic of a seminar I audited yesterday.

As usual, here is the related information:

Why Do Mothers Breastfeed Girls less than Boys? Evidence and Implications for Child Health in India [click here to download PDF]
Seema Jayachandran (Stanford University)
UPF Labor, Public, and Development Seminar

Well, although I'm a kind of being addicted to theoretical works, this kind of empirical is still attractive.  She analyzed the development of children health with respect to mothor's breastfeed behaviors. As she said, it is like "quality instead of quantity", and mothers are paying more and more attention to ensure the health of their children.

Fortunately, one guy who lives nextdoor to me has been worked in the field of public health for the last year, and he has lived in India for several years, thus before the seminar he told me many things about the real situation in India. In this way, far beyond listening to the seminar itself, I have learned more about public health, or health economics which is really hot now.

The forever pain for an econometric or empirical analysis is the accuracy of data. She was using NFHS (National Family Health Survey) and DHS  (Demographic and Health Survey) data. However, my nextdoor told me that there were some problems in the questionnaire used, which will influence the reliability of the data, or the answer from those women. He doubted whether it would have potential influence on the regression analysis.

It was also interesting to hear that so many audience there had different kinds of interesting questions.  That is partically why I love economics seminars. But still the question for me is that I am still not able to follow all of the presentation. Apparently I need to store more knowledge.

By the way, I also listened to another seminar this week, it is

Misspecification in Models for Trends and Cycles
Andrew Harvey (University of Cambridge)
UPF Occasional Econometric Seminar

Well, it was a theoretical work, but it was theoretical economics, and I have no idea with that. Unfortunately, I went there early and sat down in the second row of the classroom. Therefore, I had to pretend that I was focusing on what he talked about, but in fact I could not understand anything except HP filter. Oh my poor econometrics, and my poor time series analysis. It also proves that I need to learn more to completely understand the top seminars. Carry on!

At last, I'd like to copy some conclusions from Semma's paper here.

  • the duration of breastfeeding negatively correlates with the mother's likelihood of a subsequent birth.
  • breastfeeding increases with birth order; if parents have a preference for sons, then boys are breastfed more than girls; children with older brothers are breastfed more; these gender e ects are smallest for high and low values of birth order
  • breastfeeding the current child helps prevent or delay a subsequent pregnancy, and a subsequent (perhaps unwanted) pregnancy often causes mothers to wean the current child.
  • If contraception crowds out breastfeeding, then policy makers might consider pairing contraceptive campaigns with promotion of breastfeeding or improvements in water quality. Conversely, if contraception enables a mother to breastfeed her children longer because she can space them further apart, then policies that expand access to contraception might have an added bene t of encouraging breastfeeding.

Survival Analysis, health economics and alive economics

This post is only used to record that I have listened to this seminar:

Survival Extrapolation with applications in health economics [click here to download PDF]
Nikos Demiris (Agricultural University of Athens)
UPF Statistics and Operational Research Seminar

However, I didn't understand anything except the exponential equation. I had thought I would understand more, since I know a little about survival analysis (What I have learned on the Chinese R conference), and a little about health economics. But on the seminar I realized that it was beyond my scope of knowledge. Anyway, there were only 4 people (before I showed up), and my dear math teacher Lugosi was there! Lugosi, YOU ARE AMAZING!

Since I didn't actually benefit something from the lecture, it is impossible to leave any comments here. Therefore, I'd like to talk about something about economics. Well, I always wonder whether I can always pick up someone to go to the seminar together with me. It is not important whether you are able to understand what they are saying, but the atmosphere can help you adapt to the way that economists think in. Without doubt, there are only few people who prefer seminars to those time-consuming problem sets. Thus, the final conclusion is that I should go anywhere alone.

I have no right to force anyone to do anything. It's their own choices - and I choose seminars. Maybe it is a bad habit, since it is probable that I cannot get anything useful from a seminar. But I love the way they talk and communicate in the seminar, and I love the different stories that the researchers show in their presentations.

As what I wrote on the front page of this website ( -  "Love life, love economics", I always regard economics as a living object. Economics is from the real world, not the mathematical equations. Although we need math that complicated enough to describe the economic world specifically, the economics is, and will always be focusing on the real world. Apart from those who are engaging themselves into developing new mathematical methods for economics (e.g. theoretical  econometricians and mathematical economists), who want to explore the world of economics should pay attention to the real life of economics.

Fine, anyway I should judge anything. At least, for me, economics is always alive around everybody and everyday.

comment on Prestigious Law School

This Tuesday was a disaster for me, because I stayed in the classroom and listened to different things for nearly 10 HOURS! But I shouldn't complain anything, anyway it was my own choice.

At that busy and painful day, I listened to a seminar. As usual, here is related information.

Title: The Returns to Attending a Prestigious Law School

Author: Paul Oyer (Stanford University)

UPF Labor, Public, and Development Seminar

Ok, I should admit that I was too tired to follow and focus on all points of the presentation. Anyway, it was a typical empirical work from labor economists, or education. However, the paper is available online here:

Copy some conclusions from his paper:

  • In 2002 and 2007, those lawyers that went to top 10 law schools made, on average, 25% more than those that went to schools ranked 11-20 and over 50% more than those that went to schools ranked 21-100.
  • Graduates of Top 10 schools were also much more likely to work in large law firms in leading law markets.
  • We find that at least a third of the large returns to law school reputation are due to selection, but this selection is almost entirely on a single variable - quality of the undergraduate institution attended.
  • Our results are consistent with a reasonably large causal effect of attending an elite law school, but the exact size of the premium depends varies with assumptions about the role of unobservables.

For anyone interested, please download the paper and read. I was just interested in the topic, and it indicated it is true that "selection" is the main factor, which means those who attend top law schools are expected better than the rest peers. In the other word, they may be also successful even they are in other areas...