Bryan Caplan  

Education Signaling in China

PRINT
Does Facebook Restrict Liberty... Bob Murphy's Critique of Bill ...
Yang, a Manchester student from China, emailed me some interesting observations about education signaling in China.  Reprinted with his permission.


Professor Caplan, 

Your signalling model is illuminating. Allow me to furnish you with some data from China. 

After secondary schooling, students in China compete for placements in higher education by participating in Gaokao. The number of universities you can apply to is limited. To someone who has done pretty good on the exam, the following options are presumably open: 

a. Enter a lousy and useless department in a big name university. 
b. Enter a renowned department in a specialised and less famous university.  

In terms of attraction, a>b will correspond to the signalling model, a<=b will confirm human capital theory.  
 
In reality, a completely destroys b. 

As someone who have experienced both Chinese and Western higher education, my impression is that a Chinese university education is even more useless generally in terms of job relevance. Even more severe than the heavy taxpayer subsidy for higher education in America, we have a nationalised higher education system which exercises stringent price control. The result is more and more students graduate into unemployment.  

And I agree with you, technology is not going to solve this.


In a later email, Yang adds:


A bit more of my observation: on the employers side, I am more familiar with the hiring behaviour of 'Shiye Danwei', a form of non-public institutions which are controlled by the state (For example almost every newspaper and tv station in China is organised in this way) and the good old state-enterprises. Excluding many who get in these institutions through back door, the qualification requirements are usually high. Good university degree is a must. And a master's degree is increasingly becoming a must, too. I am guessing a mixture: signalling, heads of these units competing for prestige, and general culture reverence for learning? 

Signalling: The quality of master programs, especially that of liberal arts, is just non-existent. HOWEVER, to be chosen onto a master program, you have to sit a national exam, the preparation of which entails more than a year of memorizing and preparing. That means even if you sit through the three years, you have already proven yourself a super hard-working, super conforming person by getting there. 



COMMENTS (9 to date)
Saturos writes:

But doesn't the fact that this system is enforced by the government make it invalid as evidence for the signaling theory more generally?

Anecdotally, I can tell you that it's much the same in India - only India probably suffers worse shortages of places in eg. medicine.

Yang writes:

It would be interesting to see if any fellow Chinese would my challenge my narrative of a undervalued good-department-in-a-bad-college. China is a big place and Gaokao does differ regionally.

But the Exam for a master's Program is national. I don't think anyone who has gone through or see people go through a Chinese master's program would tell the exam is easy.

david writes:

Yeah, a labour system where connections matter far more than merit rewards a college where you can network more than a place where you can build up human capital. This is not a surprising result. Add in the structural unemployment and this is consistent with college-as-artificial-rationing, not college-as-signalling. It's basically an expensive lottery/queue combination.

It may be worth comparing to Hong Kong, Singapore, and Korea where high school graduation exams are also heavily standardized and crucial for college entry, but the graduate labor market is plausibly less distorted.

Yang writes:

People may have overrated 'connection' here. As I have said the area I am familiar with is not private sector, which is far more important and cost-sensitive.

People with real connection don't throw away three years of life to finish a master degree. They go abroad or have a job anyway. Still, employers are requiring master degree more and more.

David Jinkins writes:

I don't see this as open-shut evidence of signaling (note: it is an anecdote anyway, but still...). If you go to an ivy league, you will meet lots of high flyers. These guys will help you in your career later on. If you want to improve your skill set, you can take classes in other "good" departments, which are the best available in all of China. You also might be able to transfer to a better department.

Bryan, in general, I want to know how your signaling model can explain this and this together. The supply of college educated labor is going up, and the college premium is also going up. People who study this sort of thing (Acemoglu) explain these pictures with skill-biased technological change. You don't believe that college creates skill, so what do you think is going on?

Glen S. McGhee writes:

David -- There is nothing to explain here. The first chart only tracks growth in the number of credentials produced, an the second is out of date.

This is more accurate:
http://www.heldrichpodcasts.com/Chasing_American_Dream_Report.pdf

David Jinkins writes:

Glen -- Right, why is the credential production number going up? The second chart doesn't say that it is great to have a diploma. It says that the wage premium for having a diploma is increasing.

Seth writes:

What's strange to me from this and the original model is why the cost of the signal rises with productivity and leads to education that is "useless generally in terms of job relevance".

Charlie Vidal writes:

As a Public Policy major at the University of Chicago, I chose option A.

Comments for this entry have been closed
Return to top