Nova workboard

a blog from young economists at Nova SBE

Signalling vs. Human Capital Model: The Economics of College Majors

Under the human capital model (HCM) different years of schooling leads to different earnings, given other variables constant between individuals. Thus, assuming that studying one major instead of another does not affect human capital value for individuals, then a student graduating in one major should receive the same earnings as another student graduating in a different major. However, this is not true.

Using the American Community Survey data for the period 2009-2011, Carnevale, Cheah and Strohl (2013,2012) showed that there is high disparity in earnings across majors. In particular, they showed that scientific majors are the ones with the highest earnings.

How is this possible?

From a theoretical viewpoint, when the HCM does not directly explain an empirical fact it is proper to check whether its historical enemy, the Signalling model, can explain it. To answer this question, it is necessary to verify that signals sent from the agent (student) to the principal (firm) are costly (i.e. credible).

To measure this cost it is necessary to consider a variable that quantifies this cost and that is specific for each major. A good (proxy) variable is represented by the graduate completion rate. The lower the graduate completion rate; the lower the probability that a student will graduate in that major. Thus the difficulty of graduating in that major is high. As reported by Chang et al.(2010) scientific majors have the lowest completion rate.

Given that studying one major instead of another does not affect the human capital value for individuals, then a person graduating in Mathematics, surviving from the hard natural selection process,  sends a costly and so credible signal. Thus, using this signal, firms give higher wage to him than to other students. Under this logic, the influence of human capital is absent.

This result is supported empirically by two facts. First, when we compare earnings of recent college graduate in different majors, we obtain that scientific majors are the highest. This difference should not exist under the HCM, because, especially in the first years (where abilities do not have time to be transferred in higher wages, because of rigidities), workers with the same years of schooling should earn the same wages.

Second, when we compare earnings of an experienced college graduate with scientific majors with earnings of graduate degree holder with humanities majors, the former is higher than the latter. This directly contradicts the HCM where higher number of years schooling leads to higher wages.

In a world without signalling and where majors are qualitatively equal, students should not choose to study Mathematics due to its cost. In fact, they study it because there is a premium for that cost, which is not based on the quality of knowledge, but on the cost of achieving it. Thus signalling effect prevails.

In conclusion, a principal weakness is the assumption considered (i.e majors are qualitatively equal). However, if this assumption holds, then the logic holds. Despite the results supporting the Signalling model, we need to take into consideration that “years of schooling” is just a proxy for human capital accumulation and it does not reflect its real value. In fact, different majors can potentially lead to different increases in human capital, making the human capital theory relevant.

Riccardo Passeggeri 677 Economics

The data source of the following graphs are Carnevale, Cheah and Strohl (2013,2012).




Mitchell Chang, Sylvia Hurtado, Kevin Eagan and Josephine Gasiewski, “Degrees of Success: Bachelor’s Degrees Completion Rates among Initial STEM Majors”, Higher Education Research Institute at UCLA, January 2010. Available on 13/11/2013 at:,%20Eagan,%20Chang%20-%20Degrees%20of%20Success.pdf

Anthony P. Carnevale, Ban Cheah and Jeff Strohl “Hard Times: College Majors, Unemployment and Earnings. Not All College Degrees Are Created Equal”, Centre on Education and the Workforce, Georgetown University, January 2012. Available on 13/11/2013 at:

Anthony P. Carnevale and Ban Cheah, “Hard Times: College Majors, Unemployment and Earnings”, Centre on Education and the Workforce, Georgetown University, May 2013. Available on 13/11/2013 at:


Author: studentnovasbe

Master student in Nova Sbe

Comments are closed.