A review of Bai, J. J., Ma, L., Mullally, K. A., & Solomon, D. H. (2019). What a difference a (birth) month makes: The relative age effect and fund manager performance. Journal of financial economics, 132(1), 200-221.
I. Research question and its importance
This paper investigates how early childhood experiences relating to the month of birth affect the confidence and performance of mutual fund managers. The topic is especially important because it directly relates to one of the oldest questions common to both asset pricing and corporate finance: why some firms succeed and others fail. Recent academic attention has turned toward the individual characteristics of firm managers, namely, their overconfidence. Hence, the authors attempt to proxy overconfidence by fund managers’ childhood experience, a relatively novel approach to this question.
II. Method, finding, and limitation
There are two studies in this paper. The first study was based on observational data, via ordinary least squares (OLS) regression and fixed effects, estimating the correlation between age and manager overconfidence. The second study was an online experiment where profile pictures of a sample of relatively older and relatively younger managers from LinkedIn were presented to Amazon’s Mechanical Turk respondents (although MTurk has been dubious for marketing literature, it seems generally acceptable in finance literature. See Owens & Hawkins, 2018), as the research aims to confirm the finding from the first study. Their conclusion, in brief, confirms the variation in birth months is associated with differences in adult labour market outcomes in the mutual fund industry.
Nonetheless, variation in month of birth not only affect one’s schooling, but also many other things, such as military conscription. Thus, the findings might not directly reflect to just a more confidence experience in school.
III. Future research
To answer the research question as posed by the authors (i.e., whether month of birth affect fund manager performance) it’s always desirable to devise an experiment. But such circumstance is impossible as no one, except God, is able to change one’s month of birth. Keeping this in mind, future work might adopt a different yet parsimonious design. Namely, a natural experiment where exogenous macroeconomic change (e.g., stock market collapse) is an identification strategy, to assess the different performance amongst fund managers of different months of birth. In this manner, one can show whether month of birth decidedly permits the manager to have a different characteristic facing a crisis, pertaining to the overconfidence effect. To conclude, research question where dependent variable is not manipulatable, requires more creative research design rather than simply descriptive statistics. The authors’ use of MTurk was interesting, but it has nothing to do with age. Future research might revise research design.
IV. Reference
Owens, J., & Hawkins, E. M. (2018). Using Online Labor Market Participants for Nonprofessional Investor Research: A Comparison of MTurk and Qualtrics Samples. Journal of Information Systems, 33(1), 113-128. https://doi.org/10.2308/isys-52036