Nova workboard

a blog from young economists at Nova SBE

Observing the Unobservable: Orchids and the Stress Gene

Establishing the causal impact of education on earnings has always been a difficult task. While it is widely accepted that higher levels of education are linked to higher earnings, it is unclear that this is due to education itself. It is possible, and likely, that individuals that obtain higher levels of education are unobservably different to those that obtain lower qualifications, and for this reason alone reap higher incomes. For instance, people that do well in school may have higher earnings simply because they are more intelligent or better adjusted[1].

Essential to answering this question, then, is finding a way to control for these so-called “unobservables”. A popular way of doing so entails comparing identical twins. This method allows researchers to control for genes and upbringing[2]. However, it also makes a strong assumption: that identical twins do indeed have identical ability, and that any differences in educational attainment are exogenous, or random. Another way to establish causality is to use the instrumental variables method, whereby the endogenous variable (educational attainment) is replaced by a variable with which it is correlated, but at the same time does not directly affect income. Angrist and Krueger (1991) use quarter of birth for this purpose[3]. However, as with most IVs, this method has its pitfalls: the extent to which quarter of birth affects schooling is debatable, and there have been doubts as to its exogeneity. In fact, research has shown that winter births are disproportionately realised by teenagers and the unmarried[4].

Recent developments in psychiatry and genetics may provide a way to control for this ineffable factor that affects both educational outcomes and income. Scientists have isolated a part of a specific gene that is responsible for how individuals react to stressful situations. A meta-analysis by Caspi et al. (2010) shows that individuals with a specific variant of this gene are more likely to be neurotic, reactive to threatening stimuli, and prone to depression[5]. They conclude that there is sound evidence for genetically-driven individual differences in stress sensitivity.

Ellis and Boyce (2008) extend on these findings and conclude that there are two types of individuals with respect to “biological sensitivity to context”: dandelions and orchids[6]. Dandelions, they argue, are less sensitive to context, and therefore more likely to thrive irrespective of their environment. Orchids, on the other hand, are highly reactive to context, and are consequently more likely to be damaged by bad environments and life events. Nonetheless, if exposed to a nurturing environment, orchids are more likely than their peers to develop exceptional talents and pro-social behaviour.

In short, individuals with a certain variant of a certain gene are more likely to act a certain way, given a certain context[7]. This can bring us closer to uncovering the causal impact of education. If researchers were to use the interaction between an individual’s gene variant and their upbringing (for instance, parents’ education), then it would be, at least in part, possible to control for what was hitherto deemed “immeasurable”.

By Margarida Madaleno

[1] Card, D. “The Causal Effect of Education on Earnings.” In Handbook of Labor Economics, by O. Ashenfelter and D. Card, 1801-1863. Elsevier Science, 1999.

[2] Ashenfelter, O., and A. Krueger. “Esimates of the Economic Return to Schooling from a New Sample of Twins.” The American Economic Review, 1994: 1157-1173.

[3] Angrist, J., and A. Krueger. “Does Compulsory School Attendance Affect Schooling and Earnings?” The Quarterly Journal of Economics, 1991: 979-1014.

[4] Buckles, K. & Hungerman, D. (2013). Season of Birth and Later Outcomes: Old Questions, New Answers. Review of Economics and Statistics, 711-724.

[5] Caspi, A., A. Hariri, A. Holmes, R. Uher, and T. Moffit. “Genetic Sensitivity to the Environment: The Case of the Serotonin Transporter Gene and its Implications for Studying Complex Diseases and Traits.” American Journal of Psychiatry, 2010: 509-527.

[6] Ellis, B., and T. Boyce. “Biological Sensitity to Context.” Current Directions in Psychological Science, 2008: 183-187.

[7] Dobbs, D. “The Science of Success.” The Atlantic, December 2009.


Author: studentnovasbe

Master student in Nova Sbe

Comments are closed.