The promises and pitfalls of genoeconomics

In keeping up with cutting-edge economics research, GSE chose a controversial topic for its opening seminar in the microeconomic series for this academic year:  Genoeconomics.

Daniel Benjamin came to the UPF campus on September 30th to give an introduction into this brand new field of research. The areas is currently opening up in light of the cheap DNA data now available to researchers. Given, as Benjamin rightly stated, it is natural for economists to seize on this new opportunity for analytical enquiry, he has been testing the question: does our genetic code influence our economic behaviour?

Benjamin gave us a brief overview of the kinds of effects his work his modeling, which I will summarise even more briefly: over 99% of genetic data is the same from one person to another, however along the genome there are certain locations where variation is more likely.  One such variation is a Single Nucleotide Polymorphisms (SNPs – pronounced ‘snips’ – for short); there are around 10 million such variations in the human genome. Although there is also genetic variation of other types, this is the most common. Benjamin and his many collaborators are studying SNPs.

There are two main forms the work has taken to date:

i)              modeling the correctness of predictions of behaviour, given a known genetic code, and

ii)             ii) genome-wide association studies (g-was, for short).

The former predicts ex-ante outcomes and tests the correctness of these predictions. For example, if a certain SNP has been tested in the field of medical science to be associated with brain development, can we test improved cognitive ability in members of the population with this SNP? Most work thus far – both in medical and economics fields – has been under this approach, because of the large sample sizes needed for the latter.

Many of the known interesting SNPs have now been tested, and genetic data is more cheaply available. This has led to an increased attention on the second form of enquiry, which seeks SNPs associated with a given characteristic of interest. For example, in a sample of individuals with high-scoring IQ tests, which SNPs are crop up most often in the data set?

Benjamin explained in some detail the challenges of the techniques being used, and the relative strengths and weaknesses in terms of statistical power. There was some heavy skepticism regarding the models used, such as accounting for gene-on-gene effects and environmental effects. Since the tests are currently being run on individual SNPs, rather than in an aggregate model, these were considered to be dealt with – although, they would present a problem in aggregate models.)

The immediately obvious question popped up: If we know certain genes were to code for certain economic behaviours, how would this influence policy decisions?

We are a good while – and much more research in many disciplines – away from being able to answer this question, but it’s not difficult to imagine a world in which your genetic data is readily available to you, your government, and potentially the marketplace, which makes it an important one to think about upfront. Would we be happy with a world in which John and Jane are sent to pre-schools based on a sample of their spittle? Or, one in which marketing is tailored not only to Google searches and recent online purchases, but your genetic code?

It’s extremely early days for genoeconomics, but there is potential for a raft of research in this direction as more data become available, at a lower cost, and with better information regarding the physical attributes of each SNP. As with any new area of research, there will be some tough years ahead, but I’m confident that economics is an old dog, which likes new tricks. Let’s face it: we need them.

For further reading on this topic please see:

Ebstein RP, Israel S, Chew SH, Zhong S, Knafo A. 2010. Genetics of human social behavior. Neuron, 65:831–44

Beauchamp JP, Cesarini D, Johannesson M, der Loos M, Koellinger P, et al. 2011b. Molecular Molecular genetics and economics. J. Econ. Perspect. 25(4):1–27

Benjamin et al., 2012. The Promises and Pitfalls of Genoeconomics. Annual Review of Economics 4:627-62.

5 thoughts on “The promises and pitfalls of genoeconomics”

  1. It takes a lot of courage to work in such a new area of economics. The problems are clear, but the answers might never appear. I’m glad people like Daniel Benjamin take that risk.

    However, all of the effects are going to be so small and inconclusive at best, fallacious at worst. Will it just lead to endless data mining and no real answers? The theory does not seem to be there yet so it does not seem like economics, but maybe someday. Although, people said the same about Vernon Smith and experimental economics too.

    1. I agree there will a lot of data mining upfront, but I disagree that this isn’t driven by theory: if our genes can describe behaviours, why not economic behaviour?

Comments are closed.