Sunday, 20 June 2010

Freakonomics, Bananomics and Lovenomics

Economics is good fun. Not always, for sure, but often it really is. Those of us who have chosen to pursue an academic career in this discipline of course know it – and so we frequently fall into passionate discussions whether a particular relationship might be endogenous, we marvel at the elegance of a new theoretical proof, and we travel to conferences to boast about our new and exciting research to our colleagues and peers. Yet, the rest of mankind often fails to appreciate the beauty of economic and considers it rather boring (and I suspect that even some of our students may feel that way!). But every now and then, an economist comes up with a new article or book that makes even the non-economists smile and think.

Freakonomics was just such a book. Published in 2005 by a University of Chicago economist Steven Levitt, it succeeded in convincing the popular public that economics can be fun and also that it can shed light on important real-world questions at the same time. The book encompasses a broad range of topics, from analyzing results of school tests or sport events to identify cheating, through arguing that abortions helped lower crime in the US, to analyzing the economics of drug dealing. The book quickly became a best seller – and soon a sequel, blog and movie followed on the coat tails of its success.

Of course, Steve Levitt is not the only academic economist capable of producing research that is publishable and fun to read at the same time. As the title of this short contribution suggests, I want to talk about two other examples. And only one of them is shameless self-promotion.

I am writing this on my way from an international conference where I have seen a presentation by Richard Jong-A-Pin from the University of Zurich entitled “Are Banana Republics Banana Republics?”. The term banana republic is used to denote countries marked by high degree of political instability and dysfunctional state institutions (Woody Allen’s movie “Bananas” depicts the stereotypical image of a banana republic rather well). So Richard in his paper asks a simple and intuitive question: do countries producing bananas really suffer from higher political instability? This question, however, is more complicated than the first look suggests, because there is a possibility of reverse causality: what if politically unstable countries are more likely to produce bananas? Richard resolves the question of endogeneity by means of instrumental variables – he uses a set of third variables, unrelated to political instability, to predict whether a particular country produces bananas and then uses the predicted banana production rather than the actual one to explain political instability. This approach ensures that the regression analysis only picks up causality going from bananas to political instability and not the other way around. Doing so, he finds that banana republics indeed have higher political instability than countries that do not produce bananas.

Now this result begs the follow-up question why banana republics are banana republics. In other words, why are bananas apparently detrimental to political developments? Richard's paper does not address this question but a plausible explanation was suggested by Ronald Wintrobe (University of Western Ontario) who commented on Richard’s paper at the same conference. Bananas, as some other agricultural crops, perish quickly and this creates fertile ground for rent seeking. Let me explain. Suppose I want to buy a truck load of bananas and find a farmer – in an anonymous banana republic – who is happy to sell his bananas to me. We agree the terms of the deal months in advance – but when the bananas are ready to be delivered, I renege on our original agreement and insist on a substantial discount. The farmer now has few options other than accept my ultimatum: he will have a hard time finding a different buyer at such a short notice while taking me to court would take months – and his bananas would rot in the meantime. My bargaining position is therefore much stronger than his and in the end I can get almost any discount I want. The eventual outcome of this, however, may be that banana-producing countries develop a culture of rent seeking – and this eventually translates into political life too. Hence, producing bananas indeed breeds political instability. Watches, on the other hand, are not perishable and therefore Switzerland is not a banana republic.

And now for something completely different. Studying the formation (and dissolution) of relationships has a long tradition in economics, especially since the publication of Gary Becker's seminal “Theory of Marriage” in the Journal of Political Economy in1973. In my recent paper entitled “Anthropometry of Love,” written jointly with Michèle Belot from Oxford, we analyze a peculiar – and at a first sight puzzling – feature of interethnic marriages: men and women belonging to the same ethnic minority group are not always equally likely to intermarry with whites. Among blacks, it is black men who intermarry much more often than black women. The reverse holds for East Asians (for example the Chinese): in this case, women are much more likely to have a white partner than men. These asymmetries could be due to a number of factors. For example, it may be that black men and East Asian women are more numerous in the British society due to the supply of immigrants and/or due to past immigration policies. But that does not appear to be the case, as a quick look at the summary census figures confirms. Another possibility is that men or women of a particular ethnic group are more likely to possess characteristics that make them attractive in the marriage market, such as being more educated or richer. But again, data fail to confirm this view. Instead, we propose an explanation based on physical characteristics, in particular height. There are sizeable differences in height across the different ethnic groups in the UK: the whites and blacks are the tallest, followed by Indians, Pakistani and Chinese while Bangladeshi happen to be the shortest. This can have important implications for marriage-market participation of ethnic minorities and for interethnic marriage as across all ethnicities, we can observe a strong preference for the husband to be taller than his wife. This implies that men and women from the same ethnic group face different distributions of potential marriage partners in the white segment of the marriage market. Specifically, relatively tall men and short women have a wider choice of white partners than short men and tall women. Hence, black men, being relatively tall, are ‘on the top of the food chain’ (to quote an unnamed Brunel student), unlike black women. And, similarly, Chinese and other East Asian women have a similarly wide choice, unlike their male counterparts. So far so good but do the data support this explanation? Using data from the Millennium Cohort Study, we estimate regressions explaining the probability that a member of an ethnic minority has a white partner. We include in our regressions controls for the share of suitable partners – defined for men as those who are shorter and for women as those who are taller than the individual in question – both within the white population and within one’s own ethnic group. Doing so improves the quality of the regression and helps reduce the unexplained variation in the patterns of intermarriage. Height, however, is not the only explanation of intermarriage patterns; cultural norms, for example, are certain to matter too. But it is one of the factors underlying the observed differences. Furthermore, the importance of height for inter-ethnic marriage has some unexpected implications. Firstly, as social attitudes towards interethnic marriage become more liberal, this increases the number of potential marriage partners for black men and East-Asian women but may reduce it for black women and East-Asian men. Secondly, immigration has a marriage-market effect which is different for men and women. And how about policy implications? I am afraid there are none: love and the marriage market are (so far) largely free from government interference. Should this change, I would suggest subsidies for couples involving East-Asian men and black women.

[Originally published in Economics and Finance Newsletter, Brunel University, April 2010]

No comments:

Post a Comment