Subscribe Twitter Twitter

Thursday, June 27, 2013

The very deep roots of economic development

Could historical factors, determined 10,000 years ago, have some explanatory power in predicting today's levels of economic development between countries? I would not have thought so. It's too long ago that the Neolithic transition, colonization, the industrial revolution, and globalization would have wiped out any predictive power these factors had. Except this paper by Ashraf and Galor which I have just read argues that genetic diversity in a society determined 10,000 years ago, resulting from the exodus of homo sapiens from Africa, are strongly correlated with development outcomes today. What? They are using data from the Human Genome Diversity Cell Line Panel.
The level of genetic diversity within a society is found to have a hump-shaped effect on development outcomes in the pre-colonial era, reflecting the trade-off between the beneficial and the detrimental effects of diversity on productivity. Moreover, the level of genetic diversity in each country today (as determined by the genetic diversities and genetic distances amongst its ancestral populations), has a non-monotonic effect on income per capita in the modern world. While the intermediate level of genetic diversity prevalent among Asian and European populations has been conducive for development, the high degree of diversity among African populations and the low degree of diversity among Native American populations have been a detrimental force in the development of these regions.
I have my doubts mainly about the quality of the data they are using, but the paper is pretty convincing, making use presumably of the most of what historical data we have. In almost 50+ regressions where they control for other potentially confounding factors, the measure of genetic diversity remains a statistically significant predictor of economic development today. This result seems robust.

That history matters was once a strange concept in economics because the neoclassical model predicts convergence among economies. Of course, empirical papers have since provided strong evidence that particular events in history like colonization, slavery, etc. have long shadows and continue to manifest their effects even today. I suppose I just didn't think something that happened a long, long, looong time ago could still matter.

Tuesday, June 25, 2013

Why do Legal Permanent Migrants Return?

Is it because of consumption, investment, or employment in the home country? Which factors are salient? Do home country conditions even matter at all? These are questions I attempt to answer in my new paper on the return motivations of legal permanent migrants. I am pleased to finally put up a draft. 

I focus on legal permanent immigrants in Australia and their motivations for return. Are these migrants more likely to be target earners, who move abroad in order to accumulate resources to invest in some business at home, or are they life-cycle consumers, who primarily balance the marginal benefits and costs of staying abroad? One can distinguish between the two by their reactions to shocks. For instance, target earners are thought to cut their stays abroad shorter when their purchasing power for the home country increases while life-cycle migrants react by making their stays abroad longer.

I use the 1997 Asian Financial Crisis as a quasi-experiment. The figure above encapsulates the empirical strategy (exchange rates are normalized to 1 in 1996). In particular, the crisis generated varied and substantial exchange rate shocks between home country currencies of migrants and the Australian dollar. Moreover, this crisis was largely unexpected, hence plausibly exogenous. Immigrants from Australia come from different countries so it was as if they were randomly allocated different exchange rate shocks, shocks to their purchasing power. The basic approach simply looks at whether those who obtained more favorable exchange rate shocks (depreciations in their currency) returned more or less compared to those who obtained less favorable shocks.

In sum, I find that a 10% increase in the exchange rate (a home country currency depreciation) leads to a 0.37 percentage point reduction in the probability that a migrant returns. The 2-year permanent return rate in this period is small at 4.1% so this effect is almost equivalent to a considerable 10% of the return rate. That these migrants continue to be sensitive to home country conditions is a somewhat surprising result, given that these individuals are granted permission for indefinite stay in Australia. The result is robust and consistent with the story that migrants return due to life-cycle considerations. A substantially larger effect is found for migrants who have pre-determined that they would want to return, evidence that migrants optimally time their return to favorable conditions. Moreover, I show evidence that this exchange rate shock effect is not merely a proxy for the influence of other macroeconomic conditions, such as GDP per capita growth or the change in unemployment in the home country. This suggests that return is primarily a function of purchasing power and consumption rather than employment possibilities in the origin country.

 More details are in the paper. I welcome comments and suggestions at

Sunday, June 16, 2013

What I've Discovered About the Steak We Eat...

is that most of it is corn and petroleum. This I learned from Michael Pollan's, The Omnivore's Dilemma, as he talks about his visit to the factory farm thus:
“In my grandfather’s time, cows were four or five years old at slaughter,” Rich explained. “In the fifties, when my father was ranching, it was two or three years old. Now we get there at fourteen to sixteen months.” Fast food, indeed. What gets a steer from 80 to 1,100 pounds in fourteen months is tremendous quantities of corn, protein and fat supplements, and an arsenal of new drugs.
Economic efficiency indeed in its most disgusting form. I still find it hard to believe that we have managed to force feed cows corn because its faster than grass. Recommended. I've also learned much about the economics of US agriculture in the process.

Partly as a response, I have replaced all my junk food with peaches for this week's groceries. Not sure if I'm ready to replace that steak though just yet.

Saturday, June 15, 2013

Nothing is Self-Made

A while back, I wanted to write a post about how the best feeling in the world is knowing that you are "self-made." I certainly felt this way the first time I moved into my apartment in DC. It was a thrill to be in my 20s, to realize that the furniture around me was paid by my own income, and to know that my achievements were brought about by my own hard work.

I had come all the way from Manila, armed with only my college degree, from a school only a few have heard about. But somehow I landed a good job in the city.

I started feeling uneasy in the middle of composing this piece. I would never have made it in the US, if not for my dad who decided to apply to become a nurse in this country. It must have been a hard decision; he was a practicing physician back home. But this was the way to obtain green cards for the entire family. And he wanted his children to have the best opportunities, like the one I am taking advantage of now, studying in one of the leading institutions in Economics and Public Policy in the world. Now my sister is doing well at NYU on her way to being a doctor, while my younger brother is busy with an internship in the same city before he finishes up next year at Georgetown. It was because of my dad's sacrifices that we had enjoyed our successes.

That piece was never finished. It hit me, nothing is self-made.

A happy father's day to my dad.

Wednesday, June 5, 2013

Can Pegging Teachers' Salaries to Attendance Reduce Absence?

The answer seems yes and I would like to see it done in a larger scale and evaluated. This has become one of my favorite papers on education by Duflo, Hanna, and Ryan:
Many developing countries have expanded primary school access. These improvements, however, have not been accompanied by improvements in school quality. For example, in India, a nationwide survey found that 65 percent of children enrolled in grades 2 through 5 in government primary schools could not read a simple paragraph (Pratham 2006). These poor learning outcomes may be due, in part, to teacher absenteeism. Using unannounced visits to measure attendance, a nationally representative survey found that 24 percent of teachers in India were absent during school hours (Kremer et al. 2005)... 
We use a randomized experiment... In treatment schools, teachers’attendance was monitored daily using cameras, and their salaries were made a nonlinear function of attendance. Teacher absenteeism in the treatment group fell by 21 percentage points relative to the control group, and the children’s test scores increased by 0.17 standard deviations.

Monday, June 3, 2013

On Return Migration: An Anecdote is Not A Trend

An ABS-CBN news article today cites the trend of returning overseas filipino workers due to the booming Philippine economy. This is good news, except it cites *gasp* an anecdote as evidence of a trend. Hence, sample size = 1 and this is supposed to move your priors.

Note, I am actually not against Filipino migrants returning home. It may very well be happening, though there is no data at present. But I want to call out poor journalism, on a topic I am trying to specialize in, when I see one. You see, I've been watching Newsroom lately. And what I've learned is, aside from the fact that seeing Olivia Munn as economist is awesome, journalists often use misleading anecdotes, perhaps to entice emotion, but it shouldn't necessarily speak to the truth.

One of my favorite academic articles on why migrants return is Yang (2008). The author utilizes an unexpected event, the 1997 Asian Financial Crisis, when substantial and varied exchange rate shocks were realized between the Philippine peso and foreign currencies to figure out why temporary migrants return. Now, Filipino migrants work in a diverse set of countries abroad so it was as if each of them were randomly allocated different exchange rate shocks during this time. By comparing the behavior of Filipino migrants who attained greater or smaller shocks, the paper establishes how exchange rates affected the decision of these migrants to return home. The finding is that Filipino migrants are primarily driven by life-cycle considerations, prolonging their stay abroad when they experience favorable changes in their purchasing power at home (exchange rate depreciations). They return mainly because of consumption reasons and not primarily for investment.

I have an upcoming paper using the same natural experiment but this time looking at a mirror image of legal permanent residents in Australia and their motivations to return. My finding is similar, that exchange rate appreciations (unfavorable rates) encourage migrants to return, except I am able to additionally test whether this effect of exchange rate shocks simply embody other macroeconomic conditions at home, that is, is it growth rates, employment opportunities, or purchasing power that are driving return. I find evidence consistent with purchasing power as the main driver of migrant return.

Watch out for that paper. After reading this ABS-CBN article, I am even more excited to finalize it.

Sunday, June 2, 2013

The Economics of Names

The NYTimes explores the economics of names:
What’s in a name? What isn’t, these days? Baby naming has become an industry — with paid consultants, books, Web sites brimming with trend data, and academic studies exploring correlations between baby names and future success. The once-simple task of coming up with a monogram for the baby blanket has evolved into a high-stakes exercise in personal “branding.”
Surely one would think these correlations between names and success are merely correlations, right? And that names do not actually have a causal influence on economic outcomes. But you are wrong.

In their most famous study, Bertrand and Mullainathan run a field experiment where they send out fake resumes where they randomly assigned names, holding everything else constant. The question was simple: do whiter sounding names like Greg or Emily get more callbacks than names like Lakisha or Jamal, names more common for blacks. They find that white names get 50% more callbacks than black names. There are other such studies. One I like is by Carlsson and Rooth who do a similar experiment but look at the disadvantage of immigrant-sounding names in Sweden. Similarly, Swedish sounding names are 50% more likely to get a callback for a job interview than other names. It seems like there are significant economic arguments for choosing a baby's name.

Note: I do not know whether there are any economic arguments to changing your name to Metta World Peace though.