The vertical axis shows the fraction of adults inside each country with a tertiary degree. The horizontal axis shows the fraction of each country’s adults who got a tertiary degree in that country and later moved to an OECD country. Each black arrow shows how one country moved between 1990 and 2000. Panel (a) shows all developing countries where data are available; panel (b) shows the same data ‘zoomed in’ on the clump of countries with low emigration rates. (The dotted line in panel [b] retains slope –1; the horizontal axis is stretched.) The data do not cover the year 2010 because currently-available estimates of skilled migration for that year are by country of birth, not country of training.
Countries should move ‘southeast’ in this graph, roughly paralleling the dotted line of slope –1, if the direct effect dominates. That would imply something close to a one-for-one tradeoff between skilled workers leaving and skilled workers inside. But the data look nothing like that. The vast majority of countries either moved to the left (skill stocks abroad fell) or they moved to the right and up, with positive slope (87 of the 100 in Figure 2). Skill stocks inside developing countries are primarily shaped by other forces, separate from the direct effect.
(graph above) shows stocks of tertiary graduates inside and outside all developing countries in 2010—this time by country of birth, not country of training. Developing countries are divided into terciles by the fraction of the adult population with a tertiary degree, where the first tercile is the lowest. On the vertical axis, the light area shows the average tertiary graduates per adult inside countries in that tercile. The dark area shows average tertiary graduates per adult outside, that is, living in an OECD country in 2010 but born in the origin country. (The fraction trained in the country of origin is much smaller.) The overall height of the bar, light and dark, thus shows the hypothetical stock of tertiary graduates that each tercile would have if—somehow—all emigrants with tertiary degrees were suddenly obliged to return to their countries of birth. The horizontal dotted line shows the average stock of tertiary graduates in advanced economies.
Again we see that skilled emigration does little to explain low stocks of human capital in developing countries. The gap in graduate density between poor countries and rich countries would be little affected even by the draconian measure of obliging 100% return of all tertiary- 5 educated people born in poor countries, even if they grew up and acquired skills abroad. A more moderate but still drastic policy of obliged return for, say, half of skilled emigrants who acquired their skills at home—a limited subset of skilled emigrants—would change homecountry human capital stocks by a small sliver of the dark bars in Figure 3. Human capital shortages in poor countries would be barely affected, even mechanically.
The most internally-valid evidence for these effects comes from natural quasi-experiments in single countries, with circumscribed external validity.
- Batista et al. (2012) find that the success of skilled emigrants from Cape Verde, as determined by exogenous economic shocks in migrant-destination countries, substantially raises demand for secondary schooling by other family members.
- Chand and Clemens (2008) find that a large and sudden skilled emigration from Fiji, due to a shock specific to one ethnic group, caused offsetting investment in tertiary education by only that ethnic group.
This evidence suggests that even very large and sudden emigration by skilled workers can, in some settings, cause human capital investment that offsets the departure.
considering the introduction of education as a selection criterion to recruit Nepalese men in the British Army. This change raised educational investments by 1.15 years and increased the average education of men living in Nepal even after allowing for emigration. Despite not being selected in the British Army or emigrating elsewhere, these non-migrants benefited directly from additional schooling.
2) Technology transfer: Skilled migrants are known to act as conduits for the transfer of new technologies to their countries of origin.
- Kerr (2008) shows that when patents filed in developing countries cite a patent filed in the United States, the patent filed in the United States is more likely to have been filed by a researcher whose ethnicity corresponds to the developing country. For example, US patents cited by researchers filing a patent in India are relatively more likely to have been filed in the US by ethnically-Indian researchers.
- Comin et al. (2012) find that person-to-person interactions have been an important channel of technology diffusion among nations over the last 140 years.
- Bahar and Rapoport (2015) show that developing countries with larger stocks of skilled emigrants in a country that produces a certain good are more likely to subsequently begin producing and exporting that good themselves. That is, skilled emigration may be involved in changing the comparative advantage of nations and the complexity of their production capabilities—which is a strong predictor of subsequent development (Hidalgo and Hausmann 2009).
- The share of foreign direct investment that developing countries receive from the United States is strongly associated with the stock of college graduates from that country present in the United States (Kugler and Rapoport 2007; Docquier and Lodigiani 2010; Javorcik et al. 2011). This relationship is stronger for high-skill migrant stocks than for low-skill migrant stocks, thus it signifies something beyond the bonds of migration in general.
- Constant and Tien (2010) find that African countries whose leaders studied abroad attract more Foreign Direct Investment. High-skill migrants also tend to remit more cash to their countries of origin than low-skill migrants (Bollard et al. 2011).
It has been known for some time that the more migrants from developing countries live in rich countries, the more trade occurs between those countries (Rauch 1999). In recent research it has become clear that this relationship, too, is stronger for high-skill migrants than for low-skill migrants (Felbermayr and Toubal 2012; Aleksynska and Peri 2014).
- An intriguing recent literature, reviewed by Ivus and Naghavi (2014), finds that migrants from developing countries are involved in the transfer of institutional norms to their countries of origin. These include informal norms such as social conventions on fertility (Beine et al. 2013). Here again there appears to be a special role for high-skill migrants in particular.
- Spilimbergo (2009) finds that developing countries with more students abroad in democratic countries tend to become more democratic themselves.
- Beine and Sekkat (2013) find a relationship between the lagged extent of high-skill emigration from developing countries and later improvements in governance quality in the country of origin, while Mercier (2013) finds that African leaders with foreign education govern more democratically.
Many beneficial interactions with the high-skill diaspora would have been extremely difficult to foresee and deliberately create. In fact, well-intended policy might have eliminated them.
- It is hard to imagine how Sudanese officials working to reduce high-skill emigration could have predicted the development effects of Mohammed Ibrahim. He quit his job as a technician at Sudan’s telecommunications administration and emigrated; several years thereafter he brought billions of dollars in investment and technology transfer to Sudan and 13 other 11 African countries with his firm Celtel.
- It is hard to imagine that Beninois officials limiting
skilled migration could have predicted the development effects of Léonard Wantchékon. He
emigrated from Benin with its top tertiary qualifications in mathematics and physics, and recently
founded the highly-regarded African School of Economics in Cotonou.
- It is hard to imagine how Indian officials condemning the “drain” of high-skill engineers could have predicted the development effects of Vivek Paul. He emigrated after becoming a highly-trained engineer, and would later return to transform the Indian firm Wipro into a multibillion-dollar global company and an engine of technology transfer to India.