FRIEDRICH-EBERT-STIFTUNG 34 in the medium and long term and may affect the stock of human capital in the migrants’ country of origin, including its potential. The term most commonly used to describe this phenomenon is»brain drain«, but the meaning of this term is far from obvious. It is derived from an approach developed in the 1960s and 1970s, which focused on the negative effects of the outflow of highly educated people, including the lost fiscal outlay on the education of(future) migrants or lost productivity(Grubel and Scott 1966; Bhagwati and Hamada 1974). This»traditional approach« was predominantly pessimistic and sought solutions to stop the outflow of well-­educated people from less developed countries. However, in the 1990s a new approach sprang up that substantially modified previous paradigms. The so-called new economics of brain drain rests on the pretty obvious assump­tion that migration may be seen through probabilistic lenses (migration as a probabilistic event); that is, the members of a given population may exercise the option, but they are not compelled to. In simple terms, it is assumed that the migra­tion option is available but that its probability is usually lower than 1. If we also assume that in specific circumstances the possibility of going abroad may induce people to invest more in human capital(expecting a higher return from human capital utilised/employed abroad), paradoxically, even large­scale migration of highly educated people may lead to an increase in the human resource capital in the country of ori­gin. This phenomenon is dubbed in the literature brain gain, or beneficial brain drain(Stark 2005; Mountford 1997; Beine et al. 2001) and clearly exemplifies the»modern approach« to the mobility of highly skilled people. One of the most convincing theoretical models that explains the possible effects of highly skilled mobility has been proposed by Beine et al.(2001). The model assumes that those who migrate are mostly educated and if so, that migration chances depend strongly on education(in the basic model not on real abilities or competencies; in other words, there is no self-selection based on unobservable characteris­tics). Furthermore, it is assumed that human capital is trans­ferable and there are higher returns to human capital abroad than in the country of origin. If so, from the theoretical model it follows that the share of well-educated people in a given society will depend on migration prospects and economic growth will depend on the share of well-educated people(in a positive way) and migration(negatively). Following this line of reasoning, it is possible to distinguish two effects related to short- and long-term outcomes of the migration process under consideration. First, there is a traditional»drain effect« (ex post or static effect) related simply to the outflow of hu­man capital. Second, there is a novel effect –»brain effect« (ex-ante or dynamic effect) attributable to the increase in human capital stock as an outcome of migration prospects, with a critical role of incentives to learn. Finally, beneficial brain drain(BBD) emerges when the brain effect dominates, which remains a purely empirical issue. Even if the brain gain theories are appealing, the findings of empirical studies looking at the impacts of highly skilled mobility are far from unambiguous. This is mainly due to the complex transmission mechanism between migration pros­pects and incentives to learn and to the time dimension of the story(ex-ante and ex-post effects). Several studies have assessed the brain drain/brain gain issue at the macro level. Docquier and Rapoport(2008) used a cross-section of 127 developing countries and found that migration prospects positively and significantly impact the stock of human capital. Beine et al.(2008) attempted to assess the short-run net effects of brain drain. According to their analysis, brain drain has decreased the level of human capital in around 53 per cent of countries in the sample and those were mainly small and medium sized countries with extremely high rates of highly skilled emigration(over 50 per cent). The evidence of brain gain(or beneficial brain drain) was found in countries with well-educated emigration rates of around 20 per cent. In a recent paper, Beine et al. (2011) used panel data to show that the emigration of skilled persons exerts a positive impact on human capital formation (based on panel data analysis). Similar to previously quoted studies the overall effect depends strongly on the size of a country and the highly skilled emigration rate. To understand better the mechanisms underlying the relationship between highly skilled mobility and human capital formation it is necessary, however, to refer to micro level studies. Kangasniemi, Winters and Commander(2007) analysed a cohort of Indian doctors who were practicing abroad, finding that around 30 per cent of the doctors surveyed reported that the prospects of migration affected their level of effort during their studies(clear sign of migration-­driven incentives to obtain more human capital). Commander et al.(2008) assessed the IT sector in India, pointing to positive externalities related to well-educated mobility(this sector is commonly presented as a virtuous circle of migration and development). Lucas(2004) confirmed the close links between migration prospects and educational choices by young Filipino students. Last but not least, Gibson and McKenzie(2010) designed a large-scale survey among the best students and their teachers in four less developed countries(Ghana, Micronesia, Papua New Guinea and Tonga) and New Zealand. They found that the proportion of students who made a particular effort with regard to their education due to the prospect of migrating abroad varied from 8 per cent in the case of New Zealand to over 30 per cent in the case of Ghana. At the same time, between 6 and 35 per cent of teachers reported that they taught different things as a result of expectations that some students will go abroad. This kind of outcome strongly supports the brain gain hypothesis. On the other hand, a growing number of papers emphasise other effects. McKenzie and Rapoport(2008) suggested that, paradoxically, a reduction in educational attainment in the areas characterised by higher emigration rates is possible as a consequence of very low(or even zero) returns to human capital in the destination countries. This effect is commonly termed»brain waste«(Mattoo et al. 2005; Kaczmarczyk and Okólski 2008). On the other hand, discussing the short-term effects of post-accession migration, Kaczmarczyk and Okólski (2008) challenged the concept of brain drain and suggested the term»brain overflow«, instead(see below). All in all, recent empirical evidence tends to suggest that the scope for a beneficial brain drain is diminishing substantially(Egger and Felbermayr 2007). Following this line of reasoning and considering the con­troversies attached to the term»brain drain«, while referring