Selected Examples of Emigration Patterns. 

Context in source publication

Context 1

... course, another reason of choosing Field 4 less frequently might be that it leads to a higher risk of over-education (hence, to lower average wages of this field’s graduates). This is indeed the case, as reported in Table 7. Note also that Field 3 has the lowest probability of having to move to an unqualified position after graduating it, but this does not increase its popularity. Finally, we consider effects of field and profession popularity among friends on the choice of field of study. For this purpose, we run a multinomial logistic regression, the dependent variable being four fields of study (see Table 8). Firstly, note a quite high pseudo R-squared of 0.34 coming exclusively from the popularity of the field or the corresponding profession (these being the only regressors, we did not take into account friends’ labour market characteristics). Field popularity among friends increases the probability of choosing Fields 1 and 2 and decreases that for Fields 3 and 4. The popularity of the corresponding profession among friends increases the likelihood of choosing Field 3, decreases that of choosing Field 4, but has a quite minor influence on Fields 1 and 2. Figure 5 presents examples of emigration patterns encountered in the simulation. In vast majority of cases, there is a single leading target country for emigrants, which attracts around 60 to 80 per cent or all emigrants (see the left panel). Rarely there are two target countries, each attracting 40 to 45 per cent of migrants (the middle-left panel) or three target countries with similar chances to attract migrants from the given home country (the middle-right panel). The right panel shows the typical pattern encountered when the target country is selected only based on its relative average wage. Comparing these graphs, one can conclude that the use of the complex criterion, when social network aspects influence decision-making, brings more structure to emigration patterns. Table 9 shows who tends to emigrate from the native land. While the chance of finding an unskilled job are much lower than that of finding a skilled job, the unskilled are the least likely to emigrate, due to lower probability of making the emigration decision. Graduates in Field 3 are also quite unlikely to emigrate, this time because they have lower requirements for being admitted to work and, hence, higher chances to find one. Quite surprisingly, we find that graduates in Field 4 are less likely to emigrate than graduates in Field 1 or Field 2 (the differences in means are statistically significant). In this paper, we study a model that allows to assess the macro-dynamics emerging exclusively through (1) interactions within social networks and (2) the quality of match between the agent’s characteristics and that required by profession. Overall, our model gives plausible results about education and labour markets. This, however, does not mean that all results are as expected. For instance, surprising results were found among effects of popularity factors on the choice of field of study, where effect directions differ by field, or in emigration, where fewer emigrants are educated in a more difficult field of study than expected. Statistics of agent behaviour in choosing field of study and succeeding in the labour market was very similar in all four countries, which was expected. Nevertheless, it was shown that social networks play a significant role in shaping emigration patterns. The current work might be extended in several directions. Firstly, heterogeneous countries might be introduced. Secondly, one could analyse the explicit and implicit requirements for jobs in different (broad) professions to make the -characteristics and the requirement matrix more ...