![Sample our Urban Studies journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110826/TOC-Subject-Sample-2021-Urban-Studies.jpg"})
Abstract
Ciriaci D. Does university quality influence the interregional mobility of students and graduates? The case of Italy, Regional Studies. This paper analyses the role that university quality holds, among other socio-economic factors, in the sequential migration behaviour of Italian students and graduates. Regardless of the recognition of universities as key drivers of economic development, empirical evidence on the effects of their quality on the attraction and retention of students and graduates is still limited, and mainly related to Anglo-Saxon countries. This study aims at partly filling this knowledge gap. Its results indicate that both university research and teaching quality influence migration choices and, consequently, affirm that universities are ‘supply’ tools for policy-makers, in order to counterbalance the negative effects of brain drain on regional human capital accumulation.
Ciriaci D. 大学素质是否会影响学生与毕业生的跨区域流动？意大利的案例，区域研究。本文分析大学素质以及其他社会、经济因素，在意大利学生及毕业生的连续迁徙行为中所扮演的角色。儘管大学已被认为是经济发展的关键驱动者，但有关大学素质对于吸引或留住学生及毕业生之影响的经验研究仍相当有限，且多半仅关乎英美国家的经验。本研究旨在部分填补此一知识阙如。研究结果指出，大学的研究与教学素质，皆影响了迁徙的选择，并因此确认了大学做为政策制定者制衡人才外流对区域人力资本积累所产生之负面影响的‘供给’工具。
Ciriaci D. La qualité des universités, est-ce qu'elle influe sur la mobilité interrégionale des étudiants et des diplômés? Étude de cas de l'Italie, Regional Studies. Cet article cherche à analyser le rôle que joue la qualité, parmi d'autres facteurs socio-économiques, quant au comportement migratoire séquentiel des étudiants et des diplômés italiens. Malgré que les universités soient reconnues comme forces motrices du développement économique, les preuves empiriques au sujet des effets de la qualité sur l'attraction et la rétention des étudiants et des diplômés restent limitées, et se rapportent principalement aux pays anglo-saxons. Cette étude cherche à compléter à moitié les connaissances. Les résultats laissent voir qu'à la fois la recherche et l'enseignement universitaires influencent les choix migratoires et, par conséquent, affirment que les universités constituent des outils ‘de travail’ pour les décideurs politiques, pour compenser les effets négatifs de la fuite des cerveaux sur l'accumulation du capital humain régional.
Ciriaci D. Hat die Qualität von Universitäten einen Einfluss auf die interregionale Mobilität von Studenten und Absolventen? Der Fall Italien, Regional Studies. In diesem Beitrag wird die Rolle der Qualität von Universitäten neben anderen sozioökonomischen Faktoren für das sequenzielle Migrationsverhalten von italienischen Studenten und Absolventen untersucht. Trotz der Anerkennung von Universitäten als zentralen Faktoren der Wirtschaftsentwicklung gibt es nach wie vor nur begrenzte empirische Belege über die Auswirkungen ihrer Qualität auf das Anwerben und Beibehalten von Studenten und Absolventen, wobei diese Belege zudem vornehmlich aus dem angelsächsischen Raum stammen. Diese Wissenslücke soll mit dieser Studie teilweise geschlossen werden. Aus den Ergebnissen geht hervor, dass die Qualität sowohl der Forschung als auch der Lehre von Universitäten einen Einfluss auf die Migrationsentscheidungen ausübt und dass Universitäten für Politiker somit ‘Angebotsinstrumente’ darstellen, mit denen den negativen Auswirkungen des Braindrains auf die Ansammlung von regionalem Humankapital entgegengewirkt werden kann.
Ciriaci D. ¿Influye la calidad de la universidad en la movilidad interregional de estudiantes y licenciados? El caso de Italia, Regional Studies. En este artículo se analiza el papel que desempeña la calidad de las universidades, junto a otros factores socioeconómicos, en el comportamiento de la migración secuencial de estudiantes y licenciados italianos. Aunque la universidad es reconocida como un impulsor principal del desarrollo económico, sigue habiendo pocos datos empíricos sobre los efectos de su calidad a la hora de atraer y conservar estudiantes y licenciados; además, estos datos se refieren principalmente a países anglosajones. El objetivo de este estudio es cubrir parcialmente este vacío de conocimiento. Los resultados indican que la calidad de tanto la investigación como la docencia universitarias influyen en las decisiones migratorias y, en consecuencia, confirman la hipótesis de que para los responsables políticos las universidades representan ‘herramientas de suministro’ que compensan los efectos negativos de la fuga de cerebros sobre la acumulación del capital humano en el ámbito regional.
Acknowledgments
The author is especially indebted to Alessandro Muscio for his support in gathering the VTR data. The author would also like to thank Francesco Bogliacino (European Commission, IPTS/Joint Research Centre), Michele Cincera (Solvay Brussels School of Economics and Management, Université Libre de Bruxelles), Abraham Garcia Torres (European Commission, IPTS/Joint Research Centre), Elisabetta Marinelli (European Commission, IPTS/Joint Research Centre), Andrea Nuzzi (McKinsey), Daniela Palma (ENEA), and two anonymous referees for their useful review comments on previous versions of the paper. This paper also benefited from the discussion that followed its poster presentation at the Impact of Higher University Institutions on Regional Economies Conference (Edinburgh, UK, November 2009), during a seminar at the Almalaurea Consortium (Bologna, Italy, July 2011), and from the comments received at the 2010 Italian Labour Economics Association Conference (Chieti-Pescara, Italy, September 2010). In particular, the author wishes to thank Andrea Cammelli (University of Bologna and Almalaurea), Giancarlo Gasperoni (University of Bologna), Carlo Giannella (European Commission, IPTS/Joint Research Centre), Federico Lucidi (European Commission, DG ECFIN), and Paolo Naticchioni (University of Cassino).
Notes
1. The author is grateful to both the Director of ISTAT and the Director of Laboratorio Adele for having granted access to the Safe Center where the micro-data of the latest survey on the labour market entry conditions of the Italian graduates of 2004 were elaborated. The ideas and the views expressed in this paper should not under any circumstances be regarded as stating an official position of ISTAT and/or of the European Commission. The results and any errors are entirely the responsibility of the author alone.
2. In this paper central–northern provinces include those of Lazio, Tuscany, Umbria, Marche, Emilia-Romagna, Liguria, Valle d'Aosta, Piedmont, Lombardy, Trentino-Alto Adige, Friuli-Venezia Giulia and Veneto; and southern (or Mezzogiorno) provinces include those of Campania, Abruzzi, Molise, Puglia, Basilicata, Calabria, Sicily and Sardinia.
3. Their migration choice is modelled controlling for individual characteristics, field of study, wage differentials (Kwok and Leland, Citation1982), and a number of regional push-and-pull socio-economic factors (Mixon and Hisng, Citation1994b).
4. The author is grateful to an anonymous referee for this comment.
5. Individuals travelling abroad to study or work are not included in the sample.
6. It follows that five migration paths can be identified: the non-migrant (the student/graduate who studied in their province of origin and after graduation stayed there); the return migrant (the student who moved to study and after graduation moved back to the province of origin); the repeat migrant (the student who moved to study and after graduation moved back to the macro-area of origin but to a different province with respect to that of origin); the university stayer (the student who moved to study and after graduation stayed in the province where the university was located); and, finally, the late migrant (the student who graduated in his/her province of origin and moved only after graduation) (Faggian and McCann (2009).
7. Regarding the latter choice, which could be seen as a drawback of the research, it is worth stressing that it is rooted in the aim of the study itself, that is, understanding if, notwithstanding the persistent socio-economic differences existing between central–northern and southern regions, there is still room for university quality as a cause of Italian graduates' migration choices. However, for the sake of completeness and to check for the robustness of the results, the analysis was carried out also considering a different definition of migration, that is, regional migration at the NUTS-2 level. Results are robust and in line with those obtained by analysing the south/centre–north migration flows, and are available from the author upon request. All in all, compared with the results obtained if south/centre–north flows are considered, in the case of regional migration flows the impact of research and teaching quality on ante- and post-lauream migration is statistically significantly higher.
8. If regional migration (and not migration between macro-areas) were considered, the percentages of ante- and post-lauream migrants would have been 23.0% and 23.7%, respectively.
9. Both the methodological approach and the specification strategy adopted (see the next section) allow these choices to be tackled correctly.
10. It is likely that a shorter ‘distance’ might contribute to explain why the percentage of migrants between the two macro-areas is higher for those central Italian regions located on the border. This aspect will be accounted for in the econometric analysis.
11. This definition allows for solving the problem about whether the respondent may decide to study in a different province belonging to the same macro-area of origin because there is not a university in his/her original province.
12. The employment status of the graduates was not inserted as a control given its endogeneity: it was not possible to assess precisely whether the individual's migration decision was made before or after employment was found. In any case, regional differences in employment opportunities were controlled for.
13. The migration definition used does not allow a vector of distances between the province of origin and the destination to be inserted. In fact, a zero value of the dependent variable is always associated with a zero distance between provinces belonging to the same macro-area, but a value of the dependent variable equalling ‘1’ is always associated with a distance different from zero for provinces belonging to different macro-areas.
14. Clearly, endogeneity is a problem if the wage of the individual is included as an exogenous variable. In fact, a higher wage might be a consequence of the migration choice (and of individual characteristics, university background, etc.).
15. Correlation was checked for all variables used in the analysis. The results are available from the author upon request.
16. In addition, three or more X variables can be multicollinear together without having high pairwise correlations.
17. It was decided to include these dummies instead of the number of students in each university to avoid correlation problems with the proxy used for teaching quality.
18. Enrolment of over 50% of the students interviewed by ISTAT was split across fifteen university institutions. In the sample, 36.3% of graduates attended a mega-university (that is, over 40 000 students), 48.0% attended a large university (that is, 15 000–40 000 students), 7.7% attended a medium university (that is, 10 000–15 000 students), and 8.0% were enrolled at a small university (up to 10 000 students).
19. Results without individual weight are available from the author upon request.
20. Dummy variables for large cities (Milan, Rome, Naples) were initially inserted into the model, but they showed a strong correlation with the value-added ratio, and the other regional controls (the regression's mean VIF increased from 1.82 to 9.14). As such, they were excluded from the final model to avoid multicollinearity.
21. This implies that although the level of education of a student's father does not directly affect a student's post-lauream migration choice, it does it indirectly through its effect on the first-stage migration choice.
22. The set of dummies inserted to account for a father's profession were excluded due to the high level of missing values (once inserted the number of observations decreases from 45 389 to 32 448).
23. From the year 2000 on, Italian universities changed from offering a four-year module to a 3+2 model, hence respondents who graduated from a four-year course are significantly older than those who graduated from a first-degree course as they enrolled, at the latest, mostly in 1999 (although some Italian universities have yet to introduce the reform).
24. This might be due to the effect of ‘Bologna process’ which has changed the Italian degree structure: some of the students graduating from a three-year course might have preferred to continue studying (+2) elsewhere.