![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
This article assesses the impact of human capital endowments on international competitiveness in Europe, with special reference to transition economies in Central and Eastern Europe (CEE). The analysis uses longitudinal data for twenty-seven European economies over the period 1995–2010. In line with orthodox theory, a positive relationship is found between the labor force’s level of educational attainment and competitiveness. While in the European Economic Area (EEA17), tertiary education is the only significant education-based determinant of the export market share, in CEECs; both the shares of the workforce with secondary and tertiary education are significant with the former having a greater impact. Some evidence is found for the hypothesized impact of the quality of education.
Acknowledgments
We thank the anonymous referees and the editor for their helpful comments and suggestions.
Notes
1. Given the restricted availability of data for certain indicators, the empirical assessment is limited to selected transition economies: Bulgaria, the Czech Republic, Estonia, Hungary, Lithuania, Latvia, Poland, Romania, the Slovak Republic, and Slovenia.
2. Although the transition process has been declared to be complete by the World Bank (Citation2008) for all the Central and Eastern European countries in the sample, these are still regarded in the literature as transition economies.
3. This period was also associated with an authoritarian administration of education institutions—a strictly centralized traditional curriculum with no emphasis on creative judgment and problem-solving skills and very limited monitoring of learning outcomes (OECD Citation2011).
4. PISA—Programme for International Student Assessment.
5. TIMSS—Trends in International Mathematics and Science Study.
6. PIRLS—Progress in International Reading Literacy Study.
7. As a robustness check, an alternative regression analysis employing this measure was conducted, these results are available on request. The results for the variables of interest were very similar.
8. In addition to the export market share, a modified version of Balassa’s (1965) revealed comparative advantage index (RCA) was considered. However, given the various criticisms of its inability to capture fully the theoretical concept of competitiveness, as well as its questionable statistical features, it was not used in the estimations.
9. The following tests were used by Hanushek and Woessmann (Citation2012) in the calculation of the indicator: First International Mathematics Study (FIMS); First International Science Study (FISS); Second International Mathematics Study (SIMS); Second International Science Study (SISS); Third International Mathematics and Science Study (TIMSS); and Programme for International Student Assessment (PISA).
10. Even though the quality of the human capital of the adult labor force has started to become part of various International Adult Literacy surveys, their time span and country coverage are still very limited, making it unsuitable for this analysis.
11. Education quality dataset v2.2, available at https://sites.google.com/site/nadiraltinok/home/datasets.
12. Data for some of the countries in the sample are missing for some years/intervals, thus making interpolation inappropriate.
13. Research and development expenditure (% GDP) and patent grants could not be used in the estimations due to the large amount of missing data.
14. While there is a correlation of 0.72 between the economic freedom index and the transition indicator in the CEECs sample, this did not affect the final estimation results. As Wooldridge (Citation2009) points out, if the degree of correlation between any control variables does not affect (i.e., is not correlated with the variables of interest), the partial effects of the latter can be determined without any difficulties.
15. It also covers the imputed bank service charges, import duties, and any statistical discrepancies noted by national compilers as well as discrepancies arising from rescaling (WDI—World Bank Citation2014).
16. Albania, Croatia, Bosnia and Herzegovina, Kosovo, Macedonia, Montenegro, and Serbia are excluded from the investigation, due to lack of data.
17. The impact of the quality of education proxied by three different measures is assessed in alternative model specifications.
18. The Chow tests were based on fixed-effects estimations. Models were run on a pooled sample of countries, including the explanatory variable set as well as their interactions with the transition dummy. The joint significance of the interaction terms is then tested by an F-test. See in the appendix section.
19. Although this approach has been used in many empirical analyses, it has also been criticized. Greene (Citation2011) argues that the new method is the same as the LSDV estimator and that there are no apparent efficiency gains. Moreover, he has strongly criticized step 3 of the procedure by arguing that it produces very small standard errors, and therefore, it should not be carried out. However, the authors of FEVD have addressed the issue of very small standard errors in their updated Stata ado file. According to Breusch et al. (Citation2011), if there is an indication of potential endogeneity—with time-invariant variables being correlated with the unobserved effects—the FEVD estimator will be inconsistent.
20. The estimation results for the time-varying variables are generally consistent across the estimators, with the exception of FEVD estimates being statistically insignificant.
21. While the magnitude of the effects of the two variables seems quite different, it should be noted that these are measured differently. Cskills is a standardized index with the minimum of 4.54 and maximum of 5.19, whereas meanscores reflects mean test scores in the range of 507 to 608.