Vocational versus general education: Employment and earnings over the life course in Switzerland

(2019) - Maïlys Korber, Daniel Oesch

Journal: Advances in Life Course Research
Link:: https://linkinghub.elsevier.com/retrieve/pii/S1040260818300662
DOI:: 10.1016/j.alcr.2019.03.003
Links::
Tags:: #paper #Transition #school-to-work #LifeCourse
Cite Key:: [@korberVocationalGeneralEducation2019]

Abstract

After the Great Recession, vocational training was strongly promoted across the OECD as a measure to fight youth unemployment. Vocational degrees may give workers a head start in the labour market. However, these degrees may also become obsolete sooner and leave older workers vulnerable to technological change. We thus compare the evolution of employment and earnings over the life course for holders of vocational and general education at the upper-secondary level. We use a cohort design for Switzerland, the OECD country with the highest share of youth undertaking vocational education. Based on the Swiss Labour Force Survey 1991–2014 and the Swiss Household Panel 1999–2015, our results show that employment prospects remain as good for vocational as for general education over the second half of workers’ careers. However, vocational education is associated with substantially lower earnings once workers enter their thirties, and this disadvantage is larger among women than men. While vocational degrees protect against unemployment, they come at the cost of flat earnings curves over the life course.

Notes

“After the Great Recession, vocational training was strongly promoted across the OECD as a measure to fight youth unemployment.” (Korber and Oesch, 2019, p. 1)

“ompare the evolution of employment and earnings over the life course for holders of vocational and general education at the upper-secondary level.” (Korber and Oesch, 2019, p. 1)

“ur results show that employment prospects remain as good for vocational as for general education over the second half of workers’ careers. However, vocational education is associated with substantially lower earnings once workers enter their thirties, and this disadvantage is larger among women than men” (Korber and Oesch, 2019, p. 1)

“ational degrees protect against unemployment, they come at the cost of flat earnings curves over the life course” (Korber and Oesch, 2019, p. 1) Is this an example of risk mitigation? Can people afford to risk general education for the chance at higher pay or unemployment?

“In the short run, vocational education and particularly apprenticeships have several attractive features. As employers contribute to the development of degrees and teaching curricula, the vocational skills obtained are immediately instrumental in the labour market. This helps young people to transit into the world of work and leads to lower youth unemployment (Shavit & Müller, 1998).” (Korber and Oesch, 2019, p. 1)

“may lay the foundation for a successful career. On the contrary, prolonged youth unemployment may have a scarring effect on subsequent work experiences (Gangl, 2006) – and securing a stable job may be more difficult for young people with general schooling than apprenticeships.” (Korber and Oesch, 2019, p. 1)

“everal influential studies have recently taken up this issue, notably Forster, Bol, and Van de Werfhorst (2016) and Hanushek, Schwerdt, Woessmann, and Zhang (2017). They examine the age-employment curve associated with different types of education and find a higher employment probability for VET than general education at the start of workers’ career, but a reversing pattern in later lif” (Korber and Oesch, 2019, p. 1)

“On this question, we distinguish three competing expectations: (i) cumulative advantage in favour of vocational education, (ii) reversal of fortunes in favour of general education, (iii) identical outcomes for general and vocational education once heterogeneity is taken into account.” (Korber and Oesch, 2019, p. 2)

“rk-based VET has the key advantage of integrating practice into the learning process and thus ensuring close correspondence between the acquired skills and the actual requirements of firms. Thereby, it smooths the school-to-work transition (Wolter & Ryan, 2011: 523)” (Korber and Oesch, 2019, p. 2)

“Avoiding a prolonged period of youth unemployment may be crucial for the subsequent career as an early failure in the labour market possibly triggers a mechanism of cumulative disadvantage.” (Korber and Oesch, 2019, p. 2)

“Unemployment early on in the career may leave scars on workers because it reduces their human capital, confidence and psychological readiness for work, and may thus make them less attractive to prospective employers (DiPrete & Eirich, 2006: 287). Life history data from” (Korber and Oesch, 2019, p. 2)

“A second expectation derives from the argument that vocational training provides specific skills, whereas academic schooling produces general skills (Becker, 1964). In this view, vocational training programmes that provide students with a narrow and specific set of skills should lead to a better initial match in the labour market and to higher starting wages.” (Korber and Oesch, 2019, p. 2)

“While general skills may not prepare well for entry into the labour market, they are probably more adaptive and transferable, thereby providing a stronger basis for further learning” (Korber and Oesch, 2019, p. 3)

“A third expectation does not anticipate different labour market outcomes for holders of different types of education, as opposed to different levels. According to this view, both the advantages and disadvantages of holding a vocational degree are overdone. To begin with, occupational mobility is large and vocational education does not lock workers into a single occupation” (Korber and Oesch, 2019, p. 3)

“Different labour market outcomes for vocational and general education may primarily result from two sources of heterogeneity: heterogeneity of vocational degrees and population heterogeneity, that is, the selection of different people into different educational tracks.” (Korber and Oesch, 2019, p. 3)

“first source” (Korber and Oesch, 2019, p. 3)

“(i) firm size: larger firms provide more demanding workplace learning than small businesses; (ii) sector: the public sector tends to offer more structured VET programmes than the private sector; (iii) level of intellectual requirement: requirements” (Korber and Oesch, 2019, p. 3)

“he second source of heterogeneity relates to selection effects. Young people (and their parents) choose a given educational track based on their ability, social origin and interest” (Korber and Oesch, 2019, p. 3)

“If there is strong heterogeneity among individuals who take up vocational and general education, it is possible that VET pays off for some individuals but not for others. Based on the British National Child Development Study, Dearden et al. (2002: 269) find that the wage premium from vocational qualifications is twice as high for individuals having low ability in reading and mathematics than for those having high ability.” (Korber and Oesch, 2019, p. 3)

“We impose two restrictions on the analytical sample. First, we limit our analysis of the SLFS to a pseudo-cohort of respondents born between 1954 and 1966. This specific cohort has obtained upper-sec-” (Korber and Oesch, 2019, p. 4)

“econd, we limit our analyses to individuals who have no more and no less – than upper secondary education, either general or vocational. This allows us to compare individuals where only the type, but not the level or duration of education varies.” (Korber and Oesch, 2019, p. 4)

“Our analysis focuses on individuals’ labour market outcome measured with two dependent variables: employment and work income.” (Korber and Oesch, 2019, p. 4)

“Our key independent variables are education and age” (Korber and Oesch, 2019, p. 4)

“We check for the varying quality of vocational degrees by distinguishing six types of VET based on the occupation for which respondents were initially trained: (1) agricultural and construction occupations; (2) technical and industrial production occupations; (3) technical office and computer science occupations; (4) commercial and clerical occupations; (5) occupations in private consumer services; (6) occupations in health and social services.” (Korber and Oesch, 2019, p. 4)

“As the descriptive results will show, employment and earnings do not follow a linear trend, but evolve in a polynomial pattern over the life-course. Therefore, we estimate a model with age squared, cubed and to the power of four (agei2, agei3, agei4) with the respective interactions terms between education and age: educ age i i2, educ age i i3 and educ age i” (Korber and Oesch, 2019, p. 5)

“ince age is entered as a polynomial term with several interactions, we make sense of the results by plotting in Fig. 2 the difference between vocational and general education (that is, we show the average marginal effects).” (Korber and Oesch, 2019, p. 5)

“active for other reasons (notably linked to family or health)? We introduce all our control variable into a linear probability model on unemployment and plot the difference in predictive margins between holders of vocational and general education.11 The result shows that the two educational groups do not vary in their unemployment risks over the life course (see Fig. A.1 in the appendix)” (Korber and Oesch, 2019, p. 6)

“we find a higher employment rate of holders of vocational education until the age of 26 for women and until the age of 33 for men (see Fig. A.2 in the appendix). In terms of employability, vocational education seems to have remained as instrumental for younger as older cohorts” (Korber and Oesch, 2019, p. 6)

“eady at the age of 30, men and women with no more than upper-secondary general education receive higher hourly work income than those with upper-secondary vocational training. However, both educational groups remain well beneath the earnings of workers with tertiary education.” (Korber and Oesch, 2019, p. 6)

“er the age of 30, the earnings of workers with upper-secondary general education continue to increase, whereas the earnings curve of vocationally trained workers is almost as flat as that of individuals without post-compulsory” (Korber and Oesch, 2019, p. 6)

“oling. As a result, the gap in hourly earnings widens over the life course in favour of workers with general education” (Korber and Oesch, 2019, p. 8)

“Up to now, we have treated VET as a homogeneous type of education. However, the labour market value of a vocational degree may depend on the occupation in which it was obtained” (Korber and Oesch, 2019, p. 8)

“General education is associated with higher hourly earnings than every single one of the six fields of VET that we distinguish. The first runner-up are vocational degrees in technical and industrial production occupations followed by vocational degrees in commercial and clerical occupations. By contrast, the lowest hourly earnings are paid for vocational degrees in private consumer services that include occupations such as vendors, waiters or hairdresser” (Korber and Oesch, 2019, p. 8)

“Finally, the question arises as to whether individuals with vocational training succeed in working longer hours and thereby compensate for their lower hourly earnings. We address this question by cal-” (Korber and Oesch, 2019, p. 8)

“Overall, the earnings gaps between workers in the two educational groups are similar when we use annual instead of hourly work income. However, at the beginning of careers, annual work incomes show a clearer advantage of VET for men and a smaller advantage of general education for women.” (Korber and Oesch, 2019, p. 8)

“hy would vocational education be more instrumental for men than women? Vocational education was originally devised to train craftsmen and is thus still more relevant to, and possibly more rewarding in, traditionally male- than female-dominated occupations (Murphy & Oesch, 2016)” (Korber and Oesch, 2019, p. 9)

“In policy terms, these findings caution against the overly enthusiastic endorsement of vocational education. While vocational de-” (Korber and Oesch, 2019, p. 9)

“n other words, vocational training reduces the risk of unemployment and downwards social mobility, but may also limit the likelihood of entry into the upper-middle class and upwards social mobility (Breen & Goldthorpe, 1997: 283)” (Korber and Oesch, 2019, p. 9)