The growth of employer-supported adult education and its role in mitigating unequal chances to participate

• In Event: Adult education: practice, theory, and debate

Proposal

In this paper, results of a multivariate analysis of the determinants of employer-supported adult education based on 2012 data are summarized. In turn, estimates of the growth of employer-supported adult education by different types of jobs and socio-demographic characteristics of workers since the 1990s are provided to ascertain the impact of the growth on the distribution of participation over time. Overall rates of participation in adult education have increased substantially since the 1990s. Already in the 1990s, Heckman, Lochner, and Taber (1998) estimated that firm-based and other adult training accounted for over half of lifelong acquisition of human capital.
Following the theory of human capital, the continued increase may be attributed at least partly to the increasing interest by employers to invest in adult education probably due to its substantial benefits. Lerman (2015) for example, documented the positive impacts, such as industry productivity and innovation, wage increment and tax benefits of continued investment in learning. As investment grows, a key research and policy question becomes who gets employer support to participate in adult education, which is the central research question of this paper. This is a particularly relevant question since unequal chances to participate may exacerbate social inequalities of various kinds. This follows from the theory of human capital and an assumption about employers’ general behaviour of seeking benefits over costs (Becker 1964), whereby participation in employer-sponsored training is likely to be selective. For example, adults with higher levels of education tend to be perceived as more capable by employers and thus tend to selectively train them for advanced skill development to boost productivity (Ci et al. 2015). Accordingly, results are discussed in relation to cross national variations in the growth of adult learning systems and unequal chances to participate.
In terms of methodology, growth is measured as the percentage change in participation rates for most countries that took part in both the 2012 Survey of Adult Skills (also known as Programme for the International Assessment of Adult Competencies - PIAAC) and the 1994-1998 International Adult Literacy Survey (IALS). Twelve countries are included in the analysis, namely: Belgium, Czech Republic, Denmark, Finland, Ireland, Italy, Netherlands, Norway, Poland, Sweden, United Kingdom and United States. A small number of other countries participated in both studies but due to restricted data access and/or comparability concerns, these countries are excluded from the analysis (e.g. Australia, Canada, Germany).
Both IALS and PIAAC were primarily designed as international comparative assessments of literacy proficiency, which were administered to nationally representative samples of adults aged 16 to 65 (large sample sizes ranging between 2,000 to 5,000 cases per country). However, IALS was effectively the first large scale international comparative study of adult learning ever undertaken which offers an important baseline measurement of the extent and distribution of adult learning in the 1990s for a wide range of OECD countries. Similarly, PIAAC collected detailed information on a range of education and training activities undertaken by adults in the 12 months preceding the interview including formal education programmes and other non-formal learning activities such as workshops, seminars, on-the-job training as well as leisure and civic related courses. Therefore, with both datasets it is possible to empirically assess the extent of growth in adult education since the 1990s by a range of job-related and socio-demographic characteristics.
The paper presents a multivariate analysis of the determinants of employer-supported adult education based on the 2012 PIAAC data. The model builds on Boudard and Rubenson’s (2003) research examining the determinants of adult education which includes most of the same predictors and is based on the IALS data. Afterwards, changes in the probabilities of receiving employer support since the 1990s by the various characteristics included in the model are examined to ascertain changes in the distribution of employer-supported adult education considering its rapid growth.
There are two broad types of factors that are hypothesized to affect the probability of participating in employer-supported adult education, namely socio-demographic factors and job-related factors. The former reflects a range of individual characteristics, namely: gender, age, immigrant and language status, highest level of educational attainment, literacy proficiency and parents’ highest level of education. The latter reflect a range of job-related characteristics, namely, type of occupation, firm size, the frequency and variety of reading at work, and earnings. Both the individual and job-related characteristics are included in the same binary logistic regression model. The dependent variable is whether an adult participated in employer-supported adult education or not.
Results of the analysis strongly suggest that employer support for adult education is playing an important role in mitigating unequal chances to participate. These findings are contrary to what was expected. The expectation was that as the role of employers becomes more important in extending support for adult education, that this would exacerbate the unequal chances to participate in adult education. This is because employers are expected to channel more support to more trainable and efficient trainees who tend to have advantaged characteristics such as already higher levels of education and skills which follows from Becker’s (1964) theory that the decision to invest in human capital is a function of the cost/benefit ratio. Vignoles et al. (2004), for example, had found evidence to suggest that employers channel support to workers who are most likely to gain from adult education. However, the analysis presented in this article, which was conducted based on high quality and representative data across many countries with very different institutional configurations and starting points, almost conclusively demonstrates that the trend in the growth of adult education is the reverse of this expectation. This may be an indication that the cost/benefit ratio associated with employer investment in adult education is increasingly favorable even among the most disadvantaged adults and those in the lowest skilled jobs, which would be consistent with a broader trend of upskilling across the skill spectrum in a range of advanced economies.

Author

• Richard Desjardins, University of California, Los Angeles