On‐the‐job training: a skill match approach to the determinants of lifelong learning

• Published date: 01 January 2021
• Author: Michael Tåhlin,Tomas Korpi
• DOI: http://doi.org/10.1111/irj.12317
• Date: 01 January 2021

On-the-job training: a skill match approach

to the determinants of lifelong learning

Tomas Korpi Professor of Sociology and

Michael Tåhlin Professor of Sociology

ABSTRACT

A recurrent f‌inding in on-the-job training research is the ‘training gap’in formal train-

ing: the positive correlation between initial education and continuing training. This

f‌inding is here examined from the perspective of two important distinctions: (i) be-

tween employee skill supply and job skill demand and (ii) between formal and infor-

mal training. Less-educated workers may hold jobs with low skill requirements

demanding little further formal training because the use of high skills is irrelevant,

jobs that moreover provide little informal training. Exploring these issues on repre-

sentative Swedish survey data using the educational mismatch (overqualif‌ied, the

rightly qualif‌ied and the underqualif‌ied) model, we f‌ind that job requirements are

strongly correlated with the incidence of both formal and informal training. Rather

than, as has previously been argued, employee training decisions being the cause of

the gap, this suggests that employer decisions regarding how to structure jobs and

whom to hire are the primary factors behind the training gap.

1 THE INTEREST IN LIFELONG LEARNING

Lifelong learning has in recent years become a primary policy concern in many coun-

tries, as it is believed to facilitate the adaptation of workforce skills to rapidly changing

economic circumstances as well as enable the equalisation of inequalities in initial ed-

ucation. A recurrent issue in this on-the-job training (OJT) literature relates to training

gaps in lifelong learning, differences in the incidence of formal workplace training be-

tween various groups. One dimension where such discrepancies have been identif‌ied

relates to individuals with different levels of initial education, with less educated re-

ceiving less further training [refer to e.g. Organisation for Economic Co-operation

and Development (OECD) 1999, 2003, 2012; European Commission, 2015]. This ed-

ucational training gap is pervasive and casts doubt on the possibilities of achieving hu-

man capital compensation over the life course of individual workers.

A straightforward explanation for the training gap is that less educated workers

may hold jobs with low skill requirements, jobs demanding little further formal train-

ing because the use of high skills is irrelevant. Moreover, these jobs may be supplying

little informal training as work task involves little learning. Although training supply

and demand have been issues in the training gap literature, attempts to separate them

❒Tomas Korpi, Swedish Institute for Social Research (SOFI), Stockholm University, Stockholm 10691,

Sweden and Michael Tåhlin, Swedish Institute for Social Research (SOFI), Stockholm University,

Stockholm 10691, Sweden. Correspondence should be addressed to: Tomas Korpi, Swedish Institute for

Social Research (SOFI), Stockholm University, 10691 Stockholm, Sweden; email: tomas.korpi@sof‌i.su.se

Industrial Relations Journal 52:1, 64–81

ISSN 0019-8692

© 2021 The Authors. Industrial Relations Journal published by Brian Towers (BRITOW) and John Wiley & Sons Ltd.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License,

which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial

and no modif‌ications or adaptations are made.

are rare and have relied on questionable data. This calls for additional analyses as the

question is of major policy relevance: successfully furthering lifelong learning among

underserviced groups depends on disentangling the two factors.

The purpose of this paper is to extend the analysis of the training gap by pursuing

the distinction between the supply and the demand for skills using a model in which

further education and training is dependent on the match between existing educa-

tional qualif‌ications and the educational requirements of jobs. More specif‌ically, we

study training using the so-called overqualif‌ied, the rightly qualif‌ied, and the

underqualif‌ied (ORU) model (Duncan and Hoffman, 1981) in which the employee’s

educational qualif‌ications are related to the educational requirements of the job. This

juxtaposition generates three categories: the overqualif‌ied, the rightly qualif‌ied and

the underqualif‌ied. The relationship between these three match categories and wages

has been extensively studied, with remarkably consistent results across both countries

and time periods (Rubb, 2003). We extend this well-established line of research by re-

placing wages as outcome measure with workplace training.

We conduct these analyses for both formal and informal workplace training. Most

research on the topic of lifelong learning has focused on formal training, and reviews

have lamented the fact that there still are large gaps in our knowledge with regard to,

for example, the relationship between informal and formal training as well as their

impact on labour market outcomes. Comparisons of the two forms of training may

also be informative regarding the causes of the training gap. Explanations of the for-

mal training gap have for instance posited that greater training participation among

higher educated workers may be related to differences in workers’ability to pay for

or in their expected returns to training. Although these seem plausible reasons for dif-

ferences in formal training, they would appear less relevant with regard to informal

training as this rarely is associated with any direct costs. Furthermore, uncertainties

regarding aptitude for and potential returns from learning would seem less of an issue

when training is obtained while carrying out assigned work tasks. A training gap in

formal training combined with the absence of a gap in informal training would there-

fore suggest that the causes for the gap in formal training lie with the employee. In

contrast, similar training gaps for both formal and informal training would question

these employee-based explanations.

These analyses of formal and informal training may therefore further our under-

standing of the training gap by providing indirect evidence on the relevance of previ-

ously suggested causes, despite the fact that we do not perform a directly causal

analysis. The latter would necessitate modelling, for example, employer choices re-

garding production technology and job content, something beyond the current study.

In line with most ORU research, the analyses instead consist of standard ordinary

least squares (OLS) regression.

The next section, Section 2, provides a brief review of the theoretical and empirical

training literature, highlighting the limited attention given to both job requirements

and informal training. Section 3 presents the data, drawn from the Swedish Level of

Living Survey (LNU). Section 4 discusses the methods used while Section 5 contains

the results. The f‌inal section, Section 6, relates the results to ongoing policy debates.

2 MODELS AND MEASURES OF ON-THE-JOB TRAINING

Analytically, it is possible to distinguish between many different forms of OJT. The

primary distinction is between formal and informal training, where formal refers to

65On-the-job training

© 2021 The Authors. Industrial Relations Journal published by Brian Towers (BRITOW) and John Wiley & Sons Ltd.