The impact of worker-financed training: Evidence from early- and mid-career workers in Japan☆
Introduction
In recent years, continuing education has become an important policy issue in developed countries as the acquisition of knowledge and continual upgrading of skills has become an essential strategy for surviving and thriving in a knowledge-based economy.1 In Japan, even though a shrinking working-age population has increased the importance of a well-educated labor force, Japanese workers since the 2000s have had fewer opportunities to improve their work-related skills at the workplace.2 This is due to an ongoing structural change in the Japanese labor market associated with the decades-long economic downturn and gradual dismantling of the lifetime employment system that has led to more than one third of the labor force now working as non-regular employees under fixed-term contracts or on a part-time basis.3 However, this phenomenon is not unique to Japan and has also been observed in other OECD countries as well and, further, this increasing flexibility in the structure of the labor market has been shown to be negatively correlated with work-related skill development in many advanced economies.4 Due to these societal changes, self-financed continuing education is expected to become increasingly important in many countries.
The importance of continuing education for a contemporary labor force is already broadly recognized, with promotional policies having been in place in some countries since the 1990s. In Japan, for example, the Educational Training Benefit System (kyouiku-kunren-kyufu-seido in Japanese) was introduced in 1998 under the Employment Insurance Law to provide direct financial support to individuals for worker-financed training. This system is a co-financed voucher program that pays an employment insurance benefit to reimburse a portion of the costs incurred by individuals who take up and complete at their own expense a job training course designated by the Minister of Health, Labour, and Welfare. More recently, the Professional and Practical Educational Training Benefit (senmon-jissen-kyouiku-kunren-kyufu-kin in Japanese) was introduced in October 2014, providing an even larger subsidy to workers who undertake professional or practical training at approved educational organizations, including graduate schools.5
However, even though continuing education has become an increasingly important feature of human resource development in contemporary advanced economies, little is known about its determinants or effects because of its broad and complex nature, consisting of all learning opportunities throughout a person’s life that foster the continuous development and improvement of the knowledge and skills needed for employment and personal fulfillment. Accordingly, this study concentrates on work-related continuing education, of which there are two main types available to workers: formal or informal training provided by a firm during work hours (firm-provided training), and training initiated by workers themselves to improve job-related skills and knowledge outside of work hours at their own expense (worker-financed training). As firm-provided training has been researched extensively,6 this study focuses on worker-financed training (jiko-keihatsu in Japanese), using a unique Japanese worker dataset.
The literature on worker-financed training in Japan remains in its infancy, with the few extant studies providing only a rudimentary understanding of its determinants and effects. Kurosawa (2001), analyzing microdata from establishments in the industrial district of Kitakyushu in the southern part of Japan, finds that worker-financed training does not affect wages significantly, and Yoshida (2004), using panel data on Japanese women and the Heckman et al. (1997) difference-in-differences (DD) non-parametric matching extension procedure, finds limited evidence of a positive impact of worker-financed training.7,8 This study differs from these previous studies in that the dataset includes both male and female workers and covers all city areas nationwide. Further, we examine both the determinants and effects of worker-financed training from a broad perspective using a wide range of outcome variables: wage growth, changes in job skills, changes in job tasks, and job mobility.
For the purposes of this study, worker-financed training is defined as any work-related learning activity undertaken after completing full-time education and entering the labor market. The analysis sample consists of Japanese people aged 25–44 who are labor force participants and not unemployed.9 As worker-financed training is a form of continuing adult education, with participants worldwide and of global policy relevance, the results of this study are relevant not only to Japan but are applicable generally.
The other characteristics of the study are as follows. First, when estimating the effect of worker-financed training, we apply a growth estimation model to control for any potential unobserved individual heterogeneity, meaning that growth in wages, job skills and job tasks are taken as the dependent variables to eliminate any individual specific factor fixed over time. This approach, which is crucial for measuring the effects of job training, is explained in detail in Section 5.1. However, while this method is effective in controlling for time-invariant heterogeneity, it may not be able to control for time-variant unobserved heterogeneity. To the extent that this is true, the estimates in this study might have an upward bias, with the results best interpreted as defining the upper bound of any worker-financed training effect. However, we address this concern by controlling for the proxy variable of basic academic skills10 because it can be assumed that changes in economic outcomes caused by participation in training might be higher for an individual with high basic skills than one with low skills.
Second, the dataset employed in this study, the Survey on Work and Learning, is particularly informative relative to the literature. The Survey includes questions about educational history, work history and life events every quarter after graduation from junior high school, providing a panel structure that allows us to control for any time-invariant unobserved effects as well as information about workers prior to training.11 This pre-training information is important, as Heckman et al. (1997) point out that the DD non-parametric matching extension procedure requires that the treatment and control groups need to be constructed to be as equal as possible, and this dataset allows us to do this. Additionally, as the Survey includes information not only on wages but also on many other outcome variables such as skill development, job tasks and job changes, this allows us to provide a much more comprehensive picture of the effects of worker-financed training than the extant literature.
Lastly, in addition to analyzing the effects of worker-financed training, this study also examines its determinants, which is important from a policy perspective. Although continuing education may be important, we must work to ensure that there is an equal opportunity to partake in it, for the opportunity is likely to be strongly biased in favor of the more advantaged members of the labor market. Therefore, this study examines which workers are more/less likely to participate in worker-financed training. Further, it is well known in economics that information is an important motivating factor for a wide range of economic activities, so this study also explores the effects of information by identifying which workplace environments are likely to promote worker-financed training.
The main findings of the study are as follows: firstly, workers at small companies and temp staff are not likely to participate in worker-financed training, suggesting that those who have less access to firm-provided training are also less likely to undertake self-initiated and privately-funded work-related continuing education.12 Secondly, workers who have received either formal firm-provided training or informal guidance from their supervisors about required skills at the workplace are more likely to engage in worker-financed training, which implies that this work-related continuing education is not completely unrelated to experiences at the workplace but is affected by information received there. Thirdly, those who work long hours are least likely to participate in worker-financed training, so relaxing these time constraints could be critical for encouraging more working people to take up this activity. Fourthly, after controlling for unobserved heterogeneity, we do not find that worker-financed training leads to any statistically significant immediate increase in wages but, because it may improve job skills and job tasks, worker-financed training could lead to wage increases in the future. Finally, worker-financed training does not typically cause workers to change jobs in the short run because the skills acquired are likely to have a large general component.
The rest of the paper is organized as follows. Section 2 briefly reviews the theoretical and empirical literature and Section 3 describes the dataset. Sections 4 and 5 describe and discuss the determinants and effects of worker-financed training, and Section 6 summarizes and concludes the paper.
Section snippets
Theory
In this section, we discuss the theoretical mechanism as to why a worker might engage in worker-financed training and its effects. First, human capital theory predicts that human capital investment in the form of formal education or job training will lead to higher earnings because the increase in human capital, which roughly corresponds to knowledge and skills, will improve an individual’s productivity. This question as to whether human capital investment actually does play a role in improving
Data
This study uses data from the first and second waves of the Survey on Work and Learning conducted by The Japan Institute for Labour Policy and Training (JILPT) in 2008 and 2011,19 each wave providing cross-sectional data for analysis. The Survey was implemented using the area
Determinants of worker-financed training
This section reports the determinants of worker-financed training. As the literature on work-related training in the private sector (which is broader in scope than worker-financed training) has shown that those most likely to participate are 1) male, 2) highly-educated, 3) young, and 4) employees of large companies.29 We examined whether
Effects of worker-Financed training
Now turning to the effects of worker-financed training, as explained in Section 2, economic theory predicts that a worker who participates in training will acquire higher skills and knowledge which will, in turn, improve the range and level of the worker’s job tasks, leading to a higher wage than before training and also potentially affecting job mobility. In this section, we examine whether this theoretical expectation of worker-financed training was supported empirically by the four outcome
Conclusion
This study analyzed the determinants and effects of worker-financed training. In Japan, policies promoting worker-financed training have been in place since the late 1990s, precipitated by the need to create a ‘smarter but leaner’ workforce due to the decline in the working-age population associated with an aging society. The Educational Training Benefit introduced in 1998 and the more specialized Professional and Practical Educational Training Benefit in 2014 are voucher-based systems
References (29)
- et al.
Training subsidies and the wage returns to continuing vocational training: evidence from italian regions
Labour Econ.
(2012) - et al.
The returns to voucher-financed training on wages, employment and job tasks
Econ. Educ. Rev.
(2016) The impact of firm-provided training on productivity, wages, and transition to regular employment for workers in flexible arrangements
J. Jpn. Int. Econ.
(2014)- et al.
The impact of training vouchers on low-skilled workers
Labour Econ.
(2014) An empirical analysis of the impact on wages of workers’ self-development
Jpn. J. Labour Stud.
(2004)- et al.
Firm-provided training and temporary contracts
Spanish Econ. Rev.
(2005) - et al.
On-the-job training in europe: determinants and wage returns
Int. Labour Rev.
(2010) - et al.
Worker characteristics, job characteristics, and the receipt of on-the-job training
ILR Rev.
(1991) - et al.
Training and labour market flexibility: is there a trade-off?
British J. Ind. Relat.
(1998) Investment in human capital: a theoretical analysis
J. Pol. Econ.
(1962)
Private sector training and graduate earnings
Rev. Econ. Stat.
Temporary jobs: stepping stones or dead ends?
Econ. J.
Reexamining the returns to training: functional form, magnitude, and interpretation
J. Human Res.
Firm-specific human capital as a shared investment
Am. Econ. Rev.
Cited by (1)
Workers in the Namlea market
2019, International Journal of Scientific and Technology Research
- ☆
I would like to express my gratitude to two anonymous referees and the following people for their helpful comments: Masaru Sasaki, Akihito Toda, Keisuke Kawata, Tomohiro Machikita, Jiro Nakamura, Fumio Ohtake, Akihisa Shinozaki, Keiko Yoshida, and Daiji Kawaguchi. I also thank the participants of the 14th Labor Economics Conference and the 2011 fall meeting of the Japanese Economic Association. This research was supported by a grant from the Tokyo Center for Economic Research and Grants-in-Aid for Scientific Research (C) (grant numbers: 25380371, 16K03711) from the Japan Society for the Promotion of Science. Editorial assistance was provided by Philip MacLellan.