Printer Friendly

Participation in workplace employer-sponsored training in Canada: role of firm characteristics and worker attributes.

I. INTRODUCTION

To be successful in the highly innovative and internationally competitive knowledge-based global economy, Canada must produce, attract, retain, and upgrade the well-educated labor force. In addition to producing new graduates and attracting skilled immigrants, renewing and upgrading skills of the existing labor force remain one of the most challenging and important tasks. Employer-sponsored training is one important vehicle for skills upgrading.

On the one hand, employer-sponsored training in Canada has been falling short of international standards (Government of Canada 2002a, p. 59) but is increasingly demanded across industries (Government of Canada 2002b, p. 41). This is of particular importance considering the Canadian aging population and smaller future cohorts of new workers who would enter the labor force in the years and decades to come.

On the other hand, as illustrated in this work, international evidence indicates that increased market competition, organizational changes, research and development, and technological innovation have raised the demand for job-related training in the United States. But the empirical evidence for Canada is quite limited. Many existing studies on employer-sponsored training are primarily based on household-based surveys (such as the Adult Education and Training Survey [AETS] for Canada) where the information on firm characteristics is not as rich as that in firm-based surveys (such as the Workplace and Employee Survey [WES] for Canada).

Lin and Tremblay (2003) note that many existing Canadian studies have examined employer-sponsored training in programs and courses from the perspective of households but few studies have examined directly workplace job-related classroom and on-the-job training from the perspective of firms. Many studies have examined the relationship between worker attributes and participation in employer-sponsored training based on surveys that contain limited information on firm characteristics (e.g., firm size, industry, and union status) but few studies have examined the role of other critical firm characteristics such as market competition, research and development, technological innovation, and management practices. The WES data link these firm characteristics to their workers' attributes and record workplace classroom and on-the-job training. Therefore, the WES data enable us to better understand workplace training.

This work adds to the literature in the following ways. First, we attempt to evaluate the role of firms' training provision in workers' participation. We find that when firms provide more training, their workers tend to participate more in workplace training. This finding has an important implication to firms and their training decisions. Second, we try to examine how workers' participation is correlated with changes in market competition, organizational changes, and technological innovation. The new evidence from the WES data indicates that changes in market competition, organizational changes, and technological innovation affect workers' participation in workplace training. This finding explains in part why workers in some firms participate more in workplace training than those in other firms. These new findings suggest that there is a strong and direct relationship between those important firm characteristics and workplace training.

The remainder of the work proceeds as follows. In Section II, we review the existing literature and state our key hypotheses about workplace training participation. In Section III, we describe the WES data and highlight some observations based on the statistical analysis of workplace training participation with reference to all of the firm characteristics and worker attributes. In Section IV, we use the econometric models to analyze workplace training participation by taking into consideration all firm characteristics and worker attributes so that we can identify and interpret the net marginal impact of each of these determinants on workplace training participation. The work closes with some concluding remarks in Section V.

II. EMPLOYER-SPONSORED TRAINING: WHAT WE DO AND DO NOT KNOW

Generally, there are three main interdependent components of human capital--early ability (whether acquired or innate); qualifications and knowledge acquired through formal education; and skills, competencies and expertise acquired through training on the job. All of these components are essential for productive capacities. However, the provision and utilization of employer-sponsored training are dependent upon the rational decisions of both the firms and their workers in question. (1) It is possible that firms consider job-related training beneficial and hence offer training to workers but workers may or may not participate, or that workers believe job-related training beneficial but firms may or may not offer it to workers. Observed data on workplace training reflect rational choices made by both firms and their workers.

The labor economics literature recognizes the necessity for firms to offer or sponsor job-related training for various reasons. Javanovic (1979) notes that job-matching difficulties in the labor market lead to a high turnover of workers. Barron, Black, and Loewenstein (1989) show that it is the process of job matching in a heterogeneous labor market that explains the necessity of job-related training. Stevens (1994) also identifies a natural link between training and labor market imperfection.

The literature also recognizes the economic implications of job-related training to firms and their workers. Loewenstein and Spletzer (1998) analyze how firms and workers share both costs of and returns on general training, and how the general training financed by previous employers has a larger wage effect than the general training financed by the current employer. Ace-moglu and Pischke (1998, 1999a, 1999b) note that firms provide general training in addition to specific training and that workers, including those with minimum wages, finance their own general training. Acemoglu and Pischke (1999c, 2000) find that the compressed wage structure may motivate firms to finance general training and that it is the rent collected by firms via monopsony that motivates general training for workers. According to Audor (2001), training helps firms to attract higher ability workers and lower wages of the workers who are trained at workplaces. Diaz-Vazquez and Snower (2003) propose a theory showing that employer-sponsored training influences firing costs that would occur to firms in question.

Given the necessity of providing job-related training on the part of firms, does training provision encourage workers" participation? Altonji and Spletzer (1991), Hui and Smith (2004), and Lillard and Tan (1992) find that employer sponsorship may help workers' involvement in job-related training. However, Barron, Black, and Loewenstein (1987) and Lynch (1992) note that the evidence on the role of employer sponsorship is not strong. Hence it is critical to know more about the role of training provision among Canadian firms. For this purpose, we wish to test an important hypothesis: Does firms' provision of workplace training encourage workers' participation? The WES data contain the information on firm characteristics including firms' training provision. As such, we were able to test this hypothesis using the WES data.

In general, if firms can gain net benefits from offering training to their workers, they will offer it. The net benefits can result from employees' greater capacity in dealing with increased market competition, organizational changes, and technological innovation. Hence, these challenges may be important drivers for more job-related training provision and participation at workplaces. Knoke and Kalleberg (1994) note that market competition pressures are a non-trivial factor for firms to train their workers at workplaces. Bartel and Lichtenberg (1987) find that the rapid technological change causes firms to provide more training to production workers. Mincer (1989) notes that employer-sponsored training becomes increasingly important as an economy becomes more knowledge based. Bre-snahan, Brynjolfsson, and Hitt (2002) find that information technology, complementary workplace reorganization, and new products and services constitute a significant skill-biased technological change affecting labor demand and hence, employer-sponsored training. These findings are confirmed by OECD (2003). But the empirical evidence directly from Canadian workplaces is quite limited. Therefore, we wish to test another important hypothesis: How do changes in market competition, organizational changes, and technological innovation affect workers' participation in workplace training in Canada?

To test the above two hypotheses, we also need to take into consideration other plausible factors that are important to workers' training participation. Existing empirical studies suggest that employer-sponsored training can also be affected by the following worker attributes and firm characteristics: age, gender, marital status, presence of pre-school children, schooling/education, job status, occupation, job tenure, income, industry, firm size, union membership, and region (provinces and metro centers).

On the role of worker attributes. Blinder and Weiss (1976), Weiss (1986), and Polachek and Siebert (1993) note that older workers rake less benefits from investment in human capital and hence participate less in training. Heckman and Smith (1999) find that adult female workers in the United States obtain less training. Holtman and Idson (1991) show that marital status in the United States is a significant factor influencing workers' participation in job-related training. Greenhalgh and Stewart (1987) find that the presence of children affects workers' participation in job-related training in the United Kingdom. Brown (1990), Lillard and Tan (1992), Lynch (1992), Barnow, Giannarelli, and Long (1996), Barron, Berger, and Black (1997), Betcherman, Leckie, and McMullen (1998), Lynch and Black (1998), Holzer and Reaser (1999), and OECD (2003) show that adults with higher education attainment participate more in adult training than those with lower education attainment. (2) Hui and Smith (2004) find that white collar workers tend to get more training in general. Simpson (1984) and Bishop (1991) note that workers with longer job tenure receive more training, although Hui and Smith (2004) find weaker evidence on this in Canada. Lillard and Tan (1992) find that disadvantaged groups such as low-income, nonwhite, and part-time workers have lower training incidence.

On the role of firm characteristics, Lillard and Tan (1992) and Turcotte, Leonard, and Mont-marquette (2002) find that patterns of training vary across industries in both the United States and Canada. (3) Barron, Black, and Loewenstein (1987), Holtmann and Idson (1991), Barron, Berger, and Black (1997), Betcherman, Leckie, and McMullen (1998), Lynch and Black (1998), and Holzer and Reaser (1999) show that smaller firms offer less job-related training in the United States and Simpson (1984), Jennings (1996), Lin and Tremblay (2003), and Hui and Smith (2004) observe the same for Canada. Mincer (1983) finds that union membership reduces training incidence in the United States but Lynch (1992) and Lillard and Tan (1992) find that U.S. unionized workers are more likely to participate in apprenticeship and on-the-job training. Dustmann and Schonberg (2004) find that union members receive more on-the-job training in Germany. Whereas Simpson (1984) finds that union membership does not affect training incidence and durations in Canada, Hui and Smith (2004) find that Canadian union members have lower training incidence and that no patterns can be deciphered in terms of training duration.

The WES data are particularly useful for the purpose of our hypothesis testing relative to the other household-based survey data (e.g., the AETS data) because the WES data contain more detailed information on firm characteristics. The two hypotheses (the role of training provision and that of firm structural characteristics) can be readily tested based on the WES data.

III. DATA SOURCE

A. The WES and Key Variables

The WES is a firm-based survey conducted by Statistics Canada, which has two target populations, firms and their workers. The firm population comprises paid workers from all business locations in Canada. The worker population is derived from the Canada Customs and Revenue Agency T-4 supplementary forms of employees working in the selected business locations. The 1999 WES provides the data for 6322 firms and 23,540 workers. The 2001 WES has the data for 6,223 firms and 20,377 workers. In our statistical analysis, we use the final sampling weights that account for both the multilevel sampling procedure and nonresponses.

The WES data exclude private households; religious organizations; employers in public administration; and employers in crop production, animal production, fishing, hunting, and trapping. It also excludes Nunavut, Yukon, and Northwest Territories.

The use of the WES data has an advantage over the AETS data on which a large part of the existing Canadian literature is based. The WES data link firm characteristics directly to worker attributes in the sampling process so that researchers can analyze jointly the role of firm characteristics and that of worker attributes. In particular, we were able to examine the role of firms' training provision and that of changes in market competition, organizational changes, and technological innovation.

In the WES data, workplace job-related training takes two different forms: classroom training and on-the-job training. The data contain not only worker attributes such as age, gender, marital status, presence of pre-school children, education attainment, and so on, but also richer firm characteristics which include, in addition to the usual firm characteristics (firm size, industry, and union status), changes in market competition, organizational changes, and technological innovation. See Table Al for the variable definitions and Table A2 for the participation rates of employer-sponsored training across various covariates.

B. Basic Statistics

The overall participation rates of employer-sponsored training in 1999 and 2001 are 54.8% and 53.8%, respectively. Provincial variations are large. Quebec has the lowest rates in both years (46.6% in 1999 and 47.6% in 2001) and British Columbia the second lowest participation rates in both years (50.9% in 1999 and 47.6%- in 2001). It is helpful to consider these Canadian data in some international context. According to Lerman, McKernan, and Riegg (2004), the participation rate for informal workplace training in the United States is about 95% of workers in establishments with 50 or more employees based on the 1995 Survey of Employer-Provided Training (SEPT). The same survey shows that 70% of workers in the U.S. establishments with 50 or more employees received formal employer-provided training. The 1995 National Household Education Survey (NHES) found that the incidence rate in the United States is about 37% among members of households rather than workers at workplaces.

In the WES data, workplace training participation does not vary much by gender, marital status, and the presence of pre-school children. However, as will be noted later, the presence A. of pre-school children affects workers' participation in workplace training when all other determinants are controlled for. Better-educated workers have higher participation rates. As noted later, when all other determinants are con-trolled for, less schooling is associated with a higher tendency to participate in workplace training. Full-time workers tend to participate more in workplace training than their part-time counterparts do. By occupation, professionals and managers have higher participation rates, whereas production workers and marketing and sales employees have lower participation rates.

By industry, workers in labor-intensive tertiary manufacturing, real estate and rental and leasing services, construction, and retail trade and consumer services participate less in workplace training, whereas those in finance and insurance, and communication and other utilities participate more. Workers employed in larger firms participate more in workplace training. Unionized workers also participate more. Employees with higher incomes participate more in workplace training than those with lower incomes do. However, as will be shown later, income loses its significance when all other determinants are controlled for.

The WES data provide insights into how the business environment affects workplace training participation. More specifically, workplace training participation is positively correlated with skill requirements. Workers facing higher skill requirements have much higher training participation rates. Workers in firms with organizational changes have higher participation rates. These changes take various forms in relation to the knowledge-based economy: greater integration, reduction in managerial levels, greater research and development, collaboration, reengineering, and adoption of flexible working hours.

More market competition is correlated with more workplace training. Workers in firms competing with firms beyond local markets or with internationally owned firms have more workplace training. Similarly, workers with firms recognizing these competitors have more workplace training.

However informative these observations from the basic statistics may be, they are obtained by examining each determinant in isolation. To identify the net marginal impact of each of firm characteristics and worker attributes, we now turn to econometric modeling where impacts of all determinants are properly controlled for.

IV. DETERMINATION OF WORKPLACE JOB-RELATED TRAINING

A. Model and Variable Specification

We are interested in the determinants of workplace job-related training. As participation decision can be characterized by a binary variable (taking the value of 1 if a worker participates and the value of 0 otherwise), we used the logit model to analyze training participation. We present the model so that the log of odds ratio is expressed as a linear function of worker attributes and firm characteristics. The log of the odds ratio is a monotonic function of the odds ratio, which, in turn, is a monotonic function of the probability of training participation. This approach permits a straightforward interpretation of the slope parameter estimates in terms of odds ratios. That is, a slope parameter estimate is presented as the estimated net marginal impact of a change in an explanatory variable on the odds ratio for training participation.

The explanatory variables in the WES data are province, metropolitan center, age, gender, marital status, presence of pre-school children, education, job status (full-versus part-time), industry, occupation, job tenure, firm size, union status, and income. These variables were also analyzed by the existing literature based on the AETS data. However, the WES data provide additional important firm characteristics such as technological complexity, amount of training, availability of training, skill requirements, human resource practices, various forms of organizational changes, innovation, and market competition.

Interpretation of an explanatory variable's contribution to the odds ratio should be made with reference to the baseline case, which is specified in Table 1. When the contribution to the odds ratio is equal to 1, (4) there is no impact from a change in the associated explanatory variable. When the contribution to the odds ratio is greater (less) than I, the impact from a change in the associated explanatory variable is positive (negative). The more the value of the contribution deviates from 1, the greater the contribution to the odds ratio will be.
TABLE 1

Participation Models: WES Data

WES 1999 2001

Logistic regression No. of 20662 No. of obs 19398
models obs.s.

 LR 703.32 LR 664.35
 ch[i.sup.2] ch[i.sup.2]
 (42) (40)

 Prate > 0.00 Prob > 0.00
 ch[i.sup.2] ch[i.sup.2]
 Pseudo 0.11 Pseudo 0.12
 [R.sup.2] [R.sup.2]

 Log -12577.54 Log -11693.12
 likelihood likelihood

Independent
variable
Participation

 Odds ratio p-value Odds ratio p-value

Atlantic Canada 1.42 .00

Quebec 0.82 .02

Ontario 1.23 .01 1.40 .00

Alberta 1.54 .00 1.20 .10

Manitoba 1.30 .05

Saskatchewan 1.35 .07 1.28 .10

British Columbia Baseline Baseline

Age 0.97 .00 0.97 .00

Male 0.80 .00

Pre-school children 0.79 .01 0.80 .08

Grade 0-10 3.22 .16

Grade 11-13 1.28 .01 1.38 .00

Some PS, PS 1.21 .00 1.17 .05
certilicate/diploma

University Baseline Baseline

Full-time 1.58 .00

Managers 1.19 .13 1.35 .03

Professionals 1.40 .00 1.57 .00

Technical/trades 1.24 .02

Production worker Baseline Baseline

Tenure 1.00 .08

Union member 1.17 .06

Wage 1.00 .08

Labor-intensive 0.48 .00 0.36 .00
tertiary
manufacturing

Primary product 0.72 .00 0.55 .00
manufacturing

Secondary product 0.61 .00
manufacturing

Capital intensive 0.66 .00
tertiary
manufacturing

Construction 0.77 .05 0.81 .15

Transportation, 0.78 .04
warehousing-
wholesale

Communication and 2.01 .00
other utilities

Retail trade and 0.81 .05 0.77 .06
consumer services

Finance and 1.88 .00 1.58 .01
insurance

Real estate, rental 0.63 .01
and leasing
operations

Information and 0.76 .04 0.68 .01
cultural industries

Natural resources Baseline Baseline

1-19 Employees 0.63 .00 0.69 .00

20-99 Employees 0.88 .13 0.86 .06

100-499 Employees

500 Employees or Baseline Baseline
more

Tech complexity 1.29 .00 1.27 .00
high

Tech complexity
equal

Tech complexity low Baseline Baseline

Training time high

Training time equal 1.22 .00 1.70 .00

Training time low Baseline Baseline

Avail training high 2.06 .00 1.92 .00

Avail training
equal

WES 1999 2001

Avail training low Baseline Baseline

Skill required high 2.31 .01 2.30 .00

Skill required equal 1.64 .14 1.35 .20

Skill required low Baseline Baseline

Training decision by 1.19 .01
supervisors

New soft/hardware 1.27 .01

Greater integration 1.13 .12

Downsizing 0.85 .04 1.17 .11
Greater reliance

on part-time workers 0.82 .03

More overtime 0.80 .05

Less management levels 1.28 .06

More job rotation 0.77 .00

Total quality control 1.23 .01

Greater reliance on 0.86 .09
external suppliers

More R and D 1.18 .06 1.18 .08

Competition from 0.69 .09
Canadian firms

Competition from
local firms 0.55 .01

No competition 0.81 .14

Competition from other 0.56 0.02
international firms

Importance of 1.53 .06
Canadian competition

Importance of 0.79 .00 1.47 .11
local competition

Importance of other 2.18 .00 1.40 .00
international competition

Notes: The models in this table, which are selected on the basis of a
model search process, do not include all listed explanatory variables.
There are no odds ratio estimates for these excluded explanatory
variables.


B. Empirical Results

The estimated results for the final specifications of the logit models for 1999 and 2001 are presented in Table 1. In this table, we report the estimated net marginal effects of the explanatory variables on the odds ratio of training participation. We proceed with our discussion of these results in turn.

Effects of Worker Attributes. Provincial differences are substantial in both 1999 and 2001 when all other determinants are controlled for. The incidence of workplace training in Atlantic Canada and that of Manitoba are not so different from that of the baseline case of British Columbia in 1999. Quebec, however, has the lowest participation rate in 1999. The participation rate of Quebec in 2001 is not so different from that of British Columbia but all other regions in Canada have higher marginal participation rates than British Columbia. (5)

Age indeed plays a significantly negative role in workplace training participation. The older the worker becomes, the less likely he/she participates in workplace training. Because older workers have fewer years remaining in their working lives, their returns on training investments are expected to decrease with age from both the employer and worker perspectives.

Compared with their comparable female counterparts, male workers participated less in workplace training in 1999 but not so in 2001. Marital status is not a statistically significant determinant for workplace training. When all other determinants are controlled for, the presence of pre-school children clearly lowers the probability that workers participate in workplace training in both 1999 and 2001.

Less education is correlated with more workplace training when all other determinants are controlled for. This finding differs from that in the literature based on the household surveys, from which we note that more educated people tend to study more or have more training. But this finding is based on workplace behaviors of firms and their workers and is consistent with the fact that workplace training is driven primarily by the gap between the job functions performed by workers and the education backgrounds of these workers, everything else being equal. Full-time workers had a higher marginal participation rate in workplace training in 1999 but this was not the case in 2001.

Relative to the baseline case of production workers, managers, professionals, and technical/trade workers have higher marginal participation rates in workplace training. Job tenure is neutral (in 1999) or statistically insignificant (in 2001). Union membership and wage are also neutral in influencing workplace training.

Effects of Firm Characteristics. Now we turn to firm characteristics. Relative to the baseline case of workers in natural resources, workers in finance and insurance have a higher marginal participation rate in workplace training in both 1999 and 2001, whereas workers in communication and other utility have the highest marginal participation rate only in 2001. Workers in all other industries have lower marginal probabilities of workplace training. Larger firms tend to have higher participation rates in workplace training. This robust finding is consistent with that of Baron, Black, and Loewenstein (1987), Holtmann and Idson (1991), Simpson (1984), Jennings (1996), Lin and Tremblay (2003), and Hui and Smith (2004).

Does firms' provision of workplace training encourage workers' participation in Canada? For this first hypothesis, we note in Table 1 that the net contributions of high training availability to the odds ratio of training participation are 2.06 and 1.97, respectively, in 1999 and 2001. That is, when the availability of the training is high, the workplace training participation will be substantially higher than that of the baseline case. This net impact is also statistically significant in both 1999 and 2001. Clearly, this finding has some implications to firms and their decisions in providing more workplace training to their workers.

How do changes in market competition, organizational changes, and technological innovation affect workers' participation in workplace training in Canada? Regarding this second hypothesis, we also find some important evidence from the WES data. That is, workplace training incidence is indeed positively correlated with international market competition, organizational changes, and technological innovation. The WES data offer some unique perspectives on this hypothesis.

More competition leads to more workplace training. For example, recognizing the pressure from international competition motivated firms to train more persistently in both 1999 and 2001. This is reflected in Table 1 which shows that the contributions of other international competition to the odds ratio of training participation are 2.18 and 1.40, respectively, in 1999 and 2001. National and local competitions are also important factors influencing workplace training, although they are not as persistent.

As shown in Table 1, workplace training is also correlated with a series of organizational changes. Most noticeably, the organizational changes that are positively associated with workplace training are greater integration (in 1999), downsizing (in 2001), less management levels (in 2001), total quality control (1999), and more research and development (1999 and 2001). Research and development activities appear to be one of most important driving forces behind workplace training, as the contribution of research and development activities to the odds ratio of training participation is 1.18 in both 1999 and 2001.

Table 1 also illustrates that the contributions of high technological complexity to the odds ratio stand at 1.29 and 1.27, respectively, in 1999 and 2001. This means that when technological complexity is higher, the participation rate in workplace training is higher. The contributions of high skill requirements to the odds ratio are 2.31 and 2.10, respectively, in 1999 and 2001. This shows that the higher skill requirements lead to higher training participation. These findings for Canadian firms arc consistent with those found elsewhere.

Overall, we have observed some important empirical results that support our two hypotheses. That is, in Canadian firms, the provision of training matters to workers' participation and increased international market competition, organizational changes, and technological innovation are all positively correlated with workplace training participation.

V. SUMMARY AND CONCLUDING REMARKS

In this work, we empirically investigated how firm characteristics and worker attributes are associated with workplace job-related training using the Canadian WES data. Some of these findings can be of great interest to decision makers of firms and economic policy markers for a region or nation.

We found that workers in Quebec had the lowest workplace training in both 1999 and 2001 than those in other provinces. Workers in British Columbia had the second lowest workplace training in both years.

Among various worker attributes, age is negatively associated with workplace training. Fulltime workers participated more in workplace training. Workers with pre-school children participated less in workplace training in both 1999 and 2001. Workers with less schooling participated more in workplace training, everything else being equal, as job functions dictate the need for training for those who have less education attainment but must do required work. This firm-survey-based finding is different from the finding based on household surveys where more educated people tend to have more training and/or education at and beyond workplaces. Relative to the baseline case of production worker, managers, professionals, and technical/trade workers had higher participation rates in workplace training.

Among various firm characteristics, in addition to industry, firm size, and union status, our empirical results show that the firms' provision of training can lift workers' participation significantly and that increased international competition, organizational changes, and technological innovation are positively associated with workers" participation in workplace training. Organizational changes are embodied in greater integration, downsizing, less management levels, total quality control, and more research and development. Technological innovation is primarily reflected in high complexity of technology and high skill requirements. These new findings were not previously available without using the WED data.

This study also shows where workplace training is lacking. More specifically, lower participation rates occurred among workers in some industries (such as labor-intensive manufacturing, construction, retail trade, and real estate), workers in some firms (such as firms with under 20 employees, firms with low technology, firms with little research and development, and firms facing little competition), older workers, workers with pre-school children, part-time workers, and production and marketing/sales workers.

REFERENCES

Acemoglu, D., and J.-S. Pischke. "Why Do Firms Train? Theory and Evidence." Quarterly Journal of Economics, 113, 1998, 79-119.

__. "Beyond Becker: Training in Imperfect Labor Markets." Economic Journal, 109. 1999a, Fl 12-F142.

__. "Minimum Wages and On-the-Job Training," NBER Working Paper No. 7184, 1999b.

__. "The Structure of Wages and Investment in General Training." Journal of Political Economy, 107, 1999c, 539-72.

__. "Certification of Training and Training Outcomes." European Economics Review, Papers and Proceedings, 44,2000,917-27.

Altonji, J., and J. Spletzer. "Worker Characteristics, Job Characteristics, and Receipt of On-the-Job Training." Industrial and Labor Relations Review, 45(1), 1991, 58-79.

Audor, D. H. "Why Do Temporary Firms Provide Free General Skills Training?" Quarterly Journal of Economics, 116(4), 2001, 1409-48.

Barnow, B., L. Giannarelli, and S. Long. Training Provided by Private Sector Employers. Washington DC: The Urban Institute, 1996.

Barron, J. M., D. A. Black, and ML A. Loewenstein. "Employer Size: The Implications for Search, Training, Capital Investment, Starting Wages, and Wage Growth." Journal of Labor Economics, 5(1), 1987, 76-89.

__. "Job Matching and On-the-Job Training." Journal of Labor Economics, 1, 1989, 1-19.

Barron, J. M., M. C. Berger, and D. A. Black. 1997. On-the-Job Training. Kalamazoo, Michigan: W.E. Upjohn Institute for Employment Research, 1997.

Bartel, A. P., and F. R. Lichtenberg- "The Comparative Advantage of Educated Workers in Implementing New Technology: Some Empirical Evidence." Review of Economics and Statistics, 69(1), 1987, 1-11.

Becker, G. Human Capital, 2nd ed, New York: National Bureau of Economic Research, 1975.

Betcherman. G., N. Leckie, and K. McMullen. "Barriers to Employer-Sponsored Training in Canada." Canadian Policy Research Network, July 7, 1998.

Bishop. J. "On-the-Job Training of New Hires," in Market Failure in Training? New Economics Analysis and Evidence on Training of Adult Employees, edited by D. Stern and J. Ritzen. Berlin: Springer-Verlag, 1991, 61-98.

Blinder, A. S., and Y. Weiss. "Human Capital and Labor Supply: A Synthesis." Journal of Political Economy, 84(3)," 1976, 449-72.

Bresnahan, T. F., E. Brynjolfsson, and L M. Hitt. "Information Technology, Workplace Organization, and the Demand for Skilled Labor: Firm-Level Evidence." Quarterly Journal of Economics, 117, 2002, 339-376.

Diaz-Vazquez, P., and D. Snower. "On-lhe-Job Training, Firing Costs and Employment," Mimeo, Universidad de Santiago de Compostela. Departamento de Funda-mentos del Anilisis Economico, Spain, 2003.

Dustman, C, and Uta Schonberg. "Training and Union Wages," IZA Discussion Paper No. 1435. December 2004.

Government of Canada. Achieving Excellence: Investing People. Knowledge and Opportunity, Ottawa: Industry Canada, 2002a.

. Knowledge Matter: Sills and Learning for Canadians, Ottawa: Human Resources Development Canada, 2002b.

Greenhalgh, C, and M. Stewart. "The Effects and Determinants of Training." Oxford Bulletin of Economics and Statistics, 49(2), 1987, 171-90.

Heckman, J., and J. Smith. "The Pre-Program Earnings Dip and the Determinants of Participation in a Social Program: Implications for Simple Program Evaluation Strategies." Economic Journal, 109(457), 1999, 313-48.

Holtmann, A., and T. Idson. "Employer Size and On-the-Job Training Decisions." Southern Economic Journal, 58(2), 1991,339-55.

Holzer, H., and J. Reaser. "Firm-Level Training for Newly Hired Workers: Its Determinants and Effects." Research in Labor Economics, 18, 1999,377-402.

Hui, S., and J. Smith. "The Determinants of Participation in Adult Education and Training in Canada," Report Prepared for Human Resources and Development Canada, Ottawa, Canada, 2004.

Javanovic, B. "Job Matching and the Theory of Turnover." Journal of Political Economy, 87. 1979, 972-90.

Jennings. P. 1996. "Employer-Sponsored Training in Canada: Evidence from the 1994 Adult Education and Training Survey," Working Paper Series, W-96-4E, Applied Research Branch, Strategic Policy, Human Resources Development Canada.

Knoke, D., and A. L. Kalleberg. "Job Training in U.S. Organizations." American Sociological Review, 59(4), 1994, 537-46.

Lerman, R. I., S.-M. KcKernan, and S. Ricgg. "The Scope of Employer-Provided Training in the United States," Chapter 7 in.lob Training Policy in the United States, edited by C. J. O'Leary, R. A. Straits, and S. A. Wanduer, W.E. Upjohn Institute for Employment Research. Kalamazoo, Michigan, 2004, 211-43.

Liilard, I ... and H. Tan. "Private Sector Training: Who Gets It and What Arc Its Effects?" in Research in Labor Economics, edited by R. Enrenberg. Vol, 13. Greenwich. CT: JA1 Press, 1992, 1-62.

Lin. Zhengxi, and J.-F. Tremblay. "Employer-Supported Training in Canada: Policy-Research Key Knowledge Gaps and Issues;" Ottawa: HRSDC-IC-SSHRC Skills Research Initiative Working Paper 2003 B-01. 2003.

Locwenstein. M. A., and J. R. Spletzer. "Dividing the Costs and Return to General Training." Journal of Labor Economics. 16. 1998, 142-171.

Lynch. L. "Private-Sector Training and the Earnings of Young Workers." American Economic Review, 82(1). 1992, 299-312.

Lynch. L. M., and S. E. Black. "Beyond (he Incidence of Training." Industrial and Labor Relations Review, 52(1), 1998, 64-81.

Mincer, J. "Union Effects: W;ages. Turnover and Job Training," in Research in Labor Economics: New Approaches to Labor Unions, edited by J. Reid. Jr. Supplement 2. Greenwich. CT: JAf Press, 1983. 217-52.

__. "Human Capital Response to Technological Change in Labor Market." NBER Working Paper No. 3207. 1989.

OECD. Beyond Rhetoric: Adult Learning Policies and Practices. Paris, France: OECD Publications Service, 2003.

Polachek, S. W. and W. S. Siebert. The Economics of Earnings, Cambridge and New York: Cambridge University Press. 1993.

Simpson, W. "An Econometric Analysis of Industrial Training in Canada." Journal of Human Resources, 19(4), 1984, 435-51.

Stevens. M. "A Theoretical Model of On-the-Job Training with Imperfect Competition." Oxford Economic Papers, 46, 1994. 537-62.

Turcotte, J. A. Leonard, and C. Montmarquette. 2003. "New Evidence on the Determinants of Training in Canadian Business Locations." in The Evolving Workplace Series, Ottawa: Statistics Canada and HRDC, Catalogue No. 71-584-MIE.

Weiss, Y. "The Determination of Life Cycle Earnings: A Survey," in Handbook of Labor Economics, edited by Ashenfelter and R. Layard. Vol. 1, New York: North-Holland, 1986. 603-640.

ABBREVIATIONS

AETS: Adult Education and Training Survey

NHES: National Household Education Survey

SEPT: Survey of Employer-Provided Training

WES: Workplace and Employee Survey

(1.) Becker (1975) suggested that individuals and firms invest in training when the discounted expected benefits outweigh the discounted expected costs. Contrarily, if the discounted expected benefits are less than the discounted expected costs, individuals and firms will not invest in training.

(2.) Among OECD countries, Portugal is an exception where adults with middle education attainment have the highest participation rate.

(3.) For details, see the summary provided in Lin and Tremblay (2003).

(4.) When participation and nonparticipation are equally-likely, the probabilities of these two actions are the same (.50%). The odds ratio is therefore.50/.50 = 1. When the two probabilities are.90 and.10, respectively, the odds ratio.90/. 10 = 9 suggests that participation is more likely. When the two probabilities are 10 and.90. Respectively, the odds ratio.10/.90 = 1/9 means that nonparticipation is more likely.

(5.) This finding for overall training participation is consistent with that in Hui and Smith (2004). According to the detailed statistical analysis of the authors, we note that in Quebec, classroom training tends to be higher than that of the baseline case (British Columbia) but on-the-job training tends to be lower than that of the baseline case.
TABLE A1

List of Variables from 1999 to 2001 WHS

Key Variable Variable Variable
of Interest Description Name

classroom job-related Received Classroom PARTCLRM
training training Length of DURCLRM
 first course taken

On-the-job Received on-the-job PARTOTJ
training training

 Time spent on-the-job DUROTJ
 training

Region Region DOM_REG

 Atlantic ATLANTIC

 Quebec QC

 Ontario ON

 Alberta AB

 British BC
 Columbia

 Manitoba MB

 Saskatchewan SK

Age Employee AGE
 birth date

 Age groups AGE_GRP

 Age group: AGELT25
 less than 25

Key Variable of Variable Variable
Interest Description Nam

 Age group: 25-34 AGE25_34

 Age group: 35-44 AGE35_44

 Age group: 45-54 AGE45_54

 Age group: 55-64 AGE55_64

 Age group: 65 and older AGEGT64

Sex Gender MALE

Marital status Marital status MARRIED

Dependent Dependent children PRESCH
children

Schooling Highest grade of ED UC
 Ele. or Hs completed

 Schooling: less than GRADE10
 10 yrs

 Schooling: 10-13 yrs GRADE 13

 Schooling: Some SOMEPS
 post-secondary

 Schooling: University UNIVER
 and above

Employment status Terms of employment FULLTIME
Industry
 WES industry aggregation DOMJND

 Forestry, mining, NATRESRC
 oil. and gas extraction

 Labor-intensive MANUL3RD
 tertiary manufacturing

 Primary product manufacturing MANV1ST

 Secondary product MANU2ND
 manufacturing

 Capital intensive MANUK3RD
 tertiary manufacturing

 Construction CONSTRUT

 Transportation, warehousing, TRANSWHS
 wholesale

 Communication and COMMUTIL
 other utilities

 Retail trade and RETAIL
 consumer services

 Finance and insurance FIN1NSUR

 Real estate, rental and REALEST
 leasing operations

 Business services BUSISRV

 Education and health services EDHTHSRV

 Information and INFOSRV
 cultural industries

Occupation groups WES occupation group OCP_GRP

 Managers MGMT

 Professionals PROF

 Technical/Trades TECH

 Marketing/Sales SALES

 Clerical/Administrative ADMIN

 Production workers PRODWKR

Job tenure Number of months previously TENURE
 worked for employer
 Job tenure groups TEN_GRP

Firm size Size FIRMSIZE

 1-19 employees FZ20LS

 20-99 employees FZ20TO99

 100-499 employees FZ1HT05H

 500 employees or more FZ5HPLS

Union Covered by CBA UNION

income Wage WAGE

Complex technological Complexity of technology TECFLCOM
change
 Remained about the same CMPLXEQU

 Increased CMPLXH1

 Decreased CMPLXLO

Subjective view/on Amount of training AMTRAIN
training motivation
 About right for the AMTTREQU
 demands of the job

 Too little for the AMTTRLO
 demands of the job

 Too much for the AMTTRHI
 demands of the job

 Not applicable, no AMTTRNND
 training required

Key Variable of Variable Description Variable
Interest Name

 Availability of training AVTRAIN

 Increased AVTRHI

 Remained about the same AVTREQU

 Decreased AVTRLO

 Overall skill requirements SKILL

 Increased SKILLHI

 Remained about the same SKILLEQU

 Decreased SKILLLO

Human resource practice Training decided by groups DSCNBY

 Training decided by nonmanagers BYCOWKR

 Training decided by work group BYWKGRP

 Training decided by work BYSUPER
 supervisor

 Training decided by manager/owner BYMGMT

 Training decided by people BYOUTSID
 outside workplace

Organization change Greater integration among INTEGRATE
 different functional areas

 Reduction in the number of LESSMGNT
 managerial levels

 Greater reliance on job ROTATION
 rotation, multiskilling

 Implementation of total QUALITY
 quality management

 Greater reliance on EXTERNAL
 external suppliers of prod./serv.

 Greater interfirm collaboration RANDD
 in R&D. production

 Other, specify OTHER

 Increase in degree of CENTRALL
 centralization

 Downsizing DOWNSID

 Decrease in degree of DCENTRA
 centralization

 Greater reliance on TEMPWKR
 temporary workers

 Greater reliance on PTWKR
 part-time workers

 Reengineering REENGINE

 Increase in overtime hours OVERTIME

 Adoption of flexible FLXBHOUR
 working hours

Technology use-computer Implementation of new NEW_SOFT
 software application or hardware

Technology use-other Implementation of other OTH_TECH
technology technologies or machinery

Innovation Innovation types INOVTYPE

 Innovation: improved processes IMPV_PRC

 Innovation: improved products IMPV_PRD
 or services

 Importance of innovation INNOV

 Innovation: new processes NEW_PRC

 Innovation: new products NEW_PRD
 or services

Competition Competitions with CMP_CAN
 Canadian-owned firms

 Competitions with CMP_LOC
 locally rowed firms

 No competition from other firms CMP_NONE

 Competitions with other CMP_OTH
 internationally owned enterprise

 Competitions with CMPJJSA
 American-owned firms

 Level of competition from LEV_CAN
 Canadian-owned firms

 Level of competition from LEV_LOC
 locally owned firms

 Level of competition from LEV_OTH
 other internationally owned firms

 Level of competition LEV_USA
 from American-owned firms

Weights Sampling weights for employees EMP_FINA

TABLE A2

Participation Rates of Employer-Sponsored Training

 1999 2001

 Rate of Rate of
 Participation Participation

Variable groups All training All training

Total 54.78% 53.76%

Region and province

Atlantic 51.21% 54.25%

Quebec 46.60% 47.62%

Ontario 59.75% 58.34%

Alberta 59.28% 53.34%

British Columbia 50.90% 47.64%

Manitoba 56.15% 54.65%

Saskatchewan 60.65% 53.22%

Age group (years)

Age less than 25 53.44% 55.02%

Aged 25-34 58.45% 59.59%

Aged 35-44 57.35% 53.09%

Aged 45-54 53.48% 51.87%

Aged 55-64 43.55% 40.67%

Age greater than 64 25.01% 20.83%

Gender

Male 56.36% 53.49%

Female 53.06% 54.02%

Marital status

Married/common-law 54.60% 53.26%

Other 55.18% 54.79%

Pre-school children

Without 54.72% 53.74%

With 55.08% 53.86%

Education

Below high school 39.34% 34.47%

High school graduates 47.04% 45.54%

Some university or 56.16% 56.35%
post-secondary

University or above 63.48% 64.95%

Type of job

Full-time 56.06% 54.39%

Part-time 42.71% 46.88%

Occupation

Managers 60.60% 58.13%

Professionals 68.06% 68.19%

Technical/trades 51.45% 51.24%

Marketing/sales 43.08% 44.71%

Clerical/administrative 54.83% 51.92%

Production workers 44.63% 41.54%

Job tenure

1-12 Mo 54.49% 53.59%

1-5 Yrs 58.87% 57.50%

6-10 Yrs 63.08% 51.24%

11-20 Yrs 60.83% 50.15%

Union membership

Yes 58.29% 58.18%

No 53.42% 52.03%

Income

Under 15,000 48.68% 47.67%

15,000-19,999 58.34% 52.88%

20,000-29,999 58.12% 44.67%

30,000-39,999 60.08 60.93%

40,000-49.999 64.86% 66.25%

50,000 or more 67.64% 66.91%

Industry

Forestry, mining, oil, 61.94% 59.91%
and gas extraction

Labor-intensive 38.45% 35.00%
Tertiary manufacturing

Primary product 52.64% 47.67%
manufacturing

Secondary product 59.30% 52.19%
manufacturing

Capital intensive 61.41% 55.04%
Tertiary
manufacturing

Construction 43.08% 43.28%

Transportation, 55.43% 50.27%
Warehousing
wholesale

Communication and 66.01% 73.55%
other utilities

Retail trade and 45.33% 44.99%
consumer services

Finance and 75.39% 75.97%
insurance

Real estate, rental, 43.46% 40.14%
and leasing
operations

Business services 58.61% 61.49%

Education and health 61.39% 62.76%
services

Information and 56.05% 55.36%
cultural industries
Firm size

Less than 20 43.62% 43.43%
employees

20-99 Employees 54.46% 52.32%

100-499 Employees 61.47% 58.20%

500 Employees or 65.76% 66.62%
over

Complexity of
technology

Remained about the 46.31% 46.07%
same

Increased 62.53% 62.88%

Decreased 50.29% 39.85%

Amount of training

About right 54.93% 57.72%

Too little 54.28% 56.97%

Too much 57.16% 41.33%

Availability of
training

Increased 71.24% 70.62%

Remained about the 45.83% 45.02%
same

Decreased 52.05% 50.98%
Overall skill
requirements

Increased 63.99% 64.17%

Remained about the 44.63% 3.71%
same

Decreased 36.81% 34.48%
Training decision
markers

By non managers 43.67% 45.14%

By work group 56.67% 61.03%

By work supervisor 58.49% 55.57%

By manager/owner 54.90% 52.86%

Types of innovation

New processes 62.59% 60.34%

New products or 58.15% 58.05%
services

Organization change

No greater integration 50.82% 51.01%

Greater integration 63.67% 63.19%

No reduction in 53.61% 53.06%
managerial levels

Reduction in 62.75% 64.63%
managerial levels

No greater reliance 53.85% 52.91%
on job rotation

Greater reliance on 57.28% 57.42%
job rotation

No total quality 53.01% 52.58%
management

Total quality 61.10% 60.49%
management

No greater reliance 54.41% 53.41%
on external
suppliers

Greater reliance on 56.41% 56.62%
external suppliers

No greater interfirm 52.73% 52.26%
collaboration in R&D

Greater interfirm 62.98% 64.45%
collaboration in R&D

No increase in degree 53.11% 51.73%
of centralization

Increase in degree of 62.36% 65.61%
centralization

No downsizing 54.22% 52.49%

Downsizing 57.06% 61.27%

No decrease in degree 53.61% 53.38%
of centralization

Decrease in degree of 63.34% 58.51%
centralization

No greater reliance 54.33% 53.17%
on temporary
workers

Greater reliance on 58.81% 59.51%
temporary workers

No greater reliance 54.78% 53.52%
on part-time workers

Greater reliance on 54.78% 55.80%
part-time workers

No reengineering 49.64% 49.45%

Reengineering 61.71% 62.52%

No increase in overtime hours 53.78% 52.56%

Increase in overtime hours 59.24% 60.08%

No flexible working hours 54.43% 52.78%

Adoption of flexible 56.35% 60.64%
working hours

New computer/technology

No new software 55.49% 52.49%
application or hardware

New software application or 59.48% 63.94%
hardware

No other technologies or 54.71% 53.68%
machinery

Other technologies or machinery 55.73% 55.16%

Competition

Not with Canadian-owned firms 53.70% 53.00%

With Canadian-owned firms 55.96% 54.69%

Not with locally owned firms 60.67% 60.17%

With locally owned 51.03% 49.07%
firms

With any other Arms 55.15% 54.21%

Without any other firms 49.45% 45.68%

Not with internationally 53.03% 51.17%
owned firms

With internationally 60.67% 62.79%
owned firms

Not with 53.13% 51.83%
American-owned firms

With American-owned 57.69% 57.43%
firms

Importance of competition

Competition from 53.47% 52.74%
Canadian-owned
firms not important

Competition from 56.31% 55.13%
Canadian-owned
firms important

Competition from 60.24% 58.96%
locally owned
lirnis not important

Competition from 50.79% 49.53%
locally owned
firms important

Competition from 52.77% 51.20%
internationally
owned firms not important

Competition from 62.51% 63.23%
internationally
owned firms important

Competition from 53.03% 51.64%
American-owned
firms not important

Competition from 58.12% 57.97%
American-owned
firms important

Note: The participation rates are for employer-funded courses training
and on-the-job training at workplaces.


doi: 10.1111/j.1465-7287.2010.00204.x

KUAN XU and ZHENGXI LIN *

* Views expressed in this work are those of the authors and as such, do not necessarily reflect those of their affiliations. Kuan Xu acknowledges financial support from the Skills Research Initiative under the HRSDC-IC-SSHRC partnership (Reference Number: 5013873). We thank Maxime Fougere and Andre Leonard for helpful comments on a previous draft, and anonymous referees for constructive comments and suggestions. We also thank Arden Bell, Neil Buckley, Sai Choi Chua. Lucy Chung, James Chowhan, Heather Hobson, Emmanuelle Petard, and Phyllis Ross for informatics support. We are solely responsible for any remaining errors.

Xu: Department of Economics, Dalhousie University, Halifax. Canada BAH 3J5. Phone 902-494-6995, Fax 902-494-6917, E-mail kuan.xu@dal.ca

Lin: Policy Research Directorate. Human Resources and Skills Development Canada, Place du Portage Phase IV, 140 Promenade du Portage, Third Floor, Gatineau. Quebec, Canada K1A 0J9. Phone 819-994-3699, Fax 819-953-8868. E-mail zhengxi.lin@hrsdc-rhdsc.gc.ca
COPYRIGHT 2011 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Xu, Kuan; Lin, Zhengxi
Publication:Contemporary Economic Policy
Article Type:Report
Geographic Code:1CANA
Date:Jul 1, 2011
Words:7839
Previous Article:Legislator incentives in a fragile democracy: evidence from budget allocation in Thailand.
Next Article:Do unions impact efficiency?: evidence from the U.S. manufacturing sector.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters