Incidence, coverage and employee knowledge of participation in financial incentive schemes: evidence from US cases

Abstract

By using new linked firm and individual data for cases in one US region, including firms in banking, we contribute to three issues concerning the incidence of financial incentive (FI) schemes: (1) firm-level data reveal the existence of substantial differences in incidence across firms in similar industries in this region and provide weak evidence of a role for urban versus regional location in accounting for differences in implementation of FIs; (2) individual-level data, gathered from face-to-face surveys for almost 900 employees in the cases, show substantial within-firm heterogeneity concerning the distribution of particular FIs among employees, although probit estimates do not uncover consistent evidence concerning the key determinants of the reasons behind these mainly managerial choices for all FIs; (3) the individual data also indicate the existence of substantial within-establishment heterogeneity in employee knowledge of the coverage of FIs. In accounting for the determinants of this variation in knowledge, for some plans, sometimes probit estimates provide evidence in support of hypotheses concerning age, experience and education. We also find roles for some attitudinal factors relating to discretionary effort, peer monitoring and employee involvement, and thus augment findings reported in earlier work that typically have used more restricted data. One general implication of our study is that research of issues surrounding FIs may be best served when data collection operates in tandem at different levels of aggregation.

Keywords:

• bonus
• employee involvement
• employee ownership
• financial participation
• profit sharing

Acknowledgements

The authors are grateful to Russell Sage, Rockefeller and NSF SES-0522117 for their support. They also thank Jeffrey Pliskin and two anonymous referees for their valuable comments.

Notes

 1. The five cases are part of a larger sample for firms in the same region. That sample is discussed in Jones, Kato and Weinberg (2003). The cases in this paper are chosen because we have individual data (employee responses to a questionnaire) only for these five cases.

 2. Our larger sample of 12 cases reveals similar diversity in the presence of a programme at the firm/establishment (see Jones, Kato and Weinberg 2003).

 3. For a discussion of offline teams for this case, see Jones and Kato (2011).

 4. The strongest evidence on US workers wanting more innovative work practices that are typically provided, especially EI programmes, is provided by Freeman and Rogers (1999).

 5. On-line teams are what is usually meant by a “team”. They occur in cell or modular production and emerge after workplaces have been transformed by using self-directed production units. By contrast, members of an off-line team are drawn from different parts of the business. Often offline teams are modelled after Japanese small group activities (e.g. QC circles) and were designed to problem-solve or enhance employee involvement.

 6. In some instances, such as ESOPs, since there are legal requirements concerning levels of participation, managers do not have much choice.

 7. If some scope for employee choice exists then we expect that similar factors would govern their decision whether or not to choose to participate in an EI/FI programme.

 8. For blue-collar workers in Canada, Jones and Pliskin (1997) find evidence that some FI schemes are negatively related to unionism.

 9. Applebaum et al. (2000) present evidence on the link between commitment to the organization and business performance.

10. Of course, we realize that such differences, especially in attitudes, might, to some degree, result from the existence of these different HR policies and practices.

11. To keep the paper (and the number of tables) to a manageable length, we are forced to choose and to make selected comparisons for certain FI or EI programmes both within and across firms. In the main, the analyses we undertake reflect both the availability of data (which potentially permits meaningful analysis) and feedback from our interviews (as to which HR features are believed to be most important in different firms).

12. This is not explained by some respondents not meeting the vesting period since the average tenure for respondents who believe they are not covered is 17.6 years.

13. For the remaining case for which we have data for individuals, Materials, the near existence of universal coverage for most schemes and the relatively small number of observations for this firm (71) prevents us from undertaking this kind of analysis as well as any kind of multivariate analysis.

14. Age and education have been found to play similar roles in many other workplace issues. Thus, a recent study on workplace safety (Gyekye and Salminen 2008) finds that higher educated workers recorded the best perceptions on safety. And DellaVigna (2009) reports that a growing body of field evidence exists that show how, in their decision-making, individuals often show concern for the welfare of others.

15. Broadly speaking, in unreported exercises we find that similar patterns exist regarding basic characteristics and attitudes of employees who do/do not perceive that they are in a PSP at Large Metal.

16. We report probability coefficients. Marginal effects have the same signs as the reported coefficients but their magnitudes depend on the characteristics of the employees.

17. One implication of this finding that geographical location potentially plays an important role in accounting for variation in the incidence of FIs, is that when using large firm-level data sets to test learning hypotheses (as in Ben-Ner and Lluis 2009), one must be cautious when constructing appropriate measures. Our evidence is suggestive that simple measures such as distance from a single major city may be too one-dimensional and that measures need to be more nuanced (and include dimensions such as the extent of ‘ruralness).