The differing effects of individual and group incentive pay on worker separation: evidence using Finnish panel data

Abstract

We investigate the role of individual incentive (II) and group incentive (GI) pay as determinants of worker separation using a large panel data set from Finland during 1997–2006. For white-collar workers, GI pay is associated significantly with an increased probability of separation (diminished employment stability), but in large firms only. For blue-collar workers, II pay is associated with a decreased probability of separation (enhanced employment stability), in both small and large firms. By providing results for different forms of performance pay in a single study, some of our findings are novel. In accounting for differences in our empirical findings compared to those in earlier studies, our results suggest that outcomes depend on the differing institutional contexts found in coordinated market economies (such as Finland) and liberal market economies.

Keywords:

• Job mobility
• performance related pay
• profit sharing
• wage inequality
• worker separation

JEL Codes:

• J33
• M52
• J31
• J62
• J63

Acknowledgments

We would like to thank the editor and the anonymous referees for their valuable comments. We are indebted to the Confederation of Finnish Industries (EK) for providing the permission to use their data on which the empirical analysis is based. We also thank Tomi Kyyrä, Gabriel Burdin and Jeff Pliskin, as well as participants at several conference including the International Association for the Economics of Participation 2012, the SOLE/EALE World Conference 2015 and the Leeds festival of Industrial Democracy 2017 for helpful comments. We are grateful to the Emil Aaltonen foundation for financial support. Part of the study was conducted by Mäkinen at Aalto University. The usual disclaimer applies. This paper is based on a Jones, Kalmi, Kato, and Mäkinen (2017a) (https://helda.helsinki.fi/bof/handle/123456789/14936).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 To be able to take into account this difference is an important issue because the presence of PRP (0/1) is a rather coarse measure for the magnitude of monetary incentives PRP provides for an individual worker, in particular when deciding whether or not to continue an ongoing employment match.

2 Due to some differences in the way the wage records are set up, our data are not fully comparable between white-collar and blue-collar workers. It is also likely that the determinants of the separation process differ between white-collar and blue-collar workers. We therefore analyze white-collar and blue-collar samples separately.

3 Relatedly, female labor market decisions (e.g., job mobility) may be affected by the presence of young children and women who have just given birth may decide to interrupt their employment spell in order to stay at home with their children – for example, Erosa, Fuster, and Restuccia (2002) find that fertility decisions produce important gender differences in employee turnover rates. Perhaps more importantly, our data do not allow us to distinguish between “true” female exits from the ongoing job match and female exits from the ongoing job match due to family reasons such as a decision to stay at home to take care of their young children.

4 In so doing we are aware that a worker’s firm ID code can change for reasons other than worker separations. For example, mergers and acquisitions of firms often result in firm ID changes for those workers affected, in spite of the fact that they are still working in the same workplace under the same employment contract. To respond to this potential problem, we verify whether all workers exit from a firm in a given year (i.e., the firm ID changes for all the workers of the firm). If so, we exclude these worker-year observations from the sample (but keep all the preceding years of observations).

5 As discussed later, note that we control for regional economic conditions by regional unemployment rates and heterogeneity across firms by firm fixed effects. We also exclude worker separations due to plant closures from the sample. Further, we try to address this in our unreported regressions (see pp. 20 and 22) where we consider individual firm performance (proxied by growth of total earnings paid by the firm) as an additional covariate and our results were largely unaffected.

6 We do not include individual earnings in the basic separation model, since it is potentially an endogenous explanatory variable with individual earnings probably correlated with individual innate ability (in the error term) as well as worker separation (dependent variable). However, in unreported regressions, we do estimate specifications in which we also include individual earnings as an explanatory variable. The reported key findings are largely intact.

7 In the EK data information on worker education come from Statistics Finland. The level of education consists of six categories (unknown, primary, upper secondary, lowest tertiary, bachelor, and master or higher). We combine the two lowest-level education categories (i.e., unknown and primary) in our analysis.

8 Somewhat surprisingly, in our sample 26% of blue-collar workers may have received both group and individual incentive pay in a given year. For empirical analysis, this combined performance pay variable is a potential concern since it may be highly correlated with both individual and group incentive pay, making the estimates less precise. Also, the association of this combined variable with worker separation is an aggregate of individual and group incentive effects, which would make its interpretation difficult. We therefore do not include this combined performance pay variable in our models. We did, however, some additional robustness checks to respond to this issue (see p. 22).