Elsevier

Labour Economics

Volume 65, August 2020, 101864
Labour Economics

Why do firms (dis)like part-time contracts?

https://doi.org/10.1016/j.labeco.2020.101864Get rights and content

Highlights

  • Administrative data on the universe of Italian private-sector workers and firms are used to estimate the part-time / full-time wage differential.

  • Regression estimates control for worker, firm and match fixed effects.

  • A fairly uniform part-time wage premium is found for all subgroups of workers and firms.

  • We explored different mechanisms for the estimated part-time wage premium.

Abstract

This paper investigates the costs for firms of employing women full-time versus part-time, in terms of differential hourly wages. To this end, we use administrative matched employer-employee data on the universe of female workers in Italy over a period of 33 years and rely on regression models that control for worker, firm, and job match fixed effects, in addition to several worker-, job-, and firm-level time-varying factors. We find that, when a worker switches from a full-time to a part-time contract within the same firm, she benefits from an increase in the hourly wage. Over the last three decades, these wage premiums have significantly reduced, although they remained positive and significant up to 2015. We also find that the part-time premium is pervasive and stable across many different labor market segments and independent of the workers’ intrinsic productivity levels. These and other findings appear to be compatible with developments in wage bargaining institutions, whereby more generous conditions can be granted to part-time workers. Coupled with the detrimental effect of part-time work on firm productivity documented by Devicienti et al. (2018), our results contribute to explaining why firms are often unwilling to concede part-time positions to those employees who request such arrangements.

Introduction

Many experts have stressed that part-time work is a valuable work-life balance instrument, since it allows people to better conciliate their work with the needs of their private lives (Eurofound, ILO, 2019, Eurostat, 2009, OECD, 2017). However, both anecdotal and the available statistical evidence suggest that workers who wish to switch to a part-time work schedule often encounter resistance, if not outright opposition, from their employers. A quick search on the Internet confirms this: there is a multitude of online forums where workers complain that their employers have not allowed them to work a reduced number of hours. It is not unusual for workers, particularly females, to be forced to leave their jobs following a denied request to switch to a part-time work arrangement. Accordingly, Gasparini et al. (2012) reported that only about 30% of full-time employees in EU-15 feel that their employer would favorably consider their request to reduce their working hours.

There may be several reasons why firms are generally unwilling to satisfy the requests of their workers to switch to a part-time contract. The communication and start-up costs associated with part-time work and the difficulties of optimally staffing part-time employees might lead to efficiency losses in a firm’s organizational structure. In a recent study, we have found, after accounting for a large number of worker and firm-level characteristics, that part-time work is, in fact, linked to significantly lower firm productivity, and that this result holds for different categories of firms (Devicienti et al., 2018). If firms can compensate for this productivity gap by offering lower wages to part-time workers, they should be indifferent between employing the workers according to a full-time or a part-time schedule. However, this is usually unfeasible in most industrialized economies, where the law dictates that part-time workers should have the same monetary (and non-monetary) benefits as comparable full-time workers. In some countries, such as Italy, the law even allows for a more favorable treatment of part-time workers. The emergence of a zero or positive part-time/full-time wage differential, coupled with the lower productivity associated with part-time arrangements, could help to understand the reluctance of firms to satisfy the requests of workers to switch to part-time schedules. This paper, which tackles the analysis from a firm’s viewpoint, is aimed at investigating whether this happens or not.

In order to estimate whether part-time schedules have a different cost from full-time arrangements, we use a very large data set, which covers the universe of private-sector employees in Italy over a period of more than 30 years. This data set is based on administrative data from the Italian Social Security System (INPS) and links each employee to the firm he/she works in, thereby allowing a large amount of longitudinal worker-, firm-, and job-level information to be exploited. This multiple-level information is crucial to assess the presence of any wage differentials associated with part-time arrangements, since a multitude of factors can simultaneously determine wages and the part-time work status. We concentrate on the impact of a change in the working time arrangement of the same worker within the same firm. This allows any confounding effects, due to the presence of unobserved fixed worker, firm, and job match heterogeneity, to be removed. At the same time, we control for an ample set of time-varying observable factors that might also confound the effect. Particular attention is devoted to controlling for the employee’s work history, with specific reference to accumulated experience in the labor market, in part-time and full-time work. We also pay attention to any possible contemporaneous endogeneity stemming from maternity and from the employer undergoing periods of economic crisis, which are likely to simultaneously influence a change in the working schedule and wages.

In this study, we focus on females. This decision was taken for several reasons. First, females constitute the vast majority of part-time workers. Second, they represent the most relevant segment for the aims and policy implications of this paper. Females are those that are more likely to ask for (temporary) transitions to part-time work, often to conciliate work with family commitments. A denied request to switch to a part-time arrangement may entail withdrawal from work, with well-known long-lasting consequences in terms of earnings and the possibility of successfully re-entering the labor market. Third, part-time work for males is a very heterogeneous phenomenon and, unlike what happens for women, it is mostly involuntary (i.e., most men working part-time would prefer a full-time position). Fourth, we concentrate on females for comparative reasons, since most studies that have examined part-time/full-time wage differentials have focused on women.

Several papers have analyzed part-time/full-time wage differentials, but most of them have investigated the issue from the workers’ viewpoint. These studies were mainly interested in assessing the determinants of part-time/full-time wage differentials, and how the part-time status influences the future earnings and career trajectories of workers. Particular attention has been devoted to gender issues, either by concentrating the analysis on women or by juxtaposing part-time earning differentials with gender wage gaps (Manning, Petrongolo, 2009, Matteazzi, Pailé, Solaz, 2018, Mumford, Smith, 2009, Pacelli, Pasqua, Villosio, 2013); to the impact of switching to a part-time contract on the future earnings and career prospects of the workers (Connolly, Gregory, 2009, Fernández-Kranz, Rodríguez-Planas, 2011, Paul, 2016); and to the presence of a possible heterogeneity of part-time wage differentials along the wage distribution (Gallego Granados, Nightingale, 2019, Simon, Sanroma, Ramos, 2017).

This paper contributes to the existing part-time literature in several ways. It is one of the few studies that explore the wage effect of switching to a part-time contract while remaining with the same employer. Moreover, the use of administrative data on the universe of workers and firms over a period of more than three decades allows us to explore the long-run dynamics of the part-time/full-time wage differential, and to run separate analyses on many different categories of workers and firms, based on, for instance, age, migration status, parenthood, job duration, occupation, and the firm’s size, industry, and location. The dimension of our data entails that, for each of these analyses, we can remove any confounding factors related to worker, firm, and match-specific unobserved heterogeneity. Furthermore, in order to explore the mechanisms at play, we investigate whether the effects vary between short and long part-time work (as in Paul, 2016) and between switches from full-time to part-time work and switches from part-time to full-time arrangements (as in Booth and Wood, 2008, and Day and Rodgers, 2015). We also explore the relevance of the mechanisms related to workers’ commuting to work and differential rent-sharing by part-time status within the firm.

Fernández-Kranz and Rodríguez-Planas (2011), who used longitudinal matched employer-employee data to estimate part-time/full-time wage differentials in Spain are, to the best of our knowledge, the only scholars who have controlled for both unobserved individual- and firm-level fixed heterogeneity, as we do in this paper.2 They reported significant part-time wage penalties for female workers, which remained after controlling for individual and firm fixed heterogeneity, and which were particularly pronounced for temporary workers. On the contrary, our estimates point to the existence of pervasive wage premiums associated with part-time work schedules, which are transversal to many different segments of the labor market. These results, together with other additional results presented in this paper, are compatible with part-time premiums that stem from the relatively higher protection accorded to (female) part-time workers by unions and sectoral collective agreements.

The rest of the paper is structured as follows. Section 2 briefly discusses the mechanisms that can lead to a wage differential between part-time and full-time work schedules. Section 3 reviews the existing empirical literature on the part-time/full-time wage differential along two main dimensions: estimation methods and cross-country evidence. Section 4 outlines our empirical model, the type of effect that we identify, and its relationship with previous empirical works. Section 5 describes the data, Section 6 shows and discusses our results and, finally, Section 7 outlines the implications of our findings and draws conclusions.

Section snippets

Conceptual framework

There are several theoretical explanations as to why firms may pay full-time workers different hourly wages from part-time workers.

One mechanism is related to productivity differentials between the above two categories of workers, which - wage rigidity being absent - should be reflected in wage differentials. Part-time workers may be less productive than full-time workers, due to the daily start-up costs, whereby the individual labor productivity is lower during the first hours of work and only

Previous empirical literature

The existing evidence on the presence of part-time/full-time wage differentials is mixed and depends on the type of data and estimation methods that were used. Many studies have focused on cross-sectional surveys of workers, and have generally found wage penalties associated with part-time work, which often remain after controlling for a series of individual-, firm-, and job-level characteristics (Bardasi, Gornick, 2008, Elsayed, de Grip, Fouarge, 2017, Hardoy, Schøne, 2006, Jepsen, O’Dorchai,

Empirical model and identification issues

As discussed in the previous sections, the part-time/full-time wage differential may be the result of several intervening factors, and working a reduced number of hours per se is only one of them (Paul, 2016). In order to motivate our empirical model, clarify the nature of the estimated parameters, and discuss identification issues, it could be useful to quickly recap the numerous confounding factors at play that have emerged from previous empirical works.

A relevant fraction of the raw

Data

We use administrative data from the Italian Social Security System (INPS), which collect labor market histories for the 1983–2015 period of each employee working for at least one day in any private-sector firm in Italy. INPS assigns unique identifiers to the workers and firms, which allow us to track them longitudinally. It is also possible to know in which firm a given worker is employed at each point in time. Hence, we have a longitudinal matched employer-employee data set on the universe of

Main results

Table 1 shows the main results derived from the estimation of Equation (1). Here, we concentrate on the 2009–2015 period. Following the discussion in Section 4, we present different versions of the estimated part-time/full-time wage differentials, in which controls are gradually inserted. All the estimations report robust standard errors clustered at the worker and firm (i.e., at the job match) level.

The first row in the table shows a raw part-time/full-time wage differential of -0.233. This is

Conclusions

In this paper, we have used matched employer-employee data on the universe of female Italian private-sector employees over a period of 33 years and analyzed the costs for employers, in terms of wage differentials, of transforming a work contract from full-time to part-time. Our research aim was motivated by the fact that employers are often reluctant to concede switches to part-time arrangements, as reported in official statistics and anecdotal evidence. Assessing the reasons for such a

References (39)

  • D. Fernández-Kranz et al.

    The part-Time pay penalty in a segmented labor market

    Labour Econ

    (2011)
  • E. Wolf

    Lower wage rates for fewer hours? a simultaneous wage-Hours model for germany

    Labour Econ

    (2002)
  • D. Aaronson et al.

    The effect of part-Time work on wages: evidence from the social security rules

    J Labor Econ

    (2004)
  • J.M. Abowd et al.

    High wage workers and high wage firms

    Econometrica

    (1999)
  • E. Bardasi et al.

    Working for less? Women’s part-Time wage penalties across countries

    Fem Econ

    (2008)
  • Y. Barzel

    The determination of daily hours and wages

    Q J Econ

    (1973)
  • A. Booth et al.

    Back-to-Front down under? part-Time/full-Time wage differentials in australia

    Industrial Relations. A Journal of Economy and Society

    (2008)
  • D. Card et al.

    Rent-Sharing, holdup, and wages: evidence from matched panel data

    Rev Econ Stud

    (2014)
  • S. Connolly et al.

    The part-Time pay penalty: earnings trajectories of british women

    Oxf Econ Pap

    (2009)
  • I. Day et al.

    The premium for part-time work in australia

    Australian Journal of Labour Economics

    (2015)
  • F. Devicienti et al.

    Collective bargaining and the evolution of wage inequality in italy

    British Journal of Industrial Relations

    (2019)
  • F. Devicienti et al.

    The impact of part-Time work on firm productivity: evidence from italy

    Industrial and Corporate Change

    (2018)
  • A. Elsayed et al.

    Computer use, job tasks and the part-Time pay penalty

    British Journal of Industrial Relations

    (2017)
  • J.F. Ermisch et al.

    Wage offers and full-Time and part-Time employment by british women

    J Hum Resour

    (1993)
  • Eurofound et al.

    Working Conditions in a Global Perspective

    Technical Report

    (2019)
  • Eurostat

    Reconciliation between Work, Private and Family Life in the European Union

    Technical Report

    (2009)
  • D. Fernández-Kranz et al.

    Part-Time work, fixed-Term contracts, and the returns to experience

    Oxford Bullettin of Economics and Statistics

    (2015)
  • Gallego Granados, P., 2019. The Part-Time Wage Gap across the Wage Distribution. DIW Discussion Papers, No....
  • G. Gasparini et al.

    Full-Time or Part-Time Work: Realities and Options

    Technical Report

    (2012)
  • Cited by (2)

    • Tastes for discrimination in monopsonistic labour markets

      2022, Labour Economics
      Citation Excerpt :

      Moreover, this effect is stronger in some specifications and more significant among blue-collar workers. The presence of wage penalities for full-time women at firms where more flexibility is available is a novel evidence, given that most studies focus instead on penalties (or premiums) among part-timers (e.g. Devicienti et al., 2020; Elsayed et al., 2017). Hoever, this finding is coherent with studies on women’s preferences for shorter schedules (e.g. Booth and van Ours, 2013; Del Boca, 2002), and with studies that suggest that part-time contracts are costly for firms, thus that they should be modeled as a non-wage amenity (Devicienti et al., 2018).

    1

    We kindly acknowledge the funding support and data access provided by the VisitINPS program. Francesco Devicienti and Davide Vannoni also gratefully acknowledge the funding support of MIUR - Ministry of Education, University, and Research (Financing Fund for Departments of Excellence) and of the PROWEDEC (Productivity, Welfare, and Decentralized Bargaining) project, financed by the Compagnia di San Paolo. We would like to thank Edoardo Di Porto and Paolo Naticchioni for their general help with INPS data usage. Thanks are also due to Michelangelo Filippi and Roberto Quaranta for sharing their knowledge on the INPS variables related to part-time work with us, and Ainoa Aparicio Fenoll, François Rycx, and Claudia Villosio for their helpful comments. We gratefully acknowledge the insightful comments and suggestions made by the Editor and two anonymous referees. The opinions expressed by the authors are solely their own and do not necessarily represent the views of INPS.

    View full text