Key Words: compensating differentials, propensity score matching, public sector
In this article, propensity score matching (PSM) methods are applied to data from the 2005 International Social Survey Program Work Orientations (ISSP-WO) survey to examine the public sector pay premium in New Zealand. Taking account of a wide range of worker characteristics and attitudes, job attributes, and the effects that jobs have on workers and their family life, there appears to be a pay premium from working in the public sector of 17-21%.
The rising public sector wage bill is a key feature of the New Zealand labour market. This reflects not only the growth of the public sector, but also improvements in remuneration for public sector workers. For example, according to the Quarterly Employment Survey (QES), for the decade prior to the current Labour Government's election in 1999 average private sector wages were at least 80% of those in the public sector. Since then there has been a steady decline in pay parity, with average private sector wages being below 75% of public sector wages since 2005. There have also been improvements in non-wage benefits for the public sector, including the State Sector Retirement Savings Scheme which since 2005 has provided 3% matching employer contributions to employee retirement savings; well before and well above the level of employer contributions for other workers under KiwiSaver.
Some analysts have suggested that this rising remuneration for public sector workers reflects an asymmetry in employment relations between the public and private sectors. Since governments have statutory power to raise taxes, with large tax surpluses in the recent New Zealand case, they may not face the same financial pressures that inhibit wage rises in many private firms. This asymmetry also reflects the difficulty for taxpayers, who are the ultimate employer of public servants, to ensure that they are well represented in the wage negotiation process (Grimmond, 2007). Moreover, bureaucrats may have both a supply and demand role since they can influence the size of public sector employment, and hence face less of a trade-off between wage increases and employment than do other workers (Dahlberg and Mörk, 2006). Finally, since many public sector services are essential, an inelastic product demand contributes to inelastic labour demand, providing more scope for union activity to raise public sector pay. Therefore, it is unsurprising that unions have higher membership rates in the public sector than the private sector (Gregory and Borland, 1999).
However, improvements in the relative pay of public sector workers may also reflect changes in skill demands and job attributes between public and private sectors. These differences in job attributes have been found to account for much of the pay difference between public and private sectors in the U.K. (Bender and Elliott, 2002). However, other studies find that fringe benefits, such as holiday allowances, job security and pension schemes, are more generous for public sector workers (e.g., Poterba and Rueben, 1998), and that overall job satisfaction is higher (Demoussis and Giannakopoulos, 2007) so compensating differentials would imply lower public sector wages to offset these more favourable job conditions.
Since workers may choose to work in the sector that best suits their mix of observable and unobservable characteristics, any evaluation of the net advantages of public sector employment also needs to take such selection into account. For example, Bellante and Link (1981) find that public sector employees are more risk averse than their private sector counterparts. Therefore statistical methods used to estimate the public sector pay premium should compare public sector workers only with similar workers from other sectors. Such a comparison should also control for differences in productivity-related characteristics and in the positive and negative features of jobs that give rise to compensating pay differentials. …