Academic journal article Australian Journal of Labour Economics

Job Mobility and Earnings in New Zealand1

Academic journal article Australian Journal of Labour Economics

Job Mobility and Earnings in New Zealand1

Article excerpt


The movement of workers between jobs may play an important role in determining both the average level and overall dispersion of earnings in an economy. This study provides some initial empirical results on this topic using administrative data provided by Statistics New Zealand (Linked Employer-Employee Data). We find that the movement of workers to different employers is extensive, and job mobility is associated with a narrowing in this earnings gap. Earnings growth is higher for those who change jobs than for those who stay with the same employer. However, this result disappears once we control for a wide variety of other determinants of earnings growth. Firm characteristics appear to play an important role in the relationship between job mobility and earnings. A move to a larger firm (i.e., one with more employees) and a firm that pays higher average earnings to all its employees can result in a substantial increase in individual earnings. Earnings growth is also found to be negatively related to the time interval between jobs, and the initial earnings of the individual. Once individual and firm characteristics are held constant, however, job changes by themselves lead to a relative decline in earnings growth.

1. Introduction and Overview

We know little about the extent and nature of job mobility in New Zealand. Similarly, the link between job mobility and earnings has never been thoroughly investigated in this country. The purpose of this study is to provide some preliminary empirical findings on the movement of workers between employers, and the consequences of this job mobility for the earnings received by these individuals. The underlying notion is that job mobility may play an integral role in the determination of the average level of earnings, as well as the dispersion of earnings, in the economy. This study illustrates how linked employer-employee data derived exclusively from administrative records can be used for analysing job mobility behaviour.

The structure of this paper is as follows. Section 2 provides a brief review of the relevant overseas literature in this area. Section 3 summarises the characteristics of the Linked Employer-Employee Data (LEED) provided by Statistics New Zealand for this project, discusses the advantages and disadvantages of these data for the specific research questions addressed in this project and presents some descriptive statistics on job mobility and earnings Section 4 presents the results from a regression analysis on the potentially complex links between job mobility and the earnings of individuals. Section 5 summarises the findings of this study.

2. A Brief Literature Review

Two basic facts appear to characterise job mobility in most countries. Most new jobs end early (i.e., frequent early job separations), and long-term employment relationships are common (i.e., lengthy job tenure). For example, nearly one-quarter of US workers in 1996 between the ages of 20 and 64 were in their current jobs for less than one year. Yet, more than one-quarter of US workers examined in the same year between the ages of 45 and 64 had held their jobs for at least 20 years (Farber, 1999).

These facts are not contradictory. They simply suggest that there is a great deal of churning of workers in new jobs (especially early in the work life). Yet, employee-employer matches that survive these first few years tend to be relatively long-lived. As a result, hazard rates (the probability of a job ending conditional on some elapsed time in the job) decline substantially with tenure.

There are competing hypotheses for the observed decline in hazard rates. They could be attributed to worker heterogeneity (mobility could be largely relegated to workers who are inherently predisposed to high levels of job turnover), or they could be attributed to state dependence (early mobility may itself directly lead to subsequent mobility). The literature suggests that both explanations have merit, but that unobserved heterogeneity alone can not entirely explain the mobility patterns that are found (see, Farber, 1999; and Munasinghe and Sigman, 2003). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.