ECONOMIC DEVELOPMENT QUARTERLY / November 2001
workers by employers was positively related to the provision of good benefits packages (pension and health insurance benefits) to their regular, full-time employees.4
EMPLOYMENT SECURITY AND RETENTION
The shift to more flexible forms of production and the intensified poaching of skilled workers appears to be creating problems for many smaller manufacturers in the form of increased turnover, workforce instability, and breakdowns in internal skill-development systems (Appelbaum & Batt, 1994; Luria, 1996). Workforce instability does not just hurt workers; it also hinders the ability of employers to plan work orders and production timetables and can cause productivity losses.
Employers experienced higher rates of turnover in the 1990s than they had in the two previous decades (Bureau of National Affairs, 2000). In particular, turnover increased sharply among those businesses with fewer than 250 workers. Small manufacturers reported losing about 40% of their workforces annually (Siekman, 1998). The entry-level job market in particular is characterized by considerable churning. A survey of 500 small employers revealed that in half the firms surveyed, the majority of entry-level workers stayed with the employer for 1 year or less (Regenstein et al., 1998). Another study found that young employees now work for more employers and have shorter tenures at each job site (Bernhardt, Morris, Handcock, & Scott, 1998). Rapid job churning is most pronounced among workers with less than a high school education (Monks & Pizer, 1998). Even in manufacturing, where workers tend to be older and presumably less mobile, job tenure is much shorter than it once was—the odds of a job separation in manufacturing are 30% higher for workers in the 1990s than for workers two decades ago (Bernhardt et al., 1998)
Of course, turnover may be either employer-initiated (firings and layoffs) or worker-initiated (resignations and retirements). These separations are often referred to as involuntary or voluntary, respectively. Although there is an obvious difference between choosing to leave and being forced to leave a job, especially in terms of one’s eligibility for unemployment insurance, dissolving an employment relationship is most often a joint decision. The greater the threat of being laid off involuntarily, for example, the greater the likelihood of voluntary separation (Stoikov & Raimon, 1968).
Employees are working for individual employers over shorter tenures—whether by choice or because their jobs are being created and destroyed at a rapid rate. Despite evidence of record job creation in the United States in the 1990s, urban and rural economies still have experienced large- scale layoffs arising from plant closings, downsizing, and mergers and acquisitions. Following a surge in downsizing in the recession of the early 1990s, permanent job loss remained quite high throughout the decade (Hipple, 1997; Valletta, 1998). In particular, layoff incidence increased sharply between 1994 and 1995, due in part to a delayed response by defense-dependent contrac- tors to declines in the military procurement budget. Even in the late 1990s, voluntary job leaving constituted a relatively small share of the unemployment incidence (14% on average) (Economic Policy Institute, 1999). Although small manufacturers often find ways to retain their most valued employees during downturns, pressures to lower costs force companies to do more with less.
What explains the increasing frequency of employment turnover in the economy? A certain amount of turnover is to be expected from a dynamic economy; individuals choose employers and positions, and employers decide which applicants are suited to filling available jobs. Specific envi- ronmental factors, however, contribute to higher rates of turnover. Voluntary turnover tends to fol- low the business cycle. During periods of economic prosperity, workers have greater confidence in their ability to migrate to other, often better paying, jobs (Economic Policy Institute, 1999). At the same time, when a strong economy leads to labor shortages, workers filling entry-level positions are more likely to be young, have little work experience, and few proven skills. These workers have a higher propensity to quit or be fired (Cappelli, 1999). The opposite is true during recessions.
A number of studies have explored the causes of employee turnover. In their summary of this lit- erature, Cotton and Tuttle (1986) noted several factors that appear to be positively correlated with decreases in turnover. These include age, job tenure, number of dependents, wage levels, job satis- faction, union presence, and aggregate unemployment rates. This supports earlier findings that