Understanding Labor Turnover: The role of Labor Stability
(Published in Jigyasa (New Delhi), 22006)
Martin G. Evans
Abstract
In nearly all of the research on labor turnover there has been little or no focus on a second aspect of worker mobility: labor stability. By labor stability we mean the proportion of workers who have remained with the organization for a given period of time (usually one year). In this paper we show how different combinations of stability and turnover can co-exist and that turnover has quite different causes and consequences depending on the labor stability of the organization or its sub-units.
A major human resource management problem faced by many organizations is that of understanding and managing labor turnover. There are two diagnostic questions that a manager needs to ask in order to develop a program of turnover control: how high is labor turnover; and what kinds of people are leaving? Most research has focused on the former question (and this paper adds a further consideration, labor stability, to the issue); the second question has only recently come under consideration (e.g., Staw, 1983; Dalton et al 19xx)
In most of the discussions of labor turnover (e.g. Mobley, Griffeth, Hand, & Meglino, (1979)) there has been little consideration about the pattern of turnover within an organization. Pfeffer (1983), with his focus on organizational demographics, has asked whether or not turnover is spread evenly across people with different lengths of tenure or whether turnover is spread evenly across all departments in the organization. This is based on some earlier work (cited in Van Der Merwe & Miller (1973) that examined survival curves in organizations (Brissenden & Frankel, 1922; Greenwood, 1919; Lane & Andrew, 1955))
One metric that helps to identify turnover in different parts of the organization or at different stages of tenure is the measure of labor stability. Labor stability is usually defined as the proportion of workers (in a particular category or in the firm as a whole) who remain for one calendar year in the organization:
# employed on both Jan 1st. and on Dec 31st.
Stability = ___________________________________________ X100
Average number employed during year
Similar indices could be developed for different time periods. Whether or not turnover is concentrated in a few positions or spread widely throughout the organization has quite important implications about the causes and consequences of labor turnover. Pfeffer (1983) has argued that tenure distributions may be quite important to organizations for understanding turnover. McCain, O'Reilly, & Pfeffer (1983) found that in a university setting turnover was associated with quite uneven distributions of tenure. It is likely that socialization and attachment will be equally difficult in other organizations with similar gaps in demographic continuity between organizational members. A newcomer entering the organization who is the first newcomer for five or six years (characterized by high prior labor stability) will have difficulty in identifying with her/his fellow workers and therefore have little attachment to the organization.
Combinations of stability and turnover
These examples indicate that the role of labor stability in understanding labor turnover and its causes and consequences is not trivial. Across different situations, labor stability may differ markedly: such differences are associated with quite different types of labor turnover, and require quite different managerial responses.
There are three extreme types (1): a) high stability, low turnover; b) high stability, high turnover; c) low stability, high turnover. The fourth combination (low stability, low turnover) cannot occur, except in organizations which are expanding rapidly: in the more typical situation if turnover is low then stability, by definition, is high. The opposite is not true: high stability can be coupled with high or low turnover.
Three combinations of stability and turnover
High stability, low turnover. (2) This is perhaps the most desirable condition; although some turnover is often beneficial in terms of weeding out deadwood and introducing new ideas to the organization (Staw, 1983). In this situation, only a few positions are implicated, few vacancies occur in each. In this case, there is no crisis. It is this steady state that organizations strive to achieve and of which organizational researchers strive to understand the causes.
High stability, high turnover (3). Here most positions in the organization are filled with experienced people. Nevertheless in a few positions it is almost impossible to retain an employee. No sooner is the position filled than it is vacated; it is then refilled, but the new employee does not stay; and so the process continues. The crisis is that of socializing newcomers. The long tenure people seem committed to the organization; the newcomers are not.
This pattern of stability and turnover seems to have been ignored by both managers and researchers. They implicitly assume that there is a negative correlation between turnover and labor stability. The existence of this stability/turnover pattern would upset such a simple relationship. Clearly we should be alert to its possible occurrence. I have stated that in this condition there is an organizational crisis. Clearly this is an overstatement. A pattern of this type could be developed through a conscious organizational strategy. If the position is one in which a) highly skilled personnel are required, and b) learning time is brief, but there is c) little intrinsic interest in the work and d) little opportunity for advancement, the organization may wish to staff the organization with what a colleague once called "bright young birds of passage." Here high turnover is accepted as a cost of staffing. However, if this is not the result of a deliberate strategy, the organization has to focus on the problems of selection and socialization (see below).
Low stability, high turnover.(4) In this condition, many positions are becoming vacant; many hires are being taken on each year. The whole organization is in a state of flux. Turnover is spread over many positions, quite likely people at all stages of tenure will be leaving the organization. Thus the organization is facing two crises. The first is the crisis of retaining people with experience and high tenure. The second crisis is that of selecting, socializing and retaining new hires into the organization.
Implications
When turnover is narrowly concentrated in a few positions (and hence among those with low tenure) the problem faced by the organization in identifying those who may leave is a different problem from that faced by the organization when turnover is spread more broadly among high and low tenure employees. In the first case, the focus is on new hires. The problem is to identify the factors that increase or reduce their commitment to the organization. In the second case, the problem is more complex: the focus has to be both on new hires and on long tenure employees.
This is not the place to go into an exhaustive treatment of the factors that differentiate attraction to the organization for short and long tenure employees. Clearly such differences do exist. The basis for commitment to the organization of short tenure workers and long tenure workers is quite different. For persons with short tenure, commitment is a function of the nature of the initial decision to join the organization (Salancik, 1977; O'Reilly & Caldwell, 1981). When individuals had many choices, selected the alternative with few external pressures (high volition), and made irreversible decisions then commitment to the organization would be high. When these factors were absent (i.e. few choices, severe pressure, and a lack of irreversibility) then commitment would be low. A low level of commitment coupled with good opportunities for alternative employment would result in high levels of turnover among the recently entered group.
On the other hand, commitment by long tenure employees is based upon a whole stream of prior commitments made: the extent to which the person has developed firm-specific knowledge and skills including 'local' knowledge of the organization itself (when this is high, the decisions become irrevocable), the extent to which the person is locked in by non-portable compensation plans, and the extent to which current rewards and benefits are in line with aspirations, though this may always be high as aspirations adjust to jibe with the rewards and benefits available.
A number of researchers have found that different factors affect turnover at different levels of job tenure. Werbel and Gould (1984) have argued that organizational commitment is unrelated to turnover at the earliest stages of tenure (less than one year). In addition, the role of job characteristics (Katz 1978), the type of need sought, (Hall and Nougaim (1968), the role of the work group (Feldman, 1976) all differ in their association with attraction at different levels of tenure in the organization. Turnover research needs to take very seriously these different relationships. As a start a closer integration between the socialization and turnover literatures would be appropriate.
Not all turnover is alike. Staw (1983) has alerted is to the functional aspects of turnover: opening up positions so that new people with new ideas can enter the organization. This paper points to different causes for high turnover. These different causes can be easily identified by considering the level of labor stability:
- if stability is high; problem is in socializing newcomers
- if stability is low, problem is with the commitment of existing members of the organization
Footnotes
1The data described here are provided as illustrations of the phenomenon. They are not viewed as representative of the current state of labor turnover or stability. Nevertheless, the suggestions made for managing turnover in these different situations are sound. The organization will need to undertake its own diagnosis of its current situation to determine what the appropriate course of action should be.
References
Brissenden, P.F., & Frankel, E. (1922). Labor turnover in industry: A statistical Analysis. New York: Macmillan.
Dalton, D. R., Krackhardt, D. M., & Porter, L. W. (1981). Functional turnover:an empirical assessment. Journal of Applied Psychology, 66, 716-721.
Evans, M.G. (1963). Supervisory attitudes: A case study from the packaging industry, with an evaluation of the method used. Unpublished M.Sc.Tech. Thesis. Department of Industrial Administration, University of Manchester Institute of Science and Technology.
Evans, M.G. (1965) Supervisors' attitudes and departmental performance. Journal of management studies, 2, 174-190.
Feldman, D.C. (1976). A contingency theory of socialization. Administrative science quarterly, 21, 433-452.
Greenwood, T.F. (1919). Problems of industrial organization. Journal of the Royal Statistical Society, 82(2), 186-221.
Hall, D.T., & Nougaim, K. (1968). An examination of Maslow's need hierarchy in an organizational setting. Organizational behavior and human performance, 3, 12-35.
Katz, R. (1978). Job longevity as a situational factor in job satisfaction. Administrative science quarterly, 23, 204-223.
Lane, K.F., & Andrew, J.E. (1955). A method of labour turnover analysis. Journal of the Royal Statistical Society. Series A (General), 118(3), 296-323.
McCain, B.E., O'Reilly, C.A., III, & Pfeffer, J. (1983). The effects of departmental demography on turnover: The case of a university. Academy of management journal, 26, 626-641.
Mobley, W.H., Griffeth, R.W., Hand, H.H., & Meglino, B.M. (1979). Review and conceptual analysis of the employee turnover process. Psychological bulletin, 86, 493-522.
Pfeffer, J., (1983). Organizational demography. In L.L. Cummings & B.M. Staw (Eds.), Research in organizational behavior Volume 5, Greenwich, Conn.: JAI Press, 299-357.
O'Reilly, C.A., & Caldwell, D.F. (1980). The commitment and job tenure of new employees: Some evidence of post-decisional justification. Administrative science quarterly, 26, 597-616.
Salancik, G.R. (1977). Commitment and the control of organizational behavior and belief. In B.M. Staw & G.R. Salancik (Eds.), New directions in organizational behavior. Chicago: St Clair Press, 1-54.
Staw, B.M. (1980). The consequences of turnover. Journal of occupational behavior, 1, 253-273. ?1983?
Rice, A. K., Hill, J. M. M., & Trist, E. L. (1950) The representation of labour turnover as a social process; studies in the social development of an industrial community (the Glacier Project). II. Human Relations, 3, 1950, 349-370.
Werbel, J.D., & Gould, S. (1984). A comparison of the relationship of commitment to turnover in recent hires and tenured employees. Journal of applied psychology, 69, 687-690.
1. These extremes are not just logical constructions. They really have been observed. In my first piece of organizational research (Evans, 1963, 1965),1 I investigated turnover in a northern plant of a British paper packaging company (1700 employees). Although turnover was associated with managerial practices (Evans, 1965), there were also some interesting demographic effects. Turnover was in general confined to very specific groups of employees: young women with very short tenure in the organization. There were also some marked inter-departmental differences.
Overall, the establishment studied had an annual turnover rate of 63%. At that time, in the early 1960s, unemployment in the UK hovered at about 2.5% so the probability of a leaver obtaining another job was quite high. On the other hand, in the year studied, a surprising 70% of the work force present at the end of the year had also been present at the beginning of the year. In other words, labor stability was also very high. Most positions were filled with experienced people but 30% of the positions exhibited very rapid turnover: 200% per year!
2. In the Evans (1963) study, one department exhibited this pattern: labor stability was about 87% for both male and female employees; labor turnover was about 18% which was low for this organization at this time.
3. A second department from our early study (Evans, 1963) illustrates this situation. The labor stability of the male employees was 60%, yet labor turnover was 76%. This meant that for each of the 40% of positions at risk during the year about two employees had to be hired to keep the department fully staffed. Similar results were found for the female employees in the department (stability 71%, turnover 80%; so that 29% of the jobs were at risk, and each had to be filled by four persons during the year).
4. One of the departments studied (Evans, 1963) was in this position: it had a labor stability of 37% for its female work force, yet the labor turnover was 128%. This meant that for each of the 63% of positions at risk, two persons had to be hired during the year in order to keep the department fully staffed.
