Myth: It’s All About the Benjamins
Understanding What Really Drives Turnover Decisions
Benjamin is troubled. Generally speaking, he likes working for the organization, but lately he’s been feeling like his boss doesn’t really appreciate him. Because of this, he is worried about his opportunities for advancement. Benjamin decides to start looking for other possible jobs. He currently makes $40,000 per year; thus, his job search only targets opportunities that make $40,000 or more. He would have to be REALLY unhappy to consider taking a pay cut! After a few months of searching, Benjamin finds another opportunity that pays $46,000 per year. After weighing the pros and cons of staying or leaving, he decides to take the new job. When his boss asks him why he is leaving, Benjamin says it is because he has found a new job paying 15% more. Benjamin’s boss approaches Manager Green. She says that a valued employee is leaving because of his compensation; is there anything the organization can do? Now Manager Green is troubled. He would like to retain Benjamin, and the organization could probably counter-offer a few thousand dollars. However, Benjamin is already near the top of his pay band, and the organization is hesitant about setting precedents that encourage employees to seek outside offers for the purpose of leveraging counter-offers. Maybe it is time to revamp the compensation system! Manager Green approaches Manager Savvy to discuss compensation system design, but Manager Savvy asks a very insightful question: is Benjamin really leaving because of his pay?
Kernel of Truth
When we ask managers, executives, HR professionals, and students why people quit jobs, pay is almost invariably the first or second reason provided. It is true that some people quit because they are unhappy with their pay. It is also true that people often quit in order to take higher paying jobs elsewhere, although, as described in the example above, this can be misleading. Benjamin likely would never have been searching for that higher paying job if he had felt his boss was looking out for his best interests. It is also true that there are ways of structuring pay that can affect retention. For example, evidence shows that how widely pay is dispersed among employees affects the tenure of managers and the likelihood of their leaving the organization such that turnover is higher when the spread between average employee compensation and high employee compensation is wider.1 We discuss creative ways to structure compensation to manage turnover in more detail in Chapter 10. However, the bulk of the research evidence suggests that pay may not be nearly as important as many managers believe.
What the Research Says
We reviewed the most up-to-date research summarizing the results of hundreds of peer-reviewed studies of predictors of individual turnover decisions (summarized in Figure 2.1).2 Out of 41 predictors, level of pay was tied for the 30th strongest relationship with turnover. Ok, so maybe it is not the level of pay that matters. After all, some people might be quite happy and satisfied with relatively modest pay. Others might be very dissatisfied even with very high pay, especially if the person working near you doing the same job makes more. Evaluations of pay are probably relative. Surely, then, how satisfied or dissatisfied I am with my pay should be a better predictor of turnover? Nope. Out of 41 predictors, pay satisfaction showed the 35th strongest relationship with turnover!
Note: A plus (+) indicates that the predictor is positively related to turnover. (As the predictor increases, so does turnover.)
A minus (–) indicates that the predictor is negatively related to turnover. (As the predictor increases, turnover decreases.)
Figure 2.1. Voluntary turnover predictors.
Source: Reprinted with permission, SHRM Foundation.
Conclusion: despite the widespread belief that pay is an important driver of turnover, pay level and pay satisfaction are relatively weak predictors of individual turnover decisions. So, what are stronger predictors of who is likely to quit? The studies cited above suggest three primary categories of predictors that are more strongly related to turnover: the withdrawal process, key job attitudes, and the work environment.
Research consistently demonstrates that the predictors with the strongest relationships with turnover are those related to the withdrawal process, notably turnover intentions, individual mobility, and job search. Although some individuals may quit jobs quickly and impulsively, most go through one or more steps of psychological or behavioral withdrawal first.3 For example, individuals may experience thoughts of quitting, search for alternatives, evaluate possible alternatives against their current job, and develop intentions to quit, meaning they definitely plan to quit as soon as a preferable opportunity presents itself. Being aware of the importance of the withdrawal process is important for managing retention for two reasons. First and foremost, it allows savvy managers to intervene in the withdrawal process before it is too late. Discovering that a valued employee is leaving after they have found a preferable alternative and decided to leave is often too late. However, there are methods of measuring withdrawal process variables that would enable managers to take action before employees decide to quit. Second, there is considerable research evidence about the drivers of turnover intentions and job search, and the good news is that this research is quite consistent with the drivers of ultimate turnover.
Research also consistently demonstrates that key job attitudes, notably job satisfaction and organizational commitment, are relatively strong predictors of turnover and the withdrawal process. Job satisfaction is a positive emotional state resulting from the appraisal of one’s job or job experiences.4 Organizational commitment is the employee’s psychological attachment to the organization.5 Many organizations already measure these attitudes. If you do, good! If you don’t, the research suggests that you should. Our experience suggests two key improvements many organizations can make in how they assess job satisfaction and organizational commitment. One is to be sure to use well-developed measures with substantial validation evidence. Many organizations rely on a single question to tap a complex multifaceted idea such as job satisfaction (e.g., an employee could be quite satisfied with their tasks but dissatisfied with their work environment, or vice versa) or make up their own questions without putting substantial research behind them. There are many excellent measures of both satisfaction and commitment available in the published research domain that have already been subjected to extensive testing and validation and are freely available. (Validation refers to the process of ensuring the soundness of a particular measure and for predicting outcomes important to organizations such as turnover.) The other is to measure more frequently and link individual responses to important outcomes. Many organizations assess attitudes only once a year, on anonymous surveys, with results only analyzed at the department or business unit level. Although linkage research of this nature has some value, we have found that more frequent assessment at the individual level allows for more timely and targeted interventions. There can be large differences between what is happening across business units and what is happening to the individuals within any given unit. This methodology requires building up substantial trust in the workforce and/or working with outside researchers to collect such data.
The next strongest class of turnover predictors includes key variables related to the work environment, notably aspects of leadership, work design, and relationships with others. Have you heard the truism “people leave bosses”? There is considerable research evidence to back this up. The strength of the relationship an employee has with their immediate supervisor/manager is one of the most consistent predictors of turnover. Research suggests that leaders help shape the roles that followers will have, and this process of role determination is called leader–member exchange (LMX).6 This research suggests that leaders treat some subordinates as part of their “in-group” with extensive trust and access to resources, and others as part of an “out-group” with a more transactional relationship. Those in the in-group are substantially less likely to leave. Leaders made aware of these important distinctions in their interactions with subordinates and the subsequent impact on attitudes and turnover can better manage their relationships with their followers.
In terms of work design, research suggests that role clarity and role conflict are two of the most consistent predictors of turnover. Role theory7 suggests that individuals hold multiple roles that come with expectations that influence behavior. When there is a lack of clarity (such as ambiguous department goals) concerning role expectations or when role expectations are in conflict with each other (such as having to report to two bosses with competing agendas), individuals can experience stress, burnout, and dissatisfaction, and are more likely to quit. Leaders and organizations need to ensure that expectations are clearly communicated and supported. The other element of work design that is directly relevant to the case of Benjamin at the start of this chapter is that opportunities for advancement are also consistently related to turnover decisions.8 Individuals who believe that there are future opportunities for growth and advancement are more likely to stay—even if they are not completely satisfied with their present circumstances. Organizations would benefit from proactively managing career paths and opportunities, and leaders need to communicate with their employees about these opportunities.
Beyond supervisors, research evidence is also beginning to accumulate that relationships with others in the workplace are important for retention.9 Satisfaction with co-workers and work group cohesion are two of the more consistent predictors of individual turnover decisions. As described in more detail in Chapters 3 and 8, employees can become embedded in a network of relationships at work that make it less likely they will leave. Managers can create opportunities for interaction and design work to foster cohesion. For example, research demonstrates that designing socialization tactics so that newcomers interact with other new hires and have positive interactions with experienced organizational members increases the sense of being embedded in the organization, which increases retention.10
Evidence-Based Management Implications
In summary, extensive research evidence on individual turnover decisions suggests several key points for savvy managers to keep in mind.
•Pay level and pay satisfaction are relatively weak predictors of individual turnover decisions. The use of rewards to manage retention is discussed in more detail in Chapter 10.
•Indicators of the withdrawal process are the strongest predictors of individual turnover decisions. Organizations should measure and manage employee mobility, job search, and turnover intentions.
•Job satisfaction and organizational commitment are key attitudes that are consistent predictors of individual turnover decisions. Organizations should measure and manage both satisfaction and commitment. There is extensive research evidence available for the savvy evidence-based manager on the drivers of these attitudes.11
•When measuring attitudes and withdrawal, organizations should use well-developed measures with validation evidence, assess more frequently than annually, and link individual responses to individual behaviors and outcomes.
•The nature of the relationship with one’s immediate supervisor is a consistent predictor of individual turnover decisions. Organizations should provide leadership training to all supervisors and managers, and hold leaders accountable for retention. The practical implications of leadership for retention are explored in more detail in Chapter 11.
•Individuals with clear role expectations, minimal role conflict, and opportunities for growth and advancement are less likely to quit. Organizations should train managers on the importance of providing clear role expectations, design organizational processes to minimize role conflict, and develop and communicate career paths, especially to highly valued employees.
•Individuals linked by positive relationships with others in the organization are less likely to quit. Organizations and managers should work to foster positive relationships among co-workers, provide opportunities for interaction, and help newcomers form and develop relationships. The practical implications of socializing newcomers on retention are discussed in more detail in Chapter 8.
Final Thought
It should be good news that pay is not the most important driver of turnover. Revamping compensation systems, paying considerably above market, and throwing money at valuable employees can be complicated and expensive. Many of the recommendations provided in this chapter are less expensive to implement, more likely to have a positive impact, and thus, likely to provide a greater return on investment. Returning to the case of Benjamin at the start of this chapter, research suggests it would have been more effective and less costly to proactively communicate with Benjamin about future opportunities than to try to come up with a counter-offer or risk losing a valued employee.