Joel Brockner: Can Job Autonomy Be a Double-Edged Sword?

This post was originally published on the Psychology Today blog.

“You can arrive to work whenever convenient.”

“Work from home whenever you wish.”

“You can play music at work at any time.”

These are examples of actual workplace policies from prominent companies such as Aetna, American Express, Dell, Facebook, Google, IBM, and Zappos. They have joined the ranks of many organizations in giving employees greater job autonomy, that is, more freedom to decide when, where, and how to do their work. And why not? Research by organizational psychologists such as Richard Hackman and Greg Oldham and by social psychologists such as Edward Deci and Richard Ryan, has shown that job autonomy can have many positive effects. The accumulated evidence is that employees who experience more autonomy are more motivated, creative, and satisfied with their jobs.

Against this backdrop of the generally favorable effects of job autonomy, recent research has shown that it also may have a dark side: unethical behavior. Jackson Lu, Yoav Vardi, Ely Weitz and I discovered such results in a series of field and laboratory studies soon to be published in the Journal of Experimental Social Psychology. In field studies conducted in Israel, employees from a wide range of industries rated how much autonomy they had and how often they engaged in unethical behavior, such as misrepresenting their work hours or wasting work time on private phone calls. Those who had greater autonomy said that they engaged in more unethical behavior on the job. In laboratory experiments conducted in the United States we found that it may not even be necessary for people to have actual autonomy for them to behave unethically; merely priming them with the idea of autonomy may do the trick. In these studies participants were randomly assigned to conditions differing in how much the concept of autonomy was called to mind. This was done with a widely used sentence-unscrambling task in which people had to rearrange multiple series of words into grammatically correct sentences. For example, those in the high-autonomy condition were given words such as, “have many as you as days wish you vacation may” which could be rearranged to form the sentence, “You may have as many vacation days as you wish.” In contrast, those in the low-autonomy condition were given words such as, “office in work you must the,” which could be rearranged to, “You must work in the office.” After completing the sentence-unscrambling exercise participants did another task in which they were told that the amount of money they earned depended on how well they performed. The activity was structured in a way that enabled us to tell whether participants lied about their performance. Those who were previously primed to experience greater autonomy in the sentence-unscrambling task lied more. Job autonomy gives employees a sense of freedom which usually has positive effects on their productivity and morale but also can lead them to feel that they can do whatever they want, including not adhering to rules of morality.

All behavior is a function of what people want to do (motivation) and what they are capable of doing (ability). Consider the unethical behavior elicited by high levels of autonomy. Having high autonomy may not have made people want to behave unethically. However, it may have enabled the unethical behavior by making it possible for people to engage in it. Indeed, the distinction between people wanting to behave unethically versus having the capability of doing so may help answer two important questions:

(1) What might mitigate the tendency for job autonomy to elicit unethical behavior?

(2) If job autonomy can lead to unethical behavior should companies re-evaluate whether to give job autonomy to its employees? That is, can job autonomy be introduced in a way that maximizes its positive consequences (e.g., greater creativity) without introducing the negative effect of unethical behavior?

With respect to the first question, my hunch is that people who have job autonomy and therefore are able to behave unethically will not do so if they do not want to behave unethically. For example, people who are high on the dimension of moral identity, for whom behaving morally is central to how they define themselves would be less likely to behave unethically even when a high degree of job autonomy enabled or made it possible for them to do so.

With respect to the second question, I am not recommending that companies abandon their efforts to provide employees with job autonomy. Our research suggests, rather, that the consequences of giving employees autonomy may not be summarily favorable. Taking a more balanced view of how employees respond to job autonomy may shed light on how organizations can maximize the positive effects of job autonomy while minimizing the negative consequence of unethical behavior.

Whereas people generally value having autonomy, some people want it more than others. People who want autonomy a lot may be less likely to behave unethically when they experience autonomy. For one thing, they may be concerned that the autonomy they covet may be taken away if they were to take advantage of it by behaving unethically. This reasoning led us to do another study to evaluate when the potential downside of felt autonomy can be minimized while its positive effects can be maintained. Once again, we primed people to experience varying degrees of job autonomy with the word-unscrambling exercise. Half of them then went on to do the task which measured their tendency to lie about their performance, whereas the other half completed an entirely different task, one measuring their creativity. Once again, those who worked on the task in which they could lie about their performance did so more when they were primed to experience greater autonomy. And, as has been found in previous research those who did the creativity task performed better at it when they were primed to experience greater autonomy.

Regardless of whether they did the task that measured unethical behavior or creativity, participants also indicated how much they generally valued having autonomy. Among those who generally valued having autonomy to a greater extent, (1) the positive relationship between experiencing job autonomy and behaving unethically diminished, whereas (2) the positive relationship between experiencing job autonomy and creativity was maintained. In other words, as long as people valued having autonomy, the experience of autonomy had the positive effect of enhancing creativity without introducing the dangerous side effect of unethical behavior. So, when organizations introduce job autonomy policies like those mentioned at the outset, they may gain greater overall benefits when they ensure that their employees value having autonomy. This may be achieved by selecting employees who value having autonomy as well as by creating a corporate culture which emphasizes the importance of it. More generally, a key practical takeaway from our studies is that when unethical behavior is enabled, whether through job autonomy or other factors, it needs to be counterbalanced by conditions that make employees not want to go there.

BrocknerJoel Brockner is the Phillip Hettleman Professor of Business at Columbia Business School. He is the author of The Process Matters: Engaging and Equipping People for Success.