CAMBRIDGE, Mass., December 1, 2011—There are many explanations for the persistence of workplace inequality such as glass ceilings and old boys’ clubs, yet little is known about the role managers play in shaping employee performance assessments, a common procedure used by the majority of companies in the U.S. to distribute rewards and make career decisions. New research by MIT Sloan School of Management Prof. Emilio J. Castilla sheds light on this piece of the puzzle, finding three distinct ways that managers’ influence might affect workplace inequality.
“Managers play a key role in employees’ careers, as they decide employee job and project assignments, training and development possibilities, pay raises, promotions, and terminations. But how exactly do managers impact the workplace treatment and assessment of workers below them over the course of their careers?” asks Castilla.
To answer this question, he analyzed personnel data and thousands of actual employee performance evaluations at a large private U.S. company over the course of several years. He examined the potential effects of three mechanisms of potential managerial influence: 1) the social network influence between employees’ current and former managers; 2) the demographic similarity of current and former managers in terms of gender, race or nationality; and 3) the demographic similarity between the current manager and employee.
Castilla found evidence of all three mechanisms of managerial influence on the outcome of managers’ disagreement in the performance evaluation ratings of the same worker. Specifically, his research showed that managers in the same social network―who were structurally connected within the organization―tended to disagree less in their performance ratings of the same employee than independent managers. He also found that disagreement in the ratings of the same employee was reduced among demographically similar managers.
When managers were structurally connected within the organization, the probability of obtaining the top performance rating was 9.6% higher for all demographic groups. When former and current managers were both male, the probability of obtaining the top performance rating jumped to 20.7% higher. When those managers were both white, the probability of a top rating was 21.1% higher compared with managers who were racially different.
As for manager-employee demographics, Castilla found that white male and female employees born in the U.S. had the greatest advantage. “When all three mechanisms are studied together …, the highest agreement scenario occurs when former and current managers are connected, and both managers and the evaluated employee are all U.S.-born white males,” he writes, noting that if a worker obtained a top performance rating the prior year, the probability of getting the same rating was 68.1% higher compared to an employee with independent managers who were demographically different from each other and the employee.
In this particular company setting, Castilla found that African-American and Hispanic women working for female independent former and current managers were the least advantaged with only a 14.4% probability of obtaining the top performance rating. As managerial influences in the assessment process in turn affect workers’ final performance ratings on which rewards are typically based, Castilla also discusses the implications of his study for understanding pay disparities, especially in companies that use performance-reward practices.
“My results stress how managers’ evaluations of their employees do not occur in a vacuum but within the social context that surrounds them,” writes Castilla. “My study also underscores that these mechanisms complement each other and may operate simultaneously in many organizational settings and procedures.”
He adds, “Within the limits of my study’s (non-experimental) design, these findings highlight the complex social and organizational web in which managers and employees interact, which is particularly relevant in light of ongoing diversity efforts to increase the representation of women and ethnic minorities in management. While researchers have documented a diverse range of explanations for the persistence of workplace inequality in spite of organizational policies and even legislation meant to promote fairness and increase diversity in the workplace, the potential influence of managers should not be overlooked.”
Castilla’s article, Bringing Managers Back In: Managerial Influences on Workplace Inequality, was published in the October issue of the American Sociological Review and can be found at: http://asr.sagepub.com/content/76/5/667.abstract
For more information on Prof. Castilla, please visit: http://ecastill.scripts.mit.edu