When companies go public with salary information, the average productivity of their employees does not decline. But small shifts in work output are highly individualized, and they may reflect workers’ responses to how closely they feel their efforts align with the pay.
A recent study in Strategic Management Journal is one of the first to investigate the effects of pay transparency in the field. The authors drill down to how individual workers respond when they see their salaries stacked up to those of their peers—both in proximal positions and up and down the vertical hierarchy.
“In the organizations we’ve studied, we find pay transparency has a negligible impact on aggregate productivity,” said study coauthor Todd Zenger, PhD, the N. Eldon Tanner Chair in Strategy and Strategic Leadership and Presidential Professor at the David Eccles School of Business at the University of Utah. “But the interesting story is the individual variance in response.”
The research team found that those responses depend on what transparency reveals to individual employees about their pay and its fairness. Workers who discovered they were ‘underpaid’ in comparison with their peers tended to respond with a slight decrease in output, while ‘overpaid’ workers appeared—at least temporarily—to start working harder.
“Employees who found they were paid more than their performance warranted increased their productivity, likely to justify their elevated compensation,” observed lead author Cédric Gutierrez Moreno, PhD, an assistant professor in the Department of Management and Technology at Bocconi University in Milan, Italy.
Among organizations that are hesitant to publish salary data, a common concern is that transparency will lead to a decline in work output, Dr. Gutierrez explained. But he and his colleagues assert that these concerns may be unwarranted. They instead emphasize the importance of clearly delineating the effects driven by equity—rewarding workers fairly in accordance with their contributions—as opposed to homogenized equality.
Their study was limited to workers in academia: it tracked the productivity and salary data of nearly 20,000 individuals employed by public universities across 116 institutions in eight U.S. states. They ranged across academic domains, in fields from engineering and biological sciences to business, languages and the performing arts.
Using data from university workers provided an advantage that may not be feasible in some other fields—for example, one couldn’t scour individual performance reviews for productivity indicators among office workers. Not only is the information confidential, it’s also subjective and prone to supervisor bias, and what counts as “productive” could vary widely depending on the scope of the job.
But productivity in academia is easier to observe—when it’s measured in academic achievements. The researchers generated their productivity index by aggregating the individuals’ published academic articles, academic awards, and published books or book chapters.
“Because the sheer number of articles may mask important heterogeneity in quality, we scaled this measure by the impact factor of the journal in which they were published,” the investigators noted.
Coauthor Tomasz Obloj, PhD, an associate professor and Weimer Faculty Fellow at Indiana University’s Kelley School of Business, illuminated some of the limits of these productivity metrics. The team’s available data were mostly related to research activities or awards by professional organizations, and didn’t include, for example, teaching performance or institutional service.
“While there are important productivity metrics that we don’t observe in our data, we capture most of the research output that R1 institutions are likely to rely on to evaluate their tenure-track faculty’s productivity,” Dr. Obloj explained.
The research team framed their analysis around the timing of transparency “shocks,” periods when it was safest to assume that most employees knew their institutions’ salary reports had become available for viewing. For each state in their study sample, the researchers identified the date when aggregated salary data for university employees were first made public.
In principle, salary information for employees of most public universities has ostensibly been “public record” since the 1967 Freedom of Information Act, the investigators noted. But before the advent of online access via newspaper articles, think tanks, and state agency websites, getting a hold of that information cost effort and money.
“Such aggregating repositories gathered a lot of public and media attention, making those events salient and thus potentially triggering the mechanisms that we seek to test, thus serving as valid exogenous shocks,” the authors observed. “For example, The Sacramento Bee’s University of California salary database received 6 million hits in the first 2 months after its launch.” Within a 6-month window, Google searches for the Bee and the Texas Tribune, newspapers that published salary data in the two largest states, were highest around the date of those publications.
With each state’s salary reveal serving as “treatment,” the researchers were able to compare how productivity changed as “untreated” employees became “treated.” The team recognized that the degree of change could be influenced by myriad factors—such as an institution’s level of emphasis on academic output, or the university’s rank among other schools across the country. They sought to identify and control for as many of these variables as possible.
To account for wage inequality, the investigators categorized academics’ salaries based on the median salary of all academics present in the same department and holding the same academic rank—for example, assistant professor, associate professor, or full professor.
“Our results suggest that individuals primarily responded to wage inequity rather than inequality,” said Dr. Obloj. “By inequity, we mean unfairness in how pay reflects performance, not just differences in pay levels.”
“Faculty adjusted their productivity based on whether their compensation fairly reflected their performance, not simply based on having a relatively high or low salary compared to peers,” Dr. Obloj continued. “Individuals who were ‘overpaid’ demonstrated an increase in productivity between 5% and 13% depending on the productivity output.”
In addition, the team’s results suggest that employees ramp up their productivity when salary reports reveal that a promotion could come with a substantial salary increase.
Importantly, the researchers also found heterogeneous responses across academic ranks. ‘Overpaid’ faculty increased productivity regardless of rank. Among ‘underpaid’ individuals, the negative productivity response was strongest among tenured full professors, but virtually nonexistent among assistant professors. These responses suggest that job security plays an important role in how individuals respond to pay transparency, Dr. Gutierrez noted.
The research findings add a rich layer of empirical data to the growing body of literature on the effects of pay transparency. The authors hope that transparency practices will not only continue to shine light on pay inequities, but also serve as a catalyst for change that will in turn steer future positive productivity.
“An initial productivity response may reflect what each employee discovers about how they are treated,” Dr. Gutierrez explained. “But if, in response to the transparency, the pay structure changes, those initial productivity responses may dissipate as inequities are addressed.”
