PRINCETON, N.J.--In the era of big data, transparency has become a popular policy tool for addressing potential problems. But publicly disclosing personal information -- such as government officials' income -- may result in unintended consequences.
Using California as a case study, a researcher from Princeton University's Woodrow Wilson School of Public and International Affairs shows that city managers -- typically the highest-paid city employees -- saw an 8 percent reduction in pay after their salaries were disclosed to the public. These cuts also triggered a 75 percent increase in the quit rate among city managers.
The findings, released as a working paper by the National Bureau of Economic Research, suggest that top salaries are cut because they appear excessive, regardless of whether the reductions in pay are good policy. Additionally, the research suggests that media exposure restrained high wages in cities where the top salaries were already disclosed.
"On the surface, transparency seems unambiguously good. Why wouldn't we have transparency?" said Alexandre Mas, professor of economics and public affairs and author of the paper. "This paper shows that there may be unintended effects from these policies. If the public has an averse response to large salaries, regardless of whether these salaries are justified, there might be adverse consequences."
The average salary of a city manager is around $200,000, and one made as much as $800,000. City managers are responsible for the day-to-day operations including managing the city's budget and promoting economic development. In California, there is no legal limit to what a city manager can earn.
Mas selected California for a case study because, in 2010, the state issued a mandate requiring all cities to post municipal salaries online. While some cities had previously disclosed municipal salaries, the mandate introduced disclosure in the remainder of the cities. Mas used an Internet database of historical webpages and newspaper archives to research which cities disclosed wages, and he used public records requests to gather payroll information. He then compared the evolution of wages between cities that had and previously had not disclosed city manager wages. Once their wages were disclosed, city managers saw an average pay cut of about 8 percent, according to Mas' calculations.
While the timing of the policy coincided with the aftermath of the Great Recession, Mas shows that the reductions came not through citywide furloughs or budget cuts but from nominal pay cuts over time.
"One explanation is that city managers, in general, are overpaid, so public disclosures forces compensation to be appropriately corrected," Mas said. "A second explanation is that the public views these salaries as excessive, regardless of whether they are warranted. This reaction, in turn, pressures the city council to lower the salaries." Mas finds that the evidence points more toward the second explanation.
Mas also wanted to determine whether publicly disclosing salaries caused city managers to leave their positions. Using state records and press reports, he found that, following the 2010 mandate, there was a 75 percent increase in the quit rate of city managers. These numbers suggest that city managers might have been at their tipping point on the salary scale.
"The salaries these city managers earned were enough to keep them there, but the pay cuts were enough for them to leave," Mas said. "It suggests that these high-level officials had better outside opportunities."
There was no response in pay setting among cities that, prior to the mandate, already disclosed their compensation to the press. As Mas argues, this suggests that the press already had a role in restraining top salaries.
Interestingly, Mas found that wage declines mostly came from male managers. On average, compensation did not decline for female managers.
"We can really only speculate why this might be the case," Mas said. "It is possible that male managers did more to use secrecy to inflate their pay, but it is also possible that city councils believed there to be a higher risk of a lawsuit by female city managers on the basis of discrimination, so they didn't cut their wages."
Mas said this work sets the stage for further research related to the private sector -- for example, whether chief executives of publically traded companies have lower compensation than they would otherwise because their compensation is disclosed by the Securities and Exchange Commission.
"More work needs to be done to investigate other effects of pay disclosure, including whether it closes gender and race gaps and whether transparency changes the relative bargaining ability of workers versus employers when wages are being set," Mas said.
The paper, "Does Transparency Lead to Pay Compression?" was published online October 2014 as a working paper by the National Bureau of Economic Research.