The Morning Report
Get the news and information you need to take on the day.
Tuesday, July 21, 2009 | Given San Diego’s continuing spiral toward fiscal insolvency, some elected officials and local political observers have been engaged in an important debate about the generosity of city worker compensation. Those who claim that public employees and their unions are responsible for many of our financial problems have been encouraged by a controversial analysis of city payroll records recently published in the Union-Tribune.
Yet, to date, no one has thoroughly addressed the key question at the heart of the debate: Compared to their peers elsewhere, are San Diego city employees paid too much? By considering the compensation practices of nearly 900 major cities across the country, I hope to take a first step to finding the answer.
The tables below compare the public-sector wages of San Diego employees to those in other cities. The data include all salaries, wages, fees, commissions, bonuses, and awards paid to employees during the month of March 2007. (One important caveat that the reader should keep in mind is that the data do not include non-wage compensation. While such statistics are collected by the Bureau of Labor Statistics, they are confidential and not available to the public.)
The first table reports San Diego’s total payroll expenditures per 100,000 residents, the most substantively interesting statistic, and compares it to the averages across other cities. Because cities differ greatly in the types of services they provide — some own their own utilities and ports, other run local schools — I break the totals down by major government functions to facilitate comparison. Focusing on payroll by function provides the best apples-to-apples comparison. In addition, I report information on the number of government employees, and the average wage per employee.
The final column reports San Diego’s percentile ranking: For example, our monthly payroll expenditures on libraries (per 100,000 residents) put us in the 33rd percentile, meaning that 33 percent of cities spend less while 67 percent spend more.
Several important numbers stand out. First, San Diego’s monthly payroll costs fall below the average for most major categories of government services. Second, the numbers suggests that we employ far fewer employees than other cities — but pay those we do hire significantly more.
Limiting our examination to the 25 largest cities in California, some similar trends emerge. Again, San Diego’s total payroll costs are significantly lower — indeed, even more so. Compared to other cities in the state, however, the average San Diego employee also makes significantly less than their counterparts. There is no clear trend for employment levels.
However, even these comparisons may not provide a fair benchmark. Different cities vary in the preference of their residents for government services, their financial capacity, the conditions of their local labor markets, and their needs. All of these factors could — and in a functioning democracy should — affect how many workers they employ and how much these workers are paid.
In an effort to compare like to like, I repeated the above analysis by considering cities from across the country that are most similar to San Diego. In particular, I selected cities with the most similar population density, unemployment levels, per-capita city revenues, and average per-capita income. The results are reported in the table below and are generally similar to those discussed previously.
Collectively, these tables point to several important conclusions. First, San Diego’s payroll does not suggest that the city spends more on the salaries of its workers than other local governments across the country. If anything, our overall payroll costs are toward the lower end of the spectrum.
Second, the analysis highlights the importance of using the right benchmark to make claims about local policies. Comparing San Diego to other California cities, for example, may lead to different conclusions than comparing us to cities outside of the state. We should be especially wary of efforts — including Councilman Carl DeMaio’s study of employee fringe benefits — that contrast San Diego to the “national average.” Any such comparison implicitly assumes that San Diego is an “average” city. Given our atypical size, population, and wealth, this assumption should be problematic on its face.
Finally, the aggregate data point to the danger of drawing inferences from the experiences of individual workers. Focusing too much on the trees may make us miss the forest. Though, individually, San Diego workers may indeed earn more than public employees elsewhere, the city’s total payroll costs still appear to be lower. One possible, but far from the only, explanation is that San Diego city employees are simply more productive, and are rewarded for this productivity with higher pay.
I end with a word of caution: None of the above discussion is meant to suggest that the city’s compensation practices are correct, fair, or equitable. Indeed, labor economists have long pointed out that public sector workers tend to make more than their colleagues in private practice. Decisions about the proper wages for city workers should involve wide-ranging debate and discussion, and are most appropriately left to local officials who are elected to make contentious and weighty decisions on our behalf.
The data do, however, suggest that efforts to balance the city budget by winning significant wage concessions face a very real risk of creating a “brain drain,” with long-time city workers leaving for other jurisdictions that offer more competitive pay. It’s a risk that we all should take seriously.
Notes on methodology: The data on government wages and levels come from the 2007 Census of Government Employment. Some cities do not report any payroll expenditures for certain categories of services, either because they do not provide these services or because they contract their provision out to other government and private entities. These cases are not included in the comparisons. Population figures are from 2005 and come from the Legal Landscapes Database. Information on population density, unemployment, and city revenue are from the 2000 County and City Data Book. The per-capita income numbers are from the 2006 American Community Survey. To identify the most “similar” cities, I created ordinal rankings of cities for each variable. I then included the ten cities immediately before and immediately after San Diego on that ranking, repeating the process for each variable. The main dataset used in this analysis can be downloaded from here.
Vladimir Kogan is a doctoral student at UCSD’s Department of Political Science. He is a co-author of a forthcoming book about San Diego politics and its pension crisis. His e-mail address is email@example.com.