Ever since I started tabulating and analyzing data from the Governments division of the U.S. Census Bureau in the early 1990s, I’ve found that the results are something that no one of any political persuasion has wanted to see. My former supervisors at City planning once presented the data, along with a bunch of other information, to former NYC economic development czar John Dyson. He was so upset that most of NYC’s extra local government taxes as a share of personal income, over and above the U.S. average, went to categories of spending that no business person would consider “public services” that he started ranting “New York stinks” at a conference that very evening. When presented with the same data his boss, former Mayor Giuliani, was not happy to see that NYC’s staffing and spending levels were so high for police and low for education, given that he wanted to make them higher and lower respectively. A housing advocate and city planner grew upset with me when I pointed out New York City’s comparatively high spending on housing and low spending on services such as education and parks, accusing me of immorally “trading off one need against another,” as if that wasn’t what the data showed had already happened. And when I presented the data to the City Planning Commission as part of the preparation of the charter-mandated Planning and Zoning Report, former commissioner Ron Shiffman felt the comparison with the U.S. average made NYC look unfairly bad. “At least compare us with other cities for God’s sake,” I recall him saying. Very well Professor Shiffman, this post and the attached tables contain the comparison you asked for.
As noted in my first post on local government and payroll data for 2007, comparing city governments is misleading, because some cities also have local government services provided by separate counties, school districts and special districts. There are, however, several older cities where the city boundary is co-terminus (or nearly so) with an entire county (or its equivalent), making a direct central city comparison with New York City possible. The cities of San Francisco and Philadelphia swallowed up all of San Francisco and Philadelphia counties before their annexation ceased. The city of Boston is nearly co-terminus with Suffolk County in Massachusetts, off by only a few small communities. St. Louis and Baltimore got themselves completely separated from St. Louis and Baltimore Counties, and have data provided separately from them. The District of Columbia includes not only all the functions of local government but also all the functions of state government (the data for it in the table excludes state-only functions such as Social Insurance).
The attached spreadsheet includes data for all of these counties/cities, along with New York City and the U.S. average. Included is data on March 2007 local government employment per 100,000 residents by category of public service, and payroll per employee for that month relative to the national average. Private-sector payroll per employee, with and without the high-paid Finance and Insurance sector included, for a comparison with relative local government pay. I asked the relevant state departments of labor for data on private-sector industries related to local government, and received cooperation from Maryland and Pennsylvania. That data is presented for Baltimore and Philadelphia, but not the other cities.
So what does the data tell us?
Low public school employment per 100,000 residents, relative to the national average, is common in most older central cities. Poor cities such as Philadelphia and Baltimore have poor schools that drive parents with choices out of town; those children who remain are often powerless and ill served. In some cases the schools have been by reputation pension-rich work-poor sinecures for well connected people living in the suburbs or in certain areas of the city. Middle class and affluent residents of these cities do not send their children to public schools; note the above average number of private school workers per 100,000 residents in Baltimore and Philadelphia. In more affluent cities, such as Washington, Boston and San Francisco, there are often relatively few children of any racial and economic background, as families are outbid by the college-educated young, affluent empty-nesters, gays and others less likely to have children. In San Francisco the city is making an effort to attract more families with children, without success. Having relatively few children to educate, however, saves San Francisco a lot of money, which is spent elsewhere.
It appears that the cities of St. Louis and San Francisco operate substantial institutions of higher education. In St. Louis, the city school system is considered by the rest of the state to be mismanaged, and last I read a state takeover was being debated. St. Louis had the highest non-instructional employment per 100,000 residents in this table, although the number for St. Louis is not as high as in Upstate New York.
Not surprisingly, a relatively high number of police officers per 100,000 residents is typical of all the older central cities, and not just New York City. In all of these cities except Washington DC, however, the relative number of police officers was much lower than in New York City. And in all of them police officers were much better paid, except perhaps in St. Louis where the average police officer earned 13.2% less than the national average but the average private sector worker in the metro area as a whole earned just 2.5% less than average. Evaluating fire department employment is difficult because much of the country is covered by volunteer rather than professional fire protection. Compared with other central cities with professional fire departments, however, New York’s fire department employment per 100,000 residents was typical.
Most of the other cities didn’t have as much public welfare employment as New York City, probably because the public welfare function is handled primarily by state governments there. Washington DC was as high as New York, but that is combined city-state government. San Francisco was also very high. Looking to the private sector, both Baltimore and Philadelphia had massive hospital employment relative to population, likely because their hospitals serve the metropolitan area and not just city residents. To be fair that is also true, to a lesser extent, in NYC. Neither of those cities had comparatively high public hospital employment on top of the private hospital employment, as in NYC. Private employment in Philadelphia and Baltimore was lower than in NYC in the home health care and social assistance industries. Much of NYC’s excess social assistance employment is actually Medicaid-funded personal care aides for senior citizens. Neither Pennsylvania nor Maryland have anything like New York’s level of Medicaid spending, and neither burden poorer communities by shifting Medicaid costs to the local level.
New York’s very high employment in housing and community development, relative to the national average, was similar to Washington, Baltimore, and Boston, with Philadelphia also well above average. These cities have significant public housing authorities; Washington also has a massive redevelopment program. In St. Louis, there is plenty of private housing available at low cost, making public housing undesirable and increasingly abandoned. St. Louis was one of the first cities to tear down a public housing development, the Pruitt-Igoe development, in 1972. Last time I checked St. Louis had lost a higher share of its peak population than any other U.S. city, though Detroit may pass it this decade.
Like Upstate New York, San Francisco had a lot of people working on the streets and roads, with employment 70% above the national average. Employment in the San Francisco Muni system was higher, relative to the population of that city, than New York City Transit’s employment was relative to NYC, even though BART and CALTRANS also serve the City by the Bay. According to a left-wingish transit fan from that city I correspond with, work practices are negligent and absenteeism high on the Muni, bad enough to make a lefty willing to privatize the whole thing or at least declare war on the union. All of these cities save Boston have many more people than NYC working on their water and sewer systems, though I am uncertain as to whether their water departments also serve the suburbs.
All of the other cities save Washington DC picked up the trash with fewer solid waste management employees, relative to population, than NYC. All of the other cities had significantly more local government workers in parks, recreation, culture and libraries than NYC, except that parks employment was even lower in Boston.
Employment in public health and general bureaucracy in most of the other cities was similar to NYC and the national average, except in Washington DC, where the inclusion of state government functions provides an explanation, and in San Francisco, which has a high level of public health employment. According to the tabulation of state government employment attached to this post, the relatively high level of local government judicial and legal workers in Philadelphia and San Francisco was offset by a relatively low level of state government workers in that category.
Overall, the ratio of local government employment to population was similar to NYC in San Francisco and St. Louis, higher for Washington DC (which also includes state government functions), but lower in Baltimore, Boston and Philadelphia, where state governments shoulder more of the load and, in the case of Philadelphia, pensions suck up more of the money. Philadelphia’s public services may provide a good indication of where New York City’s public services are going, once the cost of all the recent pension enhancements can no longer be deferred. In education, that would mean the number of teachers would fall by 36.6%.
As in the different parts of New York State, in these cities the overall level of local government payroll per employee for most of these cities differed from the U.S. average to approximately the same extent that the average private sector pay for the related metropolitan area differed from that average. In metropolitan Boston, for example, the average private sector worker (excluding those in the Finance and Insurance sector) earned 32.2% more than the national average; the average local government worker in Suffolk County earned 28.3% more than average.
Overall San Francisco local government workers were paid more relative to their private-sector counterparts, while St. Louis local government workers were paid less. As a result of a 100 year-old law, however, any changes to San Francisco’s public employee pensions require a public referendum, whereas in New York the state legislature could pass a law tomorrow granting every public employee immediate retirement with a 100% pension, including themselves, and it would be legally and constitutionally binding. Some public employee pension changes have passed over the years, but many others have been voted down. Most public employee pension systems in California are in deep trouble. Last time I read an article on the subject, San Francisco’s has much healthier.