So it has come to this. Bad financial reporting by local governments elsewhere has so delayed the release of final local data from the finance phase of the 2007 Census of Governments that local government employment data for March 2008 has been published first. I’ve compiled it for New York City, the rest of New York State (by subtraction), New Jersey and the United States, along with some related and relevant private sector data, in the attached spreadsheet for those interested. Far more geographic detail was available from the employment phase of the 2007 Census of Governments, and little changes from year to year in New York’s relative priorities as evidenced by the data. Data and a discussion of the employment of states for 2007 is in this post. A very detailed spreadsheet for local government employment in 2007 is here, with a comparison with 2002 here, and other comparisons here, here and here. The rest of this post is a discussion of how the 2008 data was compiled (mostly copied from a post on a similar exercise for the March 2006 data). I’ll write an analysis when I can, but reading over the 2007 posts will almost certainly do as well.
Some background on the data. Every five years, most recently in 2007, the U.S. Census Bureau conducts a Census of Governments, which allows data to be compiled on the employment, payroll, revenues, expenditures and debts of all the local governments, added together, in each county in the country. Because this data includes everything – municipalities, counties, school districts, special districts – it can be directly compared with New York City, which is all of the above.
In non-census years, the Bureau conducts a survey of local governments, and provides estimates for all local governments totaled together for each state and the United States. That is what this data is. Individual unit data for the City of New York and Port Authority is always included as well, and this allows me to create data for New York City, and thus for the rest of the New York State by subtraction. (The Office of the State Comptroller could match the Governments Division and provide comparable data for every county in the state every year if it chose.) To do so, I have to divide up the Port Authority employment and payroll between New York City (and state) and New Jersey, adjusting state totals accordingly (the Census Bureau assigns the whole agency, including operations in New Jersey, to NYC local government).
To make the data comparable, additional adjustments are required. I divide full time equivalent employment by population to get the number of local government employees in each category per 100,000 residents. The division between state and local government also has to be accounted for, to the extent it can be. Most transit agencies are classified as “local government,” but those in New Jersey and the portion of New York State outside New York City are classified as state government. I add that employment back to the local government totals. New Jersey and some other states have taken over schools, also classifying their employment and payroll as state government. To make the per-capita figures comparable with New York City, I re-classify that data as local government also.
The government sometimes acts through the private sector (by paying for private health care via Medicaid for example), and some private industries substitute for public employment. For example, New York City has 589 Transit employees per 100,000 residents, far above the national average of 78, but that is because New York City has a lot of transit. On the other hand, the city had only 428 employees per 100,000 residents in industries related to the private automobile, compared with a national average of 1,308. What New Yorkers must pay, in taxes and fares, to support all those transit workers is offset through the ownership and use of fewer private cars. As we shall see, however, that isn’t true in the hospitals industry. I show, on the same page, employment and wages data for 2008 for “government-funded” and “government substitute” private sector industries next to the March 2008 Census Bureau data on local government, although the two are not directly equivalent
Overall, New York City had 5,187 full time equivalent local government employees per 100,000 residents in March 2008, compared with a national average of 4,080, with 5,061 in the rest of New York State, and with 4,417 in New Jersey. When considering the average for the rest of New York State, recall that parts of the state do not have public water, public sewer, public trash collection, and professional fire departments. The public housing, public hospitals, and public transit are far less extensive there than in New York City as well. So the “rest of state” figure is remarkably high due mostly, the data shows, to public schools. In New York City, in addition to transit, employment per 100,000 residents is unusually high for police, public hospitals, housing and community development, and social services. School employment is low.
I’ve have also calculated the percent more or less than the national average each category of public employee was paid in March 2006. The data is total payroll divided by the number of full time equivalent employees in March. These figures are not really comparable unless one considers the higher cost of living (associated with higher average private sector wages) in Downstate New York.
For 2008, the average private sector worker in Downstate New York earned 30.9% more than the national average if the high-paid Finance and Insurance sector is excluded. That figure has generally been about one-third more in the past. In New Jersey, the average private sector worker earned 20.2% more than the national average excluding finance, while in the portion of New York State outside New York City (including the suburbs this time as well) they earned 1.3% less than average. Because the annual data does not allow a separation of low-cost Upstate from the high-cost Downstate Suburbs, payroll data for the rest of the state is not that useful in non-census years. Review the 2007 data above instead.