State and Local Government Finance: Census Bureau Data for FY2014 Compared with FY 2004

The Governments Division of the U.S Census Bureau released its detailed state and local finance data for FY 2014 on January 31, and I have compiled it and produced a couple of large tables – one for all state governments and one for all local governments by state – comparing that year with FY 2004. The data shows, by category, the amount of revenues (property taxes, federal aid), expenditures (public school spending, police department spending) and debt for every state in the country and, at the local government level, for New York City and the Rest of New York State separately.

To be comparable across states and across the years, the data is presented per $1,000 of the personal income of all the residents of each state. Think of it this way. Your household else spends X percent of its household budget on food, X percent on housing, etc. And, via the taxes and government fees you pay, X percent of its income on public schools, X percent on police, etc. The data is presented per $1,000 rather than as a percent to make the data for small categories easier to see. I plan to write a series of posts, with additional tables and charts, for different aspects of state and local government finance separately. But in the triumph of hope over experience, in this post I will explain where the data comes from and how it was compiled, provide the whole database for download up front, and invite people to look at the numbers themselves at the same time I do, and decide for themselves what the data means. Before getting my take on it.

The state and local finance data is compiled by the Governments Division of the U.S. Census Bureau, and has been since the late 1950s, except for years when budget cuts eliminated it. You can download the original data, which you can see below in the tables in my spreadsheets, here.

I used the state by type of government – public use format file. Compared with the data the Bureau puts out in its other tables by state, it contains additional detail on state and federal aid by category. I want to see how much of local government spending on, say, public schools or social services or parks and culture is funded by local taxes, and how much is funded by federal or state aid, or by charges for services. Taxes only account for about one-third of local government revenues nationwide.

The data for the personal income of each state’s residents, and the residents of New York City and the Rest of New York State, is taken from the Bureau of Economic Analysis Local Area Personal Income data series.

Every five years the Bureau conducts a Census of Governments and tries to collect data from every local government in the country, with the most recent for FY 2012 and the next one for FY 2017. Detailed information on local government is available for those years. I did an extensive tabulation of FY 2012 Census of Governments data, with local government data for every county in New York State and New Jersey and for selected counties elsewhere in the U.S. Those spreadsheets, and the links to the first post in the series (from which the rest may be read by following the link at the bottom of the one before), may be found here.

During the other years, the Bureau samples local governments and provides estimates of state and local government finances for the U.S. and each state only.   Since the City of New York and the Port Authority of New York and New Jersey are always included in the sample, however, I am able to produce local government data for New York City, and get the Rest of New York State by subtraction. That is the information in the tables provided today – all the states, and New York City and the Rest of New York State separately.

With less data at the local level, I’ve decided to analyze more states, and place more emphasis on the change over time. Generally I focus on the Northeast, and other places I think of as being somehow comparable with New York City.

The Census Bureau’s data requires some adjustments, over and above the presentation per $1,000 of personal income, to be comparable across places and time.

In Hawaii all public schools are operated by the state government, rather than by local governments, and this reduces the U.S. local government school spending totals. Other states have taken over some local school districts from time to time. Although now part of the same MTA, a state agency, for historical reasons New York City Transit is still classified as City of New York local government. Most other transit agencies are also classified as local government. The Long Island Railroad, Metro North, and all major Upstate Transit agencies, on the other hand, are classified as “state government,” as are New Jersey Transit and some other major transit agencies around the country. To make things comparable across states, all state public school and transit expenditures are added to the local government totals for each state.

The Census Bureau also assigns the entire Port Authority of New York and New Jersey to New York City local government, and thus to New York State local government, inflating New York expenditures and leaving New Jersey with virtually no expenditures on Airports and Seaports. So I divide up each line of revenue or expenditure for the Port Authority between New York City and local government in New Jersey with a simple percent. Looking at the original data, and in particular categories in italics, you can see exactly what I did. The term “adjusted” is used for the NYC and New Jersey local government data for that reason.

Pensions need to be handled separately, and will be the subject of a separate series of posts I will write later this year, data availability allowing. The Census Bureau compiles public employee pension data from the point of view of the pension fund. Thus taxpayer payments into the funds, rather than being considered government “expenditures,” are considered “revenues,” and they are generally not available for individual government functions (schools separately from police). While pension benefit payments are considered “expenditures.” It is as if the money the City of New York paid private insurance companies for its employees’ health insurance were counted as City of New York “revenues,” and the amount the insurance companies themselves paid to health providers on behalf of those city employees were counted as City of New York expenditures.

Since the City of New York has its own pension system, moreover, the Census Bureau doesn’t include its taxpayer pension contributions in its general government finances data at all, since it considers them “payments by the city to itself.” Meanwhile, local government employees in the Rest of New York State are included in the New York State pension system, and payments by local governments in the rest of the state to the New York State pension system are counted. Using Census Bureau data specifically on pensions, I adjust for this and provide what I can. For other states, I suggest paying attention to the state plus local government total alone for pensions, since many local government employees are covered by state pension systems.

Also excluded from the tabulation of expenditures on most individual functions for New York City, thus making their expenditures seem lower than they really are, are employee benefits other than pensions. For the City of New York, most of those end up in the “General Expenditures – Other” category, and account for NYC’s massive spending in that category relative to other places. According to the Bureau’s Government Finances manual, “General – Other,” among other things, includes “multi-functional general sector activities that cannot be separated into specific functions” including “lump-sum contributions for employee benefits (retirement, unemployment and workers’ compensation, health and life insurances, etc.) other than transfers to own insurance trusts.” That is, payments by New York City to New York City’s pension funds are not counted there either.

It should be noted that the Bloomberg Administration included, in its Budget Documents, tables that showed pension and employee benefit expenditures for the largest government functions individually and separately. The DeBlasio Administration has eliminated those tables.

So it is no longer possible to see how much pensions and employee benefits cost the police department and public schools and sanitation department separately, nor which agencies account for the most legal judgments and claims against the city.  Although the Bloomberg Administration’s allocations were estimates, this is a move away from transparency.

One more funky aspect. In FY 2014, unlike past years, local governments in New York State reported sending vastly more money to the State of New York (mostly for the local government share of Medicaid, but also to the MTA) as intergovernmental expenditures, than the State of New York reported collecting from local governments as intergovernmental revenues. And the State of New York also reported vastly lower “local government aid” for the most recent year than for past year. Perhaps someone at the state level got tired of hearing just how much of New York’s local tax burden is actually state taxes in disguise, because of this virtually unique “local to state aid,” or perhaps its just a screw up. I’m pretty sure of the local share of Medicaid remains, and that if it had been eliminated I’d have heard about it. So the local government data is probably correct for FY 2014.  I’ve informed the Census Bureau of this issue, but there will probably be no correction until revised data comes out in a year, so I moved ahead without it.

One generally needs to focus on “direct expenditures” in this data. A single dollar may be sent from the City of New York to the State of New York as part of the local government share of Medicaid, then sent from the State of New York to the City of New York as a Medicaid state aid payment for services by the New York City Health and Hospitals Corporation, and then actually spent on health services by the HHC. Later, in the individual post on health and social services, however, I’ll show how much New York and other states spend in the category including state to local aid, to make the data more comparable between states that carry out all these functions themselves and those that pass money on to (generally) counties to do it.

Fair comparisons over time face an additional issue. Some years are better for the economy than others, and when the economy is bad personal income goes down and taxes and spending a share of personal income go up. That represents circumstances, not policy. A fair comparison over time, therefore, requires data for somewhat similar years.

From a political perspective, I might have liked to compare FY 2014 with FY 2002, to get the changes over the entire administration of former New York City Mayor Mike Bloomberg. But FY 2002 was the bottom year of a recession, and thus not comparable with FY 2014, four years after the bottom. Post-9/11 overtime also inflated NYC police, fire and sanitation expenditures for FY 2002, and I’m tired of explaining that away. I also might have liked to compare FY 2014 with FY 2010, to see what the impact of New York Governor Andrew Cuomo’s first term was. But once again FY 2010 was a recession year, and not comparable with FY 2014. So I decided to compare FY 2014, four years after the bottom of the Great Recession, with FY 2004, two years after the bottom of the tech wreck recession.

The tables, in the State Output and Local Output tabs, are found in this spreadsheet.


The individual categories are on the rows, and the data for each state for FY 2004 and FY 2014 are across the columns. The spreadsheet is set up to print the tables including data for the U.S., New York State (or New York City and the Rest of New York State), New Jersey, Connecticut, and Massachusetts. But data for all the other states may be found to the right. To keep the U.S. average and the row titles in place, you can use the Window/Freeze Panes or View/Freeze Panes commands, depending on your version of Excel or other spreadsheet program. If frozen panes are causing a problem for you, then you can unfreeze them.

As I’ve said for decades, when you actually look and see how different places compare, and how things change over time, in state and local government finance and employment, much of what you hear in state and local government politics is nonsense.   Propaganda, generally put out by the already-privileged, demanding more.

Just because it’s the national average doesn’t make it right, because different places are, in fact, different. Of course New York City spends more on transit and less on roads than local governments in Idaho, for example. But some of what you see, in fact more and more what you see, flies in the face of what people have been led to believe.

And when, for each place, this year’s taxes and spending are only compared with last year’s taxes and spending for the same place, without any other frame of reference, that comparison implicitly assumes both the losers and the winners deserve to maintain that status – or even get further ahead or behind – in perpetuity. And frankly, I recent decades that’s the way things have often gone, based on whoever has the most power and feels the most entitled in each individual place.

In the face of this, it disturbs me that when I use search engines to see who is tabulating and writing about this dataset, after the pages of the Census Bureau itself the only thing that seems to come up is posts I myself has written. Any public or private organization that wanted to, as part of its work, do exactly what I do in my spare time, and more. But the totality of the data doesn’t suit anyone agenda. The real deceit, as always, is the unsaid.

And given the shift away from transparency that we are beginning to see across government, and the reduction of funding for agencies that collect factual information, I have begun to fear for the integrity of the database itself. The Republican Party, or most of it, has long been against such transparency, always wanting to so away with public statistical information and defund it. In this Republican dominated-era at the national level they didn’t want evidence that most Americans, and younger generations of Americans, have been left worse off.

But thanks to the soaring cost of public employee pensions, which has caused the Democrats to be driven out of statehouses across the nation, I get the feeling that the Democrats have no changed sides, particularly with regard to data on state and local finances. A change that seems to be happening right now.

For the winner/predator special interests backing/funding both parties, apparently, there are things the serfs shouldn’t know.  Or, perhaps, for Generation Greed, still in charge of our public and private institutions, there are things that those coming after should not know.  What?  Download the spreadsheets and take a look.

The next post will be a detailed discussion of state and local taxes, by state and category, with the others to follow as I have the time produce charts and write them.

1 thought on “State and Local Government Finance: Census Bureau Data for FY2014 Compared with FY 2004

  1. Frank

    Hi Larry I’m trying to understand the difference between row 113 in the local tab and row 35 in the state tab. In the local tab row 113 adds rows 449 and 516 then divides by income, but row 449=516+553+554. The local government contribution in the state tab is equivalent to row 449/income. Might you be double counting? If not, could you walk me through what I’m looking at?

    Also, the numbers for Oregon’s pension contributions in 2004 are rather high. I know the state and some local governments have issued bonds to make lump sum contributions to the pension fund. Do those POB lump sum payments show up as a line item somewhere?

    A couple other local stories of note:

    >”We’re beyond crisis,” Katy Durant, chair of the Oregon Investment Council, said in an interview after last week’s meeting. “We should have been addressing this 20 years ago and it’s just been building. It’s a little bit like a Ponzi scheme. Sooner or later it’s going to catch up with you.”

    >In other words, to pay down the debt over 20 years, the increase in pension payments over the next two years should be more than three times higher, or nearly $2.7 billion.

    Just about the size of the $3 billion dollar sales tax that’s not a sales tax that voters shot down in November.

    >PERS fix won’t come until schools ‘go off the cliff,’

    Apparently they didn’t have a good backup plan in case the tax didn’t pass.

    A little history on where the debt came from:

    >The Legislature in 1975 decided to ensure that PERS members, all of whom were contributing a percentage of their income to the regular account, would enjoy favorable investment returns by granting a guaranteed rate of return. The balance of the regular account would grow each year based on investment earnings, never falling below the assumed earning rate in any given year. In 1975, the assumed earnings rate was 5.5%; over the years the rate was increased and is now 8%.

    >The booming stock market of the 1980s resulted in PERS investment returns of greater than 8%. The PERS Board, populated at that time mostly by PERS members, credited member regular accounts with the actual returns on investment. The PERS board granted credits above the assumed rate in 19 separate years, sometimes crediting member regular accounts with as much as 20% in a single year. PERS employees retiring on money match could retire and earn more in annual benefits than their final salary, with payments guaranteed for life and an annual cost of living adjustment (COLA) to boot. The generous PERS benefits had become excessive

    Thanks for putting the spreadsheet together,

Comments are closed.