I recently had the pleasure of writing a blog post for the London School of Economic’s US Centre Blog. It covers a recent publication in State and Local Government Review on the implications of overlapping local governments on local public employment in the US. You can find more information about this piece here. See below for the full text of the post as well as a link to the original.
How a type of local government you may have never heard of may be costing you a bit more
Special districts are one of the most common forms of local government in the US. Aimed at tackling a specific issue in a certain area, these districts can often overlap with city and county governments. In new research Christopher Goodman looks at whether this overlap can mean an over-provision of public employees. He finds that while this overlap can often lead to a small increase in the number of public employees, the effect is much greater on local revenues – i.e. taxes – which can be up to ten percent greater.
Special districts are a unique and very popular form of local government in the United States. They can overlap any other forms of local government such as cities or counties, take on nearly any geographic shape, and provide public services similar to cities. This arrangement creates the potential for the over-provision of public services due to a lack of coordination between numerous overlapping governments. This has been found to be the case with local government revenues as well as public debt, but the question remains if this over-provision extends to local government’s most common cost, employees. In general, the answer is yes, but the size of the “overlap effect” is quite small.
Special districts are the single most numerous form of local government in the US as of 2012. Their unique attributes make them attractive for solving all manner of local issues. Need to run a metro-wide transit agency? A special district fits the bill. Need to provide a particular public service to a particular neighborhood? A special district works here too. This flexibility allows areas to tailor the collection of public services they receive to fit their needs. There are of course those who think this kind of fragmentation of public service provision is bad and can lead to all kinds of ill effects.
Back the original claim. Why would the lack of coordination among many special districts lead to an over-provision of public services? Ideally, it wouldn’t; however, we do not live in an ideal world. The single service nature of many special districts is the source of the problem. Most special districts have one job to do: provide the service they were created to provide. This naturally leads elected officials and public managers to want to do that job more and better. Potentially so much more than it becomes more than what would ordinarily be provided. If the various services provided by a collection of special districts were instead provided by a single city government, all those services (likely organized into departments) would have to compete with each other in the annual budget process for funds. This competition would keep things in line. Because each service is provided by (and funded by revenues from) a separate and autonomous special district, there is no budget competition and the potential for over-provision is born.
Note what I just described requires no one, elected or appointed officials, public managers, or voters, to do anything wrong. All that is required is for those groups to do what they normally do. Elected officials favor policies that will get them reelected. And it’s likely those same elected officials are at least somewhat interested in the public service the district provides. Both are completely rational, but also likely result in more public goods or services being provided than would be in a city government.
Most local government services are highly labor intensive; therefore, if there is an over-provision problem, it should show up in more local public employees than would ordinarily be employed. This is accomplished by comparing the level of local public employment across all types of local governments in areas with high number of special districts per city such as St. Louis City, Missouri to areas with low special districts per city like Baltimore City, Maryland.
Of course, these areas may employ more or less public workers for reasons other than special district overlap such as the prevailing wage (local governments tend to hire fewer workers when wages are high), revenues from state or federal government (local governments often hire more workers when someone else is partially paying for them), or the wealth of the community.
In particular, I examine this relationship in all county areas in the United States from 1972 to 2012. On average, a county has about 40 full time equivalent local public employees per 1,000 residents and roughly 2 special districts per city. Overall, the results of the statistical modeling suggest there is an upward bias in the number of local public employees in areas with more special district overlap, the size of the effect is quite small. Moving from the 25th percentile of overlap (0.5 special districts per city) to the 75th percentile (2.3 special districts per city) increases in level of local public employment by about one percent or less than one full time equivalent employee. When compared to the overlap effect of local revenues (approximately a 10 percent increase), these findings appear quite small indeed.
So what?
It’s generally thought that competition between many local governments is a means to constrain inefficient behaviors of those governments. However, this theory was built on competition between horizontally arranged local governments (e.g. cities); governments that do not overlap one another. While informative, most of the growth in US local governments over the last 50 years has been in the vertical direction by numerous overlapping special districts. This is a fundamentally different arrangement where it appears the competitive forces that constrain cities do not apply.
While the “overlap effect” does not appear to be particularly large for local public employment, it is for other local outcomes (revenues, specifically). This disconnect begs the question if these two results are truly the state of US local governance, where are the excess tax dollars being spent? The results of my research suggest not on employees (at least not directly) so there’s more of the puzzle to be completed.
This article is based on the paper, “Jurisdictional Overlap and the Size of the Local Public Workforce” in State and Local Government Review.