We spend a lot of time here talking about the local regulations that harm cities, from parking minimums, to height limits to restrictions on mixed-use development. I’ve been thinking recently about another policy that impacts cities at the federal level: the tax code. I bring up this topic not to stoke the political debate on redistribution, but to think about the outcome that results if tax burdens push and pull people out of cities.
I’ve often heard libertarian and conservative types say that they prefer suburban or rural living because city residents are more dependent on government services, but is it really true that city dwellers depend on government programs more so than others? City residents clearly pay higher local taxes and receive more local government services, but I would argue that this isn’t as important as federal taxation and spending because local taxes benefit the taxpayers more directly than the wider net of federal redistribution.
In Triumph of the Cities, Ed Glaeser points to the mortgage interest tax deduction as an important factor that pulls people to the suburbs by way of home ownership, and I would certainly agree that all federal policies with the goal of promoting home ownership or facilitating easy credit harm cities. Another much-talked-about tax expenditure for ethanol fuel blenders has benefits that fall on drivers, and this is on top of all of the other tax benefits that domestic oil companies receive.
Several government programs subsidize those who choose to live in inconvenient places. The USPS delivers many packages for a flat rate fee, while I doubt that FedEx would find this pricing profitable. Similarly, the American Recovery and Reinvestment Act provided funds for expanding broadband to areas of the country that could not otherwise support the service.
Does anyone know of existing research on this topic? The Tax Foundation provides at least two data sets that are somewhat relevant, but neither answer the question. They show here which states receive more money than they spend and here the per capita federal tax revenues by municipality.
You’re right that federal and state subsidies for public housing do pour a certain amount money into cities, but it’s really a double-edged sword fiscally-speaking. Residents of the affordable housing typically require public services well beyond any revenue generated, leaving the rest of the residents in the jurisdiction to fill in the gap – thus giving an economic disincentive for the middle-class to live there. I’m a supporter of affordable housing programs for those who need them, but it’s generally accepted that the responsibility for providing this safety net should be distributed around the region rather than placed fully on the city. This is something HUD is working on, but previous investments make this is a slow-moving process. Paradoxically, you may want to put these funds on the other side of the ledger, as elements of the tax code that relatively disadvantage urban areas.
There is also the additional cost of providing certain services to more spread out areas like the suburbs. The economies of scale provided by density can reduce per capita costs, sometime dramatically. This is particularly true of mail delivery and road maintenance, as well as—in many cases—police and fire services.
It would seem to me that a tax reform that would remove all taxes on buildings and, instead, place a property tax on the site location value under buildings would be a great boon to urban life.
I’m interested in learning more about moving from property taxes to land value tax. Do you recommend any specific resources on this topic?
That’s a good point — much of the federal spending that goes to cities comes with unfunded state and local mandates or at least carries incentives for increased spending at lower levels of government. Personally, I’m also sympathetic to arguments from scholars like James Bartholomew who question whether programs like public housing even help those who are receiving these benefits, but certainly it is questionable whether these programs benefit cities on the whole.
You can go to the Council of Georgist Organization’s site and find many links to organizations that promote Land Value Taxation. see http://www.cgocouncil.org/
A few years ago, you could crib data for per-capita spending in each county by going to the Census Factfinder. But they’ve since updated the data to more recent years (see e.g. here) whereas the Tax Foundation still uses 2004 data, so it’s harder to make comparisons.
Back when the Factfinder had 2005 data, I compared the numbers per county and found that the New York CSA was a net tax donor to the tune of $93 billion a year, the SF and LA CSAs were net donors to the tune of $40 billion a year each, and the Washington-Baltimore CSA was a net recipient to the tune of $67 or $76 billion a year, I forget which. I had data for every county in NY, CT, and NJ, but lost it when I lost my laptop. Very, very few metro areas that aren’t major capitals (Washington, Albany, Sacramento) are net recipients – Buffalo is the only one I saw, and even it is a tax recipient by a much smaller amount than capitals or rural areas.
Cities contribute mightily to the economy of the surrounding suburban, exurban and even rural communities but are expected to take on the costs that come with that economic dynamism all by themselves. So suburbanites get the benefits of cities without paying the higher sales and property taxes that help make them possible. Those of us who live within city limits are paying a premium, in part to subsidize the lifestyles of wealthy suburbanites.
In short, I think you’re right in thinking the tax code doesn’t do cities any favors. Looks like Alon Levy found some stats to back that up.
Thanks for pointing out the Census expenditure data and sharing the results that you found looking into this issue. It seems like the taxation side would be easier to track than spending, so it’s surprising to me that it’s not available through the Census Factfinder as well. If the Tax Foundation updates their study, it would make a great extension to look at how cities fair on net.
You can use tax revenues, I suppose, but the Tax Foundation does more than that, and makes an effort to allocate e.g. cigarette and alcohol taxes, which should count as falling partly on consumers and partly on the tobacco and alcohol industries.
By the way, to clarify, although major metro areas are huge tax donors, the same is not true of major cities. New York City was a $10 billion/year tax donor. This is roughly comparable to Suffolk, Nassau, Westchester, and Fairfield County, each. But the Tax Foundation does not allocate spending, so ESA would count as city spending even though the beneficiaries are people in Long Island; this skews the numbers in favor of suburbs, since suburbanites use city infrastructure more than the reverse (and metro areas use rural infrastructure, e.g. Interstate truck routes, more than the reverse).