Thursday, May 24, 2012
Excessive Big City Zoning Is a Party Foul
In this post, I argued that the demise of pork spending by Congress is likely the result of increasingly coherent political parties at the national level. In my next few posts, drawing on this draft article, I will argue that excessive zoning in big cities is frequently the result of the lack of coherent parties.
A bit of set-up: Scholars of all stripes in land use have generally assumed that while rich suburbs may use zoning exclude undesirable building, big cities run by "growth machines" coalitions will allow rampant building because developers are the biggest players in local politics. This once was true -- big cities did allow lots of building to match demand -- but it no longer is in many cities. Price increases in Manhattan were once followed by lots of new housing starts, but no longer are. D.C. actually saw a decrease in new housing building permits as prices increased. Because supply and demand apply equally to housing markets as they do everywhere else, restricting supply in the face of increasing demand causes huge price increases. The average cost of a Manhattan apartment is over $1.4M (and the average rent is above $3400), and in D.C., even a small 2 bedroom in the poorest area of town costs $1300 a month. In cities with fewer restrictions like Houston and Atlanta, we do see supply increases when there are demand shocks, prices are much, much lower and population flows to these areas despite lower wages. But many big urban areas, in Ryan Avent's choice term, have become "Gated Cities."
But it's not like big developers in these cities have become politically powerless. So what gives? More specifically, what is the mechanism through which big city politics sometimes can produce a restrictive atmosphere for building despite the influence of the Bruce Ratners and Donald Trumps of the world? The key, I argue, is the absence of political party competition at the local level.
One notable thing about urban politics is that it lacks competitive party politics. Often this is by design, with formally non-partisan elections; other times it is due to dominance by one party. Trying to explain why one party dominates a level of government for half a century with virtually no competition -- like say, the New York City Council -- turns out to be quite difficult, and is something I have tried to do here and here and here. But to understand how we have ended up with excessive zoning in big cities like New York and D.C., the only key is that we don't party competition in local issues.
As noted in the last post, the need to for party leaders to promote a healthy-sized caucus (in order to keep their jobs) causes them to propose policies that will maximize electoral benefits to their caucus (which explains why caucus members are willing to delegate power to them even though it means they will not always be able to propose amendments that they would prefer). They do so by structuring votes in ways that will promote the party brand generally across the electorate. In the absence of any relevant party-based competition, there is no one in the legislature that has an incentive to promote generally-beneficial legislation as opposed to district-specific goods (or to strike deals between members to forgo district-specific goods in favor of a greater good and make them stick). Further, in the absence of a party structuring the voter order, the formal rules governing procedure are likely to determine the results between Arrovian cycling preferences.
All that positive political theory sets up two likely results. First, in a world without parties, we are more likely to see distributive politics -- pork, or in land use, the ability of neighborhoods to determine land uses with little attention to broader citywide needs for housing supply. Second, in the absence of parties, procedural rules can have substantive effects. In land use, the structure of the land use amendment process leads to a situation where the interests of big developers and small developers are divided. This limits incremental increases in the housing stock. Both increase prices.
Land use procedure -- through the traditional Standard State Zoning Enabling Act and modern innovations like New York City's ULURP -- insists on peculiar procedure. (Don't worry -- this isn't going too deep.) First, cities develop "plans" laying out land uses generally and then "maps" defining land uses and heights for specific parcels. Changes from these maps are done seriatim through "amendments" or "variances." Cities sometimes do create sometimes create compeltely new maps or plans, but this is relatively rare -- NYC's last did so in 1961. The rest of the time, changes are made in specific geographic locations one at a time.
The seriatim nature of land use procedure has two central effects. The first is that it entrenches distributive politics. We see excessive pork spending when legislator preferences take a specific "prisoner's dilemma" form -- preferring say lower rather than higher taxes, but preferring spending in your own district even more -- and where a "universal log-roll" norm develops to manage those preferences. Land use fits the distributive politics model perfectly. Definitionally, NIMBYism suggests that people wouldn't mind more development as long as it was in someone else's backyard. Deciding changes geographically-specific amendment by amendment serves to entrench a norm inside legislatures that individual Council members are the key and perhaps only player in deciding land use questions in their districts. This has been described as the "Ironclad Principle of Aldermanic Privilege." If you want to get something built, you need to get the support of the local councilmember. For council members who don't want new building in their district unless it's happening elsewhere as well, the developer faces a difficult task of striking deals across projects and time, something far harder than merely influencing a majority coalition in a legislature dominated by parties. The need to have local support for a new project basically turns big cities into a bunch of exclusive suburbs, focused on local externalities and ignoring the citywide benefits of increased housing supply. This drives up the cost of housing.
The second big effect of land use procedure on land use policy is to divide the interests of big developers and small developers. When a big new project is proposed, we know what the politics looks like -- neighborhood groups fight against developers. But when cities propose "downzonings" or reductions in the size of the "zoning envelope" to current uses -- i.e. removing the ability to develop as of right -- the politics are tilted in favor of neighborhood groups. This is a function of land use rules that force amendments to be considered one-by-one (rather than collectively, the way taxes and spending are in a budget). Neighborhood groups get concentrated benefits from stopping new building; housing consumers each suffer very little harm from each new downzoning and connected developers don't care because they haven't invested yet. Even if the overall harm outweighs the benefits, downzonings go through easily due to these tilted Olsonian politics. And downzonings matter. While big developers can fight and buy their way out of the restrictions of the zoning envelope, going through the many-month-long and lawyer-heavy zoning amendment process is just too costly for small builders. The fixed cost of achieving a zoning change is just too much for incremental new granny flats or small apartment buildings. Notably, this may have a "Curley Effect," shaping the electorate in ways that make future policy more restrictive. When a building on a property hits the zoning envelope, its owners know they won't be able to build in the future (because they won't be able to pay the fixed cost of getting through the amendment process). As a result, they go from being voters with mixed motives -- both neighbors to potential development and potential builders themselves -- to merely being NIMBYist neighbors. Further, the cost of achieving zoning amendments may help big builders by inhibiting incremental development (it's competition, after all). But if the issue were presented in a different way, you might see coalitions of big developers and small developers instead of our current system. Land use procedure restricts incremental building and thereby drives up the cost of housing.
Can anything be done when cities start to shrink wrap themselves? If my analysis is right, and procedure is sometimes part of the problem, procedural changes may be part of the solution. By posing issues differently, our democratic process may produce different results. But that will have to wait until my next post, (but if you're interested, you can see some reform ideas in the paper.)
Posted by David Schleicher on May 24, 2012 at 03:10 PM | Permalink
TrackBack URL for this entry:
Listed below are links to weblogs that reference Excessive Big City Zoning Is a Party Foul:
The comments to this entry are closed.