Let’s revisit our discussion of zoning, but reimagine zoning codes as cartel enforcement contracts and see how well the analogy fits.
- Zoning codes are obsessed with setting limits on housing supply. Some of the rules that limit housing supply include maximums (but almost never minimums) on height, interior square footage, lot coverage, dwelling units, numbers of unrelated persons allowed to dwell in a single unit.
- Cities have boards, usually filled almost entirely by homeowners, who adjudicate disputes about whether homeowners have special circumstances can allow them to waive rules when special circumstances mean that rules don’t allow them to build their “fair share.” Particular groups of neighbors join together into neighborhood associations to advocate for strict enforcement of zoning rules in their neighborhood.
- Where denser housing is allowed, it is usually the housing least likely to be a substitute for existing homeowners’ supply: sites on greenfield at the periphery of the city, far from existing housing, near disamenities like high-traffic corridors, or in areas dominated by university students.
- Many of the same neighborhood associations which work to enforce zoning cartel rules also push for inclusionary zoning — forcing new multi-family housing supply must cross-subsidize dedicated low-income housing. This cross-subsidy drives up the cost of the market housing which competes with cartelized single-family housing, while providing low-income housing only for those market segments which can demonstrate they will not be in the market for cartelized single-family housing in the first place.
This doesn’t mean that zoning has no other function than enforcing a cartel! But it is clearly one of the most powerful cartel enforcement mechanisms that exists.
Welfare Losses From Cartels
Cartels are generally criminal for a reason. Monopolies reduce total welfare, at least as reckoned by economists. The reason is not that monopolists are able to charge their customers a higher price. Each extra dollar paid by one of its customers (a loss) is offset by the extra dollar received by the monopolist (a gain); the gains and losses balance out, and so the monopoly’s price effect is merely to redistribute money from consumers to monopolists, not reduce total welfare. The central problem with monopolies instead is that monopolists invariably choose to sell fewer widgets than the sellers in a competitive market would sell. The real losers from a monopoly are the consumers shut out of the market altogether. They do not get to enjoy the widget they would have liked to purchase. These costs are readily apparent from our comparison of Affordaville and Restrictia in Chapter 2: among those housed, prices are higher. Some of those higher prices represent wealth transfers to incumbent landowners, while some of them represent deadweight loss from multi-family housing being forced into higher-cost techniques prematurely.
In what is perhaps the worst loss, many people who wish they could live in a neighborhood find that they no longer can. Without cartel restrictions, more people would have the opportunity to live in the places that they want. For some, this means living near school or work and spending less time commuting. For some, this means living near friends and being better connected socially. For some, it means living in a neighborhood of opportunity — the theater district, say, where you bump into the people you need to meet to build connections and a career.
Special Welfare Losses From Urban Cartels
As noted above, land-use cartels produce additional welfare losses that are not accounted for by a simple supply-and-demand diagram. Tight zoning of an exclusive neighborhood does not merely affect the residents of the neighborhood or those who would live there. It affects those in adjoining neighborhoods, who must contend with those priced out of the exclusive neighborhood, and those in more distant neighborhoods, who must contend with longer commutes in worse traffic. Practiced broadly enough within a metropolitan area — as is the case of our most expensive coastal markets — zoning cartels often produces spillover welfare losses by inhibiting the free flow of labor between markets. And by artificially holding population to a fixed level, zoning cartels can also interfere with the increasing returns from urban agglomerations. Consequently, zoning cartels should be viewed with greater suspicion than ordinary cartels. The land value trigger we propose would relax quantity controls in high-priced areas, while leaving other zoning regulations intact as long as they do not function as surreptitious quantity controls.
I’ve been working on a chart to illustrate why a developer who enjoys pricing power would sell above land cost, and how this makes housing (specifically, the land under the house) cost more. The next step is to show that neighbors will want to maintain the price point, and how an upzoning will cause them to defect, and lower the profit (land value) for all of them.
In the chart below, I assume construction cost of $125,000/unit and land cost = value of land for alternate commercial/agricultural use = $25,000/acre. Developer faces marginal cost of $125,000/unit after the first unit. (He’s got to eat the whole cost of land no matter how many houses he sells, so land doesn’t factor into marginal cost after the first unit).
I assume a 12-acre development.
The demand schedule is given by the price column. (I.e., the developer is not a price taker but has pricing power.) Demand has the downward slope in part because there are no good substitutes for the neighborhood and in part people value having a bigger yard. (Yards shrink as total number of units increases.)
The profit maximizing price is $237,500 for a 2-acre lot. (This is where Marginal Revenue = Marginal Cost.) The effective land cost is $62,500/unit (or $31,225/acre) over “cost”.
We assume this quantity is stable because the developer negotiates zoning specifying a minimum of 2 acres per lot.
Now when the neighborhood is upzoned to 1 acre/lot minimums, neighbors “defect” and develop one of their two acres. At least the first 5 do. The 6th one won’t build anything because it would be at a loss. (MC > Price when n = 12.) The value of land is driven back down to approximately $25,000 per acre. This increases consumer surplus. Specifically, it adds the consumer surplus from buyers 7-11, who weren’t able to buy into the neighborhood under the old zoning.
Needless to say, neighbors organize to oppose the upzoning. Functionally, they are asking the city council to act as their cartel “fixer.”. They complain about traffic, noise, water quality, fireflies and land values (the latter being a genuine complaint)).
A few points:
- This chart illustrates one of the central fallacies when thinking about land values and zonings: upzoning one lot might make that lot more valuable and earn the owner more profit. (E.g., the 7th house is still quite valuable and quite profitable to build; so an owner that got the right to build an extra housing unit would be making a bundle.) But a broad upzoning of the neighborhood could still lower the total land value. (Put more technically, demand for one lot may be elastic while demand for all the lots is inelastic.)
- Relatedly, neighbors will object to a modest zoning of even one additional house, because it reduces total land value. But this also shows why, even if there were only 5 houses to begin with, and even though it would maximize land value to add a 6th, it’s impossible to do so while making everyone happy. You can’t give everyone the right to build ⅙th of a house. They get one or none at all. So even though the neighborhood as a whole would be worth more with 6 houses, the neighbors will fight to keep it at 5 rather than upzone to 10 (which is the quantity allowed by a minimal, uniform upzoning).
- Price expectations. If you increase zoning to allow 12 rather than 6 units, what does the first new house built — the 7th unit — sell for? Assuming perfect information, not $218,000! A buyer will recognize that the prices will fall as four other units are built. The buyer will pay the new equilibrium price of $152,27 plus the value of having the unit for a month or six months earlier (i.e., the rental value for that period). Put differently, rezoning itself will lower the equilibrium price, even before the new supply comes on line.. Doesn’t it follow then that an anticipation of automatic upzoning will lower the equilibrium price? (Or is that stretching it?) This is why the LVT is in some respects similar to a zoning expiration date.
- If you assume intensity of neighborhood opposition is proportional to the amount of land value (profit) at stake, a neighborhood that starts out with 8 houses will not fight nearly as hard as a neighborhood that starts out at 6 houses. Moral: always max out zoning initially!
- The neighbors can still make money selling all 12 houses if they can price discriminate. I.e., shake down each new homebuyer (or their proxy, the developer) for their reservation price.
6. Affordable housing is price discrimination and could preserve the prevailing price for the first 6 houses. Sell the affordable units to people who wouldn’t pay market price anyway, leaving the base of market-rate customers intact.
The Neighborhood is the market
We argue that zoning cartels operate at the neighborhood level. Some variation in the package of amenities offered by neighborhoods and small towns may be socially desirable. And, in most neighborhoods, the package of amenities has little impact on home prices. The neighborhoods may not be very nice, so that the value of the neighborhood’s amenities is zero or negative. The neighborhoods may have many other close substitutes, so that no one is willing to pay a premium for their amenities. Or the neighborhoods may enjoy particularly attractive amenities, but the zoning allows practically unlimited development, preventing homeowners from capturing the value of those amenities (i.e., the cartel allows a supply in excess of the market-clearing level). Consequently, the cartelization of supply through zoning within a given neighborhood may not impose any welfare loss at all.