Minimum lot size rules are a mainstay of contemporary U.S. zoning. Despite their ubiquity, we know very little about how they work in practice. Do these rules simply reflect what the market would create without regulation? Or do they actively force lot sizes to be larger? In a new paper for the Mercatus Center at George Mason University, my co-author Salim Firth and I decided to investigate whether minimum lot sizes are, in fact, binding. That is to say, do these rules actually affect the decisions made by developers?
Let’s start with some basic background: minimum lot size rules determine how small a developer can subdivide a lot. Lot subdivision is a wonky issue, but it matters a lot for the form that neighborhoods take. Small, 1,400 square foot lots allow for rowhouses and and shotgun houses. These are building types that keep housing cheap and burden on infrastructure low by economizing on land; they also tend to produce more tax revenue, acre-for-acre, to fund local government services. Their higher densities also make walkable, mixed-use urban neighborhoods viable. Large, 5,000 square foot lots, on the other hand, are the standard in US post-war suburbs. These larger lots mean that fewer homes can fit in a given area, which encourages auto-orientation and the segregation of land uses like residential and commercial.
Unlike most zoning rules, minimum lot size rules have been around for centuries. Indeed, unlike many design rules, they serve a clear purpose in many cases: In urban areas, tiny or highly irregular lots can make property rights fuzzy. In rural areas, you need a large lot to safely accommodate things like septic tanks and wells for drinking water. But like many zoning rules, these historically reasonable rules have often transformed to serve dubious ends: in most US cities, large minimum lot sizes are used to drive up housing prices or mandate spread-out development.
But do minimum lot size rules in fact increase lot sizes? On the face of it, this might seem like a weird question: why, after all, would cities have these rules on the books if they weren’t increasing lot sizes? But it could be the case that planners (or more accurately, planning consultants) are simply rulemaking for the sake of rulemaking, setting minimum lot sizes at roughly what the market already wants to provide. If that’s the case, all the research about how minimum lot sizes increase home prices or drive sprawl could be misdirected.
To answer this question, we looked at four Texas suburbs which have experienced explosive growth and development over the past 30 years: Round Rock, Pflugerville, Frisco and Pearland. We picked these cities because each had clear minimum lot size rules, high quality parcel data, recent population growth, and undeveloped land that remains available for development.
With parcel data in hand, we then compared the actual size of every single parcel to its zoned legal minimum size. What we were looking for was an unusual concentration of lots whose size was exactly, or very close to, the legal minimum.
The logic behind this can be illustrated by analogy. Imagine a grocery store that sells apples, and you can go in and pick them out a la carte and buy whatever number you like. If you were to survey customers who bought apples on how many apples they bought and graph the results, you might expect to see a pretty diverse distribution of results. On the other hand, suppose for some reason that this store imposed a minimum purchase of three apples. You can buy as many as you like beyond that, but the store will not sell you any apples at all unless you buy at least three. We would now expect to see an unusual concentration of customers buying exactly three apples, because that number would include those who might otherwise have opted for one or two apples.
In the same way, if there is a concentration of homes built on lots that are almost precisely the legal minimum area, that is a sign that developers might otherwise have chosen to develop smaller lots, but were legally barred from doing so.
If a lot’s area was below or just barely above (less than 110 percent) the legal minimum, we inferred that the minimum lot size regulation had a binding effect on the lot’s size. These cases indicated either that the developer had to pursue regulatory relief (e.g., a variance) or just barely met the minimum requirements. The developer presumably might have opted for a smaller lot absent the rules. If a lot was slightly above (110 to 120 percent) the minimum, we interpreted the rule as potentially binding. If a lot was well above the minimum (over 120 percent), we interpreted this as evidence that the rule wasn’t binding; i.e. it was not likely the deciding factor in lot size.
So what did we find? Round Rock and Pflugerville, two Austin suburbs, each had one minimum lot size for single-family homes—6,500 and 9,000 square feet, respectively—and the results were straightforward. In Pflugerville, fewer than one-in-five lots were substantially larger than the zoned minimum. In Round Rock, this number sat at right around half. That is to say, a large share of subdivisions in both towns are either below or just above the zoned minimum lot size, indicating that the rules are binding and thus forcing up lot sizes.
Frisco, a suburb of Dallas-Fort Worth, was more complicated, with various different single-family residential zones setting distinct minimum lot sizes ranging from 7,000 to 16,000 square feet. Frisco was much harder on non-conformance: few lots were permitted to dip much below the zoned minimum. But like Round Rock and Pflugerville, surprisingly few lots were far above the zoned minimum. The lion’s share of lots sat right at the zoned minimum, again indicating that regulations were forcing developers to plat larger, more expensive lots.
Pearland, a suburb of Houston, had multiple different single-family housing zones like Frisco, with minimum lot sizes as low as 5,000 square feet. Unlike Frisco, or any of the other towns we surveyed, there was little evidence of a binding effect in Pearland. In nearly every zone, a majority of the lots were clearly unbound by the regulation, with lot sizes extending well beyond the zoned minimum. Only in higher zones like R-2 (7,000 square foot minimum) and R-1 (8,800 square foot minimum) was there any evidence of binding, but even here, it was low relative to the other cities surveyed.
Why did this happen? One possibility is that the town offers a zoning district with a relatively low minimum lot size of 5,000 square feet. This is the lowest minimum in any of the cities surveyed and may roughly track onto the market-clearing lot size in an outer suburb like Pearland. Another possibility is that the demand for smaller lot sizes is absorbed by nearby Houston, which permits lot sizes as small as 1,400 square feet.
All in all, we found strong evidence that minimum lot size rules do, in fact, drive up lot sizes. This has important implications for planning practice. For starters, planners interested in improving housing affordability should apply a critical eye to any minimum lot size rules much higher than 5,000 square feet, at a minimum. With land as a major driver of housing costs, zoning regulations that require excess land could be playing a large role in increasing housing costs and exacerbating shortages.
Furthermore, planners concerned about excessive outward growth—and the high cost of all the additional infrastructure—should critically evaluate all minimum lot size rules. Overly strict rules could be inhibiting desirable infill development where infrastructure is already built, pushing new housing development out to where land is cheap but infrastructure is missing. As any regular Strong Towns reader should already know, building and maintaining this new infrastructure won’t come cheap.