Houston: Minimum Lot Size Reform (1998 / 2013)

Houston is the largest city in the United States with no conventional zoning code. There is no map that divides the city into residential, commercial, and industrial districts the way nearly every other American city does. Voters rejected zoning at the ballot box three separate times, most recently in 1993. Because of this, Houston is often used as a natural experiment for what happens to housing when one of the most common land-use controls is simply absent.

But the absence of zoning did not mean the absence of rules. Houston still controlled development through its subdivision ordinance, and the single most important rule in that ordinance was the minimum lot size: the smallest piece of land on which a builder was allowed to put one home. For decades that minimum was 5,000 square feet across most of the city. A lot could not be carved into smaller pieces than that, which meant that even without a zoning map telling you what to build, the lot-size rule effectively capped how many homes could go on a given block. As the Mercatus Center documents, this single number was the binding constraint on density. Change the lot-size rule, and you change how much housing the city can hold, without ever drawing a zoning district at all.

This is the lesson Houston offers that is easy to miss. People point to Houston as proof that "no zoning" produces abundant housing. The more precise truth is that Houston produced abundant housing only after it reformed the one rule that was actually limiting supply. The lever was minimum lot size, not the absence of zoning.

In 1998, the Houston City Council approved a major change to its subdivision rules, effective in 1999. Inside the I-610 Loop, the inner ring of the city, the by-right minimum lot size was cut from 5,000 square feet to 3,500 square feet, according to the Mercatus Center. That alone allowed roughly 40 percent more homes on the same land.

The reform went further for qualifying subdivisions. Where developers met certain conditions, lots could be subdivided down to an average of about 1,400 square feet per home, per the same Mercatus analysis and the Bipartisan Policy Center. A lot that once held one house could now, in the right configuration, hold three. That is a roughly 72 percent reduction from the old 5,000-square-foot floor, and it is the change that quietly reshaped large parts of inner Houston over the next two decades.

It is worth being precise about what this did and did not do. It did not legalize duplexes or apartment buildings by changing a use rule. Houston has no use rules of that kind. What it did was shrink the unit of land on which a single fee-simple home could sit. A builder could now buy one old lot, split it into two or three narrow lots, and sell each one as a separate home with its own deed. The home type that fit this pattern was the three-story townhouse, and that is what got built.

The 1998 change applied mainly inside the I-610 Loop. For roughly fifteen years, the rest of Houston, the larger outer city where most land actually is, kept the older, more restrictive rules. In 2013, the City Council extended the small-lot provisions across the entire city wherever municipal wastewater service was available. The vote was 13 to 3, according to the Mercatus Center.

The practical effect outside the Loop was large. Before 2013, the HUD Cityscape account notes that a builder could put four houses on a 16,000-square-foot lot; after the reform, with one shared driveway counting toward the required open space, the same lot could hold six or seven homes. That is a roughly 50 to 75 percent increase in the number of homes a single parcel could carry, achieved by changing the lot-size and open-space math rather than by rezoning anything.

The 2013 extension is the more important policy event for anyone studying reform, because it took the idea out of one favored inner district and applied it to an entire major city. It also produced the cleanest before-and-after data, which is what later allowed researchers to measure the effect on land prices.

Houston did not force smaller lots on every block. Both the 1998 and 2013 reforms came paired with a block-level opt-out. Homeowners on a given block could petition and vote to keep the old 5,000-square-foot minimum for their own street, a tool Houston calls a "special minimum lot size" or minimum-lot-size block designation. If a block opted out, the small-lot rules did not apply there.

This design choice is one of the most copied parts of the Houston model. It defused the most common objection to density reform, the fear among existing homeowners that change would be imposed on them. By letting each block decide for itself, the reform let the people most worried about density protect their own street while leaving the rest of the city open to new building. In practice, the opt-out was used but did not swallow the reform: about 16 percent of eligible blocks took the special minimum lot size and kept the old 5,000-square-foot floor, according to Market Urbanism, and the blocks that opted out skewed more affluent and white. The remaining 84 percent stayed open to smaller lots, which is where the new housing went.

The results are the reason Houston gets cited at all. Since the reforms, nearly 80,000 homes have been built on small lots in Houston, according to the Mercatus Center. Narrowing to the clearest measure, the academic record finds that between 1999 and 2016, 25,269 new residential parcels smaller than 5,000 square feet were developed, as reported in the Mercatus review of Houston's subdivision reform. These are not small numbers at the margin; they represent a meaningful share of all the new housing the city produced over that period.

The affordability data is the part that matters most for policy. The homes created by this reform came in cheaper than the alternative. The Pew Charitable Trusts reports that townhouses built through lot conversions had a 2020 median assessed value of about $340,000, compared with about $545,000 for other new single-family homes. The same Pew analysis counts more than 5,300 conversions of detached single-family homes into townhouses between 2005 and 2020. Lower prices were not an accident of the market; they were a direct result of fitting more homes on less land, so each home carried a smaller share of the land cost.

Houston as a whole stayed comparatively affordable. The HUD Cityscape account notes a Houston median home value of $258,055 against a national median of $327,514, and a median house price equal to about 3.3 times median income, the most affordable among comparable Sunbelt metros it examined.

A common worry about any reform that lets owners build more on their land is that it will simply make the land more expensive, with the gain captured by landowners rather than passed on to buyers. Houston is one of the few places where this question has been tested directly, because the 2013 citywide extension created a clean natural experiment.

Economist Emily Hamilton of the Mercatus Center studied land values before and after the 2013 reform. Across many statistical specifications, she found that the 2013 reform had no detectable upward effect on land values, and she found some evidence that it actually reduced them, according to her Mercatus working paper. The same finding was published in HUD's Cityscape, the federal housing research journal, which gives it weight beyond a single advocacy source.

The likely reason is supply. The reform did not just permit more homes in theory; builders actually built them, in large numbers, which added enough housing to offset the upward pressure that more development rights might otherwise put on land. This is the single most useful empirical result Houston offers any other city: under the right conditions, letting people build more housing per acre can deliver cheaper homes without inflating land prices.

Houston's reform is not a model of everything. It is important to be clear about what it produced and what it did not. The reform unlocked fee-simple townhouses, individually deeded narrow homes, not duplexes, triplexes, or apartment buildings in the way that "missing middle" reforms in other cities aim to do. Because Houston has no use zoning, the lever it pulled was purely the size of the lot, and the building form that fits a very small lot with its own deed is the tall, narrow townhouse. The Mercatus Center describes the dominant outcome as the shared-driveway townhouse, where the common driveway counts toward each lot's required open space.

This narrowness draws real criticism. The townhouses are frequently three stories, built to a similar template, and the design quality is debated. The shared-driveway layout, while efficient, produces a streetscape some residents find monotonous. And because Houston relies on the lot-size rule alone, it does not directly produce the gentler two- and three-unit buildings that fit into existing neighborhoods, which is the form many other reform efforts are specifically trying to enable. Houston shows that a single rule change can unlock enormous volume, but the form that volume takes is shaped, and limited, by exactly which rule you change.

British Columbia's Bill 44, the 2023 Small-Scale Multi-Unit Housing law, pulled a different lever than Houston. Bill 44 changed the unit-count rule, requiring most municipalities to allow three to six homes on lots formerly limited to single-family or duplex use. Houston changed the lot-size rule and let the unit count follow from how small a lot could legally be. These are two different tools aimed at the same goal of more homes per parcel.

The contrast is the lesson. The Mercatus Center sets Houston's nearly 80,000 homes against Minneapolis, which permitted only 104 duplex or triplex units in the first two years after its widely publicized 2019 reform. Minneapolis legalized more units per lot but kept other rules, including lot sizes and building-form limits, that made it hard to actually build them. Houston attacked the rule that was truly binding. The result was a difference of three orders of magnitude in homes produced.

For BC, the takeaway is not "copy Houston." It is that legalizing more units on paper is not enough on its own. If the lot is still too large, the setbacks too deep, the parking minimums too high, or the permitting too slow, a generous unit-count rule produces very little. Bill 44's success will depend on whether the surrounding rules, lot dimensions, site coverage, parking, and approval timelines, actually let those three to six units get built. Houston is the clearest evidence in North America that the precise rule you reform, and whether it is the one actually stopping construction, determines whether reform produces tens of thousands of homes or barely a hundred.