Reduced To A Study

There is a sentence in the building code that says a residential building above three storeys needs two separate exit staircases.

No other wealthy country's code says this. Several of those countries have better fire outcomes than the United States. The rule adds something like ten percent to the cost of a mid-sized building, and it is the reason a great many small lots in American cities have sat empty for decades.

This piece is not really about staircases. It is about what happened when people finally tried to change it, because that part is a management story and it is one you will recognise.

The rule nobody can source

The International Building Code is the document nearly every American state and city builds its own code on top of. It requires two exit stairs above three storeys, reachable from every point on every floor.

Despite the name, it is not an international document. It has been adopted across the US, its territories and the Caribbean, and it has influenced codes in Mexico, Abu Dhabi and Haiti. It plays no part in the building codes of other OECD countries.

So the requirement is not a global safety consensus that America happens to share. It is closer to a local convention that acquired the authority of one, and the countries that never adopted it are not burning down. Several of them do better than we do.

Two stairs in a small building is not a detail, it is the floor plan. You need a corridor to connect them, which forces single-aspect flats, long dark hallways, and on a narrow lot an arithmetic that frequently does not close at all.

What happened when people tried to change it

Here is the sequence, and it is the whole point.

California. AB 835 originally required the State Fire Marshal to propose a code change permitting single-stair buildings up to six storeys. At the request of the California Professional Firefighters, the committee amended it to require a research report instead.

Minnesota. SF 3538 would have required the Labor and Industry Commissioner to amend the code to allow single-stair buildings up to seventy-five feet. It died in committee. A study passed instead, inside a budget omnibus.

Washington. SB 5491 originally carved single-stair reform out of the requirement that cities adopt codes at least as strict as the state's. In committee it was reduced to a study requirement.

Three states, three real bills, three studies. The reform was not defeated. It was converted into homework, and homework does not have a deadline that anybody loses an election over.

The studies came due

Those commissioned reports were due by 2026, which is the interesting part, because 2026 is now.

In March, the California Office of the State Fire Marshal published a report urging caution. Its language is measured and, to be fair to it, entirely reasonable: "The presence of an additional exit is important for maintaining safety in the face of unforeseen failures, fire spread, or structural compromise."

That is what a study produces when the study was the compromise. Not a decision. A considered restatement of the concern that caused the study to be commissioned in the first place.

Meanwhile, the states that just did it

Not everybody took the homework option, and the contrast is instructive.

Tennessee wrote single-stair buildings up to six storeys directly into the state building code, leaving local governments to adopt the provision as they update their own.

Colorado passed a law in May 2025 allowing multifamily buildings up to five storeys with one staircase in municipalities over 100,000. An earlier attempt had failed in 2024.

Michigan signed two bills on 21 July 2026, raising the single-stair limit from three storeys to four, and permitting it under six storeys where an accredited fire department serves the area. The state's own estimate is a construction cost reduction upwards of ten percent.

Seven states enacted some form of this in 2025.

What Colorado did differently

Colorado failed first, in early 2024, and the second attempt is the case study.

Representative Andrew Boesenecker of Fort Collins described the approach plainly: "We knew that we would need a long runway, based on the conversations that had transpired with the previous bill that did not pass."

What followed was a listening tour with the people positioned to kill it. The Fire Marshals Association of Colorado. The state chapter of the International Code Council. The Associated Builders and Contractors. It produced nearly two dozen amendments to the bill: staircases at least forty-eight inches wide, a maximum of four apartments per floor, smoke alarms meeting National Fire Protection Association standards, sprinklers and materials still held to full code.

Read that list again. It is not a watered-down bill. It is the same bill with the objection designed into it, which is why it is now law and California's is a report.

The pattern, which is not about buildings

Every organisation has a two-staircase rule.

A required approval nobody can source. A mandatory field on a form. A review step that exists because of an incident in 2019 that four of the five people involved have since forgotten. It was adopted for a real reason, it is defended on safety grounds, and it quietly taxes everything you ship.

And when somebody proposes removing it, the same thing happens. Not a refusal, because refusing is confrontational and nobody wants that meeting. Instead: let's study it. A working group. A review. A recommendation due next quarter, which arrives as a considered restatement of the original concern.

The tell is that the study is commissioned by the party that did not want the change, and its scope is written by them too.

The playbook

1. Ask who owns the rule, not whether it is a good rule

Debating the merits is the trap, because the merits are contestable forever. The productive question is who would have to agree, and what they are actually protecting. In this case it was fire safety, held by people who run into buildings, which is a concern worth taking seriously rather than out-arguing.

2. Design the objection in before you table the proposal

Colorado's two dozen amendments were not concessions extracted in a fight. They were gathered before the fight, from the people who would otherwise have opposed it. That is the entire difference between the states with a law and the states with a report.

3. Treat "let's study it" as a decision

Because it is one. It is a decision to keep the current rule while sounding open-minded. If a study is genuinely warranted, it gets a scope, an owner, a date, and a pre-agreed statement of what result would change the outcome. Without those four things it is not research, it is a filing cabinet.

4. Find out what your rule costs, in a number

"Upwards of ten percent of construction cost" is why this reform moved at all. Nobody repeals a rule because it is inelegant. Put a figure on the tax and the conversation changes character immediately.

5. Look at who does not have your rule

The strongest fact in this entire debate is that other wealthy countries do not do this and are not worse off. If everyone in your category has a constraint and nobody outside your category does, the constraint is probably a convention rather than a law of nature.

6. Expect adoption to lag the decision

Tennessee changed the state code and left adoption to local governments. Passing the reform and getting the building are different events, often years apart. Whatever you change internally, the announcement is not the outcome.

Our position

The most useful thing here is not that a staircase rule is being relaxed. It is the demonstration that a rule can be genuinely load-bearing in cost, genuinely weak on evidence, and still survive for decades because the people who benefit from removing it are diffuse and the people defending it are organised and sincere.

That describes a great deal of process inside a great deal of companies.

The lesson from Colorado is not that the objectors were wrong. They were not, particularly. It is that the reform only shipped when somebody stopped treating the objection as an obstacle to get past and started treating it as a design constraint to build around. That took a failed attempt, a long runway, and two dozen amendments.

Everywhere that skipped it got a study.

Sources

Multifamily Dive, Ryan Kushner, "Single-staircase reforms aim to boost middle-scale housing in Michigan" (2026-08-06), for the two Michigan bills signed 21 July 2026, the three-to-four storey change, the accredited fire department provision, the ten percent cost estimate, the seven states in 2025, and the California State Fire Marshal quotation.

Niskanen Center, "Understanding single-stair reform efforts across the United States" (August 2024), for the IBC requirement and its limited international adoption, the fire outcome comparison, the Seattle, New York City, Honolulu, Chicago, Dallas and Vermont exceptions, the Tennessee code change, and the legislative histories of California AB 835, Minnesota SF 3538 and Washington SB 5491. Dated 2024; its state counts describe that moment and are not blended with the 2025 and 2026 figures above.

APA Planning Magazine, Joe Tedino, "How a Single Stairway Can Take Affordable Housing to a New Level" (2025-11-17), for the Colorado law, the failed 2024 attempt, the Boesenecker quotation, the listening tour, the two dozen amendments and their specifics, and Stephen Smith of the Center for Building in North America on per-building cost.

City Observatory, Joe Cortright, "Single Stair Reform and Housing Affordability" (2026-03-14), and City That Works, Zak Yudhishthu and Alex Montero, "Small Lots and the Case for Single Stair Reform" (2026-04-28), for the floor plan and small lot consequences and the distinction between zoning code and building code.

One connection we did not assert. California's March 2026 Fire Marshal report and the research report commissioned by AB 835 may well be the same document, and the timing fits, but our sources do not state that explicitly and we have not claimed it.