Excitement about zoning and land use changes usually isn't preceded or followed by the use of solid ... [+] data and evaluation.

I often get calls or emails from people and organizations that claim they’re doing something transformative in the housing economy. Whether it’s regulatory reform or methodology or materials or financing these people will say something like, “This can solve the housing crisis.” My usual internal response, “Yeah, right” and my out loud response, “Sure. Let’s hear it” seems preemptively cynical. Here’s why I have the reaction I do and how we might get to a place where such claims might be better evaluated.

I was on a Zoom call not long ago with an advocate in a meeting about reducing regulatory overreach as a condition of creating more housing. I pressed the group that more than anything else, before introducing any measure or intervention we thought might benefit housing production and people with less money in the economy, we need to be sure that the policy was both necessary and sufficient. I mentioned my analysis of the accomplishment of allowing three units per lot in Austin as an example of necessary but not sufficient to change the housing economy.

We didn’t bicker necessarily over the issue, but I made it clear that we have a paucity of solid data sources on the housing economy down to the level of cities and submarkets. I pointed out that there really isn’t a city in the United States that has public facing platforms to query, for example, “How long does it take, on average, for a project to get a certificate of occupancy from time of permit application?” I posted about this problem a while back , noting that lack of this data makes it impossible to know, for example, how closely changes in population, job creation, rents and prices correlate with permitting. Prospective work I’ve done has indicated that to some degree, those relationships are predictive. That is, there are retrospective data points that show how peaks and valleys in permitting.

Data is critical for understanding which interventions will improve housing prices and rents and ... [+] determining whether those interventions are succeeding.

“We looked into our apartment database and plotted the following chart. We looked at the metro’s over 35 years of history, when vacancy is lower than 20-yr average, market rent growth took off! So faster inventory growth would help tame the double digit rent growth seen over the last two years, especially when vacancy dipped to just 1.9% by 2022.”

Well, that validates what we all know about supply and demand. In Cincinnati, the same firm produced this graph of permits in Cincinnati and rents.

There is a relationship between permits, population, and price, understanding that relationship and ... [+] applying that to policy and land use is central to improving housing affordability.

They said this about the data and the graph.

What’s important here is that this very interesting data came from this firm’s proprietary methodology. I have no idea all the under-the-hood stuff behind these analyses. We simply can’t do this at will in markets across the country and state and local government certainly can’t either.

In the discussion on the Zoom call, my interlocutor I think literally shrugged. “We don’t need all this data to know something works.” Really? So, if we reduced minimum lot sizes in Albuquerque and Cincinnati or allowed accessory dwelling units by right or any number of so called, YIMBY (yes in my back yard) measures, we just know that it will be effective to lower prices and rents. We have no obligation to determine whether those interventions worked or not.

I used to think that if we just could convince people that supply and demand was as real an operative as a heliocentric solar system is in physics and astronomy , that our troubles with housing policy would be solved. I was wrong. Everyone throws around supply now as a buzzword and justifies various policy changes at the state and local level as monumental accomplishments that will lead to lower rents and prices. Sometimes, people will point to modulations in quarterly rents and suggest that a flattening or drop in rents is “proof” that an intervention “worked” without any justification in the permitting or any other data.

My eyes roll every time I get messages about something “solving the problem” somewhere. Cottages, modular construction, online lending, smaller units, bigger units, office conversions, and the list goes on. Maybe these interventions solve a problem , but they aren’t solving the problem . How do I know? I don’t because we don’t have good longitudinal data for markets that show endemic trends in permitting or policy and how those impact price one way or another. If I don’t know they aren’t working, proponents can’t prove they are. Again, they might be necessary, but we can’t know they are sufficient until we have retrospective data that helps us understand where the best place to make changes might be.

My dad and I used to go to the race track from time to time where I’d place stupid bets to have some fun like $2 on a long shot horse to show, or a favorite to win. Sometimes I’d win and sometimes I’d lose, but the stakes were low. I’d see the guys who had all sorts of racing forms spread out in front of them, maybe drinking a beer and smoking. This was their job. How in the world could they make bets on animals running around track with a little guy sitting on them and know how to bet and win? Data. These guys had experience (qualitative data) and lots of retrospective information (quantitative data) about each horse and jockey.

Real estate professionals are the same way. They can look a corner and tell if building on it will make money. They can read a market and make a decision about whether their clients should invest hundreds of millions of dollars in fixed real estate assets or build or do something different. Minor changes and tweaks to zoning codes might make a difference or not. But people with bets on real property don’t just guess the way government and YIMBY activists do and then celebrate wins without any clarity whether the bet really paid off. We have to raise our expectations on housing data. Otherwise we’re making random bets with the money of people who actually have to buy housing, and people earn less money and pay more for housing need more than just a shot in the dark.