Building software to demystify land use regulations
We’ve seen a string of YIMBY wins in recent years at the state level in California and elsewhere and a growing number of local victories in, for example, Sacramento, Austin, and Minneapolis.
Self-imposed housing scarcity has negative ramifications for our wellbeing across many axes, from public health to climate change, so these wins are heartening. There’s momentum in the right direction.
But actual new housing development is still painfully slow. The regulatory changes are necessary but not sufficient. We also need lots of property owners to choose to build more housing.
One factor holding back more housing supply is that the opportunities to build aren’t widely enough known or understood.
Case in point: my brother recently purchased a house in South Pasadena, one of the NIMBY-est cities in California. He and his partner are planning to gut-renovate and expand the place. They were unaware until I informed them that they’re allowed to build two ADUs by right.
My proposal is to build software that provides a clear and persuasive answer to the question, "given a parcel, what’s the maximal housing I’m allowed to build on it?"
I intend to begin with just one city, because there's so much variance between different geographies, and each is unique in many ways. My longer term goal is to do this statewide and then nationwide.
To do this well will require assimilating and interpreting parcel data and state and local zoning and building codes and case law, as well as thoughtful user experience and graphic design and some financial modeling.
I can build an MVP using ordinary & vanilla software tooling and hand-parsing and interpretation of the relevant artifacts.
Because curiosity and self interest are powerful motivators, I intend to frame the site in terms of the self-interest of land owners and/or potential land purchasers.
Upshot: I believe well designed software tools can help bridge the gap between policy changes and real-world housing supply by appealing to self interest, and I want to try to do that at scale.
Describe why you think you're qualified to work on this
I've focused my professional efforts on trying to achieve socially positive outcomes along avenues suggested by thoughtful academic research.
I was an early employee at Opower, where I helped design and develop software-powered direct mail programs that have so far yielded more than 31 terawatt-hours of energy usage reductions.
More recently I founded Vote Forward (https://votefwd.org), a 501(c)(4) nonprofit that organizes volunteers to send handwritten letters to their fellow citizens encouraging voter turnout. Inspired by the work of folks like Don Green and Todd Rodgers on social pressure for voting behavior, I started with a small RCT, sending 1000 letters from my kitchen table, and saw a startlingly large (>3 percentage points) effect on turnout. On the strength of that result I scaled it up, first with help from friends, and later with support from donors to hire a team. So far more than a quarter million people have written a total of tens of millions of “please vote!” letters through Vote Forward. Overall I can say with high confidence that we’ve had a significant influence on the makeup of the electorate, including in some close and highly consequential elections including the 2020 Presidential & Senate contests.
So two notable successes in using software to drive real-world outcomes at a large scale — energy efficiency in one case, voter turnout in the other — with rigorous impact measurement.
Earlier this year I passed the baton to a new Executive Director of Vote Forward. I’m still a board member, but my duties are fairly light, so I’m now exploring options for a next big project.
My abilities and proclivities are well suited to this one. I’m a competent software developer in my own right, so I personally can do a lot of the work I’m proposing. I also have a track record of taking inklings of the sort that I have here now, building first versions in scrappy ways, and then scaling them up.
Most importantly, I feel quite motivated to do this. I find the choices we’ve made as a society about land use to be utterly maddening. I think I have a solid idea here for how to make some progress and help a lot of people join me in seeing the situation more clearly, and get more housing built, so I’m eager to give it a try.
Other ways you can learn about me
I’m @sjforman on Twitter and sjforman.me on Bluesky. I also have a (recently published, so pretty minimal) personal website at sjforman.me. I haven’t historically been much of a poster so these aren’t that rich of a source of info, but will suffice to validate my identity.
How much money do you need?
I think I need $15k to take a swing at this. I loosely anticipate spending about $5k to pay for some expert advice on the land use legal landscape, $5k for user experience and graphic design support, and $5k for software development assistance.
Estimate your probability of succeeding if you get the amount of money you asked for
I estimate a high probability that, given funding, I can manifest some version of the idea I'm describing here, perhaps 80% likelihood.
A stricter definition of success would be demonstrating that I've caused at least one additional unit of housing to come into being. I'm not sure how or on what timeframe to measure that though, so I'm reluctant to assign it a probability.
Cautionary Note
I am eager to make a foray here, but this is a new domain for me — I'm not a planner, architect, or real estate pro. So there is a significant chance that — as with all early stage projects of this kind — I will discover as I dive deeper that my initial idea doesn't make sense, or doesn't have the impact I hope, for reasons that I can't foresee. In other words, funders should be comfortable with the possibility of pivoting.
Possible Future Directions
I suspect the data and analytic tools I'm envisioning to help demystify land use regulations at the level of individual parcels could be useful in more systematic ways.
Building Political Support
I want to be able to say to users: “If Senate Bill X is enacted this year, you’d be allowed to build a third ADU on your property. Even just passing the law would increase your property value by approximately $10k, and if you decide to build such an ADU, you stand to gain $100k to $200k in property value. You can call your assembly member to express your support at…” The converse message could be sent for down-zonings.
Helping to Shape Policy
I also envision eventually using these analytic tools to provide policymakers with increased visibility into the real-world impacts of legal and regulatory changes under consideration. Matt Yglesias recently:
"The thing I keep saying YIMBYism needs next is better analytic capacity to assess the actual impact of specific policy changes on housing outcomes. Right now, I think advocates are to an extent flying blind in terms of which things are really worth fighting for and which are less important."
I agree, and I think what I’m going to build could be quite powerful in helping to navigate between alternative policy pathways or to further arm advocates with hard data to make their case.
Application status
App status across various funders