From the Ground Up: A New Report Analyzing LVT Shifts in Cincinnati
Six years ago, as a Junior in college, I started a club called the Notre Dame Student Policy Network. The thesis went something like this: most students studying political science enter college without understanding what a job in policy looks like. Meanwhile, governments and think tanks are understaffed; a student group could partner with governments and think tanks to provide students hands-on, real world experience. The club has grown substantially and continues to do real, impactful work every semester. And somehow, I have ended up on the other end of a Student Policy Network project, being the think tank that partners with the students.
Over the past half year, students at Notre Dame have been investigating Cincinnati’s property tax structure and modeling a land value tax. This week, the Center for Land Economics released the report in collaboration with the Student Policy Network.
The report has the following high level findings:
About $1.87 billion in land value, nearly one in five parcels, sits vacant or underdeveloped, much of it in the city’s most valuable neighborhoods.
The median single-family homeowner pays less, with roughly half paying less and about one in four paying more.
Two-thirds of multi-family parcels see tax cuts above 10%, with the median multi-family property saving about 16% to 17%.
Vacant parcels see a median tax increase of about 126%, and every vacant lot sees its bill rise more than 10%.
The burden shifts progressively: the lowest-income neighborhoods see median cuts near 17%, while only the highest-income quintile sees a median increase of about 11%.
The property tax system in Cincinnati creates the wrong incentives. It penalizes those who develop, and subsidizes those who make money from speculating on raising land values. On one intersection in Cincinnati, low income housing is taxed 10-20x more than surface parking and vacant land on the same block.
The report then modeled a 4:1 shift to land value taxes, where the city collects the same property tax revenue on their general operating millage, but land is taxed at four times the rate as buildings. In Cincinnati, the result is a 10% decrease in taxes for the median single family house. Meanwhile vacant land would nearly double in property taxes.
Across neighborhoods, the shift in property tax burdens is progressive. Residential properties in the lowest income neighborhoods see tax decreases on the median of 16.7%, while those in the highest income neighborhoods see tax increases of 10%.
As a quick aside, that increase for the top quintile is heavily concentrated in a couple of wealthy neighborhoods. Those wealthy residents may end up being political opposition, and by nature of their wealth, may have outsized local political power.
If you are a frequent reader of this Substack and our work at the Center for Land Economics, you will be used to seeing the figures above, but they are worth showing again and again: shifting property taxes to land value taxes is a pragmatic policy that solves real incentive problems in our urban cores.
This report comes with the backdrop of a statewide push to abolish property taxes. Lincoln Institute put out a good report last month; to abolish property taxes in Ohio would eliminate over $20 billion for schools and municipalities annually unless municipalities were to increase their pass 13% sales taxes to offset the decreased revenue.
Republican State Senator Louis Blessing has been in vocal opposition to the abolition of property tax proposals; instead, he has put forward a proposal to enable cities to implement split-rate taxes (he spoke about it in our webinar last month). This report is an analysis of what could be possible if this proposal were to pass.
Want to create similar data analysis in your city? We released software that helps anyone model LVT in their local area, and I have a fresh update on that!
Put It On A Map: LVTShift Edition
Lars has a mantra: Just Put it on a Map. It is much easier to notice trends and understand data analysis when you visualize geospatial attributes on a map. Yet, when I launched LVTShift 2.0 (you can now vibe code LVT modeling), it had no map creating feature!
In part, this is because I was in the habit of just bringing its raw exports into mapping software to visualize the results manually. But no longer!
From now on, LVTShift also outputs an HTML file complete with an interactive parcel map on top of a basemap (plain or satellite view). Each parcel can be individually clicked on to reveal parcel information. Further, at the bottom of the webpage is an easy way to also see all the graphs created by the LVT Shift analysis. I have a feeling this is the first step toward something a lot cooler… an interactive visualization of LVT in every city.
Down With Google and Other CivicMapper Updates
How many Slack messages from Lars does it take for me to finally let him get rid of the Google sign in on CivicMapper? It turns out it was more than a couple! But, with no strong reason to keep it, and with increasing updates to CivicMapper, I have finally caved, and we have de-Google-ified CivicMapper.
Now, anyone can view CivicMapper without any sign in. And with so many new cities, we needed a better organized front-end to select the city you want to see.
And as the new map view reveals, CivicMapper has gone international! We have just added Tallinn, Estonia and Copenhagen, Denmark. Both of these places have available open data, and we thought it’d be interesting to explore them. Estonia is particularly interesting as it administers a national land value tax.
More visualization controls
We’ve updated the visualization controls with a couple features requested by users. First, you can disable the 3D view and get a pure top-down, non-extruded 2D map. Next, we took the previous maximum height and made that be the new 50%, so you can really stretch things out now. Finally, we added new color scaling options – adding “logarithmic” to our previous choices of linear and quantiles, and now you can adjust the positions of the colors themselves so you can get the perfect screenshot, just drag the sliders back and forth. You also have the option to invert the color ramps.
Neighborhood selectors
We’re also starting to slowly roll out a new feature – regional overlays AKA “neighborhoods.” These are derived from shapefiles that carve up a city into different areas, and are sourced from open data sources.
This feature was added at the behest of Strong Towns Houston, and the feature is best exemplified in that city:
This is only available in a few cities right now, but where it’s enabled, click on the eye icon to show or hide that region, and click on the rectangle to show or hide the overlay. As you do this, whatever value you’ve currently selected will show an updated summary of the currently visible sum.
The hope is that this feature will aid in live “story-telling” where activists are able to use Civic Mapper to tell stories such as “here’s how much land value is in Downtown alone”. This is particularly effective when combined with the parking lot tab that visualizes how much of that land value is dedicated to surface parking lots:
OpenAVMKit: Marching along
It’s been a while since we have posted about OpenAVMKit–an open-source automated valuation model to value every parcel of a city, and the groundwork of research we are undertaking at the Center for Land Economics. We shipped a major release back in June that not only fixed a bunch of bugs and new features, but also added a ton of robust documentation, which is particularly useful if you want to use AI agent tools to drive the library. You can see the full changelog here. We plan to have another release out sometime in the next month.
Hey, what format and data would you need to add Birmingham Alabama to Civic Mapper? (We can do the basic analysis)
Thanks for the good news for land value mapping. How to do this for my town of Chambersburg, PA? and also for our LVT projects in Costa Rica, Nigeria and Colombia?