Accurate Land Valuation Comes to Madison
The assessor's office improved land valuations, why this matters, and what it could enable
The 2026 assessments were released in May in Madison, WI. At the surface, it’s business as usual. City-wide assessments are up 6.1% and residential assessments are up 4.8%, which I would consider normal. If you’re a homeowner and you looked a little closer, you’ll notice something very different this year. My improvements went down for the first time in the last 10 years and my land increased so much that my property’s total value still rose by 8.5%.
For my house, the land allocation (the % of land value of the total value) increased from 26% to 38%! Was this just me or did this happen elsewhere? We’ll get a sense by checking how land allocation has trended for all residential property assessed in the city.
This is a sharp change from the 2024 and 2025 values (~25% and ~23%). On average, land makes up almost 29% of residential property values in 2026, and we know that residential total values have increased, so land values must have changed significantly.
As a Georgist, this really excites me. I appreciate the assessor’s office’s continued work to provide accurate assessments, with their recent market correction on land values and adjustments in land allocation. We’ll look at how this plays out across the city and what the implications are for land to be valued more accurately.1 First, let’s talk about property taxes.
Property Tax 101
Most people assume that increasing property values result in higher property taxes, but it’s more nuanced than that. First, the city sets a levy, which is the total amount to be raised by property taxes.2 Once we have the levy, it’s split across all taxable parcels in the city based on the assessed value. Properties with higher value bear a higher proportion of the levy and vice versa. So when you look at your new assessment, you need to understand how your property’s value increased relative to all other taxable property in the city. Think of it as your property having a slice of the levy pie. The last piece of the puzzle is the mill rate: divide the levy by the total value of assessed properties, then multiply by your property’s value to get the tax bill. If you’d like to read more, I recommend this post!
Above I mentioned that on average, property increased by 6.1% and my property value rose by 8.5%, so my slice of the pie increased this year compared to last year. Absent levy increases, I can expect my property taxes to rise by ~2%. This is fine by me because I love property taxes. If your property value increased by less than 6%, you might even see your taxes fall with levy increases.
Your property’s value is divided into two pieces: land and improvements. For residential properties, improvement comprises the house/building/whatever you want to call it. The land is the dirt. For our current property tax regime, the difference doesn’t have a lot of meaning at face value because we’re (likely) required by state law to distribute the levy based on the total value, but we’ll come back to this later. Land value is where all of the interesting assessment changes happened in 2026.
Land Values in 2026
Let’s take a closer look at how assessments changed for residential properties, including the land and improvement components. I’ll generally show the aggregate weighted means (matching the assessor’s summaries) and the medians given the distribution shapes in the data.
As I’ve said before, residential property values are up 4.8%, and the uncharacteristic parts: improvements are down 2.5% and land is UP 28.7%. Using my house as an example, my improvements decreased by 9% and my land increased by 59% (total value up 8.5%). We can see that our median land value change is around half the mean, indicating a long right tail on the distribution. This makes sense. The reassessment is a correction to market land values, catching up on a multi-year drift that was concentrated in high-demand areas.
Land is a big deal here.3 You might be asking, what the heck gives land value?
We’ll keep this simple and focused on residential property, but know that there are more details to dive into. Improvements are expected to be valued based on the cost of construction for the structure minus depreciation. Land, on the other hand, does not depreciate. Land’s value is derived from its proximity to amenities and scarcity (aka location location location). In Madison, you’ll find land value is gigantic downtown because of the proximity to lakes, jobs, restaurants, etc. As you move away from downtown, land values typically drop.
Let’s think through my own property. My house was built in 1940 and is lovely and I’m lucky to have it, but the depreciation aspect is real. The windows are old, the insulation isn’t the best, my roof is almost 20 years old, so it’s reasonable to expect that there’s depreciation at play. My land (aka location), on the other hand, is great. I’m on the near west side of Madison, very close to the Southwest Commuter bike path, Hilldale Mall, and the Sequoya library. I’m lucky to live in an amenity-rich area, and my land should reflect that value. To illustrate where land is most valuable in Madison, take a look at this map.
I’ve animated the map of land values/sqft to illustrate land value changes across 2024, 2025, and 2026. If you look at this, you can see which spots get darker on our color gradient. Notice how the map doesn’t change between 2024 and 2025. 2026 truly represents a change in how land values are assessed. We see that central neighborhoods on the near west/east sides saw the largest increases in residential land value per square foot. These are high demand areas and that demand will naturally be reflected in the land value. The value is derived from the accumulation of jobs, nice restaurants, improved infrastructure, etc. All things contributed by the community or tax dollars.
How Vacant Lots are Impacted
We have 1,417 vacant residential parcels in Madison, which makes up ~2% of our total residential parcels (72,361). I won’t comment on how this compares to other cities, but nevertheless it presents at least some opportunity for increasing housing supply.
You can reasonably expect that between two parcels that are the same shape, same size, same block, and same allowed use, they should have a similar land value. This is true between improved parcels and vacant parcels, which are only made up of land value. With our improved parcels now having higher land values, does it hold true that the vacant parcels now have higher total value? Yes! We see this to be true city-wide, where vacant parcels’ total value is up by a lot.4
The vacant parcel data is highly susceptible to outliers, so I’m showing the aggregate weighted mean (79.1%) to match the type of calculation the assessor would do, and median (17.9%) to reflect the typical vacant parcel. Again, we have a long tail on the value increase distribution because we’re seeing a market correction that has accumulated over years. This means that many vacant residential parcels will see higher property taxes next year because now their slice of the levy pie is larger than it was last year.
We actually see that residential properties as a whole will see a decreased share in the levy, while those vacant lots will experience a larger share of the taxes compared to 2025. I did a crude estimate with vacant lot sizes and their respective zoning to make a guess on if they could build a duplex and a duplex ADU (both allowed nearly city-wide).5 Approximately 2,900 units could be built as a combination of duplexes and ADUs. This doesn’t consider splitting lots or single unit ADUs, but rather I want to convey the magnitude of the potential.
Why Accurate Land Valuation Matters
Now I’m going to do a little dive into an old, but resurging, set of ideas from Henry George. In the late 1800s, Henry George observed that during a time of progress with the industrial revolution, there was an ever increasing rate of poverty. He surmised that land owners in a thriving area saw incredible financial windfalls as industry surged around them. This progress increased demand to live in these industrious areas, and with it rents rose due to rising land values. Landlords were able to privatize the increased land value without contributing to the conditions of success, essentially sapping the wealth generated by the community. Similarly, owners of vacant land can hold onto it waiting for it to have a future higher value and sell at a profit, again without any effort required to reap the benefits of rising land values. This is called land speculation and I don’t think it’s controversial to say that this is undesirable behavior. Henry George’s suggested solution was simple: tax the rental value of the land, aka land value tax (LVT). The structure of this tax means that the speculative gains from land are taxed and justified by the fact that the land value is a result of the community’s effort and tax dollars, and as such, that value should be returned to the community.
Back to modern times, we see land speculation happening in holding vacant land. I’m not here to say that the people that hold this land are maliciously participating in land speculation, but the fact remains that by holding this land unused as it gains value, land speculation is happening regardless. By accurately valuing land, we’re increasing the holding costs of land, which puts pressure on vacant land owners to sell to someone willing to use the land. Property taxes are partially land value taxes already, so any increase in the holding costs for land directionally reduces land speculation. Even though vacant parcels are going to see higher taxes, the tax bill will still be smaller than nearby improved parcels.
Let’s take a look at two lots on Gorham St in downtown.
Here we have a rental 3 unit townhouse next to a vacant lot (used for parking). The total value of the townhouse has been rising steadily over time, while the vacant lot has been largely static over the last few years and has been valued lower than the land component of the improved lot. The 2026 land value assessments show a break in the trend. For one, the land values match between the two lots now, which makes sense because they’re the same size, in the same location, and allow the same use. The total value of the townhome increased slightly above the city-wide average, so taxes will increase slightly. The vacant lot on the other hand, will see a more substantial tax increase because its total value increased by 36%. The important takeaway here is that the overall value of the improved lot increased by ~8.5%, while the overall value of the vacant lot increased by 36%. This will increase the tax burden on the vacant land.
If we consider the tax bills for these properties in 2025, the 3-unit townhouse paid $9,957.01 in taxes. The vacant parking property next door paid $2,363.74. The townhouse gets charged more than 4 times the amount for property taxes. Bluntly, the parking lot gets a tax break by not having a building on it. Is this really the financial incentive structure we want in a city that needs more housing?
Property Taxes are 50% Good
I said before I love property taxes, but that’s only 50% true. The tax on land is great. The tax on improvements, not so much. A significant source of housing unaffordability is due to constrained supply, and for anyone that has spent any time reading about housing, there are familiar barriers to this like zoning. A tax on improvements is a lesser known disincentive on housing supply. If you tax something you get less of it. Taxing buildings means we’ll get less building. Land on the other hand, is in fixed supply, so taxing land more will not result in less land, but more efficient land use. For the existing homeowner, this means that you’re disincentivized from adding a new bedroom/bathroom because your property taxes will probably increase. If you’re from Madison, you’re probably familiar with the very old rental houses on W Washington Ave downtown. The proximity to downtown lets the landlords maintain higher rents, especially absent adequate supply. The landlords can charge market rent because of demand, but if they maintain the houses adequately, that will increase the improvement value, which can increase their property taxes. The current property tax structure is punishing people for building and improving, and rewarding people for neglecting and holding.
While it’s not a full LVT, I’m in support of a universal building exemption (0% tax on buildings and 100% on land) or split-rate tax. This would have different rates for taxing improvements and land, pushing us in the direction of discouraging land speculation and encouraging good use of land. These are revenue neutral approaches, so taxes are shifted based on the property’s new slice of the pie from different build/land tax rates. One precursor to these reforms is accurate land valuation, so we’re making progress!
Pennsylvania is the classic example of municipalities that have had split rate taxes, where land is taxed at a higher rate than improvements. Harrisburg implemented a split rate tax in 1975 and the mayor at the time credited the policy with a decrease in vacant structures from ~4,200 in 1982 to fewer than 500 by 2001.6 Recently, Virginia and Kentucky have passed legislation to enable implementation of LVT in some form.
LVT’s Future in Wisconsin
The road to LVT in Wisconsin is likely a long one (read more from my fellow Madisonian, Josh Olson, on this)7, but I can’t stress enough how important it is to value land accurately. As I said before, this is a prerequisite for the policy, but it also allows for analysis on the impact of enabling LVT shift. We could model who pays more, who pays less, and how the shift lands across renters vs. owners and across income groups (likely a future post). In the meantime, it’s important to keep pushing for the standard YIMBY playbook on liberalized zoning, faster permitting processes, anything that will facilitate building so that vacant lots or end-of-life structures can actually turn over. In fact, I think the new land valuations help make a strong case for lot size reform: if land is such a large piece of property value, you can makes gains on affordability by having a smaller lot.8 With these policies in place, the economic feedback mechanism of LVT can be met with feasible development projects.
Appendix - Distributions
Some of you probably want to see some distributions instead of averages/medians alone, I’m here for you:
I’ve stored the scripts/visuals for this work in this GitHub repo if you’re curious.
Wisconsin limits levy growth by indexing it to net new construction (new buildings - demolished buildings) or requires levy increases to be approved by a referendum. This means that a municipality that doesn’t build much will have a harder time keeping up with inflation for its expenses. Madison recently passed a slew of referenda for the city’s operating budget and the school district’s capital/operating budget (the school referenda will make up the largest future increases in our property taxes).
Land Is a Big Deal is also a great book to learn more about this stuff.
The reassessment of vacant parcels shows a total value of $68.7M in 2025 that jumps to $119.4M in 2026, an effective increase of $50.7M on the tax roll. This only considers existing vacant parcels that did not have a previous assessment of $0.
I estimated that parcels that meet the mapped zoning’s minimum lot size and are at least 6,000 sqft can fit a duplex main structure and duplex ADU. For parcels under 5,000 sqft, I assume a duplex fits. Parcels that are below their zoned minimum lot size are considered unbuildable, which should take care of abnormally small parcels. This misses lot splits and lots with unbuildable dimensions, but I’m just trying to get a broad sense of opportunity.
Neither Josh nor I are lawyers, so this is just our interpretation around the legal precedents. That being said if you’re a lawyer and want to interpret the precedent around Wisconsin’s uniformity clause, we could certainly use your help!
Strong Towns Madison will likely be making a case for lot size reform and I hope to follow up with more analysis on the impacts it can have on housing affordability.