Philadelphia taxes buildings and land at the same rate, so an empty lot in a rising neighborhood costs little to hold. This study follows ten places and asks what a different tax on land could, and could not, change.
The ask: a Council hearing, a funded audit of land values, a resolution seeking state enabling law, and open data. Read the four asks ↓
Between Vine and Pine, from the Delaware to the Schuylkill, more than two dozen large lots in Center City are used for public parking. Many were paved during the urban renewal of the 1960s. Philadelphia's zoning code does not treat surface parking as a permanent use: owners need a special license, renewed every three years, and renewals have become routine.
The lots are already zoned for towers. What they lack is a reason to change. Some owners hold them with no mortgage, so their main costs are insurance and property tax, and an empty lot carries very little tax.
The pattern is not limited to downtown. Along commercial corridors in every council district there are parking lots, shuttered drugstores, tire shops and vacant parcels on land zoned for apartments over shops. Each one has its own history. Together they show how the tax system shapes the city we see.
On the 700 block of Chestnut Street, the owner of a parking lot tried four times to partner on an apartment tower. Each deal fell apart at the last minute. The lot is still parking.
Zoned for up to 258,120 sf of building. Four tower deals have collapsed. Parkway · Inquirer
The 2019 property tax on a 140-space lot zoned for up to 422,208 sf, against $91,346 in net parking income. Flynn Co. listing
A tower planned on this Parkway lot was dropped in 2023. The lot sold to a new owner in 2025, with no project confirmed. Philly Land Sites
Two counts are given because the city's records are imperfect. The larger counts every parcel the Office of Property Assessment classes as vacant land or a parking lot. The smaller keeps only those with no building value and a vacant or parking description. A field check of the 63 mid-sized lots is the next step; the checklist, with a City Atlas and Street View link for each, is published with the study.
The waiting land is concentrated where the market is hottest. District 1, which takes in Old City, Northern Liberties and Fishtown along the Delaware, holds 50 of the city's 120 CMX-3 vacant and parking parcels. Those are the places where land prices have climbed fastest.
The records also show how long land waits. Half of these parcels have no recorded transfer in nine years or more. A few have been held since the 1970s.
Citywide, the City's records list 4,478 parcels as parking lots: 757 acres, more than a square mile, holding $1.68 billion in land value. Of that, $1.05 billion is privately held and taxed. Another 9,700 acres are recorded as vacant land, $6.3 billion in land value, about half of it publicly owned.
District 1 alone has $621 million of land under parking. Districts 2, 3 and 5, which share Center City and University City, have most of the rest.
Every property tax bill in Philadelphia has two parts: the assessed value of the land, and the assessed value of whatever stands on it. Today both are taxed at 1.3998%. A split-rate tax, often called a land value tax, puts a higher rate on land and a lower one on buildings, collecting the same total. This study uses one setting throughout: land taxed at four times the building rate, a moderate ratio within the range Pennsylvania cities have used. Other ratios can be tested in the workbench (chapter 11).
Every number on this page depends on the City's land values. The Office of Property Assessment usually values land by allocating a share of a property's total value to it, often around 20%, rather than valuing the land directly. The City Controller found in 2019 that about half of single-family land values were set at a fixed 15% or 30%.
So the rates are a range. Across plausible land shares, the 4 : 1 land rate falls between 2.7% and 4.1%, and the building rate between 0.7% and 1.0%. In every case the building rate is well below today's 1.4%.
For 90% of all properties with a building, and 95% of owner-occupied homes, the City's land value is exactly 20% of the property's total value. It is a formula, not a measurement. Land is valued on its own mainly when nothing stands on it.
That makes block-by-block comparisons revealing. On the 11,042 block faces that have both empty and built parcels, the empty lots' land is assessed per square foot at about the same level as their neighbors' in the median, but in more than a quarter of block faces one is valued at more than twice the other. On the 0 block of Laurel Street, assessed land ranges from about $10 to $195 a square foot; the Laurel Street lot is at $31.
Unlike Baltimore, where the Center for Land Economics found vacant lots valued at a fraction of their built neighbors, Philadelphia's empty lots are not systematically under-assessed per square foot. On mixed-use corridors they are often assessed higher than the built land around them, because the built land is set by the 20% formula. The problem is less that one kind of land is under-valued than that most land is never valued at all. Larger lots also normally sell for less per square foot, which this comparison does not adjust for. Data: same_block_land_rates.csv.
An owner holding a lot earns two things: whatever the parking brings in, and the land's rising value as the neighborhood around it improves. The yearly cost of waiting is mostly the property tax. When that cost is low and the neighborhood is hot, waiting can pay as well as building.
A land tax raises the cost of waiting every year, whether the land is used or not. It also lowers what land sells for, because buyers price in the higher tax; the buyer who builds keeps that difference as a lower cost of entry.
But the tax can only bite on the value it sees. The Laurel Street lot is assessed at about $31 per square foot; a lot a few blocks away sold for $185 per square foot. At today's assessment, a land tax changes the owner's math only a little. Assessed at market value, it changes it a lot. Accurate land values come first.
A building gets built when the rent it can charge covers what it costs to build, with a return for the risk. On the Laurel Street lot, a six-story building of 101 apartments needs more rent than the neighborhood pays under every tax policy tested. The tax policy matters. Construction cost matters more.
About the 20-year abatement: the 2026 proposal would extend a 20-year abatement to office-to-residential conversions, not to new buildings on empty lots, and had no bill as of late September 2026. It is shown here as a hypothetical, "if extended to new construction," because it is the policy most often offered as the alternative.
One waiting place in each council district, drawn from the City's records, with the story of its land. Pick a district, or tap a site on the map. District lines are the City's official 2024 boundaries.
Dots: every privately held vacant or parking lot of 4,000 sf or more in CMX-3 (orange) and CMX-4 or CMX-5 (gray). Dashed lines: council districts (City of Philadelphia, 2024).
An abatement is paid for by everyone else, including the School District, which receives about 56 cents of every property tax dollar. A land tax builds the incentive into the rates themselves, and collects from land that sits idle.
A revenue-neutral shift does not change what the city collects. It changes who pays. Using every parcel in the City's records, placed in its 2024 council district, here is how a 4 : 1 shift would move the bill, before any land prices adjust.
About a third of all assessed value in Philadelphia is exempt from property tax, but that figure mixes three different things. Fully exempt parcels, the land a split rate can never reach, hold about 13% of all value and 18% of all land value. In District 3, which includes University City, it is 30% of land value; in District 9, 6%.
A split rate makes that land matter more. Today the fully exempt parcels avoid about $445 million a year in tax. At 4 : 1, the forgone tax on their land more than doubles while the forgone tax on their buildings falls, so exemptions cost about $41 million a year more, paid through everyone else's rate.
Most exempt land is public: City land and parks, the School District, the Housing and Redevelopment Authorities and the Land Bank, state and federal land, transit and the port. That land should be counted honestly, but most of it serves the public and should stay untaxed. Some of it is waiting land too: the City, its authorities and the Land Bank hold about 3,100 acres of vacant and parking parcels. A land tax cannot reach it; public land policy has to.
Universities and hospitals hold about $1.2 billion of exempt land and $5.4 billion of exempt buildings. Taxed, they would pay about $93 million a year today. Boston asks its large institutions for 25% of their would-be tax and collected over $87 million in 2020; at that rate Philadelphia's would pay about $23 million. A payment based on land value alone fits the logic of this study: it asks for a share of the value that streets, transit and city services create, without penalizing institutions for building. At today's rate, their land alone would owe about $17 million a year.
A land tax has serious critics, and some of their points are right. Here are the strongest objections, and what this study finds about each.
True. Pittsburgh taxed land at nearly six times the building rate until 2001. A court-ordered reassessment, after years of frozen values, sent land values in many neighborhoods up by hundreds of percent, and owners of modest houses on newly valuable land faced sudden jumps. Council repealed the split rate within weeks.
What this study takes from it: keep assessments current, phase in any shift, and design the homestead protection before the first bill.
Partly true. Oates and Schwab found Pittsburgh's building boom of the 1980s far outpaced 15 comparison cities, but credited the land tax with a meaningful, not sole, role alongside a downtown development push. Banzhaf and Lavery found Pennsylvania split-rate towns added housing units faster than trend, about 4 to 5 points a decade, but did not claim the tax reduced sprawl. Most of the evidence comes from small Pennsylvania cities.
What this study does: it does not forecast construction. It shows the mechanics and leaves the size of the effect to the research.
Today, largely true for vacant land. This study compared assessments with 1,256 single-parcel sales of vacant and parking land since 2023. The median assessment matched the median sale, but values were widely scattered (a coefficient of dispersion of 50, against an industry standard of 20 to 25 for vacant land), and cheaper lots were assessed high relative to sales while costlier lots were assessed low. And for 90% of built properties, the City does not value land at all: it sets it at 20% of total value. A land tax on those values would magnify the errors.
What this study takes from it: a funded land-value audit comes before any rate change (Ask 2).
True today. At a Council hearing, the Revenue Commissioner testified that Philadelphia lacks the authority to enact a land value tax, and warned that a court loss could force refunds that would hurt the General Fund and the School District. Other Pennsylvania places split their rates under state statute. Whether a Philadelphia split rate would face a challenge under the state constitution's uniformity clause is a legal question this study does not resolve.
What this study takes from it: the path runs through Harrisburg (Ask 3).
It can. With the homestead exemption applied as it is today, average owner-occupied homes in Districts 3, 7 and 9 would pay 13% to 24% more. Long-time owners on land that has become valuable, and neighbors who keep a side lot, are the cases critics raise.
What this study finds: spreading the exemption across land and building reverses the result in every district (chapter 07). Phase-in and existing longtime-owner and senior programs add further protection.
Unlikely, though the transition needs care. Land gets cheaper because it costs more to hold each year, so a lower price is not a discount. It lowers the cash a small builder, land trust or first-time owner needs to get in, and it makes stockpiling idle parcels expensive, which is how land is held off the market. Ownership of waiting land is already fragmented: the 196 study lots have 174 different owners, only 4 own three or more, and citywide the 10 largest owners of large private vacant and parking parcels hold about 16% of their land value, and the largest by area are railroads, utilities and institutions rather than developers. Shell companies can hide common owners, so these figures are floors.
Guardrails: phase the shift in; give the Land Bank, land trusts and community developers a first look at public and tax-delinquent land; and report ownership concentration each year (Ask 4).
Mostly not, for the part that falls on land. Land does not go away or move when it is taxed, so a landlord who could charge more rent is usually charging it already; the tax shows up as a lower land price instead. Germany's municipal land taxes, for example, were found to capitalize into land prices with no effect on rents (Buettner, 2003). The evidence is not unanimous: two Danish studies disagree, and taxes on buildings do pass partly to renters: about 15% in Boston-area estimates, and fully within three years in one large German study.
What this study finds: because the building rate falls, rental buildings pay 4% to 8% less on average under a 4 : 1 shift (chapter 07). The shift lowers the tax that can be passed on.
Agreed. In this study's own model, a 15% change in construction cost moves the rent needed more than any tax policy does (chapter 04). A land tax is not a substitute for cheaper building, better codes or zoning that allows housing.
What this study argues: unlike an abatement, a land tax does not spend public revenue to close the gap, and it adds a cost to waiting.
Pennsylvania has the country's longest record of cities taxing land and buildings at different rates. Dozens of its towns have tried it over a century; some kept it and some, like Pittsburgh, dropped it. Philadelphia never has, and it would need the state's permission to start.
Pennsylvania enables split-rate taxes for Pittsburgh and Scranton. Pittsburgh phases one in.
Third-class cities are authorized. Harrisburg, near bankruptcy, adopts a split rate and later taxes land at four times the building rate.
After a countywide reassessment, Pittsburgh repeals its split rate within weeks.
Altoona completes a phase-in to a tax on land only, the first U.S. city to do so.
Philadelphia's mayor calls the state's uniformity clause a barrier to tax reform; a repeal amendment is introduced in Harrisburg.
Of 33 Pennsylvania places that have adopted or debated it, 16 still use a split rate and 5 have rescinded it.
Pennsylvania cities phased their split rates in a step at a time. 4 : 1 can be read as one stop on a longer road whose end, a tax on land alone, advocates call a universal building exemption. Each step moves more of the bill from buildings to land. How homeowners fare along the way depends on the exemption design: spread across land and building, they pay less at every step; left on the building, they pay more, most in Districts 3, 7 and 9.
Audit land values for vacant and commercial parcels, and test valuing land directly, so the base is sound before any rate moves.
Authority for Philadelphia and its School District to set separate land and building rates, with phase-in and homestead design written in.
Shift the ratio a step a year, spread the homestead exemption across land and building, and keep longtime-owner and senior programs.
Sources: Lincoln Institute · Lincoln Institute, Pennsylvania (Yang, 2016) · Legal Clarity on Pittsburgh's repeal · WHYY, April 2024 · WHYY on the Council hearing
This study does not ask Council to adopt a land tax. It asks for the steps that would let the City decide on good information.
On land assessment and split-rate options, with the Office of Property Assessment, Revenue, Law, the School District and the Controller. Use the district tables in chapter 07 as a starting point.
An independent ratio study of vacant, parking and commercial land to the industry standard, then a pilot of direct land valuation: one land rate per square foot for each block or small area, set from sales; a check that similar lots on the same block carry similar land values; public review of the land map before values are final; and revaluation every one to three years. Open tools such as OpenAVMKit, from the Center for Land Economics, already do this work in other cities.
A Council resolution asking the General Assembly to authorize separate land and building rates for the City and School District, with phase-in and homestead options, and to fund a state study of the effects.
Publish how each parcel's land value was set, an annual report of vacant and parking land by district, including who owns it and how concentrated ownership is, and the official boundaries and queries behind it, so studies like this one can be checked.
Revive payments in lieu of taxes for large universities and hospitals, as Philadelphia did from 1994 and Boston does today, but base the request on land value: a share of the value city services create, without a penalty for building.
For reviewers: the figures most worth challenging are construction cost, market land prices, the land share, and the strict definition of a waiting lot. Each is marked on this page and documented in the appendix.
For planners, staff and anyone who wants to test the numbers: the full model behind every chart above. Pick a site, change any assumption, try other ratios, and see how the developer, the landowner and the public fare under each policy.
Every number on this page can be traced, rerun and challenged. The queries, the code, the tests and the outputs are published as an open package. Open any section below for the detail.
The Waiting Lots: land, tax and time in Philadelphia. Pilot study, AIA Philadelphia Housing Committee. Draft for peer review, October 2026.
This appendix describes every data source, filter, formula and assumption behind the published page, so that any reader can check the work, rerun it and propose changes. Code and data are in the accompanying repository. Where a choice follows prior research, the source is named.
Version 2.2, October 29, 2026. Version 2.2 replaces the grid-based incidence with a parcel-level run on all 583,716 parcels and adds four analyses from the full parcel file: how the City sets land values, a same-block comparison, land under parking, exempt land by owner type, and ownership concentration (section 5.3). Version 2.1. Version 2 was revised after an internal skeptic review; version 2.1 adds the phase-in ladder, an independent cross-check with LVTShift, the renters' pass-through evidence, a named method for the land-value audit, district briefs, and prepared queries for three analyses that need the full parcel file. The changes are listed in section 15.
The study asks three questions:
The study is a set of transparent calculations, not an econometric forecast. Behavioral claims (that a split-rate tax increases construction) rest on the published research in section 9, not on new estimation here.
Parcels. Office of Property Assessment (OPA) records, table opa_properties_public, queried from the City's public Carto SQL API on 2026-10-28. The queries are in queries/ with their exact download URLs.
cmx_lots.csv: every parcel zoned CMX-1 through CMX-5 whose category is vacant land or whose building description is a parking lot, with lot area over 4,000 sf (787 parcels). Fields: parcel number, address, zoning, OPA category and building descriptions, lot area, market value, taxable and exempt land and building values, last recorded sale date and price, year built, and coordinates.citywide_totals.csv: citywide sums of taxable and exempt land and building value (583,718 parcels).
Citywide incidence: every parcel, aggregated by a 0.002-degree grid cell (about 170 by 220 m) and property class, with sums of taxable and exempt land and building value and counts by land share (queries/incidence_grid.sql); homestead parcels separately (queries/homestead_grid.sql); fully exempt parcels separately (queries/exempt_grid.sql).
queries/vacant_sales.sql).Council districts. The City's official 2024 boundaries, from the City of Philadelphia ArcGIS feature service Council_Districts_2024 (layer 0, field district_num), downloaded 2026-10-28 as GeoJSON simplified to about 40 m (geo/council_districts_2024.geojson). Every featured site, and every study parcel within 150 m of a district line, was then checked by a point query against the full-resolution service. No study parcel changed district from the earlier reconstruction (geo/council_districts_approx.geojson, rebuilt from 2019 election divisions), but one featured site did: the Logan Triangle lies in District 8, not District 9, and District 9's story was replaced (section 15).
Neighborhood outlines for the base map: Abaca Labs, CC BY 4.0, via Azavea.
Market inputs. Rents, retail rents, cap rates and land prices come from published listings and market reports cited in section 6. Construction costs are placeholders pending project bids.
From cmx_lots.csv, src/assign_lots.py keeps parcels that are:
CMX3);NON PD PKG LOT COMMERCIAL, PD PKG LOT COMMERCIAL, PKG LOT NON COMMERCIAL);Duplicate parcel numbers are removed. Each parcel is placed in a district by point-in-polygon on its OPA coordinates, against the official boundaries.
Result: 196 parcels, 203 acres, $695.8 million in assessed value, $9.74 million a year in tax at 1.3998%. Of these, 120 are CMX-3; 63 of those are "mid-sized" (8,000 to 60,000 sf), the population used for the apartment estimate.
Strict definition. OPA's classes are imperfect, so the study also reports a strict count: parcels with no taxable building value whose building description names a parking lot or vacant land. 159 of the 196 parcels pass (171.8 acres, $8.13 million a year in tax); 47 of the 63 mid-sized CMX-3 lots pass, enough land for about 4,200 apartments at six stories, against about 6,000 for all 63. The page reports both counts as a range. data/derived/verification_checklist_mid_cmx3.csv lists all 63 mid-sized lots with a City Atlas link, a Street View link and empty columns for a reviewer to record the use seen on the ground.
Known limitations of OPA records. OPA's "vacant" category includes some parcels whose building description is not vacant (for example a shopping center recorded as vacant land); these were kept because OPA classifies them as vacant, and a reviewer may reasonably drop them. Recorded sale prices under $100 are transfers between related parties, not market sales (nominal_sale = true). One recorded price can cover several parcels sold together, so sale-to-assessment comparisons for individual parcels are indicative only.
Philadelphia taxes assessed land and buildings at one rate, t = 1.3998% (School District 0.7839%, City 0.6159%; the School District receives 56% of collections). A split-rate tax sets a land rate rL and a building rate rB with rL = k·rB, where k is the ratio. Holding citywide revenue constant:
rB·(1 − s) + rL·s = t, so rB = t ÷ ((1 − s) + k·s)
where s is land's share of taxable value. From the October 2026 records, taxable land is $44.50 billion and taxable buildings $121.30 billion, so s = 26.84%. (The 2020 Tax Shift Explorer used 24%.)
| Ratio k | Building rate | Land rate |
|---|---|---|
| 2 : 1 | 1.104% | 2.207% |
| 4 : 1 | 0.775% | 3.102% |
| 6 : 1 | 0.598% | 3.586% |
| Land only | 0 | 5.215% |
This is the same revenue-neutral construction used in Pennsylvania's split-rate cities and in the Lincoln Institute's Detroit study (Anderson et al., 2022), which modeled land taxed at five times the building rate.
Long-run rates. A higher land tax lowers land prices (section 7.3). If assessments follow prices, the land base shrinks and rates must rise to stay revenue-neutral. With full capitalization at land yield y, each dollar of land value today becomes f = (y + t) ÷ (y + rL), so revenue neutrality requires rB·(1 − s) + rL·s·f(rL) = t, solved by iteration. This is an upper bound on the long-run adjustment, since buildings also rise in value and assessments lag. For a land-only tax there is a closed form, rL = t·y ÷ (s·(y + t) − t), which has no solution when s·(y + t) ≤ t: at a land share of 20%, the capitalized land base cannot carry the whole levy.
| Land share s | Land rate (static) | Building rate (static) | Land rate (long run) | Building rate (long run) |
|---|---|---|---|---|
| 20.0% (OPA's usual rule) | 3.50% | 0.87% | 4.11% | 1.03% |
| 24.0% (Tax Shift Explorer, 2020) | 3.26% | 0.81% | 3.85% | 0.96% |
| 26.84% (City records, Oct 2026) | 3.10% | 0.78% | 3.68% | 0.92% |
| 30.0% | 2.95% | 0.74% | 3.49% | 0.87% |
| 35.0% | 2.73% | 0.68% | 3.22% | 0.80% |
All at 4 : 1, y = 5%. In data/derived/rate_range.csv. The page uses the static rates at the City's recorded land share and presents the range as the uncertainty.
Caution on s. The land share is only as good as OPA's land values. OPA's tax year 2025 methodology states that it usually allocates 20% of a property's value to land and 80% to the structure. The City Controller's 2019 review found that about half of single-family land values were set at a fixed 15% or 30% of total value, and that land values on the same block varied by as much as $22 per square foot. The split rates above should therefore be treated as provisional until land is valued directly (section 8).
For each parcel, src/assign_lots.py computes tax today, (land + building) × t, and under each split rate, land × rL + building × rB, using current assessed values. For the 196 study parcels at 4 : 1, tax would rise from $9.74 million to $21.40 million a year. This is a static first-round calculation: it ignores the fall in land prices that a land tax would cause (section 7.3) and any change in use.
To see who would pay, every parcel in the City's records is classed as an owner-occupied home (any homestead exemption), other residential (single family, multifamily or apartments without a homestead exemption; mostly rentals), business (commercial, industrial, mixed use), vacant land, or other. Parcels are summed by grid cell and class on the City's server, because district polygons cannot be joined to 580,000 parcels within the API's time limit; each cell's parcel-weighted centroid is then placed in a district. Cells straddling a district line are assigned whole, so district totals near lines are approximate; 1,131 parcels (0.2%) in cells whose centroid falls just outside the simplified city edge are left out. Class totals reproduce the citywide taxable land ($44.39 billion of $44.50 billion) and building values.
For each district and class, tax today is t × (taxable land + taxable building) and at 4 : 1 it is rL × taxable land + rB × taxable building. The totals are revenue-neutral citywide ($2.31 billion either way). This is a static first-round calculation, before land prices adjust.
Two homestead designs. Philadelphia's $100,000 homestead exemption is recorded against the building value (OPA's exempt-building field), leaving the land fully taxable. Under a split rate that design taxes the land of a modest home at the higher rate. The study models:
Results for the average owner-occupied home (about $295,000 market value; $194,900 taxable after the exemption): $2,728 a year today, $2,800 at 4 : 1 as applied today (+3%), $2,493 with the exemption spread (−9%). By district, the as-applied-today change ranges from −5% (District 5) to +24% (District 7); the spread design lowers the average in every district, by 7% to 11%. About 181,000 of 242,500 homesteads would pay more under the first design, and fewer than 100 under the second. These counts depend on OPA's residential land values, which are often set by rule at about 15–20% of total value; if land were valued directly and proved a larger share in some neighborhoods, more homeowners there would pay more. Results are in data/derived/incidence_by_district.csv.
Parcel-level run (version 2.2). The full parcel file (queries/opa_parcels_for_lvtshift.sql, 583,716 parcels, downloaded 2026-10-29; its taxable and exempt totals match citywide_totals.csv to within two parcels) was assigned to the official 2024 districts by point-in-polygon (167 parcels just outside the simplified boundaries were assigned to the nearest district; 112 have no coordinates), and every taxable parcel was run through LVTShift's model_split_rate_tax at 4 : 1 under both homestead designs (src/parcel_prep.py, src/lvtshift_parcels.py; results in data/derived/lvtshift_parcels_by_district.csv). The rates match section 4 (3.1018% and 0.7754%; spread design 3.2447% and 0.8112%). Citywide, 78% of owner-occupied homes pay more with the exemption on the building (median +$190 a year, +9%), and none with it spread (median −$165, −7.3%). By district the share paying more under the first design runs from 54% (District 5) to 98% (Districts 7 and 9). Businesses' total tax rises 7.9%, but only 12.5% of business parcels pay more; the median business pays 11% less. 11% of rental parcels pay more; the median pays 11% less. District and class totals below now come from this run.
Independent cross-check. The same district and class sums (data/derived/incidence_inputs.csv) were run through LVTShift, the open-source split-rate model by Greg Miller of the Center for Land Economics (src/lvtshift_crosscheck.py). Its revenue-neutral solver reproduces this study's 4 : 1 rates (3.0975% and 0.7744% on these sums, which omit the 0.2% of parcels left unassigned) and every class total exactly (data/derived/lvtshift_crosscheck.csv). LVTShift applies exemptions to building value first, matching the first homestead design. Because the solve is linear, this confirms the arithmetic but not parcel-level medians; src/lvtshift_parcels.py runs LVTShift on every parcel, by district, under both homestead designs, once the parcel file from queries/opa_parcels_for_lvtshift.sql is downloaded.
The phase-in ladder. The page shows how each group's tax would change at 2 : 1, 4 : 1, 6 : 1 and land only (sometimes called a universal building exemption), under both homestead designs, from the same sums with rates set from the citywide totals (data/derived/phase_in_ladder.csv). With the exemption spread, owner-occupied homes pay less at every step: −4% at 2 : 1, −9% at 4 : 1, −11% at 6 : 1 and −21% at land only. With the exemption on the building, the average home pays 1% to 6% more and the hardest-hit district 11% to 53% more. Vacant land rises 57% at 2 : 1 and about 270% at land only.
Exempt land. Fully exempt parcels (taxable value zero, exempt value positive) hold about $9.85 billion of land, 18% of all land value. The share ranges from 6% in District 9 to 30% in District 3. No split rate reaches this land.
How land is valued. For 90.5% of parcels with a building, and 94.6% of owner-occupied homes, OPA's land value is exactly 20% of total market value; for commercial parcels, 52%. Land is valued separately mainly for vacant parcels. Because the homestead results depend on this rule, every homestead's change under the spread design is the same −7.3%.
Same-block comparison (src/same_block.py, data/derived/same_block_land_rates.csv). A block face is a street, hundred block and side. On the 11,042 block faces with at least one vacant or parking parcel and two built parcels, the ratio of the built parcels' median land value per square foot to the empty parcels' median has a median of 0.97 (10th to 90th percentile, 0.42 to 2.09); on 27% of faces one is more than twice the other. On faces with a CMX-zoned empty parcel the median ratio is 0.65: empty lots are assessed higher per square foot than built neighbors, whose land is set by the 20% rule. Unlike Baltimore, Philadelphia's empty lots are not systematically under-assessed relative to their neighbors; most land is simply not valued directly. The comparison does not adjust for lot size, which normally lowers value per square foot on large lots.
Land under parking and vacant land (src/exempt_and_parking.py, data/derived/parking_vacant_land_by_district.csv). Parcels whose building description is a parking lot: 4,478 parcels, 757 acres, $1.68 billion of land value ($1.05 billion taxable, $0.62 billion on exempt parcels). District 1 has $621 million. Vacant land (other than parking): 39,642 parcels, 9,719 acres, $6.32 billion, about half on exempt parcels. Parking that serves a building on the same parcel is not counted.
Exemptions (data/derived/exempt_land_by_owner_type.csv, exempt_land_top_owners.csv). Of $81.4 billion exempt (33% of $247.2 billion in assessed value): fully exempt parcels $31.8 billion (land $10.3 billion, 18% of all land value; 33,275 parcels); homestead exemptions $23.2 billion; other partial exemptions $26.4 billion, by subtraction, mostly abatements. Fully exempt parcels avoid about $445 million a year today and would avoid about $486 million at 4 : 1 (land at 3.10%, buildings at 0.78%). Grouped by owner name, which is approximate, exempt land is: City and city agencies 38%; other nonprofits 12%; universities and schools 11%; companies, mostly in state tax-incentive zones, 8%; state and federal government 8%; Housing Authority, Redevelopment Authority and Land Bank 7%; religious 4%; School District 4%; transit and port 3%; railroads and utilities 2%; hospitals 1%. Universities and hospitals hold $1.24 billion of exempt land and $5.39 billion of exempt buildings; fully taxed they would owe about $93 million a year, or $17 million on land alone at today's rate.
Ownership concentration (src/ownership_concentration.py, data/derived/ownership_concentration.json). Owner names are normalized (upper case, punctuation removed) but not linked across companies, so shell companies with common owners are counted separately and concentration is understated. The 196 study lots have 174 owners; 4 own three or more (3% of the area); the 10 largest hold 47% of the area and 33% of the land value; Herfindahl index 400 by area. Privately held (taxable) vacant and parking parcels of 4,000 sf or more citywide: 3,294 parcels, 2,768 owners, 3,088 acres; the 10 largest owners hold 32% of the area and 16% of the land value (Herfindahl 244); the largest by area are railroads, utilities and institutions.
The ten one-page district briefs (brief/d1.html to brief/d10.html) report the spread-exemption design only, the design this study recommends, from a parcel-level LVTShift run (src/district_brief_numbers.py, data/derived/district_brief_numbers.csv). For the briefs, vacant land and parking-lot parcels are grouped together as "vacant and parking lots", so business parcels exclude parking lots; "small business properties" are business parcels with a market value under $1 million; "typical" is the median change. Each brief also states the result under today's exemption design, and that the estimates depend on the City's land values, state authority and a phase-in. Citywide under the spread design, 88% of taxable parcels, every owner-occupied home, 91% of small business properties and 88% of rental parcels pay less.
A mid-sized building allowed by right in CMX-3: a ground floor of retail and lobby covering 75% of the lot (the CMX-3 maximum occupied area for an interior lot), and five residential floors above. That is a floor area ratio of 4.5, under the 500% base allowed in CMX-3, so no mixed-income bonus payment is needed. A building of this height is typically built as wood frame over a concrete podium.
| Parameter | Value | Basis |
|---|---|---|
| Upper-floor efficiency | 82% | Typical double-loaded corridor; placeholder |
| Average apartment | 665 sf | 80% one-bedrooms at 650 sf, 10% studios, 10% two-bedrooms |
| Retail share of ground floor | 60% | Remainder lobby and services |
| Construction cost | $300 per gross sf | Placeholder, hard and soft, excluding land; to be replaced with bids |
| Residential vacancy | 5% | Metro vacancy 4.6% in Q2 2026 (Northmarq) |
| Operating costs | 30% of income | Before property tax; placeholder |
| Retail rent | $18–$28 per sf per year, net | Fishtown average about $27 (TenantBase); metro in-line $19.98 NNN (CBRE Q2 2026) |
| Retail vacancy | 10% | Metro retail availability 8.1% (CBRE Q2 2026) |
| Market rent | $2.20–$3.20 per rentable sf per month | Fishtown 23 West one-bedroom $1,895 for 646 sf (Inquirer, 2025); Rittenhouse one-bedroom average $2,329 (RentDataNow, Mar 2026); Class A metro asking $2,214 (Northmarq Q2 2026) |
| Exit cap rate | 5.5% | Philadelphia deals averaged mid-5% (Northmarq Q2 2026); 6.08% (Matthews Q2 2026) |
| Developer's required return | 10% profit on total cost | Common feasibility threshold; tested at 15% |
| Developer's discount rate | 8% | For valuing abatement savings |
| City's discount rate | 4% | Roughly a municipal borrowing rate |
| Construction period | 2 years | Land tax only during construction |
| Land assessment | The City's recorded assessment | Laurel St: $678,600 (about $31/sf). Tested with land reassessed at market price |
| Market land price | Placeholder, $125/sf for Laurel St | A 25,907 sf lot at 943 N 2nd St sold for $185/sf in 2017; tested at $60 and $185 |
| Parking income | Placeholder, $80,000 a year for Laurel St | Tested in the workbench |
| New building assessment | 100% of construction cost | OPA assesses new construction at market value; tested at 75% |
On the Laurel Street lot (21,858 sf, modeled as 22,000 sf) this gives 99,000 gross sf, 101 apartments and about 9,900 sf of retail.
All equations are implemented in src/model.js and tested in tests/test_model.js.
Stabilized net operating income before property tax:
NOI₀ = units × rent × 12 × (1 − vacancy) × (1 − operating ratio) + retail sf × retail rent × (1 − retail vacancy)
For each policy, NOI after full tax is NOI₀ − land tax − building tax. Value at stabilization is NOI ÷ cap rate, plus, for an abatement, the present value at the developer's discount rate of the building tax forgiven over its schedule. This follows the Controller's 2018 abatement analysis in treating the abatement as a capitalized benefit to the developer.
Return on cost = (value − land price − development cost) ÷ (land price + development cost). The rent needed is the rent per rentable square foot at which the return equals the required 10%, solved in closed form:
NOI needed = cap rate × ((1 + required return) × total cost − abatement value)
rent needed = (NOI needed + land tax + building tax − retail NOI) ÷ ((1 − operating ratio) × (1 − vacancy) × 12 × rentable sf)
The test suite confirms that plugging the rent needed back into the model returns exactly the required return.
A higher tax on land is capitalized into a lower land price, the standard result in the property tax literature and the reason a land tax is thought to have little deadweight loss. The model applies a one-time adjustment:
The model separates the land's market price M (what a buyer pays) from its assessment V (what the tax sees). In equilibrium today, M = (R − t·V) ÷ y, where R is the land's yearly rent and y the land yield, so R = y·M + t·V. Under the land tax:
On the Laurel Street lot, the recorded assessment is about a quarter of the placeholder market price. At that assessment a 4 : 1 land tax lowers the price about 8% and raises the yearly tax on the empty lot from $9,499 to $21,048; the owner's return from waiting falls only from 7.6% to 7.3% a year. Reassessed at market, the tax is $38,495 today and $67,379 under the land tax, the price falls about 21%, and the return from waiting falls from 6.5% to 5.6%. Accurate land values matter more to the waiting decision than the rate ratio does.
The land yield (the pre-tax return buyers require on land) of 5%, tested from 4% to 7%. Yang and Hawley (2021) found that Pennsylvania's split rates lowered land values but not drastically, consistent with this adjustment.
The 20-year abatement proposed in 2026 would apply to office-to-residential conversions, and had no bill as of late September 2026. It is modeled here only as a hypothetical extension to new construction on vacant land, because it is the policy most often offered as the alternative to a land tax. A conversion is a different pro forma (existing structure, different costs and assessment) that this study does not model.
Land is taxed in full under every abatement.
Tax from the site for 30 years under each policy, in constant dollars, discounted at 4%. The School District share is 56%. Two holding cases are included: the lot stays parking under today's tax, and under the land tax.
The landowner's yearly return from waiting = (parking income − property tax) ÷ market price + expected appreciation, with the tax computed on the assessment. It is compared with the developer's 8% discount rate as a benchmark for the return from building.
Following the IAAO Standard on Ratio Studies, each sale's ratio is current assessment (market value) ÷ sale price. The sample is every parcel classed as vacant land, or with a parking-lot building description, whose last recorded sale was on or after 2023-01-01 at $10,000 or more, excluding any sale whose date and price appear on more than one parcel (a likely multi-parcel sale): 1,256 sales. Outliers beyond 1.5 times the interquartile range were trimmed before computing the coefficient of dispersion (COD) and price-related differential (PRD).
| Group | Sales | Median ratio | COD | PRD |
|---|---|---|---|---|
| All | 1,256 | 0.99 | 50 | 2.24 |
| Zoned CMX | 170 | 0.98 | 58 | 1.59 |
| Zoned residential | 1,028 | 1.00 | 48 | 1.81 |
| Zoned industrial | 46 | 0.79 | 62 | 3.16 |
| Lots of 4,000 sf or more | 123 | 0.83 | 68 | 2.32 |
| CMX lots of 4,000 sf or more | 26 | 0.63 | 77 | 1.31 |
| Parking lots | 55 | 0.69 | 59 | 4.46 |
The IAAO standard calls for a median ratio of 0.90 to 1.10, a COD of 5 to 25 for vacant land depending on market, and a PRD of 0.98 to 1.03. The median is on target, but dispersion is two to three times the standard, and a PRD well above 1 means cheaper lots are assessed high relative to their prices and costlier lots low. Large CMX lots and parking lots are assessed at well under their sale prices.
Limits. The OPA table holds only the last recorded sale, so the sample is small and not verified as arm's length. Assessments are current while sales span almost four years of a rising market, which biases ratios down for older sales. Some sales prices include approvals or plans. This is a screening study; the City's own ratio study for vacant and commercial land, with validated sales, is the recommended next step. Results in data/derived/sales_ratio_summary.csv; query in queries/vacant_sales.sql.
Built-in checks (tests/test_model.js, 11 tests). Split rates are revenue-neutral for every ratio and land share, static and long run; a 1 : 1 ratio reproduces today's tax; abatement schedules match their definitions; the rent-needed solution returns the target return; the land price matches its closed form under both assessment bases; today's tax on the Laurel Street lot matches the recorded bill; the prototype envelope gives 6 stories, 99,000 sf and 101 apartments.
Sensitivity (src/sensitivity.js, data/derived/sensitivity.csv). Each main assumption was varied one at a time. At the baseline, the rent needed per apartment is $2,654 with no abatement, $2,537 with the 10-year abatement, $2,473 with a hypothetical 20-year phased abatement and $2,421 with a 4 : 1 land tax, against a placeholder market rent of $1,995. A ±15% change in construction cost moves the rent needed by about ±$410 a month, more than any tax policy. With long-run rates, the land tax needs the same rent as the 20-year phased abatement ($2,473). In every case tested but one (a 35% land share, where the two are within 2%) it leaves the public more revenue from the site over 30 years than the hypothetical 20-year phased abatement, though less than the current 10-year abatement and less than an unabated building, because buildings are taxed at a lower rate permanently; a revenue-neutral land tax recovers that difference from other parcels, while an abatement's cost is forgone.
Recommended next steps.
src/same_block.py compares, on every block face, the median land value per square foot of vacant and parking parcels with that of built neighbors. Center for Land Economics studies found built lots carrying several times the land value of comparable empty ones in Spokane, Baltimore and Cincinnati.src/exempt_and_parking.py totals the land value under every surface parking lot and vacant parcel, by district, for a citywide map.src/lvtshift_parcels.py. Next: link owners across shell companies using mailing addresses and state corporate filings, to measure concentration properly.verification_checklist_mid_cmx3.csv.It does not estimate how much new construction a land tax would cause citywide; it relies on the research above for that. It models one homestead design choice but not phase-in schedules, senior or longtime-owner programs, or payments in lieu of taxes. It treats each site in isolation and holds rents and costs constant. Its land values are OPA's, which sections 4 and 8 show are partly rule-based and widely dispersed for vacant land. It does not resolve the legal questions: whether the General Assembly would authorize a split rate for a city of the first class and its School District, and whether such a law would face a challenge under the state constitution's uniformity clause.
The notes beneath each site's maps and aerial photographs were read from the images by the study team. Labels on the 1942 and 1962 land use maps are small and sometimes illegible at site scale, so the notes use cautious wording ("appears to") and omit uncertain uses. Corrections are welcome.
Requirements: Python 3.10+ with shapely, pyproj and pyshp; Node.js 18+.
git clone --depth 1 https://github.com/azavea/geo-data vendor/geo-data
# geo/council_districts_2024.geojson is downloaded from the City's ArcGIS service (queries/README.md)
python3 src/build_districts.py # (optional) the earlier approximate reconstruction
python3 src/assign_lots.py # data/derived/*.csv
node src/sensitivity.js # data/derived/sensitivity.csv
node tests/test_model.js # model checks
To refresh the data, open the URLs in queries/README.md and save the CSVs to data/raw/.
Construction cost ($300 per gross sf), market land prices for every site except District 4 (whose 2026 sale price is used), parking income, and market rents for the district prototypes are placeholders. They are marked on the page wherever a figure depends on them.
Version 2.2: parcel-level incidence; voter-facing district briefs (section 5.4); land-value rule, same-block, parking, exemption and ownership analyses (section 5.3); District 9 historic maps and aerials; an optional fifth ask on land-based payments from large institutions; an ownership card among the objections.
Version 2.1: a phase-in ladder to a land-only tax (section 5.2); the LVTShift cross-check (section 5.2); the renters' pass-through objection on the page and in section 10; a named valuation method in Ask 2 (section 9); one-page district briefs (brief/d1.html to brief/d10.html on the site); and prepared queries for the same-block check, the land value under parking, and a parcel-level LVTShift run (section 9).
Council_Districts_2024), queried October 2026. Every featured site and every study parcel within 150 m of a district line was checked against the full-resolution boundaries.