Miller & Hoskins — College-Town Land Value Tax Shift Simulations (South Bend, IN and Princeton, NJ)

Summary

This entry covers two linked, revenue-neutral parcel-level tax-shift simulations, jointly announced on the Progress and Poverty Institute (PPI) blog: "New Reports: Land Value Taxes in College Towns." Greg Miller, president of the Center for Land Economics (CLE), led Toward Smarter Taxation (dated April 25, 2025), a South Bend, Indiana study with Notre Dame's undergraduate Student Policy Network (SPN), modeling land taxed at 4x the rate of buildings. Stephen Hoskins, a PPI researcher, authored A Land Value Tax Shift for Princeton, NJ, splitting Princeton's 2024 municipal rate of $0.578/$100 of total value into $0.860/$100 of land value plus $0.291/$100 of improvement value. Both PDFs were fetched and read in full (South Bend: 60 pages; Princeton: 10 pages). Dating resolved: both reports are 2025 publications, not 2026-cycle work. The South Bend report's own title page states "April 25th, 2025." The Princeton report carries no in-text date, but the Internet Archive Wayback Machine's earliest capture of the original (pre-"-2") Princeton PDF URL is dated 2025-05-09, confirming it was already live by then; the joint PPI blog announcement (article:published_time meta tag: 2025-05-16, cross-posted to Substack 2025-05-09) frames both reports as published "in the past month" — consistent with an April 2025 origin for Princeton alongside South Bend. The currently-linked Princeton file (the "-2.pdf" URL, fetched after the original 404'd) carries Canva export metadata dated 2026-01-27 and an internal filename of "NEW Updated Land Value Tax Shift for Princeton NJ [RGM].pdf" (Producer: Canva; Author field: Daniel Mejia) — evidence this is a later re-export/reupload of the report (the original URL's last successful Wayback capture was 2025-09-12), not evidence the report itself originates in 2026.

Both organizations are explicitly Georgist advocacy bodies: PPI's lineage runs through the Schalkenbach Foundation ecosystem, and CLE's stated mission is promoting land value taxation. Neither report is peer-reviewed; SPN is an undergraduate club whose report was prepared for CLE and South Bend's city planning office as a client deliverable.

The Core Argument / Findings

South Bend (4:1 land-to-improvement ratio, $16.7M revenue held constant via a hypothetical circuit-breaker-exempt referendum levy). Vacant land's median tax bill rises 127.8% and parking lots' 96.6% (a separate per-square-foot table gives smaller figures using a different denominator: +110.92%/+34.93%). Large multi-family (20+ units) properties see the biggest median cut (-15.5%). Among owner-occupied homesteads, 52.53% see an increase versus 47.47% a decrease, averaging +4.10%; among non-homestead (mostly rental) single-family parcels, 84.25% see a decrease, averaging -11.31%. Tax change correlates positively with neighborhood income and negatively with percent-minority population. The report's own caveat: 92% of South Bend parcels already sit at Indiana's circuit-breaker caps, so an ordinary split-rate conversion would mostly be absorbed by existing refunds — the reported effects apply only to the report's alternative referendum-levy scenario.

Princeton (land taxed roughly 3x the improvement rate, 7,950 taxable parcels analyzed). Apartments' average municipal bill falls from just under $16,000 to about $12,740 (-20%); commercial properties fall from $12,690 to $11,980 (-6%); vacant land rises from $1,150 to $1,710 (+50%). Residential parcels see a mixed +1% average change (+$40, to $4,940): of 7,135 residential parcels, 2,411 see their bill fall (typically ~10%) and 4,724 see it rise (typically 10-25%). Hoskins's stated rule of thumb: "homeowners whose house is worth more than half of their total property value will enjoy a tax cut, whereas those for whom land is the majority of property value will see their taxes rise." One named example: a permitted-but-unbuilt redevelopment site (12% building share) rises 37%, to $15,288 — the largest single-property dollar increase in the dataset.

Miller's shared framing for both cities is "LVT is a Density Discount": the studies are designed to show land taxation shifts burden off built, occupied, rented property and onto vacant, under-built land.

Relation to the Georgist Case

These are advocacy-institute simulation studies, not independent or peer-reviewed research — the same genre as Bowman & Bell's Virginia parcel studies: a static, revenue-neutral, parcel-level re-allocation of an existing tax base under a simulated rate schedule. Unlike Bowman & Bell, a National Tax Journal replication designed to test whether a prior finding generalized, Miller and Hoskins work for the advocacy organizations commissioning the reports and wrote them to demonstrate that a shift benefits development and (in South Bend) lower-income areas. They are cited here for the shift arithmetic and as a record of the advocates' own analysis and framing ("LVT is a Density Discount," Miller's phrase) — not as independent confirmation that an LVT shift causes development, migration, or welfare effects. Both are directionally consistent with the Pennsylvania split-rate incentive-effect literature (see Split-rate taxation increases urban construction), but neither models any actual behavioral response; both are pure current-year tax-bill reallocation exercises.

Nuances and Limits

• Static incidence, not behavioral prediction. Neither report models how landowners might redevelop, sell, or pass costs to tenants — the same limit the wiki flags for Bowman & Bell and England & Zhao.
• Advocacy provenance. Neither report underwent peer review or independent government review; the shared announcement states its purpose is to show "land value taxes are a good policy for these cities" — a conclusion held before, not derived neutrally from, the modeling.
• South Bend's circuit-breaker caveat limits real-world applicability, since the modeled effects depend on an alternative referendum-levy mechanism rather than an ordinary tax-code change.
• University tax-exemption angle is thin in both reports. Neither quantifies its host university's own exempt-land footprint. Resolved for South Bend: core Notre Dame campus buildings are excluded — the report's Methodology section states "religious and educational buildings...are fully exempt from property taxes, so their exemptions are the total value of the property," and its modeled split-rate scenario explicitly "still uses the typical exemptions that would still apply under a referendum," so fully-exempt religious/educational buildings stay untaxed under the simulated shift too, not just under current law. Not resolved for Princeton: the report states 7,950 of 8,836 total parcels are "taxable (non-exempt)" — implying roughly 886 parcels are exempt — and its Appendix 1 map of "total property values across Princeton" shows "the large parcels owned by Princeton University clearly visible," but the text never states whether those university parcels fall inside the 7,950 taxable set used for the tax-bill simulation or the ~886 exempt set. Checked: full text searched for "exempt," "Princeton University," and "non-profit" — no passage answers this directly. Given Princeton University's ordinary property-tax-exempt status as a nonprofit educational institution under New Jersey law, its core campus buildings are more likely among the exempt parcels excluded from the simulation (paralleling South Bend), but this is an inference, not something the report confirms.
• Small, self-selected base: two college-town case studies from the same two-author collaboration, announced together; no claim to generalizability is made.

Bears On

• Benefit: A land value tax can be progressive — the South Bend study's income-and-minority regression is directionally supportive, but as an advocacy-commissioned, non-peer-reviewed single-city simulation it belongs alongside, not in place of, the peer-reviewed Bowman & Bell/England & Zhao evidence already on that page.
• Objection: LVT hurts the "asset-rich, cash-poor" — the Princeton and South Bend homestead figures (4,724 of 7,135 Princeton residential parcels rising; 52.53% of South Bend homesteads rising) are relevant raw material for this objection's steelman: a meaningful minority-to-plurality of owner-occupied homes pay more under both modeled shifts, even where the citywide average is small or negative.
• Concept: Split-Rate Taxation — both reports are worked numerical examples of a 4:1 (South Bend) and roughly 3:1 (Princeton) split-rate design applied to real assessment rolls.

See Also

• Split-Rate Taxation
• A land value tax can be progressive
• Bowman & Bell — Distributional Consequences of Converting the Property Tax to a Land Value Tax
• Split-rate taxation increases urban construction
• Objection: LVT hurts the "asset-rich, cash-poor"
• Notre Dame SPN's Cincinnati LVT Analysis (2026) — the same SPN/CLE partnership's third parcel simulation, this time for a full city rather than a college town

Sources

1. Greg Miller and Stephen Hoskins (2025), "New Reports: Land Value Taxes in College Towns," Progress and Poverty Institute. Article — used for the joint announcement framing ("LVT is a Density Discount"), the summary figures for both cities, and confirmation of Miller's Center for Land Economics affiliation and Hoskins's Progress and Poverty Institute affiliation; fetched and read in full this session (page metadata dates the post 2025-05-16).
2. Student Policy Network (University of Notre Dame), prepared for the Center for Land Economics and the City of South Bend, Indiana (2025), "Toward Smarter Taxation: An Analysis of a Split-Rate Property Tax Structure in South Bend, IN." PDF, landing page at landeconomics.org/southbend — used for all South Bend methodology, tables, and figures (Fig. 7, Fig. 8, Fig. 9, Table 3, Table 4, Table 6, the circuit-breaker caveat, and the Conclusion); fetched and read in full (60 pages) this session.
3. Stephen Hoskins, Progress and Poverty Institute (2025; report re-uploaded as a "-2.pdf" file with 2026-01-27 Canva export metadata), "A Land Value Tax Shift for Princeton, NJ." PDF — used for all Princeton methodology, Table 1, Table 2, and the three named example properties; fetched and read in full (10 pages) this session. The URL given in the task and in the original blog post (.../A-Land-Value-Tax-Shift-for-Princeton-NJ.pdf, no "-2") returned HTTP 404; this "-2.pdf" URL, located via web search, returns the same report. The Wayback Machine's earliest capture of the original (non-"-2") URL is dated 2025-05-09, confirming a 2025 origin despite the 2026-dated file metadata on the current reupload (see Summary for the full dating analysis).
4. Bowman, John H. & Bell, Michael E. (2008), "Distributional Consequences of Converting the Property Tax to a Land Value Tax," National Tax Journal 61(4). Wiki summary — used as the comparison genre for static, revenue-neutral parcel-shift simulations and for the "jurisdiction-dependent, not proof" framing applied to this entry's findings.
5. David Roberts (Volts podcast, June 2026), "Want less sprawl and more urban infill? Try a land value tax!" — interview with Greg Miller (Center for Land Economics) and Spokane City Council member Kitty Klitzke. volts.wtf — further listening: the lead author presenting the same LVT case to a general audience (podcast, navigation-tier; not used for any claim).