Tenant Turnover and HPD Violations: What the NYC Data Shows

It's intuitive that a building with a high rate of HPD violations would also see more tenant turnover — nobody renews a lease happily in an apartment with recurring heat outages. But intuitive isn't the same as demonstrated, and this piece is upfront about which side of that line it's on: this is a mechanism and a research plan, not a result.

Part 1: The Plausible Mechanism

There are a few distinct channels by which HPD violation density could plausibly drive higher turnover, and they're worth separating because each has a different evidentiary signature. First, habitability issues directly reduce tenant satisfaction and are a documented reason tenants give for not renewing — a unit with recurring problems is less pleasant to live in regardless of any public record of that fact. Second, repeated emergency conditions (Class C violations in particular — heat and hot water outages, for example) can be severe enough to justify a tenant breaking a lease or declining renewal outright, independent of any broader pattern at the building. Third, public visibility effects — a building's appearance on a list like the Public Advocate's Worst Landlord Watchlist, or simply a tenant's own awareness of the building's HPD history via the city's public violations lookup — could make re-leasing harder even for units that aren't themselves affected, since prospective tenants increasingly check a building's public compliance record before signing.

Part 2: The Data That Would Actually Test This

Testing this mechanism rather than just asserting it requires cross-referencing at least two data sources that don't currently talk to each other in any published analysis DERALO.AI Research is aware of: HPD's violations dataset (NYC Open Data, Socrata dataset csn4-vhvf), which provides building-level violation history including class, date, and correction status over time; and a turnover-adjacent proxy data source, since NYC does not publish a direct building-level turnover metric. The most plausible proxy candidates are rent-stabilization registration filings (which can indicate unit-level tenancy changes over time for stabilized units) and, with appropriate limitations, other registration-adjacent public records. A real analysis would need to define a turnover proxy carefully, match it to buildings by BIN or BBL, control for building size and rent level (since both plausibly affect turnover independent of violations), and test whether violation density predicts the turnover proxy after those controls — not just whether the two move together in a raw correlation, which could easily reflect confounding factors like building age or neighborhood rather than a direct causal link.

If that analysis were run, the output would look like a correlation coefficient or regression result describing the relationship between violation density (likely normalized per-unit, matching the Worst Landlord Watchlist's own approach) and the turnover proxy, across a defined citywide sample, with appropriate controls. None of that exists yet in this article, because the analysis hasn't been run.

Part 3: What Landlords Report Anecdotally

Independent of any formal data analysis, it's a common industry observation — reported informally by property managers and landlords across conversations, not attributed to any specific building or owner here — that units with a recent history of unresolved maintenance complaints are harder to re-lease at the same rent, and that buildings with a visible public compliance problem see more turnover inquiries turn into non-renewals during the same lease-renewal cycle. This is pattern-level anecdote, not data, and it's presented as such — it's the kind of observation that motivates running the actual analysis described in Part 2, not a substitute for it.