What have migration patterns of high‑income households been to and from NYC since 2019 and how have they affected city tax revenue?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
Since 2019 New York City and State tax-data and IRS migration series show a pandemic-era spike in high‑income households leaving—peaking in 2020—and then a reversion toward pre‑pandemic migration rates by 2023, with the share of million‑dollar tax returns changing addresses rising above 6% in 2020 and falling to just under 3% more recently [1] [2]. Analysts agree that outflows of high earners matter because those households pay a disproportionate share of PIT and city revenues, but estimates differ on how large and persistent the revenue hit has been [3] [4] [5].
1. What happened: the migration pattern in numbers
Matched tax‑return and IRS SOI data reveal that part‑year filers and out‑of‑state address updates drove most of the observable movement: more part‑year taxpayers moved out than moved in across income thresholds, and pre‑pandemic net annual outflows of part‑year filers averaged roughly 28,700 filers (2015–2019) with especially large outflows at higher income levels in 2019; these flows grew during the pandemic and then normalized by 2023 according to NYC Comptroller analyses [6] [2] [7].
2. Who moved and where they went
High‑income households—measured by New York State AGI of $1 million or more—experienced a higher rate of address changes than average filers, and the pandemic period concentrated movement to lower‑tax Sun Belt and Florida destinations, a pattern corroborated across IRS and county‑level studies showing affluent in‑migrants to growing counties and strong flows toward Florida [1] [8] [9].
3. How migration affected city and state tax revenue
High earners pay a disproportionate share of PIT revenue, so their departures lower the tax base; New York’s comptroller and policy groups explicitly note that out‑migration of wealthy households reduces revenues and fiscal capacity, and statewide losses of net AGI have been large over the decade according to cross‑state trackers [3] [4] [10]. At the same time, multiple analyses find the behavioral effect of tax differentials on migration is measurable but often quantitatively modest, with much of the observed fiscal shock in 2020 driven by temporary pandemic dislocations rather than only tax policy [5] [2].
4. Why people left (and why some returned or stayed)
Researchers identify a mix of drivers: pandemic‑era work‑from‑home enabled relocations, quality‑of‑life and affordability concerns, and tax differentials—especially attractions to Florida—played roles; studies and the NYC Comptroller stress that tax changes like the federal SALT cap and state PIT increases can influence behavior but likely only partially explain migration relative to the broad pandemic shock and housing/remote‑work dynamics [8] [5] [2].
5. The debate over magnitude and permanence
There is a sharp, documented debate: some institutions and media point to large, sustained losses of high‑income AGI and consequent fiscal strain, citing cumulative net AGI shifts in the tens of billions [10] [3], while careful state and city tax‑data analyses show migration rates for millionaires have come back toward pre‑pandemic norms and that tax‑policy‑driven migration effects are generally small and concentrated in flows to Florida, implying the long‑term revenue impact may be more modest than headline claims [1] [5] [2].
6. What remains uncertain and why it matters for policy
Available sources provide strong evidence of a pandemic spike and of high earners’ above‑average mobility, but they also disclose limits: IRS SOI and state filings measure address changes and AGI but cannot fully capture second‑home arrangements, tax‑avoidance strategies, or the counterfactual economic growth that would have occurred absent the pandemic—so policymakers must weigh the documented short‑term revenue declines and disproportionate contribution of top filers against research showing tax differentials alone produce modest long‑run migration [9] [6] [5].