The Vests That Follow You

Equity compensation is treated as compensation income when it vests, and the portion of it that is United States source is determined by the work performed during the period between grant and vesting. Sources describing the federal position for a non-resident put it in terms of a count: income is United States source based on the number of workdays spent in the United States during the vesting period. That is a fundamentally different test from the one governing most things on this site. Residency does not decide it. Where you were sitting on the vest date does not decide it. What decides it is a historical fact about where the work was actually done over a period that may have started years earlier and may finish years after you leave.

The consequence of the previous figure is the thing people do not plan for. Unvested awards held at the point of departure do not stop being connected to the United States because their holder has left — each future vest carries its own allocation, computed over its own grant-to-vest window, and a window that included American workdays produces United States source income however far away the person is by then. On a four-year vesting schedule that can mean a tail running for years after the final payslip. The instinct that the relationship ended with the last day of employment is exactly wrong: the employment ended, and the tax consequence of the award did not.

This is the part that surprises people who have already satisfied themselves about the federal position. States apply their own allocation, and sources describe formulas based on workdays in the state against total workdays, with returns filed as a non-resident. California is described as taxing stock-option and unit income by reference to the services performed in the state rather than to where somebody was resident when the award was exercised or sold, and New York is named as operating a comparable workday allocation. So a former resident can be filing a non-resident state return years after moving away, on income from an award that vested in a country the state has nothing to do with. The intuition that leaving a state ends its interest is the intuition this figure exists to remove.

The two allocations sit on different statutes, are administered by different authorities, and can produce different numbers from the same facts. That matters practically rather than academically: somebody who has taken advice on the federal position, or whose employer has handled federal withholding, can reasonably believe the question is closed and be wrong about a state that is still owed a return. The reverse also holds. Treat them as two questions with two answers, and when asking anybody for help, ask about both explicitly rather than assuming that a conversation about tax covered the one you did not name.

Sources describing the position for a non-resident are explicit that only the portion tied to United States workdays is taxed there, and that the holder's home country may tax the rest. Two things follow. The arithmetic does not produce a gap in which some of the income escapes: it produces a split, and both halves have a claimant. And the two claims are computed under different systems that need not agree about periods, valuation dates or characterisation, so an overlap is entirely possible. Whether relief is available for that overlap depends on the treaty position between the two countries and on how each treats the award, neither of which this page establishes. It is the specific question to put to somebody who works across both systems rather than to one adviser in one of them.

Federal allocation runs on United States workdays during the vesting period. State allocation runs on in-state workdays against total workdays. Both are arithmetic performed on a historical record of where somebody actually was and worked, potentially across several years and several awards. Almost nobody keeps that record, because at the time there is no reason to — and reconstructing it years afterwards, from another country, against calendars and travel that may no longer be accessible, is the part of this that turns a manageable filing into an expensive one. If a departure is coming, the single most valuable thing to assemble before leaving is the boring one: grant dates, vesting schedules, and a defensible record of working days and location across the periods those awards span.

Where post-departure vests carry United States source income, the compliance consequence is a filing obligation in the years those vests occur rather than only in the year of departure. Sources describe the state limb of this in the same terms, with a non-resident state return as the mechanism. That is worth anticipating for two reasons beyond the filing itself. It requires keeping access to documents, statements and a preparer in a country somebody has left, which becomes progressively harder as time passes. And it is the kind of obligation that is easy to be genuinely unaware of, since nothing arrives to announce it and the employment relationship that would have prompted a question ended years earlier.

Nothing on this page is a computation of what any award would attract. The inputs — grant and vest dates, the workday record, the plan's own terms, residency and non-resident status for two different systems, and the treatment in the destination country — are facts this page cannot see, and several of them are technical determinations rather than things somebody can read off. It reproduces no form, threshold or deadline, and names no employer, broker, plan administrator or advisory firm. Worth stating plainly: most of the accessible writing on this subject is published by equity-specialist and cross-border tax practices whose product is advice on exactly this. That is not a reason to discount them — they are the only people writing about it in operational detail — but it is a reason to notice that the field tilts toward complexity, and to weigh any recommendation against how the person making it is paid.