San Francisco No-fault Terminations and Relocation: Where Specialist Review Starts

The instinct to treat "no-fault termination" as a single category is understandable and wrong. San Francisco's Rent Ordinance, layered on top of state law, actually recognizes several separate legal mechanisms for a landlord to recover a unit or its rental income without any tenant fault, and they don't substitute for one another. Choosing the wrong one, or assuming a step from one process applies to another, is where landlords most often end up in a dispute they didn't need to have.

Owner move-in and Ellis Act withdrawal are the two paths that actually end a tenancy over the tenant's objection. Capital improvement work doesn't end anything; the tenant stays, and the landlord recovers renovation cost through a separate rent mechanism. A buyout is different again: it's not a legal termination at all, it's a negotiated agreement the tenant has to voluntarily accept, with its own disclosure and rescission rules precisely because it isn't backed by a notice period or a Rent Board order.

Confusing these categories tends to happen in a specific way: a landlord planning a genuine capital improvement project assumes it gives them a right to end the tenancy once work is scheduled. It doesn't. A landlord considering a buyout assumes a signed agreement is final the moment ink hits paper. It isn't, for 45 days.

Owner move-in terminations exist for a narrow purpose: the property owner, or a qualifying family member, genuinely intends to live in the unit. This path is anchored entirely to that person's actual occupancy, which is also where it most often runs into trouble. San Francisco requires the named occupant to move in and stay for a minimum period, and state-level guidance issued by the California Attorney General now requires the unit be offered back to the displaced tenant if the named occupant doesn't actually move in within a defined window after the tenant leaves.

This path doesn't require the whole building to be involved, unlike Ellis Act withdrawal, and it doesn't compensate the tenant the way a capital improvement passthrough compensates the landlord. It's tied to one person's genuine intent to occupy one unit, and the documentation trail proving that intent is what protects a landlord if the termination is ever challenged as pretextual.

Ellis Act withdrawal is the path landlords reach for when the goal isn't recovering one unit, it's exiting the rental business at that property altogether. Under state law and San Francisco's implementing ordinance, a landlord can withdraw an entire building from the rental market, but the emphasis is on entire: this isn't a mechanism for ending one lease while continuing to rent the rest of the building.

Notice periods differ by tenant category: most tenants receive 120 days, while tenants 62 or older or with a qualifying disability receive a full year. Relocation payments follow a structure set on a periodic schedule by the Rent Board, paid in two installments, one with the notice and the remainder on vacancy, with an additional amount for qualifying senior or disabled tenants. The actual dollar figures on that schedule move over time, so a landlord evaluating this path needs the Rent Board's current Form 577 rather than a number carried over from an earlier year, including any number that might appear elsewhere in this article series.

What surprises landlords most is what happens after withdrawal. Re-renting any unit in a withdrawn building within a set number of years triggers restrictions: rent has to track what the displaced tenant was paying, adjusted only by otherwise-allowable increases, and that tenant holds a right of first refusal for a longer window still. Ellis Act withdrawal is not a way to clear a unit and re-rent it at market rate shortly afterward.

This deserves its own section precisely because it gets mistaken for one. A landlord planning a substantial renovation, a new roof, seismic retrofitting, replacing building systems, sometimes assumes that scope of work comes with an implicit right to clear the building. It doesn't. Capital improvement work in San Francisco is a cost-recovery mechanism, not an eviction path, and it's structured entirely around the tenant staying.

The process requires filing a petition with the Rent Board before any rent increase notice goes out, and the petition has to be filed within a set window after the work is completed. The passthrough itself is capped: for smaller buildings, roughly $30 or 5% of base rent annually, with the full certified cost eligible to pass through; for larger buildings, a similar dollar-or-percentage cap but typically only half the cost passes through unless tenants elect an alternative. The passthrough doesn't become part of the tenant's permanent base rent and it phases out once the amortization period ends.

If a renovation genuinely requires the unit vacant, that's a different situation, governed by different provisions with their own notice and compensation requirements, not this passthrough process. Conflating the two is a common and avoidable mistake.

A buyout is fundamentally different from the first two paths because it isn't backed by a notice period, a Rent Board order, or any legal requirement that the tenant leave. It's a negotiated agreement, and San Francisco regulates the negotiation itself before regulating the outcome.

Before any buyout discussion can begin, the landlord has to provide the tenant a Pre-Buyout Negotiation Disclosure form, explaining the tenant's rights, including the right to simply decline to negotiate at all. The landlord then has to file a declaration confirming that disclosure was provided, with the Rent Board, before actual negotiations start. The Rent Board separately catalogs completed buyout agreements by amount and neighborhood, which functions as a public record of what buyouts in a given area have actually looked like.

Even after an agreement is signed, it isn't final. The tenant holds an unconditional right to rescind within 45 days, returning any funds received and walking away from the deal entirely, no justification required. An agreement missing required disclosures or formalities can be rescinded even later, and a tenant has up to four years to sue over a defective agreement. A landlord treating a signed buyout as closed the moment it's signed is working from an incorrect assumption that can unwind the entire transaction months later.

None of the four paths above are something to execute directly from a general overview, and that's intentional. Each one has procedural requirements, current dollar figures, or eligibility questions that change over time and depend on the specific property and tenant situation, which is exactly the kind of determination a general compliance article shouldn't make on a reader's behalf.

What this article can do is narrow the question before a specialist conversation starts. A goal of owner occupancy, not a building-wide exit, already rules out Ellis Act. A tenant who needs to stay in place rules out both owner move-in and Ellis Act, pointing toward capital improvement work instead. Both sides potentially agreeing to an exit, rather than one side needing to force it, points toward a buyout conversation instead of a notice-based process. That narrowing turns a vague "how do I end this tenancy" question into a specific, answerable one.