Bribes and Blind Spots: SF’s Corruption Problem

Last month, Supervisor Shamann Walton accepted a $5,500 oil portrait of himself from a nonprofit called Urban Ed Academy gift him . He also accepted a $1,679 conference trip from a second nonprofit, Bay Area Community Resources. Then, without recusing, he voted along with the rest of the board to increase BACR’s city contract by $4.8 million. For accepting gifts above the legal limit and failing to disclose and recuse, the Ethics Commission fined him a measly $4,500.

San Francisco does not have one corruption scandal. It has a pattern — but the pattern is more complex than “officials sell votes for cash.” Admittedly, some of what follows is exactly that: contractors handing over gold watches and construction favors in exchange for city business, the kind of thing the FBI spent six years and a grand jury unwinding. Some of it is a different and more common failure: officials who had a financial relationship with an interested party, didn’t recuse, and whose regulators found no evidence of intent to trade the vote for the money. Both are real problems.

The Same Two Doors, Two Different Failures

Every scandal in this city runs through one of two chokepoints: the moment a contract is awarded, or the moment a project is approved. But what happens at each door varies in severity.

In June 2026, a joint audit by the Controller and the City Attorney found that Tajel Shah, the city’s chief assistant treasurer, spent years steering a $7 million contract to a friend’s company, Mechanical Orchard. She never disclosed the friendship, let the firm run a sole-source “discovery project,” fed it information no other bidder had, and altered the scoring method while adding $1.7 million in unsubstantiated costs to competing bids to make her friend look cheaper. Auditors called it the appearance of a “pay-to-play” tax system. This is closer to the bribery end of the spectrum: active, repeated manipulation of a process, not a single missed recusal. And yet — notably — no fine or criminal charge against Shah has been reported anywhere in the audit or the coverage of it; she was simply released from her position that November.

In February 2026, the Ethics Commission fined Planning Commissioner Kathrin Moore $12,000 for voting four times on projects involving Skidmore, Owings & Merrill, the architecture firm paying her about $15,000 a year in retirement income. The projects she voted on totaled nearly $900 million. The Ethics Commission’s own charging papers state there is “no evidence to suggest that she deliberately intended to benefit her financial interest” — her violation was failing to recuse, not steering votes for money. That is a serious violation of conflict-of-interest law. In the same batch, Film Commissioner Franco Finn was fined for his own disclosure violations.

These are comparatively small in relation to the case of the Mohammed Nuru scandal at Public Works: a six-year FBI probe that implicated more than two dozen people. Harlan Kelly, the former head of the Public Utilities Commission, went to prison for four years after a jury convicted him of bribery and fraud for taking bribes from a contractor chasing city work. Nick Bovis, a restaurateur, funneled bribes to Nuru through a children’s charity and got nine months; the judge called him “the face of private corruption in San Francisco for this scandal.” Sia Tahbazof, a developer, bribed city building officials for eighteen years to speed his permits and pass his inspections, and got three years of probation and a $75,000 fine, not a day in prison. Florence Kong, a contractor, gave Nuru a roughly $37,000 gold Rolex for favorable treatment on recycling contracts and got a year and a day plus a $95,000 fine.

Permits. Inspections. Contracts. Approvals. The Nuru scandal was not a story about one crooked director. It was a stress test of the same two doors, and — unlike the Walton and Moore cases — this one really was about bribery, proven at trial and admitted in plea after plea. It’s the clearest example in this piece of the two doors actually being sold, not just left unattended.

What’s Actually Wrong Here

This isn’t one bad apple or one unlucky year. In the past few years, San Francisco has caught a sitting supervisor, a chief assistant treasurer, a planning commissioner, a film commissioner, and a PUC general manager all failing the same basic test — money touched their decisions, and nothing stopped them from voting or steering anyway. That’s not confined to one office or one kind of official; it runs across the legislative branch, the treasurer’s office, planning, and utilities. A pattern that wide, that recent, and that consistent is the actual scandal.Three things keep letting it happen: concentrated discretion, detection that depends on luck, and penalties too small to deter.

Concentrated discretion is real at both doors. Shah could personally rewrite the scoring rubric on a $7 million bid. Moore could cast the vote on $900 million in projects without a second sign-off. Tahbazof’s scheme worked because individual plan checkers and inspectors held make-or-break power over his buildings. Concentrated authority is a necessary condition for both bribery and unrecused conflicts — but it doesn’t by itself explain which of the two you get, and Moore’s case shows the discretion going unchecked for over a decade: city attorneys warned her in a 2012 memo that her votes could breach ethics rules, and she kept voting anyway until reporting on the memo became public in 2025.

Detection here mostly depended on outsiders. The Shah scheme surfaced only after a whistleblower took it to reporters. The Nuru empire cracked because the FBI arrived. The city’s own auditors said the treasurer’s office should have built “stronger controls” to detect and deter misconduct, and the Controller called organizational culture the city’s “strongest line of defense” — an admission that the formal controls weren’t doing the job.

On penalties, the honest headline isn’t a single “going rate” — it’s an inconsistent range with little visible relationship to the harm. Walton’s disclosure and recusal failures cost $4,500. Moore’s four un-recused votes on $900 million in projects cost $12,000. Tahbazof’s eighteen years of bribing inspectors cost $75,000 and no prison time. Florence Kong’s Rolex-for-contracts bribe cost a year and a day in prison plus $95,000. And Shah — the person who ran what may be the largest single scheme by dollar value here — appears to have paid nothing at all beyond her job. That last gap is the more damning data point: it’s not that the fines are uniformly small, it’s that whether you’re punished at all seems to depend more on whether a whistleblower or the FBI got involved than on what you did.

What Might Actually Help

None of this points to one silver-bullet fix, but a few things plausibly help on their own terms.

Reducing discretionary chokepoints. A project that already complies with every local law arguably shouldn’t need a commissioner’s discretionary vote at all, since a discretionary vote is one more thing that can go unrecused or, in the Nuru cases, be bought outright. GrowSF made this argument after the Moore case. On the contract side, a transparent, harder-to-tamper-with scoring rubric is a narrower and less contestable fix.

Better detection that doesn’t rely on luck. Voters already built a tool for this. In November 2024, Proposition C created the city’s first Inspector General, with subpoena power reaching vendors and nonprofits seeking city money. The Controller hired Alexandra Shepard, a former federal prosecutor who was part of the team that put Mohammed Nuru away (she has clarified in interviews that she joined the case after charges were filed and handled the sentencing phase, rather than building it from the start). The office is new enough that its record is still to be written; the case for funding it is that it replaces “reporters or the FBI happen to notice” with something more systematic, not that it has already proven it will.

More consistent penalties. Whatever the right dollar amount is, the more striking problem in these cases isn’t that fines are small — it’s that they’re inconsistent, and that the person plausibly running the largest scheme here paid the least. Closing that gap matters more than any single number.

San Francisco has two real problems, not one. It has a bribery ring that took the FBI six years to unwind, and it has officials who keep failing to recuse from decisions involving people who pay them. Both move through the same two chokepoints — contract awards and project approvals — and both got caught more by luck than by design.