Closer scrutiny ordered for Santa Rosa’s anti-rent control petition
Sonoma County elections officials say they will need to conduct a full review of the signatures gathered in the petition drive to overturn Santa Rosa’s new rent control law, a delay that will leave the contentious issue in limbo through late December.
After initially concluding Oct. 14 that opponents submitted enough valid signatures to force a referendum on rent control, the Registrar of Voters now says after further review that it’s too close to call and a full hand count is required.
“We made an error,” Bill Rousseau, the county clerk and registrar of voters, said Tuesday. “It was our error and we caught it, and the city of Santa Rosa caught it as well.”
The reversal caps nearly two weeks of back-and-forth discussions between city and county officials over whether the 12,524 signatures submitted by paid petition gatherers funded by real estate industry interests were sufficient to force the City Council to either repeal the law or send the issue to voters.
If enough signatures are validated, the delay will hand the decision about whether to repeal the rent control law to a new City Council. That means the Nov. 8 vote still could affect the city’s rent control law, which remains suspended pending the validation process and potential referendum.
The city received five boxes of petitions on Sept. 26, and, lacking the staff or expertise to count them, asked the county registrar for help.
In addition to the boxes, the city turned over to elections officials letters from 151 people who requested their names be removed from the petition, most of whom claimed they were duped into signing the petition by gatherers who mischaracterized the effort.
Those 151 people proved to be a curve ball for elections officials.
“This is the first time people had withdrawn their names from a petition, at least in these numbers,” Rousseau said.
The withdrawal requests were tricky because elections officials at that stage weren’t counting every signature, but doing a statistical analysis meant to predict if enough valid signatures had been gathered. Opponents need 8,485 valid signatures on their petition to force an election on rent control.
Initially, that seemed a no-brainer. Opponents had submitted 4,039 more signatures than needed.
But a closer look at 501 randomly selected signatures revealed problems. Elections workers found 127 signatures that didn’t pass muster: 46 listed addresses out of the district, 41 were not registered voters, 34 were registered at different addresses, four could not be identified and two were registered late.
That left 374 of the 501 signatures as valid, or just under 75 percent, according to data provided to the city by deputy clerk Elizabeth Acosta. Stated differently, the random sample had an error rate of just over 25 percent.
Elections officials commonly use such statistical techniques to quickly estimate whether petitions have enough valid signatures.
When the 74.65 percent “sufficiency rate” was multiplied by the total signatures – minus the 151 signatures of people who asked to be removed – the analysis estimated there were likely 9,236 valid signatures in the batch.
That’s 751 more than necessary to force the referendum. But it was low enough to trigger a full hand count.