San Francisco's violent crime fell 21 percent in 2025 while property crime dropped 29 percent, according to a new analysis published June 17, 2026, by the Pacific Research Institute. The report, written by senior fellow Steve Smith, argues that the city's turnaround from what he calls years of "urban dysfunction" demonstrates that public policy, political leadership, and voter expectations can still reverse disorder when officials respond to community demands.

The data breakdown shows dramatic improvements across multiple crime categories. Compared with 2019, violent crime is down 34 percent and property crime is down 54 percent. Homicides fell to just 28 in 2025, the lowest number recorded in San Francisco since 1954. Car theft declined by 45 percent, representing the largest reduction among 35 major American cities studied by the Council on Criminal Justice. The report notes that San Francisco's reductions in violent and property crime exceeded those of Los Angeles, Chicago, Dallas, San Antonio, Las Vegas, and Atlanta, suggesting local policy choices are driving improvement rather than broader national trends. Beyond crime statistics, Union Square retail vacancies have declined from their post-pandemic highs, and artificial intelligence companies are leasing millions of square feet of office space, with OpenAI, Anthropic, and other firms helping drive the strongest office leasing market the city has seen since before the pandemic.

The report emphasizes that "cities do not decline by accident, and they do not recover by accident either," with recovery following years of public frustration and political change. Smith, who describes himself as "one of San Francisco's harshest critics" during the city's downturn, documented rising crime, retail theft, and open-air drug markets under recalled district attorney Chesa Boudin's tenure. The analysis states that during those years, "property crime surged, retail theft became a national symbol of urban disorder, overdose deaths reached record levels, and confidence in city government collapsed." According to the report, the lesson is not that San Francisco's problems were exaggerated but that "they were real," and the city's recent progress suggests that "public policy still matters, leadership still matters, and voters still matter."

The report traces the turnaround to specific political and policy shifts. The recall of District Attorney Chesa Boudin, growing pressure on city leaders to address open-air drug markets, renewed attention to quality-of-life offenses, and a broader focus on public safety all signaled changing priorities among residents who had long tolerated rising disorder. Smith writes that the city's criminal justice philosophy had "seemed more focused on reducing incarceration and prosecution than on protecting public safety," but voter demands for change altered that approach. The analysis also points to the election of Mayor Daniel Lurie as aligning with the city's improvements. While the report acknowledges that problems remain unsolved—particularly addiction and overdose deaths in the Tenderloin neighborhood—it concludes that San Francisco has become "a case study in what happens when voters demand change." The city's experience, Smith argues, serves as both a cautionary tale about what happens when ideology overtakes outcomes and a reminder that even deeply troubled institutions can change course when leadership responds to community needs.