In 2001, San Francisco voters passed Measure B, allowing the city to sell $100 million in bonds to install solar panels on city-owned rooftops. With the blackouts of 2000 on their minds, and the prospect of becoming the nation's first "solar city," the measure passed with 73% in favor.
To date, the city of San Francisco has sold exactly zero bonds.
By law, the San Francisco bonds can only be sold to finance a solar project if the cost is cheaper than the price to generate power using traditional electric systems. Since the price of solar panels and related technology tends to exceed the price of fossil fuel-generated electricity, San Francisco hasn't sold any of the bonds.
I checked some other Web sites monitoring the state of the solar energy industry, and they all expect the price of solar energy systems that interconnect to the grid to become competitive "soon," with the most hyped coming from entrepreneurs with new companys and venture capital funding. But "competitive" means solar averages $3 per watt compared to a regular utitily's $1 per watt. No one is making predictions about solar ever becoming "cheaper" than traditional systems (except for off-the-grid remote systems, say, in rural Africa, where it soon will be cheaper).
San Francisco has been forging ahead with its commitment to solar anyway. The 60,000 square feet of photovoltaic panels on the roof of Moscone Center, for example, generates 65kW (the equivalent energy to power approximately 8,500 homes) is the larged municipally owned solar facility in the U.S. But it wasn't funded from the bond measure. The project, like projects at SF General, the airport, and Pier 96, are paid for out of Willie Brown's Mayor's Energy Conservation Account from PUC revenues.
So what's the statute of limitations on a bond sale? I'm thinking that by 2020, the cost of solar could drop to the threshhold to trigger the sale, but by then, SF may already filled its rooftops with PV panels.