Executives of the failed solar-panel maker Solyndra LLC are refusing to answer questions at a congressional hearing today, a sign that the first serious financial scandal of the Obama administration has arrived.

Several issues cry out for investigation: Did government officials rush to approve a $535 million loan guarantee to the company two years ago? Did company officers paint an overly rosy picture of Solyndra’s finances after trouble set in? Why, earlier this year, did the Energy Department agree to let private investors get their money back from Solyndra before taxpayers did?

Lawmakers should dig deep for answers. But their most pressing duty at this stage goes beyond finding partisan dirt. Members of Congress need to figure out a more effective way to support the quest for cheaper alternative energy.

Right now, the U.S. government underwrites less than $2.5 billion a year in fundamental energy research, far less than the $3.6 billion it spent in 1980. Stepped-up support for research is overdue. By spreading small grants among hundreds of labs, the government can help ideas that are too tentative to win corporate support today but can turn into epic breakthroughs years from now.

Meanwhile, an infatuation with loan guarantees is scattering government money in unproductive ways. Since 2009, the Department of Energy has awarded or conditionally committed more than $30 billion in guarantees to support construction of 42 alternative-energy projects. About $9 billion of proposed guarantees are awaiting final approval by Sept. 30. But in this build-out phase, private-sector financing works far better than government backing.

Energy-related loan guarantees arose from the stimulus legislation of 2009. Policy makers thought a huge infusion of low-cost loans would create many thousands of jobs at solar-panel factories, alternative-energy power plants and the like. There was an implicit assumption that most of these ventures would succeed.

Barring fraud, Solyndra’s failure reflects the company’s bet on an inadequate technology. Its tubes, coated with an unusual four-metal compound, were supposed to cut power costs more than 20 percent. That wasn’t nearly enough. Production costs fell much faster for a rival technology, conventional flat silicon panels, and Solyndra couldn’t compete.

Other giant loan guarantees have likewise been steered toward technology that turned out to be unsuccessful. Solar Millennium AG of Germany got $2.1 billion in U.S. loan guarantees to build power plants that used sunlight and mirrors to create thermal energy. In August, Solar Millennium walked away from those guarantees, in favor of cheaper photovoltaic power.

Such sudden, unpredictable shifts epitomize the way progress happens. No single person, company or agency can predict which technology will prevail. The winner emerges through constant competition between different approaches and the workings of a free market as it allocates capital.

The government’s rightful role in this competition is at the beginning -- and the end. We favor government support for research labs that can put hundreds, even thousands, of interesting ideas in play. Trying to pick winners in the midst of the action is ill-advised. The government can accomplish more, with less risk, by simply becoming a big, reliable customer for solar, wind and geothermal power.

A well-conceived alternative energy program could save the U.S. many hundreds of billions in the years ahead. If the Solyndra debacle gets U.S. policy pointed in the right direction, the loan-guarantee losses won’t have been totally in vain.

To contact the Bloomberg View editorial board: view@bloomberg.net.