There are plenty of ideas kicking around out there about how Congress can help promote a new transportation infrastructure that doesn't rely quite so heavily on oil. One oft-mentioned proposal involves having federal agencies buy up plug-in hybrids and flex-fuel cars for their fleets, in order to help jump-start the market for those vehicles. One snag, though, as Kimberly Kindy and Dan Keating of The Washington Post illustrated in a must-read piece over the weekend, is that the government can mangle these mandates very, very badly—and actually end up making the problem worse. Especially when it comes to flex-fuel cars.

The story goes like this: Back in 1992, Congress passed a law requiring all federal agencies to buy alternative-fuel vehicles for 75 percent of their light-duty fleet. The catch was that, while the agencies had to buy the cars, they didn't actually have to use the alternative fuel. So a lot of agencies ended up purchasing cars that could run on propane, compressed natural gas, or E85 (an 85 percent ethanol, 15 percent gasoline blend), and them shipped them to areas that didn't actually have any alternative fueling stations—the infrastructure just wasn't in place. Fewer than 0.1 percent of fueling stations in the United States even offer E85. That meant most flex-fuel cars were running on plain old gasoline, and, since these vehicles generally have larger-than-average engines, they actually end up using more oil and emitting more carbon dioxide. The Postal Service used 1.5 million additional gallons of gas last year because only 1 percent of its 37,000 flex-fuel vans were actually running on ethanol.

This whole debacle partly came about because the law—and the flex-fuel incentives Congress offered to automakers—were badly designed. In one longstanding and egregious provision, automakers are allowed to get extra credit under fuel-economy laws if they spend $50-$100 to convert the fuel tank of an SUV that gets, say, 16 mpg into something that can handle alcohol-based fuel. Congress considers this roughly equivalent to producing a 24 mpg car under CAFE rules. So, when federal agencies started looking around for flex-fuel cars to buy, the cheapest options available were these often big, gas-guzzling sedans and SUVs that could technically run on ethanol—except that ethanol usually wasn't available. This is a festering issue with flex-fuel car incentives in general, but it's compounded when you have government agencies that are required to buy and drive these vehicles.

Now, you wouldn't necessarily see the same type of fiasco with plug-in hybrids, which remain a much more promising technology—car companies, after all, can't just cheaply convert an inefficient SUV into an electric-powered model, and it's harder for a federal agency to buy a plug-in hybrid and then never plug it in, relying only on the gasoline engine. Plus, it's not like mandates are always unworkable: If the federal government had been better about sharing data with private companies about where it actually needed E85 infrastructure built, you might have had better synchronization. (Of course, there's the more fundamental question of whether E85 is actually better, from a climate perspective, than regular gasoline—there's excellent reason to think that, when you consider all the indirect impacts, corn-based ethanol doesn't make sense, period.) Still, the whole Post piece is worth reading as a crucial cautionary tale.

--Bradford Plumer