Despite the excellent reasons to reject the GM bailout, consider this: a strings-attached investment that tweaked GM’s production model could reap huge climate benefits — perhaps bigger than anything else we do to autos in the near term. That’s because the biggest opportunities in fuel economy are at the low end of the fleet, not in FutureCars.

Remember: you save more fuel switching from a 15 to 18 mpg car than switching from a 50 to 100 mpg car. (The explanation is here and here.) For a company like GM that’s based on building fuel-wasting behemoths this has huge implications. Seemingly minor tweaks can yield colossal returns. Let’s take a look at some specific changes to the GM fleet:

  • The Hummer H3 averages 15 mpg. Making an H3-equivalent that got just 18 mpg would be the fuel-saving equivalent of turning a Prius into a 100 mpg hypercar.
  • Grist thanks its sponsors. Become one.

  • The GMC Yukon Denali is even worse: it averages 14 mpg. Turning that tank into a 20 mpg truck would save more fuel than turning two Toyota Tacomas (22 mpg) into two Honda Civics (29 mpg).
  • The Chevy Trailblazer is worse yet: it averages 13 mpg. If we make Trailblazers that got 22 mpg, we’d save as much fuel as we would by taking a Toyota Corolla (31 mpg) off the road entirely.
  • Grist thanks its sponsors. Become one.

  • Making the Cadillac Escalade (14 mpg) get just 18 mpg would save more fuel than turning a 50 mpg car into a 500 mpg car.

It’s easy to get distracted by cool fads like hypermiling and the technological promise of Magic-Future Cars. I’m all for these things, but they are much, much, much less important than the drudgery of making improvements to the bottom end of the fleet. That’s where the real gains are to be had.

Of course, there’s another possible strategy: Just let GM fail outright. Then instead of upgrading Hummers to make them go from shockingly-horrible to just horrible, we might have a chance to go from shockingly-horrible to decent in terms of fuel efficiency. And personally, I’d like to retool a lot of those factories that are currently used to make planet-crushing tank-trucks. In their next iteration, maybe those Heartland factories could flourish as the manufacturing backbone of new infrastructure for renewable energy and truly efficient autos. That’d be groovy too.

* Please note: in all the comparisons above, I assume that the cars are driven an equal amount. Also, for any make and model there are many different packages and features that affect fuel efficiency; in the examples in this post, I picked specific real cars, but I didn’t try to average across all the available configurations of every make and model.

This post originally appeared at Sightline’s Daily Score blog.