General Motors is completely hosed. Its cars are ugly, they run on too much gas, Americans don't like them any more, and today the reckoning comes on top of extraordinary pensions and health care commitments to their workers that will drown the company for the foreseeable future.

That's the tone of most things I'm reading from smart people I trust. But can we imagine a scenario in which GM could come out of this thing looking OK?

Robert Gordon is an economics professor (and professional pessimistic) at Northwestern University, and he sees a light at the end of the tunnel. His argument:

Before the recession, annual U.S. automobile sales were about 18 million vehicles. They've dropped to half that, an insanely low--and unsustainable--level. At this rate, the average car would have to last 25 years.

Gordon's story is a tale of two graphs, both drawn most clearly by the Calculated Risk blog. The first graph shows the utter collapse of auto sales in the last year, which are approaching 1982 lows:


The second graph, also put together by Calculated Risk, shows the turnover ratio for the US auto fleet (total registered divided by annual sales rate).


As Calculated Risk points out, this picture is reason to smile cautiously. Most vehicles do not last a quarter century and that graph has to dip down from its highest point in the last 50 years. How? Removing vehicles from the fleet or, more likely, a sales increase.

Gordon and CR both expect annual auto sales to rebound to around 15 million by as early as next year. That's about 83 percent of pre-bubble figures. At the same time, GM's current restructuring, Politico reports, will "reduce the number of cars it must sell in a year to break even from the current 16 million down to 10 million." That's 62 percent of pre-bubble figures, which suggests that there's a lot of space there for profit.

Think of it this way: at best, GM could be like that movie The Rookie. In short: Dennis Quaid plays a washed up former-MLB pitcher who spends him time coaching and pitching to high schoolers, but he's so dominant at a high school level that they convince him to try out for the MLB again. It turns out that he's still got a surprising fastball, and he works his way up through Double-A and Triple-A ball until he finally makes it back to the majors.

Optimistically, that's what we can do for GM. It's out of the big leagues for now, and being scaled down to make do a much more humble level. If Gordon and CR are correct and auto purchases are set to soar, post-recession demand could help GM thrive on a leaner budget and ultimately work its way back. This is, however, a possibility rather than a destiny. Ultimately, GM is going to have to rediscover that fastball on its own.