If Ben Bernanke won't act, principal reductions just might be the most powerful tool we have for breaking open the housing recovery

Remember housing?

There was a bubble. Then were was a bust. Now there's a mess that's holding back the recovery. Even if there is nascent housing recovery underway, there are still two big stories holding back a big housing rebound. The first is the Fed not doing enough. Unless Ben Bernanke & Co. signal that they'd be willing to tolerate more inflation for a bit of time, credit will remain tight. The second is the awful legacy of the burst bubble -- underwater mortgages, foreclosures, and shadow supply -- acting as a persistent drag on the sector.

What is to be done? Stopping foreclosures is the first step. That prevents shadow supply from piling up, prices getting dragged down, and more owners falling underwater. In other words, it means modifying mortgages.

There are three big ways to modify a mortgage. You can reduce the principal, you can reduce the interest rate, or you can extend the length of the loan. Banks prefer the latter two options because they maximize the value of loans -- which is another way of saying that they don't do as much for borrowers. 

That's not an opinion. That's what the evidence says. Laurie Goodman, a senior managing partner at Amherst Securities, collected and presented the following empirical data at a recent panel at the left-leaning Economic Policy Institute (EPI). The charts below show the redefault reates on mortgage modifications for subprime and prime borrowers -- broken down by how much pay relief they received on their modifications. For example, the 20-40 columns show what percentage of borrowers who saw their monthly payments fall by 20 to 40 percent subsequently defaulted.

Remember, the lower the redefault rate, the better. 

Principal reductions trumped rate reductions for every class of borrower for every amount of pay relief. This is good, but it's not completely clear whether it's of the ambiguous or unambiguous variety. The data isn't granular enough to say. It may be that the pay relief on principal reductions is bigger than the pay relief on rate reductions within each quintile. Possible, however unlikely.

That leaves us with the unambiguously good story -- which happens to square with our intuitions. Principal reductions work better than rate reductions -- even when borrowers are saving the same amount -- because their balance sheets are in better shape. And that improves their incentives. In other words, you're more likely to pay off a $100,000 mortgage on a $100,000 house than a $200,000 mortgage on $100,000 house, even if your monthly payments are the same.

Balance sheets matter. Especially upside down ones. That's a big part of what has made the Great Recession so great. Underwater borrowers have pulled back and kept pulling back five years later.

Getting them rightside up would go a long way towards doing the same for the economy. That might be the most important thing someone not named Ben Bernanke could for the economy.