I have a piece in the new Atlantic arguing that reducing the cost of failure actually makes us all better off:

Americans' public attitude toward the bankrupt, however, is not nearly as generous as our law. Every move to make things easier for debtors meets with fierce resistance, not merely from creditors, but from ordinary people who are making payments on time. As this article went to press, the Senate was scrambling to find a compromise on a long-stalled House proposal that would allow bankruptcy judges to reduce the principal of home loans where the value of the property has fallen below the mortgage's outstanding balance. Known as a "cramdown," the idea was popular with most congressional Democrats, but apparently not with the voting public, which was telling pollsters in ever higher numbers that they thought the whole housing-bailout package was unfair.

And isn't it? Most people didn't take out giant loans with tiny down payments or do repeated cash-out refinancings. Yet the cramdown plan would make the sober, steady majority foot the bill for other people's mistakes. First they would pay as taxpayers by helping to subsidize troubled loans. Then, the next time they needed a mortgage, they'd be charged a higher interest rate to compensate for the risk that they might declare bankruptcy and ask a judge to cram down their loan. And maybe they'd have to pay a third time, again as taxpayers, by bailing out banks that got too many of their loans crammed down. Meanwhile, the guy down the street who took out a second mortgage he couldn't afford, to remodel, would be sitting pretty in his $60,000 kitchen.

It isn't fair. But by the time someone is in bankruptcy, the time for fairness is already long past. Bankruptcy is the legal recognition that someone lacks the resources to meet financial obligations. Our system works so well precisely because it mostly sets aside our instinct for just deserts, and instead focuses on minimizing the costs to everyone. It lays out clear and predictable rules for lenders and borrowers, so that they can plan for disaster, and escape as quickly as possible if it arrives. Still, it's plain as day that, in the current crisis, a whole lot of people are getting help they haven't earned. As a result, commentators, academics, and legislators presiding over hearings have diverted much time and energy away from hashing out the ugly details of rescue efforts and toward making the one point on which we can all agree: these relief measures don't seem fair.

The moral of the article:  spend less time on punishment, more time on trying to minimize the social cost of the failure.

Since I wrote that, of course, I've been blogging a great deal about Chrysler.  So it seems fair to ask:  does the Chrysler deal do that?  After all, it minimizes the cost to the unions, the largest group with substantial direct exposure to the collapse.

In fact, bankruptcy frequently does shortchange creditors if a going concern can be gotten out of the ashes.  But the Chrysler deal went a little beyond that.  It didn't just pay wages on a gamble that a reorganized firm would be worth more than a liquidation; it reaffirmed the union's pension obligations, and, AFAICT, much of the rest of their current contract.  This is unprecedented.  Moreover, it was not necessary to keep the firm going.  Detroit autoworkers do not have a lot of other exciting opportunities to make $40 an hour in compensation.  Had the bankruptcy followed a more normal course, most of them would have kept going.

That makes it a pretty straightforward transfer from senior to junior creditors.  And that's costly.  The seniority rules have no particular moral priority; like traffic rules, they matter because they are the rules.  People make decisions based on what the rules are, and if you change the rules without warning, you get nasty accidents.

Right now, senior debt is cheaper than junior debt; secured debt cheaper than unsecured.  If you declare that there are no priority rules, because the government will step in and arrange things so its friends get to cut in line, then everyone will have to pay something close to what the most junior unsecured creditors get now.  Not only will interest rates go up, but terms will shorten--no one wants to lend into a period when default risk can't be calculated.  And companies that are particularly likely to have administration "friends"--union shops, when Democrats are in office; maybe oil companies or defense contractors for Republicans--are going to find it harder to do debt financing at all.

Most people underestimate just how economically valuable the rule of law is.  A roughly stable investment environment that doesn't maximize social justice is undoubtedly better than an unpredictible one that tries to--just as the billions of poor people who live in states that have tried to exchange the former for the latter.  The winners were not the dispossessed.

This can be exaggerated--markets have memories, but they don't last forever, and robust democracies can survive a fair amount of skullduggery at the margins.  But on balance, I doubt that deal maximized utility compared to an alternative where the government stayed out of it and Chrysler made whatever deal it could.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.