People who favor greater regulation of banking say that loose or ineffective regulation was a major cause of the recent crisis. They propose thousands of pages of regulations on dozens of different subjects, allegedly to prevent future crises.

None of the new laws and regulations will address effectively the egregious conduct that led directly to the collapse: Grossly defective mortgages were sold in huge volumes, at prices greatly exceeding their value, to ill-informed (in fact, systematically misled) investors. The new risk-retention requirement, with its underlying "qualifying residential mortgage" definition, aims in the right general direction to prevent more of this, but it is very far off target.

Anyone who has examined origination and servicing files for defaulted mortgages from 2005-7 has seen many that would have failed any reasonable underwriting or credit management process. These are mortgages with unsubstantiated and unreasonable stated income; with available evidence of multiple property ownership (hence, not primary residences); properties with negligible or no investment by the owner; and loans which at prevailing rates would within a few years have required payments greatly disproportionate to borrower income — to give just a few examples.

Why were responsible underwriting processes abandoned? Because it appeared to market participants, including banks, that easy money could be made by taking bad mortgages and passing them on to a greater fool, an "investor," preferably through a series of other intermediaries.

This was an activity in which many thousands of people engaged. They were very well paid to do so. The great majority have kept their winnings. Can anyone doubt that many of these people and others would be delighted to do the same again?

Last time, some institutions didn't see that when the music stopped they would be left holding enough toxic mortgages in pipeline to sink them. Others overleveraged and keeled over when the first soft breeze blew.

But then there are the rest, who lost some money, got some help from taxpayers and may be ready soon to have another go at it. If they can do that, then what will likely follow within a few years is another housing bubble.

Rates are historically and artificially low. Investors will again reach for yield. Available money for bad mortgages can again drive up demand and prices for homes. Prices can again reach levels that are unsustainable when — after some months or years of appreciation appear to vindicate the junk securities — another wave of defaults occurs.

It is ironic that in the long and turbulent wake of the crisis, a motley assemblage of public authorities now seeks to impose servicing requirements — but not underwriting and loan quality requirements. Bad servicing was not a cause of the crisis.

Assuring effective servicing is going to be extremely difficult and complex. All the more so when perverted by political and macroeconomic considerations, with a perceived imperative to keep everyone (evidently including fraudsters) in their homes.

So, rather than obsessing over servicing, why not focus on loan production, which is far easier to regulate and far more relevant to the future of banks and our economy? How about actually giving priority to avoidance of the next crisis, rather than maintaining a maudlin and unproductive emphasis on cleaning up the debris from the last one?

The core question is a very simple one: Should an institution protected by federal deposit insurance be allowed to originate or buy any damned thing it chooses, in unlimited quantities? Is that OK as long as it retains x% of the risk — surely a wrongheaded and backhanded way of protecting the foolish from their folly?

I say no, for three reasons.

First, banks as an industry have unequivocally demonstrated that they are incapable of acting on any distinction between good and bad mortgages. Originating toxic mortgages will inevitably be fatal to at least some banks — and the more risk they retain, the more rapidly this will be fatal.

Second, origination and spraying out of bad mortgages exposes banks to immense liabilities. The courts are slow, they are expensive, they are unpredictable. Nonetheless, no prudential regulator should be comfortable imagining that if a bank has enough capital and can keep the junk off its balance sheet, then it can't fail.

Third, even if their various exemptions and taxpayer support could enable banks to get and sell bad mortgages with utter impunity to financial risk and legal liability, their doing so has been shown to have catastrophic consequences for the wider U.S. economy.

Why are we talking about QRM? Why aren't we talking about explicit underwriting standards and reporting and examination processes that vindicate industry and public interest in preventing banks from participating in the creation of bad and mislabeled assets — whether or not they do so with the intent to sell them? Start with mortgages and move on from there.

Andrew Kahr is a principal in Credit Builders LLC, a financial product testing and development company. He was the founding chief executive of Providian Corp. and can be reached at akahr@creditbuilders.us.com.