If you focus on an invented, unreal and poorly stated "problem," then there's no limit to the number of alternative "solutions" you can gin up. All of them having no practical use. That's where we are now, in the confused debate that wrongly centers on the GSEs.

The financial crisis was precipitated and then propelled by costs arising from underpricing of risk assumed by the government-sponsored enterprises (hence, over $100 billion so far in expense to taxpayers); irresponsible mortgage lending fuelled by the profitability of selling mortgage securities to ill-informed buyers; and immense losses on credit-default swaps (AIG).

These three causes all have the same ultimate origin: operations that shifted risks from a lender to others or created new risk — at immense, though sometimes only temporary, profit to the entities that produced and marketed the risk. And all these causes have had the same result: the taxpayer pays. Banks hate retaining even 5% of the risk, but they're overjoyed to stick the taxpayer with 100% of it.

The obvious way to prevent this from recurring is not by imagining that we are all — willing or not, incented or not — going to underwrite better, service better, leverage less and market more responsibly. That's an unlikely scenario. No, the way to avoid a repeat is to discourage and reduce the spreading, creation and assumption of risk to those who are not lenders. Since the risk bearer of last resort is the taxpayer, I will propose using the tax system to eliminate the taxpayer's risk.

Some people make it sound as if homes would become unaffordable to many (hence empty?) if government did not assume default risk. That's obvious nonsense. There was an immense volume of non-GSE, "private" mortgage securities. And let's give these securities credit for including many of the worst loans — those that even Fannie Mae and Freddie Mac wouldn't touch. The existence of the unsupported issues demonstrates that it's easy to achieve a mortgage market lacking either overt or implicit government support — and hence without taxpayer risk.

Would retail mortgage rates be higher? Surely. Maybe a half of a percent higher, maybe 1% higher. The risk premium is alleged to be in that range.

So, higher rates. So what? Over recent decades rates have fluctuated in a very wide band, for instance between 4% and 8%. Somehow or other, when rates were high, houses were nonetheless built and sold. Their prices even continued to rise reliably. This was not a bubble deflated by high rates. A 1% increase in the rate is a minor perturbation.

The government wishes to pursue a social policy of encouraging homeownership. Great — I guess. But there is a right way to do it. Guaranteeing mortgages is a wrong way to do it. It's inefficient, poorly targeted, concentrates taxpayer risk and creates bizarre and politicized entities that destroy value.

To make houses more affordable, increase — guess what — the much-maligned mortgage interest deduction, or better yet, replace it with a corresponding tax credit. If the average mortgage rate over time is, for instance, 7%, of which 1% is attributable to the absence of a government guarantee, then provide a homeownership tax credit, refundable in cash, equal to one-seventh of the taxpayer's mortgage interest expense. Avoid interaction with other elements of the tax structure.

This could be limited to those with modest incomes, like the earned income credit is now. It could be paid out monthly, via the servicer, since we now have cheap electronic payment capability.

To subsidize the cost of homeownership for certain categories of households, just go ahead this way and directly, efficiently, transparently subsidize it, leaving control in the hands of Congress.

The subsidy, if it is reliably assured, will be treated in mortgage underwriting as additional income. It will enable those with less income to own homes. It will have important countercyclical benefits as well, since, as to new mortgage transactions, the subsidy can be increased or diminished in order to deflate bubbles or maintain demand, and to avoid an unduly harsh impact from market rate changes.

To render ownership feasible for those with income but who can't make a down payment, just encourage rather than seek to prohibit the private market from offering loans at high loan to value, even 100% — but only to those whom the social policy targets. "Encourage" does not mean "guarantee." Lenders have already shown an invincible propensity to whittle away down-payment requirements.

These innovations would enable us to phase out a whole alphabet stew of agencies and entities, starting with Fannie and Freddie. Combining this with changes that assure unambiguous, supple alignment of servicer with owner interests and responsible packaging and marketing, we could achieve an efficient and honest market for mortgages and mortgage securities that effectively serves government policy without unnecessary cost or risk to taxpayers.

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.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.