Last week's announcement that the Federal Housing Administration would take a $1.7 billion draw from the U.S. Treasury to cover losses in its Mutual Mortgage Insurance Fund marks the latest chapter in the story of a mortgage market structure that is fundamentally broken.

Having a federally guaranteed FHA market in competition with what turned out to be a full federally guaranteed conventional conforming secondary market was destined to fail sometime. The tipping point came with irrational underwriting that created a super-heated environment that in turn amplified lenders' propensity to send their worst loans to FHA. FHA lacked the ability to understand this risk due to vast underinvestment in risk infrastructure and a clear mission conflict between maintaining actuarial soundness while promoting affordable housing objectives. This conflict contributed to the demise of the GSEs.

Lurking below the surface of FHA's MMI woes is an adverse selection problem that remains unless actions are taken to remove that risk. Today, lenders have built into their origination systems "best execution" capabilities that slot a loan to a guarantor providing the lender with the best all-in price. FHA was often the best execution for high-loan-to-value, low FICO loans. For instance, in the fourth quarter of 2008, 59% of FHA's business had FICO scores below 640 and 23% were below 580. FICOs below 580 are almost a sure bet for delinquency and provide a vivid reminder for why FHA now has to go hat in hand to Treasury.

Moreover, during the same year, nearly 58% of FHA's LTVs were above 95%. Such high concentrations of low FICO and high LTVs are a risk manager's nightmare scenario for risk layering. That FHA wound up with loans that had no business ever being originated is a story in itself. It features an inability to price this risk out due to a combination of resource and technology deficiencies and a market that allowed lenders to send FHA loans it had no ability to manage.

Therein lies the seed for change. The various proposals in both the House and Senate miss the mark for comprehensive housing reform by not recognizing where FHA competes against the GSEs in the mortgage market. Not giving the agency the same resources as the GSEs to manage risk places FHA at a competitive disadvantage to other guarantors. We should not be surprised at the results. The Corker-Warner proposal's government corporation replacement for Fannie Mae and Freddie Mac needs to include FHA  for the following reasons. Placing the FHA inside the Federal Mortgage Insurance Corp. would provide singular oversight and accountability for balancing the post-GSE mortgage market and a fully federal guaranteed market for specified market segments. The FMIC could establish specific requirements for the FHA market that do not overlap with the private market.

However, since it would have control over both markets, FMIC would be able to scale up the FHA business at times when the private market fades and vice versa. In that way FMIC would be able to more efficiently address procylicality when markets ebb and flow. The FMIC would be staffed and resourced appropriately, thus providing better management of the MMI fund. Moreover, there would be synergies in having one agency manage both the catastrophic insurance fund for the largely private mortgage market and the MMI fund for the FHA.

This recommendation would align with the House PATH proposal that would spin the FHA out of HUD and into a separate agency. But doing so does not address the future potential for FHA to be adversely selected. Without meaningful changes to FHA, the agency will fall into the same trap over and over again, much like Bill Murray in "Groundhog Day."

What about FHA's affordable housing objectives? Some form of them can remain if FHA were combined with FMIC. Agency goals could still be established statutorily and administered by FMIC and reported on to Congress. But no longer would a taxpayer insurance fund play second fiddle to a set of policy objectives that play well politically, but jeopardize the integrity of the MMI.

Some level of subsidy to low- and moderate- income segments and first-time homebuyers must remain an integral part of the housing finance system. We found out the hard way that, while a worthwhile goal, not everyone is ready for the financial obligations of homeownership. Rent-to-own programs may actually provide a sensible way to graduate new borrowers into the mortgage market without subjecting them to the financial shocks that can come along with that responsibility. Maintaining the FHA in its current form, however, assures future problems for FHA and the specter of taxpayer-funded bailouts for the insurance fund.

Clifford Rossi is the Professor-of-the-Practice at the Robert H. Smith School of Business at the University of Maryland.