The fate of the Federal Housing Administration is a hot issue in Washington.

In December the Clinton administration floated a trial balloon about eliminating the FHA. Later it pulled back from the idea; now it is considering some public-private partnership.

The FHA has many tasks, most notably insuring single-family mortgages of low-to-moderate-income borrowers. There are several players in the private sector, such as GE Mortgage Insurance and MGIC Investment Corp., that also insure mortgages of borrowers with low-to-moderate incomes.

The theory behind reengineering the FHA is that the government should not offer a product that is already being offered by the private sector. The private mortgage insurers have been adamant that they could insure 60% of the borrowers currently served by the FHA.

Perhaps the Clinton administration's original plan to eliminate the FHA completely is the optimal solution. The private sector is willing to underwrite the majority of the borrowers who are served by the FHA.

It is important to note, however, that 40% of the FHA customers will likely not be served by the private sector - and thus would probably not be able to own a home. Not coincidentally, this 40% is made up of higher- risk borrowers.

Is the answer to have a private-public partnership whereby the private sector gets its 60% and the FHA gets the higher-risk 40%?

This proposal might lower administrative costs, because managing a smaller insurance portfolio should not require so large an infrastructure.

However, there would be added costs. The 40% higher-risk portion of the portfolio would incur the highest credit losses. The current system allows premiums from the lower-risk 60% to subsidize the losses from the higher- risk portion.

Though administrative costs might be reduced if the lower-risk portion were turned over to the private sector, the savings could be eaten up by credit losses. Premium revenues might no longer cover the risk.

The other proposal is for a risk-sharing agreement between the private sector and the FHA. The basic tenet of this proposal is that the private sector would administer and insure most of the mortgages currently serviced by the FHA.

The private mortgage insurers would receive most, if not all, of the premiums for the insurance. Then, if there is significant default experience, the private companies could be made whole by the FHA.

In this scenario it appears that the FHA, and thus the American taxpayer, would share in the risk, but not in the premium.

Risk sharing has been the backbone of the FHA's multifamily loan portfolio. The credit losses in this portfolio have been tremendous. As you would expect, the private entities take the rewards and the government gets stuck with the risk.

The risk-sharing plan might reduce administrative costs at FHA, but it will be important to assess once again what the credit-loss costs would be.

In the FHA's 1994 fiscal year, it insured about $100 billion of mortgages. The FHA's insurance portfolio is about $332 billion, and it has received about $7.5 billion in premium revenue to insure this risk.

If risk sharing is going to be considered as an option, the American taxpayer must be told how much risk the FHA will insure and what the new premium revenue stream will be.

The taxpayer should be wary of another Fannie Mae/Freddie Mac situation whereby the government insures all the risk without receiving any premium.

Ultimately, the taxpayers must decide what type of housing insurance program they want to pay for.

If they are comfortable with spreading credit risk among all mortgage borrowers and nonhomeowners, then perhaps, the current system is optimal.

But if voters were telling Washington last November that they don't want to subsidize risk arbitrarily, the current discussion of risk sharing does not approach a real solution.

The real question that must be addressed is: With the private mortgage market as efficient as it is, is there really a role for the federal government in housing?

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.