More Accurate Loss Reserving May Speed Housing Market Recovery

By John Leamons and Cecil Rhodes

The pace and sustainability of the ongoing economic recovery will depend to some extent on the way private mortgage insurers analyze and calculate the level of their loss reserves. The economy is dragging a large anchor — the housing market. The speed with which that anchor is hauled in will largely depend on the ability of Fannie Mae and Freddie Mac — and others if they can be induced to participate — to provide liquidity for mortgage loans.

Following the collapse of private-label mortgage securitization, Fannie Mae and Freddie Mac, supported by funding from the U.S. Treasury and the Federal Reserve, regained their dominance of the mortgage market. However, their ability to support a recovery in housing will depend in part on their ability to purchase loans having loan-to-value (LTV) ratios greater than 80 percent. By law, they may not purchase such loans without credit enhancement, which in large measure has traditionally been provided by private mortgage insurers.

The Federal Housing Finance Agency (FHFA), the regulator and conservator of Fannie Mae and Freddie Mac, recently declared that private mortgage insurance "is critical to the nation's housing and mortgage markets."

Safeguarding Claims-Paying Ability
No private mortgage insurers survived the Great Depression. When the industry began to revive in 1957, states were determined to make the industry robust. Fourteen states require mortgage insurers to maintain a risk-to-capital ratio no greater than 25:1. All states require insurers to build up reserves during profit­able periods so they can pay claims in the trough of a housing cycle. Mortgage insurers must hold half of their earned premium in a contingency reserve for ten years and can draw on that reserve only to pay claims and only in years in which losses exceed 35 percent of earned premiums. Also, they must establish a loss reserve (a case reserve) for insured loans that have been reported delinquent. (Mortgage insurers differ from property/casualty insurers in that they establish case reserves in advance of claims.)

Not all loans that become delinquent will result in a claim. Mortgage insurers must also establish an unearned premium reserve. And when the present value of expected future losses exceeds the present value of expected future premium and existing reserves, they must establish a premium deficiency reserve.

In hard times, the capital and reserve requirements intended to safeguard insurers' claims-paying ability can limit insurers' ability to write new business and thus limit the support they can provide to the housing market.

Case Reserve Analysis
In the wake of the mortgage meltdown, private mortgage insurers have had difficulty raising private capital and have been disappointed in their efforts to obtain capital from the U.S. Treasury. They have, however, benefited from a modest relaxation of state regulatory standards. And they have sought relief in improved loss reserving.

The level of the case reserve depends on the expected delinquency-to-claim rate and on the expected claim loss. Insurers differ in the way they calculate the case reserve, but all of them rely heavily on recent experience in forming their expectations of future claims. In particular, all insurers consider their recent experience in rescinding coverage on loans. Where coverage has been rescinded, no claim will be paid unless the insured successfully challenges the rescission. Expected losses are reduced accordingly.

The risk-in-force eliminated through rescissions has risen dramatically in recent years. Insurers contend that, particularly in 2006 and 2007, loan originators often misrepresented the character of the loans for which coverage was sought — especially loans destined for Wall Street securitizations.

Adjustments to the case reserve based primarily on the recent pace of rescissions are liable to error because of the extreme volatility of that parameter.
In an SEC filing, one insurer reported a near threefold increase in rescinded risk-in-force from one quarter to the next, followed by a two-and-a-half-fold increase in the following quarter. A more reliable approach would reduce reliance on the recent history of rescission activity by expanding the use of forensic loan audits.

One way of proceeding would be to assign each insured loan to a bucket based on its relevant characteristics — delivery channel, vintage, originator, loan product type, area house price appreciation (HPA) rate, original LTV, borrower debt-to-income (DTI) ratio, geographic region, and so forth. Analytic techniques necessarily accord different treatment to static characteristics, such as original LTV, and dynamic characteristics, such as current HPA. The desired granularity of such characteristics as HPA, LTV, DTI, and geographic region would determine the total number of audits necessary to obtain a representative sample for each bucket.

Based on the number of audits that uncover grounds for rescission, the probability of rescission for a random loan in each bucket could be calculated. Subsequently, when a loan from a given bucket was reported delinquent, the amount reserved could be discounted by the probability of rescission associated with the bucket to which the loan belongs. The amount reserved for the loan would be (C)(L)(1 – P(r)), where C is the expected delinquency-to-claim rate, L is the expected loss, and P(r) is the probability of rescission associated with the loan's bucket. The reduction in reserves resulting from the rescission discount would be available to support new business.

If private mortgage insurers become better able to support the housing market, the market will in turn strengthen them, creating a positive feedback loop that benefits the general economy.

John Leamons is an analyst and Cecil Rhodes is a principal consultant at ISO's NIA Consulting unit. NIA Consulting specializes in forensic financial risk analysis. Formed in 1982, NIA performs quality-control and fraud audits for some of the largest, most successful companies in the real estate, mortgage, and financial services industries.