Mortgage insurance protects the lender against loss in the event that the borrower defaults. The borrower pays the premium, but the lender receives the protection.

Mortgage insurance has no connection to any kind of life insurance, and pays no benefits to borrowers. The sole benefit received by the borrower is that, with mortgage insurance, lenders are willing to make loans with down payments smaller than 20% of purchase price or appraised value.

Lenders require private mortgage insurance (PMI) on mortgages with down payments less than 20% because the risk of default and loss to the lender is greater on loans with smaller down payments. The reason that the borrower pays for the coverage, however, is more historical accident than anything else.

When the modern PMI industry began in the late 1950s, many states had legal ceilings on interest rates. If lenders paid for mortgage insurance and passed on the cost to borrowers as a higher interest rate, they might have bumped up against those ceilings. If the borrower paid the premium, this potential roadblock was avoided.

Unfortunately, a borrower-pay system is much less effective than a lender-pay system. Borrowers do not shop for mortgage insurance but are locked into arrangements established by lenders, who decide the insurance carrier with which they want to do business.

When the borrower pays, lenders have little interest in minimizing insurance costs to the borrower because these costs rarely influence a consumer's decision regarding the selection of a lender. Insurers do not compete for the patronage of consumers, but for the patronage of the lenders, who select them. Such competition is directed not at premiums but at the services provided by the insurers to the lenders. Its effect is to raise the costs to insurers, and ultimately the cost borne by borrowers.

Under a lender-pay system, lenders would shop for the lowest premiums. Because lenders buy in bulk, they would have the market clout to push premiums down. (Even if borrowers shopped for insurance, their single-policy purchases wouldn't give them the same clout.) As a result, the higher interest rates under a lender-pay system would be lower than the combined cost of interest plus insurance premiums under the current borrower-pay system.

A lender-pay system also would eliminate confusion over when insurance can be terminated. Under the existing system, until very recently, the borrower could terminate insurance only with the permission of the lender. The lender, however, had no financial incentive to agree other than to please the borrower. Some lenders allowed PMI termination under certain specified conditions. Others had more stringent conditions. Still others did not allow it at all. Many borrowers, furthermore, were unaware of the possibility of terminating insurance, and paid premiums for years longer than necessary.

If lenders paid for mortgage insurance, they would decide when to terminate it, based on whether or not they felt the insurance was still needed. Some lenders would probably reward borrowers after terminating the insurance. Borrowers could choose between two-tier rate plans and single-rate plans. The rules would be set in the market rather than by government.

The one rule needed from the government is that lenders must purchase mortgage insurance.