After a devastating cycle of bubble and bust, the U.S. housing sector is on the road to recovery. New homes are being built at the fastest rate in years and prices are increasing across the country. Foreclosures are down and the number of “underwater” mortgages has declined by almost 12 percent since the peak at the end of 2011. Even Fannie Mae and Freddie Mac, the mortgage-finance companies in conservatorship since 2008, are reporting record profits.

What’s wrong with this picture? None of this would be possible without massive government support. Today, the government owns or guarantees about 90 percent of new mortgages, up from about 50 percent in the mid-1990s. It isn’t sustainable, let alone fiscally acceptable, for the U.S. to have such a domineering presence in what should be a private-sector function.

The housing recovery now under way creates a perfect opportunity to plan for the future of U.S. mortgage markets. Several recent innovations in mortgage finance by economists and academics are worth considering.

First, it helps to understand the origins of today’s situation. Before the New Deal, people bought houses by borrowing for a few years at a time. They only paid interest until the loans matured, at which point they would make a large payment or refinance. That worked well enough until house prices collapsed during the Great Depression. Lenders refused to refinance, hoping to get paid in full. Many borrowers defaulted; about 10 percent of homes ended up in foreclosure.

Downward Spiral

To prevent another downward spiral, the U.S. came up with the self-amortizing, long-term, fixed-rate mortgage. It enticed lenders into offering these products by promising to buy mortgages that conformed to certain underwriting standards. That’s where Fannie Mae and Freddie Mac come in: They bundle loans into securities, then sell them to private investors. For a fee, the government absorbs the risk of borrower default.

As long as house prices were relatively stable, the new system worked. But once prices soared, only to collapse a few years later, scores of homeowners defaulted. A cascade of foreclosures further depressed prices as more houses were dumped onto the market. Economists say this was responsible for 20 percent to 30 percent of the decline in home prices from 2007 through 2009. It was the Great Depression all over again.

The biggest challenge going forward is separating the choice to buy a house from the decision to make a leveraged bet on housing prices. Right now, when a borrower puts down $50,000 to buy a $500,000 house, she doubles her equity if the value of the house goes up to $550,000. The lender, however, has no claim to any of that appreciation. Alternatively, if the price declines to $400,000, the borrower is suddenly in the hole. She has a strong incentive to default, leaving the lender in the lurch.

Outside the U.S., floating-rate mortgages, where monthly payments rise and fall with the short-term interest rate, help borrowers deal with some of this volatility. Interest rates generally move in line with the health of the economy, so these mortgages are more flexible for both borrowers and investors. This approach effectively allows borrowers to refinance even if they are underwater, yet it does nothing to reduce the risk of default and foreclosure associated with negative equity.

The U.S. must figure out a way to better manage these risks if it is to turn housing back over to the private sector. Fortunately, economists have lots of ideas. The common theme is that mortgage principal should be keyed to economic conditions, and monthly payments should rise and fall proportionately. These features ensure that borrowers have a stake in repaying their loans, while also making it easier for them to do so when times get tough.

Continuous Workouts

These new mortgages would also damp the swings in spending that come from the wealth effect. Robert Shiller, the Yale University economics professor and co-founder of the widely used Case-Shiller index of home prices, and colleagues have modeled a few versions of a product called a “continuous workout mortgage.” In essence, the Shiller loans would allow borrowers to pay higher interest rates upfront in exchange for the right to lower principal and monthly payments when house prices go down.

These loans might be right for some people, but we prefer another idea: Mortgages with principal and monthly payments that move with an index of neighborhood home prices. As prices rose, so would monthly payments. Conversely, if prices fell, monthly payments would, too. These might be more attractive to borrowers since they wouldn’t have to pay higher rates upfront. Instead, they would compensate lenders by passing on the gains from house-price appreciation.

Borrowers would still have an incentive to maintain their property because they would keep (or lose) any change in the value of their house relative to the prices of their neighbors’ homes. Investors’ demand for these products would probably be strong, given their demonstrated eagerness to gain exposure to single-family house prices by buying them outright.

The government and the private sector have an interest in this sort of financial innovation. Right now, investors have little appetite for mortgages that lack government guarantees, partly because they were badly burned by misrepresentations the last time around. But if the U.S. ever hopes to reduce Fannie’s and Freddie’s dominance in the marketplace, as its regulator, the Federal Housing Finance Agency, recommends, the country needs to accept that the 30-year fixed loan — a financial product from our grandparents’ generation — has outlived its usefulness. We can create a better housing market by encouraging the development of more resilient mortgages that don’t depend on federal default insurance.

To contact the Bloomberg View editorial board: view@bloomberg.net.