The housing collapse the U.S. has endured over the past few years has forced the industry to question some of its most basic assumptions. For decades, the market believed that housing couldn't broadly decline in value, particularly not substantially. It was wrong. We now know that a bubble can actually result in overinflated housing prices and subsequent national home price declines. As banks and lenders adjust, they may have to revisit the mortgage products they used to believe were very safe. Will the 30-year fixed-rate mortgage survive?
Look Out: Default Risk!
The big surprise the mortgage market has to cope with is default risk. Even though the government-sponsored enterprises like Fannie Mae and Freddie Mac used to guarantee many mortgages, the private sector often didn't feel that guarantee was necessary. After all, there wasn't significant default risk. Sure, interest rates might change, and prepayments could come in more quickly than anticipated. But lenders wouldn't actually lose money on mortgages -- they just might not make as much as they thought they would.
Now that they know they could lose money, they must adjust accordingly. One way to do this would be to just have the government guarantee all mortgages. That approach is championed by some. But in this political climate, nationalizing the mortgage industry might not be easy. It's pretty likely that the government will at least want to make it possible for the private sector to originate some mortgages without a federal guarantee. In order to do so, however, what can lenders do to avoid losses if homes lose value?
Option 1: High Down Payments
The most obvious solution is a high down payment, of at least 30%. This would mean that the home could decline in value by up to that amount the day after the loan is written, and the lender would remain protected. Even though some localized markets saw prices decline more than 30%, if this was a universal standard prior to the boom, losses would have been much, much smaller. Indeed, most of the subprime loans never would have been originated, so the bubble might not have been created in the first place.
A mortgage product with an extremely large down payment could also result in a smaller term. After all, if you put down 30%, you might not need 30 years to pay off the balance -- 20 might be plenty. If the down payment was 50%, then 15 years might be long enough. Lenders and borrowers alike might find shorter maturities attractive if down payments rise considerably.
Option 2: The "DJI" Mortgage
There's a problem with giant down payments: not everyone has that kind of savings. And it's likely that public policy will dictate that home ownership should be spread more widely than to only those who can afford to save up a pile of cash to put down. So another type of new mortgage product should be sought that captures the risk of default when the loan-to-value ratio is higher than, say, 70% or whatever other value the market settles on.
I would recommend what I'll call the "DJI" Mortgage. DJI stands for Declining Jackhammer Interest. It's sort of like an option adjustable rate mortgage, but turned upside down.
So let's say you've got a borrower who can only afford 5% down. That makes the loan pretty unattractive to lenders due to the default risk they now fear. So initially, a very high interest rate would be necessary to guard against that risk, say, 8%. That way, the interest payments that come in would help to cushion the lender against potential losses in principal if housing prices decline.
But the thing about mortgages is that you can always refinance them. So the bank certainly doesn't want to be in a situation in ten years or so, where the borrower has paid down a significant amount of principal and can go refinance at 5% with somebody else. So it must build a mechanism into the mortgage to reset at a lower rate once the default risk shrinks.
Enter the Declining Jackhammer Interest mortgage. A jackhammer starts at ground level, then digs deeper and deeper as it breaks through layers of rock until you hit the level you're aiming for. That's how this mortgage product's interest rate would work. As new levels of principal are reached, the rate would decline.
Going back to the previous example, here's how that might work with a home worth $100,000, with 5% down, at 8% initial interest, and a 30-year term:*
- Year 0: 5% equity; interest rate begins at 8.0%; monthly payment: $697.08
- Year 5: 10% equity; interest rate drops to 7.5%; monthly payment: $667.67
- Year 8: 15% equity; interest rate drops to 7.0%; monthly payment: $641.56
- Year 11: 20% equity; interest rate drops to 6.5%; monthly payment: $618.24
- Year 13: 25% equity; interest rate drops to 6.0%; monthly payment: $597.38
- Year 15: 30% equity; interest rate drops to 5.5%; monthly payment: $578.77
And there it would remain for the life of the loan.
There are a lot of ways you could design this product, with either more or fewer trigger points. Home value would also matter, as you would probably technically base it on loan-to-value ratio, requiring periodic re-appraisals.
Something like this structure should both minimize the risk of loss from default faced by the lender and prevent borrowers from seeking to refinance the loan as they pay down principal. Declining default risk premiums are baked in. Of course, some borrowers may still refinance -- you can't eliminate prepayment risk entirely. Indeed, this sort of mortgage would encourage prepayments until the final trigger is hit. But from the bank's standpoint -- that's okay. Once a sufficient level of equity is achieved, it no longer has to worry much about default.
Of course, if the government does choose to guarantee all mortgages, then there would be no point to the DJI mortgage. Then, lenders would never have to worry about default risk. But in unlikely event that the private market doesn't end up relying on such federal insurance, the DJI mortgage could allow people who don't have a lot of money saved up buy a home, while making such a loan attractive to a bank.
*There was a tiny correction made to the payment amounts shown in the bullet points here due to a small mistake in the model. It revised the monthly payments down by about $1 each starting in year 5. Everything else remains the same.