That's the premise of analysts at Amherst Mortgage Securities, who recently argued that without more governmental intervention, 11.5 million borrowers would be in danger of losing their homes. They now argue that "the housing overhang is not caused solely by the number of non-performing loans that exist in the market. The problem also includes the high rates at which re-performing loans are re-defaulting."

CNBC.com

They also argue that the high rate at which "deeply underwater loans that have never been delinquent are going 2 payments behind for the first time" is only adding to the potential for far higher losses than the market currently predicts.

We know that more than half of the government modification trials go bad again. Big bank proprietary modification data is harder to get, but JP Morgan Chase's

JPM loading...

Quote  |  Chart  |  News  |  Profile

[JPM  Loading...      ()   ] re-default rate on proprietary modifications as of September 2010 was around 35 percent. And therein lies the problem...according to Amherst, bond investors and housing analysts "are focusing solely on non-performing loans; they ignore re-performing loans and seriously underwater borrowers."

I have argued many times that just because a loan is underwater (value of loan is higher than value of home) it doesn't necessarily mean that the borrower will stop making timely payments. Yes, the incentive to abandon the home is there, but for most homeowners, their home is their community, their daily life, and not just an investment. Most probably think the value will come back over time, and unless they desperately need to move, they have no reason to stop paying.

Amherst analysts disagree with me. "Borrower equity status is the single most important predictor of success," they claim. To explain their premise, they use two definitions of performing loans: A "successful" loan is one that is always performing, re-performing or voluntarily prepaid. A "clean success" takes out the re-performing loans. Here's what they found:

...for loans with equity, 88.9% were successful after 2 years, and 84.4% represented a clean success. For loans with CLTV >120, only 53.6% of loans were successful and only 40.9% represented a clean success.

I think I stand corrected.

We talk a lot about the shadow inventory of foreclosed properties overhanging the market and weighing down inventories, but the inventory of potential new defaults is clearly high; that potential, even with steady economic recovery, exists and must be factored into the equation.

The latest home price reports are not good, and even though sales appear to be bottoming in some markets, prices always lag. Also, many of the sales are foreclosures (around 30 percent), so that knocks the price recovery premise on its head as well.

The math is changing yet again, and investors should take note.