I am confused by this entry from Lending Tree:

Recently, Morgan Stanley revealed there is a rate of about 12 percent of mortgage defaults that are made up of homeowners who do have the means to pay for the mortgage but don’t. The company’s recent report showed borrowers who had high credit scores and loans were more likely to stop paying their mortgage when the rate was higher than what the home was worth. As of March, online real estate database reported about 23 percent of single-family homes had loans that were more than what the properties value was.

link: Fannie Mae Will Deny Loans to Deliberate Defaulters

So does this mean that high credit scores are measuring something other than credit-worthiness the way we’ve always been told? What’s the latent variable here? Or latent variables? Or I am not reading this correctly?

An article last year from the LA Times:

Reporting from Washington — Who is more likely to walk away from a house and a mortgage — a person with super-prime credit scores or someone with lower scores? Research using a massive sample of 24 million individual credit files has found that homeowners with high scores when they apply for a loan are 50% more likely to “strategically default” — abruptly and intentionally pull the plug and abandon the mortgage — compared with lower-scoring borrowers

link: Homeowners who ‘strategically default’ on loans a growing problem – latimes.com

So is this a question of all those risk-assessors getting trapped in their own logic? Hypotheses:

a) People with higher scores were more likely to be allowed to over-invest than those with lower scores–thus winding up underwater more often.

b) People with higher scores are more likely to do the math, and look to the bottom line, realizing that they are better off financially getting banned for a few years and taking the credit hit than they are continuing to pay on a house that will never come back to the value they paid; or

c) those with high credit can take the hit on their credit and not have it hurt their chances of getting housing or other credit than those with lower credit.

Interesting hypotheses. Too bad I have no idea how study any of them.