Sunday, May 20, 2018

Housing: Part 298 - FICO Scores

There isn't much new here.  I was just putting together a chart from the New York Fed Household Debt and Credit Report, which tracks mortgage originations back to 1999, by FICO score.  It just still amazes me how little there is here.  How an entire, passionate point of view about the housing bubble grew out of a supposition that has no backing in the data.

There was a boost in mortgage refinancing in 2003 because households with strong credit tend to tactically refinance when rates are low.  For the entire rest of the housing boom, there was no shift in lending, by FICO scores.  The general rise in the value of originations followed along with rising home values, but then it leveled off.  So, not only was the private securitization boom not associated with a rise in lending to low-FICO scores, it wasn't associated with a rise in originations at all.   2018
Source: New York Federal Reserve
From the end of 2003 to the end of 2005, the value of residential real estate rose by about 40%.  We might expect mortgage originations to scale with aggregate value, even if it is rising value that is driving lending, not the other way around.  But, that didn't even happen.  While valuations increased by about 40%, originations remained flat.

There is a tiny bump up in originations to borrowers with FICO scores under 620 toward the end of the securitization boom.  The idea that this could form the basis for any noticeable shift in the aggregate market is a stretch, to say the least.   2018
Source: New York Federal Reserve
Sometimes you hear hand-waving about this.  One explanation is that rising home prices were falsely bloating  FICO scores.  The problem with that explanation is that prices in the Contagion cities tended not to rise until 2004, so this wasn't the case there.  And, the housing stock in the Closed Access cities is overwhelmingly owned by households with high incomes while households with lower incomes were moving away from those cities in droves.  The Closed Access housing markets are not a good example of markets driven by low credit quality borrower activity.

But, really, someday, I think people will look back and wonder how such widely held and extreme beliefs developed from empirical conclusions that were built entirely from modifications to data that, in its raw form, provided little or no basis for the conclusion.  This data is collected in a fairly timely fashion.  When the broad consensus formed in 2006 and 2007 that collapse was inevitable and curtailing lending at the margin was an important part of the process that we needed to enforce that collapse, this data was available.  Did anyone dare point that out at the time?

There is a lot of latent potential lending out there. (Averages are quarterly.)

 I am sure that many readers might think that lending was excessive throughout the 1999 to 2007 period, so that I can't really treat the average originations for that period as a normal baseline.  Let's say they are correct.  Then, my question for them is, what explanation do you have for borrowing that rose evenly across FICO scores from 1999 to 2007, but then only dropped for FICO scores under 720 after 2007?  Was lending too generous, and were prices too high, for all types of borrowers before 2008, but for some reason lending and relative prices only needed to decline for low FICO scores and low tier housing markets?  Can someone explain that to me?


  1. Wasn’t it more a situation of lower standards on the asset side of the underwriting rather than the borrower? For example, didn’t advance rates (loan to value ratios) grow well beyond previous averages?

    1. Yes. If there was a shift in risk, it was in the terms of the loans, not the borrowers. That is why it has been a terrible mistake to impose strict new lending standards that have sharply reduced lending based on borrower characteristics.

    2. Easier to scapegoat the customer than admit to flaws in the business plan!

    3. I would argue that the prime scapegoat is the lenders/business plan. The places where terms decayed were places where rent affordability has been extremely high because of an endemic shortage of housing. That created a natural demand for mortgages that were less affordable. Those loans were a part of the destabilization, but there was nothing inevitable about the process that brought on the destabilization.

      The CFPB doesn't keep you from paying 50% of your income on rent. It only keeps you from taking on a mortgage. And, we don't track households that become homeless or move to less expensive cities because their rents are high with the same specificity that we track delinquencies and foreclosures.

      This leaves the impression that mortgage risk is the problem, but it is rent affordability that is the problem. Since we scapegoated the lenders, this problem has only gotten worse.

  2. Great post. Yes housing and rent affordability are certainly key issues along the West Coast and elsewhere.

    The macroeconomics profession seems unable to wrestle with the policy implications.

  3. OT

    “Everyone recognizes the stability and security of the U.S. market more than ever before,” Genton says. He adds that up to 70 percent of a 59-unit Four Seasons Private Residences project in Beverly Hills likely will be foreigners.


    I understand not everyone subscribes to the "large current account trade deficits equates to rising house prices" idea. There is some serious scholarship in this area that suggests as much, but there is serious scholarship backing up every point of view in macro. So set that aside.

    But the optics! We build housing on the Westside of LA, just not for Americans.

    Various versions of these bad optics scenarios play out, even domestically. We build (somewhat less desirable) housing in downtown LA too, just not for middle-class people. Even there, you better be well-set financially, if you want to buy.

    Of course, the solution is not xenophobia, or social welfare-housing programs, but un-zoning property. But that is the solution not on the radar.

    Given context, the optics are perfect for creating class envy, obstructionism and xenophobia.