Tag Archives: loan characteristics

2-4 Unit Properties: Housing’s Middle Child

Posted on by

photo by Kgbo

The Urban Institute’s Laurie Goodman and Jun Zhu have posted Do Two- to Four-Unit Properties Have Higher Credit Risk? An Analysis of Default and Loss Experience to SSRN. The abstract reads,

Two- to four-family properties make up 19% of all rental housing but receive almost no attention. Using a unique dataset from Freddie Mac and Fannie Mae, we show that, for any given set of loan characteristics and compared with one-unit properties, two- to four-unit properties are more likely to default, its owner-occupied (investment) properties are less (more) likely to liquidate, and all two- to four-unit properties are more likely to have a higher loss severity upon liquidation. Historically, these patterns have led to higher losses on two- to four-unit loans. Current tighten credit results in loss rates much closer to those on one-unit owner-occupied properties, indicating that policymakers can relax the credit requirements of two-to-four properties to better serve affordable rental housing.

It is great that the authors are looking at the neglected, middle child of the rental housing market. Providing 19% of the rental housing stock is nothing to sneeze at, even if other segments of the housing stock provide more.

It is particularly interesting to me that owner-occupied 2-4s do better than investor-owned 2-4s in terms of liquidation, even while overall 2-4s are roughly on par with 1-unit owner occupied properties in that regard. There are a lot of other interesting tidbits about this housing stock in the paper, such as the fact that these properties are more likely to be owned by lower-income households and that 2-units have the highest default rates of 1-4 unit properties.

The authors make the case that

though predicted losses on two- to four-unit production are now on par with one-unit owner-occupied properties, the low volume suggests that many borrowers (who are disproportionately likely to be low and moderate income and minority) are getting squeezed out. In the interest of expanding credit to these underserved populations and expanding, or at least preserving, the supply of affordable rental housing, the government-sponsored enterprises (GSEs) could relax the current loan-to-value requirements. If this relaxing were coupled with counseling for landlords, we believe it would make financing more available for this critical part of the market, with little additional risk to the GSEs. (3)

This all sounds good, although I am somewhat skeptical of the claim that reduced financing costs for owners will be passed onto tenants in the form of lower rents or rent increases. There are a lot of factors that go into rent levels, and costs are just one of them. The local demand for housing as well as the competing supply cannot be ignored. Owners may be able to keep all of those reduced financing costs as additional profits, depending on those local conditions.

The main question I am left with after reading the paper is — why haven’t Fannie and Freddie, whose data the paper is based upon, already reached the same conclusion about loosening credit for this type of housing? Do they know something about it that the author’s don’t?

Frannie v. Private-Label Smackdown

Posted on by

Eric Armstrong

S&P posted a report, Historical Data Show That Agency Mortgage Loans Are Likely to Perform Significantly Better Than Comparable Non-Agency Loans. The overview notes,

  • We examined the default frequencies of both agency and non-­agency mortgage loans originated from 1999­-2008.
  • As expected, default rates for both agencies and non-­agencies were higher for crisis-­era vintages relative to pre­-crisis vintages.
  • The loan characteristics that were the most significant predictors of default were FICO scores, debt­-to­-income (DTI) ratios, and loan­-to­-value (LTV) ratios.
  • Agency loans performed substantially better than non­agency loans for all vintages examined. The default rate of agency loans was approximately 30%-­65% that for comparable non-­agency loans, whether analyzed via stratification or through a logistic regression framework. (1)

This is not so surprising, but it is interesting to see the relative performance of Frannie (Fannie & Freddie) and Private-Label loans quantified and it is worth thinking through the implications of this disparity.

S&P was able to do this analysis because Fannie and Freddie released their “loan-level, historical performance data” to the public in order to both increase transparency and to encourage private capital to return to the secondary mortgage market. (1) Given that the two companies have transferred significant credit risk to third parties in the last few years, this is a useful exercise for potential investors, regulators and policymakers.

It is unclear to me that this historical data gives us much insight into future performance of either Frannie or Private-Label securities because so much has changed since the 2000s. Dodd-Frank enacted the Qualified Mortgage, Ability-to-Repay and Qualified Mortgage regimes for the primary and secondary mortgage market and they have fundamentally changed the nature of Private-Label securities. And the fact that Fannie and Freddie are now in conservatorship has changed how they do business in very significant ways just as much. So, yes, old Frannie mortgages are likely to perform better, but what about new ones?

Loose Credit. Plummeting Prices.

Posted on by

"Durdach Bros Miller Lite pic4" by MobiusDaXter

Christopher Palmer has posted Why Did So Many Subprime Borrowers Default During the Crisis: Loose Credit or Plummeting Prices? to SSRN. While this is a technical paper, it is clear from the title that it addresses an important question. If it can help us get to the root causes of the foreclosure crisis, it is worth considering. The abstract reads,

The surge in subprime mortgage defaults during the Great Recession triggered trillions of dollars of losses in the financial sector and accounted for more than 50% of foreclosures at the height of the crisis. In particular, subprime mortgages originated in 2006-2007 were three times more likely to default within three years than mortgages originated in 2003-2004.

In the ensuing years of debate, many have argued that this pattern across cohorts represents a deterioration in lending standards over time. I confirm this important channel empirically and quantify the relative importance of an alternative hypothesis: later cohorts defaulted at higher rates in large part because house price declines left them more likely to have negative equity.

Using comprehensive loan-level data that includes much of the recovery period, I find that changing borrower and loan characteristics can explain up to 40% of the difference in cohort default rates, with the remaining heterogeneity across cohorts caused by local house-price declines. To account for the endogeneity of prices — especially that price declines themselves could have been caused by subprime lending — I instrument for house price changes with long-run regional variation in house-price cyclicality.

Control-function results confirm that price declines unrelated to the credit expansion causally explain the majority of the disparity in cohort performance. Counterfactual simulations show that if 2006 borrowers had faced the price paths that the average 2003 borrower did, their annual default rate would have dropped from 12% to 5.6%.

Ok, ok — this is hyper-technical! The implications, however, are important: “These results imply that a) tighter subprime lending standards would have muted the increase in defaults, but b) even the relatively “responsible” subprime mortgages of 2003–2004 were sensitive to significant property value declines.” (40) It concludes that, “In reality, cohort outcomes are driven by both vintage effects (i.e. characteristics bottled into the contracts at origination) and path dependency in that exposure to economic conditions affect cohorts differently depending on their history.” (40)

So, the bottom line is that loose credit and plummeting prices were both causes of the defaults during the crisis. Mortgage underwriters and policymakers are on notice that they need to account for both of them in order to be prepared for the next crisis. This paper’s contribution is that it has quantified the relative impact of each of those causes.