|
Review of Home Price Movements
Over the last five years, the direction of home prices in the United States experienced a dramatic turnaround after appreciating considerably over the first half of the decade. While the annualized home-appreciation values tend to be more volatile, most charts tell the same story. After significant increases in home prices during the early part of the decade, prices declined precipitously over a period of approximately three years before showing signs of stabilization during the last several months. Depending on the index, home prices declined anywhere from 31.76% to 33.52% from peak to trough, and Las Vegas, the hardest hit MSA of the twenty tracked by Case Shiller, declined 55.60% peak to trough.
As mentioned previously, the government has taken enormous measures to help stabilize the housing market. These steps include direct purchase of mortgage-backed securities to make financing more affordable, offering tax credits to qualified first-time homebuyers, and injection of capital to the largest US lenders to ensure their survival. While it may be impossible to quantify precisely how much impact these policies have had, it is hard to imagine they haven’t had significant impact. In fact, the Case-Shiller 20 MSA Seasonally-Adjusted Index posted a month-over-month gain for the seventh consecutive month in its December reading. Because of these data points, many speculate that the housing recession has ended. Conventional wisdom suggests that since the housing market has suffered such a steep decline during this recession, we are nearing a bottom in terms of home values. However, this is a dangerous assumption to make for a couple of reasons. Home price declines and the loss of equity in the home have demonstrated some of the strongest correlations to mortgage defaults. As a result, large supplies of distressed inventories still exist all across the country, and they should continue to weigh on the housing markets going forward.
The negative employment situation should also adversely impact the housing market and hold back a recovery for some time. According to the Bureau of Labor Statistics, the national unemployment rate is 9.7%. The negative feedback loop between home prices and unemployment should continue to weigh heavily on the market for at least the near future.
Against this backdrop, performance for residential mortgages in the near future across all sectors is expected to be poor. Before the downturn in the housing market, HPA (Home Price Appreciation) masked the underlying credit risks associated with many of these mortgages as borrowers were able to tap into the equity in their homes and refinance out of problems or sell their homes to completely repay their mortgages.
Cure rates and the probability that delinquent loans would not result in a loss were very high during the housing boom. According to one study performed by Amherst Securities Group (Amherst Mortgage Insight September 23, 2009), cure rates in the first quarter of 2005 were 84% for borrowers behind one payment, 66% for those behind two, and 41% for those behind three. Now that borrowers no longer have the option to tap the equity in their homes to save themselves from default, borrowers who fall behind two payments are almost certain to default and those who miss one payment will default the majority of the time.
JPMorgan Chase releases default projections by sector and year of origination on a monthly basis, assuming that the most recent six-month average of transition rates (rates at which one loan of a particular loan type moves from one level of delinquency to another) stay constant over time. For loans originated in 2006 and 2007, even the best performing sector – Prime Fixed – is expected to default in large numbers.
In the March 2010 report, of the current population, 30% of the borrowers are expected to default. The expected default rate increases to 54% for those who are one payment behind, 66% for those who are two payments behind and 70% for those three payments behind. A total of 77% of loans in foreclosure are expected to go through liquidation. Ninety-nine percent of the loans in REO status are expected to suffer losses when liquidated. The projected lifetime default for Prime Fixed loans originated in 2006 and 2007 is expected to be approximately 35% of the Prime Fixed population as of March 2010.
As we move forward, what areas should we examine to determine whether we have moved out of the housing recession? In Deustche Bank’s December 11, 2008 report, the bank’s research analysts cite four reasons why they believed home prices were likely to “overcorrect” despite record affordability at the time. They include:
1) Auto-correlation – home prices are auto-correlated and in a market that depends so much on “comparables,” price momentum is a big driver in future home values.
2) Tight credit – while affordability may be high according to traditional measures, due to the elimination of many of the affordability products such as IOs, 40-year amortization loans, and 80/20 lending, fewer buyers can find financing to purchase new homes. Underwriting continues to tighten even in products that were once considered a haven for the less creditworthy such as FHA/VA.
3) Economic recession – low home prices and record-low interest rates are not enough to compel potential homebuyers if they are unemployed, underemployed, or fear becoming under or unemployed. In fact, as is commonly known, the 9.7% unemployment rate understates the severity of the current job market. If the underemployed and those who have quit looking for work are included, the rate of those unemployed and underemployed would be 16.8%.
4) Distressed supply – serious delinquencies continue to rise. Servicers have done a better job of managing their pipeline of distressed inventories, and doing so has prevented distressed sales from being a bigger percentage of actual home transactions recently.
Examining these four areas now, only the auto-correlation factor has arguably moved from a negative to a positive trend. Even that assertion can be debated, because while the seasonally-adjusted month-over-month changes have been positive, the non-seasonally adjusted number has been negative for the past three months. It is understood that once Spring arrives when the housing activity picks up, absent an unexpected negative exogenous macro-economic event, prices should move back to positive if seasonal patterns hold. However, seasonal patterns may not be enough to offset the negative situation.
Any discussion regarding the path of home prices without mentioning the impact of modifications would be incomplete. Whether the modifications were initiated by the servicers themselves or completed as a part of a broader government program, they have kept many foreclosures from hitting the market. Completed and future modifications guarantee that many more that have yet to hit the market will do so over an extended period rather than within a tight timeline. Keeping a flood of distressed properties off the market undoubtedly has had a positive impact in the housing market as evidenced by the increases in home prices during most of the second half of 2009. In fact, during the months in which home prices increased the most, foreclosures as a share of total home sales were low during these months relative to what the market experienced during the worst of the housing crisis.
However, liquidating distressed properties over a long period should prevent any sharp rally in home prices. We should expect servicers to take advantage of any uptick in home prices and sell into a rally, thus preventing any robust recovery in the housing market. In addition, while the re-default rates are expected to be lower for modifications completed through the government-led Home Affordable Modification Program (HAMP), the recidivism rates experienced from previous modification efforts suggest that a large percentage of them will probably re-default. Thus, the supply of distressed properties is expected to be quite large for the foreseeable future.
For instance, Barclays Capital’s February 5, 2010 Securitization Research Report, “An Update on HAMP Modifications,” reported that the average payment reduction for HAMP modifications in December of 2009 was approximately 42%. In a separate Barclays Capital’s report in October 2, 2009, “Loan Modification: Recent trends, re-defaults, and future projections,” a chart shows that cumulative re-default rates for modifications in 3Q08 with payment reductions of greater than 40% reached more than 50% by the ninth month after modification.
Admittedly, there are many factors that impact the re-default rates for modifications such as sector, borrower payment history, the delinquency status of the borrower at the time of modification, FICO score, MTM CLTV, and date at which it was completed. However, even if we were to assume that the re-default rates would be half of that of the population in the chart below, more than 25% of the modifications would re-default by month nine.
In fact, modifications completed after the 3Q08 have experienced lower re-default rates during the immediate months after modification, the trajectory still seems to indicate that they will be lower, but not significantly lower. The re-default rates for modifications completed in 1Q09 is almost 40% by month six, slightly more than 10 percentage points lower than that for modifications completed in 3Q08. In short, early low re-default rates do not necessarily guarantee significant performance improvement for the life of the mortgages.
Modifications should have the dual effect of stabilizing home prices while preventing any sharp rally in the housing market absent any robust economic recovery and significant improvement in the re-default rates for modifications. Without any improvement in the recidivism rates, these efforts will only delay the inevitable for many borrowers and do little to the supply of distressed properties in the long run. However, by properly managing the delinquency pipeline, servicers can prevent a large demand/supply imbalance in the housing market at any one point in time.
Going forward, despite recent positive news, the housing market should feel some pressure in the near term. First, the Federal Reserve has discontinued the MBS purchase program in March. The end to the program should push mortgage rates higher reducing the purchase power of many potential homebuyers. More importantly, the unprecedented levels of distressed inventory will increase supply and distressed sales should make up a higher percentage of sales adding further pressure to markets.
Finally, it is unlikely that the negative economic environment will abate anytime soon. In aggregate, these factors are expected to overwhelm the recent positive developments in the housing market. And even if home prices do not suffer a sharp decline in the near future, the distressed supply overhang of distressed properties may prevent any meaningful recovery as servicers slowly unwind their inventories over an extended period of time. The unwinding of the distressed supplies may take a very long time and home prices may continue to remain subdued absent any unexpected, robust economic recovery.
|