Into the Woods: Mortgage Credit Quality, Its Prospects, and Implications

A special Moody's Economy.com study incorporating unique data from
Equifax and Moody's Investors Service.

SUMMARY

This Moody's Economy.com special study utilizing unique data from Equifax and Moody's Investors Service assesses how much further credit conditions will deteriorate and exactly where the credit problems will be most pronounced. The study also evaluates the threat mortgage credit problems pose to the housing market and broader economy, and the scenarios under which these problems could result in an economic recession.

View the Table of Contents

KEY QUESTIONS

• How serious is the erosion in mortgage credit quality?
• Where across the country is quality weakest and where is it eroding fastest?
• Are the credit problems only in subprime lending?
• What is driving the surge in mortgage delinquencies and defaults?
• How should mortgage credit problems be modeled?
• Will the fallout on the housing market and broader economy be significant?
• Will global financial markets be significantly impacted?
• When will mortgage credit quality stabilize?
• What could make the credit outlook measurably better or worse?
• Should there be a policy response to the credit problems?

FEATURES

Projections are provided for first mortgage delinquency and default rates for the nation, states, and the nation's 200 largest metropolitan areas. Delinquency, foreclosure, and cumulative loss rate projections are also provided for securitized subprime adjustable rate mortgage pools by quarterly vintage. All of the projections have a quarterly periodicity, and forecasts are provided through the end of 2010. These projections are produced under a baseline, or most likely, outlook for the economy and housing market, and under three alternative, more negative scenarios.

SOURCES

Two different historical data sets are used in the study. The first is based on a large random sample of credit files taken at the end of each quarter from the credit bureau Equifax. Historical information is available for 30-, 60-, 90- and 120-day delinquency rates and the default rate is based on the number of loans and dollar value of outstanding balances for first liens, home equity lines of credit, and closed-end second mortgages at the state and metro area levels. Historical data are now available through the first quarter of 2007. The second data set is compiled by Moody's Investors Service from pools of securitized subprime adjustable-rate mortgage loans. Historical information is available for quarterly vintages through the third quarter of 2006.

For more information on the data sets behind this study, please contact Richard Bloom at 212.553.6893 or richard.bloom@moodys.com.

MODELS

The study describes in detail the specification, estimation, and validation of econometric models of mortgage delinquency and foreclosure rates from the Equifax data set, and foreclosure and cumulative loss rates from the Moody's Investors Service data set. Mortgage credit conditions are determined to be influenced by a wide range of factors, including but not limited to unemployment, employment growth, house-price growth, household debt service burdens, and various measures of lenders' underwriting standards. The modeling effort is shown to be useful in determining how sensitive mortgage credit quality is to these various factors, and how these sensitivities vary over time and across different regions.

These models are used in combination with economic and housing market forecasts created using Moody's Economy.com large-scale macroeconometric models to produce projections of mortgage credit quality. Moody's Economy.com produces forecasts of a wide range of economic and housing market variables for the nation, states, and metro areas each month. A baseline, most likely, scenario is produced, as are three more negative alternative scenarios, including 25%, 10% and 4% scenarios.

FORECAST SCENARIOS

The baseline scenario is designed to be in the middle of the distribution of possible economic outcomes over the next three years. Growth slows through midyear under the weight of the housing correction, but the expansion remains firmly intact. The downside 25% scenario is designed so that there is a 25% probability that the economy will perform worse, and 75% probability that it will perform better. The economy remains recession-free in this scenario, but the housing downturn is more severe and the economic fallout results in higher unemployment than in the baseline.

The downside 10% scenario is designed so that there is a 10% probability that the economy will perform worse, and 90% probability that it will perform better. This is a recession scenario, the severity of which is similar to the 2001 and 1990-1991 downturns. Finally, the downside 4% scenario is designed so that there is only a 4% probability that the economy will perform worse, and 96% probability that it will perform better. Another way to think about the 4% scenario is that it represents an economic outcome that will occur once every 25 years. The catalyst for this scenario is a global financial event emanating from the hard-pressed U.S. mortgage-backed securities market.

ECONOMIC FALLOUT

The nation's housing and mortgage markets appear to be very fragile. While the ongoing correction in these markets appears to be evolving in an orderly way, if just a few things fall the wrong way, then the current corrections may very well devolve into a crash. The resulting fallout on the economy would be substantial. This study ends with a careful consideration of this possibility.