Page 245 - untitled
P. 245

             FINANCIAL CRISIS INQUIRY COMMISSION REPORT



         Mortgage Delinquencies by Region
         Arizona, California, Florida, and Nevada—the “sand states”—had the most
         problem loans.
         IN PERCENT, BY REGION
         16%
                                                                     13.6%
                                                                     Sand
                                                                     states
         12
                                                                     8.7%
                                                                     U.S.
                                                                     total
          8
                                                                     7.0%
                                                                     Non-sand
                                                                     states
          4



          0
             1998     2000     2002    2004     2006     2008    2010
         NOTE: Serious delinquencies include mortgages 90 days or more past due and those in foreclosure.
         SOURCE: Mortgage Bankers Association National Delinquency Survey



         Figure .

            Serious delinquency also varied by type of loan (see figure .). Subprime ad-
         justable-rate mortgages began to show increases in serious delinquency in early ,
         even as house prices were peaking; the rate rose rapidly to  in . By late ,
         the delinquency rate for subprime ARMs was . Prime ARMs did not weaken un-
         til , at about the same time as subprime fixed-rate mortgages. Prime fixed-rate
         mortgages, which have historically been the least risky, showed a slow increase in se-
         rious delinquency that coincided with the increasing severity of the recession and of
         unemployment in .
            The FCIC undertook an extensive examination of the relative performance of
         mortgages purchased or guaranteed by the GSEs, those securitized in the private
         market, and those insured by the Federal Housing Administration or Veterans Ad-
         ministration (see figure .). The analysis was conducted using roughly  million
                                                                  
         mortgages outstanding at the end of each year from  through . The data
         contained mortgages in four groups—loans that were sold into private label securiti-
         zations labeled subprime by issuers (labeled SUB), loans sold into private label Alt-A
         securitizations (ALT), loans either purchased or guaranteed by the GSEs (GSE), and
         loans guaranteed by the Federal Housing Administration or Veterans Administration
         (FHA). The GSE group, in addition to the more traditional conforming GSE loans,
               
   240   241   242   243   244   245   246   247   248   249   250