My last post is about knowing what is being measured when reading or hearing media reports about surveys declaring utopia has arrived or Armageddon is upon us.
Case in point, the weekend edition of the Wall Street Journal contains a story that the foreclosure rate is increasing. I inferred from the secondary headline "Rates for Many Categories Jumped ...6.6% of Mortgages Were at Least 30 Days Past Due" that they were referring to all mortgages. I know that some areas are bad, but 6.6% just seems unbelievable to me. The survey is from the Applied Analytics unit of Lenders Processing Service, Inc. Continuing with the story, there are some quotes and breakdowns of the data in by loan categories. About half way through, to the WSJ's credit, they describe the survey in some detail. Most importantly it is a survey of about 22 million mortgages originated between January 2004 and August 2007. Not all mortgages, just mortgages made in a 3 ½ year period. I presume the mortgages are the ones serviced by Lenders Processing, but I could not find any reference to the survey on their website. Also, what is the geographic makeup of their portfolio? Is it representative of the country as a whole or is it weighted disproportionately to one geographic area? Who knows?
Until I got to the description of the methodology, I thought the 6.6% was of all or most mortgages because it didn't say otherwise. How many other readers thought that as well or just read the headline? The surveys done by the Mortgage Bankers Association of America, http://www.mbaa.org, cover many more mortgages and show similar trends, but significantly smaller percentages.
One thing that all the surveys that I've read show is that about 50% of all foreclosures are in California, Florida, Arizona and Nevada. So if you're there, it's bad. The rest of us though, are in a different neighborhood.
Comments(4)