The Federal Reserve Board recently released the latest in a series of occasional reports on mortgage lending based on data collected under the 1975 Home Mortgage Disclosure Act (HMDA). The law, broadly amended in 1989 and several times since, requires most mortgage lending institutions with offices in metropolitan areas to disclose information about the geographic location, the income, sex, and race or ethnicity of individuals applying for credit, the disposition of the application, and other characteristics of the home loans they originate or purchase. The 8,850 lenders currently covered by the law account for an estimated 80 percent of home lending nationwide.

For 2005, lenders reported information on about 30.2 million home-loan applications. 11.7 million of those applications were for purchasing one-to-four family homes; 15.9 million were for refinancing, 2.5 million for home improvements, and the balance were applications for financing buildings with five or more residential units.

The report is primarily designed to look at patterns of higher priced home lending but along the way it reveals a lot of information about the home mortgage market in general.

For example, even though mortgage companies (as opposed to depository institutions) represented only about one-fifth of the companies submitting data, they accounted for 60 percent of all home-loan applications. However, those mortgage companies that were affiliated in some way with a depository institution tended to be very active lenders. The 576 affiliated companies processed 24 percent of all of the applications in 2005.

The survey revealed that, while different types of lending institutions tend to specialize in certain types of home loans, this is less true than in the past. The most notable change is in the origination of government-backed originations. Mortgage companies account for about 64 percent of the applications for these loans; as recently as 2002 they had an 83 percent market share. Still the segmentation is strong; depository institutions extended 71 percent of home-improvement loans and about 88 percent of multi-family loans while commercial banks account for about half of the loans for manufactured housing.

The home-loan business is surprisingly concentrated. The data showed that 60 percent of the reporting organizations provided information on fewer than 250 loans or applications each, accounting for just 1.6 percent of all the reported data. However, 6 percent of institutions each provided information on 5,000 or more loans or applications - 88 percent of the reported data.

Depository institutions are easily identified by asset size where such information does not exist for mortgage companies. Within this limitation it was found that asset size and lending activity are highly correlated. The 707 depository institutions with $1 billion or more in assets reported 86 percent of all applications while the 4,236 institutions with assets of less than $250 million each accounted for only 5 percent of the data.

Even more striking, where data from affiliated companies were aggregated to the highest level of corporate organization, the top 25 organizations accounted for 54 percent of the 2005 data. This was similar to findings in 2004.

Of the 30,146,893 applications covered by the study, 24,665,065 were acted upon by the lender and 6,691,068 or 27.1 percent of the loans were denied. The rate of denial was influenced greatly by the type of loan and the lien position. Only 16.4 of conventional first mortgage loans for home purchase were rejected by the lender and 17.9 of comparable junior liens. Government backed purchase mortgages had a rejection rate of 12.5 percent for first liens and 11.3 for junior liens (there were less than 1,200 applications for the junior liens.) One third of applications for conventional senior loans for the purpose of refinancing were denied and 29.8 percent of junior loans in that category. Home improvement loans were turned down at rates of 35.5 percent and 41.9 percent for first and secondary liens respectively. The winner in the loan denial sweepstakes, however, was manufactured homes; 52.6 percent of applications for purchase and 54.2 percent of applications for refinancing were denied.

Multi-family purchase loans fared best of all - only 9.5 percent of these applications were denied and only 10 percent of requests for multi-family refinancing.

The report commented on a belief that the strength in the housing markets recently could be accounted for by a growing number of sales to investors and to those seeking second homes. In 2005 lenders reported about 3 million applications for nonowner-occupied (as principal residence) properties. About half of these were applications for conventional purchase money first mortgages. HMDA data reported a fairly steady percentage of this type of loan from 1990 through the middle part of that decade, generally in the range of 4.5 to 6.0 percent of both number of loans and total dollar value. However, after the mid-1990s the figures began to rise and now represent about 16 percent of the total number of loans. This was up 1 percent from 2004 figures. The nonowner-occupied applications are not spread evenly across the country - Florida, Nevada, Hawaii, South Carolina, and Vermont are among the leaders in this type of lending - probably for obvious recreational and climate reasons.

The report also looked at the incidence of "piggyback" lending. We have talked in the past about this mechanism of obtaining first and second mortgages at the same time, the second mortgage being used to bring the down payment on the senior lien up to a level where the buyer is not required to purchase private mortgage insurance. While piggyback loans (also called 80-10-10 loans) are not identified as such in the HMDA data, researchers were able to extrapolate their existence through dates, lien position, and census tract data. For 2005 the report estimates that a total of 1.37 junior loans were used for the purpose of providing a down payment for home purchase. This is an increase of 74 percent from data analyzed in 2004. The conclusion is that 22 percent of reported senior liens used for home purchases on one to four family owner occupied homes involved a junior-lien or piggyback loan reported by the same lender, up from about 14 percent in 2004.

We will continue with an analysis of this report which also talks about lending for manufactured housing and provides some insights into loan pricing and just who gets those high-priced loans.