Changes in Characteristics of Adjustable-Rate Mortgage Borrowers between 2001 and 2004

Article excerpt

To examine whether the characteristics of adjustable-rate mortgage borrowers changed over the 2001-2004 period, the study compares the distributions of these demographic characteristics in the 2001 and 2004 Survey of Consumer Finance (SCF) using two methods. However, we find that there are different results when we apply the two methods. The study explains the differences between these two approaches and concludes that a small change exists in the use of adjustable-rate mortgages (ARMs) by lower income, younger, single-headed, low-education families. Meantime, a noteworthy increase of ARM use exists in higher income or wealthy, well-educated and white families. Keywords: Mortgage market, adjustable-rate mortgages, mortgage financial education


Despite rising house prices, more and more people became homeowners. The homeownership rate increased from 68.0% in 2002 to 68.8% in 2005 as shown by U.S. Department of Housing and Urban Development. A recent study performed by Finke, Huston, Siman and Corlija which used Survey of Consumer Finance (SCF) data set from 1989 through 2001 suggests that households holding adjustable-rate mortgage shifted to financially vulnerable families such as lower income, younger and less educated families. Their study urged those vulnerable groups to be targeted for special financial education in the long run interest.

Have the characteristics of adjustable-rate mortgages (ARMs) borrowers changed over the 2001-2004 period? In order to examine the change in the proportion of ARMs that is in the mortgage market and in the characteristics of ARMs borrowers respectively, the present paper that followed the study of Finke et al. compares the proportions of ARMs in the SCF between 2001 and 2004. The two different results are found by two different methods. We explain the differences in the two results and then find the changes in the characteristics of ARMs borrowers. The analysis of the surveys indicates that the proportion of ARMs, as a proportion of total loan, substantially increased in the mortgage market over the 2001-2004 period. But the proportion of ARMs held by low income or low wealth families presented a small change during the same period. Finally, we support the suggestion that it is necessary to implement the financial education for these financially vulnerable groups.

Literature Review

Fixed rate mortgages (FRMs) have a fixed rate cost for the length of a loan, while the rate on an ARM fluctuates with the market. The appeal of an ARM is the low initial interest rate, easing borrowing constraints. However, the danger is an unanticipated increase in the rate could hurt some families' ability to pay off (Curry, 2004) if the rate is reduced the borrowers benefit. But with rates going up, hundreds or even thousands of dollars may be added to a borrower's annual interest expense and monthly payment. In addition, Hagerty (2004) believes if rates do rise and stay high, most borrowers will reduce other kinds of spending because they need to pay off the increasing monthly payments. That would have a negative impact for the economy as a whole.

One reason for the substantially rising homeownership rate is low mortgage rates experienced during 2001-04 period. The adjustable rate of mortgage dropped 1.85% from January of 2001 to April of 20043. With respect to mortgage type, ARMs have become more popular among consumers in recent years. The share of ARMs, as a percentage of all mortgages, accounted for 34% of mortgage market in 2004 (FREDDIE MAC ARM annual survey, 2004). According to the Mortgage Bankers Association's (MBA) 2004 Single-Family Mortgage Activity Survey (2005), in the second half of 2004, 46% of new mortgages were ARMs. About half of these ARMs were hybrids which had initial low fixed-rate periods of at least three or five years, followed by an adjustable rate.

A common accepted opinion is that households with more stable income, lower risk aversion, and higher probability of moving prefer to choose ARM (Campbell and Cocco, 2003). …