Empirical studies of amortisation behaviour

Empirical studies have been concerned with the choice of maturity of a mortgage and thus implicitly consider the rate of amortisation. This research has covered mortgage choices in the US (Dhillon et al. 1990), Canada (Breslaw et al. 1996) and the UK (Leece 1997). For example, Dhillon et al. (1990) evaluated the choice between a 15- and a 30-year fixed rate mortgage and estimated a simple probit to represent this choice. The empirical results stress the importance of affordability and tax subsidies. For example, the demand for contracts with short maturities was likely to be less when interest rates were high. Wealthier households, more able to adopt higher rates of amortisation of debt, were more likely to choose the shorter maturities. An interesting result was that liquidity constrained households maximised gearing, and consequently chose mortgages of a long duration. However, affordability has not been the major finding in all of the studies and cost minimising behaviour over the life of the mortgage is also apparent (see Breslaw et al. 1996).16

Breslaw et al. (1996) reported research on the Canadian mortgage market. The authors made an important distinction between the term and the maturity of the mortgage debt. The term is the time period for which the interest rate is fixed on the FRM. The maturity is the usual life of the mortgage over which payments are calculated, and implicitly represents the rate of amortisation. It is worth remembering that Canadian fixed rate mortgages are similar to those obtaining in the UK, that is rates of interest are typically fixed for between 1 and 5 years. The choice between this term and the maturity of the debt (that is the rate of amortisation) was modelled simultaneously. The estimation involved the use of an ordered bivariate probit model and covered the period 1980 to 1988.

Breslaw et al. identified different consumer strategies in the joint choice of mortgage instrument and mortgage term. The results indicated the importance of risk aversion, and mortgage cost minimising behaviour. For example, risk averse households facing a comparatively high mortgage cost adopted fixed rate debt and repaid quickly. This result depends upon the absence of a binding borrowing constraint. Such a constraint would encourage risk averse borrowers to cover the higher cost of fixed rate debt by lower payments, through an extension of maturity. Interestingly, there is a stream of largely normative literature that examines the implications of the choice of mortgage instrument from a cost minimisation point of view (Milevsky 2001; Tucker 1991; Sprecher & Willman 1993; Templeton et al. 1996). This literature examines lifetime costs over the given maturity of a mortgage, generally finding against the FRM. This research will be discussed in Chapter 7 of the book.

Zorn & Lea (1989) considered a range of possible payment behaviour for Canadian ARM holders. This included an analysis of partial prepayment and thus implicitly amortisation rates. Zorn & Lea found partial prepayment to be sensitive to the opportunity cost of equity in a property and the mortgage interest rate. The elasticity of partial repayment with respect to the opportunity cost of equity was negative, so that low opportunity cost increased repayment; a result compatible with theoretical predictions of Plaut. Higher incomes led to lower amortisation, a result that might be a proxy for the impact of wealth. Wealthy individuals might have a higher opportunity cost of equity that reflects more sophisticated portfolios. It might be worthwhile examining the behaviour of low and high income groups separately as in the mortgage demand models of Follain & Dunsky (1997).

Leece (1997) estimated a multinomial logit model of choice of debt maturity, using UK data. Three discrete choices were modelled: a standard 25-year contract, contracts greater than 25 years, contracts of less than 25 years. The study used a sample of 2033 mortgage holders taken from the British Household Panel Survey (BHPS) for 1991. The empirical analysis used personal characteristics (age, education, marital status) and income from both employment and investments. A unique aspect of the specification was a qualitative variable indicating that a household was experiencing problems meeting their housing costs. The results of the research suggested the importance of a household's perception that it faced financial problems - leading to longer maturities. Overall the results emphasised affordability, perhaps not surprising given the recession of the 1990s.

The Plaut model suggested the basis upon which a borrower would prefer a balloon mortgage, that is zero amortisation of debt. The balloon mortgage is not widespread in the US forming only 7.7% of the primary mortgages of households in the 1999 American Housing Survey. For the UK this choice corresponds to the endowment mortgage. The endowment represents a corner solution, that is when the preferred rate of amortisation is zero an endowment is optimal.17 Though some research suggests the dominance of affordability in driving endowment choice (Leece 1995b), there is also indicative evidence that the perceived rate of return on the endowment was a determinant of choice, offering some confirmation of the insights of the Plaut model (Leece 2000b).

The theoretical predictions of the Plaut model are not always confirmed by empirical work. For example, Dhillon et al. (1990) found a negative rather than a positive sign on the mortgage interest rate with respect to the choice of mortgage maturity. There may be some difficulties in translating the theoretical restrictions of the Plaut model directly into empirical practice. Actual mortgage choices may reflect constant repayment scheduling which in a two-period model would set an absolute constraint on amortisation. It might also be advantageous to identify constrained and unconstrained borrowers. Cost minimisation may be more important for some households. Amortisation is a part of a complex picture of multiple mortgage choices including the level of housing demand, gearing, portfolio choices and choice of mortgage instrument. The increase in the availability of flexible mortgages may ultimately generate the necessary data for more detailed study of amortisation behaviour.

Ultimately the prevalence of a truly flexible mortgage instrument may assist in overcoming liquidity problems and facilitating optimum life cycle and portfolio planning. Other issues would also arise, including the possibility and limitations of flexible scheduling being used to minimise default risk. Behavioural finance offers an interesting perspective on repayment flexibility that suggests that this facility may not be unproblem-atic. Thaler & Shefrin (1981), and Shefrin & Thaler (1988) argue that individuals struggle between impulse and planning and that some regular savings contracts represent commitments that discipline economic agents. This was an often-used rationale for the sale of the rather inflexible endowment mortgages in the UK. Commitment and fulfilment of regular savings contracts may also have information content for the lender. Thus, whether truly flexible amortisation schedules reduce or add to default risk, due to the removal of this financial discipline, might be a further interesting research question. For the UK recent evidence suggests it is the wealthier and more sophisticated borrowers that are taking advantage of increased payment flexibility.18

Emergency Money

Emergency Money

Within my 50 page guide, "101 Ways To Raise Emergency Money," you'll learn all the tricks, tips, and cost-cutting strategies you need to start saving and earning emergency funds. The report downloads straight to your computer and you can get started raising that cash in less than an hour from now!

Get My Free Ebook

Post a comment