Equilibrium rationing separating equilibrium and liquidity constraints

This section examines the evidence for equilibrium credit rationing (Sti-glitz & Weiss 1981; Williamson 1986, 1987). Given that empirical research in this area is sparse, then both UK and US work are discussed together. The US is particularly interesting in that default risk is fully insured for the Federal Housing Association, but not for the alternative conventional lenders. This has led to a number of interesting studies. For example, Duca & Rosenthal (1991) use time series data to explore the FHA/conventional loan decision, and the persistence of equilibrium rationing. Ambrose et al., (2002) analyse variations in default risk and the proportion of FHA/con-ventional borrowing by state to detect credit rationing. The latter study exploits the fact that the FHA is not allowed to vary its lending criterion on a geographical basis. UK research in this area is even more sparse but there is some tentative evidence of a separating equilibrium, where the choice of mortgage design might signal a group of borrowers that continued to be rationed after financial deregulation (Leece 2000b).

The extent to which non-price terms are used to offset default risk will depend upon the ability to vary interest rates according to each individual borrower's risk (Stiglitz & Weiss 1981; Riley 1987; Duca & Rosenthal 1994). This redlining (Stiglitz & Weiss, 1981), or rate sorting, will reduce the need to use non-price terms for credit rationing. In this case non-price terms will have little or no influence on observed household behaviour, for example the size of borrowing, tenure choice and life cycle consumption and saving. However, some studies have shown that non-price terms do impact upon household mortgage choices (Duca & Rosenthal 1991, 1994; Cox & Jappelli 1993; Perraudin & Sorensen 1992). Duca & Rosenthal (1994) analysed data on FRM rates (1981 to 1983) taken from the 1983 Survey of Consumer Finances and found no substantive evidence of red lining. The major explanation for this was the existence of fair lending laws constraining price discrimination.

A key issue is how far credit rationing attributable to variations in default risk is likely to persist after financial deregulation? Duca & Rosenthal (1991) conduct a time series analysis of the extent of credit rationing pre-and post-financial deregulation. They use the unique institutional feature of the US market that FHA loans are fully insured against default risk. When the likelihood of default risk increases overall in the national economy then any credit rationing apparent in the conventional loans sector will increase the take up of more costly FHA loans, but with some borrowers remaining credit rationed. The study found that from 1973 through 1987 originators of conventional mortgage loans used non-price terms of mortgage contracts to ration mortgage credit. The FHA partly offset this rationing effect. Linnemann et al. (1997) find some evidence of credit rationing after financial deregulation in the US, while Leece finds tentative evidence for this in the UK (Leece 2000b). This form of post-financial deregulation rationing is most likely to take the form of equilibrium credit rationing, though this still requires more explicit empirical verification.

Brueckner (2000) points to the payment of higher mortgage indemnity premiums on large loans as a form of price discrimination that might induce separating equilibrium. Such differential insurance payments are also characteristic of the UK mortgage market. Harrison et al. (2004) find evidence to support separating equilibrium based upon affordability criterion though credit rationing is not necessarily an outcome. The empirical aspect of the work is based upon a 'rich data set' which contains measures of the level of default costs of borrowers, and a variety of proxies for default risk. The data pertains to originations from December 1989 to June 1991 with a recording of default status up to mid-1997. The estimated regression model used a number of interaction terms (e.g. default costs8 multiplied by self-employment) and suggested that less risky borrowers borrow more. Given this challenge to conventional wisdom then this is an area where further work would be welcome, say in different economies for different points of the business and housing cycles.

Sub-prime lending is another area that raises challenges for the detection of credit rationing. The interesting questions in relation to the sub-prime lending market concerns the efficient pricing of the credit risk, and how far the self-selection on sub-prime loans overcomes problems of asymmetric information? Analysis of this market has revealed that third party mortgage originations lead to significantly higher default rates (see

LaCour-Little & Chun 1999; Alexander et al. 2002). Thus information asymmetry may still be a significant problem in mortgage markets where loans involve third party originations, that is sales by mortgage brokers. Lack of information on the part of borrowers can also lead to 'churning', with brokers encouraging prepayment and the unnecessary origination of new mortgage contracts. Alexander et al. argued that agency risk was not initially priced during the industry growth stage in the US, but that it is correctly priced now.

Mortgage brokers in the UK have also been a source of controversy, with sub-prime lending attracting increased critical attention, but no substantive academic study as yet.9 The question remains as to whether equilibrium mortgage credit rationing is evident in the sub-prime lending market, in the US or the UK? Ben-Shahar & Feldman (2003) noted the signalling effects of credit scoring and the possibility that within each credit scoring category there is a menu of mortgage contracts that screen borrowers. Thus asymmetric information and separating equilibrium with credit rationing might still persist, a theoretical issue that invites empirical research.

There is some evidence for the UK that the choice of mortgage instrument during the mid-1980s might have signalled which borrowers were liquidity constrained, and possibly continuing to be mortgage rationed (Leece 2000b). The repayment mortgage is a more flexible mortgage instrument than the endowment mortgage and might therefore have appealed to liquidity constrained borrowers wishing to vary their payment patterns via changes of maturity, etc. Leece estimated separate truncated regressions on samples identified by this mortgage type. The research made no pretence at being a direct test of the separating equilibrium outlined in Chapter 5, though it is indicative of the potential importance of self-selection and screening in the UK mortgage market.

Repayment (annuity) mortgage holders displayed behaviour consistent with liquidity constraints, that is they responded to changes in nominal interest rates, and showed no significant sensitivity to changes in expected house price inflation. Endowment mortgage holders were more responsive to user cost arguments. The estimation indicated that income, reflecting affordability constraints, was a significant explanatory variable for annuity mortgage holders only. The evidence tentatively suggested a group continuing to be credit rationed after financial deregulation. The results might also reflect credit rationing of non-housing finance but if borrowers cannot substitute cheaper mortgage debt then some mortgage credit rationing is at least inferred.10

There is much work yet to be done on determining the extent of equilibrium credit rationing in the major mortgage markets, where the development of sub-prime lending, mortgage instruments that allow flexible amortisation, and mortgage securitisation raise the question of its continuing importance. Equilibrium credit rationing is bound up with the explanations of default behaviour, to be discussed further in Chapters 9 and 10. Whether affordability issues drive default, or if default is explained by the option theoretic approach, is also discussed in those chapters. These issues reflect upon the behaviour of low risk borrowers, and whether they borrow more or less than they would under full information equilibrium. Empirical findings may vary by time and economy, and this area merits significantly wider and more in-depth investigation.

Was this article helpful?

0 0
Sell Your Annuity

Sell Your Annuity

Do you have annuity you dont want? Discover When is it Time to Sell Your Annuity? What can I do? Where can I get the money I need? I have an annuity, but I dont know that I can sell it. Is there a good time to sell my annuity? I already have a home improvement loan, but it was used before the roof needed replacing.

Get My Free Ebook

Post a comment