Worst of the good and best of the bad. Adverse selection consequences of risk pricing
| Published date | 01 December 2003 |
| Pages | 450-472 |
| Date | 01 December 2003 |
| DOI | https://doi.org/10.1108/14635780310508621 |
| Author | Gwilym Pryce |
| Subject Matter | Property management & built environment |
JPIF
21,6
450
Journal of Property Investment &
Finance
Vol. 21 No. 6, 2003
pp. 450-472
#MCB UP Limited
1463-578X
DOI 10.1108/14635780310508621
Worst of the good and
best of the bad
Adverse selection consequences of
risk pricing
Gwilym Pryce
Department of Urban Studies, University of Glasgow, Glasgow, UK
Keywords Pricing policy, Credit institutions, Credit management, Premium pricing,
Risk assessment
Abstract Why do lenders shrink back from full risk pricing in certain credit markets, even when
a sophisticated system of credit scoring is already in place? Fear of bad publicity is the usual
reason cited but this paper offers a complementary explanation which suggests that there may be
an underlying financial process driving such behaviour. The key proposition of the paper is that
risk pricing can cause adverse selection which has the potential to mitigate any positive benefits
such a pricing strategy may bring to the lender. This explanation is developed by introducing risk
pricing into the seminal Stiglitz and Weiss model and in so doing offers the first substantial link
between the risk assessment and credit rationing literatures.
1. Introduction
Risk pricing ± the practice of charging a premium to higher risk[1] customers ±
is common in many areas of finance because it has the obvious benefit of
helping to ensure that the expected revenues from lending to a particular risk
type exceed the expected costs. Thus, higher risk car owners pay higher
insurance premiums, and less financially secure borrowers face wider interest
rate spreads than their lower risk counterparts. For risk pricing to be effective,
however, the lender has to have a risk assessment procedure that accurately
allocates borrowers to the relevant risk category. The more refined the risk
assessment procedure, the narrower the risk bands that lenders can define, and
the more specific the interest rate that can be charged.
For most lenders, this process entails some form of ``credit-scoring''[2] where
each borrower is marked on a range of indicators thought to have some bearing
on default risk. An overall score is then calculated and used to place the
borrower in an appropriate risk group. Curiously, however, mortgage markets
(particularly in the UK) have been slow to fully implement risk pricing. Even
though many mortgage lenders have been applying fairly sophisticated credit
scoring techniques for a number of years, they have been reluctant to allow the
results of the risk assessment to feed through into differentiated interest rates,
choosing rather to use the information to ration credit by excluding the worst
risks (according to Brown-Humes, 1997, three in ten people who apply for
mortgages are turned away, for example).
What is the cause of the reluctance to price risk? The most obvious
explanation is fear of bad publicity. Risk pricing in most mortgage markets
The Emerald Research Register for this journal is available at
http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is available at
http://www.emeraldinsight.com/1463-578X.htm
Consequences of
risk pricing
451
would mean that a poorer individual in less stable employment would pay
more for the same house than someone who is well off and enjoying secure
employment. The implication? ``Those who are able to pay the most are
required to pay the least'' (Barnett, 1997, p. 6). The social ramifications are
heightened by the fact that employment and income brackets tend to fall along
racial and gender lines. Hence, risk pricing in mortgage markets could be
perceived as a form of class, racial or sexual discrimination, as some of the
negative publicity surrounding the issue has recently suggested (Barnett, 1997;
Kempson, 1996; Herbert and Kempson, 1996).
However, there may be an entirely financial explanation for the lack of risk
pricing in certain markets, not based on fear of bad publicity or social concern,
but on an objective, financial decision by lenders who seek to avoid deleterious
effects on the risk of their lending portfolio. It is the articulation of this
argument that is the main concern and contribution of this paper. Whilst a
formal mathematical model underlies the theory presented here, an explanation
is attempted without recourse to formulae. A more mathematical presentation
can be obtained from the author upon request.
To summarize, the paper argues that, under certain conditions, risk pricing
may cause ``adverse selection'', a term initially deployed in insurance theory
literature, but now incorporated into common economic parlance, referring to
any process that inadvertently increases the average risk of a lender or
insurer's portfolio. Written in full, the core proposition of the paper is that, by
differentiating the interest rates charged to each identified group of risks,
lenders may inadvertently worsen the average level of risk of loans in its
portfolio as a whole. And if the lender is aware of the possible adverse selection
effects, it will think twice about risk pricing. If lenders are unaware of the
possible adverse selection effects, then this paper offers a note of caution with
regard to risk pricing.
It is worth drawing the reader's attention at this stage to an already well
established finding in the theoretical literature, first put forward by Stiglitz and
Weiss (1981 ± henceforth ``S&W'') and discussed or assumed in a long list of
papers since[3]. In a pooled interest regime (that is, where all borrowers are
charged the same rate of interest) and where there is ``asymmetric information''
(lenders do not know for sure how risky a particular borrower is, though the
borrower does), S&W demonstrated that raising the rate of interest can cause
adverse selection (inadvertently attract bad risks and repel good risks). When
there is excess demand, text-book economics tells us firms will benefit from
raising prices, but this assumption may not hold in credit markets, argued
S&W, because of the adverse selection effect[4]. This consequence of raising
interest rates may provide lenders with an incentive to ``ration credit'' rather
than raise the interest rate. So even though borrowers would be willing to
accept a loan at a higher interest rate, lenders choose to keep interest rates the
same and either limit the amount of credit offered to each borrower or refuse to
offer any credit at all to particular applicants.
To continue reading
Request your trial