Written evidence submitted by David T
Llewellyn, Professor of Money & Banking,
Centre for Financial Markets and Institutions, Loughborough University
COMPETITION, CONTESTABILITY AND EFFECTIVE
COMPETITION IN BRITISH BANKING
EXECUTIVE SUMMARY
When considering the nature and structure of competition
in banking and retail financial services, three alternative concepts
are to be identified: competition, contestability
and effective competition. A market may appear to
be highly competitive but competition may nevertheless not be
effective in the market place. Even though there may be many competitors
in a market, competition is only effective in practice if consumers
are able: to make rational and informed choices between competitors,
and to exercise choice at low transactions costs. Both may act
to limit effective competition. Even if consumers are able to
make rational choices, the transactions costs of exercising this
choice may be high. Information is the major constraint to making
rational choices, and transactions costs to executing them. Combined,
they can have the effect of limiting the effectiveness of competition
even in a market place which is contestable or which has many
competitors.
Banking is not a homogeneous business but a conglomeration
of different businesses. For this reason, the focus needs to be
on sub-markets. Banks are involved with different customers, markets,
and products in each market. Competition takes place in sub-markets
rather than generically between firms. It is necessary, therefore,
to focus on the collection of narrowly-defined markets if the
true nature of competitive conditions in the banking industry
are to be understood.
Competition can be especially powerful when it develops
from outside the traditional industry rather than developing endogenously
within the industry. For this reason, competition would develop
to consumer advantage if entry barriers were to be lowered, and
if diversity in the banking system were to be fostered. In particular,
enhancing the role of the mutual sector in retail financial services
as is allegedly the government's aim) would be beneficial.
The degree of contestability in some banking and
financial markets has increased. Many of the traditional entry
barriers have been lowered partly because of the impact of technology.
It is evidently the case that the degree of contestability varies
considerably between sub markets. The impact of new entrants is
not measured by the market share they secure which may be quite
limited. The biggest potential impact is in terms of how they
force incumbents to behave differently. Although entry barriers
have declined in some banking markets, and as a result contestability
has increased, this is evidently not the case in all banking markets.
Entry barriers may be powerful in some banking markets which means
that excess returns can be sustained for incumbents, and anti-competitive
practices can be sustained.
NATURE OF
COMPETITION
1. When considering the nature and structure
of competition in banking and retail financial services, three
alternative concepts are to be identified: competition,
contestability and effective competition.
The first-mentioned focuses on the traditional way of considering
competition in an industry and considers factors such as the number
of suppliers, the market share of the dominant players (as, for
instance, measured by Herfindahl indices) etc. This dimension
also considers the type of competition in an industry: perfectly
competitive, monopoly, complex monopoly, monopolistic, oligopolistic,
etc.
2. In some industries (and in some banking markets)
the traditional way of measuring competitive conditions has become
less appropriate because of increased contestability. A
market is said to be contestable if entry and exit barriers are
low: ie it is easy for new firms to enter the industry at low
cost, but equally easy (and at low cost) to exit (Baumol, 1982).
In this case, incumbent firms are restrained in exercising monopoly
power (such as through high costs, prices, and profits) because,
if they were to do so, new firms would enter the market. In the
extreme case of perfect contestability, even a monopolist would
be forced to behave as if it faced many competitors because of
the threat of entry by others. It is the credible threat of entry
that deters anti-competitive behaviour by incumbents, irrespective
of the number of firms actually in the market at a given point
in time. In a contestable market the number of actual competitors
is largely irrelevant in determining competitive conditions in
the market. A later section argues that contestability has increased
in some, but certainly not all, banking markets in the UK.
3. The main purpose of this memorandum is to
consider effective competition. A market may appear to
be highly competitive (as measured in standard ways) but competition
may nevertheless not be effective in the market place. Even though
there may be many competitors in a market, competition is only
effective in practice if consumers are able:
— (1) to
make rational and informed choices between competitors, and
— (2) to
exercise choice at low transactions costs.
Both may act to limit effective competition.
4. There are many reasons why in some markets
consumers are unable to make rational choices. If consumers do
not have sufficient information to make comparisons and rational
choices, competition between suppliers may not be effective in
practice. For instance, some years ago independent research commissioned
by Abbey National revealed that more than half of bank account
holders were unaware that they could obtain higher interest rates
if they switched banks. We also find a wide range of prices and
interest rates for almost identical products offered by different
banks. There are also cases where the true price and costs of
a financial product are difficult to discern (and hence for comparisons
to be made) because of its complexity and the occasional practice
of obfuscatory pricing. To alleviate some of these problems the
Financial Services Authority introduced a "Treating Customers
Fairly" requirement on financial firms.
5. Even if consumers are able to make rational
choices, the transactions costs of exercising this choice may
be high. Clearly, consumers will incur the costs of "shopping
around" only to the extent that the expected benefits exceed
the costs and there are sometimes impediments in making this calculation
in financial services (Llewellyn, 2005). In an empirical study
of the current account market in the UK, Gondat-Lerralde and Nier
(2004) find that switching costs are a key determinant of competition
in the market. The bundling of products and services may mean
that the purchase of one service may be dependent on the purchase
of other services from the bank. For instance, loans to SMEs are
often dependent on the individual firm having many other services
(payments, etc) supplied by the same bank. The Competition Commission
(CC) focused heavily on the issue of bundling (which it defined
as "the supply of a product is made conditional on the take-up
of another product or obtaining services at a reduced price when
taking out another service") in its 2002 enquiry. There may
be considerable inconvenience attached to buying a particular
product from the "best" bank if, because of bundling,
many other services need to be switched. Switching costs may be
high. The CC found that "there is only a very limited degree
of switching of main bank accounts by SMEs. This is one of the
most important characteristics of banking services to SMEs".
Furthermore, consumer inertia may inhibit switching and the search
for the "best" deal, and in some cases redemption penalties
may be high especially when long-term contracts are involved.
There may also be costs in disturbing an existing relationship
with a bank because of the information advantages gained though
a long-term association. This relationship can work to the advantage
of both the bank and the customer.
6. Research by the Consumer Panel of the Financial
Services Authority found that, contrary to the evidence, 39% of
consumers believe there are no differences in costs and charges
between banks because there is competition between them (Consumer
Panel, 2001). Clearly, if this expectation exists the cost-benefit
calculations of shopping around are adverse. In fact, the evidence
is that there are significant differences between almost identical
products offered by different banks and even by the same bank.
Research undertaken by the FSA suggests that this is most especially
the case with short-term products (Cook, et. al., 2002). In an
attempt to enhance consumer awareness in financial markets, the
FSA now publishes tables of comparative information.
7. Information is the major constraint to making
rational choices, and transactions costs to executing them. Combined,
they can have the effect of limiting the effectiveness of competition
even in a market place which is contestable or which has many
competitors.
METHODOLOGY: MARKETS
V. INSTITUTIONS
8. A central methodological issue when considering
the degree of competition in the banking industry is whether the
focus should be on the total market or on individual banking sub-markets.
In the latter case the issue arises as to how to define (and with
what degree of fineness) the relevant markets. The general conclusion
here is that banking is not a homogeneous business but a conglomeration
of different businesses. For this reason, the focus needs to be
on sub-markets. Banks are involved with different customers, markets,
and products in each market. The emphasis on sub-markets is considered
in Bikker and Groeneveld (2000) and in Bikker and Haaf (2002b)
where they define a market as "all suppliers of a good who
are actual or potential competitors".
9. A central, though difficult, research area
in banking is how to measure and define competitive conditions
in banking markets and how to make comparisons between countries.
Several alternative methodologies (eg Structure Conduct Performance,
Panzar and Rosse Model, et. al.) have been applied in an extensive
literature. A theme here is that these conventional approaches
have become less appropriate because they are too aggregative.
Evidence from the UK, and the work of the Competition Commission,
suggests that the focus needs to be on micro banking markets and
that these markets might need to be defined very narrowly. Competition
takes place in sub-markets rather than generically between firms.
Unfortunately for research purposes, there is often a paucity
of data regarding sub-markets and the profitability of banking
business in these disaggregated markets. However, some research
has been conducted applying this approach. For instance, Courvoisier
and Gropp (2002) in a 10-country study found significant correlations
between margins and product-specific measures of industry concentration.
Similarly, Bikker and Haaf (2002) found that when focussing on
a national scale, the degree of market concentration of the largest
banks is often modest and they have only limited market power.
On the other hand, when the analysis is conducted on a disaggregated
basis large banks often do have market power.
10. There are several reasons why different sub-markets
need to be distinguished: competitive conditions vary between
sub-markets, consumer behaviour is different in each market (eg
with respect to the extent that a product is regarded as commoditised,
ie purchased mainly on the basis of price), the profitability
of different sub-markets varies, there are cross-subsidies between
different markets (where prices in some markets are set high relative
to cost and risk while in others they are set low), and the way
banks compete differs between sub-markets. In an empirical study,
Corvoissier and Gropp (2002) find that different banking products
are affected differently by competition and concentration: increased
concentration leads to less competitive pricing in loan and deposit
markets but not in savings and time deposit markets.
11. Analysis of competition in the banking industry
therefore needs to focus on sub-markets rather than regard banking
as a homogeneous business. As banks operate in sub-markets, profits
are generated in such markets and the overall ROE is based on
the sum or profits in sub-markets. This was the focus of two Competition
Commission enquiries where the conclusion was that competitive
pressures vary greatly between different sub-markets, eg between,
for instance, loans to SMEs and savings deposits for personal
customers. This analysis was applied in the adjudication of the
proposed merger between two banks. The CC rejected the bid by
LloydsTSB for Abbey National not on the grounds that it would
have serious negative impacts on competition overall, but because
it would have significantly diminished competition in two sub-markets:
SME banking and personal current accounts. And yet, in the wake
of the banking crisis, the government allowed a merger between
LloydsTSB and HBOS which would create equal, if not more, competition
concerns in the same areas as identified by the CC in the earlier
case.
12. It is necessary, therefore, to focus on the
collection of narrowly-defined markets if the true nature of competitive
conditions in the banking industry are to be understood. This
also means that profitability may vary between banks within the
same country (and also on average between countries) because of
the particular business mix of individual banks, and also because
there may be substantial differences in competitive conditions
between sub-markets with such differences themselves varying between
countries. It also means that aggregate concentration figures
(eg Herfindahl indices) may give a distorted and misleading indication
of market power. Thus, for instance, aggregate concentration in
the UK banking sector is comparatively low but is high in some
sub-markets such as SME banking products and services.
ENTRY BARRIERS
AND CONTESTABILITY
13. The nature of contestability was briefly
outlined above. While competition (as, for instance, measured
by the number of competitors) may be low overall, competitive
conditions may be strong in particular markets if they are contestable.
The key (though not the only) characteristic of contestability
is that entry and exit barriers are low. In practice, there are
many entry barriers into some banking markets (Llewellyn, 1999).
14. The degree of contestability in some banking
and financial markets has increased. Many of the traditional entry
barriers have been lowered partly because of the impact of technology
(Llewellyn, 2002). The process of deconstruction (whereby
component parts of banking products are supplied on an outsourcing
basis) also has the effect of lowering scale barriers, delivery
barriers, set-up costs, skill requirements, and the barrier of
integrated processes. In particular, the process of deconstruction
and outsourcing means that a new entrant is no longer required
to supply all the components within the value chain of a product
and, therefore, is able to concentrate on that part of the value
chain in which it has a potential competitive advantage. To the
extent that consumers have become more prepared and able to unbundle,
this potential entry barrier has also weakened. Equally, the development
and increasing sophistication of credit-scoring models has made
it easier for new entrants to enter some lending markets. The
development of technology has also lowered some traditional information
barriers to entry to the extent that it has increased the supply,
and lowered the cost, of some information needed to provide some
financial products and services.
15. More generally, the development of the Internet,
and its application to banking, has had the effect of lowering
entry barriers into some sub-markets in the banking industry.
In particular, it has lowered the marginal cost of transactions,
has made distance and location increasingly less significant,
has lowered consumer search costs, has increased the availability
of information, and has lowered the cost of price discovery. It
has also raised transparency which has lowered search costs for
consumers and raised the potential for consumers to make rational
choices between alternative offerings by different banks.
16. It is evidently the case that the degree
of contestability varies between sub-markets. While in new entrants
have entered some banking sub-markets, other markets have been
untouched. The CC noted that there are few suppliers, and high
entry barriers, in the markets of general-purpose business loans
for SMEs. It also observed only limited entry by suppliers into
the provision of liquidity management services, and in particular
current accounts with overdrafts and bank loans. In fact, some
past new entrants have withdrawn from these markets. On the other
hand, there is a large number of suppliers of some other banking
services to SMEs. There have been several new suppliers of Time
Deposit accounts and asset-related loans that have entered these
markets. As noted by the CC: "this shows the relative ease
of entry into these activities on a niche basis". There has
also been some entry into the supply of current accounts without
overdrafts. Equally, in some personal banking markets (notably
savings accounts) there have been several new entrants including
Supermarket Banks.
17. The degree of contestability therefore varies
considerably between sub-markets in banking. This can be represented
in a Contestability Matrix (Figure 1) where sub-markets are defined
in terms of customer groups and products. There is no scientific
basis to the precise numbers indicated in each cell (i.e. each
sub-market) which are impressionistic and given for illustrative
purposes only. Nevertheless, they may be a reasonable reflection
of the relative contestability of each market.
NEW ENTRANTS
INTO BANKING
MARKETS
18. Competition can be especially powerful when
it develops from outside the traditional industry rather than
developing endogenously within the industry. There are several
reasons for this: new entrants have different cost structures
than incumbents, they are more prepared to challenge traditional
ways of conducting business, the basic economics of the firm are
different, alternative business models are applied, they apply
a different business strategy and business mix of products and
services, and the incumbents often do not understand the business
models of the new entrants and hence may find it difficult to
develop competitive strategies against them.
19. In various countries new entrants into some
banking sub-markets have included supermarkets, motor car manufacturers,
DIY furniture stores, the Post Office, utility companies, insurance
companies, and even well-known football clubs! They tend to have
certain common characteristics: entry barriers are low in the
relevant sub-markets, new forms of delivery are utilised, and
only a limited range of products is offered rather than the full
range of products and services offered by conventional banks.
In addition, they are focussed within the value chain by outsourcing
a large proportion of processing, and have low fixed costs (partly
because they do not need to cover the substantial costs of establishing
a processing infrastructure), low costs overall, and no legacy
costs through dated IT systems and infrastructure. Furthermore,
they often enter the market in partnership with an incumbent bank.
20. The impact of new entrants is not measured
by the market share they secure which may be quite limited. The
biggest potential impact is in terms of how they force incumbents
to behave differently. For instance, faced with competition from
some new entrants, at one time Lloyds TSB offered a Current Account
paying interest of 3.5% against its norm of around 0.5%.
21. Although entry barriers have declined in
some banking markets, and as a result contestability has increased,
this is evidently not the case in all banking markets. Entry barriers
may be powerful in some banking markets which means that excess
returns can be sustained for incumbents, and anti-competitive
practices can be sustained. There are entry problems in four main
areas: (1) deterrence to entry, (2) cost and pricing asymmetries
facing new entrants, (3) the advantages possessed by incumbents,
and (4) the difficulty for new entrants to induce customers away
from long-standing incumbents with whom they may have had a long
relationship.
22. The context of the CC study on SME banking
was that around 85% of SME traditional banking business was at
the time in the hands of only six banks and the bulk was in the
hands of only four. The CC considered in detail how entry barriers
might operate in the SME banking market. In the process, they
identified several barriers of varying degrees of intensity: information
problems for new entrants about the credit-standing of SMEs; free
banking provided by incumbents to start-up firms creating pressure
on new entrants to offer it to the majority of their customers
whereas incumbents offered it to less than 20% of their customers,
and the scope by incumbents to negotiate with customers who are
considering switching. In addition, potential new entrants often
lack access to relevant skills such as credit assessment and relationship
management. An incumbent may also have economies of scope in the
provision of a range of services: in particular, the "first-port-of-call"
advantage.
ASSESSMENT
23. Our central theme has been that different
concepts of competition need to be applied to banking with special
attention given to "effective competition". It has also
been argued that competition can be particularly powerful when
it develops from outside the traditional industry. For this reason,
competition would develop to consumer advantage if entry barriers
can be lowered, and if diversity in the banking system is fostered.
In particular, enhancing the role of the mutual sector in retail
financial services (as is allegedly the government's aim) would
be beneficial This is argued further in Richie (2010).
REFERENCES
Baumol, W J (1982), "Contestable Markets: An
Uprising in the Theory of Industrial Structure", American
Economic Review, Vol. 67, pp 809-822.
Bikker, J A and Groeneveld, R (2000), "Competition
and Concentration in the European Banking Industry", Kredit
and Capital, 33, 62-98.
Bikker, J A and Haaf, K (2002a), "Competition,
concentration and their relationship: An empirical analysis of
the banking industry", Journal of Banking and Finance.
Bikker, J A and Haaf, K (2002b), "Measures of
Competition and Concentration in the Banking Industry: A Review
of the Literature", Economic and Financial Modelling,
Summer.
Consumer Panel (2001), Consumers in the Financial
Market, Annual Survey, Financial Services Consumer Panel,
London.
Cook, M, Easley, F, Ketteringham, J and Smith, S
(2002), "Losing Interest: How Much can Consumers Save by
Shopping Around for Financial Products?", Financial Services
Authority, London, Occasional Paper 19, October.
Courvioisier, S and Gropp, R (2002), "Bank concentration
and retail interest rates", Journal of Banking and Finance,
26, pp2155-2189.
Gondat-Larralde, C and Nier, E (2004), "Economics
of Retail Banking: An Empirical Analysis of the UK Market for
Personal Current Accounts", Bank of England Quarterly
Bulletin, Summer.
Llewellyn, D T (1999), The New Economics of Banking,
SUERF Study No 5, SUERF, Vienna.
Llewellyn, D T (2002), Technology and the New
Economics of Retail Financial Services, Building Societies
Association Annual Lecture, BSA, London.
Llewellyn, D T (2005), "Trust and Confidence
in Retail Financial Services: A Strategic Challenge", paper
presented at the Building Societies Annual Conference, Building
Societies Association, London, May.
Michie, J (2010), Promoting Corporate Diversity
in the Financial Services Sector, Oxford Centre for Mutual
and Employee-Owned Business, Kellogg College, University of Oxford.
|
