Abstract
This chapter focuses on the analysis of competitive processes and their applications to banking. The main goals of this chapter are to provide an eclectic framework within which to analyse competition in general and its applications to banking; to review a pivotal competitive process in banking that takes into account the dualistic aspect of banking (double competition and risk-return optimisation); and, finally, to review the distinctive problems in analysing banking. In essence, this chapter provides the necessary analytical background for further discussion of competition in banking in chapter 5.
[A]t a theoretical level financial intermediation is still not well understood, and competition among financial intermediaries less so. Vives (1991: p. 9)
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
It is worth noting that the notion of welfare developed in SCP is narrower than in the welfare economics arising out of general equilibrium (Tirole, 1994a). The IO literature follows Marshall in confining itself to partial equilibrium (the general public are consumers, not employees nor shareholders of enterprises), and its analysis of welfare implications refers to such ideas as the trade-off between static and dynamic efficiency.
Chamberlin (1933) defines a market in which there is a given number of similar firms producing goods that are to some extent differentiated, but in the same industrial grouping. Assuming that firms maximise profits and that entry and exit are free, price settles at a point equal to the average cost of production (in the long run). Average cost includes ‘normal’ profit, i.e., a return to capital just enough to keep the configuration of firms in the industry, so this is a long run equilibrium in the sense that there is no incentive for firms to enter or leave the industry. Hence, under monopolistic competition long-run equilibrium is where price equals average cost (Robinson, 1933).
The ‘prices of production’ of classical economists, or Okun’s (1981) ‘customer’ prices.
For example, it is not necessarily the case that actions to block entry are inefficient, if entry accommodation is a prelude to tacit collusion (Milgrom and Roberts, 1982a, b).
Much of this is, of course, controversial. It is not necessarily the case that large firms are more efficient than small, if profitability is taken to proxy efficiency (Prais, 1976). But it may be true that efficiency in the development of improved products relies on the sort of investment which high price-cost margins accommodate.
In Leibenstein (1966), the distinction is made between — on the one hand — administrative, cost-reducing, or X-efficiency; and, allocative efficiency, on the other hand. Porter’ s (1988) critique of SCP, which leads to his own approach, focuses on the impracticality of the regulatory recipe.
SCP-based expositions implicitly assume that, other than in size, all firms in an (ill-defined) industry are identical in an economic sense; other differences — technology, products range, personalities, etc., are noise (Porter, 1988).
For a US summary see Berger (1995a). Representative examples of SCP work in relation to Europe include: Mooslechner and Schnitzer (1992), Molyneux and Teppett (1993), Molyneux (1993), Vennett (1993), Lloyd-Williams, Molyneux and Thornton (1994), and Lucey (1995).
It is notable of course that because of the McFadden Act (1927) US banking has remained far more fragmented at state (provincial) or even municipal level, so that the requisite structural information is available on a banking (as opposed to branching) basis.
This hypothesis is derived from the model of oligopolistic behaviour of firms which implies cartel arrangements are less costly to maintain in concentrated markets (Stigler, 1964).
Baumol et al. (1982) developed the approach as part of his expert witness defence, from a welfare point of view, of A.T.&T. in the anti-trust action which eventually (1984) led to the break-up of the company.
As Baumol et al. (1977) have shown that pricing can be expected to provide ju st enough price-cost margin to cover the overheads common to the industry.
In the two stage version the initial (Cournot) capacity commitment, reinterpreted as a Stackelberg (1952) lead, is intended to mould subsequent competitive behaviour.
General reviews can be found in Fudenberg and Tirole (1991) and Tirole (1994a). The early history of game theory can be traced through contributions by Cournot (1838), Bertrand (1883), Edgeworth (1897), Hotelling (1929), and Stackelberg (1934, 1952), to the classic work of von Neumann and Morgenstern (1944), and of course Nash (1950).
Porter (1988, 1991) argues that game theory suffers from uncomfortable assumptions, such as the heroic amounts of information utilised by all parties, simplistic strategy options and one-time games. Testing game theory is undertaken in abstract experimental situations and not in actual industries.
Among the concepts referred to in this summary are those of Bulow et al. (1985), Spence (1977, 1979) and Dixit (1979, 1980), and Fudenberg and Tirole (1984). The last provide a taxonomy of possible cases in which choice of (large-small) commitment by the incumbent can be determined, so accommodate the (enter or keep out) response by the potential entrant.
Essentially for the Milgrom and Roberts (1982 a, b) result a low-cost producer should not want to choose the equilibrium price that a high-cost producer would choose, nor a high-cost producer the equilibrium price that a low-cost producer would choose.
For further discussion see Tirole (1994a, pp. 374–6). Although there is an initial welfare benefit through lower prices, to depreciate the acquisition value of the prey, the overall welfare effect across all the stages of the game is more complex; and depends on discount rates. For a strong positive welfare effect, the sociall discount rate must be high, so that next-stage monopoly losses are small.
Stiglitz and Weiss (1981) apply Akerlof’s (1970) insight on adverse selection in a oneperiod model. While they do consider the effect of other banks on one bank’s rationing of its loan-customers, the core of the argument refers to the bank-borrower relationship. Diamond and Dybvig (1983) deal with an instance of the prisoners’ dilemma concerning depositors, and the fact that deposits withdrawn from one bank might, under stable money demand, shift to another bank is not developed. There is also earlier work on game-theoretic aspects of what financial intermediaries do by Shubik (1990, 1999).
Although there is no close parallel in the models, as in Milgrom and Roberts (1982a,b), the Broecker (1990) result derives from the fact that one strategist takes the other’s strategy as informative about cost (in this case credit quality).
In this regard the assumptions differ from Stiglitz and Weiss (1981) whose project-risks are non-diversifiable. Van Damme’s (1994) assumption, as he notes, disposes of the moral hazard problem associated with excessive risk taking.
The mathemat ics of portfolio theory are explored in detail by Huang and Litzenberger (1988: chapter 3). Other examples are Copeland and Weston (1988) and Bodie et al. (1996).
The fact that risk-return combinations can be changed by gearing a particular risky position up or down is the so-called ‘Two-Fund’ theorem of Fisher (see Tobin, 1958).
The following argument is inspired by Kim and Santomero (1988). Bank profitability goes up as the ratio of non-deposit to deposit liabilities goes down (since profits are earned on nondeposit liabilities or shareholder’s funds), but so does riskiness of the bank’s position.
This is widely illustrated, e.g., whether runs occur because bad loans are thought to be made, as in Postlewaite and Vives (1987), or do loans get to be bad because runs are thought likely to occur?
The reference is to Sir James Steuart, (1770, Book II, Chapter VII) who discusses the benefits of simultaneous competition in terms of what is sold and what is bought. “Double competition is, when, in a certain degree, it takes place on both sides of the contract at once, or vibrates alternately from one to the other.” In banking, deposits are sold to competing buyers, while loans are bought from competing sellers.
Yanelle’s (1989, 1991) tries to resolve the parallel debate of whether chicken or egg comes first — i.e., competition for loans then deposits, or the other way round. In practice, the complexity of the evolutionary process of intermediation is not perfectly understood.
This is highlighted by Vives (1991), Diamond (1984), and Leland and Pyle (1977).
Freixas and Rochet (1997: pp. 191–219) assemble a series of models that explain the whole bank run phenomenon.
This was indeed the policy concern of Diamond and Dybvig (1983), published shortly after the deposit insurance limit had been significantly raised in the US (in 1981).
Thus the BIS (1988) capital adequacy ratio promotes the ‘hiving-off’ of securitisable business (mortgages, credit card receivables, small-business working capital loans) to subsidiaries that are off the balance sheet.
See Kim and Santomero (1988), Davis (1992). and Dimson and Marsh (1995) — the latter referring, however, to securities houses rather than to banks.
Rights and permissions
Copyright information
© 2002 Springer Science+Business Media New York
About this chapter
Cite this chapter
Spajić, L.D. (2002). The Analysis of Competition and Applications to Banking. In: Financial Intermediation in Europe. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-1013-0_4
Download citation
DOI: https://doi.org/10.1007/978-1-4615-1013-0_4
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-5351-5
Online ISBN: 978-1-4615-1013-0
eBook Packages: Springer Book Archive