Mondo Visione Worldwide Financial Markets Intelligence

FTSE Mondo Visione Exchanges Index: 98,249.56 -633.34

Corporatisation of exchanges and central counterparties: the dark side will come

Date 25/06/2002

Ruben Lee
Managing Director, Oxford Finance Group

It's a brave new world. To be precise, a brave new corporatised world. The transformation of exchanges, and to a lesser extent central counterparties (CCPs), into for-profit firms has become both commonplace, and commonly accepted as desirable. As with all new visions, however, there are unexpected problems waiting to be discovered.

To date, the demutualisation and corporatisation of exchanges and CCPs has been seen almost universally as beneficial. For their previous owners, primarily the banking and brokerage communities, the corporatisation of exchanges and CCPs has been seen as a way of capitalising their long written-down investments in outdated infrastructure institutions. This is particularly valuable in today's era of shrinking revenues. To the exchanges, corporatisation has allowed them to become more nimble and efficient than their cooperative and non-profit predecessors, and to have access to the capital markets to raise the necessary finances -- supposedly. For their shareholders, the promise, as always, is of dividends and capital gains.

However, two aspects of the dark side of the corporate dream will soon come to light for exchanges and CCPs: monopoly and bankruptcy.

A CCP typically has elements of a natural monopoly because it benefits from a positive network externality which arises from two sources. The first is because the benefits of netting -- one of the key functions provided by a CCP -- are dependent on the number of traders using the CCP. The more traders that use a CCP, the more netting is likely to reduce the number and volume of trades that need to be settled. Once a CCP has been established in a market to deliver netting, all market participants are therefore likely to choose to use this CCP to net their trades over any potential competitor. In most circumstances, if participants already net most of their transactions through a particular CCP, the benefits of sending additional transactions to the same CCP in terms of the reductions in settlement instructions and volumes are likely to be greater than those achievable by netting the transactions through any alternative CCP.

A second reason why the operation of a CCP often gives rise to a network externality concerns the collateral that market participants are typically required to put up to support their trading activity via a CCP. The more assets that are cleared through a single CCP, and the more this CCP is able to offset margin positions in one type of asset against positions in other types of assets, the lower the amount of collateral that is likely to be required. This is because the risk associated with the combined portfolio is likely to be less than the sum of the risks associated with each of the individual positions, given any correlations in the returns of the relevant assets. Once a CCP starts to dominate clearing in one or more assets, it is difficult for new competitors to offer market participants similar reductions in collateral for this range of assets, while still employing appropriate risk-management procedures.

The network externality associated with netting and the calculation of collateral is a strong reason why the activities of a CCP may be provided most efficiently by a single supplier. The fact that there is only one CCP in a particular market should not, therefore, be viewed a priori as a sign that the CCP is acting anti-competitively. However, whenever a monopolistic organisation is in operation, the potential for it to exploit its position to bring about inefficient outcomes is always present.

The functions of trading and clearing in securities markets have historically been viewed as different levels in the vertical structure of the markets. An important and controversial question that has recently arisen is what is whether exchanges, namely the providers of trading systems, should also own CCPs, thereby creating what have been called 'vertical silos'.

Various benefits have been put forward in support of vertical integration between an exchange and a CCP. First, vertical integration may let a trading system platform be directly linked with a CCP, so that trades can be electronically matched and routed to the CCP's clearing system. Such straight-through-processing reduces operational risks by decreasing the manual processing of trades. Second, if an exchange owns a CCP, the exchange may benefit by obtaining a source of revenues from the CCP that is not directly correlated with those arising from other sources available to the exchange. The combined revenue streams may be less volatile than the revenue from the exchange's other businesses. Third, there may be economies of scope between the different activities undertaken by the exchange and the CCP. The extent to which these arguments are true can only be determined in practice.

However, if exchanges do operate both trading systems and CCPs, it will be unsurprising if some of their behaviour is anti-competitive. By owning a CCP, an exchange has the opportunity to exploit a monopoly to its own advantage in a manner that inappropriately restricts competition by other exchanges and trading systems. The demutualisation of exchanges means that they are more likely than before to seek to take advantage of any monopolistic power that owning a CCP might give them, compared to the previous situation where both exchanges and CCPs were operated for the most part as non-profit mutual organisations.

There are many ways in which a securities exchange may inappropriately seek to exploit the monopoly power of a CCP, if it owns one. It may seek to cross-subsidise its trading system by using the profits it obtains from its CCP. The possibility of doing this may be increased by the difficulty of distinguishing the costs of clearing from the costs of trading in a vertical integrated organisation. The ability of trading systems without access to such cross-subsidies to compete with an exchange supported by revenues from a CCP may be limited.

An exchange may also restrict access to its CCP to other competing trading systems. In order for netting to be viable it is necessary that positions can be off-set against each other in a clearing-house, or be fungible with each other. Without such fungibility, no netting is possible. The extent to which market participants will be able to net any positions they take on different trading systems is therefore dependent on whether these trading systems have access to the relevant CCP. If, for example, one exchange owns the CCP on which most clearing is done, and restricts access to this CCP by another competing exchange, market participants will not be able to net any trades they execute on the second exchange through the first exchange's CCP. The ability of the second exchange to compete with the first exchange will therefore be reduced.

A parallel for exchanges that has been widely drawn is that they are like national airlines -- every country believes it needs one. This correspondence has been given an added piquancy following both the rapid decline of Swissair, and the money the Swiss government decided in the end to put up effectively to restore its name. This event highlights another downside of the brave new world of commercial exchanges -- namely the possibility that they may go bankrupt.

Many exchanges have closed in the past, almost always due to a lack of liquidity on their trading platforms. However, in today's corporatised environment, exchanges have a much greater incentive than before to diversify their revenue sources and maximise their profits. If an exchange delivers many other services in addition to simply providing a trading platform, it is possible that losses from these other services may force the exchange to close, even with a liquid market operating. It would be a brave government that did not bail out such an exchange.

The likelihood of government money being extended is significantly greater if the exchange in question owns a CCP. The operation of a CCP goes to the heart of a country's payment system, and any government will therefore be extremely loath to allow such an institution to go under. The notion that it might be possible for an exchange to be allowed to go under, while still ensuring the financial sustainability of a CCP that it owns, assumes that sufficiently tight financial firewalls can be created around the CCP for it to survive, even if its owner fails. Given the general difficulty of supervising financial conglomerates, this is a notion that few governments will be willing to put to the test.

This will lead to a moral hazard problem. The management of for-profit exchanges are paid, and will seek, to take risks in order to maximise the exchanges' profitability. However, they will know that any serious mistakes they make are likely to be supported in the end via government subsidy. Exchange managements are therefore likely to take excessive risks.

It is true that the traditional ownership models for exchanges and CCPs, namely those of the cooperative and the non-profit organisation, had significant problems. To overlook the possibility that today's demutualised and corporatised world could also lead to significant problems for exchanges and CCPs is, however, only to see the bright side of the dream. The darker realities of the corporate world cannot be ignored.

Ruben Lee is the Founder and Managing Director of the Oxford Finance Group. The views presented in this article of those of the author, and not necessarily those of the Oxford Finance Group or any of its clients.