BORDERLESS TRADING CONFERENCE
At the Royal Institute for International Affairs
Clearing and Settlement
- the Barrier to a Pan-European Capital Market
Speech by Don Cruickshank
Chairman, London Stock Exchange
Thursday 26 April 9.20am
Good morning ladies and gentlemen.
The title of this conference is "Borderless Trading - Strategies for a Globalised Securities Market". This is certainly not the first financial conference in recent years to be linked to the theme of "globalisation", and not the last, because there are powerful forces at work ensuring that efficient globalised markets do not emerge!
Even within the European Union, and despite the endorsement by the Member States' governments of the findings of the recent Lamfalussy committee, there are many powerful players who benefit from inefficient capital markets.
Which is a pity, because an efficient European capital market would make a significant contribution to a thriving European economy. European investors spend up to 1 2 billion per year more than they need to on the direct costs of the same clearing and settlement services. If the volume of European transactions were to be run over an infrastructure that was similar to the DTCC in the US, the cost per transaction would be around 7 times smaller. European intermediaries spend more on capital - in the order of tens of billions of Euro more - to service those transactions, and that cost ends up with investors. European companies face higher than necessary costs of capital; a very small increase in the cost of capital imposes billions of Euro in extra costs on European companies - about 1 billion for each basis point increase - and I would judge that we are talking about several basis points. This cost has to be recovered somewhere, so goods and services are more expensive, and less internationally competitive. Finally, any inefficiency in allocating capital to ideas will further contribute to the lack of innovation and international competitiveness of Europe. Although the total cost to the European economy is hard to calculate it is likely to be significant.
Of course not everyone would benefit directly from capital market reform. Intermediaries would face stronger competition as would exchange organisations. But the only net losers are the organisations that own Central Securities Depositories (CSDs) who could see losses of up to 300 million euro and that's a problem as it creates powerful incentives to protect those revenue streams.
Clearing and Settlement
This morning, I want to focus on these clearing and settlement organisations. Not the most interesting issue, you might think. Kind of boring infrastructure like, say, telecomms access networks or cash machine networks.
But like many bits of apparently boring infrastructure there is more to clearing and settlement than meets the eye, and getting the provision of these network services wrong can have a significant effect on a much larger market - even the rest of the economy. The same is true of clearing and settlement.
It is my belief that getting clearing and settlement right for Europe is the most important problem facing European capital markets. The core problem that perpetuates the fragmentation of the capital market is the fragmentation of the clearing and settlement system. This is much more important than, for example, the apparent fragmentation between exchanges. Indeed, getting clearing and settlement right would allow effective competition between exchanges, within a single European capital market. Something that is sorely lacking at present.
It follows that mergers between exchanges would not be necessary to complete the single market. Indeed, given the positive benefits of competition in terms of innovation and efficiency, what Europe needs is effective competition between exchanges, not one single exchange. The fragmentation of clearing and settlement systems is a significant roadblock to achieving this.
This is why I welcome the recent initiative by DG Competition in looking at these issues, and the continuing interest being shown by DG Internal Market and bodies such as the Giovannini group.
I also think that the Lamfalussy Committee's recent conclusions point in the right direction, notably:
- objective-based Directives - vigorously implemented.
- enhanced co-operation and networking among EU securities regulators and a strong role for the Commission to ensure common implementation standards.
- no all-embracing European 'SEC' as there is no widely accepted constitutional framework within which it could operate.
- new measures to make cross-border capital raising and investment as easy as domestic.
And crucially,
- recognition that clearing and settlement is a key public policy area that the EU competition authorities should investigate as a matter of urgency.
A concerted effort to root out anti-competitive practice and structures, as well as national institutional reform, is needed to really crack these problems at the core of the capital market.
Know thy enemy
But if this is all true, and a bit of boring infrastructure is getting in the way of realising large benefits from integrating the European capital market, why hasn't something been done about it before now? After all, the problem is hardly new and the completion of the internal market was the last century's objective. To understand why, you need to know how the existing market works, the underlying economic characteristics of the problem and the idiosyncratic and, quite frankly, Byzantine, legal structures that have grown up in this area. Or, to put it more bluntly:
- the perverse incentives currently facing many existing market participants,
- the natural monopoly characteristics of clearing and settlement itself, and
- the legal barriers to effective competition.
Perverse incentives
First, the perverse incentives.
There are perverse incentives impacting on many of the market participants - including exchange organisations - who have invested in the current structure and the international providers of financial services - who have been forced to invest to overcome the market fragmentation and both of whom now risk those assets becoming "stranded".
My direct experience tells me that no industry - especially the financial services industry - likes to see valuable revenue streams eliminated by public policy intervention for the benefit of consumers. So expect some resistance to workable solutions from this quarter.
These incentives and the fuzzy governance of exchanges and clearing and settlement organisations make it harder for management to work towards the long-term interests of customers and instead focus them on the short-term benefits of exploiting inefficiency and fragmentation.
In addition, inefficient providers of services currently sheltering behind a fragmented clearing and settlement system have an incentive to resist change. Their current investments are under threat, not to say the easy life that many monopolists enjoy. No great hardship to be able to hide behind legal complexity, if not straight legal protection from competition. So expect resistance to workable solutions from this quarter as well.
An additional problem is that there is a serious lack of transparency. These costs are hidden to investors - bundled up with the other costs of the transaction - and to issuing firms raising capital.
And for the bigger transactions, the direct costs are only a small fraction of the overall total. Intermediaries are not particularly worried by the costs either. For as long as there are no realistic alternatives for investors, intermediaries can just pass on those costs to end-users - the investors themselves. And given that most of the costs are fixed - based on the number and not the size of transactions - they fall disproportionately on smaller investors.
Most exchanges - but not the London Stock Exchange - recover these costs through shared ownership of "profitable" clearing and settlement agencies. So the opportunity to cross-subsidise a potentially competitive service - exchange services - arises, and this is very handy to have.
All this explains why the market pressure to remove the barriers to an efficient European capital market has been extremely limited.
This is where the London Stock Exchange is uniquely placed.
Not just because London is the most international stock exchange and currently contributes a significant share of cross-border share transactions (2 trillion euro in 1999).
But also because our interests are firmly aligned with those of our ultimate customers - the end users of the system - crucially investors - and ...
thanks to Taurus ...
You probably didn't expect to hear a Chairman of the London Stock Exchange say that! I'll repeat it:
thanks to Taurus, the London Stock Exchange does not suffer from the burden of aligned ownership of the clearing and settlement layer.
Most other European exchanges, on the other hand, hold significant stakes in CSDs, most notably from my perspective Deutsche Brse with its 50 per cent stake in Clearstream.
Why do I say suffer - that they suffer - from the burden of aligned ownership, when those clearing and settlement agencies are such cash cows - with observed profit margins of up to 40% - for their affiliated exchanges and when their valuations are such an important part of their market capitalisation?
Well, because I believe that these are economic rents and as such are not sustainable and that their inevitable removal will do serious damage to those organisations which rely on them to underpin their current market valuations.
The vertical silo integrated through ownership is an unstable structure that cannot be relied upon - and should not be relied upon - to deliver an efficient market structure. Indeed, intervention is required to ensure that exchanges cannot distort competition in exchange services by leveraging their ownership of, or other close relationship to, clearing and settlement systems - in contravention of Articles 81 and 82 of the EU Treaty.
But in addition the side effect of the current silos is that cross border (or, more correctly, inter-system) transactions are very expensive, whether or not these prices are excessive in relation to costs. These high prices reinforce the silo, but they also induce investments by other market participants to overcome these barriers. Action by competition authorities can also impact on these areas. So an intermediary's business model that relies on the maintenance of inefficient inter-system clearing and settlement is also unstable.
Even if competition intervention cannot ensure the creation of a single integrated clearing and settlement system it will still undermine the business models based on inefficient and expensive cross border clearing and settlement. The indirect market impact is likely to be profound, and how market players react may well determine which ones thrive in the transition, and which ones don't.
A Natural Monopoly
Of course, an individual share transaction is most efficient if all three parties to the transaction - buyer, seller and company - link into the same CSD.
But because potential buyers and sellers are unknown, and because buyers and sellers want access to all potential sellers and buyers, efficient transacting requires that all shares of a particular company and all potential buyers and sellers of that security link into the same CSD.
However, individual investors tend to hold a portfolio of different securities. Given that there are fixed costs to hold accounts at CSDs, this means that they have an economic interest in holding their complete portfolio of electronic shares within the same CSD.
The combination of
- a single CSD holding all the shares of a particular firm, and
- investors wanting to hold their portfolio of shares in a single CSD
produces a market structure with only a single CSD which will tend to hold all the shares held by - and traded between - a large number of investors who invest within the limits of that CSD.
There are also likely to be pure economies of scale in operating CSDs. Like most computer-based systems, many of the significant costs are essentially fixed.
This is crucial. This economic analysis leads me to conclude that Europe is best served by a well-regulated single clearing and settlement system that would minimise market distortions and maximise the economies of scale and network externalities.
Legal Barriers
The legal framework for the relationship between the company, the CSD and the transaction is generally set by domestic company law and is different for each jurisdiction. And this legislation, coupled with opaque rules and regulations, acts to restrict further competition in the supply of these services and in doing so supports the structure of the vertically integrated silo.
It does this in various ways.
- Either by appearing to create a monopoly by saying that only one organisation can hold electronic share certificates in that jurisdiction (in the case of Italy and Spain) or to say that the CSD must be a domestic entity (in Finland and Germany).
- Or by limiting membership of the domestic CSD to prevent the establishment of more efficient links by imposing requirements for "equivalent standards of regulation and supervision" as a domestic entity. More subtle, this one. This appears to be the position in Germany and Spain.
Add to that an obligation to use a particular clearing and settlement system - either by primary law as appears to be the case in Spain or through the exchange's own listing rules as in Switzerland and France.
And differences in rules governing payment make it harder for CSDs to effect efficient links between them.
We recommend a thorough examination of these legal frameworks to root out potential breaches of Article 86 or even Article 81 of the Treaty.
For investors, the natural monopoly characteristics combined with the legal complexities create barriers to investing outside their 'home' CSD, as the costs of holding and trading these 'foreign' securities is much higher than holding 'home' securities.
And in respect of corporate actions, this may mean losing some of the rights of ownership that should flow to shareholders. In other words, companies, who face barriers to tapping into capital held by these investors, will tend to decide to hold their shares in the single CSD that gives them access to the largest pool of potential investors. As a result, not all potential investors are efficiently tapped.
This is not to say that investors cannot access electronic shares away from their 'normal' CSD; just that it's much more expensive. How much more expensive, it's hard to calculate. Part of the transparency problem, I guess. But our estimates suggest that the difference is a factor of five times the cost for a cross-border trade in which there is an effective electronic link. Up to a factor of thirty times where there isn't. Either way, pretty substantial.
The end result is that the capital market of Europe is seriously fragmented. Combinations of an exchange and clearing and settlement system listing and trading different securities, catering for a largely differentiated customer base, who choose to transact their business within, rather than between, these isolated silos is not good for Europe, and not actually in the long-term interests of those in the financial services business either.
Knock-on Effects
The monopoly characteristics of holding and transferring electronic share certificates in turn impact on the services provided by trading platforms.
As I have said, the purpose of a trading platform is to transfer ownership, and that efficient transfer requires the security, the buyer and the seller to be linked to the same CSD. It follows that a trading platform can only supply 'efficient' end-to-end transactions for those securities that are held in the CSD to which all - or at least most - of its members are connected.
In other words, the efficient scope of an exchange is significantly limited by the scope of the relevant CSD.
This means that exchanges, through patterns of usage, and reinforced where there is shared ownership, tend to support the silo model. This produces market structures where the dominant national CSD produces a dominant national exchange in a vertical relationship with a CSD.
In a world in which investors only want the shares in their "local" CSD and the companies only want "local" investors, these market structures - even though they are not exploiting the economies of scale that are available - may not produce particularly visible distortions.
However, once investors want to transact in securities outside this CSD and issuers want access to additional pools of liquidity - what we might call tapping into the European corporate market - the fragmented nature of CSDs becomes a barrier to achieving that objective and to effective competition between trading platforms.
It's an extremely complex situation but it needs to be resolved quickly because it lies at the heart of the inefficiency which is preventing completion of the single capital market.
And again, it appears that Article 82 may be being breached, and the Commission should investigate this as well.
What tools are needed to get from here to there?
So what can we do to expose the underlying economics of the situation and to eliminate the inefficiencies and reduce costs to investors?
It's at this point that you might expect someone with my background and beliefs to recommend a market with multiple competing clearing and settlement infrastructures.
But, for the reasons I have outlined, I don't think that such a market structure in the provision of clearing and settlement services for dematerialised securities offers the appropriate long-term remedy. The inherent dangers in creating a monopoly can be dealt with by effective regulation and are significantly outweighed by benefits of the monopoly capturing the economies of scale and the indirect benefits of enabling effective competition at the trading level.
As I have illustrated, that's not to say that there are no competition issues. But the economic characteristics are such that competition between organisations which carry out these activities is unlikely to be effective or stable.
This is perhaps a rather counter-intuitive conclusion, and given the perverse incentives on some market players to resist reform, the complexity of the differing legal structures, and the overall lack of transparency, devising workable solutions to this problem is no easy task. It is vital, however, to identify solutions that will work.
Possible Solutions: there needs to be one!
So what can actually be done to overcome the clearing and settlement inefficiency?
The starting point is the welcome investigation by DG Competition into the provision of clearing and settlement services in the EU. If, as the analysis I have presented suggests, clearing and settlement systems have natural monopoly characteristics then ensuring that market power is not abused is vital.
As for where the DG Competition investigation should end up, it's generally agreed that there are three potential solutions:
- a single CSD covering all the securities that (European) investors want to hold and trade;
- efficient interlinking of CSDs to reduce (even if they cannot be eliminated) the additional costs of trading securities between investors and intermediaries connected to different CSDs;
- all investors and intermediaries connected to all CSDs.
The last of these is likely to be impracticable, inefficient and expensive.
To decide between the other two it's useful to look at the US experience.
In the US - partly as a response to the slow process of reform in the system of vertically integrated silos and partly as a response to the mountain of paper that was swamping back offices, the US moved from a system of seven CCPs - each contributing a significant component of the revenues of the exchanges to which they were affiliated - to one CSD and one CCP.
The economies of scale are such that you could envisage all European volume going over one infrastructure, similar to the DTCC.
But it is also useful to look at the US experience to see why this is not likely to be achievable in Europe in the immediate future. There, the contribution of clearing and settlement systems to total revenues of their affiliated exchanges ranged from around 11% to over 50%. The existing vested interests were not particularly enthusiastic about the reform. A situation that is not too dissimilar from Europe now. Under these circumstances it took the power and weight of Congress and the SEC to force change, helped by a (reasonably) uniform legal, and language, structure.
So could such an approach be applied throughout Europe? Given the different constitutional and legal structure of Europe such a centrally imposed solution would seem difficult to achieve within any decent time-scale that the European economies need.
Our analysis is that a different approach is required, one that is consistent with Europe's existing market and legal structure and that there are two tracks to follow:
One track to address the immediate competition issues over a short time scale, the other to address the more fundamental structural inefficiencies of the market.
Track One
First, the immediate competition issues:
After an appropriate investigation, the Commission should be in a position to take action that will:
- enable efficient electronic linkages between CSDs, or at least between the major CSDs, Crest, Euroclear and Clearstream, to eliminate the need for the very expensive manual inter-system links currently used
- direct fair access to CSDs to create the services and interfaces necessary to facilitate corporate actions across multiple clearing and settlement systems
- limit restrictions on membership of CSDs
- ensure that there are linkages between CCPs
- ensure non-discriminatory access between trading platforms and clearing and settlement systems
- end undue discrimination in terms of price and services between inter and intra-system transfers
- and, crucially, prevent cross-subsidy between the provision of clearing, settlement and CCP services and the provision of exchange services.
Competition law can assist in facilitating these intermediate steps by focusing on current practices by vertical silos, which have perverse incentives to hinder the development of linkages between clearing and settlement systems and which allow the leverage of their monopoly power to benefit their affiliated exchange.
For example, by:
- tying the use of the exchange to access to the clearing and settlement system;
- bundling trading and clearing and settlement services, or offering "integrated" tariffs;
- discriminating between the affiliated exchange and independent trading platforms (or between their respective customers) in granting access and setting prices;
- charging excessive clearing and settlement prices.
All of which could be in breach of well-understood and robust concepts of competition law and in particular Articles 81 and 82 of the EU Treaty - leading to potential fines of up to 10% of world-wide turnover.
In vertically integrated companies generally, monopoly profits can be easily shifted across the business, thereby disguising cross-subsidies. As I mentioned in the US example, clearing and settlement can account for a substantial proportion of an integrated exchange's income. Likewise, Deutsche Brse Clearing (now merged into Clearstream) is the single, most important, revenue and profit generator for Deutsche Brse.
An examination of the publicly available information is not reassuring in respect of anti-competitive or abusive behaviour either. Some operating margins look very high, and prices for what appear to be very similar services vary significantly from one system to another, and are hard to relate to obvious cost differences. The pattern of efficient linkages looks a bit odd in relation to likely demand from end users. Some restrictions on who can do what, and what services will be supplied, are also difficult to reconcile to objective criteria. Of course, this appearance may all just be a side effect of the lack of transparency, and with better information it can be shown that all is well. DG Competition should get to the bottom of this - which is one good reason to welcome their investigation - and I await with some interest the results.
There is more surface level consistency with anti-competitive behaviour. For example, under the present circumstances it is difficult to see what incentives Clearstream would have to allow Crest to establish an effective link into Clearstream on fair and reasonable terms. To do so would create the possibility that some trades in securities listed in Deutsche Brse might be cheaper to clear and settle in the form of Crest Depository Interests, which, of course, would never do!
But all these changes will only lead to a distinctly second-best solution - a spaghetti bowl of interconnection links between national CSDs. Even efficient linkages will still mean that intra-system (ie domestic) clearing and settlement will be cheaper than inter-system (ie cross-border). Europe will still not reap the economies of scale that are potentially available. More importantly, there will still be knock-on effects into the capital market, making them less efficient.
This is why a second track of action is needed to tackle the fundamental structural issues.
Track Two
Let me say at once that successfully tackling these fundamental issues will not be easy. But it would, in my opinion be rash to rely on the spaghetti solution to produce world class efficiency in the European capital market, so the stakes may still be high.
The first element of track two should be to explore how the European Commission might actually impose a structural solution on the clearing and settlement industry - starting with the imposition of an enforced split between an exchange and the underlying clearing and settlement system. This would be to ensure that no exchange had control of any related clearing and settlement system, thus making it easier to ensure that any non-discrimination rule was easy to apply.
Achieving this would be easier as a result of action under track 1. Once non discrimination is applied properly, firms often find it easier to "voluntarily" break themselves up so as to isolate the monopoly, and free up their competitive activities.
Beyond that the Commission should explore if there are ways it could take positive steps to ensure the creation of a single clearing and settlement system. A co-ordinated effort would be required to persuade member states to align their company law enough to enable a single CSD to easily satisfy all the requirements of all member states.
Finally, the Commission should explore how it might actually force the move to require the future use of a single CSD.
Other action - that would help - would be for the Commission to consider if a European central counter party is so important that it would require open access to any CCP from day one. By treating such entities as if they were dominant, market forces can be harnessed that increase the probability that a single CCP is created, rather than continuing the existing fragmented structure as CCPs are created within existing silos.
Another action the Commission could take would be to encourage the increase in efficiency of clearing and settlement by benchmarking the charges for inter-system connections on the most efficient connections available, or even the efficiency of an integrated system. Such moves would undermine the economic viability of inefficient systems, which would be a powerful motivation for consolidation.
National governments will also have to play their part. National laws will need to be more aligned to make an efficient single CSD work. Not complete harmonisation, but a removal of national differences that make a single system hard to achieve.
Users - both intermediaries and investors - also have a crucial role to play in making sure that track two is successful and that Europe ends up with an efficient capital market. In the end it is they who will continue to foot the bill for the direct costs of the inefficiencies in the spaghetti solution.
For those users, the crunch comes when new investment is required to update outdated CSDs and clearing and settlement systems. Should they continue to throw good money after bad? Or should they insist that new investment takes place in systems that will provide an efficient, unified, European clearing and settlement service? In this context users are pretty powerful. A small number of big users provide most of the custom for the big clearing and settlement systems. And they all tend to operate in all the main systems. A bit of co-operative pressure - perhaps building on the valuable work of the European Securities Forum - could do wonders for the future structure of the industry.
Not an easy task. There are powerful forces that are resisting the integration of the European capital markets, and some of them will continue even after a spaghetti solution. There is, however, a bigger picture to be looked at and that's the future of the European economy, and that is a fight worth taking on.
Many of the levers are in place to create a more integrated capital market. All the more reason to expose those who exploit the current structure and protect their existing market power. A bit of "sunshine regulation" can work wonders. But in the end the really big question is whether Europe has the political will to improve its own economy, raise productivity and benefit its citizens even if that upsets some powerful, vested interests in the short-term.
The London Stock Exchange is well placed to contribute to the debate and to the solution. In this particular area we have no vested interest, and no perverse incentives. We would much rather face tough competition for exchange services within an efficient and growing European economy than exploit market power in a declining, fragmented market. Our customers have asked us to take a lead in this issue. This we intend to do, in the ways I have just outlined. I very much hope that others, with the long-term interests of the European capital market at heart, will join us.
Thank you very much.