2 May 2007

From 1 November, the Markets in Financial Services Directive (MiFID) will give investment firms more freedom in providing services across EU borders. But firms must also adhere to rules on client classification and best execution. We asked eight experts about the issues facing companies as the deadline for preparation draws nearer, and who will be the winners or losers in this new competitive environment.

What issues are most likely to delay preparations for MiFID?

IW: The areas where we're seeing most work done are client classification and best execution. For some European customers, client classification is a new concept – these banks have not been required to classify customers and protect them accordingly, so for them it will involve changes to business processes, customer information management systems and increased system integration.

Many institutions that have client classification will need to 're-paper' their customers to put them into retail, professional and counterparty groupings. This requires document management and workflow to track communications with customers and maintain a repository of documentation.

RT: The UK financial services industry believes it is on track for transposition by the end of November. Some other European countries are on track while others may have delays of around four months. There is enough information for firms to get pretty close to compliance, apart from a couple of questions, such as who should be covered in commodity derivatives. There is no consensus from regulators in terms of normalising the different formats for transaction reporting, and best execution among dealer markets is not yet entirely clear.

Richard Barden: Best execution will be difficult to implement quickly in countries where orders have traditionally gone to a single exchange. This is more to do with changes in trading culture than technology. New data sources will also cause issues. It's not yet clear who will be data aggregators and who will be data distributors. There are deep commercial issues for financial firms, such as whether data will genuinely be made available on a level playing field basis, along with cost issues like how many venues will need to be connected and what communications or development costs this will entail.

IW: Before MiFID, executing a trade on the local exchange was considered best execution. Now, best execution will be more difficult to achieve and report.

BS: While the regulation is being finalised, MiFID's implications for individual businesses need a lot of thought, time and commitment. Some firms haven't committed enough resources to fully translating the regulation into business strategy, how the strategy affects business processes, and how the processes can best be supported by people and technology.

Ralph Baxter: Delays will be caused by uncertainty over the spectrum of trades MiFID should apply to. There is also a tendency to let others take the lead in defining best and efficient implementation practices, so others make any costly mistakes first.

IS: Transaction reporting and passporting issues could cause delays – these present a significant challenge for firms with more complex cross- border corporate structures and trading operations.

PB: The devil is in the detail – January's confirmation and implementation of level three measures from each of the 25 European securities market regulators leaves just ten months for completing implementation. While the broad thrust of MiFID has been known since 1992, issues like the interpretation of best execution have been problematic.

Additionally, the highly competitive environment for institutional and hedge fund business is causing almost continuous re-engineering of order management, designated market area and algorithmic trading services. Basel II and the Risk Based Capital Directive involve a major overhaul of risk management and associated market and reference data systems. Firms are finding themselves short of implementation capacity amid all this change.

How can these issues be addressed?

Richard Barden: The Commission of European Securities Regulators and the EU need to address the issues faced by host country regulators. Best execution is primarily an educational issue, but timely delivery of the relevant technology is also key. For new data sources, the extent of integration work required will only become clear once all the systematic internalisers and multilateral trading facilities (MTFs) are known.

BS: Firms need to commit resources and recognise that MiFID is a strategic business challenge, not simply a compliance issue.

IS: Interim measures by the regulators look set to resolve some passporting and transaction reporting issues, but as it is not a level playing field in terms of individual country/regulator compliance, there may be a messy transitional regime for some firms.

Reclassification of clients, establishing an execution policy and notifying clients of changes can be addressed immediately. Firms must ensure they have the internal procedures, systems and data to support these policies.

PB: Firms are increasingly using offshore outsourcing services to boost their implementation capacity. There is also real interest in transforming legacy applications using business rules-driven and service oriented technologies such as BizTalk Server and .NET to reduce change costs and implementation times, and to make auditing easier.

IW: Microsoft is working with partners to help customers address these issues. Client classification can be achieved by implementing a document management and workflow system to track customer communication and provide document retention. We are also working with Singularity and COS to provide a complete MiFID customer classification and document repository solution.

Financial institutions are looking at providing best execution using two mechanisms. Pre-trade analytics will require that banks capture large amounts of streaming market data from multiple trading venues, coupled with speed of execution, cost of clearing and cost of settlement to determine the venue for best execution. MiFID will drive growth in algorithmic trading to achieve best execution. Many organisations will initially focus on post-trade analysis to prove that the customer received best execution. Best execution can become a point of differentiation for the bank.

What do you see as the most challenging aspect of the new directive?

PB: Internalisation and best execution appear to present the greatest challenges. Firms wanting to become systematic internalisers have the most to do. The best execution obligations will challenge firms and service providers while providing opportunities for benchmarking data, transaction cost analysis, algorithmic trading developments and better automated execution approaches, which have tended to become institutionalised. However, much work will be needed in the fragmented settlement and clearing processes in Europe, where a large part of secondary market trading costs remains.

IW: One of the most challenging aspects is in proving best execution for over-the-counter and fixed income markets, where no benchmark exists. In order to address this, we're working with SuperDerivatives, which provides benchmarking for the entire spectrum of derivative instruments.

Equities institutions will still have the problem of working out the reference price for a particular instrument, given the exchanges, MTFs and systematic internalisers that MiFID will spawn. Systematic internalisers have the additional problem of having to publish prices to the wider market to comply with MiFID's reporting requirement. Project Boat, which was put together by a consortium of banks in the UK, is addressing part of this issue.

Richard Barden: New venues generating market data present a major challenge: how many venues will have to be integrated? How much data this will generate? What are the communications costs and commercial terms for accessing this data? And what are the implications for existing infrastructure?

Equal enforcement of regulations is a major concern. The Investment Services Directive failed due to host regulators gold-plating regulations. This applies to overall MiFID regulations, but also in the application of commercial terms such as fair and equal pricing for access to market data from the new venues.

TY: MiFID's propensity to grind data warehouse operations to a halt trumps most other concerns. Corporations are already sinking under the weight of their own data, and are likely unprepared to handle the added burden of MiFID information. All the promises of MiFID could be rendered impotent when they enter the real world environment of the data centre. The solution to that lies in implementing a data management system that can analyse huge amounts of information quickly, instead of allowing it to overwhelm data infrastructures.

What major benefits will it bring for financial services firms?

PB: MiFID and other change factors are triggering re-engineering of front, middle and back office legacy applications as well as data management systems to incorporate modern technologies and business processing. In the long run this will benefit the larger sell-side firms that invest the most in new technologies.

Levelling the playing field for provision of secondary market trading services, especially the systematic internalisation regime, provides scope for big sell-side investment banks to provide lower cost execution services than those currently on offer from traditional exchanges and national central securities depositories.

MiFID is designed to provide benefits to investors through better transparency, best execution and increased competition in the provision of trading services. It will be interesting to see how things pan out in terms of more accessible liquid markets and cheaper executions.

IS: Firms will benefit from lower trading costs due to increased competition with and between the exchanges. MiFID has been a catalyst for firms to challenge existing trading mechanisms. While innovation has been led by larger firms, the wider market looks set to benefit from the downward pressure on trading costs – already, most leading exchanges have announced a review of their trade reporting charges. Larger firms will benefit from an ability to differentiate themselves on the basis of execution policy.

IW: Probably the greatest financial upside comes from being a systematic internaliser. Celent estimates that 72 per cent of the €1.6 billion revenue/savings will come from trading profits through internalisation. Some firms could make money from selling market data or running their own MTF. Banks that don't do this will still benefit from expected savings on market data and reduced exchange trading fees due to increased competition.

How can companies prove best execution in terms of technology and processes?

IW: Companies will have to adopt a best execution policy and share it with customers and regulators. This will define which venues they connect to and how they will route a particular order. Banks will have to monitor trades to ensure they comply with this published policy.

From a technology perspective, market data from exchanges, MTFs and systematic internalisers will have to be captured in real time. Price information, speed of execution, cost of clearing and cost of settlement will have to be maintained for the various venues. This will require market data feed handlers, time-series databases and reporting tools.

Richard Barden: In simple technology terms, best execution initially requires a sophisticated rules-based order management and routing application suite. This will allow differing client orders to be routed to the appropriate range of venues for execution, with each venue chosen on a per-order and per-client basis. Real-time integration to trade reporting and back office functions is needed so firms can dynamically create a seamless audit trail.

In process terms, best execution requires a firm to have a clear understanding of its clients, their categorisation and the instruments they have chosen to transact across. Using this and the MiFID requirements, clients can construct a flow chart process that combines client, asset type and venue. The process must be flexible enough to accommodate client movement between categorisation, instrument type and trading venue in real time on a per-order basis.

RT: The initial proof is in adhering to your policy. The challenge lies in proving every year that your policy is best execution. You don't need to do that on a trade-by-trade basis – you have to prove that on average, the policy is giving best execution.

IS: Firms must establish and adhere to their execution policy. From a technology and process perspective this involves the tools required to trade according to execution policy, and data archiving and reporting services that enable review of the policy and compliance with it.

Retail brokers using request-for-quote services have access to liquidity pegged to the underlying market price. This method of price discovery is easily audited given the right technology. As the market price becomes less centralised due to liquidity fragmentation, brokers offering retail quoting services are using more sophisticated technology, enabling them to offer a 'best' quote against the inside price across a number of markets for the same instrument.

Institutional brokers offering trading services must prove executions are in line with their execution policy. In exchange-traded instruments the market data required to prove compliance must be linked to the order and trade audit trail. As liquidity is fragmented brokers require a means of viewing and trading on the markets within their policy. The technology to enable interaction with this 'virtual market' is being deployed and refined to enable algorithmic execution against the firms' policies.

Will preparing for MiFID entail a high technology spend? Will the costs of compliance outweigh the benefits?

BS: A company's strategic vision will drive whether or not MiFID will entail high technology spend. If the benefits significantly outweigh costs, thin a high spend should be seriously considered.

IW: It depends on how the business sees MiFID – as a revenue opportunity or a compliance initiative. Our customers are evenly split between the two. For those that see MiFID as an opportunity and wish to become a systematic internaliser, sell market data and perhaps run an MTF, the cost of technology will be relatively high. Changes to order management systems, customer information systems, market data handlers and reporting will be required. For those that do the bare minimum – client classification, post-trade analytics for best execution and reporting, the costs will be much reduced.

Richard Barden: Generally, technology costs for best execution and data generation/consumption should not be excessive, as many required components can now be bought off the shelf. For smaller players, the costs could outweigh the benefits – they may not be able to internalise orders, so they will see little extra revenue and will not save much unless other trading venues are quickly forced to lower costs.

PB: In isolation, MiFID spend will be written down as a necessary business cost. It is one of many factors triggering investment in state-of-the-art technologies for trading, data management and risk and trade reporting systems that will stand bigger firms in good stead. More strategically it has caused bigger investment banks to fundamentally review the way trading services and market data are provided.

IW: The costs of complying can be minimised significantly if a suitable, flexible and scalable technical architecture is adopted. For many financial institutions, complying with yet another new regulation results in reinvention of the wheel. By taking a holistic view of regulation and using a service-oriented architecture (SOA) such as the .NET Framework, the costs of complying with new directives can be drastically reduced.

Are the costs of non-compliance high?

Ralph Baxter: It depends on how – or if – the regulations are enforced. The fear is that host regulators may adopt a lopsided approach to enforcement, rendering the promise of MiFID void.

PB: Firms face substantial direct and reputational costs if they are found to be significantly non-compliant. In an electronic age the costs of switching order flow are low, and sell-side firms that consistently provide best execution services across the range of prescribed instruments will be the winners.

BS: Continuing non-compliance would attract pressure from member state regulators, but any difficulty in placating them may be more than outweighed commercially by the reputational damage incurred.

Are there any potential clashes between MiFID and SEPA?

IW: These are complementary initiatives, and much of the business process and technical architecture developed for one can be reused for the other. By using SOA to build reusable components, much of the work done for MiFID can be reused for the Single European Payments Area (SEPA).

PB: This is a good example of the diverse European changes firms need to respond to. Following on from the introduction of the euro, SEPA has the goal of making European non-cash retail payments as easy as national payments, and impacts retail banks rather than investment banks. Once implemented, this could enable more streamlined retail trading in MiFID instruments, and firms will be able to compete on value added services while sharing a common payments infrastructure.

Who will be the winners and losers in the new competitive landscape?

IW: The biggest winners are likely to be large investment firms with the ability to become systematic internalisers, especially in continental Europe where internalisation is often prohibited. These firms will also generate revenue from selling market data.

The biggest losers are likely to be small and medium sized investment firms for whom the cost of compliance may be too much. They may get out of the business or outsource certain areas. Typically, they lack the scale to become a systematic internaliser.

Exchanges could suffer greatly due to increased competition and the demise of the concentration rule, which required all trades to go through the local exchange. Exchanges will lose market data revenue as well as trade fees.

BS: The winners will be organisations that see MiFID as a real business challenge and seek to use compliance with it for competitive advantage. They will aim to use increased volumes of MiFID-related data to add value to their business and not merely as a compliance necessity. MiFID is a huge competitive opportunity that should be strategically embraced at the highest level.

PB: We can expect to see much more competition and innovation in the provision of secondary market trading services now there is a level playing field among firms and established exchanges.

The move by big investment banks to provide an alternative to established execution venues is noteworthy. Big banks control the lion's share of order flow, so the liquidity barrier that dogged previous initiatives is not a problem. They can save money by systematically internalising order matching under MiFID. By operating as systematic internalisers and building a hub akin to the US national market system, they could link their liquidity pools and publish consolidated market information to meet MiFID best execution and transparency requirements in a compete/cooperate business model. The hub could be built using cheap and accessible technology and allow for other banks to join, leading to a global exchange anchored in the UK. Once critical mass has been achieved, the hub becomes the price reference point for the securities traded, and the consortium banks become net producers rather than consumers of market data. The arrangement avoids exchange transaction and reporting costs, and reduces settlement costs. This can only be good for issuers and investors.

Richard Barden: Market data 'plumbers' are also likely to benefit, including order-routing software suppliers, connectivity suppliers and market data platform providers.

IS: Competition with the exchanges is not a new concept, but MiFID makes it much more difficult to justify an execution policy that ignores alternative venues.

Participants

Richard Barden, head of business development, Tenfore.

Tim Young, director, Netezza

Ian Warford, industry director, securities and capital markets, Microsoft EMEA

Ian Salmon, Fidessa

Ralph Baxter, vice president of products, ClusterSeven

Brian Sentence, Xenomorph

Peter Bennett, business development and capital markets, HCL

Richard Thornton, partner, SunGard Consulting Services