INTRODUCTION
The regulatory environment in which the financial services sector finds itself today is intended to change how business is done. While MiFID II is just the start, there is a host of further regulation coming down the pipe – much of it aimed at creating greater transparency in the way the markets trade.
Under the revised Markets in Financial Instruments Directive (MiFID II), buy- and sell-side firms around the world are finding themselves in a very different trading environment when doing business within the European jurisdictions. In particular, their front offices are seeing trade execution and research being paid for in new ways, while the reporting of transactions becomes more challenging. The 3 January 2018 date has come and gone, leaving many firms still struggling with these details.
RESEARCH
1. MANAGING THE COST OF RESEARCH PAYMENTS
Unbundling the payment for research and execution into separate packages creates an entirely new revenue stream for banks, which also needs to be commercialised and pitched as a product. Ten years ago, many asset managers selected who to trade with based upon their portfolio managers’ collective assessment of their research function – through a broker vote.
Although there has been a gradual shift towards paying for trade execution, there has been very limited use of specific payment for research. Until now, the sell-side rarely priced research. This was mostly done by independent research firms who were paid via commission sharing agreements (CSAs), which shaved off certain commission monies to fund that research payment.
Now a price must be agreed and explicitly paid for, either from a pre-funded account set up for the client, or by the asset manager billing the end client via its P&L account. Only five investment firms have suggested they might pass on the cost of research to clients under MiFID II. Four other firms, who originally reported they would pass on the cost, changed their minds to take research costs onto their own P&L, a move which has proven to be popular with the majority of asset managers.
Whether or not it is paid for by clients or themselves, asset managers need to account and report on the cost of research. The reported charges for access to research, analysts and events are highly variable. Full access to the services is typically in the hundreds of thousands of dollars as an annual charge, reaching nearly US$500,000 for the upper levels of access, while specific asset classes can be anywhere US$10,000 to US$50,000. For example, Catalyst research has offered annual coverage for fixed income at between €5,000 and €10,0002.
At a process level, firms are incurring or passing on these costs and, therefore, are managing the cash flows associated with payment. This has posed several challenges, the first of which is the need to book the charges and allocate the cash to pay for them.
Buy-side firms that have set up research payment accounts (RPAs) are tracking what has been consumed, the cost and how it is charged. Managing individual accounts for clients is a more complicated task than paying for all research via a single P&L account. The need to track and report usage is also tied to the division of cost of research across multiple firms where it has impacted investment decisions across funds and accounts.
They are having to buy or build systems that can account for spend and assign it, while building a picture of where spend is occurring in order to spot efficiency gains that might be achieved. Whichever route they chose to take, the key is demonstrating transparency to their client base – a task that is harder for firms to prove, if they do not disclose payment to clients.
Taking the cost of research onto the P&L means that asset managers have a simpler cost management exercise, but nevertheless need to develop a system for tracking and reporting the charges and assessing costs.
2. REALISING VALUE
Any asset manager now paying for research needs to consider its spend at several levels – the fund, the broker, and the client – if the firm is to find ways to manage the cost of this previously unknown quantity. There may be savings to be found through the sharing of information or commercial pressures to be brought to bear upon uncompetitive pricing.
The last point also requires an assessment of value. As research has never been priced before, its value has never been assessed. Whilst the front office is not entirely new to the pricing of information, only recently has the ability to model its use become practical. As a result, firms have typically overbought licensing for market data in order to reduce the risk of finding a firm is under-resourced.
Where pricing and depth of market have very direct impact upon trading decisions, the effect of research upon investment decisions is less clear. It is a more qualitative process. Written documents can be digested by multiple readers without direct ownership.
This does not help the asset manager. They need to assess the value provided in order to drive down the cost of research through competitive pressure. Any system developed to look at the return on investment of research now has to take on a detailed and thorough analysis of what constitutes value, the value that funds generate through success and client inflows and the missing link – the value that the investment manager provides.
The latter point proves crucial to assessing the cost of research. Buy-side firms now need to assess their front office staff, the information provided to those staff and the decisions they make. At the same time, there is serious competitive pressure on active fund managers themselves from investors who are increasingly choosing to use passive investment funds.
Selecting stocks to go in an index based upon criteria, such as the capital weighting of stocks listed on a given exchange, rather than the estimates of a manager as to which may deliver future success, reduces the need for research and management, lowering costs significantly.
Active managers must assess their costs effectively in order to drive them down and deliver a competitive base, while at the same time optimising the flow of information in order to deliver better returns to investors.
TRADE EXECUTION
3. PROVING BEST EXECUTION
Transaction cost analysis (TCA) has long been used on buyside trading desks to understand the effectiveness of thirdparty brokers and buy-side traders themselves. Following the unbundling of trading costs and research payments, this is taking on greater importance as brokers are selected purely on the basis of their trading performance.
A challenge to the use of some TCA models has been its lack of independence. Historically, brokers offered both TCA and order routing systems. With best execution more clearly defined under MiFID II and offering greater transparency to end investors, it is incumbent on firms to demonstrate how they are arriving at trading decisions with their clients’ best interests.
Gathering the data for this requires firms to track and store trading reports in an automated manner, with the costs available in a form that can be interrogated and used to deliver information to the end investors who engage investment firms to manage their money. These reports consider the different nature of trading across multiple instruments and represent enough detail that the next issue firms face can be addressed; where trading decisions cannot be easily quantified, the logic behind the decision must be easily understood.
4. QUANTIFYING QUALITATIVE DECISIONS
MiFID I required equity trading firms to accurately assess the cost of their trades. However, ten years on, many firms still struggle to fully understand how much a trade costs, despite the explicit charges that equity trades are subject to. With MiFID II, firms are assessing trading decisions across asset classes for all listed and over-the-counter instruments; this is a significant challenge for the buy-side trading desk as there are multiple fees (from broker dealers, trading venues, and post-trade processors) to consider for each trade.
Trading across instruments is comparable because those instruments are fungible; one equity for a given company is the same as another equity for that company. The same is not true for bonds, where a single company may issue thousands of different bonds, which have different maturity dates and tenors.
For example, knowing that an IBM share traded for US$19 thirty seconds ago is likely to assist an investor in assessing IBM share value now. If an IBM bond traded under the same conditions on the other hand, it may not be the same bond that an investor was seeking to buy or sell. Moreover, where most shares are listed on an exchange which publishes the price at which stocks are trading, bonds are traded bilaterally, and in Europe no single source for pricing data exists.
The lack of fungibility between fixed income instruments make them more illiquid as the automated matching of securities or currencies that can occur in equities and FX is not possible. And as quotes must be requested from dealers to assess price, information leakage is a greater risk.
Consequently, the trading process for non-equity instruments is relatively complicated and other elements may prove more dominant in determining what to buy or sell at a given point, such as speed of execution or limiting information leakage.
TRANSACTION REPORTING
5. WHERE AND HOW TO REPORT
On 10 October 2017, the FIX Trading Community, which represents users of the front office, trade messaging standard, published its ‘Recommended Practices for Best Execution Reporting’. This 22-page document offered helpful guidance but also contained complex decision trees identifying who ought to report given trades according to where they were traded, the different counterparties and the regulatory status of those counterparties.
The reporting firm may be the investment firm, the trading venue, the counterparty or a systematic internaliser depending upon various factors. Getting this wrong could result in a failure to report, an issue which has seen many sell-side firms fined by the UK’s Financial Conduct Authority prior to MiFID II.
Given that some transactions were excluded from the wellresearched FIX Trading Community study – for example, buy-side to buy-side bond trades – because they lacked a standardised process for capturing trades3, it is clear that firms may still find the process difficult.
In a recent Tabb Group report it was noted by a European buyside respondent that “Transaction reporting is probably what someone will get fined for first. The regulators have made it very clear that there will be no forbearance on any screw ups around transaction reporting.”
Assessing whether or not reporting falls upon a firm is the first step. Compiling the data for this process – including information such as time stamping – and pulling it into a single system that can generate reports is likely to require coordination between the order management system (OMS) used within the firm, and other sources including external data providers and a data layer that can normalise the information, before it is formatted.
Then the report must be submitted and a confirmation checked and passed to all concerned parties in order to fulfil compliance checks.
CONCLUSION
The challenges posed by MiFID II are considerable for buy-side firms. There are very real risks of non-compliance at the early stages, but, more importantly, there is the risk that firms could be left behind as the commercial models for trading and investment change at a rapid pace. Having the infrastructure to track and manage costs and value will determine the capacity of those firms to evolve as the landscape changes. With such a heavy emphasis on cost of investment, finding efficiency will be key.
View a full list of references for this white paper.
