PMP Lessons Learned

3 weeks ago  •  By  •  0 Comments


Download PDF

Introduction
On September 1 and August 12, 2016, respectively, the MSRB and FINRA formally applied to the SEC to make mark-up disclosure on retail confirms the law of the land. In the 14 months since, BondWave has been actively engaged in providing a real-time Prevailing Market Price (PMP) solution to automate the required mark-up calculation and reporting function. BondWave has been in regular and on-going dialogue with all three regulators to ensure our understanding and interpretation of the complex set of steps required to calculate a PMP value.

We have also had many meetings with clients and prospects to discuss their needs and the applicability of our solution. As a part of this dialogue we always offer to perform “Proof of Concept” calculations. As a result, we have been engaged by a number of firms and have performed tens of thousands of these PMP calculations.

All of that practice and all of those discussions in advance of the May 14, 2018 implementation deadline has led to a number of important lessons learned.

  1. Negative Numbers Are Not Just Possible, They Are Probable.
    In performing these calculations over 50,000 times BondWave has found that there is a possibility of deriving a mark-up that suggests a firm has lost money on a trade even when they have not. A typical example involves a firm with no prior position that first buys a bond from one retail client before selling to another retail client. The first trade’s mark-up is calculated by traversing down the waterfall and can lead to negative numbers even in the case where the firm later sold the bond at a profit.
  2. The Clearing Firms’ Solutions Are Not Really Solutions, But They Will “Accommodate” Your Solution
    Most clearing firms’ trade processing capabilities are not engineered to accommodate a third-party modification of a trade record and most have decided not to perform the calculation themselves. Instead they will permit their clients to manually input a PMP/mark-up, or insist that the proper value is delivered at the time the trade record is delivered. This can be problematic for firms that either wish to, or are required by the rule to, perform full-day PMP calculations.
  3. Even Riskless Principal Brokers Will Need a Barrell (i.e. Go Down the Waterfall)
    Many firms that mostly trade on a riskless principal basis believe all their PMP determinations will occur at level one of the waterfall. However, as mentioned in the original MSRB filing with the SEC and in the subsequently published FAQs, if both sides of the riskless principal trade, or both sides of a bond that goes into and out of inventory in the same day, involve retail clients then the first trade will require a calculation at a lower level of the waterfall.
  4. Time is Not Your Friend (Systems Integration Takes Time)
    While solution providers like BondWave may have completed the work necessary to provide PMP/Mark-up calculations, the work of integrating the solutions with front and back office systems is a daunting one. The IT staffs at regulated firms have no open cycles and with year-end technology freezes right around the corner hard decisions to re-prioritize projects cannot be put off much longer.
  5. An Extension Will Not Come at the 11th Hour
    Any extension to the May 14, 2018 deadline must first be approved by the SEC. The SEC approval process is a time consuming, public process. So, if there is to be an extension it will be signaled well in advance of the current deadline. There was a five-month gap between the original rule change request by FINRA and the MSRB (August/September 2016) and when it was approved by the SEC (February 2017). Since it is likely that a shorter comment period would be needed for an extension request, perhaps a three-month window is more realistic. Therefore, at some point early next year, it will become too late to anticipate an extension.
  6. There Are Traps in the Calculations
    Without great care and a deep understanding of the edge cases and intricacies of the waterfall, it is possible to produce nonsensical results. The most dangerous level of the waterfall is probably Level 3, the similar bonds level. Without a well-thought-out definition of similarity any result is possible. Of course, the MSRB signaled as much in their FAQs when they counseled that firms can use a different, tighter definition of similarity for mark-up disclosure, than they would for best execution determination. FINRA was mute on this point in their FAQs.
  7. Exceptions Are Expected (But Don’t Get Carried Away)
    At the November 2nd MSRB Forum: Mark-Up Disclosure Implementation Conference the MSRB confirmed that they believe the waterfall is designed to always produce a PMP. They also confirmed that there should be an exception process for dealing with outliers produced by the waterfall. However, they stopped short of defining what constitutes an outlier. They also confirmed that it will be important to well-document reasons for determining an outlier so that firms can defend themselves against concerns of cherry-picking. Finally, without defining “too many” they stated that a process that produces too many outliers will need to be re-visited and modified so that it produces fewer outliers.
  8. Regulators Like Free Look Back Options
    Throughout this process both the MSRB and FINRA have been very open, accommodating and available to offer guidance and help with interpretation. However, they have also been very careful not to quantitatively define many terms. For example, there is no numerical definition of what constitutes a contemporaneous trade or what constitutes an outlier. This can seem maddening in a rule that is designed to produce a highly quantitative, highly specific outcome: the exact profitability on every retail trade. But it may also suggest a shift in regulatory philosophy. Beginning with the new best execution guidelines that replaced the “rule of 3 bids/offers” there has been a shift away from “rules-based” regulation and toward principles-based regulation. Principles-based regulation recognizes that hard and fast rules can have unintended consequences and potentially encourage some regulated entities to look for loopholes. By defining specific principles and allowing for “facts and circumstances” interpretations of individual cases, regulators and regulated gain a more common-sense approach, but it can come at the cost of certainty.

Conclusion:
It can sometimes feel like surgery with a blunt instrument. The concept of Prevailing Market Price (PMP) has been with us in the fixed income markets for over a decade. However, it was first introduced to assist with the post-trade determination of Best Execution. Firms used their notions of PMP to review and analyze customer executions to look for potential outliers that may be in need of price adjustments. Now PMP will be used to precisely calculate, and individually report, a firm’s compensation on each retail confirmation to which the rule applies. This level of precision was never contemplated by the original introduction of the PMP concept.

Be prepared for rule expansion. The solution that works for the initial implementation of this rule in May 2018 needs to be flexible and adaptable. Mark-up Disclosure and Market Transparency will likely expand over time. This will likely include additional retail trades that need to display PMP values and what constitutes contemporaneous. One should envision a time when all retail trades will require a PMP value. Moreover, as middle and back office systems improve, those that comply by use of end of day batch confirmations will likely need to adapt to real-time calculation. These systems will have to be automated with the ability to flag for review outliers as discussed earlier in this report.

The simple answer is not necessarily the long-term answer. Given the history of regulatory drift and market transparency expansion in the US Markets, we would urge some forward-looking analysis regarding solutions for Mark-Up Disclosure. The expedient and simple solution that “complies” with the initial spirit and intent of the rule, may not offer the best solution in the medium to long-term. A one-size-fits-all solution typically results in a one-size-fits-none solution over time. Adaptability, flexibility and a solution that can be customized and modified as the rule and enforcement of it evolves over time will provide the greatest value for those having to comply with it.