Skip to main content

2017 EXAMINATION FINDINGS REPORT

December 6, 2017

Market Access Controls

Missing content block.

As trading in the U.S. securities markets has become more automated, the potential impact of a trading error or a rapid series of errors—caused by a computer or human error, or a malicious act—has become more severe. The SEC adopted Securities Exchange Act (SEA) Rule 15 c3-5 (referred to as the SEC’s “Market Access Rule”) to require broker-dealers with market access or that provide market access to their customers to “appropriately control the risks associated with market access so as not to jeopardize their own financial condition, that of other market participants, the integrity of trading on the securities markets, and the stability of the financial system.”13 For such broker-dealers, the Market Access Rule applies to trading in all securities on an exchange or alternative trading system, including equities, options, ETFs, debt securities (including municipals and treasuries) and security-based swaps.

FINRA observed firms that provide market access implement a variety of effective controls to help satisfy the requirements of SEA Rule 15c3-5, such as maintaining reasonable documentation to support thresholds; conducting periodic reviews that assess the reasonableness of thresholds (e.g., through a credit or capital utilization review); aggregating capital or credit usage limits by assigning finely tuned or granular limits, which in total represent a reasonable threshold, or by aggregating across applicable measures (e.g., accounts and systems) on a pre-trade basis; and establishing well-defined procedures that clearly describe the process to adjust a threshold both on an intra-day and permanent basis.14

Selected Examination Findings

FINRA observed several areas where some firms that provide market access fall short of their obligations under SEA Rule 15c3-5, particularly with respect to the establishment of pre-trade financial thresholds, implementing and monitoring aggregate capital or credit exposures, and tailoring erroneous trade controls.

FINRA also found that some firms did not appropriately apply the Market Access Rule to some or all of their fixed income activities. The Market Access Rule applies to any of a firm’s fixed income trading activity directed to an alternative trading system or exchange, including from a firm’s proprietary and principal trading desks, even if such activity represents a small percentage of the firm’s overall fixed income trading activity.15

  • Establishing Pre-Trade Financial Thresholds – FINRA observed instances in which firms failed under the Market Access Rule to establish reasonable pre-trade financial thresholds (capital and credit), or to undertake reasonable due diligence to substantiate those firm-assigned thresholds. For example, in one examination, FINRA noted that a firm assigned unreasonably high financial thresholds to its broker-dealer affiliate and was unable to provide any empirical data to support those thresholds. Certain single-trader IDs within the affiliate were assigned buying power of hundreds of millions of dollars and had a combined buying power of several billion dollars. The firm also lacked any substantiation of the reasonableness of those thresholds.
  • Implementing and Monitoring Aggregate Financial Exposures – FINRA observed instances where firms did not adequately consider capital and credit usage in the aggregate.16 FINRA also observed instances where firms providing market access lacked procedures on how to request, review, or approve adjustments to capital or credit thresholds. Often such adjustments were made on an ad hoc basis (e.g., in expectation of increased order flow in response to a market event, such as an index rebalancing) and not sufficiently documented. In some cases, the firm did not reset the adjusted levels or maintain documentation to support a permanent increase in the capital or credit threshold.
  • Tailoring Erroneous or Duplicative Order Controls – Striking a reasonable balance between preventing potentially erroneous or duplicative orders while not unduly inhibiting trading can be challenging. FINRA observed instances in which firms did not appropriately tailor their erroneous or duplicative order controls to particular products, situations or order types. For instance, firms use an “away from the market” control to prevent erroneous orders.17 However, relying solely on this control may put a firm at risk when entering large market orders, as there is no limit order price reference point. An effective practice that FINRA has observed to reasonably prevent erroneous orders of this type is to employ a market impact check, which measures the size of a customer’s order compared to the average daily volume in that security. If a check of this type is used, it should be set at a reasonable level.18

    FINRA also observed situations where a firm had not considered the character of the market at the time of order entry. For instance, firms that only used the “away from the market” control may have created issues at times when the NBBO may not have been indicative of the true market.19 When the NBBO spread is above a preset percentage, FINRA has observed that one effective practice to prevent erroneous orders is for the firm to establish an alternative reference point, such as a control that measures the order price as a percentage away from last sale as opposed to the NBBO.
  • Implementing Effective Fixed Income Financial Controls – FINRA observed that in some instances, firms were not implementing the required systemic pre-trade “hard” blocks to prevent fixed income orders from reaching an alternative trading system that would cause the breach of a threshold. These firms implemented either “soft” blocks that provided warnings, but did not stop (automatically or manually) orders in breach of a threshold from being executed, or post-execution controls. One firm’s systems permitted a customer to enter an additional order that breached the customers’ credit thresholds before imposing the hard block. In some cases, firms that initially implemented controls to address the rule’s requirements failed to establish market access controls as they added new alternative trading systems.
  • Reliance on Vendors for Fixed Income Financial Controls – Firms may rely on an outside vendor’s tools, including those of an alternative trading system, to effect their financial controls, but they must have direct and exclusive control over the mechanisms that have been established and remain responsible for compliance. However, FINRA observed some firms that allowed the alternative trading system to set capital thresholds for their fixed income orders instead of establishing their own thresholds. Occasionally, firms were not sure what their thresholds were, and had no means to monitor their usage during the trading day. Some firms failed to understand how their vendors’ controls worked and could not explain them to FINRA.
  • Effective Testing for Fixed Income Financial Controls – Firms also must periodically test their market access controls, which forms the basis for an annual CEO certification attesting to a firm’s controls. FINRA found that in some instances, firms either failed to conduct any tests at all for their fixed income orders, or relied on their vendors to perform the tests without appropriate due diligence by the firm.

End Notes

13 Exchange Act Release No. 63241, 75 FR 69792 (Nov. 3, 2010).

14 These procedures included details on the approval process (who has the authority to override or change a threshold) and the steps leading up to that approval. Firms retained clear documentation to support these decisions, and for instances where a limit increase was given on an intra-day basis, procedures that addressed the readjustment of the limit.

15 While the Market Access Rule defines market access as the entry of orders on alternative trading systems and exchanges, with very limited exceptions, nearly all fixed income market access occurs on alternative trading systems.

16 The challenge of considering capital and credit usage in the aggregate generally arose where firms assigned multiple account identifiers or provided services that could create points where thresholds could be multiplied without appropriate monitoring of the aggregate impact. Scenarios that can result in a firm unwittingly multiplying thresholds include those that offer an individual customer multiple trading platforms to route orders to market centers, provide sponsored access or the use of other market center specific controls, establish multiple trading accounts for a single customer, including LLCs (Master/Sub-Accounts), and assign multiple user IDs, monikers or other identifiers to a single customer.

17 An "away from the market" control is a measurement of how far above (buy order) or below (sell order) the National Best Bid and Offer (NBBO) an order is priced. A firm typically assigns a percentage above which an order will be halted.

18 For example, one firm set its control threshold at an unreasonable 500 percent of the average daily volume of the security.

19 The general nature of trading makes the premarket session particularly vulnerable to this scenario. During the premarket, participants' quotes trickle in and the NBBO spread narrows as the regular session opening approaches.