By Jim Toes, STA President and CEO 

Paul Atkins will soon testify before the Senate Banking Committee for his potential SEC Chair nomination, with no set date. In addition to facing questions on his regulatory philosophy, enforcement priorities, and the SEC’s mandate to protect investors, ensure fair markets, and support capital formation, Atkins will face more specific inquiries.

In this fourth of a multi-part series about the topics Mr. Atkins will most likely face, we highlight the evolution of the Consolidated Audit Trail, (“CAT”), a behemoth of a project spawned under a Regulation NMS Plan which has taken on its own life.

Backstory

CAT was conceived in the aftermath of the “Flash Crash” of May 6, 2010. This event saw a rapid, trillion-dollar drop in U.S. equity markets within minutes, followed by a swift recovery. The speed of modern trading across fragmented markets made it extremely challenging for the Securities and Exchange Commission to piece together what had occurred. This incident underscored the need for a more robust, centralized system to track and analyze market activity across different exchanges and trading venues, leading to the SEC’s proposal of Rule 613 in 2010.

It would take another eight years before the new surveillance entity, Consolidated Audit Trail, LLC would process its first trade report on November 15, 2018.

The primary function of CAT is to provide a comprehensive, centralized repository for tracking all trading activity in U.S. markets for National Market System (NMS) securities, which includes equities and options. CAT’s intentions are to enhance regulatory oversight and improve market surveillance, as well as to reduce the burden of data collection by streamlining the process for broker-dealers and exchanges by reducing the need for multiple, disparate reporting systems.

Despite its noble objectives, the CAT has experienced multiple delays, cost overruns, and several controversies, with issues ranging from technical challenges to governance and operational concerns. The complexity of integrating data from a diverse array of market participants has proven more challenging than anticipated. Expenses for the industry have been higher than originally projected. Market participants have also questioned whether CAT’s data will be effectively used by regulators or if it represents an overreach, potentially stifling innovation or adding unnecessary layers of bureaucracy to trading activities.

Reporting Requirements and Their Effect on Costs

The CAT data warehouse is the largest financial regulatory database in existence, with a scale speculated to rival the NSA’s data storage capabilities. This year, CAT-reportable events have averaged 50 billion per day, with regular spikes reaching up to 65 billion—far surpassing initial estimates. While this surge partly stems from growth in equity and options exchanges and overall trading volumes, the primary driver is the SEC’s increasingly aggressive interpretation of what must be reported to achieve CAT’s objectives. Over the past five years, the SEC has adopted a stringent stance on defining CAT-reportable events and the scope of required information.

Revisit What Needs to Be Reported and How Quickly

The SEC is obligated to conduct cost-benefit analysis when issuing rules. However, the scope of CAT-reportable information has grown due to new interpretations of reporting requirements, often emerging without adequate cost-benefit analysis.

Recently, the industry welcomed the SEC’s exemptive order on February 10, 2025, which exempted firms from reporting certain personally identifiable information (PII) of investors to CAT. The SEC noted in its order that such data is ‘not necessary to achieve CAT’s objectives.’ This prompts a broader question: what other data might also be extraneous? One area ripe for consideration is verbal quotes provided by market makers on exchanges, trading floors, or over the phone. While this activity received a temporary exemption on July 28, 2023, that relief is slated to expire on July 31, 2026.

There are other temporary exemptive relief orders that are repeatedly extended for various reasons, primarily the requirements are so burdensome that no viable technological solution exists.

Quote updates by options market makers across more than 1 million strike series represent another area that warrants review from a cost–benefit perspective.

The cost-benefit analysis of reportable events should also consider required reporting timelines. Since enforcement isn’t conducted in real-time, why must quote information—valuable for detecting layering or spoofing—be reported by exchanges at or near real-time? Delaying this reporting by a few hours could generate substantial cost savings without undermining CAT’s effectiveness.

Balancing regulatory oversight, market efficiency, investor privacy, and cost management should be a top priority for any SEC Chair. As Mr. Atkins prepares to face the Senate Banking Committee, CAT stands as a critical yet contentious piece of his potential SEC tenure. CAT’s sprawling scope, escalating costs, and questionable necessity of certain data requirements—like verbal quotes and short reporting timelines — need to be revisited to refine it into a practical tool for market integrity.

By George Rosenberger, Linda Giordano and Jeff Alexander, Broadridge

Charting the Unknown Depths of Dark Liquidity 

For buy-side traders, executing orders in dark pools has long been akin to navigating vast, fog-covered oceans. Two counterparties may pass within reach of each other, unaware of the liquidity that lies just beyond their sight—like ships passing in the night. The challenge is clear: without visibility into where liquidity is pooling, traders risk missing opportunities or executing under suboptimal conditions. 

Now, AI-driven advancements are acting as sonar, detecting patterns and mapping liquidity in real-time. These tools allow traders to pierce the veil of dark pool opacity, locating liquidity beneath the surface rather than relying on hindsight-based navigation. By deploying these capabilities, traders can move beyond traditional reactive strategies and proactively adjust their course to reach the best execution destinations with precision. 

The Enigma of Dark Pool Liquidity: Trading in Murky Waters 

Dark pools offer traders critical advantages: reduced market impact, anonymity, and the ability to execute large block orders without broadcasting intentions to the broader market. However, these same benefits come with costs: opportunity cost when counterparties are insufficient and the potential for the loss of post-trade transparency. While orders in dark pools are hidden and most dark pool transactions surface only as vague “TRF” (Trade Reporting Facility) markers printed to the tape, sophisticated market participants have the ability to deidentify dark pool fills and also execution mechanisms.  

Imagine navigating treacherous waters without a map—traders today are forced to infer liquidity patterns from past shipwrecks, relying on historical activity or prior fills to guess where liquidity may be lurking. This traditional approach is like scanning an old nautical chart instead of using real-time sonar—effective only when the waters remain unchanged. But when market conditions shift unexpectedly, these outdated maps leave traders stranded, unable to adjust their execution strategy quickly enough. 

The Limitations of Current Algorithmic Strategies: Drifting Without Direction 

Many algorithmic execution strategies today act like a dark pool sonde, dropping a probe into the liquidity environment to sense the current with a small size before committing the larger order.  Other algorithms are passive navigators, waiting for dark pool prints to appear before reacting. A trade must surface before algos can respond, triggering child orders only after liquidity has already been identified. But in fast-moving markets, both of these approaches are like a ship chasing a lighthouse beam—by the time the signal is seen, the light may be gone. This creates opportunity cost. 

The problem is twofold: 

1. Delayed Response to Dark Pool Signals – By the time a dark pool print is confirmed, liquidity may have already dispersed, leaving traders adrift in open water, chasing shadows of past activity. 

2. Overreliance on Historical Data – Traditional execution models depend on patterns observed in calmer seas, failing to adjust when liquidity shifts due to market sentiment, volatility, or breaking news. 

Without real-time adaptation, traders are left sailing blind, hoping that historical patterns hold steady—a risky proposition when liquidity is as unpredictable as ocean currents. 

AI-Driven Sonar: Charting a New Course for Dark Pool Execution 

AI-driven pattern recognition is transforming how traders navigate dark pool liquidity, much like advanced sonar scanning the ocean depths to reveal unseen structures. Rather than waiting for fills to materialize and reacting after the fact, modern AI continuously analyzes market signals, pinpointing where liquidity is forming in real time. This shift from passive observation to active detection enables traders to anticipate liquidity flows instead of chasing them, leading to reduced market impact, optimized execution, and improved alpha generation. 

But the real breakthrough lies in intelligent execution pairing. AI-assisted tools copilots that support trader decision-making can identify dark pool prints in real time and determine which execution strategy is best suited to access it, dynamically selecting the optimal algorithm for each opportunity. This precision-driven approach turns dark pool execution from a guessing game into a strategic operation, ensuring traders maximize liquidity access while minimizing costs. 

In this new paradigm, traders are no longer adrift in uncharted waters, hoping to stumble upon liquidity. Instead, they are navigators equipped with AI-powered sonar, charting a course toward more efficient, data-driven execution. Would you sail into unknown waters with an outdated map? Or would you rather use real-time AI guidance to seize the best trading opportunities? 

The answer is clear.