In the dark (pools)

The structure of equity trading has undergone a profound transformation in recent years, driven by the rapid rise of Alternative Trading Systems (ATSs), which now account for more than half of the US equity trading volume. Among the most prominent ATSs are dark pools - private trading venues that offer limited or no pre-trade transparency. Their expansion has coincided with a growing aversion among institutional investors to publicly displaying their orders on traditional, lit exchanges. There are several strategic reasons for this shift.

Foremost among them is the desire to minimize market impact costs. Large institutional orders, when displayed openly, can move prices against the trader's interest before the order is fully executed. By trading in dark pools, institutions can reduce this adverse market impact while often securing executions at prices within the prevailing bid-ask spread of the lit markets. Moreover, public order display exposes trading strategies to potential imitation and invites opportunistic behaviors such as “front running” - where other market participants exploit the visibility of an order to trade ahead of it, thereby eroding alpha and increasing execution costs.

The Regulation National Market System (NMS), enacted in 2005, inadvertently encouraged the proliferation of nondisplayed trading venues by allowing broker-dealers to internalize orders or route them through proprietary dark pools. In these venues, access can be controlled and transparency minimized, enabling a layer of informational protection for institutional strategies. Crucially, many of these dark pools have expanded their participant base to include high-frequency traders (HFTs) and the proprietary desks of the brokers operating the pools, introducing potential conflicts of interest and eroding some of the protective benefits once afforded by these platforms.

While dark pools executed around 14% of total US share volume in 2020 - up from less than 10% in 2009, by December 2024, this share had reached 51.8% (!), marking the first time on record that the majority of all trading in US equities was consistently taking place off-exchange. Moreover, in November 2024, the daily volume of shares traded on ATSs reached 1.7 billion, a 36% increase compared to 2023, highlighting a substantial rise in the use of these opaque trading venues.

It’s not difficult to grasp the risks posed by this situation. A market share this high makes one thing unmistakably clear: dark pools are no longer niche tools used by a handful of wealthy players, they are now a permanent fixture of the trading landscape. And that has serious implications. With half of all trades now occurring off-exchange, a significant portion of market activity is no longer visible in the public price discovery process. From this perspective, it is clear that there are - if not already, then inevitably - (negative?) consequences for how markets function and how prices are formed.

Source: Bloomberg

However, this market is going to look even more different and that’s because we witnessed the emergence of private rooms - gated spaces inside dark pools that restrict access to select participants. And with these private rooms, opacity has reached a new level. In these venues, not only are the orders invisible to the market, but so is the existence of the room itself. Only invited participants know what trades are taking place, and with whom. And this only happens if the trade is executed between both parties, otherwise the order remains invisible. That exclusivity appeals to brokers, hedge funds, and even retail-focused firms seeking to internalize flows and escape the increasingly competitive, high-frequency dominated environment of public exchanges.

However, this development in turn gives rise to a new kind of asymmetry. As the figure below illustrates, activity within private rooms is increasingly concentrated around just a handful of truly dominant ATSs. We observe that only seven of them handle a daily share volume exceeding 1 million - a sign of significant consolidation. Over time, this could lead to a reduction in the variety of trading pools available, effectively concentrating the activity of major players within a few private rooms while pushing others to the margins.

Another important consequence is the growing participation of retail brokers in these dark pools, particularly those operated by OneChronos and IntelligentCross. What does this shift mean for the market as a whole? A gradual erosion in the dominance of wholesalers. A good case in point is Citadel Securities. For instance, Wells Fargo Securities routed over 80% of its market orders to IntelligentCross in March of this year, a substantial increase from just 53% in March 2024. In contrast, Citadel Securities received only 2.5% of Wells Fargo’s order flow this year, down 8 percentage points from the previous year.

Source: Author’s representation

It’s important to make a clarification here: the rise and growing relevance of these private rooms should not be seen as setting up a direct, zero-sum competition between market-makers operating within these venues and traditional wholesalers. That’s because, in principle, even a firm like Citadel Securities could choose to participate in these private rooms (but it prefers not to because its infrastructure is strong enough to handle this type of transactions on its own). The real distinction lies in the terms of participation. A wholesaler is obligated to fill an entire order, without discretion. By contrast, in a private room, a market maker engages only if they choose to, and they can leverage their liquidity-providing experience to assess whether or not to continue interacting with a particular counterparty in the future. If the experience is unfavorable, they can simply opt out of future trades with that counterparty, a level of selectivity wholesalers don’t enjoy. Otherwise put, brokers handling orders in private rooms typically expect to execute the trade at the midpoint of the national-best-bid and offer (NBBO). If the order cannot be filled at that midpoint, it exits the private room and is routed back to broader dark pools, which still offer a degree of discretion but involve interaction among a larger number of institutions, avoiding those counterparties and private rooms in the future.

In a nutshell, why are these private rooms increasingly preferred? Well, the logic is simple: in a private room, a broker can control who sees and interacts with their liquidity. This minimizes market impact and helps preserve the value of proprietary strategies. Some brokers use these rooms for ESG- or diversity-driven mandates, matching orders only among minority-owned firms. Others use them to keep retail order flow away from dominant market makers like Citadel Securities or Virtu Financial, often hoping for better execution and price improvement.

However, such exclusivity brings risks. As these rooms grow, critics warn of market fragmentation, reduced price discovery, and rising concerns about fairness. Since public exchanges still provide the reference prices for most trades, pulling liquidity away from them weakens the mechanism through which prices are determined. Fewer orders on lit markets mean thinner books and potentially wider spreads, affecting everyone, especially smaller investors. Moreover, additional criticisms have been raised against private rooms due to the assumption that they generate phantom liquidity, especially given that all trades executed within a private room are reported only at an aggregate level by the parent company operating the dark pool. As a result, it becomes extremely difficult - if not outright impossible - for outsiders to determine the specific trading volumes of each individual room. These granular details remain known only to the participants within a given private room. As such, while operators of ATSs are subject to reporting requirements, the details of what happens inside these private venues are largely shielded from scrutiny. Allegations have surfaced that some rooms are quietly admitting market makers to soak up order flow, undermining their very promise of confidentiality and independence. Furthermore, this order flow can be used to implement the operator’s own high-frequency algorithmic trading strategies, as was the case with ITG in 2015.

Market participants, regulators, and watchdogs are taking notice. Billionaire Ken Griffin’s Citadel Securities has called just last month on the SEC to examine the rise of private rooms and their implications for fairness, efficiency, and competition (we can only understand why Citadel fights this much :-) ). Critics fear that, if left unchecked, this trend could recreate the old boys’ club dynamics of finance, where access to liquidity is based not on merit or pricing, but relationships and gatekeeping. Moreover, it also raises a crucial question, who can say what an asset is worth if no one knows what it is being sold for?

Regulators also worry that dark trading venues, particularly those relying on derived pricing mechanisms from lit markets, may base trades on stale prices, exposing traders to adverse selection. This risk becomes especially pronounced during periods of market stress or heightened volatility, such as the COVID-19 crisis in early 2020. Yet, somewhat paradoxically, dark trading has shown resilience and stability during such episodes, reinforcing its embedded role in contemporary equity market structure.

Ultimately, the growing reliance on dark pools and private rooms signals a deeper structural transformation in how liquidity is accessed and how risk is managed in today’s financial markets. While these venues offer clear efficiency and execution benefits - particularly for large and sophisticated market participants - their rapid rise raises fundamental concerns, especially around transparency and the future structure of market liquidity.

To be clear, I believe the evolution of dark pools and private rooms is, in many respects, both organic and even desirable. These mechanisms reflect the needs of modern trading environments. However, significant questions remain about what market liquidity will look like if these venues continue their upward trajectory and begin to dominate trading even further relative to traditional lit exchanges. At 50% market share, we can already argue that certain liquidity dislocations are taking place - it’s almost a certainty. But what happens when this share reaches 60% or even 70%? We’re about to find out.