Opinion piece by Karl O. Strøm: In today’s financial landscape, a small group of powerful entities dominates stock and derivatives trading. These market giants not only handle the majority of transactions but also possess cutting-edge technology, including the fastest connections and most advanced trading software, likely powered by artificial intelligence.
These companies have gained a significant advantage by purchasing order flow data, granting them unprecedented insight into market dynamics. They can see individual orders, trader identities, positions, and accompanying instructions like stop-loss and take-profit orders. With real-time visibility across stocks, options, futures, and ETFs, these players have an almost omniscient view of what is happening. They are the dragons of the market.
This concentration of power and information raises two critical questions:
- How would you approach trading if you could see and sometimes influence the orders of most other market participants?
- Is there evidence in market behavior suggesting that this advantageous position is being exploited?
As a trader, I recognize the immense profit potential in such a scenario, both from leveraging the informational advantage and from potentially creating near-risk-free profit opportunities.
To explore these questions further, let’s examine some specific examples that might indicate the exploitation of this privileged position. But first we need to take a step back to understand what created this situation.
Payment for Order Flow (PFOF)
Payment for Order Flow (PFOF) has been a feature of financial markets since the 1980s. However, it only gained significant traction in the 2010s when companies like Robinhood used it to offer commission-free trading. This model soon spread, with many traditional brokers following suit to remain competitive.
Retail brokers often tout PFOF as beneficial to customers, citing lower or eliminated observable trading fees. Additionally, some brokers use PFOF to avoid the costs associated with maintaining direct connections (seats) at multiple exchanges.
On the other side of the deal are the buyers of the order flow. They are a select group of entities known as Market Makers (MMs). The motivation behind Market Makers’ willingness to pay for order flow is less immediately apparent. However, the fact that major MMs collectively pay billions of dollars annually for this privilege suggests a profitable arrangement. These for-profit organizations seem to thrive under this model, raising questions about the underlying dynamics.
Regulatory Landscape
Regulations governing PFOF vary considerably by region. The UK banned the practice over a decade ago, and most other European countries are following suit, with the EU set to implement a ban by 2026. In the USA, the world’s largest market by far, these arrangements are still thriving. While there is some discussion and movement towards stricter regulations, little concrete change has so far been implemented.
Regulators primarily aim to ensure transparency by requiring broker-dealers to disclose their order routing and PFOF arrangements. They also mandate adherence to “best execution” principles, which require orders to be transacted “at or between” prevailing bid and offer prices without undue delay, ensuring clients receive favorable terms. While these measures aim to maintain market integrity, the sheer scale of PFOF still raises important questions about its impact on market fairness and efficiency.
Why would someone pay to receive the orders from others? What position does it put them in?
Enter the Dragon pit
Imagine possessing user data from such a significant part of the market that you could, through extrapolation, predict the short-term behavior of all participants with near-perfect accuracy in real-time. This scenario is akin to being the combined force of Meta, Google, Apple, and Amazon in the financial markets. Yet, you remain largely undetected, like the mythical dragons guarding their golden treasure in hidden mountain caves.
This extraordinary position is held by three market makers that many of you have likely never heard of. Among these, one entity stands out as the undisputed leader. While specific names and market share figures are deliberately omitted here, curious readers can uncover this information through their own research.
It’s important to note that this isn’t an indictment of these entities; on the contrary, I find their operations quite impressive. There’s no reason to doubt their compliance with regulatory guidelines in executing customer orders. However, what often goes unexamined is the immense informational advantage they wield. By holding both individual and aggregated orders from such a vast portion of market participants, these market makers possess an unparalleled view of the financial landscape – a perspective that remains largely unexplored in financial literature.
The question that naturally arises is whether these market makers leverage this informational advantage. The truth is, I don’t know, and to the best of my knowledge, no one outside these organizations can say with certainty. However, being a long time participant in these markets myself it’s an intriguing thought experiment to play Devil’s Advocate. If I had been in possession of such extraordinary capabilities, what would I have done?
Let’s explore a few examples.
Example 1: Buy the inventory
US equity markets have regular trading hours (RTH) between 15:30 and 22:00 CET, but electronic trading of stocks open at 10:00. The liquidity in single stocks varies before the start of RTH, but it is possible to get things done. As for options the liquidity and availability are very limited outside of RTH. However, futures on the leading US indexes are the most traded financial instruments in the world for 23 hours a day. Liquidity is excellent, both during regular European trading hours and especially when US markets open at 15:30.
Now consider a scenario where market-moving “good news” is released outside of RTH. This could be surprisingly strong macro numbers or a positive earnings surprise from one of the giant “Magnificent Seven” companies.
In such circumstances, a large number of buy orders for individual stocks, options, and ETFs would likely be entered into the system by retail customers and larger players alike. Most of these orders would be set for execution at the start of RTH, resting in the systems of brokers, market makers, or exchanges until then.
Payment for order flow would result in a significant portion of orders ending up in the systems of our “Dragons” – the major market makers. These could be limit orders, market orders, or orders using execution algorithms like volume-weighted average price (VWAP).
That then? Being the dragon I would have bought much of the available inventory outside of RTH, both in futures and single stock. While monitoring the volume and price changes of incoming orders, I would continue buying until orders from other participants started to taper off. Then I would have sold this inventory to fill orders already resting in my own systems when RTH opened.
Example 2: Run the stops
Most traders use stops to protect their long- or short positions. Often these are entered together with the original buy/sell order, and the full bundle might even include a “profit taker.” All of these are resting in the systems of the market makers who have bought the orders. Using any form of technical or quantitative analysis, there will at times be situations where a large number of stop-loss orders are clustered at or around the same price. Think of a market that is near a previous bottom, have consolidated for a while between two price levels, or have gone strongly up for a while, but now seems to lose steam.
The market maker will see these clusters of stop-loss orders but with full visibility of the status of other orders they will also have a good picture of whether the market is changing direction or not.
If no clear change of direction seems imminent, but a substantial amount of inventory could change hands by triggering the stops, wouldn’t it be tempting to do just that? And remember that the risk is low. The orders are already resting in the systems of the market makers and can be executed fully within the parameters given by the customers and in a way satisfying regulators. When prices quickly move back to trend or back within range the inventory market makers bought by triggering the stops is unloaded slowly.
Would you do it? Being the dragon, I know I would.
Example 3: The Macro Whip
Another interesting situation can often be seen when important US macro news is released, typically at 14:30 CET. Regardless of whether the news is good or bad, there will always be a lot of traders taking short-term positions to profit from the expected directional move. Since prices change extremely fast once the news hits the tape, many traders place their bets minutes and seconds before the release and have a tight stop on their position. Others again wait for the news, then enter in the direction of the move.
Can the dragons eat the lunches of traders using both these strategies? Yes they can.
Remember that these players have full visibility of most of the positions that are already in the market, as well as their accompanying stops. They also have complete vision of new orders entering after the news, and the systems to react faster than all the other participants.
So what one typically see in such situations is an extremely quick price move in the opposite direction of the news, before an equally quick but often much stronger move the other way. The first whip triggers all the stops already residing in the systems, thus handing this inventory over to the dragons. The second move front-runs all the orders entering the market wanting to trade on the news.
This “macro whip” is actually a combination of the two previously mentioned moves, as it first runs all the stops, then buys the inventory in front of the primary news-driven wave of orders.
You might ask how this is possible? Well, a person cannot do it, but an AI can. Especially when the orders are already residing in the system. To satisfy the “best execution” demand from regulators, you only have to create an official price-print, then internally match the orders at this price. You don’t have to send them to an external exchange… unless it is in your interest to do so.
Who cares?
Can the examples above be explained by the usual functioning of the market, as Adam Smith’s “invisible hand” guides prices in an eternal dance of supply and demand seeking equilibrium? Yes, it can. There will always be a lot of different forces at work in the markets, with different time frames, tools and strategies. Prices will move as various actors seek do buy and sell their inventory, as it always has. Maybe omnipotent forces only exist in my imagination.
But giving very few players the power of trading with near full visibility of the other participants orders and positions raises a lot of questions. I find it more difficult to imagine that they would not use this information to their benefit, than to think that they do.
Having run a small hedge fund trading very actively with a turnover of some billions of dollars I have seen and felt the behavior mentioned above, as well as many other examples. It was being this close to the micro behavior of the market that made me notice strange things.
Holding Adam Smith’s invisible hand while dancing in the markets is challenging enough. Dancing with dragons is something else entirely.
And it makes me ask: Was this how it was intended to be?
