Last time, we talked about fact versus fiction in today's markets. Specifically, we talked about the evolution of markets and what that means for market participants: improved transparency, greater access, and lower costs. But as the market has evolved, we've also seen increased complexity, which can have negative side effects.
Why? When a market is complex there is a greater chance for it to be misunderstood and have a suboptimal structure. In turn, this lack of understanding may challenge regulatory efforts as well as cause participants to interact with the market inefficiently. In addition, suboptimal market structure may lead to unintended interactions and systemic risk.
In the past we have focused on how increasing transparency, integrating technology, and identifying new risk controls can improve markets. Now it's time to take the next step and focus on improving markets by reviewing existing rules and simplifying anything that is unnecessarily complex. But, before we can do that, we must first increase our understanding of the marketplace's numerous moving parts otherwise we run the risk of implementing change that does more harm than good.
The good news is the industry has the tools to achieve that goal. Or, more accurately, the industry has the data to achieve that goal.
Our markets are not just driven by data—they are also incredible aggregators of data. Today's electronic exchanges are capable of recording every trade, new order, and cancelled order for future analysis. So it's logical that any action taken to improve markets should be data-driven. To ignore that information and blindly implement change would be misguided. Why make subjective observations when you can draw objective conclusions? Why make broad generalizations about trading practices when you can identify specific behaviors? Why guess as to what might happen in reaction to a change in market structure when you can predict what will happen by analyzing the data generated by pilot programs?
FIA PTG isn't alone in advocating for a data-driven approach to market analysis. National regulators like the Securities and Exchange Commission (SEC), the Commodity Futures Trading Commission (CFTC), and the industry's Self-Regulatory Organizations (SROs) are all moving towards taking an analytical approach to policy-making and enforcement. The SEC now uses MIDAS to better understand market structure and inform policy—as the SEC's Associate Director, Office of Analytics and Research, Gregg Berman put it, "…If we don't diagnose the problems correctly, we certainly won't arrive at the correct prognosis". The CFTC has access to Large Trader Reports and various reports from Swap Data Repositories and recently requested comments on how to implement a 21st century surveillance system (see our suggestions here). The SROs are constantly monitoring trading behaviors to prevent manipulative trading while ensuring their trading platforms continue to meet the needs of market participants.
Armed with this data, the industry can weed out those with manipulative intent and base any new regulations on participant behaviors. Take, for instance, the conversation regarding automated trading.
Automated execution methodologies are typically attributed to proprietary firms, hedge funds, and high-frequency traders (HFTs). But the truth is that today, most market participants utilize automated execution tools in one way or another—including banks, fund managers, and brokers. If new regulations are only applied to "typical" automated traders, they would fail to capture a significant portion of automated trading activity. What's even more concerning is that if we regulate based on participant type, rather than behavior, we may expose the market to the worst kind of risk—risk that was wrongly thought to have been addressed.
That's where the data comes into play. Regulators should use data made available to them by exchanges to understand how participants interact with markets, and the risks associated with that interaction. With that understanding we can work to remove unnecessary complexity from the marketplace and implement regulation based on behavior rather than participant type. In doing so we can ensure that nobody falls through regulatory cracks while also preserving the benefits of the market's recent evolution.
