July 16, 2009

I had an interesting discussion with members of the U.S. Treasury about equity market structure. They were trying to rationalize economic market theory with the messiness of the fractured, high-speed, low-latency, dark pool insanity that we call the U.S. equity market.

In my view market theory and practice are diametrically opposed. Market theory states that a single transparent market provides the greatest opportunity for price discovery, and best price is the goal; but in practice our fragmented mess of a market is about as far as you can get from a rational centralized market.

So where does theory break down, and why is our market so complex? The trouble is, all participants act in their own longer-term self-interests, which may not be aligned toward the short-term maximization of best price or rational market theory.

Theory says a central market aligns supply and demand to determine a clearing price. While this may be true where market infrastructure and communications costs are high, today markets can be quickly set up and connectivity is virtually free.

Theory states transparency is imperative. Experience tells us transparency is overrated. Transparency is great if you are small, in which case you would want to see all limit orders and place your order at a level conducive to being executed. If, however, you are a whale, the last thing you want is to be seen.

Theory suggests equal access is good. If the market is efficient, then all orders are equal and have no information content. But practice shows that some traders are smarter than others. The smart traders can predict direction, creating challenges for larger players who may not want to interact with predictive liquidity.

Theory postulates that, everything being equal, a large order should be executed more quickly as timing delays and impact costs tend to increase the longer your order is in the market. Practice poses: What if you can't find the other side? In a post-credit crisis world, liquidity and especially bridge liquidity (market maker) is more temporal and challenging to find.

So what happened to theory? Demutualization, transparency, access and decimalization have killed the classic market. Markets are no longer owned by participants and are in business to generate shareholder value. As such, a market's revenue is the broker's cost, pitting broker against exchange and the proliferation of dark pools.

Making Sense of the Market

Decimalization, speed and access have made it more difficult for traditional brokers and market makers to take risks and make markets. Post-credit crisis, banks have reduced risk capital and have priced credit higher, making it more expensive for trading desks to take positions and reducing traditional brokers' market-making facilities. In addition, in a transparent and fast market the quickest and lowest cost can set the tightest markets. While the brokers have scale, "efficiency" and "flexibility" are not words typically associated with the largest banks/brokers. Align these two issues and we see that the bulk of bridge liquidity today doesn't come from traditional big-name brokers but from a host of smaller, more nimble, less regulated electronic liquidity providers.

As traditional brokers increase the cost of bridge/block liquidity, and liquidity provisioning migrates to micro-/millisecond-enabled electronic market makers, traditional buy-side firms are trying either to break up their whale orders into minnows so as not to be harpooned or to hide their large orders in the dark to avoid being skewered.

That, however, creates another problem: How do you find liquidity if everything is dark? Indications of interest (IOIs) solve this challenge -- market centers, dark pools and algorithms are now sending liquidity notices among the venues to let other electronic platforms know where to find liquidity.

While individually all of these technologies sound nefarious and go against classic market theory, market theory unfortunately is just that -- theory. And while I know a lot of traders who would love to have a simpler and more structured market, at the end of the day it's about getting the order done. Like it or not, the current market works given today's state of technology, the cost of credit, and frictionless connectivity and processing power. Is it a perfect market structure? Probably not. But given the current state of the world and technology, I'm not sure there is a better one.