dark and dreary pools

Topic : Investment banks and financial institutions use hidden liquidity to transact
70% of global equities business. How do these dark pools of liquidity work,
why are they needed and what impact do they have on the 30% of the
market we can see?
Presenter : David Norman
Date of presentation : 15th February 2013
‘Dark pools, or dark liquidity pools, refers to the pools of liquidity and trading volume of financial
institutions that takes place outside public exchanges’ (Reuters, 2009). The ‘what’ of dark pools is
fairly easily explained: Unlike conventional exchanges where transaction prices and participants are
published, dark pools provide a market making facility where buyers and sellers are matched
electronically in anonymity. The ‘how’ of dark pools is less easily explained. Conceptually it is
straight forward; buyers wishing to buy a position are able to access a dark pool, bid and then have
this matched with a seller. Having said this, it is clear from, conversations with traders, journalistic
reporting and on reading Scott Patterson’s book ‘Dark Pools’, that those operating within the
industry, including traders, market makers and regulators do not always fully understand the
‘plumbing’ of dark pools or the implications for their continued expansion.
The extent to which ‘dark pools’ are needed is open to debate. Dark pools have arisen with the rise
of electronic trading and have provided protection for large institutional investors placing large
orders. In open markets, these ‘whale’ (Scott Patterson, 2012: 7) orders were easy targets for
traders who would ‘front run’ the order and affect the price the large investor would achieve for
their trade. Dark pools provide liquidity by electronically matching trades and in theory ensuring an
optimum price. However, there are serious concerns that as liquidity grows in dark pools it is
shrinking in conventional exchanges (Financial Times) and that a two-tier system is developing, with
rising costs that may close off the market to all but a few powerful investors (Financial Times, David
Norman, NY Sun). Furthermore it is Patterson’s view that, the forces that have created dark pools,
acting on the market as a whole have reduced the transparency of the market overall to the extent
that now ‘the market is dark’ (Patterson, 2012: 339). The 2010 Flash Crash has shown that the
implications of this darkness are significant and they will be explored in detail in this paper.
How dark pools work
A trading pool is a venue where buyers and sellers can be brought together to trade a position in
equities, foreign exchange and futures etcetera. This was traditionally done on the floor of an
exchange like the New York Stock Exchange and facilitated by market makers (Patterson, 2012: 7).
With the developments in computer technology this system began to be seen as largely arcane and
replaced by electronic trading.
Electronic trading can be seen as disruptive technology that has had an incredible impact on markets
(Neil Crammond, 2013). It has allowed the proliferation of a multitude of new venues for trading,
each competing with traditional stock exchanges as well as each other for trading volume. Neil
Crammond, a Director at day trading firm Futex, I spoke to, with over 30 years experience likened
the proliferation of new exchanges to the Tesco Express’ popping up in towns across the country
(Crammond, 2013). Neil began trading prior to the Big Bang in the late 1980s and has seen the
development of electronic and dark pool trading first hand. He noted that conventional exchanges
were, ‘making a lot of money for doing very little’ (Crammond, 2013). Electronic trading means that
Neil can now trade a stock like Vodafone in over 40 venues and at a trading cost much cheaper than
at a conventional exchange.
The move to off exchange trading has facilitated ‘dark’ trading. In traditional ‘lit pools’ market
makers would get a sense of developments in the market from their interactions with colleagues, for
investors with access to stock exchanges there was a reasonably level playing field (Patterson, 2012:
6-7). With the move to electronic trading platforms like Island and Archipelago (Patterson, 2012:
108 & 137) ‘Bot algos’ – computer programmed algorithms, would use published trading data to get
a sense of the market and design their strategies accordingly. To combat this, dark pools were
developed to allow trades to happen in secrecy. This prevented a practice akin to front-running
where bots would buy orders in front of a large buyer then resell to the same buyer reducing the
profitability of large funds.
Given that there are around 50 dark pools operating in the United States (Patterson, 2012: 40-46)
and approximately 40 in the UK (Crammond, 2013) and that up to half a dozen people per dark pool
will be privy to how the infrastructure of dark pools – its plumbing (Patterson, 2012: 6) is designed,
the number of people who truly know how dark pools operate must be limited to the hundreds.
Neil noted that until as recently as two years ago ‘no one knew what a dark pool was’ (Crammond,
2013). This lack of knowledge is system wide and worryingly has implications for regulatory
compliance; if regulators don’t know how dark pools work, how will they be able to monitor dozens
of non-transparent off-exchange trading? The extent of the lack of knowledge is revealed in a NY
Sun article that reported Erik Sirri, Head of the SEC’s (Securities and Exchange Commission the main
financial regulator in the USA) Division of Market Regulation as stating that the ‘SEC believes the
[dark pools] are available to all market participants.’ (NY Sun 2007) This is clearly not the case, part
of the raison d’etre of dark pools is that they exclude bots in order to provide a safe place for
investment banks and financial institutions to trade, one of the earliest dark pools, David
Mathisson’s Crossfinder, was set up specifically to exclude bots (Patterson, 2012: 6) and Marex
Spectron, a UK based dark pool markets itself specifically as being ‘invite only’ (Crammond, 2013).
Another product of the development of electronic trading that has become more and more salient
with dark pools is the phenomenon of colocation. This involves dark pools placing black boxes as
close as possible to the matching engines of stock exchanges. With trades happening in nano-time
proximity to a server and the use of turbo-charged or ‘over-clocked’ computers cooled with nitrogen
can dramatically increase profitability for those with the right access to the system’s plumbing.
Patterson queries whether this ‘setup fit[s] in with the notion that electronic trading created a level
playing field [his italics] for all investors’ (Patterson, 2012: 282). This is noted in the NY Sun which
quotes a financial services analyst discussing Goldman Sachs’ dark pool, Sigma X.
"They have spent hundreds of millions of dollars on systems. They have more people in IT than they
have traders or investment bankers. They want to drive the price of trading to a level that will drive
most firms out of business." (NY Sun 2007)
This worry is shared by Senior Lecturer at Essex University and industry insider, David Norman.
Having worked as a trader since the late 1980s David hoped that the rise of electronic trading would
bring a democratization of the markets. Instead it is becoming clear that access to off exchange
trading and, to an even greater extent, a fundamental understanding of off exchange trading is
limited to a small number of participants (Norman, 2013). The major implication of this is that the
market is now, more than ever, open to abuse and that this abuse may over time reduce confidence
in markets preventing corporations from reliably financing their operations.
The need for dark pools
Prior to electronic trading, stock exchanges provided a venue for corporations to raise finance.
Facilitating the buying or selling of stocks was the job of market makers. Access to stock exchanges
was expensive and the market makers were well compensated for their work. Electronic trading
bypassed the need for human interaction, trading became cheaper and spreads, the difference
between the price to buy and sell, reduced (Patterson, 2012: 1-10).
Electronic trading has also allowed for the proliferation of trading venues, this competition has
further reduced the cost of trading. Neil Crammond, notes that by trading off-exchange, even small
traders are saving from £1000 per month over the prices charged to trade on conventional
exchanges (Crammond, 2013).
Following fast behind the phenomenon of electronic trading networks (ECNs) has been algorithmic
trading. Algo deploying bots would seek out large orders and, before they had been entirely
processed, get in ahead buy an option and then resell it to the mutual whilst continually ramping up
the price. The creation of dark pools allowed large funds to trade in secret. With prices and
participants not published until after the trade has occurred it mitigated, for a while, the ability of
bots to game the system (Patterson 2012, 5-6).
Improving liquidity, the ease of buying and selling, is then the critical attraction of the brave new
world of electronic trading. Bronislav Kandrik, Head of Trading at the RSJ algo likens it to the
improvements in efficiency in car manufacture observed after the introduction of robotic
technology.
‘Before robotic car building, everything was built manually. Now we have a lot of robots helping and
it has made the process more efficient and cheaper.’ (Krandrik in Bloomberg, 2012)
However, commentators are questioning whether the supposed virtues of electronic trading in dark
pools are as great as purported by its supporters. It is evidently clear that bots are indeed present
in dark pools and it seems the spreads taken by market-makers aren’t ‘as pancake thin’ as dark pool
owners would have us believe (Patterson, 2012: 283-284). Bots are bidding up the price of orders
that have been chopped up into small chunks so that a financial institution ‘could easily pay fifty
cents or more above the initial offer.’ (Patterson, 2012: 284).
High frequency and dark pool trading has been blamed for increases in market volatility. In the 2010
Flash Crash, the Dow Jones briefly fell 1000 points (before recovering) in one day (Paterson,
2012:271). With the global economy reeling from ongoing economic turbulence – caused by exotic
financial dealing in sub-prime mortgages – it is questionable whether financial markets need further
recourse to the reduced transparency associated with high frequency trading and dark pools.
The legitimacy of dark pool trading has been called into question. Simply put, the secrecy with
which trades are executed means that proper scrutiny of dark pools is far from guaranteed. Neil
Crammond notes that with 40 dark pool venues in operation in the UK alone it would be very
difficult for the FSA with its limited budget to properly monitor even a fraction of the trades
(Crammond, 2013). Further to this it seems that many in the industry are unclear exactly what is
legal and what isn’t. Practices such as selling to traders in the same office, bots which make then
cancel thousands of orders in a nano-second and dark pools allowing preferred customers to jump
the queue to fill an order are becoming common place and examples of the grey area in which
traders are operating in.
Even Duncan Niederauer, Chief Executive of NYSE Euronext and one of the chief architects of the
new era of electronic trading (Patterson 2012: 274) has expressed concerns over volatility, issues
with compliance and a growing trend of financial plumbing being captured by a privileged few
asking,
‘Can investors count on US financial markets to remain the most open and transparent in the
world?’ (Financial Times)
Conclusion
Traders have noted that in conventional markets, ‘volumes are dropping off the cliff’ (Crammond,
2013) and that as more and more open interest (positions that are held over night) traders move
into dark pools the future of the financial markets as we currently know them remains unclear. We
may not have long to wait to find out. Duncan Niederauer has stated that,
‘Investors have withdrawn billions of dollars from the US equity markets in recent years, at least in
part due to a lack of confidence in market integrity.’ (Financial Times)
If corporations are no longer able to reliably raise capital on stock markets then it is unclear how
future business ventures will be funded. Alongside this we have increasing stock market volatility
and a lack of clarity over the rules of the game in dark pools. For example, what happens if a dark
pool goes bust – do those trades count? Patterson has suggested that as a result of the forces
unleashed by electronic trading the entire market is dark (Patterson, 2012: 339). My research into
dark pools has not left me with confidence that agencies such as the SEC and FSA or regulatory
regimes such as MiFID II, have either the know-how or capacity to deal with these forces and that
coupled with a poorly performing global economy ongoing financial shocks on the scale of the 2010
Flash Crash are a strong possibility.