By: Jackie Connor
The modern conception of dark pools is a fairly recent development. In 1986, Instinet established the first truly modern dark pool: After Hours Cross. Instinet’s pool was soon followed by,among others, ITG‘s Posit. Worldwide, growth of dark pools has climbed in recent years. For instance, during the past three years, dark pool trades have increased by 50% within the US.
As of 2013, dark pools are most predominant in North America, notably in the United States, where they account for approximately 13% of trading volume. This number varies slightly from source to source due to rapid growth and perplexing operations. Their prevalence in Europe is similar but less significant than in the US. While dark pools exist in Asia, they are not as prominent as in the West. For example, in 2012, dark pools accounted for only 2% of trades in Hong Kong.
The forerunner of modern electronic dark pools was “upstairs trading”. Buyers and sellers met in the upstairs rooms of brokerage companies to discuss the conditions of trades that were not listed in public exchanges. In this sense, the role of upstairs trading is similar to that of the dark pools that exist today.
The prominence and growth of dark pools makes it necessary to acquire an understanding of these platforms and factor them into trading strategies. However, as their name suggests, many of the specific details on what they are and how they operate remain unclear.
What are Dark Pools?
We begin with the basics: a definition and description of dark pools.
Dark pools encompass trades and liquidity outside public exchanges and are based on peer-to-to peer electronic platforms rather than on traditional centralized networks. They are legal and exist in many forms.
The main types of dark pools are:
- independently owned
- broker-dealer and,
The NYSE’s Euronext and Direct Edge are two popular exchange-owned dark pools, while Instinet’s, Liquidnet’s, and Investment Technology Group’s pools are independently owned. Examples of broker-dealer pools are Credit Suisse and UBS Investment Bank. BIDS Trading is a well-known consortium-owned pool.
While the above are the most common forms, there are others. Dark pools differ from country to country due to specific regulations and environments. Rather than discussing each specific type, we describe several potential characteristics a generic pool may have.
Access to a pool may be provided to a combination of clients or participants, institutional investors, Alternative Trading Systems (ATSs), Multilateral Trading Facilities (MTFs), broker-dealers, and exchanges. Both ATSs and MTFs are dark pools with similar functions. The main difference is that ATSs are US-based while MTFs operate in Europe.
While prices in dark pools may differ from those on public exchanges, they often reference the best bid or offer (BBO), or its midpoint, volume-weighted spread, or time-weighted average price, as stated in public forums. Additionally, dark pools have multiple common order types, including market (bound to BBO), limit (based on a specific price), minimum quantity, and immediate or cancel (all or part of the order is executed immediately and anything remaining is cancelled) orders.
Order submission may occur: (1) directly through participants; (2) indirectly through electronic order routing; or (3) the dark pool may sweep, or adjust based on available liquidity or price and quantity limits, client orders submitted in a broker-dealer’s order book. These orders can be submitted on a continual basis, during call auctions, or according to the specific negotiations of participants.
Several additional dark pool characteristics should not be overlooked. For instance, unlike their counterparts, dark pools do not publish data such as price and actors before transactions. In general, some information, often including volume, symbol, price, time, and marketplace identifiers is released post-transaction. The rules and laws that dictate post-trade transparency vary among countries. Additionally, dark pools tend to handle higher-frequency, highly liquid, low-volatility shares.
Dark pools have grown because traders believe participating in them furthers their trading interests. The word “believe” is significant, because it does not imply factual knowledge. Despite traders’ opinions, dark pools may not truly provide sustainable, or any, benefits. If traders currently lack proper information on, or fail to acknowledge any potential disadvantages of, such pools and are subsequently hurt, then dark pools may not remain a viable force in the future.
The Purported Benefits and Disadvantages of Dark Pools
Proponents of dark pools often claim an array of benefits, many of which are grounded in the pools’ anonymity. Supporters contend dark pools are more reliable, especially when compared to public exchanges, are cost-effective, and give participants freedom and control.
Dark pools actively and effectively monitor and adjust their liquidity. Some dark pools categorize orders based on their toxicity. Barclays’s pool, for instance, categorizes toxicity based on multiple items, such as pricing or how an order compares to similar orders. Other examples include ITG’s Posit, which has a point-in-time crossing method, and Knight Capital’s Knight Match, which uses the equal split method. Each dark pool has different practices, in part to differentiate itself from the rest, so this cannot be a comprehensive list of all the monitoring techniques they employ. Nevertheless, the point is that dark pools’ methods inspire confidence among traders.
B) Enhanced Flash Crash Resistance
High frequency trading, or HFT, has generally flourished worldwide in recent years. In fact, up to 70% of orders in the NYSE are HFT orders. HFT involves computerized algorithmic trading and is most often employed by major investment banks, proprietary trading firms, and hedge funds. The algorithms track many variables to determine how stocks will move in the next fraction of a second. When they get the answers, computers and firms react almost instantaneously and automatically, often holding shares for less than a second. However, these types of orders revealed a problem in public exchanges that has unnerved many traders: human crowd behavior results in flash crashes.
One prime example is the United States’ May 6, 2010 Flash Crash. During this crash, the Dow Jones fell by nearly 1,000 points, which resulted in a trillion dollar loss of market value in twenty minutes. The subsequent investigation revealed the crash was started by a large errant sell order improperly announced, as it lacked appropriate price and time instructions. The SEC, or the U.S. Security Exchange Commission, observed the crash was exacerbated by a lack of proper market circuit breakers or trip mechanisms when a huge number of HFT computers sparked a chain reaction.
While the 2010 Flash Crash was the most memorable and regulators have added some trip mechanisms since, these flash crashes still occur at a smaller level. For instance, in 2013, when a false Twitter report from the Associated Press stated that the White House experienced two explosions, another miniature flash crash ensued. These mini flash crashes suggest public exchanges have been incapable of improving their technology to the point where crashes are trivial or nonexistent. As losses can be large, especially when the cumulative effect is considered, this failure is a significant concern.
The 2010 Flash Crash and its aftermath reveal that at best, public markets only can limit damage. In contrast, semi-private dark pools are inherently less prone to crashes because of the smaller number of traders. Dark pools can impose stronger risk management tools because they are privately owned.
2) Savings and Profitability
Dark pool defenders often reference four ways dark pools increase savings and profitability: technological improvements, decreased market impact costs, lower transaction costs, and price improvement through discretionary orders.
A) Technological Improvements
Public exchanges have not improved technology to the point where they experience fewer flash crashes than dark pools. Due to the non-private nature of public exchanges, this lack of progress is likely to remain the case in the future. Consequently, public exchanges are more costly for participants.
B) Less Unwanted Price Volatility
Anonymity diminishes unwanted price volatility for two reasons.
First, competitors do not have information about other orders. Consequently, they are unable to act in a manner that leads to unfavorable price changes for other firms. The result is that firms seeking to sell large volumes of shares at a time will be able to get bids and offers than they would have in a public exchange.
The size of a sale does not result in a harmful price movement. Should there be an order of an unusual size in a transparent platform, the price may change unfavorably when everyone reacts to the extra supply.
The savings from both these factors that result from being anonymous can be significant over multiple shares and trades.
C) Lower Transaction Costs
Dark pools’ transaction costs are often lower due to two forms of transaction savings.
First, dark pools often eliminate the transaction costs of routing to external markets. For instance, in ping destinations, where customer’s orders only contact the pool operator’s own orders, there is no need to reroute; these are often in broker dealer pools like GETCO. This is also the case with internalization, which is used in broker-dealer pools like Credit Suisse. In this case sellers and buyers are internally matched and do not have to route to external markets.
Second, dark pools discount large orders. For example, Liquidnet gives investors discounts when they sell in large quantities and in fewer transactions. The lower costs of essentially wholesaling a share or product is a reality that is reflected in many industries.
D) Price Improvement
Additionally, dark pool buyers believe their orders will receive the same, if not better, prices than in public exchanges because of discretionary orders. The latter gives brokers the freedom to determine the time and place of execution. Often, when acting for clients, brokers first check dark pools. Should there be any sell orders priced below clients’ buy requirements, then the order is executed. A successful order that matches both the sellers’ and buyers’ price requirements should be mutually beneficial, rather than a win-lose or zero-sum situation. If this scenario does not exist, then brokers purchase their clients’ orders on regular markets or at the market price. Thus, clients believe they receive beneficial prices from dark pools, even if the exact price is not revealed.
Prices in dark pools reference the BBOs of public exchanges. However, dark pools’ prices are sometimes lower due to sellers’ ability to execute larger orders. Price improvement varies from pool to pool. For instance, for March 2013, Liquidmatrix, a research company, noted average price improvement can vary from 3.7 to 5.3 basis points. (One hundred basis points represent a 1% change; one basis point is equal to 0.01%.)
These four perceived savings contribute to dark pools’ growing popularity and are a significant factor positive future prospects.
3) Increased Firm Control and Freedom to Customize
An array of choices and protections of proprietary information dark pools purportedly deliver affords the frequently cited benefit of freedom for participants.
As there are many dark pools but few public exchanges, traders and firms are more likely to find the most suitable terms and conditions in dark pools. Dark pools’ customization options allow traders to select whom they do and do not want to trade with within a pool, which can, for example, prevent hostile takeovers. This selection of trading partners is includes limiting the people who have access to them.
Dark pools often provide anonymity. One important implication is the identity of participants is ideally hidden before trades, even though some information is released afterwards. In theory, there is no way to know whether buyers and traders are brokers, traders, or firms, before trades are executed. A second benefit of anonymity: the size of orders is theoretically unknown. When these two attributes are combined, participants have more freedom to uptake, let go of large positions, or make other strategic moves according to their desires while minimizing market impact costs.
4) Other Benefits
The first three benefits above are the most referenced. Dark pool proponensts consider other smaller, less publicized benefits, which arise from pools’ unique characteristics and differences from public exchanges. For instance, as supporters focus on the benefits for large traders, the gains of small investors go unnoticed. Retail investors who use 401(k)s and mutual funds may benefit as their trades are often bundled into larger orders.
Harms and Concerns
While many people emphasize the advantages of dark pools, others—such as regulators and public activists – voice opposing sentiments. Those participating in dark pools also raise several concerns and complaints. We focus on the loss of price discovery, fragmentation of information and liquidity, lack of access to dark pools, and information leakage.
1) Loss of Price Discovery
One of the most significant criticisms of dark pools is the loss of price discovery. Price discovery is when market price is determined. Factors such as supply, demand, participants’ willingness to trade, past information, and other external events affect these prices.
Dark pools hinder price discovery because they do not reveal information before trades are executed. However, post-trade information aids price discovery by providing past information about the security. Dark pools often give such information either voluntarily or to meet regulations, but post-trade information availability is neither universal nor easy to verify. Indeed, there are claims some dark pools double or triple count their trading volumes. Thus, it is harder to establish an accurate price quote for a security. Dark pool opponents have become more vocal about this downside as dark pools have grown. They argue as more dark pools handle a larger proportion of orders, the price derived from available information becomes less representative of the true price.
This problem creates a cycle where people have less and less incentive to submit orders in the public arena. When everyone is releasing their information, participants do not believe that they are at a disadvantage by doing the same. However, when only a few people give out information, those who release proprietary information, intentions, and strategies are at a disadvantage. For these reasons, the growth of dark pools exacerbates price discovery losses.
Having accurate prices is necessary for resource allocation and efficiency. Critics believe the negative impact of dark pools on price discovery will prompt participants to start their own independent research. This research, in turn, causes superfluous time and monetary expenditures.
2) Fragmentation of Information and Liquidity
As the number of dark pools grows, each has its own rules. While this competition allows for more personalization and may be beneficial, there also is more unnecessary information and liquidity fragmentation. Traders and regulators are concerned about problems arising from the fragmentation: more difficulty in searches and exclusivity of information.
Traders cannot make decisions as easily because of a loss of price discovery. They sometimes compensate for this loss by seeking more information. However, the search becomes more challenging due to the fragmentation of information and liquidity. Not only must traders seek out information on prices, they must also learn the rules of various pools. They must search more places to find the liquidity to obtain the most desirable sale. However, as discussed, this search can incur significant time and money costs. To avoid these costs by not searching more may harm traders if fragmentation leads to improper decisions that lose money.
Second, fragmentation occurs because information is more prized. Rather than being readily available to everyone, participants compete to gain exclusive access. For example, people compete for indications of interest (IOIs). IOIs are essentially quotes in dark pools and contain private details about participant’s orders such as the buying or selling interest, size, and price range not provided to the public. These revelations are internal, and participants remain anonymous to the public. One example is a flash order seen on some US pools such as DirectEdge. When this order is executed, the venue sends IOIs to a select group for approximately half a second, a significant amount of time in modern electronic markets, before it sends it to others to try to keep the order on their market. Critics, such as the executives at the NYSE’s Euronext, claim the prioritized group has an unfair advantage, even though they concede these orders may attract extra liquidity or have lower costs.
3) Lack of Fair Access to Dark Pools
For the most part, public exchanges are not exclusive. In contrast, we have discussed how dark pools sometimes limit access to institutional investors, MTFs, clients, exchanges, broker dealers, and ATSs. While this in and of itself is not necessarily unfair, critics claim that potential participants are unfairly denied access to pools. Their reasoning is articulated and summarized by Daniel Mathisson, a Managing Director and Head of Advanced Execution Services at Credit Suisse. In a US Senate Hearing, Mathisson said, “[b]roker dealers are sometimes denied access to each other’s dark pools for competitive or capricious reasons.”
These concerns are significant because traders will need to access dark pools more in their search for liquidity as pools grow and public exchanges shrink. Exclusion can be particularly detrimental to those who are blocked from large dark pools that have a lot of liquidity. While there may be other pools these traders can access and thrive in, being blocked from a potentially more profitable pool can lead to financial opportunity costs.
4) Information Leakage and Access to Information
In an arena like dark pools where information is not readily available and where having it can mean making or losing money, information becomes highly priced. One responsibility of dark pools’ is to ensure private information is not improperly leaked. However, some observers are not confident that this happens. This concern is rooted in an inherent problem dark pools face: keeping proprietary information safe even though revealing some is necessary in order to match orders.
Dark pools have tried to resolve this issue through IOI’s, which many believe are imperfect. While IOIs may theoretically be dispersed fairly, data regarding the allocation of IOI’s is not publicly verifiable. Consequently, regulators, such as the SEC, and others worry that an exclusive group receives more, better, or earlier information., 
Private information is leaked to several actors: gamers, the dark pool itself, their prop desks, and undisclosed liquidity partners. We divide these actors into the types of information they receive: indirect and direct.
For gamers, information leakage is not direct. When they “fish,” gamers try to determine on their own how large orders may be by pinging the pool with many small orders to see the response. If the response is favorable, allowing them to determine there is a big buy order, gamers will accumulate shares in small blocks and then sell them en masse when the price is highest for a profit. By experimenting with orders, indirect information leakage reveals the size of buy orders. This in turn creates an unfair playing field, where some have greater access to information than others. Resolving this problem—for example, by removing some clients from their pools, or having minimum quantities of orders—is not necessarily in the interest of dark pools. Higher share volumes determine success. Critics hint that dark pools frequently ignore situations of information leakage.
The other three actors, the dark pool itself, their prop desks, and undisclosed liquidity partners, receive direct information. The dark pool operator receives information directly, as it sometimes participates in its own pool even though operators are expected to be third parties who protect information. Prop desks exist within firms who operate dark pools, such as investment banks. These desks use the firm’s money to trade. Public exchanges do not have such entities. Although prop desks are legally required to not know customers’ orders, they may have access to IOIs of their firm’s dark pool, if the latter lacks strict controls. Likewise, dark pool operators may leak IOIs to their liquidity partners. Many argue this gives all three groups an unfair advantage. Participants must give dark pools their information and do so with the expectation that it will not be used against them. Detractors emphasize that clients are not able to respond properly before it is too late. Therefore, critics believe dark pools increase adverse selection (where a more-informed side is able to better predict stock movement and the stock moves in a trader’s direction immediately after order execution), gaming, or other market impacts, all unfavorable to clients.
5) Other Concerns
Some worry about increased no execution risks in dark pools. Trades depend on matching and matching orders may be harder in an anonymous environment. People do not know supply and demand making adjustments to an equilibrium point more difficult. Thus, those who are on the side with more orders may not be able to get their orders filled. Dark pools’ algorithms remain confusing and it is difficult to quantitatively measure their performance. Many other worries about potential harms and abuse stem from anonymity and novelty issues, since both make it difficult to know if dark pools are truly good or bad.
Mixed Study Results
Studies on dark pools give mixed answers. For instance we can look at two studies on dark pools’ effects on market quality. A study from Ohio State University suggests that overall “more dark pool activity is associated with better market quality: narrower spreads, more depth, and lower volatility.”
Another study by the CFA Institute arrives at a different conclusion. It suggests that so long as dark pools remain less than fifty percent of total trading, they improve market quality. However, when a majority of trading takes place in dark pools then market quality decreases. These mixed results are a consequence using different methodologies.
The CFA Institute study measured 450 US stocks’ bid-offer spreads, top-of-book depth, off-exchange volumes, and other variables on a selection of dates between 2009 and 2011. The Ohio State University study, conducted in 2009, analyzed only eleven dark pools that voluntarily responded to SIFMA’s enquiries and uses daily data (EFA). Overall, studies may use different pools, be performed at different times, measure different traits, and have different procedures for evaluating their measurements, among other potential differences. Thus, studies frequently arrive at conflicting answers. Since they are so hard to compare, it is not easy to determine which are best to use when evaluating dark pools.
The fact that studies often conflict is not particularly encouraging. If we do not know exactly what is good and what is bad about dark pools, how can we analyze their ethical implications?
We do so by analyzing dark pools through two frameworks: consequentialism and deontology. These theories are important for two reasons. First, they are significant in that together, they cover both the ends and the principles of a process, which allows us to focus on the general picture. This view is important as the details of dark pools are often unknown. Second, the two frameworks represent the contrasting focus of supporters and detractors, who prefer consequentialism and deontology respectively. We have the opportunity to measure the strength of arguments of each side.
We “weigh” the arguments of each viewpoint. The term “weighing,” often used in debate rounds, involves comparing results of using the two theories.
Consequentialism, as its name indicates, holds that only the consequences of an act matter. Good consequences mean that an act or choice is morally right, while bad consequences mean the opposite.
In this case, we consider only the economic and financial outcomes of dark pools because people primarily trade to earn profits. Whether their methods are the best, or meet any other standard, is not relevant in this analysis. If participation in a dark pool costs traders money, then dark pools are not ethical. However, if dark pools generate more profits or savings for traders, then dark pools are ethical.
As we have seen, studies arrive at different conclusions on a variety of dark pool issues due to different measuring techniques and the anonymity associated with these platforms. This analysis does not attempt to determine which data to use regarding profits and losses generated by dark pools. Instead, the analysis focuses on what is generally agreed to create losses and profits to determine the economic consequences of dark pools.
First, the costs critics say pools generate: extra search costs and lost money from leaked information. Critics argue search costs increase. Supporters point out that lower transaction costs in part, derives from dark pools requiring less routing, or searching, for liquidity. As a result, the two arguments cancel out; the increased search costs is balanced by lower transaction costs.
We address the second criticism, information leakage, in much the same way. The critic’s argument hinges on how dark pools may lead to loss of money as customers lose control over their information. However, supporters hold that dark pools give companies more control over their strategic and proprietary information. Thus, the arguments on the effect of dark pools on the cost of information control are logically neutral.
We analyze the validity of two remaining arguments of dark pool supporters regarding generated savings and profitability: technological improvements and price improvement through discretionary orders.
Supporters mention there are technological improvements that improve reliability and save costs. Critics note that dark pools are not always reliable due to higher no execution risk. But critics do concede this is only an occasional risk, rather than a regular occurrence. The reliability issue seems to favor dark pools. The second issue, price improvements, is rooted in order types. The critics’ argument that order types may not be fair is not directly related to profitability or savings. Theoretically, everyone can still benefit monetarily even if one group benefits more than others.
Overall, dark pool supporters have two consequentialist arguments in their favor, one of which is moderately strong, while the other is solidly strong. Criticisms of dark pools have been shown to be insufficient. The strength of monetary based consequentialist arguments illustrate why dark pools are gaining popularity long held by public exchanges.
Deontology, unlike consequentialism, pays no heed to results. Deontologists believe that some actions, even if they have excellent outcomes, are inexcusable. As the basis for moral analysis, Immanuel Kant’s deontology relies on a duty to follow prescribed principles. He argues that nothing is intrinsically good unless it has a good basis, will, or means. Since “good means” and “principles” can be vague, we establish two ways of seeing if dark pools have a good basis. First, we analyze whether dark pools ensure players and technology abide by the principle of honesty by following rules in good faith. Then we assess if dark pools create fair rules.
Dark pools need a measure of control over their members and technology to create an honest system. If dark pool operators have this influence over their members, then the members can be persuaded to follow good rules. Critics mention one main way pools seem to fail: information leakage. Dark pools have not stopped gaming and directly give information that may be harmful. However, supporters note that dark pools’ technology is more reliable than that of their public counterparts. This reliability prevents the spread of improper or harmful information. In addition, dark pools monitor their participants. As neither side technically refutes the other, and we cannot compare them directly, this argument is nullified.
We also need to consider whether or not the rules that guide dark pools are fair. In particular, critics say that dark pools unfairly limit access to some people. In response, supporters note there are many pools and not all will be closed to potential clients. Even so, being blocked from a beneficial pool may make traders lose profit opportunities. However, supporters concede some dark pools do limit access. Dark pools fail to fulfill the principle of fairness in this respect, even if their actions might be profitable. This logic extends to the exclusivity of orders, which creates a two-tiered system. By noting that flash orders, for instance, do exist, supporters concede that pools’ guidelines are not fair.
In sum, we have two consequentialist arguments in support of dark pools: (1) their reliability and (2) price improvements. There is one deontological argument against them: (1) dark pools are not fair because they limit access. However, when we look at the strength of both, the criticism of dark pools is stronger, because even supporters agree the issue of fairness is valid. In contrast, the arguments in support of dark pools are contested.
Therefore, we conclude that dark pools are ethically questionable. This is not to say dark pools cannot be good economically, only that they are imperfect systems that can harm some traders.
It is clearly beneficial to fix the two deontological imperfections of dark pools: unfair lack of access and information exclusion.
The most aggressive approach is for regulators to outright ban the practices of limiting clients and sharing data with a select group of clients. Despite its simplicity, however, this option will likely face fierce opposition by those who benefit from such practices.
Another option involves increasing transparency, whereby dark pools provide more relevant information to regulators, who then make more moderate changes to ensure dark pools and their actors exercise good will. This approach may receive a slightly warmer reception, but regulators may have difficulties in drafting and implementing new policies because there are numerous dark pools, each with its own profile. It may be difficult to find one approach to suit them all. Using a multitude of methods for different pools may be difficult to enforce or create inequities.
Dark pools or regulators can mandate that the most transparent orders or welcoming pools receive some type of priority. This option, however, may also raise questions about a new type of two-tiered system based on transparency. Nonetheless, this may be better than the current system, which some believe fosters a two-tiered system on information or client exclusion.
A last suggestion involves helping participants. Regulators or dark pools themselves can try to give traders more information about orders or other available dark pools so that traders can respond more appropriately in the current environment. However, regulators may continue to face problems in accessing such information.
A Chinese translation of this article may be found here.
Photo: Courtesy Flickr
 Clark, Doug, Rizwan Awan, Jeremy Dietrich, Andrew Ng, and Andrew Karsgaard. Dark Pools and Dark Liquidity. BMO Capital Markets. BMO Financial Group, 6 July 2009. Web. 28 Sept. 2013. <http://qes.bmocm.com/papers/7_BMO_DarkPools.pdf>.
 Preece, Rhodri. “The Pros and Cons of Dark Pools of Liquidity.” Financial Times. The Financial Times, 6 Jan. 2013. Web. 28 Sept. 2013. <http://www.ft.com/intl/cms/s/0/b594f978-54dd-11e2-a628-00144feab49a.html>.
 Patterson, Scott. “‘Dark Pools’ Face Scrutiny.” The Wall Street Journal. Dow Jones & Company, 5 June 2013. Web. 28 Sept. 2013. <http://online.wsj.com/article/SB10001424127887324069104578527361102049152.html>.
 Preece, Rhodri. ” Pros and Cons of Dark Pools.”
 Lee, Justin. “Dark Pool Trading in Asia Increases – as Does Regulatory Interest.”Risk.net. N.p., 8 June 2012. Web. 29 Sept. 2013. <http://www.risk.net/asia-risk/feature/2183123/dark-pool-trading-asia-increases-regulatory>.
 Technical Committee of the International Organization of Securities Committee. Issues Raised by Dark Liquidity. Rep. Better Regulation, 27 Oct. 2010. Web. 28 Sept. 2013. <http://hb.betterregulation.com/external/Issues%20Raised%20by%20Dark%20Liquidity%20%E2%80%93%20Consultation%20Report.pdf>.
 Achuthakumar, Ravikiran. Dark Pools of Liquidity: Understanding & Technology Dependencies for the Future. Whitepaper. WIPRO, 2009. Web. 28 Sept. 2013. <http://www.wipro.com/documents/insights/whitepaper/dark_pools_of_liquidity_understanding_and_technology_dependencies_for_the_future.pdf>.
 SeekingAlpha. “Shedding Light on ‘Dark Pools'” NASDAQ. NASDAQ, 3 June 2010. Web. 28 Sept. 2013. <http://www.nasdaq.com/article/shedding-light-on-dark-pools-cm23529>.
 “NYSE Arca Equities Order Types.” Chart. U.S. Equities. NYSE Euronext, n.d. Web. 28 Sept. 2013. <http://usequities.nyx.com/markets/nyse-arca-equities/order-types>.
 “Sweeps and Intermarket Sweep Orders.” EliteTrader. Elite Trader, n.d. Web. 28 Sept. 2013. <http://www.elitetrader.com/vb/attachment.php?postid=2804485>.
 Technical Committee of the IOSCO. Issues Raised by Dark Liquidity.
 Williams, James. “Dark Pools Deliver Price Improvement And anonymity.” Hedgeweek. GFM, 6 June 2013. Web. 28 Sept. 2013. <http://www.hedgeweek.com/2013/06/06/185423/dark-pools-deliver-price-improvement-and anonymity>.
 Thomases, Jake. “Predators in the Pool: How Dark Pools Tackle HFT.” Watershed Technology. N.p., 1 June 2012. Web. <http://www.waterstechnology.com/waters/feature/2181865/predators-pool-dark-pools-tackle-hft>.
 Portnoy, Kyle. “High Frequency Trading and the Stock Market: A Look at the Effects of Trade Volume on Stock Price Changes.” The Park Place Economist XIX (2011): 69-75. Illinois Western University. Illinois Western University, 2011. Web. 28 Sept. 2013. <http://www.iwu.edu/economics/PPE19/8Portnoy.pdf>.
 Rooney, Ben. “Trading Program Sparked May ‘flash Crash'” CNNMoney. Cable News Network, 01 Oct. 2010. Web. 28 Sept. 2013. <http://money.cnn.com/2010/10/01/markets/SEC_CFTC_flash_crash/index.htm>.
 U.S. Commodity Futures Trading Commission, and U.S. Securities and Exchange Commission. Findings Regarding the Market Events of May 6, 2010. Rep. N.p.: n.p., n.d. U.S. Securities and Exchange Commission, 30 Sept. 2010. Web. 28 Sept. 2013. <http://www.sec.gov/news/studies/2010/marketevents-report.pdf>.
 Cox, Jeff. “3 Years Later: Learning to Live With Flash Crashes.” CNBC.com. CNBC, 6 May 2013. Web. 28 Sept. 2013. <http://www.cnbc.com/id/100709978>.
 Popper, Nathaniel. “As Market Heats Up, Trading Slips Into Shadows.” The New York Times. The New York Times, 31 Mar. 2013. Web. 29 Sept. 2013. <http://www.nytimes.com/2013/04/01/business/as-market-heats-up-trading-slips-into-shadows.html>.
 Michaels, Dave. “Stock Exchanges Seek Curbs on Dark Pools to Fight Exodus.”Bloomberg.com. Bloomberg, 04 June 2013. Web. 30 Sept. 2013. <http://www.bloomberg.com/news/2013-06-04/stock-exchanges-seek-curbs-on-dark-pools-to-fight-exodus.html>.
 Mittal, Hitesh. Are You Playing in a Toxic Dark Pool? A Guide to Preventing Information Leakage. ITG Posit. N.p., 6 Dec. 2008. Web. 22 Sept. 2013. <http://www.positjapan.org/news_events/papers/ITGResearch_Toxic_Dark_Pool_070208.pdf>.
 Bogoslaw, David. “Big Traders Dive Into Dark Pools.” Bloomberg Businessweek. Bloomberg, 3 Oct. 2007. Web. 29 Sept. 2013. <http://www.businessweek.com/stories/2007-10-03/big-traders-dive-into-dark-poolsbusinessweek-business-news-stock-market-and-financial-advice>.
 Kato, Hiroki. Quantifying Price Improvement Delivered by Dark Pools. Tech. Nomura Research Institute, 10 Mar. 2010. Web. 29 Sept. 2013. <http://www.nri.com/global/opinion/lakyara/2010/pdf/lkr201074.pdf>.
 Williams, James. “Dark Pools Deliver Price Improvement And Anonymity.” Hedgeweek. Global Fund Data, 6 June 2013. Web. 30 Sept. 2013. <http://www.hedgeweek.com/2013/06/06/185423/dark-pools-deliver-price-improvement-and anonymity>.
 Graham, George. “Understanding Theoretical Investment Values.” Value Line. Value Line, 9 June 2011. Web. 30 Sept. 2013. <http://www.valueline.com/Tools/Educational_Articles/Convertibles/Understanding_Theoretical_Investment_Values.aspx>.
 Ross, D. Keith. “10 Things People Don’t Get About Dark Pools: Ross.” CNBC. CNBC, 2 Feb. 2013. Web. 30 Sept. 2013. <http://www.cnbc.com/id/100424690>.
 Winter, Tom. “Gazing Into ‘Dark Pools,’ the Tool That Enables Anonymous Insider Trading.” CNBC. CNBC, 23 Jan. 2013. Web. 30 Sept. 2013. <http://www.cnbc.com/id/100400981>.
 Zendrian, Alexandra. “The Risks and Advantages of Dark Pool Investing.” CBCnews. CBC/Radio Canada, 26 May 2009. Web. 30 Sept. 2013. <http://www.cbc.ca/news/business/the-risks-and-advantages-of-dark-pool-investing-1.822429>.
 Technical Committee of the IOSCO. Issues Raised by Dark Liquidity.
 Clark, Doug et al. Dark Pools and Dark Liquidity.
 Testimony Concerning SEC Oversight: Current State and Agenda, 111th Cong. (2009) (testimony of Mary L. Schapiro). Print.
 Comerton-Forde, Carole, and Tãlis J. Putni?š. Dark Trading and Price Discovery.Tepper School of Business. Carnegie Mellon, 3 June 2013. Web. 30 Sept. 2013. <https://wpweb2.tepper.cmu.edu/wfa2013/wfasecure/upload/2013_PA_314317_855505_120831.pdf>.
 Technical Committee of the IOSCO. Issues Raised by Dark Liquidity.
 Lopez, Linette. “GASPARINO: Mysterious Dark Pools Are Seeing More Trading Action Than The NYSE For The First Time Ever.” Business Insider. Business Insider, 7 Mar. 2013. Web.02 Oct. 2013. <http://www.businessinsider.com/more-trading-in-dark-pools-than-nyse-2013-3>.
 Clark, Doug et al. Dark Pools and Dark Liquidity.
 Clark, Doug, Rizwan Awan, Jeremy Dietrich, Andrew Ng, and Andrew Karsgaard.Flash Orders. BMO Capital Markets. BMO Financial Group, 7 Aug. 2009. Web. 1 Oct. 2013. <http://qes.bmocm.com/papers/9_BMO_FlashOrders.pdf>.
 Testimony Concerning Dark Pools, Flash Orders, High Frequency Trading, and Other Market Structure Issues, 111th Cong. (2009) (testimony of James A. Brigagliano). Print.
 Boni, Leslie, David C. Brown, and J. Chris Leach. Dark Pool Exclusivity Matters. Working paper. Bank of Canada, May 2012. Web. 1 Oct. 2013. <http://www.bankofcanada.ca/wp-content/uploads/2012/11/David-Brown.pdf>.
 Examining The Dark Pools, Flash Orders, High-Frequency Trading, And Other Market Structure Issues, 111th Cong., 14 (2009) (testimony of Daniel Mathisson). Print.
 Technical Committee of the IOSCO. Issues Raised by Dark Liquidity.
 Boni, Leslie et al. Dark Pool Exclusivity Matters.
 Curran, Rob. “Trading in a Dark Pool? Watch for Sharks.” MarketBeat. The Wall Street Journal, 18 Aug. 2008. Web. 01 Oct. 2013. <http://blogs.wsj.com/marketbeat/2008/08/18/trading-in-a-dark-pool-watch-for-sharks/>.
 Gaffen, David. “Gaming in Dark Pools.” MarketBeat. The Wall Street Journal, 26 Dec. 2008. Web. 01 Oct. 2013. <http://blogs.wsj.com/marketbeat/2008/12/26/gaming-in-dark-pools/>.
 Mittal, Hitesh. Are You Playing in a Toxic Dark Pool?
 Zhu, Haoxiang. Do Dark Pools Harm Price Discovery? Federal Reserve Bank of New York. Federal Reserve System, July 2012. Web. 1 Oct. 2013. <http://www.newyorkfed.org/research/conference/2012/MP_Workshop/Zhu_dark_pools.pdf>.
 Buti, Sabrina, Barbara Rindi, and Ingrid M. Werner. Diving Into Dark Pools. European Finance Association. The Ohio State University, 17 Nov. 2011. Web. 1 Oct. 2013. <http://www.efa2012.org/papers/s1f2.pdf>.
 CFA Institute. Dark Pools, Internalization, and Equity Market Quality. Issue brief. CFA Institute, Nov. 2012. Web. 1 Oct. 2013. <http://www.cfainstitute.org/ethics/Documents/dark_pools_internalization_and_equity_market_quality_final_issue_brief.pdf>.
 Sinnott-Armstrong, Walter. “Consequentialism.” The Stanford Encyclopedia of Philosophy. Stanford University, Winter 2012. Web. 1 Oct. 2013. <http://plato.stanford.edu/archives/win2012/entries/consequentialism/>.
 Alexander, Larry, and Michael Moore. “Deontological Ethics.” The Stanford Encyclopedia of Philosophy. Stanford University, Winter 2012. Web. 1 Oct. 2013. <http://plato.stanford.edu/archives/win2012/entries/ethics-deontological/>.
 “As Dark Pools Darken, Threats to Financial Markets Grow.” Securities Technology Monitor. SourceMedia, 20 June 2013. Web. 01 Oct. 2013. <http://www.securitiestechnologymonitor.com/blogs/as-dark-pools-darken-threats-to-financial-markets-grow-31782-1.html>.
 Technical Committee of the IOSCO. Issues Raised by Dark Liquidity.