• —Who is the compliance manager chasing on the trading floor?
• —A runaway algo.

In addition to enabling record‐breaking data processing and storage capacity, technology is also responsible for a fair share of previously unthinkable risks. When billions of dollars are moved around the globe at breathtaking speed, the dynamics can be outright dangerous. Ensuring the legitimacy of accounts and identifying and blocking malicious behavior are very difficult tasks to execute in real time.

The financial markets used to comprise just one exchange for each class of financial instrument. If you traded equities, you did so at the New York Stock Exchange (later, NASDAQ was created with the explicit purpose of trading new technology stocks). If you traded commodities or futures, you would do so at the Chicago Mercantile Exchange. Another exchange existed solely for options. Foreign currency pairs and most of bonds never formally traded on an exchange, having been intermediated privately by banks and specialized dealers. The bottom line is that you had clarity and consistency in where you would go to trade a specific financial instrument.

This is no longer the case. Today, there are 21 “national securities exchanges” registered with the US Securities and Exchange Commission, the top regulatory body for just equities, ETFs, and equity options. The names of exchanges registered to trade equities at the time this book was written are shown in Table 3.1. The last entry in Table 3.1 is the notorious IEX exchange, which made copious news aided by a timely book by Michael Lewis, The Flash Boys.

While there are still only two exchanges trading commodity futures, Inter‐Continental Exchange (ICE) and Chicago Mercantile Exchange (CME), there are six exchanges registered to trade equity futures, including futures on ETFs and commodity ETFs, documented in Table 3.2.

For US equities, along with the 21 exchanges, we now have 36 alternative trading systems (ATS), also known as alternative trading venues or, simply, dark pools. Table 3.3 shows the distribution of trading volume of “Tier 1 NMS stocks” (the big guys) for the first quarter of 2016. Notice that IEX is present and is #3 out of all exchanges by the number of shares and by the dollar amount traded—not bad for a little startup! Dark pools trade anonymously and without displaying order information before trades are executed. The lack of displayed information is the key difference between dark pools and registered “lit” exchanges. Exchanges show the entire limit order book, down to how many shares are in each limit order, while dark pools hide all limit‐order book information. Institutional investors that trade large blocks and lack appropriate algorithmic expertise (more on this in later chapters) may go to dark pools and “hide” their large orders from other traders. The same orders would be displayed in the limit order books of exchanges, potentially scaring away other traders by the sheer size of their bets. Note that neither lit exchanges nor dark pools display the identity of traders: Both lit exchanges and dark pools are anonymous!

The explosion of exchanges and alternative trading venues is driven by a singular factor. Way back when, in the 1950s and '60s and early '70s, trading was limited to a single exchange per type of a financial instrument, and that exchange was a not‐for‐profit organization. The not‐for‐profit construct was not a mere convenience, it was a necessity—order matching, settlement, and auxiliary record‐keeping was so labor‐intensive that exchanges simply could not turn a profit. Technology enabled the exchanges not only to move away from a nonprofit model, but also to create a cottage industry of extremely profitable businesses, all backed by very powerful technological infrastructure.

Let's pause for a moment to consider the enormity of changes due to fintech in the exchange arena alone. We are not yet talking about blockchain—the next fintech train steaming over the exchanges in the next decade (yes, it is already in the exchanges' collective backyards). The changes that end customers are feeling right now are related to data. If you are of a certain age (ahem), you may remember those days when to find a quote for the stock you owned, you had to look in the newspaper. Not just any random freshly printed newspaper, but a thick newspaper that contained a section of yesterday's quotes for stocks, bonds, and everything else under the sun. Furthermore, for each stock there were only five data points from the previous trading day available: open, high, low, close, and daily volume. That's it.

Could you tell if there was a flash crash? Hardly. You had no idea what was going on beside the news articles written by people who traded rumors or relied on the same limited data set.

What about Bloombergs and TVs? Yes, the Bloomberg terminal did change things around a bit. Few people may remember, but Bloomberg started as a terminal‐renting company, leasing out personal computers at the time when they were prohibitively expensive for most banks to buy. As a convenient segue, Bloomberg gradually built its own operating systems for the terminals the firm offered. Its operating system allowed users to query a computer's databases of data instead of relying on the newspapers. Still, even with all the search power, the data was limited to open, high, low, close, and daily volume, as trade‐by‐trade data was just too expensive to store. Additionally, no one really cared what happened intraday—when the transaction costs per trade were as high as 0.5 percent of the amount traded, an intraday drop of 0.5 percent was a mere drop in a bucket, not a cause for concern. To top it off, few Wall Streeters could care about anything after lunch—the times were good, the markets kept rising at an average of 8 percent per year, and those liquid three‐martini lunches were all the rage.

What happened next? As the technological power continued to increase, and overseas manufacturers managed to reduce the costs of computer components to unprecedented lows, it became feasible to deliver data and transact with an unprecedented speed, frequency, and cost that was unthinkable at the time. Regulators took notice and adapted governance, allowing competition in the space via regulation alternative trading systems (Reg ATS) in 1999, enacted on April 1, 2000.

Reg ATS was limited to equities and equity derivatives. Not surprisingly, it affected many institutional investors working with equities. Equity trade sizes have fallen from thousands of shares per order to uniform 100‐share orders, sliced with precision by complex algorithms. Traders Magazine (2015) reported that two‐thirds of US and European long‐only investors missed bygone natural blocks, which are the pools of liquidity where hedge funds, asset managers, and wealth managers can execute large orders without retaining personnel or specialty firms to manage their order execution, like in the long‐gone days when only one exchange existed.

Despite the proliferation of trading venues, the landscape of the market is not necessarily a “wild west,” as many similarities among trading venues exist. Most trading venues deploy the centralized limit order book to record and match the orders. It is also known as the double‐sided continuous auction, or more commonly, the limit order book.

An order book is a way exchanges keep track of all buy and sell orders, order cancellations, and other communication with brokers and traders. In some sense, an order book is like a shelf in a grocery store that contains tomato soup offered by different companies at different prices. Some choose to sell their tomato soup at $2 a can, while others sell tomato soup for $5 a can. These offers of essentially the same product at different prices are what the exchange order book is about, at least on the ask side of the market. Each unit of tomato soup may be thought of as a unit of volume offered, its price a corresponding ask or offer. When customers interested in purchasing tomato soup arrive, they may choose one of the following actions:

1. Buy the best‐priced tomato soup available on the shelf at the moment, a method known as buying via a market order.
2. Create a bid on tomato soup, by leaving a ticket with the store clerk asking him to call you if the price on any tomato soup drops to $1.99—your desired price level. This method of purchasing via a bid is known as a limit order: you as a trader specify the limit on the price you are willing to pay.

The above example is, of course, an oversimplification of how the exchange's order books work, yet it illustrates a point: trade instructions at exchanges are straightforward and make sense. The actual exchange limit order books are two‐sided, meaning that not only the sellers can display their wares on a shelf but the buyers can do so, too, by submitting their bid tickets, formally known as limit buy orders. The exchanges stock all the buy and sell tickets along one shelf in the direction of increasing price, ultimately coming up with a limit order “book” like the one shown in Figure 3.1.

In a limit order book, the best‐priced bid and ask define a bid–ask spread, the difference between the best ask and the best bid. The minimum spread is always equal to one tick, one division on the limit order book number line defined by regulators and, possibly, trading venues. At the time this book was written, in equities one tick was $0.01, or 1 cent. In foreign exchange, one tick could be as little as $0.00001, or 1/100,000 of a dollar. During times of uncertainty, for instance, ahead of major news announcements, the spread typically widens as limit‐order traders avoid risk by removing orders too close to the market price to avoid being “run over” or “picked over” by traders with superior news services or analysis. Limit orders can be removed by cancellations—separate requests placed with exchanges. All limit orders collectively create “liquidity,” defined in academia as the immediacy with which a trader may execute a market order. The more limit orders are present, the deeper the liquidity, the faster a market order of an arbitrary size can be executed.

All market orders, orders to immediately buy or sell, are matched with the best available bid and ask limit orders “resting” in the limit order book at the time of the market order arrivals. A limit buy order priced higher than that of the best offer is treated as a market buy order and is immediately matched with the best offer. Similarly, a limit sell order or ask priced below the best bid is deemed “marketable” and is treated as a market sell order.

In addition to buy and sell limit and market orders, many exchanges provide hybrid orders that may restrict the display of size of the order (e.g., iceberg orders), and other custom types of orders. All custom orders tend to be more expensive than the plain‐vanilla limit and market buy and sell orders.

How do investors choose between placing market, limit, and other types of orders? To answer this question, consider an average investor, Joe, who wants to do something mundane: buy or sell a stock or another financial instrument at the market open prices. To do so, Joe has two basic methods at his disposal (other order types are typically variations of limit and market orders):

1. Joe can place a market order that tells his broker or an exchange to fill his order as soon as possible at the best price available.
2. Joe can place a limit order specifying a particular price, but no time limit for his trade.

If Joe chooses the market order route, he is guaranteed to have bought his desired security, but possibly at a much worse price than the opening bid or even ask price. During the few minutes immediately following the market open, prices strive to incorporate all of the information pent up from overnight, when the markets are closed. This information is transmitted into the markets through orders, and the disparity of views causes the prices to bounce violently up and down. This continues until traders reach a consensus on prices. Due to the volatility, Joe's market order may be filled at the worst possible price, possibly erasing Joe's projected gain from the trade.

As an alternative, Joe may choose to place a limit order and specify the price at which he is willing to buy it. Here, Joe is facing another decision, the price itself. If Joe chooses a price that is too low, his order may never execute. If the price is too high, he does nothing to outperform his market‐order scenario. How can Joe determine a price that is just right, that is both favorable and results in a timely execution?

A simple, yet effective strategy could be to place a limit order at a mid‐price—a price that is the average of the bid and ask at the market open. To do so, however, one needs a timely source of market data, from which to calculate the mid‐price. (Most brokers provide their clients with free access to data that are 15‐minutes delayed—too slow for Joe to successfully identify and execute upon his strategy.)

All orders, order executions, and order cancellations are received and processed by the majority of the exchanges in the first‐come, first‐served fashion. However, exchanges may offer variations to distinguish themselves from their competitors.

In equities, trading is further complicated by the national best bid/offer (NBBO) requirements.

The requirement, a product of regulation national market systems (Reg NMS, 2005), stipulates that all trading venues have to continuously submit to the government the best limit buy and sell prices (best bid and best offer/ask) available on their respective venues. This is done simultaneously for all securities traded. The best bid and best offer quotes then enter the security information processor (SIP) run by the US Securities and Exchange Commission. From there, the quotes are aggregated in real time, the very best bid and the very best offer are picked out from all submitted data. These NBBO numbers are then distributed back to trading venues with the identification of the exchanges that have the best quotes.

And here comes the fun part: An exchange that has a local best bid and best offer that is inferior to the NBBO in a given security cannot execute the incoming market orders for this particular security. Instead, the exchange with the inferior NBBO is required to route the market orders to the exchange that has the best NBBO quotes for market orders at that particular time. If at any time, the exchange receives limit orders that are better than the prevailing NBBO, that exchange will now own the NBBO, and all the market orders will be routed there. The order routing may or may not be free of charge, depending on the venue.

As an example, suppose that the current NBBO for IBM stock is $155.14 for bids and $155.15 for offers (a spread of the minimum tick, $0.01, is usually present in all markets, otherwise arbitrage opportunities exist; the spread is also often the only compensation that the market‐makers obtain—more on this later). Suppose further that BATS BYX exchange has the following best quotes for IBM: $155.13 bid (200 shares) and $155.15 offer (100 shares). As always, the best quotes are determined by the best buy and sell limit orders present in the limit order book: the price associated with the best buy order becomes the best bid, and the price associated with the best sell limit order forms the best offer. If a market order to buy 100 shares of IBM arrives at BATSY, the market order is executed at $155.15, since this is the prevailing NBBO. If a plain market order to buy 200 shares of IBM arrives at BATSY, only 100 shares will be executed, and the other 100 may be routed to an exchange where NBBO is present, unless that exchange is still BATSY at a different price level. If a 100‐share sell market order arrives at BATSY, it will be forwarded to an exchange where the prevailing national best bid of $155.14 exists.

Figure 3.2 illustrates the idea.

Although the idea of NBBO works well in general, imperfections exist. First, the technology still has a finite speed as far as the collection, processing, and redistribution of quotes is concerned. As such, it is possible for investor orders to “fall through the cracks” and to be matched on exchanges where NBBO no longer exists. To the SEC's credit, the SEC mandates regular revisions to the NBBO submission and redistribution frequencies, making the data collection faster and execution fairer. As of August 2015, the government guaranteed the round‐trip aggregation and redistribution time of best quotes of at most 500 milliseconds (one half of one second). Since, proposals have been made to reduce the quote redistribution speed to as little as 5 milliseconds.

Although all exchanges are obligated to observe the SEC Regulation National Market Systems (Reg NMS) that mandates all market orders are executed at NBBO or better, due to the competitive nature of the modern trading landscape, exchanges differentiate themselves by deploying different pricing and matching combinations. Some equity exchanges offer traders monetary incentives to provide liquidity in an attempt to attract limit orders, and thus deepen available liquidity. Exchanges doing so are known as normal and offer “rebates” for providing liquidity (posting limit orders), while charging fees for taking liquidity (placing market orders). Other exchanges, known as inverted, do the opposite. They charge for limit orders and pay for market orders. The NYSE is an example of a normal exchange, while the Boston OMX is an inverted exchange. A few exchange firms have offerings in each category. BATS, for example, has separate normal and inverted exchanges.

Some liquidity is considered to be “toxic,” or detrimental to investors. Typically, toxic liquidity comprises orders that are canceled rather promptly, raising other market participants' questions about the intent of the providers of that said liquidity. Fees and other properties of exchanges affect the toxicity of their liquidity. Some researchers find that, on average, the fees across all the exchanges are in equilibrium, balancing the explicit fees with implicit costs, such as observed spreads. The lower the fee imposed on “liquidity makers” providing limit orders, the higher is the observed spread on a given exchange, potentially implying higher toxicity levels. Order cancellation rates are lower on exchanges with lower liquidity maker fees (higher liquidity taker fees), also indicating lower toxicity levels.

Still, gaps persist. Besides periodic data outages on exchanges, an illegal activity called spoofing can really distort the NBBO, as described in Chapter 4.

THE NEW MARKET HOURS

In addition to new trading venues and data standards, many other changes have occurred in the financial markets over the last 20 years. The markets are undergoing continuous innovation with the ever‐expanding presence of computers on the trading floors. Some products of automation, such as high‐frequency trading (HFT), have generated unprecedented attention, while other changes, significant to investors, have largely gone unnoticed. This section focuses on just one such change, extended market hours, and discusses the implications for investors, large and small.

Many years ago, when markets were dominated by human traders, financial markets worked standard hours: 9:30 AM to 4:00 PM. The timing allowed sufficient leeway for market professionals to prepare for the market opening, including gathering the latest news and other requisite information, and organize daily trade “tickets” at the end of the day prior to departing the exchange. The market “open” and “close” prices, often reported in the next day's newspaper, corresponded to trade prices recorded at 9:30 AM and 4:00 PM, respectively.

The trading hours were designed to suit a business schedule normal for most market practitioners. News that arrived outside of market hours, however, often caused much volatility and could not be traded in a timely manner. In response, an innovation ensued: About 10 years ago, many exchanges began offering extended trading hours beginning at 4:00 AM ET and closing at 8:00 PM ET. The extended morning opening hour coincides with the market open in London, and the extended closing time suits professionals in Asia, also allowing the US‐based market practitioners to trade closer to a 24‐hour format, capturing latest news in the markets.

While extended hours provide a longer window to execute trades, they also set a stage for several issues unanticipated by large and small market participants:

1. Changes in open prices. Quantitative financial analysis has traditionally been developed and taught on daily open and closing prices. With the introduction of extended market hours, the open prices are now often recorded at 4:00 AM, not at 9:30 AM as before. As a result, financial analysts trying to develop portfolio rebalancing or trading models based on market open may need to recalibrate their approaches.
2. Significant market movement outside of regular market hours. A considerable portion of market movement now occurs from the “new” market open to 9:30 AM. Traders and investors expecting to wake up and enter the markets past 9:30 AM may be subject to the “you snooze—you lose” formula, whereby most of the relevant news has already been incorporated by the markets prior to the regular‐market open prices at 9:30 AM.
3. Corporate earnings announcements again often fall during trading hours. In the 1990s, there was a lot of concern related to corporate earnings announcements during normal market hours, and the resulting volatility and potential market manipulation around the earning announcements. To circumvent the issues, more and more public companies began reporting earnings outside of the “regular” 9–4 trading hours, often at 8:00 AM and 6:00 PM. The new extended trading hours, however, put the issues surrounding earnings announcements back on the table.

These issues are not dealbreakers for trading, or an argument to revert the market structure back to its 9:30 AM to 4:00 PM format. However, investors large and small need to be aware of the changes in order to understand and optimize risk factors in their portfolios.

WHERE DO MY ORDERS GO?

The order maze befuddles many investors. So, you press that “Submit” button, and your order is executed, right? Wrong. Plain wrong. And what happens in reality depends on a multitude of factors.

First, was the order you sent in a market or a limit? Limit orders specify the execution price. If the market price is far away from the specified limit price, the limit order may execute with a considerable delay and may never execute at all.

Market orders are orders to buy and sell here and now, at the best available price. Still, even market orders have to wait their turn. Most investors' orders first end up on their brokers' systems when submitted. Brokers are entities such as Charles Schwab, JP Morgan, Pragma Trading, and Quantitative Brokers. They carry a special designation to do the best possible job on behalf of their clients. The designation is administered by the SEC and Financial Industry Regulatory Authority (FINRA) for equities, ETFs, and equity options, and CFTC and National Futures Association (NFA) for commodities and futures. Foreign exchange brokers tend to be unregulated.

At the point when the order reaches the broker, the order contains all identifying information: who the order is from (institution or individual), your account number, and so on. Next, the broker may choose one of the five ways to handle your order:

1. Send the order to the exchange for execution.
2. Send the order to a market maker.
3. Send the order to an Electronic Communication Network (ECN).
4. Send the order to a dark pool.
5. Match the order internally with other orders sent to the same broker.

For most publicly traded equities, brokers can send the orders to an exchange, such as NYSE. Some exchanges will compensate brokers for the flow with rebates. As discussed earlier in this chapter, some exchanges pay rebates for the market orders and some do so for the limit orders. Depending on whether your order is a market or a limit order, your broker may forward the order to a different exchange to maximize the fees the broker receives from exchanges. You, the end client, may or may not see some or all of the fees your broker receives on your behalf. The exact distribution of fees is typically stipulated in the fine print of your broker services agreement.

As an alternative to sending your order to an exchange, the broker may choose to send your order to a designated market maker. A market maker is a broker‐dealer who keeps inventory on hand and is available to match orders out of their cache. A prominent example of a market maker is Knight Capital Group (KCG). Market makers may also pay your broker for bringing in your orders, and you may or may not see any of those payments.

Still further, the broker may send your order to an electronic matching service known as an electronic communication network (ECN). An ECN is an alternative trading system (ATS) that matches orders outside of exchanges. ECNs match orders electronically. Unlike dark pools, ECNs display their quotes in the consolidated quote feed (SIP tape) that redistributes NBBO. An example of an ECN is NYSE Arca. The very first ECN was Instinet, founded as an inter‐broker dealer back in 1969. ECNs may also pay your broker for bringing in flow.

A broker may also choose to route your order to a dark pool. For example, Interactive Brokers (IB) clearly states on its website:

IB maintains connections to “dark pool” ATS's (including the IB ATS) that execute a portion of IB customer stock orders. IB customers benefit from IB's access to dark pools. Dark pools provide a source of substantial additional liquidity. Dark pools charge no execution fees or lower execution fees than exchanges. Dark pools also provide fast executions and the possibility of executions at prices more favorable than the prevailing NBBO.

Source: https://gdcdyn.interactivebrokers.com/Universal/servlet/Registration_v2.formSampleView?ad=order_routing_disclosure.html

Finally, your broker may not choose to send your order anywhere at all, and instead match it internally with an opposite order on its internal books. For example, if you submit a limit buy for 100 shares of IBM at $155.14 and it comprises NBBO, and your broker receives a market sell order for IBM, the broker will match your order with an incoming market sell order without forwarding your order on. The process of intra‐broker execution is called internalization. Brokers are required to internalize all commodity and commodity futures orders by law. The law was created to avoid money laundering that was apparently happening when the “dirty” money was traded on the exchange into a “clean” account by the brokers who housed both accounts. The broker may still charge you the same commission as it would have if the broker forwarded your order on.

As long as your order stays on the broker's premises, your account information is visible and attached to the order. When the order leaves the broker, it automatically loses all its individual account identity and becomes associated only with broker ID. That's right—by the time your order reaches an exchange, it effectively becomes anonymous, lost in hundreds if not thousands or millions of orders your broker processes on a daily basis. No market participant outside of your broker knows who you are and what you trade, unless, of course, you dominate your broker‐dealer flow.

EXECUTING LARGE ORDERS

Brokers' orders may be more numerous than one would expect. Part of most brokers' business is best execution: The ability to slice large orders and massage the parts into the exchange order flow without causing panics, crashes, or market exuberance. In other words, many brokers are commissioned to process a large order while leaving as little impact as possible in the markets. How do brokers do that? Why, with the algorithms, of course!

Execution algos are computer programs that are designed to break down orders into small chunks and then optimize order timing and routing so that the order obtains the best execution. Suppose you want to sell $1 billion of British pound, GBP/USD. If your order hits the markets in one piece, two things happen: First, it will immediately wipe out all available limit orders, possibly causing a crash; second, the price of your execution will be horrendously low as you will “sweep the book,” picking up all terribly priced orders to satisfy your appetite. If, on the other hand, you break down your order into small chunks and spread those mini‐orders over time, you will give a chance to limit orders to rebuild naturally, and may obtain execution close to what the market price would be if your order did not exist at all!

Two common execution algos are used across all markets. The simpler one breaks down a large order into an equal number of pieces, where the number of resulting mini‐orders is specified by the client or is a function of the broker's secret sauce. With this algo, known as time‐weighted average price (TWAP), the number of the orders corresponds to the desired frequency of execution times the length of execution. The main advantage of TWAP is its simplicity.

The main disadvantage of TWAP is that it is very mechanical and completely ignores regularly occurring trading patterns. For instance, it is normally the case in equities to have high trading volume at the 9:30 AM market open, a slower late morning, and even slower lunch hour, and then somewhat of a resurgence ahead of the market close. In other words, the trading volume in the equity markets follows something of a U‐shaped pattern. The higher the volume, the easier it is to massage in larger orders without moving the markets with those particular orders. Furthermore, across individual stocks, the volume patterns are similar from one day to the next, allowing for a fair degree of intraday volume predictability based on its historical patterns. Enter the volume‐weighted average price (VWAP) algo—essentially, a TWAP, with TWAP timing of orders, where the size of the individual orders is modified according to the historical volume curve: higher in the morning, lower through the midday, and higher again at the market close.

VWAP has been such a hit in equities that it has become a de‐facto standard in execution, against which all other execution methodologies are measured. Of course, it is not 100 percent perfect. For one, you can outperform it with an overlay: a strategy that requires a slight modification of VWAP and potentially delivers a substantial gain. Companies like AbleMarkets deliver overlay services, among other data. Second, while the small packets of orders are mixed up in the anonymous markets, the patterns may still be clearly visible.

TRANSACTION COSTS AND TRANSPARENCY

Regulators, aware of data capabilities and the risks hidden in today's markets, have proposed higher transparency requirements on the whole industry. For example, the new pan‐European market regulations from MiFID II have a direct impact on the structure of brokerages in Europe. Specifically, MiFID II dictates that all brokerages are required to demonstrate best execution and provide full disclosure and transparency on the following items: price, transaction costs, speed of execution, likelihood of execution, trading venue selection, and so on. While these metrics seem to be obvious priorities for investor disclosure that should be adopted by the US regulators as well, these long have been the “secret sauce” of many execution brokers.

So where is the brokerage industry going under the new regulations? Technology is certainly not only enabling the requirements of transparency, it is also leveling the field as far as investors are concerned, making broker‐shopping easier. How are brokers to retain their clients?

The answer, once again, lies with technology. Smart order‐routing solutions should enable brokers to compete for clients beyond taking them to beer outings and popular concerts. A solid example of someone who has been doing this well for the past decade in US equities is Pragma Securities: leveraging PhD‐level research and the technology to deliver benchmark‐beating routing to their clients. However, even Pragma cannot fully disclose its secret sauce—doing so would make it vulnerable to the competition and likely affect its business considerably.

Research on how to enhance order routing is not straightforward and does not come cheap—retaining the brains from defecting to competition and spilling their knowledge there is not just a matter of bullet‐proof contracting. And the competition does not come just from other brokers—many successful hedge funds and prop trading shops are now setting up their own execution divisions to avoid brokerage costs and leaking information about their trades to a third party. Companies like AbleMarkets provide off‐the‐shelf solutions to beat the competition in execution, by tracking aggressive HFT activity, for example, making the job of executing brokers easier and more profitable. The long‐term future of many brokers, therefore, depends on sound investing and partnerships with the right research providers—competing on price and intangible perks like beer outings alone is a treacherous path for survival.

CONCLUSIONS

The changes sweeping the financial markets can be mind‐boggling. Most are driven by advances in technology at an ever lower cost, be it in data processing or storage. Computers take market paradigms to previously unthinkable constructs. These are changing trading and execution as a business, creating a slew of previously unknown risks, and magnifying the impact of formerly marginal risks. Investors should be aware of the developments in the market microstructure space and use the latest technology advances to protect their portfolios.

END OF CHAPTER QUESTIONS

1. What are alternative trading systems (ATS)? What categories do they comprise?
2. How does a limit order book work?
3. What is NBBO? How is it produced?
4. How do brokerages execute client orders?
5. What new regulations are proposed in the order execution space?