Many thanks to Flashboy00 over at Y! Finanace
for posting his thesis regarding the NYSE Group. It is outstanding! Jam-packed with a ton of info. The post is a bit difficult to navigate since it's divided into seperate posts. For everyone's convenience I thought I'd repost it in its entirety here. So here it is!
The following is a copy-n-paste of the original message:
(or why I am buying the global financial tollbooth)
What is an exchange?
Mulherin et al (1991) stress the definition of a financial exchange not as a market, as usually is done, but as a firm that creates a market in financial instruments and thus has the property of the price information produced. Furthermore, a security market can then be seen as a firm that produces a composite good, the exchanging of securities, which may be formed of different elements (Padoa-Schioppa, 1997): price formation, counterpart research, insurance for a good clearing, and the standardization of the good exchanged. The “production cycle” of these businesses is divided into three parts: listing, trading, and settlement. These three different stages of the
production cycle lead to the formation of different goods that the exchange can sell: listing services, trading services, settlement services, and price-information services. Many exchanges have dropped the settlement service as has the NYSE.
Exchanges: toll roads on capital markets volumes
Buffett has remarked that the best investment in the world is a toll booth. Dominant exchanges follow this toll booth model. In general, whenever an equity is bought and sold a fee is paid to the exchange. When a new company “enters” the public company world it must pay a fee to the toll booth. Also, every year that company remains public, the tool booth collects a fee. Because of the toll booths strategic location it is also an invaluable source of market information which it sells as well.
Exchanges are geared investments on capital market volumes. That said, they are less geared to market value than brokers since trading earnings are in part driven by the number of trades rather than purely their value - much like a toll road. So while this is a cyclical industry to some extent, i.e. there are less IPO’s during a bear market; the revenues of an exchange do not follow the slope of the Dow Jones or any other average. The recent bear market in Europe is demonstrates this well. Take the UK in 2002, which was the most extreme: while the FTSE fell ~25%, the value of trading only fell 4% and the number of trades grew 16%. LSE equity trading revenues were up 8%. It seems that new trading strategies are slicing and dicing orders and generating more trades and this has resulted in ever higher volumes.
Where does traditional exchange revenue come from?
Generally revenue comes from three major sources plus two eventual or
Taking an example from European exchanges these sources are listed below along with the percentage of revenue contributed by source.
(1) trading fees -both membership fees and trading fees (28%)
(2) listing fees -both initial and yearly listing fees (32%);
Listings are a very lucrative source of revenues for those exchanges that serve as a market of original listing. Companies pay significant fees to have their securities listed on the most important and prestigious exchanges. However, several issues cloud the revenue potential for listings. First, an exchange must be a preeminent exchange to charge for listings. Second, this source of revenue is geared toward market conditions. The recent bad market in 2001 has shown that IPO’s certainly drop off and companies withdraw from secondary listings(i.e., on other than their home market) listings when volumes decline
(3) information and price-dissemination fees (17%)
Market data has been a great source of revenues for exchanges and has been resistant to downward price pressures that have caused many market data vendors to become less profitable. Pricing pressure on exchanges has had little impact because of both the lack of competitive sources for the information exchanges provide and, in some cases, of data collection and distribution consortia (e.g., the Consolidated Tape System and Consolidated Quote Systems for US equities and the Options Price Reporting Authority for US options) that protect the member exchanges from price competition. Nevertheless, two trends could end the price stability enjoyed by exchanges. First, prices and quotes are less important to traders and investors than they once were. Those exchanges that provide electronic trading systems provide market information that is 'actionable' (i.e., the quotes can be hit by the viewer) as part of the trading system. This diminishes the value of quotes distributed through traditional vendors. Second, traders and dealers are able to 'shop' their executions among markets for reporting purposes. Recently the Island ECN in the United States began printing its trades in NASDAQ listed issues on the Cincinnati Stock Exchange instead of through NASDAQ. These factors may mean that revenue from data may not be as lucrative in the future. The data that comes out of a major hub like the NYSE may however retain its importance in the market.
The other sources of revenue are:
(3) settlement fees (16%)
Exchanges are transferring this activity to specialized entities where exchanges have been, among the shareholders.
(4) other revenues may come from the developing and selling of proprietary software and information technology (20%).
Where does traditional exchange expense come from:
Not all the costs contribute directly to the production of the three main “goods" sold. There are the costs for the regulation and supervision of the market that make it more efficient, and thus more attractive, for issuers and intermediaries to enter. Many exchanges have created arms length corporations (as the NYSE is proposing) to handle these serves, although the costs come out of exchange revenues. Regulation of the markets is critical for the reputation of exchanges, to facilitate daily trading and to persuade would-be traders and investors to use the exchange. However, it is difficult to find any effective method to pay for these services even though they represent a significant portion of the costs of an exchange. Therefore one of an exchange's largest costs must be paid for out of other revenues.
There are R&D costs and marketing costs, and there could be an implicit cost in charging low or zero fees to traders and issuers.
Who are the customers?
The customers of exchanges can be divided into direct and indirect customers:
Direct customers are the direct purchasers of exchanges' services. The most important are issuers (the companies on the exchange) who usually pay fees in order to be listed on an exchange; intermediaries who usually pay fees in order to be admitted to trading; and information vendors who pay fees to have the right to disseminate price information of the market (Reuters would be an example)
Indirect customers are all the entities that send orders to intermediaries to
be executed on an exchange. These include institutional investors and all other financial intermediaries.
Why they trade on the exchange
The choice to trade on any particular exchange involves several factors. A major factor is the overall “quality" of the exchange; price factors could be important as
it is possible that part or all the cost of transaction paid by the intermediary
is passed along to the customer. Also, market microstructure, either as liquidity, price discovery, or quick execution of the orders, is different on different exchanges (i.e. order or quote-driven markets.) Finally, market reputation and regulation can influence the choice.
The changing environment
For those who have been following the European exchanges over the last few years it was clear that there have been benefits for the owners to the exchanges becoming public
entities. These have seen some great market cap growth. Along with going public, exchanges have started to compete with each other for business. In an increasingly globalized environment the traditional barriers that would keep other exchanges away from companies from other countries have fallen. Exchanges were once seen as geographic entities. This idea of exchanges is less and less valid. What they are evolving into is industry specific entities. As an example, the TSX has the greatest collection of mining companies than any other exchange from multiple countries. There are advantages for companies to be listed where their peers are listed as this is where the market is for these companies. Hence an exchange from a country half way around the globe can become a major competitor for listing “native” companies of an exchange in a different country.
It is in this environment that the NYSE had found itself. Staying private and “non-profit” would have lead to reduced market share and eventually to niche markets developing on foreign exchanges (like the TSX mining example) which would have attracted companies away from the NYSE. To industry observers it should have been obvious that they would seek to compete on a level playing field and go public.
How they went public took me by surprise. The traditional exchange has gone through a process of demutualization and then going public. This can take several years. For several reasons, the NYSE did not have this time. They used a clever and savvy deal that in one shot got them one of the best electronic platforms in the industry, a 25 percent share of NASDAQ trading, plus they get to demutualize and go public, and eliminate a competitor. The deal positions the NYSE to not only retain the 25 percent share of NASDAQ trading, but also positions it to get into options and take a bigger share of ETF (exchanged-traded fund) transactions – both of which represented significant aspect of the archipelago business. This all occurred on the eve of the passage of Reg NMS and the extension of the trade-through rule to all U.S. equity markets. Not a coincidence I think. (more on RegNMS later)
A survey by the Swiss Stock Exchange found that there are no economies of scale for the
listing and non-trading services of stock exchanges. The study showed that being
bigger does not produce any material benefits in these areas. The trading function however, generates significant benefits in scale as the market for an asset or security develops added liquidity and depth as counterparties present themselves in greater volume. Liquidity itself breeds some powerful benefits, such as the ability to trade in the size, at a time, and as near to a desired price as possible. It also generates a “virtuous circle” of attracting more traders and products, which creates more liquidity, which attracts more traders and products—and so on. Due to the network nature of business, volume tends to consolidate in one marketplace, in both space and time.
How does the NYSE perform on liquidity? Currently, 90% of orders of 10,000 shares or more go through the NYSE, which suggests that it is a deeper market than any of its rivals. This also explains the paradoxical conclusion contained in a study of 100 large institutional traders conducted by Plexus, a research firm. This study found that, despite much bigger commissions on the NYSE than on NASDAQ, the total cost of a trade (ie, combining commissions with price deterioration) is on average one-third lower. This attests to the powerful competitive advantage of high liquidity. A competitor can try to start a price war by cutting trading costs – but even with much higher trading costs a dominant exchange will still have less overall costs of trading because the biggest cost to any investor is the bid-ask spread.
There will be a strong moat around the NYSEs trading functions based on the natural
Monopoly of liquidity. But what about the listing service. The listing services will benefit from a different type of moat. That being the cache value and brand of the NYSE itself. Most companies would rather have an NYSE listing as it legitimizes their enterprise.
SEC approval is necessary to start a new exchange and this will effectively keep most potential competitors from setting up shop. This represents another barrier to entry and widens the moat further.
The other aspect of what makes this a great business is that the costs of running an exchange are basically fixed – with increasing revenues not requiring significant increase in capex. Revenue growth tends to fall to the bottom line with even small amounts of revenue growth leading to relatively larger increases in net income. As this is occurring the switch to a cheaper electronic platform also will be occurring (which will continue to lower costs and increase margins) and as such net income should grow at a much faster rate than revenue. This basic theme can be seen most vividly in the revenue and net income growth of the cme. CME's first-quarter revenue grew 32% from the previous year to $224 million. And profit was up disproportionately at 51%.
To demonstrate this for yourself you can build a spreadsheet and let revenue grow by 4% per year only. If you set the net margins at 20% (similar to some exchanges) then earnings will grow long term at 11%. Small changes in revenue lead to large changes in earnings when costs are fixed.
As Alan Greenspan has recently stated, exchanges tend towards natural monopolies because of the liquidity issues noted above. Because of this fact, there is likely always to be some element of regulation in this industry. For American exchanges, regulation comes from the Securities and Exchange Commission. In fact, it was a recent SEC rulechange that acted as the catalyst for the merger of NYSE with AX and the overnight transformation of the NYSE into an electronic exchange. This rule change is known as Reg NMS (The trade through rule is the most significant rule change of reg NMS). And in fact, some believe that the whole reason for Reg NMS was to pressure the NYSE — which handles about 80% of listed stock trading every day — into moving more rapidly to increase automation and lower costs to investors.
The trade through rule basically states that a trade must go through the exchange with the best price. It is illegal for a broker to put a trade through on exchange A if exchange B is posting a lower price for the shares.
Now it is important to understand that the trade through rule has been in existence for many years and the recent ruling did not abolish the trade through rule. The recent regNMS change was to (1) extend the trade through rule to ALL markets including NASDAQ (NASDAQ is not officially an exchange and has not been subject to the trade through rule in the past) and (2) insist that these trades occur within one second and not the 30 seconds that the NYSE was taking to process these trades on the old floor. Crucially, it also proposed giving customers the right to waive the rule. Regulation NMS, will become fully effective in 2006 with the 'best price' component fully effective on June 12, 2006.
The trade through rule was instituted decades ago to protect regional exchanges from the NYSE. Ironically, it ended up protecting the NYSE from electronic upstarts. Under the rule, the NYSE would get a trading order if it has posted a better price. It then had 30 seconds to confirm or reject the transaction. In financial markets, 30 seconds is a lifetime. By that time the price has often moved and the potential ECN customer is gone. This also ignores the “depth of book” i.e. if there are only a few shares available but it represents the best price then the NYSE gets to see the transaction. But because the NYSE was so large it likely also had a lot more shares available at any given price.
The rule allowed the NYSE to see almost every transaction and retain those that it wanted. That, in turn, gave the exchange extraordinary volumes, which in turn allowed it to provide better prices. In a sense this is an extension of liquidity begetting more liquidity and the trade through rule has functioned as another competitive advantage for the NYSE.
In essence, with Reg NMS, the competitive advantage represented by the trade through rule provision remains intact; it just needs to move a lot faster than 30 seconds. Hence, the rule will eventually give priority in filling stock orders to exchanges that provide automatic execution -- defined as the ability to process an order in a second or less.
To put things in perspective, the SEC's Office of Economic Analysis found that only approximately 2% of all trades on NASDAQ and the New York Stock Exchange were trade-through in nature.
Some see the new regulation as a set back for the NYSE. This is wrong. It could have been so much worse. The nyse nightmare scenario would have been if Reg NMS adopted a reform that would mandate depth of book (DOB) order routing, a practice that could gravely weaken the New York Stock exchanges advantage in listed business.
That’s because mandatory DOB routing eliminates intermarket competition
by giving any limit order, regardless of where it was placed, the same protection.
This would effectively transform all exchanges into a large virtual electronic Consolidated Limit Order Book, or CLOB. Under this system all exchanges become a piece of a larger electronic network turning a capitalistic company into a commune. This fortunately did not occur.
The nightmare scenario for the NASDAQ was that the trade through rule got extended to include the NASDAQ and it did! Now the nyse can use their larger market in certain cross traded stocks to steal business from the NASDAQ. The rule as it came down was neutral to the NYSE (as long as it merges with AX) but is bad news for the NASDAQ which now must face the possibility of orders getting routed to the NYSE which did not occur before as it was not under the regulation of the trade through rule.
In most major industrial cities where businesses were starting up, requiring investment capital to grow and thrive, stock exchanges acted as the interface between suppliers and consumers of capital. In the UK and US stock exchanges were found in most major centers. This has changed. In Canada, the smaller exchanges have merged into the TSX leaving only Montreal as a derivatives exchange. The trend toward exchange consolidation is most active in Europe such as the unification of the U.K. stock exchanges in 1973 and, more recently, the three major Swiss exchanges in Basle, Geneva, and Zurich. The single European currency is eliminating barriers to moving capital and assets across country borders and removing both interest differential and currency variance. These trends will focus liquidity into one market space to the exclusion of local interests.
The merger between NASDAQ and the American Stock Exchange was a
recent major-scale move in this direction. Very recently another wave of consolidation has occurred and this involves the elimination of ECN competition. With Archipelago going to NYSE and Instenet going to NASDAQ, Bloomberg LP's TradeBook is one of the last free-standing communications networks. After the two deals are completed, the NYSE and NASDAQ will emerge as the two main cash equity trading centers in the United States, each competing for the other's order flow. It will leave one remaining electronic communications network (TradeBook); a handful of small, regional stock exchanges; and the American Stock Exchange, which trades mainly options and exchange-traded funds. The trend is toward creating a duopoly in the US for stock trading.
As a general rule one is much better off investing in an industry where competition is shrinking than one where it is expanding. This will only make the NYSE an even better investment.
Duopoly (the NASDAQ)
The NASDAQ will be the main competition for the NYSE. The model used by the NYSE and NASDAQ are very different and this must be understood to assess the competitive landscape.
In the NYSE model of a stock market, all trades in a specific NYSE stock are routed to a specialist and a floor broker. To trade stocks on the exchange floor, a brokerage firm must be a member of the exchange — in other words, own a seat on the exchange. Broker-dealers who don’t belong to the exchange can route their orders through a member firm. Trades are executed according to an auction style of market making
Here, the specialist is responsible for:
a) executing the trade
b) maintaining liquidity
c) maintaining the capital commitments for the stock
It the specialists job to know, with the help of the order book to which they have exclusive access, the constantly changing supply and demand for their stock. They must make certain that buy and sell orders are in balance, and that trading moves smoothly.
The NASDAQ model of an exchange is very different from the NYSE. NASDAQ is more of a model of many parts coming together as a whole. It is one of competing market-makers supported by a nation-wide system of quote screens. On average there are 16 market participants for every NASDAQ stock. There is plenty of competition—i.e. there are 323 market makers handling Microsoft trades, for example. The competing market makers are composed of 1) independent broker dealers (like market makers –but they do not commit capital); 2) market makers – major brokerage houses which commit capital ; 3) ECNs.
They all compete for trades based on best bid/ask. The problem with NASDAQ’s model is that there can be no one single dominant market participant. Competition is very high, and as ECN's have become more popular, they have eroded NASDAQ’s revenues from trade executions. When you hear of ECNs gaining market share this has been primarily a NASDAQ phenomenon and is occurring because the NASDAQ model allows this to be the case. A large portion of NASDAQ-listed stocks have been traded on regional exchanges and through ECNs in the last six years. Compare this to the NYSE who uses (or used to use) the trade through rule to grab an order from the ECN (because it listed a lower price) and hold it for 30 seconds –an eternity in the world of trading- allowing the price of shares to change before deciding if it would match the order.
The merger with AX allows the NYSE to retain the 25 percent share of NASDAQ trading.
Trading in NASDAQ-listed stocks has been fragmented since a series of reforms were enacted in 1999. NASDAQ has made an effort to defrag that market by acquiring several ECNs, among them Brut and the newly announced takeover of Instinet. In acquiring Inet, NASDAQ will gain a robust, leading-edge trading platform. Instinet has the leading technology on the planet, with response times of 5 milliseconds for incoming orders. With Instinet holding about a 24% share of trading volume in stocks not listed on the NYSE, the deal boosts NASDAQ’s market share to between 60% and 65%. But it does not give NASDAQ a footing in NYSE listed issues.
Does the instenet merger give NASDAQ any competitive advantage over the NYSE?
The NYSE has a much greater capability to trade unlisted securities than NASDAQ has to trade NYSE-listed securities. Not only is the NYSE gaining an industrial-strength E-trading platform with Archipelago's ArcaEx, but it's also gaining Archipelago's 22% share in unlisted trading volume. NASDAQ, by contrast, already had a decent E-trading platform of its own, and Inet's share of trading in NYSE-listed securities is insignificant.
Some are worried about a flood of companies leaving the NYSE to join NASDAQ but history suggests otherwise. From 2000 to 2003, 109 companies have moved to the NYSE from NASDAQ. In NASDAQ's 32-year history, only one company -- Aeroflex Inc., a Plainview (N.Y.) microelectronics company -- has moved to NASDAQ from the NYSE.
NASDAQ – And the uneasy relationship with market makers
Many of NASDAQ’s problems stem from its structure. Firstly despite being a for-profit company that has IPOd – it remains under the NASD stodgy regulators that keep it unable to compete with ECNs. Also, it jumped the gun on the IPO – and did so before actually receiving exchange status from the SEC. It is still waiting for SEC approval.
Unlike the NYSE, NASDAQ isn't an exchange -- a market with a single order book showing the trading interest in stocks among retail investors, mutual funds, pension funds, and hedge funds. At the NYSE, the vast majority of orders come together on the trading floor, where bidders fight to get the best prices. NASDAQ has no trading floor; instead it runs over a sophisticated telecom network linking about 600 customers in 1,000 locations. Quotes and orders come from three major sources as described above: market-makers, or dealers, such as the trading desks at Goldman Sachs and Charles Schwab , Capital Markets; the ECNs, some of which are backed by the big Wall Street firms; and online brokers such as E*Trade and AmeriTrade . NASDAQ collects and displays everyone's data, then consolidates them into a single tape.
Its new electronic trading system, SuperMontage was supposed to take on the ECNs but it didn’t work out this way because of NASDAQ’s conundrum that if it makes the best electronic trading system it will lose the market makers who give it its liquidity. The SuperMontage system lacked many of the features that attract big institutions to ECNs. Large traders don't dare advertise that they want to buy, say, 100,000 shares of Cisco for fear other players will bid the stock up. ECNs allow traders to trickle out orders without revealing their full hand. To avoid angering the market-makers, SuperMontage left out such features. This made SuperMontage irrelevant from day one.
The main issue for the SEC is "price-time priority," or making sure that customers who first offer the best price for a stock have their orders filled first. NASDAQ has never done that. For about a third of trades, dealers match buyers and sellers in their own order books -- without giving any other market participant a chance to fulfill the order. The resulting "internalized" trades are then reported to NASDAQ as done deals.
The SEC values price-time priority because it rewards the most aggressive buyers and sellers and produces fairer prices. It also ensures that market insiders don't ignore retail investors. The SEC has never approved an exchange that did not force orders to interact . But NASDAQ's market-makers count on these highly profitable in-house trades, on which they collect commissions from both the buyer and seller plus the spread between the buy and sell prices. If NASDAQ implements price-time priority, the whole market-maker model comes into question -- and all bets are off. Dealers are vital to NASDAQ because they keep trades flowing -- by posting quotes, maintaining an inventory in the stocks they handle, and putting up their own capital as buyer of last resort in disorderly markets. A market that depended solely on ECNs, which don't have such obligations, could break down under stress.
So NASDAQ is caught in a dilemma: The SEC may not let it become an exchange unless it alters how it works. But if it changes the way it works, megadealers such as Schwab and Knight Securities might flee, further reducing NASDAQ's relevance
The nagging ECNs
ECNs display customer orders electronically and provide access to substantial pricing information through their electronic limit-order books, which are accessible through the Internet. The firms directly match buy and sell orders without the intervention of human intermediaries such as Nasdaq market makers, the intermediaries who buy and sell NASDAQ securities. Island, an ECN specializing in the retail market, matches orders within its system. Archipelago, the fourth-largest ECN of the nine currently in operation, allows participants to interact with NASDAQ and all ECNs and allows orders to automatically go off the system if a better price exists in another venue. In addition, with no human intervention, orders are less likely to be mishandled. Market makers have the ability to manipulate the quote, delay the trade until the most advantageous time, and widen the spread. The ECNs avoid this problem by removing the market maker.
In general, automated trading systems free-ride on the process of listing, given that they generally trade only securities listed on other exchanges; furthermore, they sometimes free-ride on the price-discovery process given that members of exchanges may direct trade on ECNs to make some arbitrage or to operate in off-regular hours. But the main customers of ECNs seem to be institutional investors that generally are not allowed to trade directly on exchanges.
ECNs have taken some market share but an ECN has never taken the place of an existing exchange. Part of the problem is that while ECNs can display a better price than an established exchange is quoting, investors also want to know how many shares a price is good for: that is, how deep and liquid the market is. This is a big worry for institutional investors, which spend much time and money trying to minimize the risk that prices will move against them when they do a large trade (“price deterioration”, in the jargon). This is why ECN s, which were originally regarded as a huge threat to stock exchanges, have often failed to live up to expectations. Their prices have no depth, because they have little inventory.
Some believe the regional exchanges still have a chance. The combination of the NYSE and NASDAQ mergers and Reg NMS (with the extension of the trade-through rule to all exchanges) means that having the highest pools of liquidity will be a major advantage and this all puts real pressure on the small regional exchanges to survive. With respect to the ECNs, there is now only one ECN system left (TradeBook) (assuming the NYSE and NASDAQ mergers go through). But new ECNs may be developed and the financial barriers to entry are falling. The days of $100 million developments are over, and it is now possible to build a very nice facility, hardware and software, for under $10 million. But we should not forget that the ECNs arose in an environment that created a niche for them to exist in. They were the most successful under the NASDAQ system. They were able to take a little business from the NYSE while it was a non-electronic not for profit entity. But do they really have any chance against the NYSE now that the niche is closed with a new electronic NYSE? An instructive example is the TSX in Toronto. The TSX has been an electronic market for many years (it was the worlds first all electronic exchange). There is no real ECN competition to the TSX. Why is this? Because the TSX did not allow the niche to exist. People that worry about ECN competition for the NYSE are driving and looking through the rear view mirror and not the windshield. I have no doubt that a faster electronic system can be developed – admittedly better trading systems than the CME’s globex have been developed but they have not been able to take
business from the CME. The competitive advantages of liquidity cannot be overcome with better technology.
First let me start by stating that an analysis of the new company, the New York stock exchange group inc, is very difficult for the mere fact that the company does not exist. The main criticism of this valuation will be related to the assumptions that need to be made. This is inevitable, as the company does not exist. The biggest risk to shareholders buying into the company at this point is that the company will not exist – i.e. the merger will be derailed. With that aside I will lay out my preliminary thesis as to the value of Archipelago (AX) shares as part of the new company.
It is important to point out that the combination of these two companies is greater than the sum of the parts. To merely value each as a stand alone and to add these together misses some very important points. First, and I believe most importantly is the fact that the NYSE becomes a for profit public company as part of this merger. The present revenues of the NYSE are absolute rock bottom revenue for this company. They are essentially not trying. What “trying” will add to the revenues depends on a lot of things but mainly on whether the years it has spent in the dwindles has mortally wounded the brand and the exchanges competitive advantages. I don't think it has and as far as capital market places is concerned; this is potentially the biggest in the world. The NYSE has the potential to become a truly global marketplace. The second point is of course the efficiencies that come about from taking NYSE revenues and running them through a more efficient electronic network. Operating margins of the NYSE are 3.4% while operating margins for AX are 22.1% - so its easy to see that if we could run the NYSE revenue through AX a lot more cash comes out the other end. But of course even something as intuitively obvious as this is not a slam dunk here. On the conference call, Thain insists that the two companies will run parallel. The NYSE will run just as it has for years with floor traders and AX will be a separate entity. There is no plan to merge the liquidity pools. Specialists on the floor will have no access to AX technology…..What the F&*%K!! That's right. Then there is mention that the market will decide whether they want to use floor traders or electronic networks. All I can surmise from this is that Thain is treading carefully. He does not want to tell the specialist that their days are numbered. It may very well be he will keep the specialists on for a long time and slowly phase them out – even at the expense of profitability. This may be the price that must be paid for this merger to work.
The two exchanges will form a publicly traded company that is 70% owned by the NYSE and 30% owned by Archipelago. Owners of NYSE seats will receive 70% of the stock in the new company and about $300,000 in cash, Thain said. All told, the NYSE will be handing out about $400 million to seat-holders. Archipelago shareholders will hold about 30% of the stock in the new company In order to value a company that does not exist I think it is important to focus on what we know is actually there. That means stay high up on the income statement. That's right – the revenue. As a combined entity the revenue of this company will be 1.076 billion (from NYSE) plus 528.6 million (from AX) = 1.6 billion in combined revenue.
Now it gets tricky. Yes I will make an assumption. But I think it is an educated assumption. I will assume the net margins of a public/electronic NYSE will be much higher than they are today. Because this will be controversial let me list the operating margins of the worlds exchanges.
Singapore exchange 52.8%
Hong Kong exchange 52.2%
Australian Exchange 48.3%
London stock exchange 36.4%
Deuche Borse 34.7%
Pretty impressive margins.
Even small players like Archipelago have had operating margins of 22.1%
The operating margins at the NYSE = 3.4% !!!
The reasons for this are (1) Not for profit entity (2) Non-electronic exchange – both problems will soon be fixed.
If we look at net margins of the worlds exchanges:
TSX - 33.2 %
ASX - 38 %
London exchange – 27%
If I had to pick a net margin number that the NYSE could eventually achieve, I would choose 20%. This is much lower than most of the worlds exchanges and I hope conservative. I think there is an equal probability the number could be higher or lower than this number. Now 1.6 billion in revenue at 20% net margins will produce and income of $320 million. The proposed NYSE Group will have roughly 160 million shares outstanding (47.8 million AX shares are worth 30% of company). AX shares closed at the time of this original valuation at $29. If the company can indeed produce this kind of net income then it only needs to grow income by 3% per year to justify this price. How much the company can grow is unknown. I believe however that a 3% growth in net income would be easily achieved. This is the crux of my belief that there is a margin of safety in the present valuation.
Transaction revenue forms the biggest part of the combined companies revenue stream. The company has guided that they intend to grow transaction revenue by 15% per year.
The company thinks growth will come from derivatives, corporate bonds, and new market listings that presently are going to NASDQ ie they will form a marketplace for small cap stocks on AX.
Companies like the TSX have stated income growth goals of 10% per year long term. (so far the TSX has beaten this bench-mark) If this company could grow net income at 10% per year long term (15 years) the present fair value of AX shares would be $52. And remember, because these companies operate with large operational leverage, (expenses are relatively fixed compared to revenue) an increase in revenue much less than 10% can achieve a 10% increase in net income.
My valuation is conservative for many reasons. I suspect net margins could go higher than 20%. The company has indicated that expenses of the combined company will fall $100 million a year for at least the first three years.
Valuation Model 1- The TSX
The present market capitalization of the TSX is $1.2 billion. The US economy is 10 times larger than Canada and on this basis one would expect a market cap for the NYSE of $12 billion. At an AX price of $30 per share, this put the NYSE group capex at $4.8 billion and potentially more than 50% undervalued. Of course this is simplistic as the NYSE may operate as a duopoly while the TSX is more of a monopoly – but the NYSE also has more potential as a global marketplace than the TSX.
Valuation Model 2– The London stock exchange
The UK GDP = $1.782 trillion (2004 est.) while the US GDP 2003 is $11.75 trillion. Roughly the US economy is 6.6 times larger than the UK.
In terms of growth at the LSE. Operating profits are growing 17% 2003 to 2004 and revenue growth was 6% 2003 to 2004. Again one sees the effects of operational leverage with small incremental revenue growth leading to larger operating income growth. The total 2004 revenue of the London exchange = 237.1 million pounds (which converts to $434.2 million revenues). To adjust for size of economy multiply by 6.6 which gives you $2.86 billion. This makes the present NYSE market cap only 1.7 times sales.
Valuation Model 3 – Europe
The whole US market is about the size of all of Europe.
It is estimated that all European exchanges had combined revenues of some ~EUR 5.2bn in 2002 (which converts to $6.4 billion US ). Profit margins were 26% for the combined European exchanges. This comes to $1.66 billion US in profit for the European exchanges. If the NYSE could do half this revenue at similar margins this would result in earnings of $830 million. The recent $5 billion market cap would give the NYSE a very
low PE of 6. The euro-zone is roughly equivalent to the US economy therefore this is not an unreasonable comparison. (note some of these revenues are certainly going to derivative exchanges which in the US are dominated by the CBOT and CME – as such I do not expect the NYSE to attract these revenues and this is why I chose to half the total European revenues). Even if you thought the NYSE would only get one quarter of these revenues then this gets you to a PE of 12. Does that seem appropriate for the NYSE?
Breaking down the revenue
The above valuation based on present revenue of the NYSE. Keeping in mind that the US economy is 10 times larger than the Canadian economy, it is interesting to compare the revenue components of the NYSE and TSX.
In listing fees the NYSE has $320.9 while the TSX generates 40% of this at $127.1 million. In terms of Data processing fees The NYSE generates $220.7 million but there is no good comparison here with the TSX and this is charged by SIAC (NYSE subsidiary) to the customers other than NYSE under a cost recovery model. In terms of
Market information fees the NYSE generates $167.6 million vs. the tsx $58.7 million which is 35%. In terms of Trading fees the NYSE generates $153.6 million vs. the TSX $92.6 million – which is 60.3 % of the NYSE trading fees!! With an economy one tenth the US economy, the TSX is generating revenue that is close to NYSE revenue. Is the NYSE really trying. Can it do better as a for profit entity. I think that it can!
More and more countries are acquiring the “cult of equity”, since they have decided that the process of people putting money into the stock market and companies is the best mechanism for allocating long-term capital within their economies. Also, greater application of quantitative trading strategies is overall increasing trading volumes on exchanges.
The graph below is instructive in that even in one of the worst bear markets in history in Japan, the volumes of shares traded on the Tokyo exchange increases by 8% annually.
Other potential revenue streams that do not yet exist
There will be multiple other sources of revenue that are not presently part of the NYSE revenue stream but that will likely contribute greatly to the new public company.
1) Information Technology (IT) sales
IT sales provide an area where European exchanges have been able to leverage their IT investments. Scale is still important in this business, and IT investments at exchanges are quite heavy. Selling IT is a smart strategic play that leverages that investment over a larger base and also builds relationships between exchanges. The NYSE will be poised to provide technology to exchanges around the world, with the addition of Archipelago's top-notch IT department
2) Options and derivatives
The merger gives the NYSE an entry into the options business as AX has gained the options market of the Pacific exchange through an earlier merger. This market is growing rapidly. The idea of trying to put together cash and options into one single product or have a closer affiliation may have some cost advantages and product advantages. This allows some very interesting cross-derivative and cash-product capabilities that could draw liquidity from the ISE, BOX, Amex, Philadelphia Exchange and the CBOE. Trading simultaneously in more than one type of product could be the wave of the future for broker-dealers. This one-stop-shopping for your trading may have advantages.
Derivative volumes are rising strongly, driven by greater sophistication in risk management, new products on-exchange and modest cannibalization of the cash markets. Whilst some of the growth in derivatives is cyclical, and therefore may decline, the underlying secular trend to greater risk management is dominant.
However, historically this has not worked in the past for the NYSE so this could be a risky area. The options game, is a sector the NYSE entered and exited rapidly in 1997 when it sold its options business to the Chicago Board Options Exchange. John Thain said the exchange has been exploring where to trade options and that it will likely trade them on the ArcaEx platform. The fact is that the options business is the most competitive sector of the securities markets today— CBOT main competitor – but this could lead the CBOT to enter equity trading or partner with another exchange.
3) cross-listing (see below)
4) Growth from listing smaller companies. Smaller companies that were being turned away from the NYSE will now be listed on the archipelago exchange. This business was previously being lost to NASDAQ.
5)Hidden pricing power
On the NYSE a foreign company listing 15,000,000 shares at a price of $41 per share would pay an initial listing fee of $130,100 and an annual listing fee of $24,260(Werner and Tesar, 1997). This is as of 1997. Still $24 thousand a year is not a significant amount. Would most companies walk if these fees went up 10%? 20%?
6) Data, data, data!!
The mistake Mr Grasso thinks may have cost the NYSE most has been its failure to recognize that its greatest assets are its data. The exchange gets a mere $167m a year from selling its tape of share prices to outsiders, who then reconfigure the information and sell it on, earning, by some estimates, over $20 billion. Imagine, says Mr Grasso, if the NYSE had capitalized on this data: it would have so much money coming in that it could let people and companies trade and list free, creating even more data to sell. He intends to make up for lost time: “We are in the data and media business.”
The exchange wants to regain ownership of its share-price data by withdrawing from the Consolidated Tape Association, a mutual organization run by all America’s regulated exchanges, which sells their combined data. It is also looking for new ways to make money out of other data that it collects, including its “limit order book”, which includes all as-yet-uncompleted orders to buy or sell at a pre-specified price, and which is now available only to the floor specialists.
7) The electronic transformation
The rapid advances in communications technology will continue to have a major effect on exchanges. Eventually at the NYSE we will replace the 3,000 people on the floor of the exchange with a nice, big Unix server. The continuing move to electronic trading is likely to boost volumes and reduce costs, but also potentially increase the threat to exchanges from ECNs and other alternative trading systems
8) Changes to rules on trading fees
While average daily volume was up 4.6% in 2004, trading fees, which are paid by members and member organizations based on trades executed at the Exchange, showed a 2.3%, or $3.6 million, decline. Equity trading volume is the principal driver of trading fees; however, the current pricing structures inhibit the NYSE’s ability to generate revenue growth during times of high trading activity or to grow via the introduction of
new products such as Fixed Income. The NYSE states in their annual report that these rules will be readdressed. Therefore these fees could increase substantially.
It is a mistake view the NYSE as a US exchange. This is likely to become an international market. The old geographic concept has less and less meaning due
to the progressive integration of financial markets, instruments, and intermediaries as well as technological changes that allow for electronic trading. It no longer matters where an exchange is geographically
From 1986 to 1997, the number of U.S. companies listed in Europe decreased by one third. Over the same interval, the number of listings in Europe by non-U.S. and non-European companies rose by a modest 5 percent, while the corresponding increase on U.S. exchanges was 131 percent European companies appear more likely to cross-list in more liquid and larger markets, and in markets where several companies from their industry are already cross-listed. They are also more likely to cross-list in countries with better investor protection, and more efficient courts and bureaucracy, but not with more stringent accounting standards.
In1995, NYSE had 252 non-US companies listed from 40 countries against 406 in 1999. Cochrane, Shapiro and Tobin (1995) identify two causes for this: firms are looking for equity as a source of finance, more so now than in the past, and US investors are more receptive to invest in foreign equity.
For companies from certain countries – with less mature markets- an NYSE listing should imply better reputation in the capital market, more abundant outside equity finance and possibly lower cost of capital.
“We aim to be the number two market for every major non- US stock,” says Mr Grasso. There are now 400 foreign firms listed on the Big Board. For a few firms, notably some from Latin America, the NYSE accounts for the largest part of the global trading volume in their shares. Unlike merging, which is fraught with business and human dangers at the best of times, this strategy of attracting foreign listings depends on doing what the NYSE should be doing anyway, but has done only slowly: providing liquidity, lowering costs, delivering best prices and extracting value from its data.
The extra-territorial extension of U.S. securities law through the Sarbanes-Oxley Act to foreign companies that are either cross-listed on U.S. exchanges or listed through depository receipts has created a great deal of tension in Europe. This is likely to be a setback to the increased harmonization of securities regulation between the two
largest markets in the world. I hope that the U.S. and Europe will not get into a tit for tat.
Sarbanes-Oxley will certainly be a deterrent for foreign countries cross listing on the NYSE. I expect the NYSE to pressure congress to make changes to Sarbanes-Oxley which will make it less onerous.
Because liquidity is likely to beget more liquidity, smaller exchanges will likely die or be acquired. The NYSE merger with AX, and the proposed deal between Instinet and Nasdaq, does not bode well for regional exchanges, like the American Stock Exchange. The strength of the smaller exchanges has been in trading products like exchange-traded funds, trading that Archipelago can do faster (and for which AX already has a significant share of trading - Archipelago is the leader in ETF trading, with about 30% of the market as measured by shares traded daily-), and in listings for smaller companies, which will likely be decimated.
As far as floor trading is concerned, the floor trading will likely slowly decrease (much as the cme) but there may be some demand in some areas. Mr. Thain points out that a few hundred stocks which trade huge volumes and are highly accessible trade better electronically while less popular stocks tend to benefit from a system where individuals create demand for stocks. There is some evidence to support this assertion.
We are likely to see a 24 hour global stock market based on alliances between nyse, European and Asian exchanges. There will no longer be such a thing as a closing price for a security. If the UK continues to reject the Euro, the London exchange may end up being the last exchange to take part in the European exchange consolidation which is now occurring, this could lead to a merger of the London exchange with the NYSE to from a true 24 hour transatlantic market. The payoff of such a 24 hour exchange for individual companies will be access to broader pools of capital and more liquidity in its shares.
Such a 24 hour exchange also eliminates, the fragmentation of liquidity that, along with high costs, is such a problem when a company goes the ADR route. And it significantly reduces the costs to issuers. All of this adds substantial value to a listing. The payoff to the exchange is more annual fees from more listings. The involvement of the US market is critical for making the 24 hour exchange eventually a reality because the US market represents 58 per cent of the world’s market cap. The US regulators have not embraced this idea – but I believe they will eventually.
As a highly regulated industry, it will continue to be shaped by government policies. Exchanges must comply with the usually expensive and sometimes unpredictable requirements of their regulators. These problems are likely to increase as exchanges pursue cross-border transactions, sometimes seen as the best route to growth for an exchange that is dominant in its domestic market.
The other big issue under this category is antitrust. For the New York Stock Exchange to merge with Archipelago, an electronic market that trades Big Board stocks, antitrust regulators will have to be persuaded that the relevant field of competition is much larger. That is one reason CEO John Thain, talks of a global market, although there is little foreign competition for listing or trading American stocks.
The NYSE may, in short, be the most liquid exchange simply because it is the biggest. On the other hand, some financial institutions are so big that it would take only one or two (a Vanguard or Fidelity, say) to switch to an alternative venue to do serious damage to the Big Board’s liquidity. The NYSE could lose 30% of its volume overnight—and its dominant position, once lost, might never be regained. Without the economies of scale, its overheads would soon turn it into a decidedly high-cost provider. Internalization (The matching of orders in their own books by large broker-dealers is seen by some as one of the biggest threats to exchanges) Internalization still requires an exchange for price discovery, and may attract the attention of the regulators, but it has the potential to remove considerable volume from exchanges.
Technology has changed the world of stock exchanges. But I think the ECNs may be less of a threat going forward. I believe the ECNs came into existence because there was a niche created for them by the fact that all the worlds exchanges had not kept up technologically. The exchanges were run as government utilities and not enterprising for profit companies. This has changed and I suspect the niche filled by the ECNs is about to close quickly. Beyond just ECNs, the true threat to exchanges like the NYSE comes not only from other exchanges like the nasdaq, but also electronic market places like Ebay and Yahoo where buyers and sellers can be brought together. This is another source of competition, partially real and partially potential, by intermediaries acting as brokers that try to exploit the Internet to offer customers the chance of trading directly.
Cannibalization and destruction of Brand:
There is some concern that the listing of smaller business could cannibalize the exchange's lucrative listing business. ''N.Y.S.E.-listed companies pay a lot of money to get that brand -- the N.Y.S.E. brand. If they offer a 'N.Y.S.E. lite' listing as they are planning to do for smaller companies on the AX platform, their existing listed companies might put pressure on them to lower their fees.
Despite these risks and the competition, the SEC does not want fragmentation of the equity market. If a rule it has made is leading to more fragmentation the SEC would likely take a hard look at it and change it – markets work better when they are concentrated. It leads to better price discovery and tighter bid-ask spreads and the SEC knows this. A concentrated market is in the publics best interest as long as the fees charged aren’t too outrageous. Could you imagine the equity markets becoming like the bond market. Because of these fears I believe the NYSE monopoly would likely be PROTECTED by the SEC as an issue that is in the public interest. This is the best of all worlds – a monopoly likely to be protected by regulators.
There have been many comparisons of this business to the cme. This is understandable as the cme is the only other large public American exchange. But the nyse will never be as great a business as the cme which is a dream business – the reasons why include:
1) derivatives use is growing much faster than trading in equities
2) the cme can create a product out of thin air – yes it takes research to sort out what traders and hedgers want/need – but beyond that there is not much capital needed and it just becomes another link on a computer – the NYSE is dependent on new IPOs to create new “products”.
3) Listing on the cme aren’t fungible – i.e. if you buy it on the cme, you have to sell it on the cme - whereas equities can be bought and sold in many other ways – internal trades, other exchanges.
4) The exchanges best able to withstand the threats from internalization, ECNs and the like will be the derivatives exchanges such as CME and Euronext.liffe. And their complex, innovative products will be much harder for upstart ECNs to reproduce than simple cash trades.