Network Effects

To: Prof. Michael Gorham, Director – IIT Stuart Center for Financial Markets

Re: Network Effects and the resulting MONOPOLY

From: Madhuresh ‘Maddy’ Rai

Date: Oct. 13^th, 2009

While the finance literature is abundant in introducing and describing the potential benefits of network arrangements in terms of increased participation, liquidity, efficiency, and transaction costs, no article specifically addresses the concern regarding liquidity-driven monopoly. Through this report I am presenting certain instances where network externalities have resulted in monopolies. Also, I use examples from the telecommunications and software industry which can help us build similar cases in the stock exchanges industry.

Exchanges can be considered as kinds of firms that produce a combination of two goods: listing and trading services. The utility is maximized when there are greater numbers of customers (listing firms and trading intermediaries). Listing firms bring in more firms that want to be listed and intermediaries trade as more liquidity is created. The ensuing network effects cause more liquidity which reduces the bid-ask spread and diminishes the market uncertainty (measured by the variance of market prices)¹.

The experience of European exchanges from the mid-1990s to 2002 shows that network strategies are only quite recent phenomena. The success of Eurex is a good example of how networks can replace a trading floor in another country. Similarly, the huge invasion of new equity routing/matching/trading systems such as POSIT in United States serve as distinct examples of effective network externalities. Such externalities in financial exchange networks come from two ways: the act of exchange transaction and availability of different vertically related services necessary for such transactions i.e. the matching services of brokers. Out of these two, the first type of externality is more pronounced in financial markets².

Liquidity-driven monopoly in a stock exchange?

The case in May 2004 concerning the launch of an alternative service for trading Dutch equities by London Stock Exchange (LSE) to compete with the incumbent Euronext is a good example to demonstrate a liquidity-driven monopoly. Euronext had almost 100% of the market in Dutch equities trading until LSE launched the Dutch Trading Service (DTS). Prior to LSE’s challenge, Euronext implemented a digressive fee schedule based on the number of trades executed. Euronext implemented two rounds of reduction in their trading fees in the form of temporary rebates prior to the launch of DTS and these reductions were limited to operations on Dutch securities. The first set of price cuts (Apr. ’04 – Jan. ’05) reduced the price for liquidity providers while the second scheme (May ’04 – Jul ’04) reduced the prices for liquidity takers.

These actions helped Euronext hold on to a share of almost 98% of the market. However, since the case concerned potential pricing and other exclusionary abuses by a dominant company in the particular market with the aim of preventing loss of market share following the entry of a competitor, the Directorate of General Competition launched an ex officio investigation under article 82. The case was closed in Sep. ’05 without any action primarily because of the insufficient amount of data to prove clearly that the lowering of prices attracted any significant additional business onto the exchange. The number of relevant variables that would need to be modeled and the lack of control environments contributed to this. Importantly, it was agreed that economies of scale and scope and network externalities do give rise to ‘natural monopolies’, although these monopolies may be limited in scope and thus take the form of non-over lapping parallel monopolies for as long as markets remain segmented on the basis of the nationality of the firm issuing the security. The Commission acknowledged that though Euronext’s intent indeed seemed to have been to exclude any significant loss of market share to LSE, the pricing mechanisms employed by it could not qualified as abusive. The case team also rejected the theory that competition between exchanges is impossible because of network effects which favor the incumbent. Suitable conditions, notable effective technology can overcome such entry barriers³.

Another example from the financial exchanges industry could be the actions of LSE in order to undercut Tradepoint7 after entering into the market in 1995. With the liquidity generated by the shift to it of the trades of nearly three out of four interdealer markets, LSE slashed its fees by more than 60 percent to gather away the market share from Tradepoint7. During the same time, riding high on the network effects being generated, LSE decided to move from a quote drive system to an order-driven system for the FT100 Index8 leading shares to better compete with the order-driven European exchanges⁴.

The Godzilla: Microsoft Corporation

Network externalities produce an ‘inertia’ effect through established networks. Microsoft can and has been cited as the single most dominant example of a monopoly power through network externalities. Microsoft, as the proprietor of a locked-in de facto standard, Windows, is intensely interested in the circumstances by which lock in can end especially through disruptive technical or market change in complementary technologies.

Browser War (1995 - 1998) : Netscape’s Navigator browser was introduced in late ’94 and was technically more superior than Microsoft’s IE which was introduced later in spring ’95. By using a variety of strategies whose legality became the basis of anti-trust investigations against Microsoft later, Microsoft ultimately increased IE’s share of browser usage considerably and almost made the Navigator extinct. Microsoft also leveraged on the network effects of its Windows operating systems and ended up being a monopoly in the PC OS market. An internal document quoted the following observations by top-Microsoft management: “We have won platform battles before, we know what we have to do. We know that we need a very large installed base for our platform (browser and email client share) and we need to drive the adoption of the platform by developers (content developers, corporate developers, solution developers…) so they build contents and applications on it.”

To avoid a tip to Netscape or Netscape/Sun standards, Microsoft executives engaged in a broad campaign of influencing all of the outside agents in the positive networked effect to move to IE instead. With internet evolving, Microsoft strengthened its key asset and most important brand: Windows to win the internet war on the desktop side. It used an upgraded IE bundled into Windows to convert the Navigator installed base and eclipse Netscape’s browser market share leadership. Chillingly, this course of action was suggested in the internal documents of Microsoft as “The Internet is part of Windows”. The solution employed by Microsoft was not to make the browser itself better, but instead to create situations where its high flying Windows OS could be used to make people switch browser to IE by making making sure that they have to, in order to get the best content⁵.

It is interesting to note that during the U.S. Microsoft anti-trust trial, Judge T. P. Jackson declared Microsoft to be a monopoly! This dominance was chiefly a result of the network effects driven inertia and aggressive strategies by Microsoft. After all, Microsoft is in a particularly strong position to leverage this approach. It often bundles its new products into its OS, Office suite, IE browser, and other offerings. It used this tactic to transform once market-leader Real Networks into an also-ran in streaming audio! From Oct. ’01 to Mar. ’03, Microsoft bundled Windows Media Player in versions of its OS which ensured that the software came preinstalled on nearly all of the estimated 207 million new PCs shipped during that period. By contrast, Real Network’s digital media player was preinstalled on less than 2 percent of PCs. Such a monopolistic action by Microsoft helped in grab the majority of the market, while Real’s share fell below 10 percent! However, the European regulators later made Microsoft unbundle WMP from its OS. This legal action was brought in especially to counter two highly-controversial policies of Microsoft: its standard contract with PC manufacturers prevented them not only from removing the WMP, but even from providing a desktop icon for Real Networks! While network effects create monopolies, government’s action can dampen a firm’s attempt to leverage its advantages in ways that are designed to deliberately keep rivals from the market.

Firms with a commanding network effects advantage may also enjoy substantial bargaining powers over partners. For example, Apple, which controls 75 percent of digital music sales is able to dictate song pricing, despite the tremendous protests of supplying record labels. In fact, Apple’s stranglehold is so strong that it can dictate terms even though the Big Four record labels (Universal, Sony, EMI, and Warner) are themselves an oligopoly that together provide over 85 percent of music sold in U.S.!

Microsoft’s experience with the video console market can explain briefly the reasons as to why the monopoly characteristic evolves from network effects. Sony’s Playstation 2 dominated the video console market over the original Xbox, despite the fact that nearly every review claimed Xbox to be a much technically superior machine. The network effects for PS2 produced more users and hence, it attracted more developers offering more games. Monopoly evolves from network effects. A dominant player cannot be just removed from the market by offering a product or service that is better. Any product that is incompatible with the dominant network has to exceed the value of the technical features of the leading player, plus the value of the incumbent’s exchange, switching cost, and complementary product benefit. Hence, battling a leader with network effects becomes difficult and the leader can slowly evolve itself as a monopoly⁶.

Microsoft’s case is hence the most distinct example of a liquidity-driven monopoly. After all, it has officially been declared a monopoly by the federal government. Its dominance in the OS market helps Microsoft destroy any software products that can easily be ported to other platforms and replace them with products that will not run on anything but Microsoft’s specific implementation of Windows. Its monopoly revenue stream also provides it with virtually unlimited funds to engage in predatory pricing⁷.

Telecommunications Industry: MCI WorldCom Merger

The characteristics of network industries make them prone to dominance by a single firm. If the attractiveness of a network increases as it enlarges, consumers will tend to choose the larger network, which in turn will make it even larger and even more attractive. This can lead to a market ‘tipping’ towards a single company or standard. While it can be more efficient for the market to tip and for a single firm to dominate and become a monopoly, such a shift however creates significant barriers to entry. This is because the networked markets tend to display inertia, i.e. once a technology is known to have a substantial lead in its installed base, it is hard for it to be displaced even by a technically superior and cheaper alternative. The costs involved in acclimatizing the resources to the dominant player and investments in the complementary products act as further deterrents for any client to shift to alternatives. Moreover, the possibility of obtaining significant and sustained market power creates an incentive for a firm to engage in predatory behavior to create a tipping effect.

In the telecommunication industry, a likely form of such anticompetitive conduct would be for a firm seeking to obtain dominance to degrade its rival’s access to its network. By denying compatibility, a larger firm will have less to lose by decreasing compatibility than rival firms. Also, by denying the access to its network, the larger firm will deny its rival the benefits of network effects and raise a barrier to entry. As internet evolved, private companies started investing in their own Network Access Points to facilitate interconnection beyond the similar services set up by the government. The larger network providers set up their own dedicated connection points and began to create policies to restrict future peering arrangements will small and regional ISPs that had not invested in growing their own networks.

MCI/WorldCom transaction covered the same ground and hence sparked the concern from regulators. WorldCom was one of the largest telecommunications companies in U.S. providing telephone and internet access services domestically and internationally. With its ownership of few other network services companies, it was one of the leading providers of internet backbone transmission services. MCI was the second largest long distance telephone service provider and a leading provider of internet transmission services (iMCI). The merger between these two companies would have combined two of the four leading nationwide or worldwide internet backbones as both MCI and WorldCom offered internet transmission services to ISPs and dedicated access services to large businesses. The anticompetitive investigations against the merger especially focused on whether the powerful network effects produced through the merger will have any degrading effect on internet as the merger created a dominant player in the provision of backbone services. It was estimated that the market share of the merged-entity and its next largest competitor after the transaction could be as high as 75 percent, with the third largest competitor having only 4.4 percent of the market!

Prior to the merger, no single backbone provider reached a disproportionate amount of destinations on the internet relative to the other major players and so there was a huge interdependence. It created an incentive to support efficient interconnections because failure to do so would have caused such a degradation of quality producing a risk of losing customers to other networks. This incentive would have changed, however, if the two largest backbone providers were combined. By giving MCI/WorldCom a disproportionate large customer base, the merger would have changed MCI/WorldCom’s incentives from favoring compatibility toward favoring incompatibility affecting industry peering arrangements and interconnecting prices. The merger would have given MCI/WorldCom a great bargaining leverage to dictate the pricing and terms of interconnection, possibly leading to monopoly control of the internet.

The European and U.S. regulators hence proposed the divestiture of MCI’s entire internet business as a condition to gain their approval for the desired merger. MCI/WorldCom sold iMCI to Cable & Wireless for $1.75 billion apart from committing to certain anticompetitive clauses and finally completed the transaction. Importantly, the judgment acknowledged that though it is still possible in future that MCI/WorldCom may become the dominant player causing the concerned market to tip, such a scenario would be the result of the company out-competing the other networks, and not because it buying the customers!⁸

^{My research through many papers published in this field conferred on one common observation: although there has been extensive research on the economic functions of financial exchanges and the properties of prices determined on exchanges, there has been little research on their organization, governance, and competitive evolution. Exchanges through network externalities can acquire market dominance (a hypothesis that the evidence strongly supports) to such an extent that it creates incentives for the profit-maximizing members to exploit their power by creating related rules and organizational structures. In particular, the survivability of inefficient exchange practices such as collusive pricing of member services depends crucially on the existence of exchange market power, which is in itself derived from network effects. Such liquidity driven market share has actually let many futures and options markets gain nearly 100 percent market share for the products they trade despite the absence of any regulatory barrier precluding the creation of competing contracts9.}

^{While we interpret the market data to confirm whether the recent technological advances and liquidity effects that are creating natural monopolies for the largest exchanges, it has also been stated that though network externalities are important, they may not be all powerful, and thus it might be possible that the number of exchanges may proliferate simply because automated exchanges are so cheap to operate. Technology would continue to increase in size and decrease the costs of trading in secondary securities markets and hence lead even more financial transactions to take place through the securities markets10.}

Conclusion

Stock exchanges are natural monopolies because their scale is a direct benefit to their customers as liquidity attracts liquidity. It is interesting to note that in spite of a vibrant financial microstructure, cross-listing of financial products is not a commonly observed phenomenon in U.S. even when the law requires that the regional exchanges have to apply to only SEC in order to trade in stocks listed on other exchanges! Financial exchanges tend to deal in their own highly liquid constituencies where they can gain monopoly. In fact the most noted case of an aggressive move by a stock exchange to trade in stocks listed in other exchanges was that by LSE in the late 1980s when it decided to trade the major stocks listed in the other European exchanges. Liquidity-driven monopoly among financial products (promoted by stock exchanges) is definitely present in the current market, though it hasn’t reached a level where anti-trust provisions would be actively imposed.

Unlike markets in software and telecommunications which impact the denizens at a local level and require large set-up costs, stock markets are chiefly dominated by hedge funds, financial institutions, high net-worth individuals and other funds which do not impact the denizens at a local level. So any resulting monopoly has not been yet able to provoke active regulations as observed in the case of OS or internet backbone providers. Even as the financial markets evolve

¹Competition and Integration among Stock Exchanges in Europe: Network Effects, Implicit Mergers and Remote Access by Carmine Di Noia, The Wharton School, University of Pennsylvania (Feb. 1998)

²Do networks in the stock exchange industry pay off? European evidence by I. Hasan & H. Schmiedel, Bank of Finland Discussion Papers (2003)

³Competition between stock exchanges: findings from DG Competition’s investigation into trading in Dutch equities by Sean Greenaway (Directorate-General Competition, unit D-1), Competition Policy Newsletter (Number 3 – Autumn 2005)

⁴Competition and Integration among Stock Exchanges in Europe: Network Effects, Implicit Mergers and Remote Access by Carmine Di Noia, The Wharton School, University of Pennsylvania (Feb. 1998)⁵Network Effects and Microsoft by Timothy F. Breshnahan, Economics Dept. at Stanford University