Wednesday, September 13, 2006

The TSE says: Me Too !!

That stock & commodity exchanges have proved fabulous post-floatation investments is undeniable (excepting the NYSE where "members" & Goldman judiciously sucked out the juice a-priori). But NASDAQ, CME, CBOT, ISE, Deutsche Borse, LSE all have spun profits for the prescient and the lucky. In typical Japanese style, somewhat after the apparent peak of boom in values, the venerable Tokyo Stock Exchange is readying its own transformation into a publicly-traded entity slated for 2009, a date that reflects delays resulting from recent trading mishaps for which management has been excoriated.

But I must admit to being less than enthusiastic about the wisdom of the central exchange marketplace within a capitalist, market-driven economy as a for-profit endeavor. As I look at these words, they appear oxymoronic and even silly on the page, but my concerns are real for a number of practical reasons. For the exchange while it has many constituents - the companies who list there, the brokers that facilitate trade there, the institutional investors who transact there, the public who trades there, and public who uses the values derived therefrom for micro-level allocative decisions throught the economy - has IMHO a single pivotal role: to most efficiently assist in the allocation of resources.

My first objection begins with the question: What is the role of "the exchange" in a capitalist system? Is the objective to maximise profits for the "owners" per se or is it to grease the proverbial wheels of the economy by facilitating maximally efficient price discovery, thus allocation of scarce resources that in the case of a stock market, brings together those seeking capital with those supplying it? In my opinion, it is clearly the latter.

Proponents of publicly-listed, joint-stock company exchanges will self-servingly argue that not-for-profits underinvested, and have insufficient resources to "compete". But exchanges are natural monopolies, or could be or in my opinion should be granted such status, provided the benefits accrue to all constituents. Investing in technology takes capital, but it is not deterministic that because an exchange has excess capital or free cash-flow, it will invest, nor is it the case that because it doesn't have a war-chest, it won't. The NYSE has always had lots of money, and has recently raised turnover fees - not because of a paucity of investment, but for the sake of offering the shareholders an adequate return on the high prices they paid for their shares during the transfer of ownership from men's club to profit-seeking enterprise.

Second in the laundry list are the natural conflicts of interest. Now, we all know that exchanges were, previously, "members clubs", and were run as such so the bar is not set very high in regards to a for-profit company doing more than remunerating Dick Grasso and enriching Spear Leeds. Since they were run in the interests of the members, they maintained high guild-like barriers to entry preserving oligopoly-like profits for the members. In the NYSE's case, specialists were outrageously granted exclusive license (for which they didn't even competitively bid for) to steal money (which they did) from those transacting. In NASDAQs & LSE's case, they fought tooth and nail to preserve a system that prevented their customers from getting the best price in order to preserve their members' profits. In both cases all the other constituents lost.

Tokyo, on the other hand historically maintained a much fairer and more neutral market-structure system. While it is indeed run as members club, there is more impartiality and fairness in the structure. Even in the by-gone days of floor trading, the TSE eschewed specialists for a "Satori" whose job - like that a public servant - was to manage the order-flow and independantly match buyers and sellers, not to scalp them when it suited (like Spear Leeds). The same held true for the "maerklers" in Germany, whose role was the same. In both cases there was no principal vs. agent conflict, no privileged information, and club membership benefits were limited to the oligopoly rents it provided members to their granting a point of access to trade. Technology soon supplanted the floor, and now everything is fairly and impartially handled by computers, as they are in Continental Europe. Only the Anglo-Saxons stand out resisting the virtues of technology in order to preserve the benefits for the boys club.

The third issue is "costs". But this is not so simple as the difference fees it costs to transact. NASDAQ may historically have been cheap to transact upon from a commission point of view, but if the implicit cost of spread and impact was taken into account, it might look very expensive. France might have been cheap to transact from a commission point of view, and very transparent, but the monoply extracted high costs for settlement. The NYSE may appear cheap to trade and efficient to settle, but the cost in terms of what the specialists manage to extract from their privileged monoplies is high indeed. So what do we know? The fact is that the infrastructure costs money to develop and maintain. Surveillance and administration costs money. And constituents are will pay, one way or the other. We also know that concentration of liquidity is "good" and "fragmentation", while increasing competition, (a good sobering wake-up call for NYSE), also increases trading costs and decreases allocative efficiency. This should not be seen as an indictment of not-for-profit exchanges, but rather an strong reprimand for "crony capitalism" as evidenced by the NYSE. The bottom line here is that what is good for the shareholders (higer costs & fees) is patently bad for all the other constituents as the higher fees deter turnover, liquidty, and thus decrese the efficiency with which resources are allocated. Can one imagine a bigger conflict of interest??!?

The Tokyo Stock Exchange is not immune from such accusations, but they are different. The Tokyo Stock Exchange (complicit with Japanese Finannce Ministry Officials) has historically given petty advanatge to Japapnese Brokers. Historically they did not permit the "baikai" or the order book to electronically leave Japan, and limited the dissemintation of this information to special Japanese-owned machines, or from the floor. They were also very slow to permit electronic connectivity from abroad - again to rpeserve the oligopoly rents for members, particularly Japanese ones. But ever inventive, people found a way. Proprietary traders at a member brokerage in NY were reputed to have trained a video camera on the baikai machine to watch which they could access. Morgan Stanley, in the early days of connectivity allowed their customers to send through the orders electronically to their desk, where a Japan-domiciled clerk had the job quite literally of "pressing the button" to on-send the orders to the floor, thus meeting the letter of the law.

Further conflicts arise in for-profit organization. The TSE historically served to protect small investors by vetting the companies that listed there, as did the NYSE. Fly-by-nights were prohibited by minimum requirements for audit compliance, sales, earnings, operating history length, etc. This prevented the speculative issues from listing, thus protecting the public from scams, and their own greed. Today in the US (like in the 1920's) we are seeing listing of shells with no businesses who hope to acquire business or companies in the future. Or on smaller exchanges, MOTHERS, JASDAQ, etc. there are companies with little operating history, and very small floats, that are easily manipulated by speculators, hedge funds, and institutional investors such as Fidelity & Jardine. There is no good reason in a world awash with capital and savings why these securities should be be granted equal shelf-space and and best should carry serious warnings regarding lack of operating history, excess valuations, high volatility and the lack of liquid markets, particularly during times of stress.

Finally, there is what one might call freebie revenue. Financial data has become enormously valauble to participants wishing to review history in order to make better investment decisions. This includes current & historical time quote & sales data, volumes, short-interest, failed trades, specialists trades (for which by the way, there is NO transparency at all), etc. The exchange extracts huge tribute for these records, in many cases making availability prohibitively expensive for all but the most well-heeled participants. In a sense it's both exclusionary and anti-democratic. But the true conflict arises insofar as the high prices they demand for real-time access and data deters turnover, and thus liquidity, and so injures the interests of all other constituents, thus creating an inherent tension between the exchange "owners" and all of the other constituents.

It's ironic that a listed corporate for-profit exchange at the center of financial capitalist financial markets creates such inherent conflicts of interest, leading yo what is probably a sub-optimal allocation of economy-wide resources. The answer to the question of the best structure, as the NYSE has clearly shown, is not the Gentlemen's Member Club. And for-profit creates inherent conflicts. Even a "mutual association" is a poor choice given the diffuse and varying interests of constituents. I would offer that that only a true not-for-profit public service organization would maximize utility for all constituent while at the same contributing economy-wide benefits.

1 comment:

Anonymous said...

You make a good case. I didn't know quite this level of detail of the inner workings and potential for abuse in the exchanges.

Clearly, the overriding problem is the network effect: I can't just run out and start my own pure-tech exchange, because the very fact that no one uses it immediately defeats its purpose as a market (which should always be as wide as possible). So you are right, there is a kind of natural monopoly here.

Though maybe technology allows for a solution even with for-profit status. If one started an exchange which was linked into as many extant exchanges as possible, and in which the exchange itself would act as agent and market-maker to all participants, then both the price-discovery and liquidity concerns for bootstrapping a new exchange could be avoided.

In fact, I believe bullionvault is an example of a system that works basically like this... though for a single commodity (gold).

The only remaining problem would be trustworthiness of the exchange, which is indeed profit-seeking.

While there are technical approaches to that problem (such as making the exchange's code open source), they are not fool-proof. Competition itself would then provide the best check on the interests of market participants being fairly served.