GearGrind: What Long Island hedge fund purchased an IBM Blue Gene supercomputer and is using it to build a trading system?

In an interview we did back in May, a very well-connected, extraordinarily math fluent person dropped this beguiling gem:

A "Long Island-based hedge fund" had purchased an IBM Blue Gene supercomputer, and was using it to build a trading system. Cool!

Unfortunately, our attempts to find out exactly which "Long Island-based hedge fund" this is, have thus far failed. After much hand-wringing, we've decided to expose our incomplete reporting process on this one in order to perhaps crowd-source a winning confirmation, sorta like Seymour Hersh does when he gives speeches on tantalizing nuggets regarding U.S. foreign policy. Our reporting led us to focus on two potential funds that could, or could not be our Blue Gene owner, as ridiculous as that sounds, meaning we give you that, blog or not, that non-statement is near meaningless, lacking as it does, confirmation, denial or clarification in any way. But we're stuck.

Our best estimated guesses are: FQS Capital or Renaissance Technologies.

But again, let us emphasize that it's quite possible that neither of these funds owns a Blue Gene. Because neither will speak to us about the matter.

We know Renaissance Technologies, the well-known hedge fund based in East Setauket, N.Y., famous for its consistent, outsized returns - the first fit-the-bill fund that came to our mind - owns a Sun system, which is not necessarily a disqualification. A call that ended up being routed to an attorney for RenTec resulted in said attorney, audibly annoyed, grousing "Why are you calling me?" followed quickly by "No comment" and a scratchy click, which was him hanging up on me, after we asked him whether "RenTec" as they're sometimes known, had purchased a Blue Gene to help build a new, or improve an old, trading system.

Another source aware of the IBM Blue Gene's sale to a Long Island-based hedge fund said "all trails" lead to FQS Capital, the fund of funds founded by RenTec alum and math genius Robert J. Frey, located on the North Shore a little over two miles east down the scenic 25A, in Port Jefferson, N.Y. But this source could in no way offer any information that would confirm said trails' end is FQS; Frey stopped returning our calls and emails, including two inquiring directly about FQS and Blue Gene, so: Heck if we know. We simply have no more time to chase this one, so feel free to crowd-source a clincher.

While FQS is "headquartered" in the Cayman Islands, like most hedge funds, and its operational hub is in London, according to the firm's Web site, its founder's home is nearby the fund's Port Jefferson office. Both the Freys and RenTec founder James Simons and his wife are well-respected area philanthropists who have donated millions to local institutions; an example in the Frey's case is Mather Hospital; in the Simons' it's Stony Brook University, which, you guessed it, owns a Blue Gene supercomputer housed at Brookhaven National Laboratory in Upton, N.Y. Simons is the former chair of Stony Brook's math department; Frey is the director of Stony Brook's quantitative finance program and he still teaches there. Simons led a group of Renaissance partners and lab directors that gave $13 million to Brookhaven to save the lab's Relativistic Heavy Ion Collider from becoming a victim of a 2006 budget shortfall.

Note it's possible that, instead of owning a Blue Gene, FQS or Renaissance and Stony Brook could have deals which allow FQS or Renaissance to rent cycles on Stony Brook's "New York Blue" supercomputer - it's a growing trend for universities to structure similar arrangements; the schools view such deals as alternative methods for monetizing their resources, thereby increasing funding. Or - again that dreaded bugger of the unconfirmed - perhaps neither [or either] fund does this with Stony Brook's "Blue."

(An aside: It's important to add that such University-supercomputer rental arrangements pose ethical dilemmas for schools, including promoting money-making to maths and physics students who might otherwise be focused on solving what can be argued are the bigger or more worthwhile problems of humanity, such as eradicating cancer or building a workable energy infrastructure. Schools can also be easily criticized for allowing their big machines to be used for what some would deem rather basal purposes of bashing the big capital markets pinata until the sweet lucre falls out. Yet the universities counter that some of these agreements include intellectual property sharing that occurs between the schools and the private sector firms, which need not be and are not always financial companies; and that any private sector funds resulting from supercomputer rental can help pay for other math or scientific research aimed at achieving benefits that go more society-wide than say, a get-richer-quicker scheme.)

Know this: There are very few high-profile funds out on "the island" with enough lucre to buy a Blue Gene, which costs $1.5 million per rack or so, per server in the system cluster, making this certainly a multi-million dollar affair that could easily creep toward the $10 million neighborhood - after all: who could buy just one? So it's a fairly reasoned process of elimination, ultimately hampered by it's resolute lack of irrefutable proof, to narrow the field to these two high-profile, well-backed funds (although RenTec has exponentially more cash than FQS, that's also meaningless, as spending a few million to grow FQS' pot of $350 million in assets under management seems a ratio that's plausible). So:

Fuckles. We just dah-burned don't know.

But if we knew how old this jpg-based (smart!) road (or home) show courtesy of Zero Hedge was, we might,...might be able to eliminate RenTec as the Blue Gene owner, as it posits RenTec as a Sun shop. I say only "might" because maybe RenTec wanted to replace this Sun system, or perhaps more likely (as supercomputer replacement seems complicated, does it not?) the idea was to buy another newer system and run two (at least) supercomputers for other, or some mix, of funds or strategies? It's usually a mixed bag for big firms, although perhaps the 'once a Sun shop, always a Sun shop' reasoning applies here, if RenTec's nearly unparalleled performance is in any way attributed to its SunFire hardware?

FQS could just as likely be our huckleberry. Heck, maybe it's Billy Joel, trying to crack the top ten again by massively parallelizing a new hit through some hedge fund front company (that would be pretty cool, actually). But who knows?

Blue Gene could perhaps be the technology that a firm like FQS, a fund of funds, applies to optimize intelligence on performance (or some other metric) of the hedge or other funds in which it invests or is considering investing, in order to garner “actionable” data to both allocate, and do trading, sort of speak. (A fund of funds is simply a hedge fund that invests in other investment vehicles, such as other hedge funds, private equity, venture capital, or a hedge fund that is a portfolio of other such funds or investments.)

We've read with fascination Frey's explanation in the Financial Times and elsewhere of FQS' “more quantitative approach to strategy allocation,” which to us implies needing in-depth data on a lot, if not most, or all, of the hedge-type funds out there, to truly make “multi-strategy” worth its name. Maybe I’m wrong, and we loathe speculation, but we've been forced into it. We really wanted to discuss the aforementioned with this fella, and whether high-performance computing is playing more of a role in fund of funds strategies – the latter something that may not be immediately obvious, but that which conceivably could provide novel ways of garnering attractive returns on fund of funds - or "FoF" investment.

Alas, no one we reached out to is talking definitively on these matters.

Background: Why supercomputers? Hedge funds and other financial firms generally use supercomputers to scour data in milliseconds for real-time trading, and/or to analyze massive reams of historical data, in the latter case, to come up with strategies that they hope will make them money when they apply them to trading later, or in the future. Critics claim such supercomputers enable firms using them an unfair advantage over the average investor, who can't afford or even get access to such powerful number crunching and lightning fast capabilities. Some firms deploy them in strategies based on rapidly trading in and out of stocks to make money on small up- or down-ticks in the equities, which can be in decimals or sub-penny ranges.

Opponents say using supercomputers to conduct such high-speed trading essentially equates with allowing one set of privileged, wealthy investors to peck away at what essentially is the larger public's nest egg, or rapidly hitting a piñata of money funded in no small part by average investors who are barred by regulators from such hitting-the-piñata-trading, with sticks that move faster than the eye or brain can perceive, sticks which are reserved for wielding by the privileged few. Because who in your neighborhood can buy a massively parallel supercomputer (the stick)? Depends on the neighborhood I guess. But remember, these critics have emphasized, you're the bag getting hit by the sticks; it's your money falling out the holes.

Supporters argue that such firms and systems supply the market with more liquidity, meaning they buttress the overall system by creating more trading, and hence increase the market's overall depth and size than would be possible in their absence. No conclusive proof has emanated from the debate about whether hedge funds promote economic growth through such trading. As to whether their alternative viewpoints are enough to lead to more accurate and efficient markets: The "efficient markets" hypothesis was proven deeply flawed in real world testing by the mortgage disaster: There were too few firms taking alternative, meaning negative, views on the overblown mortgage market, resulting in a Great Recession that nearly cratered Wall Street and the economy with it. But those too few firms? Those contrarians made a shitload of money. And inefficiencies, fair or not, equal "alpha" in this game. That's Street talk for big fat piles of lucre.

(We dare anyone to prove the efficient markets hypothesis works, under any plausible, real world market structure that policymakers could conceivably spit out under industry-regulatory political realities as anything more than a delusive talking point for certain wealth management strategists to employ, aimed at tricking others and themselves into rationalizing their existence as beneficial to the whole or most, rather than the few. Beware of too elegant explanations, although they're not always wrong. We're certain no one with any knowledge ever said and believed it that the market's egalitarian.)

Note that some market-making banks and brokers and high-speed traders, acting to protect their interests against what seemed very odd trading activity, took liquidity away - stopped or slowed buying and selling and/or accepting bids and offers (prices) on stocks - from the market during the so-called "Flash Crash" of May 6, 2010, when a big sell-off of derivatives contracts linked to blue chip stocks contributed to the biggest one-day point decline in the Dow Jones Industrial Average's history. Everyone in this debacle is blaming everyone else for the problem. Fact is, the banks and traders lobbied the regulators for years to basically get the system we've got, which is extremely fractured - there is no fair access or single, level playing field - and is so obviously less than close to failsafe or sufficiently backstopped. Again, as always, it takes leadership to admit the problem, which neither the regulators, the industry nor its analysts have shown they possess one iota of yet in this debacle. Instead, finger-pointing and it's resulting cousin - confusion - reins. But it becomes immediately clear to anyone that steps back for a moment to take a reasoned look at facts of the situation, that rules that the industry asked for and the regulators either enacted or snuffed, have worked to solidify brokers as advantaged gatekeepers in trading equities and their derivatives, and have empowered high-speed traders. So the arch defensiveness of these brokers and traders is understandable, if not admirable: They have a hard road to hoe to prove their privileged positions are worth protecting. In the meantime, they might do themselves and everyone else a favor by admitting they are one class of ax wielders creating the potholes in our pockmarked road toward what could be a sufficiently workable - and fairer - trading system, versus blaming wholesale the regulators for whom such criticism should actually be shared: You get what you ask for, (and they did); 'least with this gang you do.

Is any of this "fair?"

One bit of proof - although its accuracy is ample and uncontested - that our tagline paraphrased as 'the market is not free nor fair, the vig is steep and you're an outgunned mark for bigger monied profit-takers' is true? The law.

Allegedly to protect the public from fraud, the SEC enforces rules that say that investors with less than $1 million in net worth (and this metric cannot include any owned residences), or those making less than $200,000 per year (or $300,000 jointly) over the last two consecutive years, are by law prohibited from access to professional money management aimed at garnering big returns. One also needs $5 million or more in 'investable' assets to join a fund that has more than 100 investors. The government essentially is saying that if you don't have a lot of money, you need protecting, from potentially making a lot of money. Put another way: You're not smart enough to understand the risks inherent in "complicated" hedge funds, private equity and venture capital investing, nor the risks undertaken by these funds' managers. So you get no chance to chase their whiz-kid strategies and hence, at least in some cases, their outsized returns: Simons' flagship fund, for instance, has returned nearly 45 percent annually in net returns since its inception in 1988.

This is problematic, because it's patently unfair. It can work to create, or some would say, solidify, an uber-aristocracy, because it engenders a situation where the rich are given an avenue to get richer from which average persons are barred, in the same system average persons help fund or pay for. This sounds like a critique on capitalism. This is NOT a critique on capitalism: We like and support capitalism. And we think that if you're smart and able, you ought to have the freedom to go out there and be successful, and if that means making money in the markets: go for it. This is just a critique on how our capital markets are run. This is a critique on the claim that the markets involved are free, or as fair as they should be, and by that, we mean that it's our wager that if our market system were put on trial in the civil courts, a jury of reasonable citizens would with a reasonable degree of certainty declare the market insufficiently fair by a reasonable degree. We point you to blogger and software consultant Jeffry R. Fisher's definition of "fair market capitalism" using the analogy of a race versus a fight. In a race, everyone moves forward albeit in varying levels of achievement in how quickly they cross the finish line. Lanes are clearly marked so it's obvious when they're crossed. And proceed from the construct there are NO performance-enhancing drugs involved. Each has a fair chance of winning, the finish can often be close and few will ever completely fail, meaning stop short of the finish line. In a fight, one person's achievement is the other's downfall. So typically, whoever's got the biggest guns, wins. The market right now resembles the fight, in which only the wealthy get the space-age firepower with which to go into battle at the expense of the less armed. (We have no idea if Fisher's analogy is original but it's a good one regardless and kudos to promoting it.) Further, this is a critique that the "free, open and fair" odious miasma that some regulators and industry mouthpieces seem always to so unctuously spit and plague us with is bullshit.

To better understand this, please read below a set of provocative questions we posed to the SEC in 2006 when it was proposing to increase the level of minimum assets required to participate in hedge fund investing to $2.5 million. The SEC might have helped itself if it were to have proposed instead to somehow segregate or wall off investments of average investors from wealthy hedge fund strategies that can and have harmed the average investor, though short of banning certain strategies, we've no idea how that could be done. Yet that is the same idea behind the major investment firms - including hedge funds - demands that the big Wall Street banks segregate or protect their assets in separate accounts walled off from the banks' everyday activities to protect these firms' assets from potentially being mired in any folly in which those daily activities could snare the particular bank involved. Such demands have risen after firms found some of their investment or "margin" trading accounts frozen at Lehman Brothers when it went bankrupt Sept. 15, 2008; some are still locked up, including in Lehman's London offices. Anyway, the SEC proposal to raise the minimum net worth required to invest in hedge funds went nowhere, but it made clearer to us that the market structure increasingly threatens to further institutionalize what seems to critics a caste system pitting the have-more-than-we-know-what-to-do-withs, against the have-barely-enoughs.

From my 2006 email:

1.) Why is it that the SEC’s solution for mitigating the risks posed by hedge funds is to bar those that are less-well-off from being able to participate and potentially benefit from the outsized returns and gains that hedge funds can enable? The SEC’s idea is to wall off these lucrative investment vehicles only to those who already have big wallets and/or assets. But by making for certain that it is only wealthy people that are allowed to seek outsize returns by engaging in hedge fund risk, how is the SEC really bolstering a truly fair and free market?

2.) Doesn’t this effectively wall off a huge source of liquidity for hedge funds to pick off and over -- namely a huge pool of average investor money unable to engage in employing various strategies and thus a sitting duck for same – as hedge funds by their nature employ techniques such as market timing, shorting and shareholder activism, which effectively works to transfer to the rich large buckets of money from the pool of money invested by the average investor who, if the SEC would have it, would be barred from accessing these aggressive techniques and thus be a potential victim to their strategies and on the wrong side of the trade, or whatever tradeoff is involved?

3.) In other words, market timing is a great example: this is wealthy money employing a trading strategy that benefited wealthy investors and their hedge fund managers at the expense of the less-well-off, average investor. Shouldn’t the SEC be seeking better ways of ensuring hedge funds aren’t enriching themselves exponentially based on strategies in which saying it plain would mean saying they’re [siphoning] hard-earned invested dollars from the poor or less fortunate, instead of pursuing a “protective” measure that effectively weakens the investment choice of these less fortunate?

Thanks, appreciate your quick response."

We never received an email back, and although the SEC spokesperson we deal with generally will call us, versus email, it's to the best of our recollection that no such call was ever received. The "market timing" we cited refers to a 2003 scandal involving certain mutual funds allowing select hedge fund clients to trade rapidly in and out of the funds, a practice these funds publicly said they prohibited and regardless was a trading strategy unavailable to general fund holders. This type of market timing can bring incremental profits for those employing the strategy, but can also come at the expense - in millions of dollars of losses - to general fund investors. What you don't know can hurt you. It did here. It will always.

Dis Pater: Quarter-end Window Dressing by Banks Open Secret for Years

FYI: Lehman Brothers' "repo 105" transactions being investigated by the Securities and Exchange Commission and similar techniques allegedly bent toward achieving similar goals - those of understating at quarter-end the level of debt these banks typically carry for the purposes of boosting in the eyes of the market their companies' values, has long been an open secret...

We wrote about it in 2003:

http://www.securitiesindustry.com/issues/20031019/5039-1.html?zkPrintable=true

Sinon VI: J.P. Morgan Securities Converts to LLC; Tax Benefit Gifted in Frank-Dodd

Banks being able to easily and legally convert to Limited Liability Companies (LLCs) is something for which domestic banks have pushed for years. They've wanted it because LLCs are only taxed once - if they are structured as partnerships or as "disregarded entities" - the tax "passes through" to the individual partners, or to what are essentially the dividends of each of the shareholders. Meaning only those dividends (or each stockholder's share of the company's profit or income) are taxed, versus taxing both the dividends and the corporation's overall income.

U.S. banks have had trouble converting to LLCs in the past due to federal law and IRS restrictions, which have been interpreted as making it difficult for banks to be anything other than "incorporated," or the legal "Inc." designation, under which the tax-reducing "S-Corp." election restricts the number of shareholders and types of stock the banks are allowed to issue, and also bars banks from the "check-the-box" taxation the IRS allows other non-bank LLC applicants, or those NOT insured by the Federal Deposit Insurance Corporation.

Well, the banks got what they've long pined for in Frank-Dodd, which explicitly provides the option to go LLC for banks, at least under the "Derivatives Clearing Organizations" subsection of the Act.

So if you're a bank engaged in derivatives clearing and seeking pass-through (lower) taxes without any restrictions on the amount of shareholders or class of stock you or they hold: go for it!

J.P. Morgan, for one, announced last week its investment banking division, J.P. Morgan Securities Inc., was converting to an LLC beginning Sept. 1. The division is now known as J.P. Morgan Securities LLC.

An email sent to the bank about whether Dodd-Frank emboldened it to go LLC and whether tax and liability issues were among reasons for converting, was not immediately returned. J.P. Morgan Securities is the part of the bank that provides customers underwriting, merger advisory, trade execution and clearing and settlement services in derivatives, stocks and bonds.

Note too that the Warehouse Trust Company, the derivatives database subsidiary of the Depository Trust and Clearing Corporation (DTCC), The Street's giant utility for processing and settling securities, was approved as an LLC, state-chartered bank by the Federal Reserve Board Feb. 2, a full five months before Dodd-Frank and the LLC derivatives clearing bank provision within it, became law. It did so by obtaining a state exemption from having to hold FDIC deposit insurance when forming as an LLC, as state law has required (such entities also cannot take deposits from the general public). Such exemptive approval presumably banks no longer need, as federal law - in this case, the Dodd-Frank LLC bank derivatives clearing provision - trumps that of the states. According to the Fed's order approval, DTCC also transitioned to a bank holding company, and was approved to do so by the board of the New York State Banking Department March 1, the day Warehouse Trust opened for business, and thus became a regulated entity of the Fed, as well as the SEC. Deriv/SERV, the parent of Warehouse Trust and MarkitSERV, which is DTCC's derivatives trade matching subsidiary, has operated as an LLC since launching at end of 2003.

Regardless, it's legally easier for banks to be down with LLC, now. To wit:

"(2) CONVERSION OF DEPOSITORY INSTITUTIONS.—A depository institution to which this subsection applies may, by the vote of the shareholders owning not less than 51 percent of

the voting interests of the depository institution, be converted into a State corporation, partnership, limited liability company, or similar legal form pursuant to a plan of conversion, if the conversion is not in contravention of applicable State law," says Dodd-Frank.

To be clear: We noticed the J.P. Morgan Securities conversion the same day as FT Alphaville, which provided us with the context that banks had lobbied for easier LLC tax status in the past. However, we are thus far the only publication to point out that the Dodd-Frank Act explicitly allows banks such LLC conversion when engaged in derivatives clearing. We bemoan how this world makes us toot our own horns. But: toot.

Not that we necessarily should, for we hasten to add that little in securities law possesses "bright line" clarity. However, of this we are certain: The subsection in Frank-Dodd that allows banks engaged in derivatives clearing to convert to LLCs has at least provided a bit more comfort that said conversions need not contravene federal law nor that of the IRS.

Among the "primary dealers" - the largest traders of government bonds as designated by the Federal Reserve Bank of New York - Banc of America Securities is the only other U.S. bank to have obtained LLC status; BofA converted over 10 years ago, on May 17, 1999. Swiss banks Credit Suisse Securities (USA) and UBS Securities are LLCs as well, according to the New York Fed's list.

Update: Thanks to briareus, commenting below, for pointing us to the New York Fed's Primary Dealers List Web site, which was updated June 1, 2011 to say that Morgan Stanley & Co. Incorporated converted to an LLC -- Morgan Stanley & Co. LLC -- effective May 31, 2011. And also according to the Fed, late last year, Royal Bank of Canada's investment banking subsidiary, RBC Capital Markets, converted from a corporation to an LLC, effective Nov. 1. We can only hope briareus uses his 100 hands and 50 heads for the forces of good. We also wonder why this story has our "pagevisits" spiking by 2,533.33 percent? Are banks now -- or have they ever been -- effectively using us for tax advice on this issue? That'd be inadvisable, although some would argue, could be oddly flattering. Or is this issue just plain keen?

Sinon V: Industry Making Sure Regulators Include Single Bank Trading Systems in Definition of "Swap Execution Facilities"

Look for the banks to register individually as "swap execution facilities" to protect their big derivatives trading revenues they garner through trading derivatives individually, or bi-laterally, with each of their customers in single, one-off transactions, even though the banks are not publicly saying they will do.

As we said back in July, despite the passage of the "Dodd-Frank Wall Street Reform and Consumer Protection Act," it's the same as it ever was regarding swaps trading, meaning these derivatives, which include interest rate and credit default swaps, will be traded the same ways they always have, including, most often, between a single bank and its counterparty over the phone.

While various mainstream news media have blithely proclaimed (and some still do) that the financial reform law passed in the wake of one of the worst market implosions in our history would push derivatives onto exchanges with multiple bidders and multiple pricing offers, or electronic trading systems where participants must request pricing quotes from multiple banks - some of which the bill appears to call "swap execution facilities" or "SEFs" - that couldn't be further from the truth.

That's not even contemplated by a reading of the Act. And it's certainly not contemplated in the definition of SEF that's being formulated in the closed-door "advisory" sesssions going on between the industry and the Commodity Futures Trading Commission and the Securities and Exchange Commission, the regulators tasked with defining such Dodd-Frank specifics, including SEFs.

Here's the proof, courtesy of MarkitSERV, a joint venture between the Depository Trust and Clearing Corporation (DTCC), The Street's giant utility for processing and settling securities, and Markit, which is owned by the biggest swaps trading banks and is the sector's pricing and data provider; it operates the well-known and most traded benchmark credit default swaps (CDS) indices, which account for about half of all CDS volume. Search "non-multiple" in the aforementioned and "1-to-many" in this document to understand that The Street is setting the agenda for defining SEFs, including as single-bank systems.

Under the law, according to this interpretation, each bank - the "1" in the aforementioned "1-to-many" phrase - can individually be considered a SEF, by providing pricing to multiple - the "many" in said phrase - customers, or mutual fund firms and the like, known collectively as "the buy-side." The bank is the "sell-side." So this interpretation of a swap execution facility includes one in which only one sell-side bank is offering derivatives prices to either multiple buy-side institutions, or other (multiple) banks. This is way different from a typical exchange system, which is "many-to-many," meaning multiple brokers (or banks) offering prices and bids on stocks to multiple buyers and sellers, meaning institutional investors, which are, again, often collectively known in the industry as "the buy-side," many of which manage our money: An example is a big mutual funds operator like Vanguard. Note that only brokers (or banks) - the sell-side - are allowed to execute trades directly on an exchange: Buy-side firms like Vanguard must trade through a broker or bank, or use a broker or bank, to trade.

Background: From nearly the start of debate over financial reform, the banks and their attorneys, including those at several of their major swaps data processing providers, have said that “single-bank” swaps trading systems would fit the definition of a SEF. Why? Well, we'll answer that with several questions: Why would a bank give up profits via a mode of execution that arguably can optimally enable those profits? Why would a bank give up a mode of execution that best promotes the bank and its own brand? Why would a bank give up any method of trading in which it makes ample revenues? Perhaps more pointedly, why would a bank give up all its single-bank execution business to competitors including exchanges, multi-bank electronic trading systems or inter-dealer brokers (the latter firms act as "brokers' brokers," or middlemen for the banks, in that they match trades between banks)?

The answers to all of these: They wouldn't. And they didn't. And they won't.

A bank likes trading with a counterparty through its own "single-bank" or "single-dealer" proprietary system because it enables the bank to have optimum control over marketing, sales and ultimately and crucially, pricing of whatever product's being traded.

Let's start with CDS: In current practice, details of CDS contracts are manually typed into electronic processing systems after the terms are agreed upon and trades executed, mostly over the phone. Work among banks begun in 2005 to digitize the back office has led to a post-trade process that has become more electronic recently. But such automation has yet to take hold in the front office, meaning in trading, or "execution."

The main reason for the lack of electronic trading of CDS is because top derivatives banks believe displaying prices on screens will reduce the spread between the price a bank pays for buying a swap and the price this same firm can receive for selling one.

Out of the difference between the bid and offer - the spread -- a dealer pockets a portion as a fee for making or facilitating a trade. Banks also make fees for creating or custom-writing swaps that are traded later. Smaller spreads reduce the size of fees banks can take in.

That electronic or screen-based prices offered from multiple counterparties can lead to higher competition and lower, more uniform prices is consistent with predictions of standard economic theory. Empirical evidence is vast that markets with multiple sellers who compete on price transparently to buyers are fairer and more efficient with cheaper goods and higher volumes - an equation that can lead to more revenues for those "providing liquidity," meaning the banks offering such trading - versus bi-lateral, or single-bank structured markets that are typically higher in cost for bank customers but represent high-profit margins yet (arguably) lower volumes for banks.

I say "arguably" because there are no accurate nor published numbers nor even clear methods to determine which volume is higher, that of single-bank platforms or the number of trades executed over multi-bank platforms. A bank could argue its own "single-dealer" platform is the one executing more trades, as the bank's single-dealer system, while connected to the multi-bank system to operate, is the one executing the trade; the bank and counterparty just met up on the multi-bank site. So, goes this reasoning, and it's been argued to us before, "it's the bank's trade," regardless of whether the trade was "clicked" or "agreed upon" over the multi-bank system, or say, via Bloomberg. Also, the banks have argued that opening markets like bonds and derivatives to streaming, executable prices will sap liquidity. However, some critics counter, it's not the market structure that's sapping liquidity in these situations, both hypothetical and otherwise, it's the banks themselves purposefully taking liquidity away: All markets make room for methods for banks and buy-side firms to unload or sell, or buy, large amounts of securities (often referred to as "block trades") in ways meant to protect the firm from the market moving (meaning other traders trading) against their need to shed or take on a particular security or risk, if said needs are somehow discerned or broadcast to other market participants. It's easy: Firms long have just split up such orders into multiple pieces and traded anonymously. But more on those issues some other time. Anyway...

One need only basic logic to understand the fundamental principles: Screen-based markets that offer simultaneous prices from multiple providers can tighten spreads, because they bring buyers and sellers together in a single place offering multiple bids and offers that different parties can see at the same time: Price competition is more apt to occur, because when banks must vie for business simultaneously from customers they share, the prices they give, or their cost of services, become paramount to the customer's decision of with whom they'll trade. Whereas in a single-bank system, the customer must essentially trust that the bank is giving the fund a good deal, or the fund must go around separately to multiple banks to compile their prices individually and then compare these prices.

Multi-bank screen trading thus creates a more efficient as well as competitive marketplace than one which can't easily be "seen," meaning, again, one requiring a person to make multiple, sequential Web visits or phone calls to each bank, from which a list of price quotes must be, one-by-one, manually compiled and put into a chart for later comparison.

Electronic markets can boost participation and trading volumes, as more players join what's viewed as a more competitive marketplace with open, transparent, cost-effective services. As more dealers get involved, fees get squeezed further, but volumes can also grow.

However, it's important to understand that the type of screen-based trading we're talking about is predominantly request-for-quote, or "RFQ." This structure does not allow for optimal competitive and transparent pricing. Whereas stock markets do, meaning, stock exchanges offer constant streaming bids and offers, or prices - a structure considered the most open in terms of competition, where brokers are forced through the electronic publishing (the result being the public streaming like that of an electronic ticker one sees coursing the sides of buildings in Times Square, for instance, or similar, with price updates flashing almost constantly), which engenders brokers (or their computers) to compete, and nearly instantaneously, on price.

The banks participating in multi-bank-to-buy-side electronic bonds and derivatives markets do NOT even see the prices of their rival banks that are also responding to an institution's RFQ. Nor do they know the names of the other banks involved in said RFQ. The rest of the participants in the market are blind to the trade's details, except of course the buy-side institution requesting the quote. And a buy-side firm need only request one quote; there's nothing in these systems forcing buy-side firms to request pricing from more than one bank.

Said another way, RFQ, in which asset managers must request a price quote from one or several sell-side shops, is a less efficient and less competitive trading model than that of an exchange: Streaming prices are manna to the buy-side, or whomever is receiving them. The models available in the over-the-counter derivatives market do NOT involve streaming prices from multiple pricing providers (which are the banks), at the same time. It's more of a serial - or one-by-one - endeavor.

As we said in July - and in fact - there is nothing in Dodd-Frank mandating that swaps be traded over an exchange or "swap execution facility." Rather, such trading would only occur of its own accord: That is, if an exchange or swap execution facility decides to list or make available any cleared swaps for trading, then those most generic swaps that clearinghouses decide to clear, can be traded over an exchange or swap execution facility. It's a rule that isn't a rule. Cleared swaps need NOT be traded on a swap execution facility.

Read the law, a la Dodd-Frank: "With respect to transactions involving security-based swaps subject to the clearing requirement of subsection (a)(1), counterparties shall — (A) execute the transaction on an exchange; or (B) execute the transaction on a security-based swap execution facility registered under section 3D or a security-based swap execution facility that is" registered with the Commodity Futures Trading Commission.

‘‘(2) EXCEPTION.—The requirements of subparagraphs (A) and (B) of paragraph (1) shall not apply if no exchange or security-based swap execution facility makes the security-based swap available to trade or for security-based swap transactions subject to the clearing exception under subsection (g)." [Emphasis ours.]

A cleared swap need NOT be traded on a SEF or exchange. Again: There is NO RULE mandating swaps trade on a SEF or exchange.

The term "swap execution facility" is vague, but the bill basically says such a system must include multiple participants. According to the banks, that means trading with a single bank complies with the definition of "swap execution facility." The longstanding method whereby transactions are completed over the phone also fits the description of "swap execution facility," meaning any such trades need not be done electronically, or executed using only the computer screen. Given all that, there is technically no trading requirement: While it's understood that if whatever swaps any one of the platforms that registers to become a "swap execution facility" chooses to offer would become required to trade over any said facility (or exchange), whether it be that of a "single-dealer" system at one of the banks; systems of the the inter-dealer brokers, which are firms that broker CDS trades between the big banks, using mostly the phone; or platforms offered by the multi-bank-to-institutional-investors [e.g., Vanguard is an institutional investor] - the way those swaps are traded will not have to change. That's because they're still allowed to trade over the phone as the majority of swaps always have, and/or via an offering of a single bank.

Plus, the types of swaps any one of these outfits may offer will be heavily influenced by their biggest customers, the banks. Some of those banks own stakes in these systems, including multi-bank-to-institutional platforms Tradeweb and MarketAxess. Regardless, all firms offering services to banks depend on the banks for their livelihoods - the banks are the main suppliers of both their products and their profits. Therefore, they will not risk losing the banks that keep them afloat by offering anything the banks don't want them to. Plus, Tradeweb offers to buy-side firms interest rate swaps trading with single banks, as does Bloomberg. It's doubtful these platforms, or the banks that use them, would give up offering trading based on single-bank inquiries and/or pricing without first fighting to keep such capabilities. We're waiting to hear from Tradeweb on it's involvement, if any, in any such negotiations.

Our queries sent to executives listed as meeting with regulators regarding SEF definitions at J.P. Morgan and Morgan Stanley went unreturned; we asked about whether these banks separately would register their own single proprietary or "single-bank" (also referred to as "single-dealer") systems as "swap execution facilities." We know not exactly how to interpret the quiet here, so understand clearly that noting it passes no judgment. Anyway, no replies yet from these banks.

Goldman Sachs said through a spokesman that it was "premature to say" whether Goldman will register with regulators as a swap execution facility. We're in the midst of trying to follow-up with Goldman regarding MarkitSERV's interpretation of SEFs as including single-bank systems. We'll update if our follow-up query is returned. (Update: Goldman spokesman said: "We aren't going to comment on MarkitSERV's view." To which one might counter: "Isn't MarkitSERV simply the messenger here, for the banks' views?")

Anyway, according to the Dodd-Frank Act:

A “securities-based swap execution facility” means “a trading system or platform in which multiple participants have the ability to execute or trade security-based swaps by accepting bids and offers made by multiple participants in the facility or system, through any means of interstate commerce, including any trading facility, that—(A) facilitates the execution of security-based swaps between persons; and (B) is not a national securities exchange.”

The bank lawyers are arguing that Dodd-Frank seems to allow for single-bank dealing as an SEF if "multiple participants" can be viewed as being the non-bank customer participants, and multiple "bids and offers" are viewed as being those coming from a single bank sent out to different non-bank buy-side customer participants; or that the multiple prices are somehow viewed as being those of the non-bank customer participants. Others, including this banking attorney, find the argument specious. Regardless...

Also, a derivatives clearing organization can be singular too, according to Dodd-Frank:

“A person that clears 1 or more agreements, contracts, or transactions that are not required to be cleared under this Act." Said person "may register with the Commission as a derivatives clearing organization." This could be one bank then, n'est-ce pas?

Yet, despite the language in aforementioned section of the law describing a clearing firm as "a person" clearing "1 or more" contracts, the Goldman spokesman said Goldman would not register as a "derivatives clearing organization." The other banks we contacted did not respond to the question.

Definition: Clearinghouses otherwise step in between counterparties’ trades to “clear” them, or ensure each has the correct terms and enough funds so a smooth and accurate exchange of monies for securities – settlement - can occur. Doing so, clearinghouses are said to mitigate risks to the whole trading market or system involved, as well as protect the trading counterparties within the structure, by pooling and centralizing funds from all participants, monies that are aimed at backstopping the system via paying or making surviving firms whole if one (or several) of the system's major players stumbles financially and fails to meet its obligations; such a failure can threaten to financially hobble the faltering firm's counterparties too (though it doesn't follow that it always does or will), potentially (but not necessarily) triggering contagion where domino-like bankruptcies occur that could rupture the whole market. Central clearinghouses are purposed to lessen that risk; and big financial firms are required by the financial reform bill to send what looks to be their most generic, or "plain vanilla" swaps trades to these central clearinghouses.

Important: The clearinghouses (read: banks) are set to decide on which swaps will be cleared. But it's crucial to understand that there is no - I repeat - NO exchange or SEF trading requirement for any swaps: Only if an exchange or SEF decides to offer any swaps over their exchanges or SEFs, swaps which the clearinghouses (read: banks) have already decided to clear, will the mandate to trade (only these) swaps be triggered: So you see, the "trading requirement" to which much of the press, analysts and industry executives have blithely and inaccurately referred, does NOT exist: It's a rule that's not a rule.