I am your biggest fan. I hope you know I am.
But do you think you can somehow
- Nicole Nordeman
This article moves my series from what's been, to what's coming - expanding the connection between the flawed economics of financial contracts to financial institutions' currently poor profitability. The next problem contract law in finance confronts: the question of legal liability for contracts irrevocably tied to computer code. This is the age of "smart" computers, and not-so-smart computerized contracts; embedded in computer code. The two central issues:
- Computers now act as agents for human principals. Examples include the questionable code/contracts that implement High Frequency Trades (HFT). There are other computer-as-agent legal issues, but this is the attention-getter.
- Computers also have been assigned the role of principal by developers of the cryptocurrency, Ethereum. Derivatives dealers show interest in Ethereum-style smart contracts on a permissioned blockchain network.
This article considers how computer-entered financial contracts led to financial markets' and institutions' descent into the darkest part of the trading forest.
Two earlier articles, here and here, examined the movement of trading from asset trading to contract trading - specifically, trading of conditional contracts, or derivatives. The superiority of contract-trading tech to asset-trading tech - accelerating trading speed, reducing trading costs and red tape - produced the beginning of the quickening revolution of trading technology, followed by the rise of computerized trading.
The markets owe much to the foresight and aggressiveness of Leo Melamed and the CME Group (NASDAQ:CME). The innovations they introduced between 1971 and 2001 brought the current trading tech revolution. This revolution has greatly benefited the trading community, but like any big change, there are cracks developing in the edifice they helped create.
The horse-race of trading tech has careened out of control, with the failure to realize that speed is interfering with the integrity of the contracts we trade. In the unquenchable thirst for faster, consideration of the average trader's interests, and that of third parties, is falling by the wayside.
Computers are no longer simply a tool. They act on their own. Once the word "tool" was a fair description. Computers were used in price evaluation and record-keeping, leaving trades to humans. But today a trader's computer enters transactions on its own. Computers have become agents, not tools. And Law has been caught flat-footed. The legal responsibilities of computer agents are unenforceable.
Computer agency creates two primary phenomena.
- Proprietary algorithms cloak illegal actions, seemingly removing responsibility for actions from human actors.
- Algorithms create unintended disastrous side-effects, since they are not programmed to consider these side effects. Ultimately, a computer will always be less sensitive to unexpected effects of their actions than humans.
As non-human agents, computers fall into a hole in the law. Agency law assumes agents are capable of criminal and civil infractions and liable to legal action.
The recent history of transactions speed and cost. Speed has always been one hallmark of profitability in trading. Consider the four steps driving every trade:
- the speed with which valuable information is received;
- the speed with which it is acted upon through transactions;
- the speed with which the market price adjusts to the trades;
- and the speed with which the resulting gains and losses are collected.
The two earlier articles show the effect of the change from asset trading to contract trading on trading speed. This change reduced the time from the acquisition of information (1.) to the receipt of trading profits (4.) The faster a trader moves from one step to the next; the greater her trading profits; the less her risks. But profitability is a relative event: if every trader is 5 seconds faster, nobody profits from greater speed, although everyone trades at less risk - until speed itself generates risk.
The year 1992 marked the beginning of electronic trading, with the launch by the Chicago Mercantile Exchange (a predecessor of CME Group), of its Globex electronic trading platform. Within a decade, the speed needed to move from step 1.) to step 3.), for the average exchange-traded instrument, had increased to the point that humans are too slow to be part of the process, for most transactions. This development went almost unnoticed by legal scholars, resulting in the current legal disarray.
The Winds of Change Blow Strong.
Clearing lags behind. Because every trading platform - equities, Treasuries, OTC trading, and futures, went electronic on the transactions side of each deal, the markets are at economic parity in the first three elements of trading speed - or at least the markets that want to be, are at parity. Thus the key speed and transaction cost issue for differing trading technologies has become step 4.).
Futures stand alone in a discussion of trading tech, since futures exchanges alone integrate trading and clearing. This, I contend, is the reason the futures exchanges dominate other markets.
As the earlier articles indicate, futures contracts have a substantial speed/cost advantage among exchange-trading technologies in clearing (step 4.). Because of this speed/cost advantage, futures have the dominant exchange-trading technology in terms of profitability, surpassing stock exchanges in the last decade of the 20th Century. That's why CME Group and Intercontinental Exchange, Inc. (NYSE:ICE) - once clearers of futures only - are the largest, most profitable, clearing organizations on the globe, having gobbled up the stock exchanges of their choice.
Other clearing processes have avoided the effects of competition, to their detriment. For clearing of exchange-traded securities, the Depository Trust and Clearing Corporation (DTCC) ambles along, at monopoly speed, clearing every US stock exchange three business days after the trade. OTC spot markets, such as Eurodollars, metals, and foreign exchange, moving at oligopoly speed, also clear at T+3 - the standard for markets with no standards - if they clear at all. OTC derivatives never clear.
Inter-dealer OTC transactions give a nod to the need for efficient clearing. Trades are not cleared, but the trading parties are replaced with a single "clearing counterparty." Thus OTC derivatives trading, and OTC markets more generally, remain in the Stone Age of clearing technology, and are a source of enormous inefficiency in financial markets as a result.
The good news is that the presence of these inefficient clearing processes has left transactions clearing vulnerable to more efficient transactions tech, as yet not introduced, that pays closer attention to all trading speeds and costs, including the costs of clearing and its effects on transaction risks.
The false promise of speed. But the advent of electronic trading has left the integrity of the entire system vulnerable. Are contracts in the day of electronic trading, where no humans are involved in the making of the majority of executed stock market agreements, for example, still meeting the fundamentals of contract economics?
- Do buyer and seller still understand and benefit mutually from their agreements?
- Are third parties left largely unaffected?
Experts challenge both assertions.
Do buyer and seller benefit from the existing transactions tech? We don't know, and have no way to find out. Which is completely unacceptable. Most transactions in the equities market, for example, are no longer conducted by people. Governed by trading algorithms, these transactions are computer-bought and -sold.
In legal terms, the SEC has simply allowed trading decisions to be hidden behind the curtain - the curtain of trading driven by proprietary computer algorithms. The SEC extradites small-time traders from London for a trading infraction. But doesn't seek to investigate the possibility of illegality on a far greater scale inside the algorithm-driven systems of high-frequency traders (HFTs).
The lesson of this SEC decision - leave illegal activities to algorithm-driven computers. Let them bury the truth. But we'll scour the world for humans breaking trading law.
Are third parties adversely affected by presence of computerized trading? The effects of HFT on markets is that they are becoming increasingly fragmented and chaotic - leading to unexplained events such as "flash crashes" - bad for the exchange-trading business and its customers. Do we know this event was driven by computerized HFT? Of course not, since we know absolutely nothing about the proprietary algorithms used.
The failure of HFT to meet the basic requirements of any legal system - that detection of criminal activity is possible - begs the question: are proprietary trading algorithms consistent with Law?
Are there ways that the intellectual property associated with proprietary algorithms could be protected, yet permit authorities to examine the legality of the algorithms? Of course. Couldn't trading algorithms be tested for the risks that they might create, of price-breaks and other market failures. The only acceptable answer is "Yes." A way must be found. This issue falls directly into the SEC's bailiwick. The HFT dealers do not have incentives to answer these questions.
But this particular SEC seems to require public pressure to act. I await a market crack-up big enough to open SEC eyes, as Congress finally feels some pressure from the average market participant, and passes it along to the recalcitrant SEC.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.