Two seemingly unrelated stories from the past week illustrate a fundamental problem with today's financial system. While this problem may seem "philosophical" or abstract, it's very real, and we won't put our economy back on a sound footing until we get a handle on it. The problem is this: Banking institutions no longer want to perform the human functions they've performed for a thousand years. Financing has become a robotic form of mass production, designed to generate ever-increasing wealth within an artificial system by draining it from the real world.
In a word, banks have lost their souls.
A recent report blames last May's "flash crash" on software run wild, while the new "robo-signing" mortgage scandal looks a lot like (and is) old-fashioned fraud. The "flash crash," which caused the stock market to plunge 600 points and bounce back within minutes, was computer-driven. In the "robo-signing" scandal the "robots" weren't machines, but bank employees who "mechanically" signed legal statements without checking their accuracy. But both are symptoms of a common disease. They both stem from the banks' insatiable desire to earn the maximum amount of money while expending the minimum amount of effort ,with the least possible real-world interaction. That's led to a mechanized form of banking that's devouring the economy.
Whether it's computerized trading or "robot" bankers, greed is the Ghost in the Machine.
Bankers with soul?
It may sound odd to speak of banks or bankers as having once had "souls." While it's true that they may not have had soul in the same way that, say, Otis Redding did when he sang "These Arms of Mine," financiers have always played a certain human role in the economy. They existed to make sure that capital was available when it was needed, whether it was to outfit sailing fleets for an voyage or plant seeds for next year's grain harvest. While bankers have never been confused with philanthropists, their role in well-functioning economies was clearly defined and useful.
Bankers throughout history lived and worked in the real world. Somebody had to inspect the granaries of ancient Egypt to see if they were full or not. The bankers financing a trading fleet had to meet the ship's captains, inspect the riggings, and make sure there was room in the holds for the treasures of the East. Back then, the money people did whatever they needed to do to see that their money was being safely handled. That's because there were never too many degrees of separation between a bank's money and events in the physical world.
That's changed. For a long time there have been financial transactions that dealt only with other financial transactions, rather than concrete phenomena. But these "abstract" transactions have grown exponentially with the explosion of derivatives and other sophisticated instruments. And it's easy money, comparatively speaking. You won't find anybody from Goldman Sachs inspecting the wheelhouse of a four-masted clipper ship bound for Madagascar.
Automated greed machines
It's natural for anybody, no matter what their level of income, to want the maximum amount of income for the minimum amount of work. This aspect of human nature only becomes a problem when society gives them too much power to indulge that urge. That's exactly what's happening today.
Take automated banking - or "algorithmic trading," as it's now known. The idea isn't evil - or if it is, then I have a streak of evil myself. Back in my systems analyst days (the early 1980's) I went to my bosses with a proposal for something similar, just like hundreds of my contemporaries probably did. I was told that it was too risky, and that automation couldn't substitute for human judgement. That was before it had become clear that large financial institutions could count on being bailed out if they got into trouble.. That was before the financial sector metastasized to gobble up 40% of the nation's profits, and before the growth of derivatives and similar transactions made the disconnect between "real world" economic activity and non-reality-based financial dealings so extreme.
Today it's a different story: Welcome to the Brave New World. "High frequency trading" - automated transactions that buy and sell massive numbers of transactions faster than the human brain can react - now accounts for a reported 73% of all US equity trades. In its report on the "flash crash," the SEC identified six different types of players: "Intermediaries, High Frequency Traders, Fundamental Buyers, Fundamental Sellers, Noise Traders, and Opportunistic Traders." Not a rigging inspector among them ...
A Wall Street Journal article traces the (de)regulatory decisions that helped turn the stock market over to unsupervised computer programs, leading to so-called "dark pools" where trading takes place outside traditional market exchanges.
The house always wins ... and the software runs the house
The result is a system of such complexity that nobody really understands how it works. It's a system where computers, and those who own them, don't just react to changing prices: They can control them. These ultrafast transactions also provide an ideal way for traders to engage in "front running," making money by placing trades for themselves a millisecond ahead of those their customers ask them to make. Since the financial market is dominated by a few large players like Goldman Sachs, each of them has enough data to manipulate the market in their own benefit without ever being detected.
(To this day, nobody has explained yet how the four largest banks - Bank of America, Citigroup, Goldman Sachs and JPMorgan Chase - went an entire quarter without losing money in their trading operations for even one day. The chances of that happening by chance in a legitimate system are infinitesimal.)
The mortgage market: Incentives to lie (and be lied to)
Mike Konczal's introduction to bank mortgage fraud includes a series of charts that nicely illustrate the separation of mortgage-backed securities from real-world economics. His illustration of the transactional layers between a homeowner and a mortgage-backed security shows how remote a trust holding these securities is from the actual banking transaction. What's more, it shows where the incentives exist to lie and exaggerate the value of the mortgages being sold.
Not only have financial institutions lost the incentive to touch and inspect the physical objects (homes) behind their loans, but many of them have had the incentive not to know if they're being lied to. That dovetails perfectly with the incentive that sellers had, which was to lie to them.
As with algorithmic trading, all that mattered was to accelerate the buying and selling process. Traders like Goldman Sachs could make money on each trade, while speculating on the overall outcome to make evey more money. In software terms, the process became the output. As a result, the speed with which these mortgages were bought and sold left the actual chain of ownership to many of these homes in question.
Why bankers become robots
Now that many of these houses are in foreclosure, lazy and fraudulent bankers chose to "robotize" themselves by signing documents for court statements without bothering to verify their accuracy. But these documents were affidavits which, as one of Yves Smith's readers points out, "is a legal document which can substitute for live witness testimony in court ... (requiring) that the witness swears to tell the truth, is competent and has personal knowledge of the facts they are testifying about ... (and) swears to tell the truth by being placed under oath by the notary."
These are not "paperwork errors," as bankers and many compliant journalists have described them. Signing an affidavit when you don't know it's true is a crime. In many cases, the banks had to know the claims in these documents couldn't be proven. In a way, they had no choice but to submit fraudulent documents. Their financial edifice was a house of cards, and without proof of their claims they were forced to add more cards to it. "Robo-signing" was the natural next step, after the "robo-lending" and "robo-betting" that built the house of cards in the first place.
This behavior is the end result of lazy, greedy, non-reality-based banking. It is the ultimate - and probably inevitable - product off a system that has turned banks into factories for the automated production of profits without any connection to the outside world.
The soul of a dead machine
The bankers signing these documents were performing a criminal act. But they were also like Mickey Mouse as the Sorcerer's Apprentice in Fantasia, running harder and harder to keep up with the creatures they had animated to do their bidding. As for the "flash crash," we've been assured that new "circuit breakers" will prevent future calamities. But we've also seen a series of subsequent "mini crashes," including one plunge in aluminum prices that was described as the "Jumpin' Jack Flash mini crash" (after the Whoopi Goldberg hacker movie, not the Stones song.)
How did the system get so irrational, so abstract, so voracious and uncontrollable? Another story this week tells us. The US Senate, acting as swiftly and invisibly as a algorithmic trading program, approved legislation that would have created new hurdles for people trying to protect themselves banks from illegal "robo-signed" documents. After a public outcry the President refused to sign the bill, but its very passage showed how a mechanized banking sector can use campaign contributions and political connections as its robotic arms and legs.
While they've been convincing us how busy and important they are, bankers have actually been doing less and less real work, with less grounding in reality as the rest of us know it. The Soul in the Machine has died, especially at the largest and most powerful banking institutions - the ones that remain Too Big to Fail and Too Inhuman to Live.
Don't just repair the machines. Give them a different purpose.
Sure, "circuit breakers" are a good idea, and so are the other reforms many of us focus upon. But while we're all debating the Basel III accord or the "finreg" bill, it's easy to lose sight of the bigger picture. Banking has become detached from human experience and turned into a mechanical, self-replicating function, one that exists only to grow and perpetuate itself.
The real solution is to return banking to its original function as a source of capital for real-life human activities. That means encouraging banks to once again become lenders, rather than merely speculators, while cracking down on all forms of human and "mechanical" lawlessness." Banks can certainly automate themselves, but "cyborg finance" will need to obey Isaac Asimov's Three Laws of Robotics, with special attention to Law #1: "A robot may not injure a human being or, through inaction, allow a human being to come to harm."
I don't disagree with Jon Stokes when he compares the entire stock market to "a single, very big piece of multithreaded software" (in an essay that will be particularly intriguing to geeks like this writer). The only problem with the analogy is that it doesn't go far enough. The entire economy is being driven by software now. Software's only as good as the intentions, knowledge, and wisdom of its programmers. Things aren't going to change until we take the source code back from the people running things now.
As programmers have always said: Garbage in, garbage out.
This post was produced as part of the Curbing Wall Street project.