How Debt Screws With Our Heads, Part 2: Distortion and Bias

This is Part 2 of a 2-part article on the human cognition of risk and debt. Part 1 can be found here.

In our last post, we talked about the loss-aversion mechanisms of the brain, and how they send us emotional signals that help us avoid unwise risk. We also noted that there were about ten or twelve cognitive biases that tend to interfere with that mechanism, keeping it from kicking-in when it should. Here are a few of the major cognitive distortions that disable our ability to objectively conceptualize the risks of debt:

Aversion to Sure Loss: “If I don’t take this risk, I can’t get back where I should be.”

Loss aversion can hurt us as well as help us, because if we feel that we are “down,” we tend to take increasingly risky behaviors to try and get “back even.” This was proven out in a serious of famous choice problems conducted by Tversky and Kahneman.

Aversion to Sure Loss is related to another bias called Social Anchoring. Social Anchoring is the idea that if you don’t take on this risk, everyone else will pass you by. Both biases make you feel like you might be “behind” by comparison. One World Bank policy working paper pointed out how the directors of the Big 5 investment banks were concerned not about the nature of the investments they took on, but about beating one another’s returns.

In his paper, How Psychological Pitfalls Generated the Global Financial Crisis, Professor Hersh Shefrin tells how UBS, trailing its competitors in 2006, got itself deep into the subprime mortgages that led to its downfall. Their decisions seemed to have less to do with the prudence of the investment than with their trailing position in the industry. They made the decision from what’s called “the domain of losses,” the same psychological sensation we feel when we’ve lost $200 at the blackjack table, and we “know we can get it back.”

Present Bias: “I’ll just sacrifice something later on to make room for this new debt.”

Present Bias says that we value the present more than we value the future. Sure, it’s okay to eat cake now; you’ll do more exercise next week to make up for it. Sure we can afford the flatscreen; we’ll give up something else for the next couple months.

How Debt Screws With Our Heads, Part 1: Pleasure and Pain

This is Part 1 of a 2-part article on the human cognition of risk and debt. The second part can be found here.

In her textbook Neuroeconomics and the Firm, Angela A. Stanton quotes psychiatrist and former trader Richard Peterson, who tells us the story of Lee:

Lee [was] a 53-year-old partner in an accounting firm, who lost part of his Orbitofrontal Cortex (OFC) as a result of surgery to remove a tumor…

After a successful operation, Lee was able to return to work and function normally except for his terrible investment decisions. He bought several expensive vacation time shares, bought penny stocks based on faxed and emailed promotional material and could not keep up his mortgage payments. The loss of his OFC took away an important part of Lee’s functional ‘loss avoidance’ system. Previously a conservative investor, he was now unable to feel ‘risk.’ Lee explained that he knew he should feel uncertain and afraid, but his highly speculative investments did not feel ‘risky’ to him.

America is a nation engorged on debt. Many of us need ten years or more to pay off hefty student loans. Many others of us never fully recover from getting too deep into credit card debt. Still others of us took mortgage debt on terms we couldn’t afford because we planned on refinancing before it became a problem.

Market theory tells us that supply and demand forces will place a rational value on debt risk. A debtor fitting such-and-such a profile, with so much collateral, borrowing for so long a time equals a precise interest cost. Of course the future is not completely foreseeable, and there’s always a risk that the borrower will not be able to pay the loan back. But the market has baked that possibility into the cost of the loan…that’s the whole point. Therefore – as far as the market is concerned – debt is a knowable, quantifiable entity.

And yet, every major financial crisis in the history of the United States has been either precipitated or exacerbated by over-leveraging. Investors and banks have borrowed to finance investments since Aristotle’s time. Yet we find ourselves living through more and more periods of financial turmoil, usually kicked off by over-leveraged or unwisely-leveraged professional investors.

Preceding the 1929 crash, investors were buying stocks on as much as 90% margin (nine dollars of loan to every one dollar of capital). Starting in 1975, the maximum debt-to-equity ratio for investment banks was 12-1 (it could borrow up to 12 times its own net worth to play the market). When the rule was relaxed in 2004, those ratios went up to 30-1 and even 40-1. At the time of its crash, the Swiss bank UBS was leveraged at 60-1. The Long Term Capital Management hedge fund, at its 1998 bailout, had an effective leverage ratio of 250-1. These outlandish leverage rates were a major contributing factor to the 2007 financial crisis.

So if the major aspects of debt risks are so quantifiable – if you can know the terms and rationally determine the risks to which you will be exposed – why does debt consistently get us into so much trouble?

Swarm Intelligence: Is the Group Really Smarter?

Swarm Intelligence, or Swarm Theory, is the collective behavior of decentralized, self-organizing systems: ants in a colony, movie raters at Rotten Tomatoes, participants in a market economy, etc. By observing these systems in nature, scientists have theorized that such systems harness a sort of leaderless, collective intelligence. By leveraging these kinds of consensus-based systems, groups of independently-acting agents can solve problems more efficiently than they could if they were centrally controlled.

Ants, for example, do not use any kind of centralized management in their colonies. Organization happens organically, through millions of interactions between individual ants who are following very simple behavior rules. Some are patrollers, some are foragers, some perform maintenance, some collect waste, and so on. A forager will not leave the colony to go find food until it’s bumped into at least four patrollers and the interactions are no more than ten second apart. The fact that these patrollers have returned alive from the same area are the cue that it’s safe to travel to that area to forage for food.

Bees choose their next hive location using a similar, self-organized system. Scout bees will fly out in all directions looking for new hive locations. When a scout finds an interesting piece of real estate, it flies back to the hive to let the other bees know about his find. It communicates to other scouts using physical motion both the location of the potential new home and his enthusiasm for it (the goodness of fit). Soon, scout bees start assembling at four or five potential new hive locations. Consensus is reached once fifteen bees arrive at any single location. Those fifteen bees will then fly back to the hive to signal that the new hive location has been chosen.

Scientists started looking at this kind of theory as early as the 40’s (John Van Neumann and John Conway did the first theoretical work on “self-replicating automatons”). The field exploded in the last twenty years with the rise of computer science and the Internet. Swarm Theory lends itself perfectly to Artificial Intelligence. Computer learning is based on cycles of testing, valuation, and reiteration using simple heuristics and leveraging computational brute force. This is analogous to leveraging the many thousands of simply-programmed individual agents within a swarm. Google uses a variation of Swarm Theory to discern authorities and rank pages.

Swarm Theory was popularized in 2007, in a National Geographic article: I also spoke briefly about Swarm Intelligence in a previous article called The Superstar Trap.

There are many examples of how society has leveraged this idea, even before we knew what it was. Adam Smith’s Invisible Hand works through the collective intelligence of Swarm Theory. Market valuation is a perfect problem for this particular structure, and the fall of the Soviet Union became a great validator of the collective wisdom of decentralized markets compared with command economies.