Why the tenets of investing sometimes fail

The current crisis in the markets demonstrates a need for investment professionals to rethink everything they were taught about the way markets operate, says Andrew Lo, well-known financial theorist and finance professor.

At the annual CFA Institute Conference in Orlando, FL, the Harris & Harris Group Professor of Finance at the Sloan School of Management at MIT told more than 1,000 chartered financial analysts (CFA), their fundamental investment teachings, rooted in theories such as the capital asset pricing model (CAPM), were incomplete.

Lo warned investors not to hold the theories of investment luminaries as truisms, saying there’s ample evidence to suggest they fail on certain occasions, while working during other periods.

Investment tenets such as diversification, buy and hold investing, the relationship between risk and reward, the effectiveness of value and growth strategies will break down because markets are sometimes efficient — but certainly not always.

In 2006, Lo proposed a new investment paradigm he calls the adaptive market hypothesis, which merges the math-based assumptions of efficient market theory with the emerging field of behavioural science. He says this theory can explain why investing traditions have broken down in this crisis.

Adaptive market hypothesis

“Financial economists suffer from a psychological disorder I call physics envy. We would love to have three laws that explain 99% of economic behaviour. Instead, we have 99 laws that explain 3%,” he says. “When we write down our models we have to assume that all models will fly in all circumstances. That’s what physicists do.”

Lo advocates investors take the mathematical teachings of investment theory and merge them with behavioural science and throw in a dash of evolutionary theory. He says markets and investors adapt their strategies so that the fundamental behaviour of markets change, and mathematics will at times be trumped by basic human emotions, particularly fear and greed.

“Sometimes markets are efficient and sometimes they are ruled by fear and greed,” Lo says. “In order to capture that interplay between two ideas I propose to look at markets as a biological system.”

The efficient-market hypothesis breaks down when it takes into account human behaviour. Lo argues human beings will adapt their behaviour to satisfy their needs, but they tend not to optimize their needs.

He used himself as prime example of why human don’t optimize by analyzing how he personally gets dressed in the morning. Lo conceded getting appropriately dressed has an impact on professional performance, but humans don’t go into a detailed benefits analysis of each piece of wardrobe they have, in fact it’s nearly impossible.

Lo has five jackets, 10 pants, 20 ties, 10 shirts, 10 pairs of socks, 4 pairs of shoes, and five belts. He usually gets dressed in five minutes. However, he has 2 million different combinations of outfits he could wear, noting if he evaluated each item outfit for even just 1 second it would take him 23.1 days to get dressed.

Instead, decisions are made based on rules of thumb usually generated by past experience. Humans have a rough estimate of what’s appropriate to wear for a given situation.

To a certain degree, the same analysis can be applied to investing. Humans will tend to invest in strategies that are working — which eventually lead to asset price bubbles and their downfall as effective strategies.

How Lo’s theory applies

“While [my] theory is not yet fully developing from a quantitative perspective it still has some very meaningful applications. For example, the fact that strategies are going to wax and wane, and markets are not going to always be rational — you have to look at the population of participants,” he says. “If a population of participants are relatively even-keeled, by that I mean they have reasonable balance between emotion and logical deliberation, you’re going to get an approximation of efficient markets. During periods of extreme stress or extreme prosperity, you’re going to have very different outcome.”

Humans during periods of exuberance can achieve the same sort of addictive high from investing that they get from narcotics.

Lo adds, “Neuroscience has observed subjects in a MRI machine who are asked to play a game and receive monetary reward. It turns out the same brain circuitry that stimulates feelings surrounding monetary reward are the same circuitry that stimulates feelings around certain drugs like cocaine.”

In addition, the human and cultural dynamics of financial firms makes avoiding a crisis fueled by asset price bubbles, like this current one, near impossible.

Lo uses a hypothetical example from 2005 when there were early rumblings that the real estate market was overvalued, a risk officer of a hedge fund or trading firm would have three options if they believed a crash in the real estate market were imminent.

The first would be to shut down real estate trading units. Most likely, the heads of the firm would shoot this idea down, since these were amongst the most profitable trading units at the time and shareholders would leave.

Second option would be to shift money to other asset classes. Lo says this would have resulted in fewer bonuses to the real estate teams doing well, since the asset pool they could use was smaller. Most likely result would be their exit from the firm, with many of the firm’s clients.

“Let’s pretend you actually have the ability to hedge the risk, you can bet against the real estate market the way that [hedge fund manager] John Paulson did successfully,” he explains. “If you did this in size in 2005 you would have lost tremendous amounts of money in 2005, 2006 and the first part of 2007. At which point you would have been fired long ago. The fact is it’s difficult to do anything when all businesses are producing record profits.”

Lo suggested five modifications to efficient market theory that investors should account for in order to reduce their risk to these inevitable market crises.

Diversification: Investors always have exposure to multiple asset classes, including liquid and high beta asset classes to survive short-term funding crunches.

Long-Only: Reducing or removing the long-only constraints on funds.

Value/Growth disciplines: Behavioral drivers can create cycles that are hard to predict and will undermine the strategies.

Stock in the long run: If returns are independently and identically distributed, then if you hold a stock long enough you’ll eventually do well. “Someone said in the long run we’re all dead, but you have to make sure the short run doesn’t kill you first,” Lo says. “The problem with the long run is you have to worry about non-stationary short-run problems.”

Risk/Reward: Lo says investors’ risk is rewarded normally. But during crises it will be punished — sometimes severely as has been observed recently.

(04/13/09)