The dangers of confirmation bias


Published:

Confirmation bias -- the tendency of people to seek and embrace information that matches their existing beliefs and paradigm -- is one of the many flaws we humans have when it comes to using data in our personal and professional lives.

Confirmation bias is one of those fundamental variables that I regularly discuss in consultation with current and prospective clients and a propensity I have to work against both professionally and personally. The reason is simple: It's a lesson often to be learned again and again, because a confirmation bias trap can exact a heavy price. Do you remember the Internet boom and accompanying bull market in the late 1990s when investors kept buying stocks at higher prices regardless of valuation?

In studying the sociological construct surrounding the nearly disastrous Cuban missile crisis, confirmation bias was also known as "groupthink." In nature, "lemmings to the sea" is a fine metaphor for confirmation bias. Because our minds are structured to seek structure in otherwise random events, confirming our biases only seems appropriate.

How about the housing market in the last decade? In 2007, there were enough signals about the coming housing market collapse that it's hard to imagine why anyone would consider buying a house or making investments in sophisticated financial instruments tied to real estate. But the stampeding bull market exerted a strong gravitational pull that enveloped some investors, analysts, homebuyers, pundits, regulators and politicians. The tulip mania in the 1600s also comes to mind. Hindsight is 20/20, but confirmation bias and its detection is less precise.

No immunity

It is hard to imagine a more public example of confirmation bias than the public reaction to the results of the recent presidential race.

Plenty has been written about why the election turned out as it did, and there's little value in going over the statistics for those of us that suffer from election fatigue. From a confirmation bias standpoint, this was an instructive moment because Republican presidential contender Mitt Romney was a venture capitalist and apparently a self-admitted man of data. According to his staff, when he realized he lost the election to Barack Obama, Romney was "shell-shocked" by the result. Romney had not prepared a concession speech because a win had been confirmed by his statistical nerds. The Romney team apparently chose to become victims of confirmation bias in the extreme.

I ask you to consider two assumptions the Romney campaign made that are directly connected to the decisions that financial advisers, and possibly their clients, make all the time.

First, the Romney team's internal polls led them to believe that their statistical modeling was accurate. Pride was involved in some of their calculations - after all, no one wants to feel they are in a losing campaign - but the assumptions also were built on the belief (as opposed to any empirical evidence) that Obama's supporters were less enthusiastic than in 2008.While the Romney expectations appeared correct to some degree, the Romney camp failed to anticipate that their candidate's base of support would turn out in numbers less than expected - a significant error. The parallel in my world is that the success of an investment based on a "gut feeling" and backed by apparent empirical support, becomes the poker equivalent of attempting to fill an inside straight -- not just once, but multiple times, which is not very probable.

Secondly, it is possible that the Romney team also misunderstood "momentum" with large enthusiastic rallies at the end of the campaign. It may not be unlike the Greek myth of Narcissus gazing with admiration into the pond at his own reflection. Furthermore, the Obama campaign was also drawing large rallies typical in the final days for all presidential campaigns. A wise financial advisor should not assume that being smart, popular or previously successful is ever enough - there is a reason why my profession regularly states that past performance is no guarantee of future results.

No person I know, including me (just ask my spouse), is perfectly rational or immune to confirmation bias. Confronting confirmation bias is discomforting because we don't like additional information upsetting our seemingly perfect and rational conclusions about the market, the quality of our decisions, or the value of our counsel.

I have engaged in high-level meetings in our nation's capital three times in the past 10 months and have become more skeptical of our seemingly dysfunctional system and its ability to cope with our nation's fiscal realities.

Imagine what significant changes to the tax code or farm subsidies will do to estate planning, corporate taxes, or commodity prices. Do you suppose Congress and the SEC might ever address the real issues of market structure, high-frequency trading or the alternative minimum tax?

It is worth mentioning that on election night Obama prepared two speeches -- one for victory and one as a concession. Even this very intelligent politician, who must have liked his re-election prospects, took nothing for granted. He did not give in to confirmation bias.

Tom Sedoric, managing director-investments of the Sedoric Group of Wells Fargo Advisors in Portsmouth, can be reached at 603-430-8000.

 


 

NHBR Poll