ALASKAJADE / 123RF STOCK PHOTO

Consider the following scenarios. Preparing for a long weekend at the lake, you’ve checked and re-checked the forecast: scorching hot under sunny skies for all three days. But it’s cold and rainy as you pack, and you find yourself throwing in a sweater and rain jacket, even though the evidence says you won’t need them.

Or, a client wants to buy shares in a tech company that just reported record earnings. Your firm’s analysis warns of new competitors entering the space as early as the next quarter. But your client is focused on the news of the day, and you find them less interested in how the company’s fortunes could change.

While the consequences of these forecasting failures are very different, the same bias may be at play. West Virginia University assistant professor Julian Givi and Carnegie Mellon University assistant professor Jeff Galak have written a paper about what they call the “future is now” (FIN) bias. The marketing professors conducted nine experiments to show that people are “systematically biased toward expecting the future to be like the present, even when the probabilities of future outcomes make such a belief unfounded.”

Their studies are based on subjects being aware of the current context (a score in a game, or today’s weather, for example) as well as the objective probabilities of possible future outcomes. The latter should make the former irrelevant, since basing forecasts on the objective probabilities alone—and ignoring the present—would be the reasonable approach.

It’s not the one most people take, though. According to the authors, that’s because our beliefs are anchored to the present, making it hard to discount current circumstances.

One study tracked forecasts for the direction of an imaginary cold front approaching the African coast. If the wind patterns stayed the same (50% chance, according to meteorologists), the front would continue on Path A the next day, the respondents were told; if the wind changed (50% chance), it would travel along Path B. Though the probability was 50% for either path, 79% of respondents said the cold front would continue on Path A. And many of the 79% expressed strong confidence about this choice.

Another study described two people playing a board game. Participants were told the winner would be determined by a coin flip. Even though the coin flip made the score up to that point irrelevant, more than 70% said the player leading before the last turn would win.

The other studies added layers of complexity, introduced incentives and substituted verbal probabilities (e.g., “most likely”) for numerical probabilities, making them easier to interpret. The authors found similar results.

The reason is that people anchor on the present circumstances before learning the probability of the future being like the present, they wrote.

“At this point, people should (rationally speaking) completely discount the present circumstances and only consider the probabilities of potential future outcomes,” said the paper.

The authors found that’s not what happens, and offer investors a lesson: “financial analysts may believe that aspects of the financial markets are more likely to remain the same (vs. change) than they truly are, and thus make poor investment decisions.”

Related biases

The authors noted that the FIN bias could be a relative of better-known behavioural finance biases.

Status quo or familiarity bias demonstrates that people choose the status quo over alternatives and prefer the familiar. Such biases deal with preferences, they wrote, while the FIN bias is about forecasting.

Recency bias, which is not noted in the paper, leads investors to give more value to recent events than older ones, which can be dangerous in a long bull market. Nobel laureate Daniel Kahneman’s WYSIATI (what you see is all there is) concept in Thinking, Fast and Slow holds that we focus on whatever evidence is most available and ignore what isn’t.

“The ‘future is now’ bias: Anchoring and (insufficient) adjustment when predicting the future from the present” is published in the September 2019 issue of the Journal of Experimental Social Psychology.