Blind spots (3): Anchoring effect, the pernicious one
“More money has probably been lost by investors holding a stock they really did not want until they could ‘at least come out even’ than from any other single reason.” — Phil Fisher
Two weeks ago, I wrote on a psychological blind spot called Confirmation Bias. This week’s blog is on a rather pernicious bias called Anchoring.
A few years back, I snap-ran a mini-experiment with a few colleagues of mine. Its results were instructive particularly given that the experiment was done on the fly with no prior preparation. Before getting into its details, let me run the same with you.
Without looking up the actual answers please answer the following questions:
· Do you think Mahatma Gandhi was older than 60 or younger than 60 when he died? Make a note of your answer on a paper
· Now, how old do you think Mahatma Gandhi was when he died? Make a note of this too.
Similarly,
· Do you think Martin Luther King Junior was older than 60 or younger than 60 when he died? Make a note of your answer on a paper
· Now, how old do you think Martin Luther King Junior was when he died? Make a note of this too.
Here are the answers:
Mahatma Gandhi was 78 years old and Martin Luther King Jr was 39 the time of their death.
Assuming that you are aware of these personalities but did not know specific details such as their birthdays and death days — in other words, you were guessing based on very little information — it is quite likely you fell for anchoring bias. The number 60 would have anchored your guesses. Try this experiment among your friends and you will witness anchoring bias in action. Answers would have been closer to 60 than otherwise. You can check this by using a control group in your experiment by not anchoring some of your friend’s guesses (ask for their life-span without asking the first question). I tried this and there was no evidence of any explicit anchoring effect for this control group.
This is a result of ad-hoc experiment designed in an office lift going up just six floors. Now imagine the impact this effect could have in carefully designed experiments. Think advertisements and politicians. Your decisions are way more influenced by anchoring effect than you imagine.
At work, how are your expectations managed with regards to, let’s say, year-end compensation hike and possibly promotions?
How does management of a publicly listed company manage equity analysts’ expectations of company performance?
When you buy stuff, anchoring effect plays a role in terms of you being drawn towards discounts. Discounts, many times, anchor you to the original high price and hence you find the discounted price attractive.
Anchoring effect is particularly strong in anything that involves negotiation like house purchases.
Supermarkets, liquor shops etc display high priced alternative options right next to the goods that they would really like you to buy. In contrast to this, where pricing is arrived on the basis of cost + margin, there is relatively lesser anchoring impact than when it is based on subjective perception of value. For example, value of wine is deeply rooted in how the wine industry anchored our value perception — older wines are considered significantly more expensive — try running a test among your social circle by offering wines across a price range (preferably of similar vintage) and ask them how would they rank them in taste. I’m sure you will not find a clear relationship with price but if you try doing this with the bottles in open display — you most likely will get a ranking in line with the pricing.
At work places, when new protect timelines are being assessed/estimated, beware of what you propose upfront as your estimation (the one with lot of assumptions) because that estimate is going to be anchored in the minds of project stake holders and you will end up with a timeline in its ball park.
Daniel Kahneman mentions an example in his book, Thinking Fast and Slow, where anchoring effect was evidenced when real estate agents were given an opportunity to assess the value of a house on market after having visited the house and reviewed a comprehensive booklet on the house that included the asking price. Kahneman observed that even professional real estate agents were influenced by the asking price to a similar extent that non-professionals were influenced. The asking price in the booklet anchored their estimate of value. Even subject matter experts and professionals are affected equally by anchoring effect just like the rest of us.
Most forecasts, either by regular people or by experts, suffer from anchoring bias.
“We try and avoid the worst anchoring effect which is always your previous conclusion. We really try and destroy our previous ideas.” — Charlie Munger
What can we do to address this?
Again like with most cognitive biases, it is hard to address this. Given that anchoring bias generally acts surreptitiously, it is difficult to even recognize it. Being knowledgeable of this bias, trying to get alternative views on your estimates, seeing if objective measures can be used to assess value and perspective, sleeping on big decisions before acting on them etc are some of the ways to address this bias. Also, Munger’s mantra “Invert, always invert” helps here too.
Also, remember that while we need to watch out for this effect on us, we can (and probably do) use this bias as a tool in influencing outcomes as well.
