PsyFi Search


Thursday, 12 November 2009

Anchoring, The Mother of Behavioral Biases

Behavioural Biases (5): Anchoring

Just as an anchored ship rarely strays too far away from its tethering point a human being prefers to stick close to the references with which they feel most comfortable. Anchoring is an easy-to-demonstrate, hard-to-eradicate behavioural bias that has all sorts of nasty implications for investors, many of them not obvious. In fact, along with availability, it has the claim to be the mother of all biases.

The fundamental investment problem lies in the difficulty in deciding what something is intrinsically worth. A skilled negotiator will start from an extreme position, such as a very high price, in order to frame the subsequent discussions. Anywhere and anytime someone presents us with a number in order to start negotiations we’re being anchored. So if it really matters then you need to start from your own number or walk away.

Random Pricing

Anchoring is almost trivially easy to demonstrate and it’s been replicated many times by many researchers in many situations. Perhaps the simplest example is to use peoples’ social security numbers as a reference. As Dan Airely shows here, by simply getting people to write down the last two digits of the number and then asking them to submit mock bids it’s possible to get people with higher numbers to bid up to twice as much as their lower number companions. Unconsciously the brain sets the social security derived number as the reference point and then adjusts accordingly.

It’s worth dwelling on that. I think it’s simply astonishing and truly shocking that highly evolved, smart primates like ourselves can be fooled into utterly stupid behaviour through a trivial piece of misdirection. And it affects us all.

Anchoring is a particularly pervasive problem and occurs in all sorts of scenarios, being particularly beloved by opinion poll surveyors, real-estate agents, stockbrokers, car salesmen, supermarkets and, well, virtually everyone who wants us to buy things. Wherever and whenever anyone presents us with a number and then asks us to do something with it you can be sure we’re be tricked into behaviour that probably isn’t in our best interests.

Multiple Choice, Multi-Buys

Multiple choice surveys on some kind of sliding scale are an especially fun source of anchoring issues since people will invariably anchor on the middle values. Ask people how many times a week they brush their teeth on a scale of 0 to 20 and they’ll be pulling out the old toothbrush somewhere in the middle of the range. Change the scale to 0 to 40 and they’ll suddenly double their efforts.

Supermarkets use this to generate more sales. Whenever you seen a multi-buy offer – buy five for the price of four, etc – you’re seeing anchoring in action. Even if they don’t take up the offer people who normally only buy one or two of the item will buy three or four. What are they thinking? Well, they’re not of course.

Anchoring Under Uncertainty

What Ariely demonstrates is that when operating under conditions of uncertainty – when we’re unable to accurately assess the value of an object – we can easily be overcome by our own behavioural biases, without even realising. Setting a price based on an arbitrary number is clearly stupid and it’s something that the majority of us wouldn’t do if the decision is clearly presented to us. But it’s not and we do.

When you start looking for it anchoring appears in all sorts of places. Consider how analysts decide whether to upgrade or downgrade a stock. Do they start from a fundamental analysis and make a clean decision? Or do they look at the previous rating and making a judgement on that basis? It’s not hard to guess that most broker recommendations are anchored and adjusted from previous ratings – which themselves may have been subject to the same effects. It’s little wonder that recommendations can drift away from reality over time.

Mystery Stock Buyers

Quite how or why many investors decide on a buying point or price is often a bit of a mystery but frequently the reason can be found in anchoring effects. Investment analyst recommendations are one possible anchor, which is a bit like a ship dropping an anchor on a submarine in a hurry, but many investors and investment managers don’t need outside assistance. The classic anchor, of course, is a buying price. We’ve seen how loss aversion often means that investors fail to sell losers and get rid of their winners but it’s a bang-on certainty that anchoring plays a part in this behaviour.

Overall portfolio value is another classic anchor. You frequently find investment bulletin board discussions peppered with references to individual investors' peak portfolio values while fund managers are often compared to such benchmarks. Yet high tide marks are, by definition, peak events: you’ll spend most of the time watching the tide sway in and out so why make yourself miserable by benchmarking on a random event?

Price or Value?

No doubt readers can think of other anchors, they’re everywhere when you start looking. However, as the Ariely example indicates, the underlying problem is that people are being unconsciously biased to assign a value to an item based on what is, essentially, a random number. Given how easy this is to do and how hard it is to stop ourselves from doing we really need to learn how to distinguish price and value.

Experienced value investors will rarely make the mistake of overpaying for a stock because they start from the basis of an intrinsic value computation. Although such calculations are themselves fraught with uncertainty by building in a margin of safety they’ll start any negotiation with a clear and reasonably objective view of what a fair price is. If you know that the case of fine wine has a real value of about $100 you’re unlikely to be biased by some unconscious trigger into bidding double or triple the true worth.


The ability to distinguish price and value is a key difference between those people we can loosely term “amateurs” and “professionals”. You’ll find plenty of amateurs in the professional investment industry and lots of professionals in the ranks of the private investor. This accords nicely with the research of John A. List indicating that experienced traders are more likely to trade an item when offered a good price than less experienced people.

Some of List’s ideas, which mainly target loss aversion, have been thrown into question by research which suggests that even professionals are loss averse. However, it’s possible that this is to do with the frame in which the research is conducted. Framing is parasitically attached to anchoring and deserves its own extended treatment but basically by presenting the same situation in different ways a frame can cause people to anchor onto different aspects of the problem.

So, for instance, if you present a problem in terms of the possible losses then people will tend to opt for the situation which avoids those losses and become risk takers. However, if you make people anchor on the gains to be protected then they’ll become risk adverse. Almost everywhere you look in behavioural finance you’ll find anchoring somewhere.

The Dumb Shall Inherit the Money

Given how easy it is to trigger anchoring and how prevalent it is in everyday investment then there are really only two ways to address the problem. One is to do the hard yards and learn how to distinguish price and value. The other is to assume that you can’t tell the difference and that, over long periods, the effects of buying too high and too low will average out: avoid the anchors completely by acting dumb.

One thing is definite – we can’t avoid the problem. The best we can do is to learn to deal with it where and when it’s important. Only you can decide as to whether that’s in stock buying, house purchases or simply the weekly supermarket sweep. However, you can be sure that where there’s a price there’s an anchor.

Previous Article: Behavioural Biases (4): Regret
Next Article: Behvaioural Biases (6): Recency


  1. One thing is definite – we can’t avoid the problem.

    As is often the case, I think you do a fantastic job of describing a very real problem and then drop the ball when it comes to helping people cope effectively, Tim.

    You hint at the solution to the problem in some earlier words from your own article:

    Experienced value investors will rarely make the mistake of overpaying for a stock because they start from the basis of an intrinsic value computation. Although such calculations are themselves fraught with uncertainty by building in a margin of safety they’ll start any negotiation with a clear and reasonably objective view of what a fair price is.

    Value investors don't anchor. They determine a fair price and then make sure to pay less than that. So they have the anchoring problem beat.

    You say that these determination are "fraught with uncertainty." I presume you mean that it is hard to pick individual stocks effectively. Okay. So it would seem to me that the answer is to employ valuation assessments but to do so without picking individual stock.

    What if we take the Value Investing concept and apply it to purchases of a broad index? The performance of the broad indexes isn't fraught with uncertainty, is it? The broad indexes in the U.S. have been paying a long-term return of 6.5 real for a long, long time (as far back as we have records). It would seem to me that that number is a solid neighborhood number.

    Add a valuation adjustment (necessary to avoid the anchoring problem) and you're set. You're a Value Investor (no anchoring problem) because you change your stock allocation in response to valuations (you do not invest passively). But you avoid making stock picks that are fraught with uncertainty by investing only in broad indexes.

    It seems to me that this is the best of all worlds. I call this approach Valuation-Informed Indexing.


  2. Hi Rob

    I understand what you mean but I don’t really focus on telling people what they should do – there are a million and one websites that attempt to do that. The main aim here is to uncover the psychological and perceptual issues that prevent people from recognising that they have a problem in the first place.

    If I touch on solutions at all it’s really only to point in the direction of ideas that might help - although I don't believe that there's a single solution that's suitable for every personal psychology out there. However, I leave the details of such solutions to those better qualified than myself ...

  3. I think you do a fantastic job, Tim. I've said it before but I see no harm in saying it again. I view this as the best written blog that I have come across. And it examines an issue of great importance as well. I think you are doing truly important work here.


  4. I am worried about 'anchoring' in relation to property prices in the UK.

    In property i think it has a stronger effect. The prices are much more public.

    They are clear overvalued in the UK but we will struggle to see them reach their true value.