Your brain’s tendency to seize upon – or “anchor” to – recent information can be disastrous for your investment decisions, says Cris Sholto Heaton. Here’s how to get around it.
“Anchoring” is one of our biggest handicaps when it comes to making sound investment decisions. The term was first coined by psychologists Amos Tversky and Daniel Kahneman in 1974, and describes the way that human beings focus too heavily on a single piece of information when making forecasts. Rather than generating ideas from scratch, we fixate on a specific number or particular event at the outset, then try to forecast by adjusting this scenario. Various studies have shown just how powerful – and damaging – this tendency can be.
Forecasts are easily influenced
In one study of anchoring, a group of currency traders were asked to estimate the future dollar-euro exchange rate. All worked for the same firm and shared the same information. So while their individual guesses might be different, you would expect them to produce broadly similar answers. But when the researchers split the group in two, they achieved very different results simply by asking the traders a question before asking for their forecasts. The first group was asked if they expected the rate to be above or below 0.6; their average guess was then 0.79. The second group was asked if they expected the rate to be above or below 1.6; they guessed 1.28 on average.
This huge gap arose because the traders anchored on the number first presented to them, even though the question contained no useful information to help them make a good forecast. Worse still, other tests show that even entirely unrelated information can have this effect. Dan Ariely of MIT carried out a mock auction with his MBA students after first asking them to write down the last two digits of their social security numbers. This clearly had no relevance to the auction process – yet the half of the group that had written down higher numbers bid 60%-120% more than their peers.
This has major implications for investors, who tend to anchor on analysts’ earnings estimates or credit ratings. But even being aware of ‘anchoring’ doesn’t prevent it from affecting your judgement. So how can you get around the problem?
How to get around anchoring
There’s no easy answer. But giving yourself multiple anchors might help. Some time ago, Chris Lobello and Connie Lacanilao, analysts at Asian brokerage CSLA, carried out a study into ‘anchoring’, using their colleagues as lab rats. The equity sales team was asked to give a forecast for the Japanese market’s closing value each Friday. Rather than pick a specific number, they were asked to choose a range within which they felt 90% certain it would close.
The team was split into two groups. One received an email that listed just the previous week’s close. This group did badly; the actual close was within their predicted range only 41% of the time. The other group were given extra data: the largest previous weekly shift in percentage terms and what index levels a similarly sized move (up or down) would equate to from the current level. This meant they had three anchors – the current level, a plausible high and a plausible low. Forecasting success jumped – the actual close was within their predicted range 70% of the time.
So a useful lesson is to get more detailed information before you invest: at least then you’ll be anchoring to relevant data. And learn from history. Looking at what has happened at similar points in the past will stop you from fixating on recent experience. Using sky-high dotcom valuations to gauge where the market “should” be is what ruined many an investor after the tech bubble burst.
Another useful approach is to turn the problem you’re looking at on its head. James Montier of asset manager GMO, once pointed out that when valuing equities, it’s easy to anchor on the current price. If we like a firm, we convince ourselves it’s worth more to justify buying. So if we’re trying to value a stock through discounted cash flow (DCF), for example, we edge up our earnings forecasts until we get the answer we want. (DCF involves calculating a present value for the future stream of cash payments that an asset is expected to generate, discounting each expected payment to reflect the risk that it doesn’t materialise.)
Montier suggested reversing this approach by calculating the growth rate implied by the current price of the stock. This can then be compared to past performance by similar firms to see if it’s actually likely to happen. Again, this takes you from anchoring on a single company and its stock price to looking instead at a wider range of possibilities.