Investment research, fast and slow

Helen Davies

Helen Davies

One of the requirements regulators have is that investment research should be based on appeal to reason not emotion. Another way of thinking about this is that it should be the reader’s System 2 that makes the investment decision not their System 1.

This System 1/System 2 framework was created by psychologists Daniel Kahneman and Amos Tversky and detailed in Kahneman’s book Thinking, Fast and Slow. Their theory was that the brain has two systems. One, System 1, is the first line of defence to any stimulus and makes quick decisions based on a pre-set list of shortcuts or heuristics. Fight or flight is one example. System 2 is more deliberative but lazy, and takes over only when System 1 doesn’t have a pre-set response that fits a situation. System 1 works most of the time. The problem is that we often use System 1 when we should be using System 2.

A recent voxpop interview by the BBC’s Nick Robinson illustrated this. He asked people on the street: “Do you think immigration should be cut?” Leave aside that the question was a leading one, what was fascinating was the clear-cut response of the people. Nobody said, or at least nobody was shown to have said “Er, it depends.” They all said “Yes”. But when they were asked “Which people would you stop coming in?” and to choose between social care workers and chefs, they were stumped.

Responses like this one are why the research regulators prohibit emotive language such as “Cheap as chips: Fill your boots”. Appeals to System 1 on complex questions are likely to yield a wrong or incoherent answer. “Cheap equals good equals buy” says System 1. Well, not necessarily says System 2. It might be cheap for a very good reason.

The problem is when to use System 1 and when to use System 2. During a panic, for example, what do you do when everyone is charging for the exit? Joining the stampede rather than assessing your options might be the right thing to do as there may actually be a fire.

For example, the counter-consensual move in 1930 was to buy, but stock prices fell precipitously for another two years, didn’t recover their value for five years, and only started rising again sustainably 20 years later. Fundamental research in that situation would have identified value, but the timescales for realising it made the advice useless for most people’s investment horizons.

The trick is to be aware that our System 1 takes on decisions that should be referred to System 2, and to know when our System 2 should refer a problem back to System 1 if circumstances merit it. In investment research one way to do this is to “delay intuition”, as Kahneman puts it, until the analyst has all the information necessary. This should work in most markets. With careful wording it can also be compliant.

Research Editors can help you word your research carefully – contact us for more details