In brief
In this Insights thought leadership piece, we explore how to recognise and manage behavioural biases that can impact investment decisions.
The context
Behavioural biases influence investment research and capital deployment. Psychologists have long known that humans face these biases.1 Artificial intelligence tools may demonstrate them too.2 But, as we will show in this article, they can be overcome. Understanding and responding to behavioural biases is core to our approach at Generation.
Over the long sweep of evolutionary history, humans have devised many mental shortcuts. To get stuff done, people cannot think like algorithms, but instead must use heuristics. Heuristics allow people to act under uncertainty, protect their wellbeing and coordinate socially. For instance, a purely algorithmic human would throw away drinkable milk because it was one day past its expiry date. A heuristic human smells it, judges it ‘fine’ and then gets on with their day.
These mental shortcuts can misfire in financial markets, however. This is where behavioural finance comes in. The most productive use of this research, in our view, is inward‑looking. It improves the design of internal processes and also cultivates humility.
Recognising and managing behavioural biases
If managed well, teams are cleverer than individuals. At Generation we believe a mix of tenures, past careers, backgrounds and nationalities drives us towards better decisions. Unfortunately, though, groups also have points of failure. We seek to design our group interactions in order to prevent them.
First step: awareness
We raise awareness of biases, including via external speakers, training sessions and book clubs. This training encourages people to look for biases in their decision-making. We also have an internal document to raise awareness of common biases: confirmation bias, halo effects, narrative fallacy, anchoring, base-rate neglect, planning fallacy, sample-size neglect and others.3 To a degree, talking about the problem can help solve it.
Awareness of the jargon of behavioural finance has another benefit. It makes biases discussable without making it personal. It is easier to hear “Are we anchored?” than “You’re biased.” A shared vocabulary lowers defensiveness and preserves trust.
Second step: meeting design
We design meetings to reduce behavioural distortions. We emphasise four design principles.
1. Encourage scepticism. This helps counter cognitive ease, the tendency to accept ideas that feel familiar. We encourage people to play the role of devil’s advocate, whose job it is to quiz an argument. More recently, we have worked with an external short seller who probes for holes in our theses.
We discourage people from offering sales pitches when talking about a company in which we might invest, encouraging them instead to identify things they are worried about. Another technique is the use of the ‘pre-mortem’ or ‘obituaries,’ where we imagine ourselves five years in the future, explaining to each other why this investment went wrong.
We encourage people to be particularly sceptical in their interactions with AI. Although AI-generated text can be convincing, it often contains logical holes or factual inaccuracies. We find that AI chatbots often confirm the initial views of the user rather than challenging them. AI tools are useful for research, but it is crucial to be aware of behavioural biases when using them.
2. Design for introverts and extroverts. Some people think best aloud. Others think best by writing. Neither cognitive approach is superior. We therefore work hard to ensure our process does not favour speed and verbal dominance. To do this, we use multiple channels to discuss ideas: pre‑reads circulated in advance, silent reflection time, structured rounds where each person speaks and written follow‑ups to allow people to express themselves when better thoughts arrive after the meeting.
3. Understand the different challenges facing junior and senior people. There is a common belief that junior investors are particularly susceptible to behavioural biases. After all, they do not have as much experience. They are keen to get ahead, so may ignore inconvenient evidence if it stands in the way of their ambition.
We argue that senior staff can be just as vulnerable to kidding themselves. They are more likely to be stuck in their ways. Their brains are less elastic. Put it this way: when you’re young, incentives work against you; when you’re old, biology works against you.
We structure meetings to account for these different challenges. Junior staff must be appropriately challenged on their arguments, and understand that it is nothing personal. But we also need mechanisms to challenge senior members of staff. One way we do this is via anonymous voting on an investment case. This helps everyone, especially junior members of staff, act on what they really think.
4. Preserve contemporaneous reasoning. The ‘narrative fallacy’ makes hindsight feel coherent and inevitable. This is highly risky for investors. You can end up continuing to hold a position, even though the investment case has changed, by misremembering why you bought the stock in the first place. How to fight against this?
Where possible we record the original rationale for an investment decision, including the notes from our morning meetings, and feed them into an AI model. This includes the assumptions and uncertainties behind the decision. Creating this record helps us strive to remain true to the original thesis – as well as tracking any deviations from it, thereby preventing ‘mission creep’ as much as possible.
Behavioural biases and portfolio managers
Different behavioural biases affect portfolio managers. These include:
- Loss aversion: the pain of a loss exceeds the pleasure of an equivalent gain.
- The endowment effect: ownership increases perceived value.
- Sunk‑cost thinking: prior effort pulls us towards ‘staying the course’ even when expected value has changed.
- Consistency bias: we want to be consistent in front of peers since there is a deeply ingrained social stigma to being a ‘flip-flopper.’
Behavioural biases in portfolio allocation are a particular challenge for a long-term investor such as Generation. We like to hold companies for many years.
We believe this gives us the ability to look through short-term noise about a company, waiting for the market to price long-term value appropriately. But it requires tremendous discipline to distinguish between, on the one hand, a meaningless blip and, on the other, news that means we should sell.
To that end, we employ a number of techniques. One simple behavioural trick helps counter the endowment effect. We ask ourselves “If we did not own this today, would we buy it at this price and size?” If the answer is no, continued ownership requires an explicit justification.
In addition, it is important to look at old investments with new eyes. Every few years analysts re-present their investment cases. This exercise allows them to look at their company from first principles, helping them overcome confirmation bias. Sometimes this exercise leads analysts to be even more convinced by the investment case, while at other times it leaves them concerned. We try to make room for analysts to feel equally comfortable going in either direction. Ultimately, the market rewards intellectual honesty.
Another technique is our stop-loss provision. This is a pre-set rule to sell an asset if its price falls to a specified level, in order to limit further losses. In the past, we had a fairly strict set of rules around the stop loss, but then we conducted data analysis on our use of this tool. We found that, under certain conditions, it might not be appropriate to trigger the stop loss. Today, therefore, we tailor the rules on the stop loss on an investment-by-investment basis.
Even that is no panacea, however. It can be tempting to override the stop loss. For that reason, we try to explain ourselves to the group whenever this happens, as well as encouraging people to challenge us.
Final thoughts
Behavioural awareness is a crucial part of allocating capital in a noisy, uncertain world. We will never eliminate behavioural bias. This is impossible. We hope instead to build a process in which the costliest biases are easier to notice, natural to discuss and less likely to dictate outcomes. This is always a work in progress. We hope that our efforts to put together this piece will further help us in this regard – and will also help you.
- Judgment under Uncertainty: Heuristics and Biases by Amos Tversky and Daniel Kahneman.
- Behavioral Economics of AI: LLM Biases and Corrections by Pietro Bini, Lin William Cong, Xing Huang and Lawrence J. Jin
-
See here for a comprehensive list of behavioural biases: https://thedecisionlab.com/biases
Important information
The Insights: How We Think About Behavioural Biases in Investing report is prepared by Generation Investment Management LLP (“Generation”) for discussion purposes only. It reflects the views of Generation as of May 2026. It is not to be reproduced or copied or made available to others without the consent of Generation. The information presented herein is intended to reflect Generation’s present thoughts on sustainable investment and related topics and should not be construed as investment research, advice or the making of any recommendation in respect of any particular company. It is not marketing material or a financial promotion. Certain companies may be referenced as illustrative of a particular field of economic endeavour and will not have been subject to Generation’s investment process. References to any companies must not be construed as a recommendation to buy or sell securities of such companies. To the extent such companies are investments undertaken by Generation, they will form part of a broader portfolio of companies and are discussed solely to be illustrative of Generation’s broader investment thesis. There is no warranty that investment in these companies have been profitable or will be profitable. While the data is from sources Generation believes to be reliable, Generation makes no representation as to the completeness or accuracy of the data. We shall not be responsible for amending, correcting or updating any information or opinions contained herein, and we accept no liability for loss arising from the use of the material.