A common misconception clients have about investment management experts is that we’re good decisionmakers. We’re not. Rather, we’re very good at repeatedly avoiding making stupid decisions.
The path to success often comes through avoiding major mistakes rather than pursuing brilliant moves. This isn’t about being passive; it’s about being thoughtful and disciplined about where and how you compete.
Charlie Munger said it best when he stated: “It is remarkable how much long-term advantage people like us have got by trying to be consistently not stupid, instead of trying to be very intelligent.”
The key to successfully investing over the long term is developing and implementing a system that helps you eradicate errors. To err is human, so my job is to understand how being human makes me a bad investor. This is called the behaviour gap.
Though there are several ways of defining what this actually means, it is most typically used to describe the cost of our poor investment decisions — the difference between potential and realised returns. While this might seem intuitive, for me it is an abstract, almost ephemeral concept.
The behaviour gap is driven by the divide between investment theory and practice.
Investment theory is machine-like, scientific and clinical, which is great but has one flaw — it leaves out the influence of emotion and psychology, and that is the whole ball game! Investment theory ignores the lived experience of investing. So, how do we deal with the behaviour gap?
Many conversations about addressing the behaviour gap often seem to assume that our aspiration should be to behave as if we are a superrational, all-seeing, and all-knowing decisionmaker. We will never reach this state, and judging ourselves against this ideal is unhelpful. We are not machines.
Most of us will also be unable to avoid the decision-making challenges that blight investing by going for decades without touching our portfolio. This might be a sensible approach, but such stoicism is an unrealistic expectation for most of the humans I have met.
This gets at a core issue of why investing is so difficult. Many of the behaviours that have made humans such a successful species also make it difficult to be good, long-term investors. Discussion about investor behaviour often seems focused on creating a long list of detrimental biases that humans suffer from as if we are just a poorly wired species, ill-equipped to make good decisions.
This is not the case — it is simply that certain ingrained behaviours that are effective in some contexts can be detrimental in others. That investment issue that you are now worrying about is very unlikely to be as vital as you believe it to be, but it is very human to act as if it is.
We should not be aiming to be the perfect decisionmaker nor someone who rarely ever decides. Instead, our focus should be on being less stupid. All most investors really need to do is make some reasonable decisions over time and they will be just fine.
There is no ideal set of choices that we can make (except with the benefit of hindsight), and striving for incredible outcomes is often the cause of our worst decisions. The aim is not to get the highest possible return, but rather the return that can be sustained for the longest possible period.
After the fact, there will always be better choices we could have made. We will also always make decisions that have disappointing results. That is inevitable and acceptable. These won’t matter that much — it is the big mistakes that will count.
The key to good investment decision-making is to understand what makes us human, and then to adapt those elements that might also make us poor investors. How we frame something matters far more than we think it should, and we must use that to our advantage.
This is where AI can be a powerful tool for investors. Beyond data collection, AI also plays a key role in analysing and contextualising information. AI in investing represents the convergence of machine learning, data science, and financial analysis. It is not a single tool or system. It’s a group of technologies that can analyse data, identify patterns, and generate insights faster and more scalable than traditional methods.
Natural language processing, for example, can be used to identify patterns in how companies are discussed across different sources. These tools can help uncover emerging risks, assess sentiment, or classify controversies by theme and severity.
Other models are designed to quantify sustainability factors at the issuer or portfolio level, transforming raw data into scores, metrics, or benchmarks that inform investment decisions. Rather than overwhelming teams with more data, AI helps prioritise what’s relevant and filter out noise, improving speed and clarity.
AI is the most powerful framing tool I have yet encountered.
By strengthening the foundation of data and sharpening the focus of analysis, AI is helping investors scale their efforts without sacrificing quality or oversight. If applied correctly, AI can help us be less stupid more often. This is what it was created for.
Not to make us into machines, nor for machines to replace us. This collaborative model allows analysts to cover more ground in less time, but it also requires guardrails. AI tools can analyse information, but experts must determine what’s material, what’s missing, and what it all means in context.
By keeping humans in the loop, investment teams gain the efficiency benefits of AI without sacrificing the depth, oversight, and accountability that sustainable finance demands.
Luthuli is investment management director at Luthuli Capital.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.