Avoiding behavioural traps in asset allocation

Active management was never meant to be easy. We can’t all outperform and behavioural traps lie around every corner. Tactical asset allocation, in particular, is often set up to fail, in my view, with a small number of big decisions made infrequently by a committee of senior people with limited individual accountability.

That’s why we built our investment process around a specialist multi asset team applying a quantitative framework, based on intuitive macro factors that respond dispassionately to only the most relevant information. No approach can be expected to outperform in every circumstance. But we believe that our approach can generate consistent added value over time in a cost-effective, repeatable and risk-controlled way.

Travelling around the world for a decade as an asset allocation strategist for a large US investment bank, I met all manner of investors. Another decade at a global asset manager allowed me to observe an existing asset allocation capability first hand, and my role gave me the opportunity to make improvements. I brought this experience to bear 10 years ago when setting up the dedicated multi asset team and investment process at Royal London.

How not to do it

In many organisations, asset allocation is decided by way of quarterly meetings of senior people, busily engaged the rest of the time in something altogether different. They get together, sometimes flying great distances, and enter into discussion. Over the course of several hours, the committee comes to a decision. None of the members feel personally responsible for the outcome.

Let’s imagine how it might happen, in the style of the iconic C. Northcote Parkinson¹.

• Person A opens the meeting saying it’s never too late to underweight UK equities, drawing on painful first-hand experience.
• Person B nods vigorously and would go further, saying stocks are outrageously expensive and they’d be in cash – the same point they have made multiple times throughout the bull market.
• Person C read a great investment bank report on the way to work and spends 10 minutes talking about their favourite AI stock. It’s a new era and Person B just doesn’t get it.
• Person D, ever the peacemaker, says the portfolio is fine as it is and all that needs to change is the investment thesis.
• Person E agrees with whatever the chair just said (they’re hoping for a raise), while Person F yawns and checks their phone.
• The Chair is thinking about tomorrow’s Board meeting and wants to get this over with. They sum up briefly, take a vote and set out new asset class targets, marginally different to the old ones.
• Billions of dollars flow across international borders on multiple stock exchanges over the next several weeks as client portfolios are painstakingly rebalanced. A job well done – until the same time next quarter when the jamboree kicks off again…

This is tongue in cheek, of course, but bells will be ringing in every financial district.

My heart goes out to asset allocation heads working under these conditions and then having to account for performance. What if your active equity manager were only able to trade four times a year on dates they didn’t choose, with only a handful of stocks to pick from and with a committee of their colleagues voting on what they should do?

A better way

Naturally, as a multi asset team we need to know what’s going on in every asset class we invest in. In this regard, there’s a huge amount to be gained from discussions with investor colleagues. We host a weekly cross-asset meeting where we review developing macro and political news and get expert input from asset class specialists. It’s short and focused. In my view, it’s a real strength as a middle-sized asset manager that our global investment team is all on one floor of one building in one location. When something unexpected happens between meetings, the flow of information is immediate.

How then should an active asset allocation function best be structured? Here are my thoughts:

• By a small, specialist full-time team, all with ‘skin in the game’ and empowered to make investment decisions within pre-set guardrails.
• Operating a limited number of strategies, reviewed and implemented daily, on the basis of in-depth research into what works and what doesn’t work.
• Informed by quantitative models responding dispassionately to the most relevant information.
• With the models having the loudest voice in the room – but with room for experience and judgement to play a role: history never repeats itself but sometimes it certainly rhymes.

Getting a balance

There’s an interesting tension between thoroughness and decisiveness. We need to be open-minded, un-rushed and thoughtful when designing an investment process but act efficiently and with conviction when implementing it.

I often think of the two systems in our brains that psychologist Daniel Kahneman² described:

• Slow thinking  takes a lot of energy. It’s careful and considered. It weighs up pros and cons. It engages in difficult calculations. You know you’re doing slow thinking because everything else stops. Your eyes dilate. What’s 127 x 32? 
• Fast thinking is instinctive but lazy. It processes readily available information and takes short cuts. It jumps to snap conclusions. On the plus side, it leads to immediate action. Coffee or tea? 

I believe that we need to harness slow thinking to back test and design tactical models, but only fast thinking can decide and trade tactical positions on a daily basis. The critical thing is to surround our fast brains with the ‘good information’ our slow brains have prepared for us, rather than falling prey to extraneous noise and behavioural traps.

There’s one last thing to say about investor psychology. Even if you design a fantastic investment process, you can easily be thrown off course by hubris or despair. Every strategy will see periods of outperformance and underperformance. A chance run of good numbers can make someone believe they’re infallible and make them take unwarranted risks. A run of bad numbers can result in style drift or resignation. What's needed is a level head.

As an active investor, the best we can realistically hope for is to be right slightly more often than we’re wrong – and to keep doing it. This is less natural than it sounds. If we cooked a good meal 55% of the time, our restaurant would be empty. If we crossed the road safely 55% of the time, we’d never leave the pavement.

Active management is hard. But a consistent, evidence-based and data-driven approach can tip the scales in your favour. Understand yourself and you can beat the market.  
 
 ^{1 ‘Parkinson’s Law; or The Pursuit of Progress (1958)’ by C Northcote Parkinson describes many classic institutional pitfalls. ‘Work expands to fill the time available’ was one of his maxims. He also described the concept of ‘comitology’: the science of how committees, government cabinets, and other such bodies are created – and eventually grow irrelevant.}

^{2 ‘Thinking, Fast and Slow (2011)’ by Nobel laureate Daniel Kahneman describes the two ways we make choices: fast, intuitive thinking, and slow, rational thinking. He reveals how our minds are tripped up by error and prejudice (even when we think we are being logical), and gives practical techniques for slower, smarter thinking, to make better decisions.}