Don't catch a falling star

clock • 4 min read

It's a tale as old as time: ‘star' fund manager grows too fast, their style starts drifting as they're increasingly stretched, and it all ends in tears. Careers ruined, investors outraged, decision-makers humiliated, and everyone doing lots of finger-pointing. The Neil Woodford saga is the classic fallen star story, with all the usual twists and turns but with some modern touches such as a hedge fund-style liquidity crisis.

Dotting the i's
It's hard to keep beating the market, even in a fully liquid strategy, beyond a certain AUM level. In search of alpha, managers of growing funds look further and wider fo r ideas - in this case, to illiquid investments. Illiquidity adds risk, as any investor knows. And when everyone runs for the door and there's limited liquidity, the whole house tends to topple.

To those of us who were in the investment industry during the financial crisis, it's deja vu. Still, what triggered the run for the door in this case, and why didn't it happen sooner?

The investors who allocated the largest pots of capital to Woodford had access to analysis of his performance and risk-taking - including traditional attribution analysis that reversed out where his performance (or underperformance) had come from, and what sorts of risk exposures he had borne in getting there. For large allocators, this stuff is bread and butter.

However, this crisis shows us that bread and butter analytics are not enough. That basic analysis would have shown the composition and liquidity characteristics of the portfolio changing as his assets grew. But what it wouldn't have provided was clarity on whether Woodford's recent streak of underperformance was actually due to an investment process breakdown, or just bad luck. And it wouldn't have shed light on the storytelling, incentive structures, behavioural patterns, and biases that enabled investors -- and Woodford -- to ignore the reality of the situation for so long.

Bias pudding
The financial incentives of wealth managers and investment product producers will likely prove to have contributed to their behaviour. It's pretty clear, though, that the inaction pervading the Woodford situation was due to something much deeper: there was an enormous amount of behavioural bias baked into this particular pudding.

It is well-documented, at this point, that human investors, whether professional or not, tend to make better buying decisions than selling decisions. Alpha generation has a life cycle. Why a given fund manager's ideas run out of alpha after a certain point is a combination of numerous factors, including AUM growth and illiquidity. But research shows that alpha decay happens even when AUM is small and portfolios are liquid. As any investor who has experienced a "round trip" - that is, a position that rises in value and then comes right back down to where it started -- can attest, when a stock or a fund has made them money in the past, they are more likely to give the management team the benefit of the doubt when things start to go south. This tendency will likely have contributed to Woodford's major investors lasting as long as they did, given their lack of visibility into his actual investment behaviour.

Once the first large allocator ran for the door, human psychology (and the very rational fear of being the last investor standing in a liquidating fund) meant that the collapse of the fund was all but certain. Looking back, could those involved have done something differently? Can fund managers and the allocators who invest in them learn something from this?

Know your bias - and control for it 
An important takeaway from this affair is to acknowledge that both fund managers and allocators are affected by judgement-clouding behavioural biases, just like everyone else. An investment manager without the ability to see their own strengths, weaknesses and biases, on an ongoing basis, is flying blind. And an allocator who entrusts assets to a manager without ongoing visibility into those strengths, weaknesses and biases is equally in the dark. Basic performance and risk analytics aren't enough.
If both Woodford and his investors had been armed with behavioural analytics, and made productive use of them on an ongoing basis, I suspect there would be far fewer tears now. Both would have been in a position to see whether he was sticking to his own strategy and process, and whether his skill as an alpha generator was intact. If he was just going through a bad patch in the market, that would have been evident in the analysis.
The allocators could stand a look in the mirror, at their own investment processes, too. I suspect Woodford wasn't the first "round trip" manager for any of them. How do they know when it's time to ask for the cheque, save for waiting until everyone else runs for the door? It starts with analysing what would have worked best in the past.
In order for managers (and allocators) to consistently generate alpha for investors, they must build more bias-mitigating discipline into their own investment processes, and use data analytics to shed a light on the efficacy and consistency of those processes. The onus is on them to prove to their investors - and themselves - that they are making the best decisions. 

Clare Flynn Levy, CEO of Essentia Analytics