Behavioural man v rational man
As one of only a handful of firms across Europe managing genuinely goal-based and outcomes-driven portfolios for individuals, our clients are in a good position to testify to one of its most important claims: it is more motivating and engaging and more likely to inspire confidence. The ‘more’ here implies that it changes behaviours: either decision-making behaviour or actions based on the way investment risk is experienced. Client surveys have confirmed this to be the case.
Whilst clients understand that goal-planned portfolios end up looking different (both from each other and from the portfolio they previously held) we tend to relate that to the effects of following a liability-driven approach borrowed form institutional practice, where it also changed portfolio construction radically. We present it as a contrast with conventional balanced management, which it is. But there is another ideas schism in finance that touches us directly which we refer to less often: the growing conflict between classical finance, in which agents are rational and make decisions based on maximising their personal welfare, and behavioural finance, in which agents are disabled by cognitive biases and emotional weaknesses.
Yesterday I found myself moderating a discussion on goal-based investing at a ‘client communicatio’n conference. On the panel was one of the world’s leading proponents of behavioural finance, Greg Davies. As both an academic and leading writer on the subject, he is unusual in having been given a powerful platform for putting the concepts of behavioural finance into practice, as head of investment philosophy and process at Barclays Wealth, no less.
What Barclays and Fowler Drew share is a belief that the mental accounting devices employed in a goal-planned approach to investing private wealth can be both motivating and engaging for individuals who are not turned on by finance or household economics. Where we are likely to differ is in the belief that investment solutions need to find an accommodation with people’s own skittish and flawed risk personalities rather than assume and impose classical rationality via the adopted solution. We appreciate the way behavioural analysis has improved our understanding of the gap between classical finance and actual decision making but it does not displace rational finance theories. Instead, we see our role as to modify behaviours on the basis that better outcomes and journey experience will follow from acting more rationally.
In shaping Barclays’ approach, Greg introduced a sophisticated process, at the start of the relationship, to test for individual risk personality. This follows the conventional sequencing of the process of taking on new clients, starting with discovery of the client’s circumstances, financial position and objectives and, as part of that, their past investment experience and behaviour around risk. I would not be surprised if Barclays’ risk personality analysis is more sophisticated and revealing for being based on the recognised cognitive biases uncovered by behavioural academics compared with, for example, the off-the-shelf psychometric personality tests used by most IFAs and many wealth managers.
By contrast, Fowler Drew clients will not be able to recall, as part of the Discovery process at the outset of their holistic financial planning, or the ‘Initial Review’, that we were obviously interested in how they took decisions in the past or that we asked any questions about their attitude to risk or tolerance of loss. These form part of the scope of the regulations in financial services against we might be tested but the rules are not prescriptive about when and how they are dealt with. There are three reasons for making these outputs of planning with the client rather than inputs:
They are likely to be changed by the mental accounting insights of assigning different pots of money different jobs.
They are likely to be changed by a planning approach to each goal that focuses on outcomes, and consequences of better or worse outcomes, rather than the experience of market risk as short-run volatility or changes in the market value of the investments.
There is plenty of survey evidence that any financial planning where before there was none changes behaviour.
So it was surprising to me (and David Anderson, who was in the audience) that Greg also believes that it is the role of advisers to modify behaviour, not just to seek an accommodation with it. Less surprising, given that bevioural finance largely uncovers weaknesses not strengths, Greg sees the modification as making them more likely to enhance personal welfare in a way consistent with classical finance. In other words, the differences between goal-based investing and conventional investing are not explained by finance theories so much as practical insights into the role and utility of professional agents. Behavioural finance can help with these insights but it can also get in the way, by appearing to be more of a philosophical challenge than is really the case.
Barclays, like us, believes that differences between goals are accurately defined by their nature (real or nominal, capital sum or stream of payments) and by duration (which Greg explained as the dollar-weighted average length, in years, of their lifetime cash-flow horizons). It sound like they are somewhat more likely to conflate different claims on cash at different stages, such as family-raising and retirement spending, into a single multi-goal portfolio by approximating its duration, but that is a small difference. We also frequently conflate, for instance, retirement and bequest goals by making the bequest a potential use of the payoffs from risk taking in the retirement portfolio, rather than create two separate portfolios.
The main difference is in the insight, which comes from liability driven investing (LDI), that the nature of the goal dictates the risk-reducing or hedging asset. Hence we divide our investment opportunities into risk free inflation-indexed government securities and equities. This allows us to project probabilistic real outcome ranges at every horizon of a dynamically-changing duration-based portfolio with greater confidence than if estimating long-term real returns at multiple horizons of a largely-static balanced portfolio. Barclays, for all its advanced thinking, is stuck with the conventional product solutions of portfolios constructed to match a target location on a risk spectrum defined by volatility.
It was from my other panel member, Leonie Woodward, who is head of global performance reporting for US investment consultants (turned managers) Russell Investments, that I needed to draw any challenge to the convention of balanced management. Based on her observations at close hand of the way LDI radically altered the investment solutions adopted by pension funds, she confirmed our view that the test of goal-based investing is that it changes the portfolio construction, introducing a specific hedging asset and eliminating some assets entirely that are otherwise relied on in balanced portfolios. Without changing the portfolio construction, you might not have changed the engagement, focus and decision-making behaviour of pension-fund trustees – which, Leonie suggested, has been one of the client-communication benefits of LDI.
Why not change the investment solution for private wealth? I had set the scene by quoting EDHEC Business School’s challenging 2009 paper on the failure of wealth managers to adopt LDI. The authors found survey evidence that we know, as an industry, that we ought to make a better connection between planning personal objectives, which we do well, and designing optimal portfolios to deliver the planning outcomes, which we don’t. As Greg conceded, you cannot overestimate the difficulties of changing conventions in a business that needs scalability and standardisation to be profitable. Unlike Fowler Drew, he did not start at Barclays with a blank sheet of paper, with no legacy thinking, products or IT systems. And unlike Fowler Drew, Greg told me afterwards, it is rare for Barclays to be sole adviser or manager, yet the entire goal-based approach presumes holistic planning and a single overarching strategy of capital efficiency for a family.