I recently sat down with Greg B Davies, Managing Director at Barclays to discuss his team’s innovative applications of Behavioral Finance. As Head of Behavioural and Quantitative Investment Philosophy, Davies and his team at Barclays are engaged in groundbreaking work at the intersection of two sub-disciplines in Finance typically treated as irreconcilable, finding creative solutions for portfolio and risk management.
What follows is a brilliant and wide-ranging Q&A encompassing topics such as the value of Technical Analysis, harmonizing Modern Portfolio Theory with Behavioral Finance, and the many unique insights the discipline offers individual investors in the aftermath of the Great Financial Crisis. Our sincere thanks to Barclays and Mr. Davies for a thoughtful and fascinating interview.
Allan Millar: Greg, to begin, could you let us know what stimulated your interest in Behavioural Finance? We can see from your first degree through to your PhD that Economics, Philosophy and Finance have been extensively covered; does this suggest an early focus?
Greg B Davies: I had always been interested in all three of those fields, but until I started researching possible topics for my PhD I had never heard of behavioural finance. My initial intent was to focus my thesis on the pure philosophy of rationality (about as far from a practical, career-oriented topic as possible). As I was reading around the topic, I came across the work that psychologists had been doing to understand practical decision-making. This field offered the tantalising opportunity for me to not only combine all my previous academic interests, but to learn about psychology, which was at that point entirely new to me.
Millar: Most of See It Market’s readers are familiar to some extent with Behavioural Finance, but could you provide a quick overview?
Davies: Academic behavioural finance is the study of financial decision-making as observed in the real world. It attempts to look beyond classical economic theory which only really works if we all put our emotions, personalities and preferences to one side and behave like the mythical Homo Economicus – a creature that obeys all the investment rules but doesn’t really exist. Applied behavioural finance should be – although isn’t always – about how to help people make better decisions, therefore, getting closer to the theory while retaining what is important about ourselves as individuals.
Millar: See It Market’s contributors and readers are active traders, most favouring technical analysis. Could you give us some insight into the behavioural element of technical analysis?
Davies: Markets are undeniably driven by crowds and their sentiment. Technical analysis attempts to provide signals about the crowd’s emotional state and so predict what it might do next. Unfortunately – aside from some high level features of market behaviour, such as momentum – technical analysis appears to have very little predictive content.
The bulk of our enhanced understanding from the field of behavioural finance relates to individual decision-making, not market behaviour. We have a much better understanding of how individuals make decisions, but have no real basis for aggregating this up to the level of the market. It is analogous to taking our knowledge of the movement of atoms and assuming that this will aggregate to an understanding of fluid dynamics. It doesn’t: we hit a wall of complexity and chaos. The system (and the market) has emergent properties that no amount of understanding of the individual components will allow us to predict. At Barclays, our focus is never on forecasting the markets, but on helping individuals make better decisions.
Millar: Looking at what you do with Barclay’s, designing and implementing Investment Philosophy.com and establishing a commercial Behavioural Finance unit seems to have prompted some other banks to follow suit. Does this reflect upon the higher profile that behavioural finance has in investing and wealth management?
Davies: Absolutely. We started our Behavioural Finance unit in 2006 and for most of that time it had been the only one of its kind globally. Since then, however – and helped by the emotional rollercoaster ride of the financial crisis – the idea that the psychology of decision-making has an important role to play in improving financial behaviour has gone from strength to strength. Not only have banks been paying increased attention to this area, but many regulatory bodies, most notably the Financial Conduct Authority in the UK, have been promoting behavioural economics as an essential part of financial regulation.
Millar: Following on from that, we can see that you have been published extensively over the last seven years. Has the reaction to behavioural finance changed in that time and is it reflected in feedback to your articles and books?
Davies: Yes, over time it has become much easier to write about behavioural finance as a practical tool for improving financial behaviour. Seven years ago, many audiences had simply never heard of the field and, among those who had, it was largely treated as a set of entertaining anecdotes, principally of academic or journalistic interest. Today, there is a much wider level of awareness and acceptance of the subject, and the sophistication of the dialogue in response to articles and books has increased significantly. I particularly welcome the healthy scepticism that now exists around the field, with many looking for practical applications and outcomes, rather than just cute stories and pictures of brains on PowerPoint presentations.
Millar: Your most recent book is “Behavioral Investment Management: An Efficient Alternative to Modern Portfolio Theory”, could you provide a brief synopsis (without giving it all away) of how behavioural finance contributes to constructing a portfolio? What would be Markowitz’s take on it?
Davies: Traditional Modern Portfolio Theory (MPT) embodies a particular assumption of how people trade-off risk and return. It is psychologically implausible, but it simplifies the maths as it allows us to assume that risk is entirely accounted for by using variance as a measure. This would be fine if all asset returns were symmetrical and there were no fat tails, but we know this is not the case.
We have reworked MPT to embody more realistic assumptions of investors’ risk-return preferences. Essentially, the result is a new measure that accounts for risk in a way that more closely resembles what matters to investors. This means that when we construct the risk-return efficient frontier, we are using the actual risk more efficiently. Looked at another way: by reducing risk, we protect investors against true risk, rather than an imprecise approximation of risk.
I’d like to think that Markowitz would quite like it! He developed MPT in the 1950s, and what we’ve tried to do is provide the portfolio theory that his thinking would have led him to if he had access to today’s behavioural knowledge and computing power.
I should also add that both MPT and our revised behavioural portfolio theory seek only to provide the theoretical right answer for investors’ long-term financial preferences. Both models ignore emotional responses and short-term preferences along the investment journey. We wanted to build the theory on behaviourally robust foundations, but it’s important to remember that constructing a real-life portfolio involves more than knowing the best allocation for your long-term financial preferences – it requires controlling your short-term emotional responses too. It’s not about pursuing optimal ‘risk-adjusted’ returns, but about attaining the best ‘anxiety-adjusted’ returns.
Millar: On a similar note, when one looks at financial theory, whether it is calculating the Cost of Capital, using the Capital Asset Pricing Model, Black –Scholes or just trying to establish a Cost of Equity, there are always a number of assumptions that have to be made. Do you think that understanding the psychology behind finance makes these models more robust?
Davies: As with our reworking of MPT, an understanding of psychology can be used to tweak the models and put them on a more behaviourally sound footing. However, this solves only part of the problem. Frequently the robustness of the models is much less important than the robustness of our use of the models. As suggested by the statistician Box: all models are wrong, but some are useful. A better understanding of psychology can make the models a little less wrong, and a little more useful. Ultimately, the biggest improvements come not from applying psychology to the models, but to the design of the processes that govern how you use the models.
Millar: Returning to your work at Barclays, you begin with the Financial Personality Assessment and then allocate assets. Are most people surprised by their risk profile and are you now a pretty good judge of people when you first meet them?
Davies: People tend not to be surprised by their risk profile (although it does happen from time to time). The reason is that most people don’t have a stable introspective assessment of their own risk profile to start with, or even a clear idea of what the term means. Indeed, while our Financial Personality Assessment does contain a stable measure of long-term risk tolerance, it also measures two additional components of risk attitude, each of which tell us something specific about the way in which investors’ respond to risk over the investment journey. Our intuitive assessment of our own risk tolerance is both unstable (tending to be higher in good times, and lower in bad), and will often be a jumble of several distinct components. The notion of ‘risk profile’ is difficult to pin down to a single thing, which is one of the reasons why we use psychometric testing in this area.
I can be a pretty good judge of people when I’ve sat down with them and asked them the right questions, but I’d always prefer to start with some objective psychometrics measures as inputs into this conversation.
Millar: One of the noticeable aspects of the work you do is that the emotional needs of the investor are catered for, as well as the financial. Would you say that this is a key principle of Investment Philosophy?
Davies: Yes. The solution of classical finance is often held to be the ‘rational’ model, but there is nothing rational about aiming for investment perfection without taking our emotions and human frailties into account. There is nothing rational about pursuing perfection if you’re likely to fail… and fail expensively.
Our Investment Philosophy strongly embodies the idea that, to be a good investor, you need to be able to make continually good decisions over the whole investment journey. To do this, it is not enough to know the theoretical ‘right’ answer; you also need to have the right level of emotional comfort to enact and to stick with it. Almost all deviations from good financial decisions are a manifestation of the investor’s innate need for short-term emotional comfort – often at the expense of long-term financial returns. The right question is not, as classical finance would have it, how to eliminate your emotions. We’re human! The right question is: how do I acquire the emotional comfort I need at the lowest cost to my long-term financial performance?
Millar: The quantitative part of your role is obviously very important. I am interested in the strategic and tactical allocations, and the review timeframes. Could you provide some insight to your thinking on this? Meeting the needs of the client and portfolio diversification is important but quantifying that must be pretty difficult?
Davies: It strikes many people as unusual that our Behavioural and Quantitative teams are joined, but for us this is essential: a practical framework requires combining the best of both classical and behavioural finance. The most important question for investors is: what is your wealth doing in the background while you’re busy fretting about short-term decisions? In other words, your actual strategic asset allocation (including your cash position) is paramount. We strive to give clients a) a strategic diversified asset allocation – covering five years or more – for their total wealth, and suitable for their risk profile, b) the emotional comfort to implement and stick with this allocation, and only then do we c) worry about tactical positioning over a three- to six-month horizon. The principle is to get the big-picture positioning right before worrying about the details.
Millar: Could you provide some detail on the Black-Litterman model and its importance in estimating future returns?
Davies: The importance of the Black-Litterman model is that it reduces our reliance on a single, realised history reflected in the data we have. Over any chosen time period, some assets will have performed better, or worse, than can be explained purely by their risk and correlations with other assets. If we use only historic data to estimate future expected risk and returns, we are essentially ‘optimising’ a portfolio for what has happened, not what might happen.
Black-Litterman allows us to, somewhat, break this dependency on historic data by ‘correcting’ expected returns towards those that would be expected in equilibrium conditions where assets over the long-term earn returns commensurate with their risks. It also allows us to adjust for subjective forward-looking views (although these should always be used with caution).
Millar: Again, without giving away all of your secrets, can you share some information on the asset classes you use and the quantitative work that goes into allocating weights based on profiles?
Davies: We take a high-level approach to selecting asset classes. A good asset class is, in our view, one that shows sufficiently distinct long-term properties to warrant being considered separately in our strategic asset allocation. We work according to nine different asset classes:
- Cash and short-maturity bonds
- Three fixed income asset classes: government bonds; investment grade credit; and high yield and emerging markets debt
- Two equity classes: developed and emerging markets
- Three alternative asset classes: real estate; commodities; and alternative trading strategies.
This is not to say that more granular divisions are not valuable in shorter-term tactical positioning but, for strategic allocation, lower level divisions are often just splitting hairs. It is more important to get the big things right, leaving space for flexible implementation, as required, over shorter horizons.
Millar: A pretty popular book in the finance community has been Kahneman’s “Thinking, Fast and Slow”. Two of the concepts I particularly picked up on were narrow framing (for readers: not considering investments in a portfolio context and overestimating the risk) and loss aversion (for readers – investors more sensitive to losses than gains). Could you share your thoughts on how these impact longer term investors?
Davies: There are two types of narrow framing that are important here: one, as you mention, is the tendency to evaluate investments one by one, ignoring portfolio effects and the larger context. The other is narrow framing in time, or myopia: the tendency to focus on the short term, even when your objectives are long term. These naturally lead investors to ignore diversification across the individual components of their portfolio, or diversification over time. The biggest effect, however, comes when narrow framing is combined with loss aversion.
Loss aversion sounds reasonable enough when you first encounter the notion; after all why shouldn’t investors be more sensitive to losses than to gains? The reason is that whether a given financial outcome is perceived as a gain or a loss is heavily dependent on the framing. The narrower the frame – i.e. the more you chop your portfolio into small pieces, and observe it over small time horizons – the higher the proportion of losses you observe for exactly the same investment. This raises the level of loss aversion, which, in turn, influences your behaviour. For example, if you hold a portfolio with 50 positions and glance frequently at the daily returns of each, you’re likely to observe a significant number of these positions flashing red on any given day. If, instead, you infrequently look at the annual total returns from the whole portfolio, you may very seldom see red at all.
Millar: On a similar topic, the last few years haven’t been a great time to be in cash or looking to purchase an annuity. This is a big problem for a lot of people just now but do you think there is a longer-term implication for Pensions if fund managers are too risk averse? Does the framework need to be reviewed as a short-term view isn’t appropriate? In the last few years we have seen a movement from equities to bonds, catering to a short-term view?
Davies: This shorter view, the myopia, is always a problem when it causes our decision horizon to be divorced from the time horizon of our objectives. And, as mentioned above, this tends to be a ubiquitous feature of human nature. Over the last five years this short-term framework, loss aversion, and the resulting need for immediate emotional comfort, has led many investors to hold too much cash. This is not necessarily irrational – as they get the emotional security they need in return – but it remains a phenomenally expensive way of getting that comfort.
Over the long run, risks are almost always lower than our immediate perceptions lead us to believe. It is vital that we proactively align our decision horizon and framing to our actual time horizon. This is a very misunderstood notion in investing, with people making up all sorts of arbitrary time horizons, when essentially one’s true horizon should simply be driven by the question of when the investor will need the money.
Millar: We are now beginning to emerge from the 2008 crisis, do you think a greater focus on behavioural finance and economics could have influenced some of the decisions in the years leading up to Bear Stearns and Lehman?
Davies: Yes, but it would be an error to think that behavioural finance will ever provide a solution to the economic cycle. Our myopic tendencies and the need for social approbation from the herd are too deeply engrained in human nature to be ‘cured’. However, a thoughtful focus on our behavioural needs can certainly help individuals, companies, or even countries weather the cycle better.
Millar: An important issue over the last eleven months has been the Fed’s taper and the prospective speed of the taper. How do you gauge the market reaction to this topic and have you seen any difference in investor behaviour? The Great Rotation didn’t really take off.
Davies: The general response to tapering – as with all aggregate market reactions – was as much about perception as reality. And all such activity is extremely difficult to predict. These responses seem to go in waves, driven more by media and fluctuations in market-wide levels of anxiety than by any identifiable outcome of the events themselves. As a result, markets fluctuate far more than they should and, unfortunately, predicting these fluctuations can be akin to throwing darts at a dartboard.
Millar: Can’t let this opportunity pass Greg, the Equity Risk Premium is an interest of mine. Could you give an overview from your perspective of how it is impacted by Behavioural Finance?
Davies: The equity risk premium exists simply because humans are generally risk averse – they won’t buy risky assets unless they believe they’ll be adequately compensated for the risk. Of course, from time to time, they get over-exuberant and underestimate the compensation required. However, on average, there is a positive premium to providing your wealth to others to put to productive use in the economy.
The real question is: why is the equity risk premium so large when, over long periods, the actual risks from investing are much smaller than it would seem to imply? The answer again comes down to myopic loss aversion. The risk premium is driven by the level of risk investors are prepared to take in the present. Since myopic loss aversion leads them to generally over-estimate this risk, they under-invest. As a result average prices get driven down and expected returns (risk premia) are pushed higher.
Millar: Thanks Greg I’d like to finish with a couple of more personal questions, hope you don’t mind answering.
How has your investment philosophy altered over the past few years?
Davies: I really try to practice what I preach and apply the things I say publically to my own portfolio – ‘skin in the game’ helps you to learn faster. Many of the basic principles have been in place in my investment philosophy over many years. However, over that time, I’ve become more aware of how complex the overall system of one’s finances really is. I have been gradually broadening my own decision frames to incorporate not just investment decisions, but the interconnectedness to financial planning, my goals and aspirations, expectations, future income and expenditure, insurance, savings and budgeting decisions. These are all interconnected and we have been refining our investment philosophy to take account of these other financial aspects of holistic wealth management. We have also been gradually refining and writing down our philosophy in order to better challenge and test it.
Millar: Most of our readers will not know this but you are also the co-creator of “Open Outcry”, a behavioural finance opera. How did you find the experience of bringing that project to life?
Davies: It was wonderful – and certainly not something I thought I’d ever have the opportunity to do as a banker! I’m a keen amateur singer, and passionate about experimental music. In the same way as Behavioural Finance allowed me to combine many academic interests, Open Outcry provided a wonderful opportunity for me to do something artistically creative with behavioural finance.
For the ‘opera’, Alexis Kirke and I created an artificial, but functioning, market, where a computer model drove the prices of three assets that were available to be traded. The catch was that the participants – who were paid according to the value of their portfolio at the end – could only trade by singing specific phrases to buy and sell each asset. The phrases were design to be harmonious when markets were rising and more traders were singing ‘buy’ phrases; and discordant when ‘sell’ phrases predominated. The singers were trading for real money, as if on an open-outcry trading floor, except their financial interactions resulted spontaneously in a musical experience for the audience. We premiered this work at Mansion House – the residence of the Lord Mayor of London – at an event sponsored by Barclays in late 2012 …for me, a terrifying, but exhilarating experience.
Millar: To finish Greg, is there one piece of advice you would like to leave active traders with?
Davies: Do less than you are inclined to.
Millar: Thank you very much Greg, I greatly appreciate it.
About Greg B Davies: Greg B Davies is a Managing Director with Barclays and has been Head of Behavioural and Quantitative Investment Philosophy since 2006. He is also an Associate Fellow at Saïd Business School, Oxford University.
Greg studied at the University of Cape Town and obtained a degree in Economics, Philosophy and Finance. He followed this with an MPhil in Economics and a PhD in Decision Theory and Behavioural Finance from the University of Cambridge.
Starting in Strategic Consulting, Greg’s career progressed to working in Decision Technology and lecturing at LSE, before becoming an Honorary Fellow – Psychology at University College London and joining Barclays.
On joining Barclays, Greg built the first commercial behavioural finance banking team in the world. His team works on portfolio optimisation and risk measurement by developing and applying behavioural insights and quantitative techniques. In August 2013, he became an Editorial Board Member of the Journal of Behavioural and Experimental Finance.
Follow Greg B Davies on Twitter: @GregBDavies