It was Benjamin Franklin who suggested that one’s happiness depends more on your inward disposition of mind than outward circumstances. The term “disposition” itself can be used to describe someone’s inherent qualities of mind (a tendency to have a pleasant outlook) as well as the way something is arranged in relation to other things (relative to a point of reference) that creates perspective.
Both descriptions help us to understand one of the most widely documented behavioural biases, the disposition effect (DE). The term was coined by economist Hersh Shefrin, and behavioural economist Meir Statman in 1985. The DE refers to the general inclination to sell off winning assets too hastily and hold onto losing ones for too long. This is rooted in two distinct states of mind, namely:
i. The aversion to losses. Trading turns a paper loss into a real one and loss aversion will mean the trader becomes more reluctant to realise this loss. The result is holding onto “losers” for too long in the hope they turn into winners.
ii. The aversion to regrets. Trading turns a paper gain into a real one that makes the trader feel good. Sometimes waiting, however, means a winning position turns into a losing one. This encourages a tendency to sell winners too quickly as the trader fears the regret of the winning position reversing.
The second part of the original DE definition provided regarding the “reference point” is also significant. A few years earlier in 1979, Daniel Kahneman and Amos Tversky published a seminal paper (that would later earn Kahneman the coveted Nobel Prize) on what they called “Prospect Theory”.
They transformed the prevailing model of preferences, being Expected Utility Theory (EUT), from a utility function to a psychological value function. Said differently they asserted that a more descriptive model of preferences requires knowledge of whether the chooser is evaluating a prospect from the perspective of winning or losing. This asymmetry is captured in the DE.
A more descriptive model of preferences requires knowledge of whether the chooser is evaluating a prospect from the perspective of winning or losing.
Momentum Investments set out to study the extent of the DE in a South African context earlier in 2022. When analysing the behaviour of nearly 10 000 execution-only traders (self-advised) on the Momentum Securities platform trading slightly more than R9-billion from 2016 until 2021, a clear DE effect is evident. Traders were 1.39x more likely to sell stocks in a gain position over a loss position during this time period which fits snugly with results in literature that vary between 1.3x and 1.5x in general. Where this gets particularly interesting, however, is when the periods are dissected into “pre-Covid” and “Covid” and further into age groupings. The highlights of these findings are:
i. During the Covid pandemic (2020 and 2021) in South Africa the DE increases rapidly to 1.9x. Said differently, traders become 1.9x more likely to trade winning positions over losing positions.
ii. Gen Xers (ages 39 to 53) during 2020 were 3.92x (392%) more likely to trade winning positions over losing ones. They account for a healthy quarter of the total population (comfortably the largest segment) and maintain the highest DE over the total period of 1.92x.
During the same period of analysis (2020 and 2021) another two large groups of investors were studied to quantify their risk behaviour. Just as selling (trading) a stock reflects a change in preferences occasionally stemming from a cognitive or emotional bias (such as loss aversion), an investment switch from one fund to another represents a similar change in preferences often from the same biases. The investment switch behaviour of over 20 000 clients in discretionary unit trusts and living annuities were analysed with the following highlights focusing on their level of engagement:
i. The number of “active” investors or those performing at least one investment switch increases by over 80% during the pandemic.
ii. The number of switches at a platform level exceeded record levels during 2020 and 2021 (increasing by over 50%).
The difference in investment behaviour over the same period that likely stems from the same bias (loss aversion sitting at the core of the DE) when the two groups are juxtaposed may cast a shadow on investment advice. The nearly 10 000 unadvised traders that clearly suffered from a high disposition effect during the Covid pandemic were significantly less likely to trade and realise these losses.
Yet, more than 20 000 advised clients in the discretionary unit trusts and living annuities were transacting at record levels incurring a behaviour tax (lower investment return) of over 6% in 2020 and nearly 4% in 2021 on average.
Loss aversion is clearly playing out different and opposing ways with the notable difference being investment advice. While this is cause for further investigation, it is very clear that the disposition of the advisor has a major role to play in creating (or destroying) investment value for clients.