The rationality of human investors has long been questioned. Behavioural biases have been broadly investigated and are known to impair human decision-making in capital markets. Our empirical study draws on these findings regarding behavioural biases and analyses the behaviour of traders on a social trading platform to expand insights on behavioural biases and human decision-making. In particular, we have analysed the influence of real-time monitoring and real-time interaction between traders and investors on traders’ behaviour.
What most investors have in common is that they are biased when it comes to investment decisions. Once a promising investment opportunity is identified and the order is sent to be executed by the broker, the most relevant decision is still to be made: what is the optimal point in time to sell? No matter if one intends to hold the asset for a long time, e.g. until retirement, or one intends to sell within the same day: the decision has to be made. Timing to find an optimal closing point is crucial as it is known to have a large influence on investment performance.
If someone can recall their first investment they might remember the mixed bag of feelings that can arise when observing the price chart of an asset they have just bought. As soon as the position shows a loss, they feverishly wait for prices to rise again. On the other hand, once a position has accumulated a substantial profit, the urge arises to close the position and realise the current profit.
These emotions lead to distortions in decision-making. Human investors tend to realise returns of their winning positions too early and let unfavourable positions accumulate losses for too long. This behaviour is known as the “disposition effect“. The disposition effect is a well-studied and long known behavioural bias affecting investors of differing levels of experience.
Delegating these decisions to a professional portfolio manager is an easy way to avoid having to personally make these decision. However, traditional active portfolio management has been shown to provide rather poor performance in comparison to a valid benchmark. These results have been replicated over time in different asset and fund classes. In addition, financial institutions in general have lost their glamorous image and reputation over the past years, especially since the last global financial crisis. Studies observed that retail investors have lost their trust in financial institutions.
However, technological progress has paved the way not only for social media and social networks but also for innovative financial services. A relatively new phenomenon gaining increasing interest is Social Trading. Social Trading platforms combine the trading functionalities of classical online broker platforms with the communication and interaction features of social networks. Next to following other users’ profile updates, a main characteristic of social trading platforms is the possibility to follow other users by automatically copying their trades or their whole portfolios.
Users (followers) can follow traders on a platform and their trades are replicated in their own online broker accounts. This set of features results in a setup that resembles traditional portfolio management, where investors delegate investment decisions to portfolio managers. Furthermore, the trader’s decisions become fully transparent in real-time and in case a follower disagrees with the trader’s decisions or performance, the user can simply “unfollow” him. The user can also provide feedback in form of star ratings for the trader and direct messages posted on the trader’s profile page – right next to the list of a trader’s performance, risk metrics and executed trades.
Traders are incentivised not only by digital fame but also by monetary compensation based on the volume of trades that are replicated in a trader’s followers’ accounts.
Followers can follow every step a trader makes in real-time and withdraw their money immediately if they want. Given this setup, we became curious if this tight monitoring and real-time sanctioning is influencing the behaviour of traders. Do they behave differently if they can reasonably expect to be closely monitored by investors? Especially investor who invested money only recently are very likely to observe the trader’s performance on subsequent days.
To address this question we obtained profile and trade data from the second largest social trading platform (www.zulutrade.com). Based on a set of 178 traders and more than 400,000 trades, we conducted a panel regression to determine the influence of changes in capital invested into a trader.
We find that on social trading platforms the traders’ sensitivity to the disposition effect is influenced by the amount of attention they receive from their followers, i.e. users who invested capital into the traders’ strategy. A trader receiving capital is increasing his win-ratio on days (number of trades with positive return) right after the invested amount of money increases. It is likely that traders are trying to please investors by closing trades that provide a positive return but might not be at the optimal closing point.
Additionally, we were able to show that the win-ratio is subject to the disposition effect. Traders’ win-ratio is negatively influenced if trades closed on that day were trades that were open for a long time. Put differently, traders let positions accumulate losses for too long. Hence, we conclude that traders’ exposition to the disposition effect increases with increases in capital invested on the previous day.
These novel insights propose a link between monitoring capability induced transparency and the disposition effect. We extend the literature on trader-investor interaction channels, especially in the context of social trading networks. The results obtained are of high relevance for regulators who have a strong focus on customer protection and financial services regulation. We show that close to real-time monitoring can actually harm the investors, because it increases the traders’ disposition effect. They also provide guidelines for platform designers, traders, investors and social trading platform operators.
Financial services are in the focus of becoming automated by algorithms or replaced by online peer-to-peer interaction. Fintechs are searching for financial services to be executed with better user experience and higher efficiency, often by relying on algorithmic automation and digital platforms. However, these are still rather new and unknown environments for financial services. As we have shown: new technological possibilities and innovative setups of online platforms might not only solve or alleviate current issues. They might facilitate undesired behaviour and mechanisms that affect our human decision-making in ways that are yet to be discovered.
- This blog post is based on the authors’ paper Effects of transparency: analyzing social biases on trader performance in social trading, in the Journal of Information Technology, 2016, pp 1-12.
- The post gives the views of its author, not the position of LSE Business Review, the London School of Economics and Political Science.
- Featured image credit: Trading stocks on a computer, by OTA Photos, under a CC-BY-SA-2.0 licence
- Before commenting, please read our Comment Policy.
Florian Glaser is a PhD candidate in Information Systems at the Goethe University, Frankfurt. His research interests are in the areas of Blockchain systems, trading networks and behavioural finance. His articles have been published in several conference proceedings and journals such as Journal of Information Technology (JIT), the European Conference on Information Systems (ECIS), the Hawaii International Conference on System Sciences (HICSS).
Marten Risius is Postdoctoral Researcher at the University of Mannheim and the Mannheim Business School. He holds a Ph.D. in Economics and Business Administration from Goethe University Frankfurt and a M.Sc. in Psychology from the University of Osnabrueck. His research interests are in the areas of data science, digitalization, and social media management. His articles have been published in several conference proceedings and journals such as Journal of Strategic Information Systems (JSIS), Journal of Information Technology (JIT), Information & Management (I&M), the European Conference on Information Systems (ECIS) and the International Conference on System Sciences (ICIS).