Neuroeconomics: An Edge When Everyone has Access to the Same Data?
Neuroeconomics: An Edge When Everyone has Access to the Same Data?
The “golden generation” at Euro 2004 – England versus Portugal in the Quarter Finals. England went on to lose the match in penalties (6-5).
We’re all looking for our own edge in markets. A friend of mine who is currently doing a PhD in biomedical sciences solely invests in niche medical stocks based on his academic knowledge. He believes this gives him an edge over finance professionals, who may in truth, have no real idea about the science.
As somebody studying Psychology, I’m increasingly interested in understanding whether neuroeconomics could be an edge.
Neuroeconomics – which is effectively behavioural economics with a sprinkling of neuroscience (what’s going on in the brain) – aims to explain why traditional assumptions of rationality in economics often do not happen in real life.
It is about understanding that human rationality is far from objective rationality, as we are influenced by mental tendencies and heuristics that push us off the path of being perfectly rational.
We may perceive our thinking as objectively rational, but the clue is in the sentence, “perceive”. What we perceive is often far from objective rationality because of the funny ways our brains work and are influenced.
Below are some fun examples I’ve encountered of human irrationality applied to the stock market.
See ‘takeaway’ section for more serious discussion about applying a neuroeconomics lens to commodity trading…
Integral vs incidental emotions:
Integral emotion = emotion in response to current situation.
Incidental emotion = emotion related to external situation.
Incidental emotions should not affect current performance as they are not related but they do - emotional overlap.
E.g. sun’s out, I’m happy = incidental, may change my integral emotions to a situation coming up.
Hirshleifer and Shumway (2003)
- Amount of sunshine (relative to amount of sunshine expected for time of year) positively correlated with market returns.
Methodology:
- The researchers analyzed 26 international stock exchanges over many years.
- They compared daily stock returns with weather data from the city where the country’s main stock exchange is located.
- Importantly, they didn’t just look at “sunny days” - they measured how sunny it was compared to the average or expected amount of sunshine for that time of year (to control for seasonal effects).
Interpretation:
- Authors argue that sunlight affects mood, and mood affects risk-taking and optimism.
- On sunnier days, investors feel more positive and may buy more aggressively, pushing up prices.
- Investor sentiment, influenced by something so trivial as weather, could it really systematically move asset prices? Weather shouldn’t affect stock valuation fundamentals…
Edmans, García, and Norli (2007)
- Statistically significant dips in market returns when national sports teams perform badly.
Methodology:
- International stock market data alongside major sporting events – in particular football (e.g. the world cup) – they then measured market returns the next day after wins or losses.
Findings in more detail:
- After a loss, the national market’s return drops significantly (about -0.5% on average the following day).
- Effect is strongest in countries where football is culturally important (e.g. Brazil, England).
- Wins don’t lead to statistically significant positive effects – pattern consistent with loss aversion, Daniel Kahneman – where “loss is felt more intensely than the pleasure of an equal gain”.
Interpretation:
- National sports results influence aggregate investor mood.
- A loss makes people feel collectively worse – leading to lower risk appetite and temporary pessimism in the market.
It’s highly difficult to prove causality from these studies, but since statistically significant correlations are there… would this be a viable way of taking neuroeconomic based punts on the stock market?
In particular, algorithms are (in essence) built around correlations in products for different periods of time. They parse variables and take lightning quick action on them if they meet certain criteria.
Could one build an algorithm using neuroeconomic research or principals? What markets could this work in?
I appreciate that the studies discussed were a bit of light-hearted fun, but they do introduce the idea that people easily deviate from “rationality” in decision making.
These studies suggest that even in a world dominated by data, humans, as long as they have an input, still leak emotion into markets.
Takeaway
Maybe the smartest approach isn’t to disregard irrationality, but to understand and map it on top of the usual fundamentals - because in a world where data becomes more transparent, understanding irrationality might remain an edge.
The growing movement from over the counter (OTC) to exchange traded products means there will be greater transparency for everyone.
Probably more data-driven trading and efficient markets as a consequence.
Less so hypotheses built from psycho-analysis of brokers and order flows, pattern recognition, hunches/gut-feelings/experience, supply-chain edge e.t.c.
However, I still believe this transparent data can be analysed with behavioural considerations because it is about understanding how everyone else is going to interpret and act on the same information.
- What are the competitors’ “perceived rationality”? / What do they consider as the perfectly rational response?
- How and why does this differ from “objective rationality”?
- What do competitors think about how other competitors will act with the same information?
- How do all of these discrepancies effect the market and provide opportunities?
In this sense, I see neuroeconomics as an important nuance that should be added on top of fundamental economics.
Do you agree?
This even applies to algorithms or big data analyses, as they are coded and refined by humans. And wherever humans have a say, you can bet they will still express tendencies in their outputs.
I appreciate the opportunities to do this in reality would be difficult and specific, requiring a lot of thinking, but that’s all part of the challenge.
Ultimately this isn't about trading off sunshine or football scores.
It's about mapping how human brains - analysts, risk managers, traders, algo designers - systematically misread or overweight information, even in efficient markets.
I will be writing about tangible examples of where you could exploit mapping out behavioural tendencies in commodity paper markets. Stay tuned to the blog…
Data is becoming more transparent – but interpretation is not.
------------------------------------------------------------------------------------------------------------------------------------------------------------------
You had me until you said you were a Chelsea fan... Anyways it's good to hear your take on how to navigate through this new age of information and the key to interpreting data.
ReplyDelete