This article introduces some thoughts and ideas about human decision-making and behaviour risk in the financial markets. The article also asks questions of the financial risk industry and looks at whether it needs to put into place changes to take greater account of the human behaviour aspect of risk. Finally it looks at whether businesses can introduce improvements and enhancements to the way they work in order to prevent and mitigate risk, and to improve the quality of decision-making in the financial markets.
In John Maynard Keynes’s celebrated 1936 book, ‘The General Theory of Employment, Interest and Money’, he used the term “Animal spirits” to describe emotions which influence human behaviour. Now, some eight decades later, research from neuroscience is shedding light on these ‘animal spirits’, and in particular, how they affect people’s decisions in the financial markets. Some of these findings are leading to questions about some of the basic assumptions of how people think and act, and are also challenging long-held beliefs and tenets central to economic theory. Whilst this has some direct consequences for the field of financial risk-management, it also provides new thinking and offers potential solutions, some of which may help to improve risk-management practices and techniques moving forward.
The traditional view from classical economics sees people as rational, utility-maximizing actors; individuals who know what they want and are consistent, methodical, and emotionless in pursuing it. Consistent with this is the view of the human mind as a machine; working like a computer and rationalizing all options through the use of people’s cognitive powers, and the supremacy of intellect. Whilst these beliefs are cornerstones of modern economics, they are increasingly being challenged by the emerging fields of ‘Behavioural Economics’ and ‘Neuroeconomics’: Backed by a growing body of research, supporters of these movements argue that: Humans have many limitations to behaving rationally and that the use feelings and emotions (Keynes’ animal spirits) extensively when making decisions. As such these behaviourists call in to question many of the long-held theories and assumptions of classical economics.
An excellent example of the limitations of human rationality is provided by some research carried out in 2010 jointly by Colombia Business School and Ben Gurion University*. The study looked into the decision-making performance of a group of highly experienced judges at parole hearings in Israel involving over 1,100 cases during a 10-month period. The findings revealed some interesting and surprising outcomes. One would surely expect the judge’s rulings to be based on ‘rational-decisions’ undertaken within the guidelines established by written laws. Actually the biggest influence in the outcomes was the time of day each hearing occurred. Prisoners who appeared before the judges early in the morning session, or during the time shortly after the mid-morning break, or immediately after the lunch break, received the most favourable outcomes, receiving parole in about 60-70 percent of the cases. Whereas, if you were a prisoner up for parole late in each session, then you were in trouble: Prisoners who appeared towards the end of each session received parole no more than 10-15 percent of the time. The research found nothing malicious or unusual about the judges’ behaviour; rather they concluded the reason was ‘Decision-fatigue’.
‘Decision-fatigue’ refers to the deteriorating quality of decisions made by people, after a long session of decision making. ‘Decision-fatigue’ occurs as more choices are made throughout a particular period of time, and throughout the day. Particularly where decisions are complex or have more far-reaching consequences. This is because each subsequent decision becomes marginally harder as the brain draws on people’s energy reserves, principally in the form of glucose. The more decisions people make, the more their glucose reserves become depleted; as glucose reserves deplete so the brain modifies the way it functions to reserve vital energy resources. In these situations people do not ‘stop-thinking’, they just think less effectively: Modified thinking behaviours may include taking cognitive short-cuts, such as relying on simple rules-of-thumb, or perhaps avoiding making decisions altogether: Many people will have experienced ‘analysis-paralysis’!
Returning to the aforementioned study, no matter how rational and high-minded the judges aimed to be, they were fighting their own human biology: The depletion of glucose to the judge’s brains changed the way their thinking processes worked. In most cases, ‘the die was cast’, and prisoners who may have received parole had their case been heard at 9am in the morning, were more often than not sent back to prison when their case was heard around 3pm.
Whilst this study highlights the limitations to human cognitive abilities, other research highlights the extent to which our emotions play a role in decision-making processes, thus further arguing against the ‘rational man’ theory of classical economics: Renowned neuroscientist Antonia Damasio has done a lot of work to highlight the primacy of emotions as a key part of decision-making. One well-known study by Damasio showed how people who had received brain injuries in which they lost of ability to feel emotions, were incapable of making even the most basic of decisions; often spending hours deliberating over irrelevant details, such as where to eat lunch. Further research into the effect of human emotions in decision-making, has turned on its head the belief that the human mind uses purely cognitive processes to reach logical conclusions. Some people now argue that beliefs and existing theories of rational decision-making are being seen as increasingly implausible.
Coming back to financial markets, what are implications from these alternative beliefs for the field of financial risk-management? And what can the industry do to improve the way it manages financial risk?
Much of the focus of risk management in the financial markets is on quantifying and measuring financial risk. A whole architecture of financial models, process and practices has arisen around this: The emergence of the Behavioural Finance movement does however question whether the financial risk management industry is on the right track. – The first question is whether the basic underlying assumptions that underpin some of these models are correct? The concept of ‘rational man’ largely underscores the long-standing view that markets are completely random, and that deviation from true value in liquid markets will be arbitraged away by ‘rational man’: Markets are however human constructs, driven by human perceptions, reactions and decisions, triggered by people’s emotions. Keynes understood the way markets worked from a behavioural perspective: In what was called the ‘Keynesian beauty contest’, he said, ‘you win not by picking the soundest investment, but by picking the investment that others, who are playing the same game, will soon bid up higher’.
It may be a stretch to say that because people act and behave emotionally rather than rationally, that therefore markets are not truly random: However, it is this emotional human behaviour which leads to trends, manias, panics and long-term distortion from value, which are NOT quickly arbitraged away by the mythical ‘rational man’. - If markets are not truly-random, then this calls into question many of the risk-management models which themselves are based off this assumption, a situation which is further compounded by over-reliance on these models. Yet even if these models are correct or simply if they are the best tool for measurement available, the issue is that they are merely tools, they do not have any predictive capabilities, they are measurements of the amount of risk being taken, or the type of risk, but there is not much focus on the quality of the risk-taking or of the individual risk-taker. The financial markets are obsessed with measuring and quantifying risk, yet this has merely enabled a situation which has led to greater amounts of risk, and ironically larger and more damaging risk events.
Anurag Vaish of the 'Final Mile' consultancy, which specialises in finding risk solutions through neuroscience and behavioural economics, sums it up well: ‘Risk is a feeling not a number; financial Institutions are highly number driven and continue to represent risk more as numbers’. The work of the ‘Final Mile’ consultancy has yielded some excellent results in risk situations across a range of industries, helping to find unorthodox solutions to conventional problems. A good illustration of their work involved an experiment on a one mile stretch of the Mumbai Rail system which was notorious for deaths from people crossing rail tracks. After researching the problem, Final Mile came up with some innovative recommendations which took account of human decision-making and behaviour. When these recommendations were implemented, deaths from rail-track crossings on the stretch of line dropped from 23 in the previous six months, to just one in the next 8 months.
This article has shown how, when it comes to making-decisions, human behaviour is not necessarily in accordance with the common-held belief of humans as rational beings; we are not as in charge of our choices as we like to think we are: Our ability to act rationally and to effect well considered decisions is limited, and our emotions, such as fear and desire, as well as mood, have a far greater effect on our behaviour than we realise. No one is immune to this; even highly intelligent people and experts make poor choices. This last point should provide some extra context to some of the high profile examples of major financial losses within trading and investment businesses in recent years; e.g. JP Morgan, UBS, Societe Generale, Amaranth, etc. In addition there have been numerous smaller, but still extremely costly, financial losses which have impacted many firms, but which will have not made the front pages or breaking news stories.
What can trading businesses do to rise to the challenge? Working on improving the monitoring of and quality of decision-making, is not merely a matter of risk-control and risk-mitigation, it is also a pro-active endeavour which can yield businesses a greater return on investment. Steps could be taken to deliver improved robustness and quality in individual, managerial and group decision-making; practices in other industries could provide some potential solutions. For example, simple checklist practices have been put into place in industries as diverse as medicine and aviation, with profound effects on safety and quality. One piece of research showed that the introduction of check-lists on post-operative morbidity and mortality rates in one hospital saw a 57% decrease in complications.** Banks could include engaging trading managers, risk managers and supervisors through ‘what-if-scenario’ exercises.
This is practiced extensively in the disaster-recovery industry and in the military. Many industries providing education, information sessions and coaching to participant’s engaged in dangerous and stress inducing industries: As part of its highly successful ‘Lamplighter’ programme, which aimed to improve employee wellness, Unilever offered a range of stress reduction programmes to employees facing high levels of stress, these included coaching and cognitive behaviour therapy to improve mental resilience. Whilst this was just one of a number of initiatives employed by Unilever, the programme saw an overall 40% increase in mental resilience, and the overall programme saw a large increase in productivity which equated to a 350% return on investment***. What about limitations to time at the screens? Many fields limit the time people can spend in high stress or highly engaged activities where human life is at risk: Aviation, transportation, surgery, they all place limitations on the amount of time practising in a day or a week. There is a valid reason for this; surely this can be translated to situations where people have huge sums of money and risk at their finger-tips.
Banks can also improve their monitoring of trading behaviours. Whilst risk management’s emphasis has largely been on quantifying the size of risks, they could easily put in steps to closely monitor trading behaviour: MI systems could be adapted to identify potential behavioural problems early, including spotting repeated patterns of sub-optimal behaviour, or signs of distress in trading P&L volatility that is outside of market norms. This can be taken further; businesses could look at their organisation or operational process to recognize organisation deficiencies in reporting lines, or to highlight limits of responsibility. Whilst, risk management could take a number of steps, it has to be cautious not to overly burden traders or be too interfering with the trading process at desk level. However, traders have a dual functionality, they are both risk seekers and risk managers, this in itself presents the first challenge: desk-level is the front line of risk-management.
In the wake of the ‘Global Financial Crisis’, and subsequent strong political, regulatory and economic forces re-shaping the financial markets, the financial risk management industry is facing many challenges. It is unfair to apportion blame to the risk-management industry for the financial disasters of recent years; however it is right to question some of its assumptions and practices, and to find out whether things could have been done better and differently. As part of this process, it may help to step away from some of the beliefs of the past, and to see if new innovative solutions could be found and applied to take the industry forward.
In John Maynard Keynes’s celebrated 1936 book, ‘The General Theory of Employment, Interest and Money’, he used the term “Animal spirits” to describe emotions which influence human behaviour. Now, some eight decades later, research from neuroscience is shedding light on these ‘animal spirits’, and in particular, how they affect people’s decisions in the financial markets. Some of these findings are leading to questions about some of the basic assumptions of how people think and act, and are also challenging long-held beliefs and tenets central to economic theory. Whilst this has some direct consequences for the field of financial risk-management, it also provides new thinking and offers potential solutions, some of which may help to improve risk-management practices and techniques moving forward.
The traditional view from classical economics sees people as rational, utility-maximizing actors; individuals who know what they want and are consistent, methodical, and emotionless in pursuing it. Consistent with this is the view of the human mind as a machine; working like a computer and rationalizing all options through the use of people’s cognitive powers, and the supremacy of intellect. Whilst these beliefs are cornerstones of modern economics, they are increasingly being challenged by the emerging fields of ‘Behavioural Economics’ and ‘Neuroeconomics’: Backed by a growing body of research, supporters of these movements argue that: Humans have many limitations to behaving rationally and that the use feelings and emotions (Keynes’ animal spirits) extensively when making decisions. As such these behaviourists call in to question many of the long-held theories and assumptions of classical economics.
An excellent example of the limitations of human rationality is provided by some research carried out in 2010 jointly by Colombia Business School and Ben Gurion University*. The study looked into the decision-making performance of a group of highly experienced judges at parole hearings in Israel involving over 1,100 cases during a 10-month period. The findings revealed some interesting and surprising outcomes. One would surely expect the judge’s rulings to be based on ‘rational-decisions’ undertaken within the guidelines established by written laws. Actually the biggest influence in the outcomes was the time of day each hearing occurred. Prisoners who appeared before the judges early in the morning session, or during the time shortly after the mid-morning break, or immediately after the lunch break, received the most favourable outcomes, receiving parole in about 60-70 percent of the cases. Whereas, if you were a prisoner up for parole late in each session, then you were in trouble: Prisoners who appeared towards the end of each session received parole no more than 10-15 percent of the time. The research found nothing malicious or unusual about the judges’ behaviour; rather they concluded the reason was ‘Decision-fatigue’.
‘Decision-fatigue’ refers to the deteriorating quality of decisions made by people, after a long session of decision making. ‘Decision-fatigue’ occurs as more choices are made throughout a particular period of time, and throughout the day. Particularly where decisions are complex or have more far-reaching consequences. This is because each subsequent decision becomes marginally harder as the brain draws on people’s energy reserves, principally in the form of glucose. The more decisions people make, the more their glucose reserves become depleted; as glucose reserves deplete so the brain modifies the way it functions to reserve vital energy resources. In these situations people do not ‘stop-thinking’, they just think less effectively: Modified thinking behaviours may include taking cognitive short-cuts, such as relying on simple rules-of-thumb, or perhaps avoiding making decisions altogether: Many people will have experienced ‘analysis-paralysis’!
Returning to the aforementioned study, no matter how rational and high-minded the judges aimed to be, they were fighting their own human biology: The depletion of glucose to the judge’s brains changed the way their thinking processes worked. In most cases, ‘the die was cast’, and prisoners who may have received parole had their case been heard at 9am in the morning, were more often than not sent back to prison when their case was heard around 3pm.
Whilst this study highlights the limitations to human cognitive abilities, other research highlights the extent to which our emotions play a role in decision-making processes, thus further arguing against the ‘rational man’ theory of classical economics: Renowned neuroscientist Antonia Damasio has done a lot of work to highlight the primacy of emotions as a key part of decision-making. One well-known study by Damasio showed how people who had received brain injuries in which they lost of ability to feel emotions, were incapable of making even the most basic of decisions; often spending hours deliberating over irrelevant details, such as where to eat lunch. Further research into the effect of human emotions in decision-making, has turned on its head the belief that the human mind uses purely cognitive processes to reach logical conclusions. Some people now argue that beliefs and existing theories of rational decision-making are being seen as increasingly implausible.
Coming back to financial markets, what are implications from these alternative beliefs for the field of financial risk-management? And what can the industry do to improve the way it manages financial risk?
Much of the focus of risk management in the financial markets is on quantifying and measuring financial risk. A whole architecture of financial models, process and practices has arisen around this: The emergence of the Behavioural Finance movement does however question whether the financial risk management industry is on the right track. – The first question is whether the basic underlying assumptions that underpin some of these models are correct? The concept of ‘rational man’ largely underscores the long-standing view that markets are completely random, and that deviation from true value in liquid markets will be arbitraged away by ‘rational man’: Markets are however human constructs, driven by human perceptions, reactions and decisions, triggered by people’s emotions. Keynes understood the way markets worked from a behavioural perspective: In what was called the ‘Keynesian beauty contest’, he said, ‘you win not by picking the soundest investment, but by picking the investment that others, who are playing the same game, will soon bid up higher’.
It may be a stretch to say that because people act and behave emotionally rather than rationally, that therefore markets are not truly random: However, it is this emotional human behaviour which leads to trends, manias, panics and long-term distortion from value, which are NOT quickly arbitraged away by the mythical ‘rational man’. - If markets are not truly-random, then this calls into question many of the risk-management models which themselves are based off this assumption, a situation which is further compounded by over-reliance on these models. Yet even if these models are correct or simply if they are the best tool for measurement available, the issue is that they are merely tools, they do not have any predictive capabilities, they are measurements of the amount of risk being taken, or the type of risk, but there is not much focus on the quality of the risk-taking or of the individual risk-taker. The financial markets are obsessed with measuring and quantifying risk, yet this has merely enabled a situation which has led to greater amounts of risk, and ironically larger and more damaging risk events.
Anurag Vaish of the 'Final Mile' consultancy, which specialises in finding risk solutions through neuroscience and behavioural economics, sums it up well: ‘Risk is a feeling not a number; financial Institutions are highly number driven and continue to represent risk more as numbers’. The work of the ‘Final Mile’ consultancy has yielded some excellent results in risk situations across a range of industries, helping to find unorthodox solutions to conventional problems. A good illustration of their work involved an experiment on a one mile stretch of the Mumbai Rail system which was notorious for deaths from people crossing rail tracks. After researching the problem, Final Mile came up with some innovative recommendations which took account of human decision-making and behaviour. When these recommendations were implemented, deaths from rail-track crossings on the stretch of line dropped from 23 in the previous six months, to just one in the next 8 months.
This article has shown how, when it comes to making-decisions, human behaviour is not necessarily in accordance with the common-held belief of humans as rational beings; we are not as in charge of our choices as we like to think we are: Our ability to act rationally and to effect well considered decisions is limited, and our emotions, such as fear and desire, as well as mood, have a far greater effect on our behaviour than we realise. No one is immune to this; even highly intelligent people and experts make poor choices. This last point should provide some extra context to some of the high profile examples of major financial losses within trading and investment businesses in recent years; e.g. JP Morgan, UBS, Societe Generale, Amaranth, etc. In addition there have been numerous smaller, but still extremely costly, financial losses which have impacted many firms, but which will have not made the front pages or breaking news stories.
What can trading businesses do to rise to the challenge? Working on improving the monitoring of and quality of decision-making, is not merely a matter of risk-control and risk-mitigation, it is also a pro-active endeavour which can yield businesses a greater return on investment. Steps could be taken to deliver improved robustness and quality in individual, managerial and group decision-making; practices in other industries could provide some potential solutions. For example, simple checklist practices have been put into place in industries as diverse as medicine and aviation, with profound effects on safety and quality. One piece of research showed that the introduction of check-lists on post-operative morbidity and mortality rates in one hospital saw a 57% decrease in complications.** Banks could include engaging trading managers, risk managers and supervisors through ‘what-if-scenario’ exercises.
This is practiced extensively in the disaster-recovery industry and in the military. Many industries providing education, information sessions and coaching to participant’s engaged in dangerous and stress inducing industries: As part of its highly successful ‘Lamplighter’ programme, which aimed to improve employee wellness, Unilever offered a range of stress reduction programmes to employees facing high levels of stress, these included coaching and cognitive behaviour therapy to improve mental resilience. Whilst this was just one of a number of initiatives employed by Unilever, the programme saw an overall 40% increase in mental resilience, and the overall programme saw a large increase in productivity which equated to a 350% return on investment***. What about limitations to time at the screens? Many fields limit the time people can spend in high stress or highly engaged activities where human life is at risk: Aviation, transportation, surgery, they all place limitations on the amount of time practising in a day or a week. There is a valid reason for this; surely this can be translated to situations where people have huge sums of money and risk at their finger-tips.
Banks can also improve their monitoring of trading behaviours. Whilst risk management’s emphasis has largely been on quantifying the size of risks, they could easily put in steps to closely monitor trading behaviour: MI systems could be adapted to identify potential behavioural problems early, including spotting repeated patterns of sub-optimal behaviour, or signs of distress in trading P&L volatility that is outside of market norms. This can be taken further; businesses could look at their organisation or operational process to recognize organisation deficiencies in reporting lines, or to highlight limits of responsibility. Whilst, risk management could take a number of steps, it has to be cautious not to overly burden traders or be too interfering with the trading process at desk level. However, traders have a dual functionality, they are both risk seekers and risk managers, this in itself presents the first challenge: desk-level is the front line of risk-management.
In the wake of the ‘Global Financial Crisis’, and subsequent strong political, regulatory and economic forces re-shaping the financial markets, the financial risk management industry is facing many challenges. It is unfair to apportion blame to the risk-management industry for the financial disasters of recent years; however it is right to question some of its assumptions and practices, and to find out whether things could have been done better and differently. As part of this process, it may help to step away from some of the beliefs of the past, and to see if new innovative solutions could be found and applied to take the industry forward.