Risk Intelligence: making decisions under uncertainty

May 10, 2016

Share this article:

What is the probability a terroristic attack will occur tomorrow? What is the probability a company goes bankrupt? And what is the probability Mayweather wins his next game? This article is based on the novel ‘Risk Intelligence: How to Live with Uncertainty’ written by Dylan Evens. By taking the reader through different situations Evens explains the importance of a relatively unknown concept, namely Risk Intelligence. The impact Risk Intelligence has got on people’s daily lives and their businesses indicates more people should be aware of this ability. What is Risk Intelligence? What influences someone’s Risk Intelligence? And frankly even more important: how can someone improve their own Risk Intelligence?

In the seventies Wall Street was flooded with gamblers, bridge and backgammon players. Twenty years later, these risk lovers were overplayed by the risk avoiders, also known as quantitative analysts. Eventually, this turned out to be one of the greatest mistakes that might have contributed to the economic crisis in 2007. Banks no longer would let their choices depend on the intuition and strategic decisions of their employees. Instead, formulas like the Black-Scholes were introduced. This formula made it possible to determine the price of financial instruments that have not been regularly traded. By comparing them with similar securities that did provide data they were able to set a price. Another group of quantitative analysts went even further. Developing the VaR (Value at Risk) enabled a company to summarize the risks of entire portfolios of financial products into this single number. The great thing about the VaR was that the dizzying variety of variables of which the investment portfolio of a market exists, clenched together into a single dollar amount that risk managers could report to directors of companies. The risk-takers initially looked at these new methods with suspicion. Nevertheless, the formulas proved to be correct over and over again and thus businessmen embraced the new way of thinking. Hereby the VaR formed an indispensable part of daily trading in the late nineties. However, slowly it seemed as if all Risk Intelligence disappeared and had to make room for pure mathematical data analysis. The problem with any mathematical technology is that there is a danger that people overlook mistakes that they previously would have seen immediately. The intermediate steps that are normally taken disappear and the order of magnitude at each intermediate step is therefore not viewed anymore. Therefore, remarkably high or low rates will be less noticeable. Hence, since directors were negligent with their computations and forsake the Risk Intelligence that they could use, their failure came as a surprise.
Risk Intelligence (RQ) is the ability to accurately estimate odds. There are more definitions of the concept. For example, Risk Intelligence can also be seen as a feature of a company and not of an individual. For this matter Risk Intelligence will be seen as a cognitive skill of individual minds. Unlike most people expect, there is neither a correlation between a person’s IQ and his RQ, or between his Emotional Quotient (EQ) and his RQ. As well as for someone’s IQ or EQ, the Risk Quotient can also be calculated through a test. The Risk Intelligence Quotient is a number between zero and one hundred. The average score of the RQ-test measured over about fourteen thousand people is 64. The test can be found on www.rq-test.nl. Each participant is presented fifty allegations with a certain probability of correctness that has to be assigned to it. For instance, 0 percent is certainly not true, 100 percent stands for certainly true, 50 percent means the participant has no idea. If someone is fairly sure that the statement is true but not entirely, a probability of either sixty, seventy, eighty or ninety per cent should be chosen, depending on his degree of certainty.

Journalists and managers in business blame both rising and falling prices on the aversion to uncertainty. This is not right. If the markets really would be in uncertainty no increase or decrease in prices would occur. Prices change only if there is an influx of information through which the uncertainty decreases and the balance of probabilities moves in a certain direction.

An objection that critics often have about the test arises from a misunderstanding. Critics find that the test should ask participants whether they are prepared to take certain risks. According to them, only estimating the likelihood of claims is too limited and too far removed from the practical reality. However, that judgement contains a confusion about the difference between Risk Intelligence and risk behaviour. Risk Intelligence is a purely intellectual ability, whereas risk behaviour is an emotional trait and has to do with how comfortable people feel taking risks. Risk behaviour has got nothing to do with right and wrong, it is better viewed as a character trait. It determines how much risk someone is willing to take, while Risk Intelligence determines the awareness of how much risk is actually been taking. Therefore, it is important to be aware of the available knowledge someone has when answering a question. For example, a participant may know little and yet achieve a high test score. This is because the test rewards people who accurately estimate their own acquaintance, and not those who just simply know a lot.
Besides a Risk Quotient, the test configures a calibration curve as well. This calibration curve is not only a visual representation of the overall Risk Intelligence that came out the test. It even gives details about the Risk Intelligence profile (i.e. whether someone shows too little or too much confidence and where the curve can show these abnormalities). The perfect calibration curve is exactly on the identity line. The surface between the calibration curve and the diagonal of the identity line is inversely proportional to someone’s Risk Intelligence. The participants with too much confidence will have a calibration curve that is above the identity line from zero to fifty percent and beneath the identity line from fifty till a hundred percent. On the other hand, people who have not that much confidence gained the opposite effect. Their curve was for the first fifty percent underneath the identity line and from fifty to a hundred percent above the line. As regards people who cherish completion, they will have a tendency to choose often zero or a hundred percent when assigning its correctness. People who are quite uncertain, not that good in making decisions and find completion not something that satisfies them, tend to stick around fifty percent.
Once, a company in the oil industry set its most clever men together in order to estimate the odds certain areas would be great resources to drill for oil. Although all of them were good at their jobs and highly regarded, their estimations turned out to be false most of the time. Therefore, the company decided to start a traineeship were the employees learned about Risk Intelligence. Showing situations the company had dealt with in the past, letting the employees assign a certain chance for oil in those areas and eventually letting the employees know how much oil there actually was, learned them how to deal with these problems. The business men actually trained their Risk Intelligence. Risk Intelligence does not have fully biological instance, by practicing everyone can improve their own Risk Intelligence. Therefore, one should know how and by what his Risk Intelligence can be influenced.
A heuristic is a rule of thumb, a cognitive trick, used by the brain to perform a specific task easier. For example, the visual system applies a number of different rules in order to estimate the size of objects. Objects that are nearby will have about the same size as large objects from a distance. In addition, a bias is a tendency to make a certain type of error that is wrong in a certain systematic way. In most situations, heuristics work well. Nevertheless under certain circumstances they take care of optical illusions. Something similar happens when people try to estimate odds. Since the risk system also has its own set of tricks, systematic errors can occur. For instance, when people are not sure whether something is true or false, they are more inclined to guess that something is true. Besides that, a fundamental feature of human psychology, the optimistic bias, shows that humans expect positive events to occur rather than negative situations. This might also let to wishful thinking, since people confuse the hope for something pleasant with the reality of what is actually going to happen.
Another heuristic that influences people’s Risk Intelligence is the availability heuristic. The probability that is given to the existence of a future event is shaped by someone’s memory of a similar event. The time and ease with which someone can find such memories, determines the approximation that is given to the event. Unfortunately, mass media affects the relationship between the frequency of certain risks and the ease with which people can remember. Unlisted things like car crashes are underestimated, whereas things that get mentioned often such as plane crashes are overestimated. The problem with this matter is that people do not have much influence on the information they get nowadays. First sociologists thought that the amount of opinions would be increased in this digital age. By contrast nothing is less true since algorithms and cookies compose people’s knowledge and even more important, their perspectives. Someone’s world is created through the opinions that are sorted out for each individual. For instance, the results of a search term on the internet differ per person, according to their previous search terms.
Furthermore people have the tendency to rely too much on the strength of the evidence and too little on its credibility. This causes that the evidence strongly argues for a particular conclusion, although the reliability of the source has too much credence attached to itself. In the same manner the confirmation bias is found. People are more likely to pay attention to information that confirms what is already believed and ignore data that is contradictory therewith.
Although all these influences seem overwhelming at first sight, there are a few instructions and methods that mine these obstacles. First of all it is essential to understand Risk Intelligence can be divided in two different abilities. The first one is to possess the skill of expressing an approximation in a numerical exact probability. The second one is to be aware of the amount of knowledge which is contained in the brain. In essence, someone needs to obtain the necessary degree of certainty. This can be considered as a cognitive virtue. The middle between two extremes, too much and too little self-confidence, must be found. Rapid increase of the Risk Intelligence awareness is the first step to a higher RQ. Those who have a high RQ seem to take into account many variables that are viewed in context to each other. Experts are not only aware of the amount of knowledge they have, even more important they are aware of the things they do not know. One technique that improves someone’s RQ overtime is asking him the same question over and over again. For instance, a weatherman is asked for the probability it will rain the next day. Theretofore an expectation has to be translated in a numerical probability. Furthermore, his estimation will have a quick feedback which turns out to be fairly well for improving his RQ. Not so surprisingly weatherman have high RQ’s through this training, when it comes to their profession. Looking back at the heuristics and biases that obstruct someone’s approximations, certain ideas should be kept in mind to be able to counteract these impediments. Wishful thinking can be corrected by learning to look at the matter from an objective perspective. There needs to be a clear difference between that what is hoped for and that what will likely be realised. Furthermore the source that is used in order to acquire certain knowledge, should be inspected. Heading back to the availability heuristic, mass media can be deceiving and therefore false and crooked. Finally, when people only listen to the facts that confirm their own theory, they should look for counterarguments. This shows their own theory might have some defaults. Generally one should try to keep thinking rational. This, of course, needs to be trained over a period of time.
During tests students take, professors could not only just ask questions about the course. In addition, they could check the assurance of the student by asking them how sure they are about the correctness of the answer they have given. In this matter students will get a grade for the subject and even for their overall Risk Intelligence which is also trained through the test. Risk Intelligence is a skill that can be improved and is highly useful for many companies that need to make decisions under uncertainty. Measuring the RQ of candidates for a job interview can ensure that people with a higher RQ rather get a job, since they are an improvement for the company itself.

Dit artikel is geschreven door Anne Dumoulin


Read more

The 100 Prisoner Problem

The 100 Prisoner Problem

Imagine if tomorrow you were abducted, and before you knew it, you were trapped in a room with 99 other people who seemed to know nothing more about what was happening than you. You notice that everyone is wearing an orange jumpsuit, which is uniquely numbered. You...

The importance of statistics in sport

The importance of statistics in sport

“In terms of merit, sports have mathematical statistics. That is how you know who the best player is”. (Norm MacDonald) Until thirty/forty years ago people would most likely not believe in this statement, but the situation has changed since the end of the 90s when...