Book Review
Title: Against the Gods: The Remarkable Story of Risk
Author: Peter L. Bernstein
Publisher: John Wiley & Sons, 1996.
Leo Kee Chye
Wednesday, July 16, 2003
The book’s cover shows a reproduction of 17-century Dutch painter Rembrandt’s Storm on the Sea of Galilee, depicting a group of men struggling to hold onto the boat and to their lives while their vessel is at the mercy of a terrifying storm.
Not just in the sea, the element of risk permeates throughout every aspect of our lives. In the past, Mankind could not but feel we are at the mercy of some Gods whose whims seal our life and death; wins and losses; rise and fall. That feeling of inevitability has therefore led some to take solace and resignation in some omnipotent beings, while others pursue relentlessly to subdue the Gods with our primordial brains. Today, the former is found among the high priests in various religious bodies, while the latter assumes the role of high priest in the world of finance, insurance and other risk management activities.
Today, people no longer are at the mercy of the gods. They have learned to quantify risk and take steps to mitigate it. They could control their own fate. Man has reduced risk into a commodity and like a commodity, risk can be sold and bought according to the risk appetite of each individual. Buying travel insurance today when traveling come as naturally as bringing an umbrella when going outdoors. Looking back, we have come far and learnt much.
“Against the Gods” retraces the historical journey of men and their battle in subduing the Gods. From the Oracle at Delphi to modern Chaos theory and Neural networks, Peter Bernstein weaves an enthralling story of the evolution of risk and how it has led to the development of modern financial economics, peppered with brief but colourful tales of the men who have contributed to this edifice.
The book could not have found a better author. Bernstein has enough credentials up his sleeves for the task. A professional investor who is no stranger to risk management, he is also a scholar and historian in the area of financial economics. Without being pedantic and too technical, Bernstein did a wonderful task of compressing 300 years of risk-related literature into an entertaining book that can be enjoyed by even a layperson.
Against the Gods increases our understanding of the evolution of risk, which is not as quite straight forward as we think. Though we have come a long way since Pascal and Fermat, the story of risk is not over. The book points out some of problems especially in the area of forecasting where a new way of interpreting and measuring risk may be needed.
This is a brilliant book which I strongly recommend as an introductory text for anyone interested in risk management. The same goes to professionals and academics, for the historical treatment of the subject matter in the book can offer a new perspective on risk.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Comments (with a focus on economics)
In the past, randomness had led the Greeks to consult their Oracles at Delphi and the Chinese to I Ching. Apart from divination, randomness also acquainted humans to an exciting diversion – gambling. Strange as it sounds, this pastime led to the first attempt by men to quantify risk. From then onwards, there’s no looking back as we now marvel at the world around us, and much of which came from the ability to quantify risk. Without this ability, the world could come to a stand still. It is almost unthinkable to live our lives without some kind of life or medical insurance, to construct a building or anything without evaluating its financial risk, or to produce any goods without sample testing for defects, and list goes so.
But these tools have their share of flaws. For example, Bernstein warns the peril of extrapolation of past data for economic system which could be nonlinear. Unlike hard sciences, to set up a controlled experiment for testing economic hypotheses is difficult, if not downright impossible. Economists have to make do with whatever data they collected. But in a nonlinear dynamic system using time-series data, the mean of the population could be in a state of flux, which renders regression to the mean meaningless.
Besides forecasting, economists also use statistics to make inference, that is, whether to reject or not to reject a certain hypothesis. Here, it is important to differentiate correlation versus causality. For example, economist William Stanley Jevons (1863a) did discover the uncanny statistical relationship between sunspot cycle and business cycle which led him to develop his Sunspot theory. He hypothesised that the sunspot cycle led to a weather cycle which in turned caused a harvest cycle and hence a price cycle. Though statistically sound, it seems far-fetched and cranky. Correlation between phenomena does not necessitate causality between them. Lighting is often seen before thunder is heard; but that does not imply lighting causes thunder. Unfortunately economists could not conduct further experiments to confirm their suspicions.
Are humans rational? Maybe not. Bernstein tells the incredible findings of two Israeli psychologists, Daniel Kahneman and Amos Tversky, who set up many experiments to record how their human test subjects assess risk when confronted with different choices. To their surprise, in many instances, the subjects showed risk-loving when the mathematical odds were obviously against them and risk-adversity when the odds were for them. This is totally inconsistent with the assumption of rational behaviour, commonly used in economic theory modelling. The asymmetric behaviour patterns of individuals between gains and losses complicate matters when many important risk management tools are based on the assumption of rational behaviour.
On a fundamental level, statistical tools like “central limit theorem”, “law of large number”, “regression to the mean” are based on limited observations of physical phenomena. They do not necessarily imply the values of sample averages will take on the shape of a normal distribution or the sample averages will tend towards the mean when the size increases. In hard sciences, the possibility of repetitive experiments under the same controlled environment has made these tools somewhat “acceptable”. But, in social sciences, it does require making a leap of faith.
At the end of the book, Bernstein warns “Those who live only by the numbers may find that the computer has simply replaced the oracles to whom people resorted in ancient times for guidance in risk management and decision-making”. (Bernstein, p336). And the economists or fund managers of today aren’t too different from the soothsayers of the ancients.
Posted By
leokeechye@gmail.com
Categories
Economics