Monday 3 October 2011

Book Review: The Rules of Risk by Ron Dembo & Andrew Freeman



The book isn't just about investing risk though the book's subtitle is "A Guide for Investors". This must have been added by the influence of the publisher since the more accurate description of the content is on the back cover: "A Dynamic Framework for Forward-Looking Risk Management". Likely the latter would have been judged too technical and jargonish to appeal to a broad book audience.

The book is a broad conceptual guide, and a useful one, to figuring out what to do about various financial risks one faces in life, like house buying, insurance or investing. There is some jargon (do you want to figure out your lambda?) but it presents mostly common sense argumentation and nothing beyond arithmetic in the numerical examples.

Despite some attempt to flesh out their methods, the book falls short of being a practical guide. I find it difficult to see exactly how I could systematically apply, in order to develop an overall financial and investing plan, all of their ideas in their multi-step process, which consists of:
1) Know the value of your holdings today - I'm ok with this one, I think, though an important caveat is the definition of "my holdings", which they do not explore. They seem to think of such holdings as consisting only of a portfolio of financial securities, which I believe is far too limiting. The single most valuable financial asset we own is ourselves, our own money-earning capacity, often called human capital. Unlike almost every other financial risk, which they explicitly define as those subject to changes in the [external] environment (i.e. that we can do nothing about directly), we can enhance, or neglect, our earning capacity through education, diet, health protection etc. I suppose the authors might respond that their book does not intend to go into that detail. And I suppose that doesn't matter as long as this book is only considered a conceptual, not a practical, guide. This book does not hand you solutions on a plate. You would have to do quite a bit of work to apply them to your own circumstances.
2) Pick an appropriate future time horizon - Though Dembo and Freeman acknowledge that most people are likely to have multiple future financial goals (car, House, retirement etc) with different time horizons, they do not flesh out how to deal with the resulting complexity when this fact is added into downstream steps below. Nor do they address at all what to do about the very real fact that life contains surprises and your best laid plans / time horizons may not be the ones that actually happen e.g. a good number of people retire sooner than they thought due to job changes or health. That really complicates the next step.
3) Choose a range of scenarios of the future, making sure to include bad extremes and assign a probability to each scenario - This is the range of future end results for "the portfolio under consideration". Of course there could or would likely be multiple portfolio possibilities an investor might want to consider. This is the point where I'd guess almost anyone trying to follow their method would give up. Huh? How do I make up scenarios and figure the odds? They touch on but do not resolve the issue on page 81: " ... there are many occasions when it is impossible to attach useful probabilities to an unknowable future, then we have to find ways to model the uncertainties we face."
4) Pick a benchmark - The benchmark seems to be a kind of base case to compare the scenario & portfolio combinations (& presumably the multiple time horizons). They say that one can choose whatever benchmark one desires.
5) Value your portfolio and benchmark for time horizons under all scenarios - a lot of mechancial work
6) Compute the appropriate risk measure based on values coming out of step 5 - Here enters one of the intriguing contributions of the book, the idea that something they call Regret is the best measure of risk to use instead of popular commonly used measures like standard deviation (aka volatility) or Value-at-Risk (VAR, used by institutional investors).

Regret, with a capital R, is what I found to be the most useful, as well as the most accessible and natural, concept in the book. It is a way of taking account of potential harmful outcomes to make better decisions. Regret deliberately embodies the emotional impact of negative possible outcomes. A wrong decision, as judged later, can gnaw away at a person forever after. Second, the Regret evaluation process of Dembo and Freeman tells us to assess the cost of preventing the big negative, which in many cases is just insurance. Then we decide whether the cost is worth it considering more the consequences of decisions rather than the probabilities of particular outcomes.

As the authors say, "It is a sensible rule of thumb that an operation should not take on positions that expose it to the worst possible outcome - that a catastrophic loss might occur, resulting in ruin." This way of thinking would be very helpful in considering, for instance, whether to buy long term care or critical illness insurance. The same loss may matter a lot more to some people than others e.g. Warren Buffett will never be interested in buying long term care insurance since he can easily afford the fanciest care imaginable.

The obverse of Regret is what they call Upside, a positive consequence that can result from a choice. The thinking process is the same - how much could the positive result amount to and is the cost of the "bet to enter the game" worth it?

The book also has several interesting or amusing bits:
  • there are a number of praiseful references (this book was written in 1998) to the revised and improved risk management techniques of major investment banks following big losses in the 1980s and 1990s, rather ironic given the 2008 debacle; the notion presented by the book that these financial institutions were really trying to manage risk and avoid disastrous financial consequences instead of going willy nilly for the gigantic gains strikes me as rather naively quaint. Dembo and Freeman's own framework provides a handy simple tool to figure out the big banker's personal risk perspective - lots and lots of personal $$$ Upside with the worst personal Regret being fired and having to look for another job.
  • Dembo and Freeman show in Chapter 3 that Kahneman and Tversky are wrong to say that people are irrational or making behavioural mistakes when they make choices that violate expected value (EV = probability x outcome) rationality. They also make buying lottery tickets an entirely reasonable and rational decision (spending a few dollars a week won't cause even poor people financial ruin but it could, despite the bad odds, make them rich, so there's lots of Upside and really no potential Regret).
Bottom line: Though the framework will get too complicated to apply to a total personal financial planning effort, it can apply to one-off decisions. The Rules of Risk is worth buying for its thought-provoking content and useful concepts.

Rating: 4 out of 5 stars

No comments:

Wikinvest Wire

Economic Calendar


 Powered by Forex Pros - The Forex Trading Portal.