In summary of my series so far on Real Risk Management, each of us subjectively has preferences. These preferences, coupled with our chosen contexts, are what creates risk. As an example, if I choose to “beat the market” I have a chosen context of financial markets, and a preference to beat the performance of that market. The risk I’ve created is that I under perform. This may sound elementary and straightforward, but this stands in starkest contrast to the variance, standard deviation, beta, covariance, VAR, et. al. versions of risk promulgated in modern finance. Throughout the series, lurking in the background are issues are temporality. That is, the effect of time on risk management. Let me redress that deficiency now.
There are multiple temporal (i.e. time-related) issues that we need to get right if we hope to engage in real risk management. To me, these are:
1. Facts (the past) are often at cross purposes with investing (the future). 2. Our choice of time horizon introduces additional preferences/risks in our investment outcomes.
Risks are Future-Oriented
Because risks are about the possibility of a failed preferred outcome, it also means that they are about the future. After all, we anticipate possible futures, evaluate those possible futures, and then choose which possible futures we care about/hope to have happen.As I have been saying for years: there is no such thing as a future fact. Facts, by definition, occur in the past, yet investing and its results unfold in the unknowable future. Facts and investing are, in many ways, at cross purposes with one another.Magnifying the effects of the cross purposes is a mistaken belief on the part of almost all research analysts and portfolio managers in their objectivity. Yet, as I have demonstrated in these articles, risks are created by our preferences for contexts and outcomes. So, why do we care about facts in investing?We care about facts in investing because we understand that most outcomes are constrained. By what? Physical laws, resources, ingenuity, and so on. This is why correct contexts are so important for investing. If we choose well, for example, then we can bypass almost all other possible, but likely unimportant outcomes; such as a star going supernova in another galaxy affecting the price of a security. We reject this phenomenon because it is so far afield of our context. [Note: we do this qualitatively.]It isn’t that “everything goes.” It is more that “somethings are more likely to go.” So, we look to the past and the panoply of previous outcomes as a guide to possible, né probable outcomes. When we do this mechanically we create additional risks for ourselves, as I discussed in my Real Risk Management: Contexts. The reason is that if we are not careful in subjectively ensuring that our data matches our preferred contexts, then it may be that our data has never logged, recorded, and measured the probability of the risk that we want to mitigate or manage. Consequently, we are vulnerable to an outcome for which we did not plan.Another, perhaps easier way to think about the effect of temporality on investing is that, by definition, the time dimension is critical in a concept of effects following causes. Thus, as investors we try and identify the critical causes that may lead to probable effects, both good and bad. If, on average, we believe that the anticipated good outcomes outweigh the bad, then we invest. All the while, we diligently pay attention to possible past causes (i.e. risk factors) that might lead to negative effects again. And, if we are wise, we constantly try and anticipate new causes and their effects on our investments.
For example, was the risk of dying in a hijacked plane in the United States greater on 10 September 2001, or on 12 September 2001? In retrospect, the answer is obvious. Once the risk revealed on 11 September 2001 occurred it became part of the risk management of 12 September 2001 and thus it significantly reduced the risk of dying in a hijacked plane in the United States. Yet, most people at the time had the emotional experience that the risk was greater after 11 September 2001. All risks are future risks.In conclusion, on this first point, time is important in real risk management because: we use the past as a guide to possible/probable futures; and understand that effects follow causes.
Time Horizon is a Preference/Risk
I believe an underappreciated point about risk management is time horizon. Yes, we certainly talk about it in investing, but rarely are the philosophical and real world implications discussed. An obvious example is that one large difference between so-called growth and value investing is time horizon. Growth investors do not want to overpay for the combination of return opportunity and risk probability of an investment any more than do value investors. However, one principle difference is that growth investors are looking for a positive outcome over a shorter time horizon. They aren’t looking for an investment to double in value over 25 years, but in 2-3 years.Again, think about how our time horizon preferences contribute to our risks. Some risks are greater if I have a long-term time horizon. For example, the risk of having at least one quarter in which I lag my index is increased if I am asking investment consultants to evaluate me over 20 quarters. Yet, some risks are actually reduced by having a long-term time horizon. What else do we call the reversion to the mean phenomena that are ascribed to so many securities? That is, if a stock is down, just wait awhile and it is likely to revert to the mean. Succinctly, if I have more time, then I have more chances for my outcome to occur, and hence, the theory underlying options pricing.
Time horizon is a risk factor and must be factored into real risk management. This is because all risks are future risks; because all of my preferences can only find fulfillment/disappointment in the future. Therefore, I need to carefully choose my time horizon preference to ensure my preferred outcomes can be fulfilled (such as positive excess return), and that I do not suffer disappointment (such as underperformance). Jason A. Voss, CFA – Your Next Excellent Hire