The Best Self-Help is Free: The Limits of Knowledge, Ideas, and Impressions

When you come to hold any idea, it is essential to understand exactly how far this idea applies and what the limits of its application are. Every idea is an idea about something, and it will work well only when applied to that, which it is about.

It is tempting to assert, on occasion, that an idea which describes or works well in one particular situation can be applied with similar effectiveness to other similar situations. It is likewise tempting to derive a lesson from a particular life experience and to think that it applies in all times and all places, for all people in all circumstances. But especial care is needed here.

Every idea we arrive at is embedded in a particular context and is based on assumptions that may or may not hold in all circumstances. Sometimes checking whether an assumption holds is a matter of fairly simple empirical observation or logical deduction. At other times, however, whether the assumption holds depends on a complex interplay of factors to observe which one might have neither the time nor the access nor the mental processing power.

This is why, despite some highly advanced mathematical models and a decent understanding of human behavior, economists and financial analysts have never been able to reliably predict the movements of asset prices; the best that can be done is to explain the movements after the fact of their occurrence or - if one is especially perceptive - to foresee some general tendencies without knowing the specific price movements or the times at which they will occur. If economists, actuaries, and financial mathematicians were able to precisely figure out how the prices of stocks, bonds, options, and other financial vehicles changed over time, they would all be multi-millionaires by now!

The more I have studied economics and financial mathematics, the more astonished I have been at how little predictive power even the best economic and financial models have. Most of them assume a ceteris paribus world, where only one variable changes and all others remain the same. While looking at such scenarios is instructive for understanding the isolated impact of one particular variable, even that often does not suffice to determine the magnitude of that impact in a given case - unless an immense host of other factors can be exactly known. Moreover, in the real world, there are literally billions of variables - large and small - affecting, directly or indirectly, any event, decision, or system. Even being aware of all of these variables is impossible, much less exactly quantifying them, modeling their impact, and making a reliable prediction that rules out future changes in the variables themselves and the insertion of yet other new and unforeseen variables.

Yet compared to other assumptions in many economic models, ceteris paribus is one of the most reasonable and innocuous. Many economic models entail assumptions that cannot possibly hold in the real world - such as perfect information, the absence of transaction costs, homogeneous goods, infinitely divisible goods, and constant preferences. Many of these in fact assume away the very problems that complicate real-world decision-making.

But does this mean that we cannot plan for or anticipate the future? Are we doomed to live in a world of perpetual flux that we cannot possibly make sense of? Of course not. We can even make some fairly reliable predictions to guide us in our decision-making. However, it is vital to understand the limitations of one's own predictive ability and to look at any possible ways in which one's predictions might end up being mistaken. What complicating factors are out there? How might events deviate from their foreseen course? If a given situation leaves open multiple possibilities for the future, how can you be prepared for all of them and benefit from as many of them as possible?

So, by all means, go ahead and make the predictions you believe will assist you in better understanding the world around you and in attaining your goals. But give yourself a nice, wide margin of error, just in case your predictions end up missing the mark. If success in any endeavor requires your prediction to be exactly right, then chances are that you will not succeed. I have done enough chemistry and physics lab work to understand that even when dealing with exact, deterministic, and fairly simple processes, it is extremely difficult to get lab results that perfectly match the output of even a well-known formula that has been verified in practice thousands of times. But if you give yourself some leeway regarding both your criteria for success and the efficacy of your predictions, then you will in most cases do well enough.

Before undertaking any action, always ask yourself, "What is the worst possible outcome?" Also ask, "What will the outcome be if my predictions are completely mistaken?" The two may not always coincide, depending on the circumstances. After all, if you make a highly pessimistic prediction, matters might turn out to be better than what you anticipated, and you might even become better off for having been wrong. Be prepared to deal with both the worst possible outcome and the greatest possible failure of your predictions.

Take particular care judging people and predicting their behaviors. You simply do not know a lot of the particular "circumstances of time and place" in those people's lives, as Austrian economist and philosopher Friedrich A. Hayek would have put it. An action that seems wasteful or irrational to you at first glance might in fact be perfectly reasonable given some factor you do not know. Even a behavior that appears egregiously immoral to you might be the best possible behavior a person can engage in given the circumstances, and all the alternatives to that behavior might be far worse. Be careful making blanket judgments that you apply to all people.

When you come to know any particular person, try to suspend your judgment initially and just gather a lot of information. Give yourself as much data as possible to make an informed impression of the person's character, behavior, aspirations, virtues, and vices.

If you do not know something about a person, do not try to extrapolate it from what you already know; people can surprise you even more than markets or science experiments. Instead, if you are curious, ask - if the question can be posed in an appropriate and respectful way. If you cannot ask, suspend judgment on the matter in question and wait for further information to reveal itself to you. Meanwhile, interact with the person solely on the basis of what you actually already know.

When you are unsure about a person's motivations and moral qualities, always give him or her the benefit of the doubt - but make sure the person cannot deprive you of your property until the doubts are resolved! Remember that the wise man is compassionate and understanding with his mind, but not with his money. If he gives his money indiscriminately because of his faith in humanity, he simply becomes an unfortunate dupe. Never condemn without extensive forethought, but also never write a check without even greater forethought!

Finally, be frank about the limitations of your knowledge. When some important information is unavailable to you, admit it openly and qualify your remarks by indicating the scope to which they apply. If you admit what you do not know, then what you do know will be seen as much more reliable by others. The only case where you ought to hazard a guess despite incomplete or unavailable information is a multiple-choice exam with no guessing penalty or an analogous situation. If you get nothing for admitting ignorance but you have some decent probability of getting something if you guess, then by all means guess. But if you stand to lose anything by guessing, it is best to abstain.

Read all chapters of The Best Self-Help is Free.