The Human Bubble, pt. 2

The Rookie: The question still remains: how does one know for sure whether one is within a bubble or else if one is within a permanent upward trend? The cases of the Palm IPO stock bubble and the more recent housing bubble had, now viewed in hindsight, obvious clues of their bubbly nature and so, as you point out, should have been obvious while within them. But I doubt that all bubbles are marked with such obvious contradictions as these two were. For example, the price of oil bubbled two years ago, but its increase seemed to be based on solid fundamentals; we were (and still are) finding new oil at a fraction of the rate of our overall consumption, and the new fields we find are generally more expensive to bring to maturity than were the old fields we're currently depleting. These fundamentals suggest an upward trend for the price of oil, but oil prices two years ago turned out to be a bubble after all.

Mr. Bishop: Yes, I suppose so. Maybe there are some contradictions going on with oil prices, but if so then they're not obvious to me. On the other hand, oil prices did rebound quicker than tech stocks or housing prices. Perhaps oil prices were indeed less bubblish.

The Rookie: Maybe so. Still, I don't see how even a keen awareness of the nature of market bubbles helps a person to avoid them, and I fail to see the value in being able to detect a bubble after the fact. I want to know when a bubble has formed before it pops.

Mr. Bishop: I suspect there is no way to know for sure, either practically or theoretically, when one is within a bubble. Consider the case in which a market's behavior is random, like how a coin toss is random. If I were to toss a fair coin one hundred times then we would expect that coin to land heads-up about fifty times. If, sometime partway through the hundred tosses, the coin had landed heads-up five times in a row, then the next flip still has an equal-likely chance of landing heads-up as it does of landing tails-up.

The Rookie: I understand that mathematically, but it seems unlikely that you would ever have six heads in a row within a mere hundred flips of a coin.

Mr. Bishop: Believe it or not, it's less probable not to have six heads-ups in a row than otherwise. But only slightly less probable.

The Rookie: No kidding?

Mr. Bishop: Yes, really. Randomness is less random than people often intuit, which is what makes it so hard to predict.

The Rookie: That's an interesting fact about the coin and an interesting example to bring up at a cocktail party, but I'm skeptical of the claim that markets are entirely random. Markets are based on real-world events, and though much that occurs in the real-world may be random, some of its events are not random. If even a small portion of a market's behavior is non-random then an aware individual should, in theory, have an advantage, however slight, in predicting future behavior, including bubbles. Right?

Mr. Bishop: No, indeed! It's possible for a market to be entirely non-random and still produce behavior that is impossible to predict. For example, imagine that a market is patterned on a fractal, such as constant number of local upswings and local downswings always occurring within a macro upswing and different, though constant, number of local upswings and local downswings always occurring within a macro downswing, and further imagine each of those local upswings and local downswings each having occurring within them the same constant number of even smaller local upswings and smaller local downswings and so on to infinity.

The Rookie: That's way too complicated for me. Can't you draw a graph?

Mr. Bishop: In a dialog? Nonsense. The point is that you're imagining some hypothetical market fluctuations being based on a small set of simple rules and, furthermore, as you zoom in on a small section of a historical price chart, you see that the pattern in the zoomed-in section is the same as in the bigger, zoomed-out section.

The Rookie: I'm still having trouble envisioning this, but it seems like if market prices followed a pattern involving simple rules and, furthermore, if I knew what those rules are then I would be able to predict future prices, at least on average over a long enough period of time. I'd surely take advantage and become very rich!

Mr. Bishop: That's just it: if the market is based upon a fractal pattern then it does not necessarily follow that one would be able to predict future prices.

The Rookie: You mean to say it's possible that market behavior is based on a simple set of rules and yet is unpredictable anyway?

Mr. Bishop: Yes, and that's the difference between something being random and something being chaotic. If something is random then it's based on no pattern and it's future behavior is thus unpredictable. If something is chaotic then it is based on a overarching pattern, but because you don't know where in the pattern you are, your ability to predict future behavior is no better than if the system were random.

The Rookie: Ugh, this sounds too complicated for me. I just want to know what I need to do to avoid making another mistake like buying a house during a temporary price run-up.

Mr. Bishop: In that case I have no advice but this: do what you can to think about one's current circumstances and environment within a historical context, and make the best decision you can without getting caught up in the various manias and depressions of the people around you.

The Rookie: That seems like sage advice. Have you used your own advice to discover any bubbles that we may currently be within?

Mr. Bishop: No, not within the context of the financial markets. If you're looking for a stock tip with which to make a quick buck then I have nothing for you. For that I can only recommend the Wall Street Journal, some push-pins, a wall, some darts, and a lousy aim.

The Rookie: Joke you may, such a stock-picking plan sounds like a good idea for those of us who seem to have a knack for picking the wrong stocks. At least achieving market average through random selection would be an improvement for us.

Mr. Bishop: Alas, market average, according to some, is not average at all but is rather the practical maximum for a minimal amount of risk over a long enough period of time. But let's ignore the financial markets for now. To answer your question another way, I have discovered a bubble, though it involves money only indirectly.

The Rookie: What bubble is that?