Friday, August 22, 2008


I'm reading a book about bubbles and other mass mistakes called Extraordinary Popular Delusions and the Madness of Crowds by Charles Mackay, published in 1841. I've only read the first two cases--the Mississippi Company bubble and part of the South Sea Company bubble, with the Dutch tulip bubble yet to come--but so far it's fascinating. Both the Mississippi Company bubble and the South Sea Company bubble deal with speculation in stock issued by companies granted monopoly by their government. The companies exaggerated the profits to be had in the New World, and people bought the stock they issued with excessive optimism. The price of the stock rose, and the companies sold more stock. In the case of the Mississippi Company, the French Royal Bank issued notes with the stock that could be used as currency, resulting in rapid inflation.

What is so fascinating about bubbles is how difficult they are to predict. No doubt many businessmen put a bright spin on their business; why do bubbles appear in some industries and not others? Why does the price go so astronomically high before anyone wises up?

Such bubbles are not always in financial instruments. There was the Beanie Baby bubble of the 1990s, the Comic Book bubble of the late 80s and early 90s, the current real estate bubble, the Internet bubble of the late 90s, and others. Why do they keep happening? Why do people repeatedly fail to see something is outrageously overvalued?

We can't really know there's a problem until it's too late. For example, the internet was a really big deal. Much of the gains were surely real. There may not have been enough information to determine that the internet was not as awesomely profitable as we hoped until it's too late. By definition, really, that's the frustrating thing about bubbles: We don't know they're going to pop until they do. They're bubbles because we don't know that they're bubbles.

It's worth pointing out that blaming speculation or greed is silly. Speculation is happening all the time in all sorts of markets, but it doesn't always lead to bubbles. Similarly, everyone who is looking for profit is, in some sense, greedy, but we don't see bubbles inflating and bursting everywhere. In fact, speculation is often essential to the proper functioning of a market. Speculators use knowledge of things that may happen in the future to make a profit, and in doing so they cause the price today to reflect that information. When a speculator learns that the orange juice crop next year will be poor, and buys up frozen concentrated orange juice futures, he drives up the price of frozen concentrated orange juice. This is good because it communicates to consumers that orange juice is becoming scarcer so they should reduce consumption, and it communicates to farmers that orange juice is becoming scarcer so they should increase production (if possible). This is exactly what prices are supposed to do. The problem in a bubble is that everyone's expectations about the future get out of whack, so prices cease to convey useful information.

This brings me to oil. There has been a lot about whether or not there is an oil bubble. It seems to me that we can now reject that idea. If it had been a bubble it would have burst. Instead, the price of a barrel of oil seems to be slowly falling. Apparently it's just ordinary supply and demand (in which expectations play a part). I'm guessing oil prices will continue to fall until the winter heating oil demand increase. Maybe oil will get very expensive again next summer--maybe we're beginning to have global "summer driving seasons", resulting in greater variance in demand.

One final somewhat technical point: When most of us think of an auction, we think of what is called an Oral Ascending Price Auction, or English Auction. In this auction bidders increase their bids by calling out or signalling, and the highest bidder wins and pays his or her bid (There are other auction structures, such as the Second Price Sealed Bid used by Ebay). The primary advantage of the English Auction is that when the value of the thing being bid upon is of uncertain value, bidders can see what other people are bidding and use that information to update their own estimate of the item's value. That is, if you see that lots of people dropped out of the auction when the price hit $100, then that is a sign that your estimate of $200 is too high. After all, you don't want to suffer the Winner's Curse, and end up paying $200 for something that is only worth $100.

The reason I bring this up in a post about bubbles is that in a bubble, this information mechanism doesn't seem to work. People don't drop out of the bidding, and their estimates of the value of the item just keep climbing as a result. I realize the analogy isn't perfect; in an auction there's one item up for sale, rather than market with many buyers and sellers, and people aren't really bidding. English auctions have been used successfully by governments in distributing sections of electromagnetic spectrum. Are there any famous examples of bubbles in English auctions?

Yes, the title of this entry is the same as that of an episode of the Powerpuff Girls.