[By Zhou Tao/Shanghai Daily]

PEOPLE frequently ask me, as someone who has written on market speculation, where the next big speculative bubble is likely to be. Will it be in housing again? Will it be in the stock market?

I don't know, though I have some hunches. It is impossible for anyone to predict bubbles accurately.

Bubbles are social epidemics, fostered by a sort of interpersonal contagion. A bubble forms when the contagion rate goes up for ideas that support a bubble. But contagion rates depend on patterns of thinking, which are difficult to judge.

Big speculative bubbles are rare events. (Little bubbles, in the price of, say, individual stocks, happen all the time, and don't qualify as an answer to the question.) And, because big bubbles last for many years, predicting them means predicting many years in the future, which is a bit like predicting who will be running the government two elections from now.

But some places appear a little more likely than others to give rise to bubbles. The stock market is the first logical place to look, as it is a highly leveraged investment - and has a history of bubbles.

There have been three colossal stock-market bubbles in the last century: the 1920s, the 1960s, and the 1990s. In contrast, there has been only one such bubble in the United States' housing market in the last hundred years, that of the 2000s.

We have had a huge rebound from the bottom of the world's stock markets in 2009. The S&P 500 is up 87 percent in real terms since March 9 of that year. But, while the history of stock-market prediction is littered with too much failure to try to decide whether the bounceback will continue much longer, it doesn't look like a bubble, but more like the end of a depression scare. The rise in equity prices has not come with a contagious "new era" story, but rather a "sigh of relief" story.

Likewise, home prices have been booming over the past year or two in several places, notably China, Brazil, and Canada, and prices could still be driven up in many other places. But another housing bubble is not imminent in countries where one just burst. Conservative government policies will probably reduce subsidies to housing, and the current mood in these markets does not seem conducive to a bubble.

A continuation of today's commodity-price boom seems more likely, for it has more of a "new era" story attached to it. Increasing worries about global warming, and its effects on food prices, or about the cold and snowy winter in the northern hemisphere and its effects on heating fuel prices, are contagious stories.

Dark horse

My favorite dark-horse bubble candidate for the next decade or so is farmland - and not just because there have been stories in recent months of booming farmland prices in the US and the United Kingdom.

Of course, farmland is much less important than other speculative assets. For example, US farmland had a total value of US$1.9 trillion in 2010, compared with US$16.5 trillion for the US stock market and US$16.6 trillion for the US housing market. And large-scale farmland bubbles are quite rare: there was only one in the US in the entire 20th century, during the great population scare of the 1970s.

But, farmland, at least in certain places, seems to have the most contagious "new era" story right now. It was recently booming, up 74 percent in real terms in the US in the decade ending with its price peak, in 2008. And the highly contagious global-warming story paints a scenario of food shortages and shifts in land values in different parts of the world, which might boost investor interest further.

Moreover, people nowadays easily imagine that the housing and farmland markets always move together, because prices in both boomed in recent memory, in the early 2000s. But, from 1911 to 2010 in the US, the correlation between annual real growth of prices for homes and farmland was only 5 percent, and the latest data on farm prices have not shown anything like the decline in home prices. By 2010, real farm prices in the US had fallen only 5 percent from their 2008 peak, compared to the 37 percent decline in real home prices since their peak in 2006.

The housing-price boom of the 2000s was little more than a construction-supply bottleneck, an inability to satisfy investment demand fast enough, and was (or in some places will be) eliminated with massive increases in supply.

By contrast, there has been no increase in the supply of farmland, and the stories that would support a contagion of enthusiasm for it are in place, just as they were in the 1970s in the US, when a similar food-price scare generated the century's only farmland bubble.

Still, we must always bear in mind the difficulty of forecasting bubbles. And, for daring investors, it is not enough to find a bubble to pile into. They must also try to determine when to cash out and put their money elsewhere.

(The author is professor of economics at Yale University and coauthor, with George Akerlof, of "Animal Spirits: How Human Psychology Drives the Economy and Why It Matters for Global Capitalism." Copyright: Project Syndicate, 2011. www.project-syndicate.org)