Millions of Europeans are about to become the subjects of a vast social experiment. What’s troubling is how little anyone understands about where it might lead.

A total of 11 European Union member states -- including France, Germany, Italy and Spain, but not the U.K. -- plan to introduce a small tax on financial transactions by the beginning of 2014. Financial institutions will pay 0.1 percent on all stock and bond trades, and 0.01 percent on derivatives. Although taxes that are at least crudely similar exist in about 40 nations around the world, the European measure will be the first introduced on such a large scale.

The idea of a financial transactions tax goes back to the economist John Maynard Keynes. In the 1930s, he argued that speculation on assets drives market instability and suggested that an appropriate levy could deter it. If small enough, the tax would have a negligible effect on long-term stock investors, who trade infrequently and focus on real economic factors in making their decisions. It would primarily deter short-term speculators who buy and sell frequently in response to temporary market movements.

The idea makes intuitive sense and could, in principle, help channel investment to productive economic activity. There’s much debate, though, over whether it can work in reality. Well-known economists such as Joseph Stiglitz and Larry Summers have supported a transactions tax. Others of equal prominence have countered that it would be likely to lower equity prices, drive trading across borders and possibly increase market volatility.

Inconclusive Evidence

The empirical evidence is inconclusive. In general, studies find that volatility tends to rise in markets with higher transaction costs. There are plenty of exceptions, though, where higher costs were associated with reduced volatility or had no effect. A study in Sweden followed the introduction of a transactions tax in 1984 -- initially 1 percent, rising to 2 percent in 1986. It found no effect at the 1 percent rate, but a significant rise in volatility at 2 percent. No one is proposing anything nearly as large as a 2 percent tax.

As so often happens in finance, the debate has been clouded by alarmist warnings and the arguments of those who, for ideological reasons, see any tax or market intervention as an unmitigated evil. This is unfortunate, because beyond the fog and controversy lie legitimate questions that deserve exhaustive study.

A tax can influence market stability through multiple and interfering mechanisms. It would probably discourage speculative activity, but it might also curb the kind of short-term trading that seeks to exploit price distortions and hence stabilizes the market. By making trades more costly, it might also drive out so-called market makers, who earn a living by standing ready to buy and sell at any moment. This could increase volatility by impairing liquidity -- that is, the ease of buying and selling securities. Teasing out all the competing effects in a theory isn’t easy.

The most detailed studies of financial taxation have used computer modeling to simulate market ecologies, with virtual investors who interact, trade and evolve their strategic behavior over time. They found, as of 2006 or so, that a transaction levy, in the simplest setting, can improve stability just as Keynes expected. The tax rate, though, matters a lot. Set it too high (as in Sweden after 1986) and markets become less stable. Further studies confirm that a tax could indeed reduce market liquidity enough to undermine any positive effects.

The general theme appears to be that a tax might do good things in a market blessed with high liquidity, but cause trouble in markets where liquidity is sometimes in short supply.

Details Matter

More recently, the economists Frank Westerhoff and Paolo Pellizzari have gone further, finding that the success of a transactions tax might well depend entirely on fine details of how the market works, including the specific mechanisms by which buyers and sellers come together. In a market with orders handled by a dealer who provides liquidity, the tax has good effects. Replace the dealer with a simple auction mechanism, where trades are executed whenever the prices of buy and sell orders match, and the tax has a malign effect.

Just to be clear, no one, especially Westerhoff and Pellizzari, thinks the existing studies do more than scratch the surface. “Right now,” Westerhoff said to me, “we do not have nearly enough convincing studies on this issue.” The real-world outcome will probably depend on the characteristics of the market in question, including how liquid it is to begin with, how trading takes place, the mix of participants and how they respond to the tax. It’s reasonable to anticipate that many other fine details of market structure will matter, too, including things no one has ever considered.

The European Commission is clearly attracted by the prospect of tens of billions of euros in revenue that a financial-transactions tax could generate. And I suppose there’s something to be said for trying out the idea, as long as you’re prepared to learn and adjust quickly in light of what happens. It’s true that not having a tax is also a grand experiment, one we’ve been trying for the past two decades with rather dismal results. I can’t help but find it odd, though, that Europe’s policy makers are taking such a risky bet with so little understanding of the likely outcome.

(Mark Buchanan, a theoretical physicist and the author of “The Social Atom: Why the Rich Get Richer, Cheaters Get Caught and Your Neighbor Usually Looks Like You,” is a Bloomberg View columnist. The opinions expressed are his own.)

To contact the writer of this article: Mark Buchanan at buchanan.mark@gmail.com.

To contact the editor responsible for this article: Mark Whitehouse at mwhitehouse1@bloomberg.net.