web server statistics
American Accounting Association

Ben Haimowitz (212-233-6170)


High volume promotes high volatility in share prices, study finds

New research spurs century-old question: Is it time for a stock-transaction tax?

Does high trading volume promote share-price volatility? While it has long been thought that the two are connected, it has not been clear whether the first causes the second or whether they both are driven by the amount of information flowing through the market. With trading volume having surged by 500% to 600% over the past two decades -- and close to 3,000% over the past half century -- the question has grown in urgency.

Now a new study employs an ingenious strategy to arrive at an answer. In research to be presented to the forthcoming Annual Meeting of the American Accounting Association (Aug. 6-10, Denver), three scholars will report that sheer volume accounts on average for about 25% of share-price volatility -- and more at the highest trading volumes.

"We find repeated and economically substantial evidence that more intensive stock trading is accompanied by increased return volatility," report the study's authors, Ilia D. Dichev and Dexin Zhou of Emory University and Kelly Huang of Georgia State University. "This relation is weak to non-existent at low to moderate levels of trading but becomes increasingly strong as volume of trading increases…The combined impression from these results is that high volumes of trading can be destabilizing, injecting a sizable layer of trading-induced volatility over and above the unavoidable fundamentals-based volatility."

So clear are the results that they lead the study's authors to raise an idea first proposed a hundred years ago in the wake of a massive wave of speculation in U.S. railroad stocks -- namely, dampening speculative excesses by levying taxes on trading. Advocates of the idea over the past century have included John Maynard Keynes, who saw it as a way of "mitigating the predominance of speculation over enterprises in the United States," and Lawrence H. Summers, who proposed it in a 1989 article that he co-authored with his first wife, Victoria, "When Financial Markets Work Too Well: A Cautious Case for a Securities Transactions Tax."

While not offering an opinion about whether such a tax should be enacted, Prof. Dichev observes that the two decades following the Summers' article witnessed the most spectacular surge in stock-trading volume in history. "They argued in 1989 that a securities transaction tax would curb speculation and raise a significant amount of money. If that was true then, it would likely be all the more true today."

The new study probes the relationship between volume and volatility by taking advantage of three natural experiments in which there is a substantial difference in volume of shares traded without a significant difference in fundamental information about the companies involved -- 1) when stocks switch exchanges; 2) when stocks are added to or deleted from the S&P 500; and 3) in cases of dual-class stocks (for example, A-class shares that feature different voting rights from those of B-class shares).

In the first of these experiments, the professors analyze 2,860 exchange switches that occurred between 1962 and 2009 and find that "change in volume produces a near monotonic ordering on change in volatility," results that are not only statistically significant but "economically substantial."

For example, among the almost one thousand switches between the New York Stock Exchange and the American Stock Exchange, the roughly 200 stocks that had the biggest drop in volume were traded on average 44% less in the month following the switch than they were in the month before the switch, and their volatility (measured by the standard deviation of daily stock returns) declined by about 25%. In contrast, the 200 stocks whose volume rose the most following the switch (on average, about 112%) registered an increase in volatility of almost 14%.

In other words, the difference in volume change between these two groups was about 156%, and the difference in volatility, paralleling the volume change closely, was about 39%. As the study puts it, "If such magnitudes are anywhere close to a guide for what one can expect in more generalized settings, it is clear that the...30-fold increases in volume [since 1960] likely have a pronounced effect on observed stock volatility." Further, the results in the second natural experiment (addition to or deletion from the S&P 500) are "remarkably similar to those for exchange switches," and results for the third experiment (dual-class shares) are "largely consistent" with the other two.

The researchers extend their probe to more generalized settings by analyzing the trading volume and volatility of all NYSE-AMEX stocks over the 20-year period 1988 through 2007. First they divide the sample into deciles on the basis of trading volume and find that, while volatility rises by more than 50% from the first to the tenth deciles, almost the entire rise occurs where turnover is extremely high. As the study puts it, "There is little reliable variation in volatility from the first decile until about the seventh decile, followed by a quick rise and hitting a high in the top decile...in the extremes of high trading."

Then the professors take the further step of controlling for companies' earnings variability, which, they find, accounts for about three times the amount of volatility that trading volume engenders. This, they observe, is hardly surprising, since "in an efficient market fundamental variability should be the only variable that affects stock volatility. What is more remarkable, actually, is that the effect of trading intensity remains economically large after controlling for fundamental variability... [so that it accounts] for about a quarter of total stock volatility."

While conceding that "theoretically, there is a solid argument that higher investor participation and trading lead to better price discovery and therefore to prices that are closer to fundamental values," the professors note that "much of the previously documented benefits of liquidity come from environments with low trading intensity…In contrast, the evidence in this study comes almost exclusively from the largest, most-traded environments and stocks of all time."

They conclude: "While the early gains from trading and liquidity are widespread, the benefits at very high levels of trading are much more specialized, accrue to a smaller circle of people, and lean in the direction of redistribution rather than the creation of new wealth. While it is helpful to be able to buy and sell sizable investment positions promptly, the race to trade on slivers of new information a fraction of a second faster than anybody else is more questionable as a value-enhancing activity at the society level."

The study, entitled "The Dark Side of Trading," will be among hundreds of scholarly presentations at the Annual Meeting of the American Accounting Association, a worldwide organization devoted to excellence in accounting education, research, and practice. More than 3,000 scholars and practitioners are expected to attend the meeting in Denver, Aug. 6-10


© 2010 American Accounting Association All rights reserved.