In an editorial published in the Wall Street Journal (“Taxing Stock Trades Will Hurt Main Street“, by Yakov Amihud and Haim Mendelson, Nov 11, 2011) the authors inadvertently make the case for imposing a stock trade tax rather than opposing it.
Their argument against a tax boils down to the claim that a fee of 0.25% on each trade (which is the amount a bill going through the Senate would levy) would depress stock prices 10%, which in turn would have a cataclysmic effect on our economy. But based on their own evidence the tax would not depress stock prices, and in my opinion would provide large indirect benefits.
I have made a living since 1986 in high frequency trading and a tax of .25% would drive me out of business as I know it. However I have long recognised the potential benefits of such a tax, my own personal situation notwithstanding. To wit, a tax would reduce superfluous trading volume, volume which in and of itself does little for our nation’s prosperity.
Dime store economists will argue that volume provides liquidity for securities markets, which I agree is an undeniable good for an economy. But at some point the marginal gain in volume doesn’t provide increased liquidity or any other benefit–it’s just churn.
It’s difficult to say where trading volumes begin to have zero marginal value, but some numbers might help give context: from 1948 to 1983 our banking sector was responsible for between 5% and 17% of our national corporate profits, while from 1998 to 2007 the range was 27% to an astounding 40%.
Meanwhile NYSE trading volume grew from 34mm shares a day in 1979, to 900mm shares a day at the peak of the dotcom bubble in 1999, to 2.5bb a day in 2007 (which is still the average in 2011). Yet between 2001 and 2011 our economy has had its worst decade by many economic measures including job creation, stock market performance, and GDP growth.
By what mechanism has all this additional trading volume hurt our economy? One way is by using valuable resources that could have been used elsewhere, taking scientists and engineers and other talented quants (not to mention history majors) out of the lab and on to trading desks. I know this because I’ve hired some of them myself. And what does that extra volume provide if not liquidity? Trades in and of themselves do not create wealth for an economy since trading is a zero sum game. The total wealth gained or lost in any market is only equal to the amount the market moves.
The number of trades is irrelevant and in fact one person’s gain is another’s loss (or profit foregone). All participants can make money if the market goes up, but the same total wealth would have been created had 1/100th (or zero for that matter) of the trades taken place. In other words, if the explosion in banking profits between 1980 and 2007 have come from trading, they have come at someone else’s expense (namely pension funds, mutual funds and non-professional participants) and have been of limited value to our economy.
Ok, so trading is a zero sum endeavour, but if higher volume means more liquidity (at least up to a point), that would indeed increase the value of all securities, which (as per the ‘zero sum’ elucidation above) does create wealth. So, how much value would our markets lose if volume was decreased? The authors against the transaction tax say 10%, though they give no evidence for how they came to translate volume change to price movement.
But they do cite studies of previous securities transaction taxes in the U.S. that found no consistent relationship between tax regime and volatility—a finding which if anything gives reason FOR a transaction tax rather than against. In my mind the reason not to impose a tax was that market makers, those providing liquidity, also known as high frequency traders, would thin considerably in the advent of a .25% tax since most of their trades have an expected return of 1/5 that. And with market makers leaving, liquidity would dry up, which I expected would lead to higher volatility, which undeniably leads to lower securities prices. But if there is no commensurate spike in volatility, then how far could security prices fall in the advent of a transaction tax?
We cannot know for certain, but with no increase in volatility the number would be closer to zero than 10%. And even if it were 10%, the resulting benefit from moving valuable resources away from non-productive trading to more marginally productive sectors would more than wipe out any one time 10% drop. After all, look at what little real gain the converse has provided us.