What is a security transaction tax (STT)? The Investment Company Institute (ICI) states, “a securities transaction tax is a tax imposed on securities transfers, including, generally, purchases and sales. The tax could apply to the value of trades in stocks, bonds, derivative instruments, mutual funds, exchange-traded funds (ETFs) and other securities.”1
Recently, STTs have received a lot of attention, largely as a result of the general scrutiny of the financial sector after the economic crisis. On Dec. 9, 2009, Peter Anthony DeFazio (D-Oregon) introduced the “Let Wall Street Pay for the Restoration of Main Street Bill” to the U.S. House of Representatives.2
Earlier, in 2010, a group of charities launched an international campaign to promote a “Robin Hood Tax,” with the objective of taxing financial transactions and generating revenues to fight poverty.3
Instead of debating such a tax, on Jan. 29, 2012, French President Nicolas Sarkozy announced that France planned to unilaterally impose a 0.1 percent tax on financial transactions starting in August.4
In this article, I will focus on one aspect of the security transaction tax: its impact on volatility. Although there are a number of important aspects of the security transaction tax deserving of study, such as its revenue implications, I choose to focus on volatility because it is central to many of the discussions taking place. Also, it’s a topic that can be explored given the brevity of this article. Proponents of the security transaction tax argue that the tax reduces “excess volatility,” or price movements unwarranted by news about fundamentals (Keynes, 1936; Stiglitz, 1989; Summers and Summers, 1989). The idea is that a security transaction tax makes it more expensive for the “noise traders” (i.e., market participants without fundamental data who trade on news or “noise”). Therefore, because they contribute nothing but noise to the price, a tax that curtails their trading should reduce volatility.
Although the argument is appealing, it is incomplete, at best. While a security transaction tax does serve to reduce the trading activities of noise traders, it also reduces the trading activities of other types of traders who are not reacting to the noise. To illustrate, let us consider a simple example where a block of shares is offered for sale. Naturally, some investors step in and demand a slight price concession for providing liquidity to the seller. The same group of liquidity providers holds onto the shares for awhile before subsequently selling them to other investors—most likely at a slight premium. Without a security transaction tax, the overall price fluctuation is relatively small. However, when the tax is levied, liquidity providers not only demand compensation for liquidity provision, but they also demand that the counterparty should cover the tax. Therefore, share prices fluctuate more during the round-trip transaction, and introduce additional volatilities.
This example is particularly relevant for today’s financial markets, where low-latency traders play the important role of liquidity provision, which in turn reduces volatilities (Hasbrouck and Saar, 2011). As low-latency traders commonly operate in a world of tenths of a basis point, a security transaction tax most likely will preclude them from participating in the market-making process, and potentially call their survival into question.
Beyond the theoretical arguments, empirical evidence also suggests that a security transaction tax does not reduce volatilities. Jones and Seguin (1997) study the change in stock volatility resulting from the introduction of negotiated commissions on the U.S. national exchanges in 1975. The commissions can be viewed as analogous to a security transaction tax. Compared with the fixed commissions, negotiated commissions effectively reduce security transaction tax by 4 percent to 20 percent, depending on the order size. In comparison to stocks not affected by the introduction of negotiated commissions (i.e., those traded on NASDAQ), volatility of the set of stocks with negotiated commissions (i.e., those traded on NYSE and AMEX) dropped by 27 percent. Among the largest 20 percent of stocks, the decrease in volatility is even more pronounced: It is nearly 50 percent.
Umlauf (1993) provides an interesting case study of security transaction tax reform in Sweden. In a response to political pressure arising from the labor sector’s envy of the high income earned by financial industry professionals, Sweden introduced a 1 percent round-trip tax on equity transactions in 1984. Two years later, the tax rate on equity transactions increased to 2 percent. Umlauf (1997) considers a pair of shares issued by the same underlying issuer, but traded on different exchanges: London versus Stockholm. In the sample period, the UK levied 0.5 percent “stamp duty tax” on stock purchases, plus a £2 surcharge for transactions over £5,000 (Table 6.1, Campbell and Froot, 1994). The transaction tax hike in Sweden introduced a spike in volatility. Compared to stocks traded in London, stocks traded in Stockholm experienced an approximately 6 percent increase in volatility.5
In the debate about a security transaction tax, volatility is certainly important for its own sake. Moreover, volatility has direct implications for an array of other market attributes. One aspect is liquidity, which is an elusive concept. Here, we narrowly define liquidity as market-determined transaction costs (excluding taxes), such as bid-ask spreads and price impact. Because transaction cost is proportional to stock volatility (equation 19, Chacko, Jurek, and Stafford, 2008), the empirical evidence on security transaction taxes and volatility suggests that a transaction tax affects trading costs as well. It is worth noting that the impact of tax on transaction costs is not limited to the direct cost of the tax increase borne by market participants. What really matters is that a transaction tax increases trading costs by indirectly increasing bid-ask spreads and price impact, thus making the security less liquid. Unfortunately, there is not much published academic research investigating the link between transaction taxes and trading costs, most likely due to lack of data.6
Nevertheless, if one views trading volume as an indication of liquidity, then the evidence is ubiquitous: An increase of a security transaction tax always decreases trading volume.7
Lastly, Schwert and Seguin (1993) recognize that a transaction tax directly affects the real economy’s cost of capital, a direct consequence of tax capitalization (Sialm, 2009). Moreover, through the channel of volatility, transaction tax also indirectly affects the real economy. The intuition is that the volatility increase induced by a transaction tax makes securities less liquid. Less liquid securities demand an additional premium in expected return (Amihud and Mendelson, 1986), thus again, increasing the cost of capital.
Overall, it is far from clear that a security transaction tax reduces volatility. Available empirical evidence suggests otherwise. Moreover, a transaction tax indirectly affects market liquidity and the real economy’s cost of capital—most likely in an adverse manner.
1) Source: http://www.ici.org/stt/ici_resources/faqs_stt
2) Source: http://www.govtrack.us/congress/billtext.xpd?bill=h111-4191
3) Source: http://uk.reuters.com/article/2010/02/10/uk-britain-banktax-idUKTRE61900U20100210
4) Source: http://www.bloomberg.com/news/2012-01-29/financial-transaction-tax-in-france-to-take-effect-in-august-sarkozy-says.html
5) Sweden eventually abolished the security transaction tax in 1991.
6) For instance, systematic collection of high frequency data in the US only goes back to 1982; and even later outside the US. Since middle of 1970’s, there has not been much change of transaction tax rate in the US. Therefore, more recently high frequency data do not offer much help here.
7) Recently, Corwin and Schultz (2011) develop an estimate of bid-ask spreads using daily high and low prices. Based on their estimates, and the empirical methodology in Jones and Seguin (1997), I estimate the bid-ask spreads changes around the introduction of negotiated commission for stocks traded on NYSE and AMEX, then compare those changes to the changes associated with stocks traded on NASDAQ. My calculation shows that, the average bid-ask spreads of NYSE/AMEX stocks decreased by about 20% between two periods: (1) April, 1974 to April, 1975 (pre-event period), and (2) June, 1975 to June, 1976 (post-even period). Interestingly, NASDAQ stocks also experience a reduction in trading cost by 9%, albeit a much smaller number. Overall, the results are consistent with earlier evidence and (the) theoretical model’s prediction: reduction in (a) transaction tax improves liquidity to the extent measured by bid-ask spreads. Future research on this is clearly warranted.
Amihud, Yakov, and Haim Mendelson, 1986, Asset Pricing and the Bid-Ask Spread, Journal of Financial Economics 17 (2), 223-249.
Campbell, John Y., and Kenneth Froot, 1994, International Experiences with Securities Transaction Taxes. In J. Frankel (Ed), The internationalization of equity markets (pp. 277-308). Chicago: University of Chicago Press.
Chako, George C., Jakub W. Jurek, and Erik Stafford, 2008, The Price of Immediacy, Journal of Finance 93(3), 1253-1290.
Corwin, Shane, and Paul Schultz, 2011, A Simple Way to Estimate Bid-Ask Spreads from Daily High and Low Prices, Journal of Finance, Forthcoming.
Hasbrouck, Joel, and Gideon Saar, 2010, Low-Latency Trading, Working Paper, Cornell University.
Jones, Charles M., and Paul J. Seguin, 1997, Transaction Costs and Price Volatility: Evidence from Commission Deregulation, American Economic Review 87(4), 728-737.
Keynes, John Maynard, 1936, The General Theory of Employment, Interest and Money, MacMillan, London.
Schwert, G. William, and Paul Seguin, 1993, Securities Transaction Taxes: An Overview of Costs, Benefits and Unresolved Questions, Financial Analysts Journal, 27–35.
Sialm, Clemens, 2009, Tax Changes and Asset Pricing, American Economic Review 99(4), 1356 – 1383.
Stiglitz, Joseph E., 1989, Using tax policy to curb speculative short-term trading, Journal of Financial Services Research 3, 101-15.
Summers, Larry H., and Summers, Victoria P. Summer, 1989, When financial markets work too well: a cautious case for a securities transaction tax, Journal of Financial Service Research 3, 261- 286.
Umlauf, Steven R., 1993, Transaction taxes and the behavior of the Swedish stock market, Journal of Financial Economics 33(2), 227-240.