Although the Security Transaction Tax (STT) has been implemented by many nations, its effects have varied by country. STT is an ad valorem tax on the value of trade of different securities and is often extended to their derivatives too. While theoretical evidence is mixed on the efficacy of imposing STT, empirical evidence suggests that it has not met with success in most countries and hence has been reduced and/or rolled back over time. STT in India was introduced in 2004 and currently stands at 0.1 % on sale and purchase of equities and 0.017% on the sale of their derivatives such as futures and options that are not exercised.
Proponents have long argued that such a tax helps in revenue mobilisation, checks for volatility in stock prices and excessive speculation/noise trading. In addition, some have argued that the tax raises asset prices and decreases cost of capital while providing an incentive for corporate managers to prioritise long-run decisions over short-term ones. Opponents, however, indicate that STT does not always fulfill its stated objectives and rather cripples stock market efficiency. For instance, revenue from STT might not actually meet its estimated potential if there is a migration of trading volume to other untaxed regions and/or untaxed substitutes. There could also be an increase in price volatility and overall decrease in liquidity. This article examines the case for STT with the help of available empirical evidence and evaluates India's experience so far.
The stock price impact of STT can be in either direction. Empirical evidence from Sweden shows that the imposition of 1% STT in 1983 led to a decline of 5.3% in the Swedish equity market within one month of the introduction of the tax. Stamp tax rate of 0.3% was introduced in China in 1990 and was doubled a few months later. The rate had to be soon revised back to its original level following a substantial fall in the stock market.
Besides, the introduction of STT also impacts volumes of trade by inducing migrations to other untaxed locations or to untaxed/lesser taxed domestic substitutes. The Swedish experience serves as the classic example in support of this argument. A massive migration of Swedish trading volumes to the London Stock Exchange followed the introduction of STT on Swedish brokerage services during 1983-91. Statistics suggest that the trading of Swedish stocks inside Sweden as a percentage of turnover in other offshore exchanges averaged 61% in 1988, falling to 56% in 1990. Moreover, STT on Swedish fixed income instruments caused a migration to other domestic substitutes such as Swedish debentures. Another case is that of Taiwan, where reduction of the tax levied on futures transactions on the Taiwanese futures market from 5 to 2.5 basis points led to a significant reverse migration of trade to Taiwan from Singapore. A few econometric studies have found that stock market trading volume elasticities with respect to transaction costs normally range between -0.5 and -1.7.
Another impact of STT is on volatility. Depending on conditions, it could be either price stabilising or destabilising. One belief is that STT has a positive impact in reducing the activities of short-term noise traders who tend to deviate from the market fundamentals and destabilise prices. In general, however, STT affects all short-term traders in an indiscriminate fashion; hence, its overall impact on price discovery depends on the composition of traders. The tax might, thus, have an adverse effect on informed traders and liquidity providers leading to distortions in pricing as well as overall efficiency. Evidence suggests that price volatility actually increased in some countries post the introduction of STT. For example, in the case of China, volatility increased significantly after the revision in the stamp tax rate from 0.3% to 0.5% in 1997.
Available evidence for India since the imposition of STT corroborates the international experience. For instance, the FII exposure in Nifty futures traded on SGX outpaced that on NSE in the previous year and this has been attributed mainly to tax arbitrage. There is also some evidence available for the migration of volumes from the equity cash segment to substitutable domestic instruments such as options as the latter have a lower STT as compared to the former. For example, turnover (in notional value) in NSE stock as well as index options--both call and put--has increased at a rate higher than the turnover in the NSE cash segment. Total turnover for NSE stock options has increased from approximately R1.6 lakh crore in 2004-05 to over R10 lakh crore in 2010-11 i.e. by more than six times. The corresponding figure for NSE index options has risen by more than 150 times from approximately R1.2 lakh crore in 2004-05 to over R183 lakh crore in 2010-11. Turnover for the NSE cash segment stood at around R35.7 lakh crore in 2010-11, just triple the turnover of R11 lakh crore in 2004-05.
The turnover ratio (turnover/market capitalisation) is used by some to indicate the speculative nature of the market i.e. lower the turnover ratio, lower is the speculation in the market. The turnover ratio for both BSE and NSE, though, has fallen from its level in 2004-05, but much of it could be due to the rise in the denominator i.e. market capitalisation. For example, the turnover ratio in NSE has declined from 71.9 in 2004-05 to 53.4 in 2010-11, approximately 26%, while NSE market capitalization has risen by more than 300% in the same period. NSE market capitalisation increased from around R15.8 lakh crore in 2004-05 to over R67 lakh crore in 2010-11. Hence, the impact of STT on speculation in the case of the Indian security market is ambiguous. Besides this, STT collection (at both BSE and NSE) of approximately R7,000 crore in 2010-11 contributed to a meagre 0.55% of total tax revenue in the same year.
It can be seen that though theoretical evidence on the impact of STT remains inconclusive, empirical evidence largely points toward its adverse effect on trade volumes, liquidity and volatility. Further, actual revenue in most countries is not commensurate with the estimated potential tax base, owing to the various distortions that the tax induces.
A thorough study of the market micro-structure of the security market before mulling such a tax, therefore, becomes necessary. In India's case, one can argue that there has been little or rather no improvement in efficiency levels of the security market as migration of volumes to other international exchanges and to lesser taxed domestic substitutes has sapped liquidity from the equity cash segment. In addition, the impact on speculation and hence, price volatility cannot be ascertained definitely. Apart from this, revenue realisation from STT in India represents a minuscule percentage of total tax revenue. The absence of migration would certainly make for higher revenues, but plugging all arbitrage opportunities for market participants is an onerous regulatory task.
The author is consultant, ICRIER