Working Paper 273

Securities Transaction Tax-

Case study of India

Neha Malik

April 2014

INDIAN COUNCIL FOR RESEARCH ON INTERNATIONAL ECONOMIC RELATIONS

 Contents
1. Introduction ................................................................................................................................... 1
2. Review of the literature on global experience with STT...................................................... 2
2.1 Volume and Revenues .................................................................................................. 2
2.2 Volatility ...................................................................................................................... 3
2.3 Liquidity and Efficiency............................................................................................... 4
3. The Indian Stock Market and STT imposition ...................................................................... 5
3.1 Trends in the Indian stock market (2000-2012) .......................................................... 5
3.2 India’s experience with STT ........................................................................................ 9
3.2.1 Migration of volumes to substitutable assets .............................................................. 9
3.2.2 Migration of volumes to other geographies .............................................................. 10
3.2.3 Revenue Mobilization ................................................................................................ 12
3.2.4 Volatility .................................................................................................................... 12
4. Empirical Analysis: The Impact of STT on Market Returns, Volatility, Volumes
and Efficiency ................................................................................................................... 13
4.1 Impact of STT on volatility in returns and efficiency ..................................................... 13
4.1.1 Impact of STT revisions on Market Efficiency ................................................................ 13
4.1.2 Impact of STT revisions on volatility in returns .............................................................. 15
4.2 Impact on volumes .............................................................................................................. 17
5. Conclusion .......................................................................................................................... 19
References .................................................................................................................................... 21
Appendix 1 ................................................................................................................................... 24
Appendix 2 ................................................................................................................................... 26

i
 List of Tables & Figures

Table 1: Total Turnover in the Cash and F&O segment of NSE and BSE (Rs. crore) .............. 6
Table 2: Delivered Quantity as a percentage of Traded Quantity in the cash segment of NSE
and BSE (2000-12) ....................................................................................................... 7
Table 3: Revised STT rates (2004-12) (per cent) .................................................................... 10
Table 4: Cost of Trading on Indian and International exchanges ............................................ 11
Table 5: STT collection at BSE and NSE (in Rs. cr) .............................................................. 12
Table 6: Volatility of Major Indices (percent) ......................................................................... 13
Table 7: Stock Returns, STT and Market Efficiency ............................................................... 15
Table 8: STT and Volatility in Stock Returns ......................................................................... 17
Table 9: Impact of STT imposition/revisions on shares traded of companies constituting the
CNX Nifty .................................................................................................................. 18

Figure 1: Combined Turnover of Cash and F&O segment of BSE and NSE (Rs. cr) ............... 7
Figure 2: Turnover Ratio (2000-12) (per cent) .......................................................................... 8
Figure 3: Explicit costs (as percentage of total) for cash deliverables (as of 2013) .................. 9
Figure 4: Nifty Futures Trading in SGX as a percentage of Trading in NSE (in terms of
volume and open interest) .......................................................................................... 11

ii
 Abstract

Security Transaction Tax (STT) was introduced in the Indian capital market in 2004. It is a
tax on transaction of equities as well as their derivatives. Despite the reduction in STT over
the years, it constitutes a large percentage (next only to brokerage fee) of the total cost of
trading. The rationale behind STT was to replace the long-term capital gains tax and create a
level playing field for all participants in the stock market. It was also seen as a way to
mobilise additional revenue. Against this backdrop, the paper examines the trends in the
Indian stock market in the past decade and attempts to quantify the impact of STT imposition
and subsequent revisions on volatility and trading volume during Oct 2003-July 2013.
Empirical results show a mixed response of volatility and volume to changes in STT. Even
though STT has contributed to the exchequer, it can be argued that the absence of such a tax
could have added more to economic growth and hence, higher revenues by promoting smooth
operation of the capital market.
_______________
JEL Classification: G10, G14, G17, G18, G19.
Keywords: Security Transaction Tax, Equity Market, Equity Derivatives, Volatility, Trading
Volume, Liquidity.

Author Emails: [email protected]

_________________
Disclaimer:
Opinions and recommendations in the paper are exclusively of the author(s) and not of any
other individual or institution including ICRIER.

iii
 Securities Transaction Tax-Case study of India

Neha Malik

1. Introduction

India’s capital market, which includes both primary and secondary markets, forms a pivotal
part of its financial system. This is reflected in the market capitalisation of listed companies
which was equal to 68.6 percent (of GDP) in 2012.1 The two main stock exchanges in India
are the National Stock Exchange (NSE), and the Bombay Stock Exchange (BSE). Both
permit trade in the equity/cash, as well as derivatives segments. A third exchange, the MCX-
Stock Exchange Limited (MCX-SX) was granted registration in the Cash and Equity
Derivative Segment in July 2012. 2

In the last decade, there was a rise in the cash turnover of both BSE and NSE. The rate of
increase in the cash turnover was however, outpaced by that of the Equity Futures and
Options (F&O) segment in both exchanges. More importantly, there has been a significant
fall in some indicators of liquidity. For instance, the NSE’s Turnover Ratio declined from
over 200 per cent in 2000-01, to 29.2 per cent in 2012-13. The BSE’s Turnover Ratio
dropped from 175 per cent, to 5.9 per cent in the same period.3 Available statistics also
indicate a likely migration of capital to the global trading platforms.

Substitution of asset classes domestically as well as migration of capital to other geographies

can be linked to an array of factors including global and domestic macroeconomic conditions,
rate of return, time differentials, transaction charge differentials, general preference of the
investors for certain asset classes, etc. The present study focuses particularly on the trading
cost differential. Cost of trading is crucial in determining the investors’ choice for an asset
class.

A critical component of the transaction charge in India is the transaction tax, more
specifically, the security transaction tax (STT). STT is the tax on transaction of equities as
well as their derivatives and accounts for a bulk of the transaction cost (at the Indian stock
exchanges) after deducting the brokerage fee. Keynes (1936) was the first to propagate such a
tax.4 Over the years, STT has been implemented (and removed subsequently) by different
countries such as the UK, Sweden, and Japan, with the main objective of mobilising revenue.

1
World Development Indicators 2014, World Bank
2
Handbook of Statistics, 2012, SEBI, April 2013.
3
Although market cap as a percentage of GDP has gone up in both exchanges, turnover ratios have declined
steeply during the period 2000-2012. Advance Estimate of GDP at market prices for 2012-13 (at 2004-05
prices) is considered. Handbook of Statistics on the Indian Securities Market, April 2013, SEBI.
4
Keynes (1936) was one of the earliest proponents of the STT. He strongly condemned speculation and
according to him the market was not run on fundamentals but by, “what average opinion believes average
opinion” of the expected price to be. Matheson (2011).

1
The STT was introduced in India in 2004 to replace the long-term capital gains tax (LCGT).
The rate of taxation has been revised several times since it was imposed and currently stands
at 0.1 per cent for both, buyers and sellers of cash deliverables.5 Despite the reduction in the
rate over the years, it still accounts for a reasonably high proportion of the total trading cost,
as will be discussed later.

One of the main arguments in favour of imposing STT in India is its potential to generate
revenue. The other was that it will create a level playing field for all participants in the stock
market.6 Most empirical literature on global experience with STT points towards an inverse
relationship between transaction tax and trading volumes. However, evidence for its impact
on volatility remains inconclusive.7 While STT has been successful in raising substantial
revenues in some countries, in others, revenues from STT have contributed to a small
percentage of overall GDP.8

The present study examines the impact of STT on different variables of the Indian stock
market, especially on volatility and volumes in the cash segment. Section 2 is an overview of
the available literature on global experience with STT. Section 3 analyses trends in the Indian
stock market over the last decade; imposition of the tax in India has also been discussed. An
empirical analysis of the impact of STT implementation/revisions on volumes and volatility
in the Indian stock market has been provided in Section 4. Section 5 concludes.

2. Review of the literature on global experience with STT

Proponents of STT argue that it reduces volatility in the market by driving away noise
traders/speculators. Opponents, on the other hand, argue that the tax has an adverse impact on
market efficiency by sapping liquidity and raising asset prices. The global experience with
STT has been reviewed by category: impact on stock volumes, volatility, and liquidity.

2.1 Volume and Revenues

A review of the literature suggests that STT leads to a reduction in traded volumes. One
reason for this is that transaction costs go up, making it costlier to trade (Wang and Yau,
2012). In certain cases, trading becomes unviable because STT does not discriminate between
good and bad realizations as opposed to capital gains tax (CGT) (Matheson, 2011). Another
factor that leads to a fall in volume in the taxed region is migration of trade to other untaxed
geographies, thus, establishing an inverse relationship between STT and trading volume
(Wang and Yau, 2012 and Matheson, 2011).

5
STT on equity futures was revised downward from 0.017 per cent to 0.01 per cent in the Union Budget of
2013-14.
6
On account of STT, even foreign investors who enjoyed a tax free return by routing their investments through
countries with which India had double taxation avoidance treaty were taxed on their transactions in the stock
exchanges.
7
Wang and Yau (2012)
8
Matheson (2011)

2
Trades within the French CAC 40 dropped by approximately 16 per cent compared to the
benchmark within 40 days of implementation of French STT on August 1st, 2012 (Haferkorn
and Zimmermann, 2013). Campbell and Froot (1994) show that trading of Swedish stocks
within Sweden as a percentage of their turnover in London, New York, and Stockholm, fell
from 61 percent in 1988, to 52 percent in 1991. Sweden imposed transaction taxes with a levy
of fifty basis points on registered Swedish brokerage services involved with the purchase and
sale of equities in 1984. According to Chou and Wang (2006), a drop in STT from 5 basis
points to 2.5 basis points on the Taiwanese Futures Markets in 2000 led to a reverse
migration of trading volume from Singapore to Taiwan. Available evidence also indicates a
shift to untaxed substitutes. In the case of Sweden it was found that taxed fixed income assets
were substituted for Swedish debentures and variable rate notes (VRNs) following the
introduction of tax on the former (Campbell and Froot, 1994).

Experience with respect to revenue is varied across the globe. In some cases, it represents a
cyclical pattern linked to the financial activity (Matheson, 2011). While calculating the
estimated potential of revenues from such taxes, the possibility of migration of trade volume
is generally not taken into account. Hence, actual revenue mobilized in most cases, does not
correspond with the estimated potential. Most studies suggest that the revenue potential is a
function of the elasticity of trading volume with respect to transaction cost/STT/Bid-ask
spread (BAS) (Wang and Yau, 2012). Various studies have tried to estimate the elasticity of
trading volume with respect to change in transaction cost; such elasticities range between -0.5
and -1.7 (Matheson, 2011). Elasticity of volume with respect to STT was estimated to be -0.5
in case of China (Baltagi et al., 2006) and -1 for the Taiwanese futures market (Chou and
Wang, 2006). Elasticity of the Swedish stock market with respect to total transaction cost was
estimated to be in the range of -0.9 to -1.4 percent (Lindgren and Westlund, 1990). Chou and
Wang (2006) found that reduction in TAIFEX transaction tax rate in 2000 led to an increase
in the revenue collected in the second and third year following the reduction.

2.2 Volatility

On theoretical grounds, the impact of STT on volatility is ambiguous (Kupiec, 1996). There
are two types of volatility: short-term price volatility and long-term price volatility
(Matheson, 2011). From the viewpoint of a transaction tax, short-term price volatility is more
pertinent. There is also a distinction between short-term price volatility and return volatility.
While STT may lower asset price volatility by causing a fall in asset prices, it simultaneously
increases return volatility (Kupiec, 1996).

According to some studies, STT decreases short-term price volatility by reducing the number
of destabilizing speculators (Wang and Yau, 2012). The other view is that the impact of
transaction tax on volatility is a function of the market microstructure and composition of
traders (Song and Zhang, 2005 and Pellizzari and Westerhoff, 2007). For instance, if the
number of fundamentalists9 or liquidity providers far exceeds that of noise traders, then STT

9
Fundamentalists are those who operate on the basis of market fundamentals.

3
will not only affect the latter but also have a disproportionate effect on the former, leading to
a fall in trade volume and hence, liquidity, implying a rise in volatility (Wang and Yau,
2012). Some have also argued that it is the variability in prices or returns on the asset that
represents volatility. The latter, however, is commonly interpreted as velocity of price
changes. Thus, STT may be desirable in some cases since it reduces the velocity of financial
markets (Miller, 1990).10

Empirical evidence, however, remains inconclusive. Roll (1989) found no direct relationship
between volatility and transaction costs while examining stock return volatility across 23
countries, from 1987 to 1989. Umlauf (1993) conducted an ex ante and ex post analysis of
Swedish transaction tax on brokerage services and found no reduction in volatility after the
introduction of the tax.11 An increase in the stamp tax rate from 0.3 per cent to 0.5 per cent in
1997 in China led to a significant rise in volatility as per Baltagi et al. (2006). Green et al.
(2000) concluded that increases in UK stamp duty were generally accompanied by higher
short-term price volatility. Pomeranets and Weaver (2011) examined changes in New York
STT from 1905-1981 and found no statistically significant relationship between STT and
volatility.

Volatility, it is argued, is also sometimes spurred by the process of trading itself (Matheson,
2011; Habermeier and Kirilenko, 2003). French and Roll (1986) examined the variability in
stock returns for all NYSE and AMEX stocks over trading and non-trading hours during
1963-1982. They concluded that volatility in stock returns was higher during the trading
hours.

2.3 Liquidity and Efficiency

The overall impact of STT on the efficiency of stock markets is determined by its effect on
volume, liquidity, asset prices and volatility. The tax, as mentioned earlier, can trigger
migration of trade volumes to other global untaxed/lesser taxed trading platforms. In the
absence of a critical level of liquidity, such migration turns the (taxed) market illiquid,
increasing the cost of trading (Campbell and Froot, 1993). An illiquid market obstructs
smooth flow of information hampering price discovery and lowering efficiency (Wang and
Yau, 2012; Habermeier and Kirilenko, 2003). The tax might also sometimes cause the
withdrawal of essential market participants such as the market makers (Habermeier and
Kirilenko, 2003). Market makers and dealers are desirable in case of stocks that are less
frequently traded. They also help to promote price stability in most cases (Madhavan, 2000).

An increase in volatility also implies a fall in efficiency of the market. Baltagi et al. (2006)
concluded that the Chinese stock market became less efficient due to quick assimilation of
volatility shocks following the increase in STT from 0.3 percent to 0.5 percent in May 1997.
Chou and Lee (2002) found that efficiency in TAIFEX futures market improved significantly
in response to the tax rate reduction in 1986.

10
Culp (2010)
11
The period under consideration is 1980-1987.

4
Transaction tax is also believed to cause a shift in the focus of the management from short-
term to long-term gains (Wang and Yau, 2012) by discouraging trade by short-term traders12
(Schwert and Seguin, 1993). This argument, however, does not support the view that long-
term decisions of the management may, in fact, be a function of the stock prices in the short-
term.

STT is widely believed to cause an increase in the cost of capital (Schwert and Seguin, 1993;
Matheson, 2011).Theoretically, as trading becomes costlier in response to the
introduction/increase in transaction tax/cost, investors demand a higher rate of return to offset
the effect of the former which leads to an increase in the cost of capital for companies
(Matheson, 2011; Kupiec, 1996). This further leads to a fall in the price of the asset (Schwert
and Seguin, 1993).13 Opponents such as Stiglitz (1989), however, argue that the tax induces a
reduction in the cost of capital and therefore, increases asset prices. 14 Empirical evidence
bolsters theoretical findings in this case. Stockholm stock exchanges witnessed a crash of
about 5.3 per cent within a month’s time of the introduction of the Swedish STT in 1983
(Umlauf, 1993). Schwert and Seguin (1993) estimated that an introduction of 0.5 per cent of
STT in the U.S. would be followed by an increase in the cost of capital.

There is, thus, a lack of consensus in theory with respect to the effect of STT on volatility and
efficiency in the stock markets. This is matched by equally inconclusive empirical findings.
As regards the impact on volumes, theoretical research suggestive of migration induced by
transaction tax is corroborated by the empirical results in most cases.

3. The Indian Stock Market and STT imposition

3.1 Trends in the Indian stock market (2000-2012)

The present analysis sheds some light on the changing patterns in volumes and turnover of
the cash and derivatives segments of the NSE and BSE over a period of time. Derivatives
here refer to only equity derivatives. Traded quantity (in lakhs) and number of contracts are
taken as proxies for trading volume in cash, and equity derivatives segments, respectively.15
Other statistics related to liquidity have also been examined.

There was a significant shift in volumes and turnover from cash to the F&O segment in both
exchanges over the last decade. The compound annual growth rate (CAGR) of the turnover in
the F&O segment of both, the NSE and BSE, was over 80 per cent for the period 2001-12.
The corresponding figure for the cash segment was approximately 18 per cent for NSE and a
meagre 8 per cent for BSE (Table 1). NSE accounted for more than 97 per cent of the total
derivatives turnover of the two exchanges for the entire period. At the same time, the

12
This argument was also put forward by Stiglitz (1989).
13
The Present Value of the asset is the discounted value of the asset.
14
It is argued that as transaction tax reduces the activities of the speculators. A reduction in such activities
reduces the amount of risk premium accounted for in the calculation of discount rate in the discounted cash
flow. Hence, it raises the present value of the asset. Culp (2010) and Baltagi, et al. (2006)
15
Please refer to Appendix A for the diagrams.

5
combined cash volume16 (traded quantity) grew at 17.2 per cent on a compound basis, while
the CAGR for the combined volume (total number of contracts) in the F&O segment was
76.1 per cent in the given period.17

Table 1: Total Turnover in the Cash and F&O segment of NSE and BSE (Rs. crore)
NSE BSE
Turnover Cash Total Turnover (F&O) Turnover Cash Total Turnover (F&O)
2001-02 513167 76764 307292 1812
2002-03 617989 339731 314073 2456
2003-04 1099534 1913234 503053 11743
2004-05 1140072 2378195 518715 16111
2005-06 1569558 4643981 816074 9
2006-07 1945287 7162459 956185 59006
2007-08 3551038 12731341 1578857 242308
2008-09 2752023 10781254 1100074 11775
2009-10 4138023 17157601 1378809 234
2010-11 3577410 28217878 1105027 154
2011-12 2810893 30372701 667498 807008
CAGR 18.5% 81.8% 8.06% 84.02%

Source: SEBI (2013)

Notes: a) For derivatives, the turnover is in notional value.
b) Notional Turnover for options= (Strike Price + Premium) * Quantity

More importantly, the combined turnover and volume (NSE and BSE) in the cash segment
show similar trends (Figures (i) and (ii) in Panel A, Appendix 1). Both show an upward trend
from 2002-03, and peak in 2009-10, falling steeply thereafter. An important feature is that the
combined cash turnover of both exchanges from April to December 2012 was almost equal to
the level in 2000-01, approximately Rs. 23 lakh crore (Figure ii), Panel A, Appendix 1). Cash
volume and turnover trends for both exchanges separately are also similar, both trading
volume and turnover increased from 2002-03 arriving at a peak in 2009-10,followed by a
decline in the rest of the period . Within the cash segment, delivered quantity (in absolute
terms) during 2000-12, increased in both, BSE and NSE but delivered quantity as a
percentage of traded quantity grew at a dismal 1.2 and 4.6 per cent (on a compounded basis)18
(Table 2).

16
Sum of traded quantity in BSE and NSE
17
Author’s calculations based on the data given in the Handbook of Statistics 2012, SEBI, April 2013
18
Author’s calculations based on the data provided in Handbook of Statistics, 2012, SEBI, April 2013.

6
Table 2: Delivered Quantity as a percentage of Traded Quantity in the cash segment of
NSE and BSE (2000-12)
BSE NSE
Traded Delivered Percent of Traded Qty Delivered Percent of
Qty (lakh) Qty Delivered Qty (lakh) Qty (lakh) Delivered Qty
(lakh) to Traded Qty to Traded Qty
2000-01 258511 86684 34 304196 50203 17
2001-02 182196 57668 32 274695 59299 22
2002-03 221401 69893 32 365403 82305 23
2003-04 385806 133240 35 704539 174538 25
2004-05 477171 187519 39 787996 201405 26
2005-06 664467 300653 45 818438 226346 28
2006-07 560780 229685 41 850515 238571 28
2007-08 986009 361628 37 1481229 366974 25
2008-09 739601 196630 27 1418928 303299 21
2009-10 1136513 363578 32 2205878 473952 21
2010-11 990776 376890 38 1810910 497367 27
2011-12 654137 255999 39 1605205 443232 28
Source: SEBI (2013)

In the F&O segment, growth in options superseded growth in futures in the latter half of the
period under consideration. Figure 1 shows a comparison of growth in the combined (both
BSE and NSE) turnover of cash, futures and options segment. It can be observed that the
options turnover which was below both cash as well as the futures turnover before 2008-09,
has outpaced the other two since then. For instance, the turnover of all options in 2007-08
was approximately Rs. 17 lakh crore, while that of cash and futures was Rs. 51.3 lakh crore,
and Rs 116 lakh crore, approximately. The corresponding figures for 2011-12 were
approximately Rs. 240 lakh crore, Rs. 34 lakh crore and Rs. 78 lakh crore, respectively. A
similar trend is observed for trading volume of options which was below that of futures
before 2009-10. After this, trading volume of options exceeded futures and also increased at a
higher rate.

Figure 1: Combined Turnover of Cash and F&O segment of BSE and NSE (Rs. cr)

Source: SEBI (2013)

Note: Turnover for both futures and options is in notional value.

7
The equity F&O segment can be further classified into stock and index futures, and stock and
index options. An analysis of trends in turnover and volume of index and stock futures (BSE
and NSE combined) suggests an overall increase in both for the entire period though there
was a fall in the period immediately after 2011-12. While the turnover of stock futures was
higher than that of index futures throughout the period, the traded volume (number of
contracts) for the former coincided with that of the latter after 2008 (Appendix 1, Panel B).
The increase in the options segment has been quite sharp both in terms of number of contracts
and turnover. This has been governed primarily by the steep rise in index options (Appendix
1, Panel B). For instance, the volume of index options in 2011-12 stood at more than 80 crore
(contracts) while that of stock options for the same year was approximately 3.6 crore
(contracts).

The Turnover Ratio,19 which is an important indicator of liquidity in the cash segment,
declined for both stock exchanges over the period 2000-12 (Figure 2). The Average Turnover
Ratio during 2000-05, was 70.5 and 113 per cent for BSE and NSE, respectively. It declined
to 24.23 and 64.3 per cent for the rest of the period under consideration. Trading Frequency,
measured by share of traded companies in the listed companies is another indicator of
liquidity. Trading frequency at the BSE has gone up since 2000. Although there has been a
rise in the number of listed companies on the NSE, the share of traded companies among
those listed has decreased since 2000-01.20

Figure 2 : Turnover Ratio (2000-12) (per cent)

250

200

150

Turnover Ratio-BSE
100
Turnover Ratio-NSE

50

Source: SEBI (2013)

It is therefore, apparent that since 2000, the growth of the equity derivative segment (in both
exchanges) in terms of turnover as well as volume has been remarkable when compared to
that of the cash segment. Within the equity derivatives, there has been an unprecedented

19
Turnover Ratio is defined by Turnover/Market Capitalization
20
Please refer to Table A in Appendix 2

8
increase in the index options since 2007-08, and they now account for the bulk of the equity
derivatives turnover. The increasing dominance of index options has been attributed to
factors such as greater diversification for investors; limited investment by way of option
premiums; structural shifts caused by the financial crisis; and revision in the STT rate
(applicable to options) in June 2008.21 The next section analyses the impact of STT on the
Indian equity market.

3.2 India’s experience with STT

The imposition of STT leads to an increase in transaction costs and makes it costlier to trade.
An increase in the cost of trade has been found to cause a decline in traded volume in most
cases globally.22 Trades executed on Indian stock exchanges are subject to high transaction
costs of which STT constitutes the bulk. It accounts for 54.64 per cent of the total trading
cost after excluding the cost of brokerage (Figure 3).23

Figure 3: Explicit costs (as percentage of total) for cash deliverables (as of 2013)

Explicit costs (% of total) incl. brokerage Explicit costs (% of total) excl. brokerage

Source: Author’s calculations

The global experience with STT (as mentioned in the previous section) has established the
possibility of migration of investment from a taxed asset class to either a substitutable
untaxed/lesser taxed asset class or untaxed geographies.24

3.2.1 Migration of volumes to substitutable assets

A rapid growth in the options turnover of the Indian stock market can, in a sense, be
classified as migration of volumes from cash (the higher taxed segment) to the F&O segment
(the lower taxed segment). This has certainly increased liquidity in the F&O segment but it
can be argued that much of it has been sapped out of the cash market. There are factors that

21
Narain (2011)
22
Wang and Yau (2012)
23
Please also refer to Table B in Appendix 2.
24
Matheson (2011)

9
make equity derivatives seem more lucrative, such as relatively small investments in the form
of margin for futures or premium for options, arbitrage opportunities in the case of single
stock derivatives, greater diversification for investors when compared to investment in single
stocks, etc. The shift can also be attributed to the tax differential between the cash and the
derivatives segment.

Table 3 shows the STT rates applicable to different segments for 2004-12. The highest STT
rate in the Indian stock market applies to cash deliverables. It can be seen that for the entire
period, the F&O segment was subject to a much lower tax rate than that levied on the cash
market. More importantly, growth in the options segment picked up momentum around 2008
when it also outpaced growth in the futures market. This could be traced to changes in the
structure of taxation in the options market. Before 2008, for unexercised options (options that
were squared off), the tax was levied on the aggregate of the notional value of the transaction
and the premium. After the revision in June 2008, the tax applied to only the premium value
for the seller. The buyer is required to pay a tax on the settlement price only if the option is
exercised. It needs to be emphasised however, that exercised options form a negligible
percentage of total options in the Indian stock market.

Table 3: Revised STT rates (2004-12) (per cent)

Date Cash Cash Non Equity Options Exercised
Deliverable Deliverable Futures (sell) Premium Options
(buy and sell) (sell) (sell)
1-Oct-04 0.075 0.015 0.01 NA 0.01
1-Jun-05 0.1 0.02 0.0133 NA 0.0133
1-Jun-06 0.125 0.025 0.017 NA 0.017
1-Jun-08 0.125 0.025 0.017 0.017 0.125
1-Jul-12 0.1 0.025 0.017 0.017 0.125
1-Jun-13 0.1 0.025 0.01 0.017 0.125
Source: SEBI and NSE

3.2.2 Migration of volumes to other geographies

There has also been a case of flight of capital from the Indian securities market to other
international markets. An example of this is the migration of FII flows in NSE Nifty futures
to those traded on Singapore Exchange (SGX). SGX launched Nifty Futures Trading around
2000-01.Trade in Nifty Futures on SGX as a percentage of trade on NSE, both in terms of
volume and open interest has increased over the years (Figure 4) . For instance, open interest
of SGX Nifty Futures as a percentage of open interest of NSE Nifty Futures increased from
approximately 34 per cent in 2008, to 57 per cent in 2012. Similarly, the volume of SGX
Nifty Futures as a percentage of volume of NSE Nifty Futures rose from 4.9 per cent to 17
per cent in the same period. The compounded growth in volume and open interest of SGX
Nifty Futures was approximately 12 per cent and 2.5 percent during 2008-12. The volume
and open interest of NSE Nifty Futures declined by 18.7 percent and 10 per cent on a

10
compounded basis.25 Although factors such as time difference and global macroeconomic
forces also play a role in determining such a transfer, difference in transaction cost is usually
an important consideration for investors. This is, however, an empirical question and can
serve as a subject of further study.

Figure 4: Nifty Futures Trading in SGX as a percentage of Trading in NSE (in terms of
volume and open interest)

2013a

2012

2011
Open interest

2010 Volume

2009

2008

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00

Source: Bloomberg Note: Figures for 2013 are till the month of July

Table 4 shows a comparison of cost of trading at global exchanges vis-a-vis the cost at the
MCX-SX in India. The mandatory cost of trading on the MCX-SX, India is as high as 98
percent of the total cost of trading leaving a small residual component, the exchange cost. For
other exchanges such as the SGX, Singapore, and Bovespa, Brazil, the exchange charges as
opposed to mandatory costs makes up the bulk of the total trading charges.

Table 4: Cost of Trading on Indian and International exchanges

Country Exchange Mandatory Cost as % Exchange Cost as a
of Total Cost % of Total Cost
India MCX-SX 98.54 1.46
Korea KRX 99.10 0.90
Mexico BMV 0.00 100.00
Hong Kong HKEx 95.37 4.63
Spain BME Spanish Ex 0.00 100.00
Thailand SET 0.00 100.00
Singapore SGX 6.54 93.46
Taiwan TAIFEX 98.30 1.70
China Shenzen Stock Exchange 93.61 6.39

25
Please refer to Table C in Appendix 2

11
 Country Exchange Mandatory Cost as % Exchange Cost as a
of Total Cost % of Total Cost
Malaysia Bursa Malaysia 76.92 23.08
Brazil Bovespa 0.00 100.00
Switzerland Swiss SIX 0.00 100.00
Canada TMX 0.00 100.00
Source: Calculations from individual exchange websites
Note: Total Cost = Exchange costs + Mandatory costs. Mandatory costs are inclusive of
Stamp duty and STT

3.2.3 Revenue Mobilization

As regards revenue generation, the average collection from STT since its implementation, has
been in the range of 0.02-0.05 per cent of GDP (Table 5). Though this may be regarded as
significant in some cases, revenue from STT in other Asian countries such as Hong Kong and
Taiwan has been around 2.1 per cent and 0.8 per cent in some years.26 Thus, it can be said
that even though the total collection from STT (in absolute terms) has risen since its
imposition, STT collection as a percentage of GDP has not been as significant as in other
Asian economies.

Table 5: STT collection at BSE and NSE (in Rs. cr)

Cash Cash Non Equity Options Exercised Total Percentage
Deliverable Deliverable Futures Premium Options of GDP
2004-05 316 56 127 0 17 516 0.02
2005-06 1,738 249 573 0 69 2,628 0.08
2006-07 2,814 362 1,185 0 168 4,529 0.11
2007-08 5,178 626 1,974 0 293 8,071 0.18
2008-09 3,510 502 1,201 67 64 5,344 0.1
2009-10 4,871 758 1,552 24 97 7,301 0.12
2010-11 4,653 602 1,675 36 98 7,064 0.1
2011-12$ 1,749 193 656 22 32 2,652 0.06
Source: Pore (2012). Note: $ as on September 2011.

3.2.4 Volatility

STT’s impact on volatility remains ambiguous both, theoretically and empirically. The
average volatility27 of the BSE Sensex and S&P CNX Nifty, the main indices of the BSE and
the NSE, rose in the post STT period (Table 6). This however, cannot be linked solely to STT
imposition. The section on empirical analysis tries to test the actual impact of STT on
volatility in the Indian securities market.

26
SEBI Bulletin, March 2012
27
Volatility is calculated as the standard deviation of the natural log of returns in Indices in the respective
period.

12
Table 6: Volatility of Major Indices (percent)

BSE Sensex CNX Nifty

2000-01 2.2 2
2001-02 1.5 1.4
2002-03 1 1
2003-04 1.4 1.4
2004-05 1.5 1.6
Average 1.52 1.48
2005-06 1 1
2006-07 1.8 1.8
2007-08 1.9 2
2008-09 2.8 2.7
2009-10 1.9 1.9
2010-11 1.1 1.1
2011-12 1.3 1.3
Average 1.69 1.69
Source: SEBI (2013)

4. Empirical Analysis: The Impact of STT on Market Returns, Volatility, Volumes and
Efficiency28

This section aims at estimating the impact of STT imposition and subsequent changes in the
rate29 on volatility and volumes in the Indian stock market.

4.1 Impact of STT on volatility in returns and efficiency

This section analyses the impact of STT imposition/revisions on the Indian stock market
using stock index returns from NSE around 2004, 2005, 2006 and 2012, when there was
either an increase or decrease in the tax rate. The period under consideration is 1st October
2003 to 30th August 2013. The impact of the tax, however, is analysed before and after
revisions in STT rates. For example, the impact of STT introduction in 2004 is assessed for
the period 2nd October 2003 to 30th September 2005, as the date of imposition was 1st October
2004.

4.1.1 Impact of STT revisions on Market Efficiency

The switching GARCH (SGARCH) model, originally proposed by Lee and Ohk (1992) to
test the volatility structure of a stock market, has been used to analyse the impact of STT on
the structure of volatility and hence, efficiency (rate of assimilation of new information) in
the Indian stock market. SGARCH can identify the structural change in return volatility, and
28
This section has been written jointly with Francis Rathinam, Research Adviser, DFID South Asia Research
Hub, New Delhi and Vijay Varadi Economic Analyst, HP. The research was undertaken when both authors
were based at ICRIER.
29
In the cash segment

13
the direction of information assimilation in the conditional variance equation. It allows
regime switching in both the mean equation and the conditional variance equation around the
time of introduction/change of STT. While the dummy in the mean equation captures the
changes in mean level of returns due to changes in STT, the dummy in the conditional
variance equation will capture the regime switching in the autoregressive structure.

Previously, Lee and Ohk (1992), using SGARCH found that listing in stock index futures has
a positive effect on the underlying index in the cash market. Li et al. (1997) used SGARCH
to analyse the impact of switching from same-day settlement to next-day settlement. Baltagi
et al. (2006) used it in the context of introduction of STT in China to show that STT had
reduced market returns and efficiency.

GARCH family models are used to analyse time-dependent volatility in return series as a
function of observed prior volatility.30 The standard GARCH (1,1) as proposed by Bollerslev
(1986) is as follows:

2
Rt t , t N (0, )
2 2 2
t 0 1 ( t 1) 2 ( t 1)

where, σ2t is the conditional variance defined as a function of the last period's squared error
and conditional volatility. The SGARCH, allowing for regime switching in both the
conditional mean and conditional variance equation, could be specified as follows:

Rt 0 1 Dt t , t N (0, t )......... .......... .......... .......... .......... .......... .......... .......... .......... .(1)
2 2 2 2 2
t 0 1 ( t 1) 2 ( t 1) 3 Dt 4 Dt ( t 1) 5 Dt ( t 1) .......... .......... .......... .......... ....... (2)

where, Dt= 0 if t≤ ť and 1 if ť ≤ t. ť is the date of introduction of STT, that is, the regime
switching point.

The null hypothesis is H 0= α 3= α 4= α 5= 0 as the coefficients of the dummy and its

interactive terms will be insignificant and equal to zero if there is no structural change in the
mean level and in the autoregressive structure in equation (2). The changes in market
efficiency could be obtained from the sign and significance of coefficients of Dtε2(t-1) and
Dtσ2(t-1).:A negative 4 and a positive 5 would imply that the effect of the immediate squared
error term is decreasing while the impact of past squared error terms is increasing indicating
that the new information is not absorbed quickly, hence, the market is less efficient than
before (Lee and Ohk, 1992; Li et al.,1997; Baltagi et al. 2006; Su, 2010). While a positive 4

30
See Bollerslev et al. (1992), Bera and Higgins (1993), Enders (2004) and Stock and Watson (2011) for a
survey of various extension of GARCH models and their applications.

14
and a negative 5 would imply that the market is more efficient, the same sign for both
coefficients would not reveal anything about the change in efficiency.31

The SGARCH model estimates regime switching in both, the conditional mean and
conditional variance equations. Regime switching in the mean equation indicates changes in
mean returns, while switching conditional variance equation indicates volatility persistence
and structural changes in information assimilation. In the mean equation (Equation 1), μ1
captures the impact of introduction of/revision in STT on mean returns. Results (Table 7)
show that increases in STT in 2005 and 2006 resulted in a reduction in average returns while
the introduction of STT in 2004 had a positive impact. Reduction in STT in 2012 improved
mean returns.

A negative 4 and a positive 5 for the year 2005, when STT was increased from 0.075 per
cent to 0.1 per cent, shows that the increase in STT negatively affected market efficiency,
that is, the assimilation of information. For the introduction of STT in 2004 and its revision in
2006, the impact on market efficiency is unclear. The same holds true for the revision in
2012.

Table 7: Stock Returns, STT and Market Efficiency

1-Oct -2004 1-Jun-2005 1-Jun-2006 1-Jul-2012

μ0 0.071 (59.82) 0.070 (363.40) 0.09 (62.44) 0.002 (201.88)
μ1 0.002 (175.15) -0.003 (214.70) -0.002 (42.79) 0.002 (219.64)
α 0 (×1 0 −6 ) 3.35 (6.82) 1.77 (13.17) 3.76 (1.55) 3.51 (2443.60)
α1 0.149 (5.29) 0.149 (8.69) 0.149 (2.98) 0.149 (13.06)
α2 0.599 (10.11) 0.599 (19.19) 0.599 (2.70) 0.599 (60.85)
α3 (×10 −6 ) -1.51 (-3.01) -1.42 (-10.63) -2.05 (-0.85) -2.29 (-662.11)
α4 (×10 −6 ) -1.35 (-116.19) -2.98 (-366.22) -1.45 (-112.97) -1.17 (-331.55)
α5 (×10 −5 ) -4.04 (-0.01) 0.99 (5.78) -0.125 (-0.03) 0.121 (0.09)
L 13442.65 13926.55 13291.41 13416.48

Notes: The z-statistic is provided in parentheses and L is Log likelihood.

4.1.2 Impact of STT revisions on volatility in returns

The second model assesses the impact of STT on returns volatility. The component GARCH
(C-GARCH) is used to decompose the short- and long-run volatility (Engle and Lee, 1993),
while the threshold GARCH (T-GARCH) is used to assess the asymmetric impact of positive
and negative information flow on volatility (Glosten et al., 1993). The analysis in this section
is on the lines of the analysis in Liau et al. (2012) which combines the component and
threshold GARCH models to estimate the long- and short-run effect of STT on returns
volatility allowing for a differential impact for negative shocks. The Asymmetric Component
GARCH (AC-GARCH) model is specified as,

31
See Baltagi et al. (2006) for a discussion.

15
Rt R , t N (0, 2 )......... .......... .......... .......... .......... .......... .......... .......... ....( 3)
1 (t 1) t
2
qt ( 2 q ) ( 2 q )d ( 2 q )......... .......... ...( 4)
t (t 1) (t 1) (t 1) (t 1) (t 1) (t 1) (t 1)
q q ( 2 2 )......... .......... .......... .......... .......... .......... .......... .......... ....( 5)
t (t 1) (t 1) (t 1)

Where, Rt and R(t-1) are stock index returns in period, t, and their lagged values, respectively,
and εt is the error term representing contemporaneous shocks. qt is the mean reversing
variance which is a constant. Asymmetric shock is captured by the dummy d(t-1) where the
dummy is equal to one if the error term is negative, that is, ε(t-1)<0, and zero if the shocks are
positive: the positive shocks have an impact equal to μ while the negative shocks have an
impact of α + γ in Equation 4. γ captures the transitory leverage effect, that is, asymmetric
effect in the variance. The fourth equation also determines the speed of mean reversion by a
factor α + 0.5γ + β as discussed in Hadsell (2006) and Liau et al. (2012). While equation 4
captures short-term transitory variance, the last equation captures the long-term permanent
variance.

AC-GARCH, captures short- and long-term changes in volatility due to STT. As Table 8
suggests, γ that captures the transitory asymmetric volatility is possessive, and marginally
increased as tax was increased during 2005 and 2006 vis-à-vis earlier lower tax regimes,
while the tax reduction in 2012 had a substantial positive effect on volatility reduction
indicating that market returns during low tax regimes are less sensitive to negative shocks.
The transitory volatility as captured by α + 0.5γ + β is mixed: when tax rate was increased
from 0.1 per cent to 0.125 per cent in 2006, transitory volatility increased indicating that
volatility is larger in the new increased tax regime, while an increment in 2005 shows a
decline in volatility. However, the last reduction in tax (2012) is associated with a decline in
volatility.

A larger ρ implies higher long-term volatility. The results, however, are mixed as tax
increment in 2005 indicates an increase in volatility, while a further increment in 2006
suggests a marginal decline in volatility when compared to the level in 2005. Notably,
reduction in tax in 2012 spurred a rise in permanent volatility relative to the level in 2006.

16
Table 8: STT and Volatility in Stock Returns

1-Oct-2004 1-Jun-2005 1-Jun-2006

1-Jul-2012
μ 0.001 (0.00) 0.001 (0.00) 0.001 (0.00)
0.001 (0.00)
μ1 0.132 (0.02) 0.131 (0.02) 0.129 (0.02)
0.121 (0.02)
q 0.001 (0.00) 0.000 (0.00) 0.000 (0.00)
0.000 (0.00)
α 0.978 (0.01) 0.980 (0.01) 0.975 (0.01)
0.984 (0.01)
γ 0.034 (0.02) 0.036 (0.02) 0.036 (0.02)
0.014 (0.01)
β 0.134 (0.02) 0.097 (0.02) 0.115 (0.02)
0.085 (0.02)
ω 0.003 (0.01) -0.104 (0.03) -0.117 (0.03)
-0.090 (0.03)
ρ 0.001 (0.01) 0.196 (0.04) 0.187 (0.04)
0.205 (0.04)
δ -0.970 (0.05) 0.772 (0.09) 0.769 (0.10)
0.764 (0.08)
L 8036.516 8047.203 8048.953
8043.911
α+0.5γ+β 0.62 0.58 0.60
0.57
Notes: The t-statistic is shown in parentheses. L is Log likelihood. α+ 0.5 γ + β captures
transitory volatility

4.2 Impact on volumes

This section tries to estimate the impact of STT imposition and revisions on traded volume of
the companies constituting the CNX Nifty Index. We use the switching first order
autocorrelation model to examine the effect during the time period, Oct 2003-July 2013.

Sinha and Mathur (2012b) used a similar model to evaluate the impact of an increase in STT
in 2006 on quantity traded, returns and volatility in returns on both BSE and NSE stocks.
They considered the time period from June 1, 2005 to May 31, 2007, a window of one year
before and after the revision in tax. Their data set comprised 302 out of 500 companies
forming the CNX 500 Index (NSE). The chosen companies were classified into four sub-
categories: NSEALL, NSELARGE, NSEMEDIUM and NSESMALL. The Switching First
Order Autocorrelation Model was used to study the impact of STT change on the traded
shares of these companies. Results were significant for NSELARGE and NSEMEDIUM
suggesting an inverse relationship between upward revision in STT and shares traded of large
and medium cap companies. Results for NSESMALL were also found to be statistically
significant but the sign of the coefficient in this case was positive, indicating a switch in the
volume of trade from large- and medium-sized to a small-sized stock portfolio.32

We examine the impact of STT imposition and all subsequent revisions33. Though we use the
daily data from Oct 2003 to July 2013, we conduct a one-year event study to test the impact
of imposition/revision, as in the section on volatility. For instance, daily data from Oct 3
2003- Sept 30 2005 has been used to evaluate the impact of STT imposition in October 2004.
Shares traded of the companies constituting the CNX Nifty Index have been taken as a proxy

32
For additional details, please refer to Sinha and Mathur (2012b).
33
In the cash segment

17
for traded volume. We use the CNX Nifty Index since it is the benchmark index for Indian
equity markets and represents overall market conditions. Data on shares traded was checked
for stationarity using the Phillips-Perron test. There was no presence of unit root. Equation
(6) specifies the model used in this analysis.

Volt = C + α1 Volt-1 + α2 Dt*Volt-1 + εt ………………………………………………………………………..……(6)

where,Volt and Volt-1 represent aggregate shares traded of companies comprising the CNX
Nifty Index in period t and t-1, respectively. Dt is the dummy variable which takes the value
of 0 if t≤ t*, and 1 if t*≤ t where t* is the date of introduction/revision in STT. Table 9 shows
the results for the impact of STT imposition and each revision.

Table 9: Impact of STT imposition/revisions on shares traded of companies constituting

the CNX Nifty

2004 2005
Constant α1 α2 Constant α1 α2
Coefficient 34554956 0.70* -0.17* 25304554 0.72* -0.04**
t-Statistic 10.07 23.58 -6.56 8.92 22.41 -1.77
Prob. 0.00 0.00 0.00 0.00 0.00 0.08
R-squared 0.65 0.50
Adj. R-squared 0.64 0.50
2006 2012
Constant α1 α2 Constant α1 α2

Coefficient 35474903 0.57* -0.03 78233479 0.52* -0.09*

t-Statistic 11.47 15.19 -1.57 12.61 13.78 -3.81
Prob. 0.00 0.00 0.12 0.00 0.00 0.00
R-squared 0.32 0.29
Adj. R-squared 0.32 0.29
Source: Author’s calculations. Note: * indicates significance at all levels. ** indicates significance at
10 per cent level only.

The results suggest a significant drop in shares traded of CNX Nifty companies after STT
imposition in 2004 (at all levels of significance) and an upward revision in 2005 (at 10 per
cent level only). As for the 2006 revision, the results indicate a decline in volumes but not a
significant one. Results for the downward revision of STT in 2012 are contradictory as they
indicate a significant fall (instead of an expected rise) in traded shares.

The model, however, is parsimonious and has its limitations. It does not control for other
relevant variables which can have a bearing on the trading volume, such as volatility in
returns, bid-ask spread, domestic/global macroeconomic fundamentals, investors’ preferences
etc. Data on companies constituting S&P BSE Sensex can also be included. Since, daily data

18
on the BSE Sensex does not provide information on aggregate traded shares of its constituent
companies, conducting a similar exercise on daily data for each company included in the
index was out of the scope of this study.

5. Conclusion

India’s stock market has grown tremendously over the years in terms of market capitalisation.
Turnover and volumes in both, the cash and equity derivatives segments of the two main
stock exchanges of India have risen persistently over the last decade. The rate of increase in
the derivatives segment, however, has been much higher than that of the cash segment. More
importantly, there has been an unprecedented rise in the turnover of index options since 2008.

A whole host of factors such as arbitrage opportunities, rate of return, transaction charges and
tax differentials can trigger the substitution of one financial asset class for another. Equity
derivatives are highly leveraged and facilitate risk management and are thus, a preferred asset
class for the hedgers. The uneven tax structure in different verticals of the Indian equity
market is also to some extent, responsible for spurring the migration of volumes from the
higher taxed segment (cash) to the one that has a lower tax rate (derivatives). There has also
been substantial flight of capital to the lesser taxed geographies, such as the spurt in FII
investment in Nifty Futures traded on SGX vis-a-vis those traded on NSE during the latter
half of 2001-12. A comparison of trading costs at some of the global exchanges suggests that
the mandatory costs component (inclusive of stamp duty and STT) of the total transaction
cost in India is significantly higher than that in other exchanges. There is, however, a need to
undertake further empirical research to discern the impact of transaction costs (particularly
taxes) on migration of capital to substitutable asset classes as well as other geographies.

Our empirical results show a mixed response of volatility and volume to changes in STT.
Results of the SGARCH model suggest that while upward revisions in STT in 2005 and 2006
caused a reduction in mean returns, the imposition of STT in 2004 led to an increase in them.
AC-GARCH results show that market returns during low-tax regimes are less sensitive to
negative shocks. The impact of STT on market efficiency and long-term volatility is unclear.
As regards the impact of STT on volumes, results of our one year event study suggest a
significant fall in the aggregate shares traded of the companies comprising CNX Nifty after
STT was imposed in 2004. Results for upward revisions in 2005 and 2006 also suggest a fall
in volumes, though not a significant one.34 A downward revision in 2012 also, surprisingly,
indicates a fall in traded volume.

Despite the reduction in STT over the years, it still constitutes a large percentage (next only
to brokerage fee) of the total cost of trading. Lowering transaction costs by reducing taxes,
especially STT, which forms the bulk of transaction costs, may help induce additional
liquidity in both cash as well as derivative markets and smoothen the process of trading. It
may also stem migration of capital to other markets although the efficacy of this effect needs

34
Results for 2005 revision are significant at the 10 per cent level.

19
to be studied empirically. A more liquid cash market is always preferred, from a growth
perspective, since higher liquidity not only facilitates trade and investment but also ensures
efficient allocation of capital to the real sector. Though, liquidity in the derivative segment
has grown faster compared to that in the cash segment, the former is nonetheless at a
disadvantage compared to its global counterparts because of the higher relative transaction
costs.

A large percentage of the savings of the Indian household sector are in fixed income
instruments, especially deposits. The securities market comprises of a negligible proportion
of such savings. Impediments such as high transaction charges and taxes in the securities
market may further incentivize other avenues of investment. A thorough research of the
market micro-structure and composition of traders in the few main segments, therefore, may
be necessary before studying this issue in further detail. Revenue realisation from STT in
India represents only a small percentage of GDP when compared with other Asian countries.
Even though STT has contributed to the exchequer, it can be argued that the absence of such
a tax could have added more to economic growth and hence, higher revenues by promoting
smooth operation of the capital market.

20
References

Baltagi, B.H., Li, D., and Li, Q., (2006) “Transaction Tax and Stock Market
Behaviour: Evidence from an Emerging Market”, Empirical Economics 31, 393-408.

Campbell, J.Y. and Froot, K.A., (1993) “International Experiences with Securities
Transaction Taxes”, NBER Working Paper No. 4587, December.

Chou, R.K. and Lee, J. (2002) “The Relative Efficiencies of Price Execution
between theSingapore Exchange and theTaiwan Futures Exchange”, The Journal of Futures
Markets 22, 173-196.

Culp, C.L., (2010) “Financial Transaction Taxes: Benefits and Costs”, University of
Chicago - Booth School of Business; Compass Lexecon, March 16.

Green, C.J., Maggioni, P., Murinde V., (2000) “Regulatory lessons for emerging
stock markets from a century of evidence on transactions costs and share price volatility in
the London Stock Exchange”, Journal of Banking and Finance 24, 577-601.

SEBI (2013) Handbook of Statistics on Indian Securities Market 2012, April 2013.
Securities and Exchange Board of India, Mumbai, India

Habermeier, K. and Kirilenko, A.A., (2003) “Securities Transaction Taxes and

Financial Markets”, IMF Staff Papers, Vol. 50. International Monetary Fund, Washington
D.C., USA.

Haferkorn, M. and Zimmermann, K., (2013) “Securities Transaction Tax and

Market Quality- The Case of France”, EFMA 2013 Meetings (Reading) & 40th EFA Annual
Meeting (Cambridge), January 4.

Keynes, J.M., (1936), “General Theory of Employment, Interest Rates and Money”
(New York: Harcourt Brace and World).

Kupiec, P., (1996) “Noise Traders, Excess Volatility, and a Securities Transactions
Tax”, Journal of Financial Services Research 10, no.2, p.115.

Lee, S.B., and Ohk, K.Y., (1992) “Stock index futures listing and structural change in
time-varying volatility”, Journal of Futures Markets12, Issue 5, 493-509, October.

Liau, Y., Wu, Y. and Hsu, H., (2012) “Transaction tax and market volatility:
Evidence from the Taiwan futures market”, Journal of Applied Finance and Banking2, no.2,
45-58.

21
 Lindgren, R. and Westlund, A. (1990) “Transaction Costs, Trading Volume, and
Price Volatility on the Stockholm Stock Exchange”, Skandinaviska Enskilda Banken
Quarterly Review 2, 30-35.

NSE (2012) Indian Securities Market - A Review, Volume XV 2012, NSE. National
Stock Exchange, Mumbai, India

Madhavan, A. (2000), “Market microstructure: A survey”, Journal of Financial

Markets 3, Issue 3, 205-258

Matheson, T., (2011) “Taxing Financial Transactions: Issues and Evidence”, IMF
Working Paper, March, International Monetary Fund, Washington D.C., USA.

Miller, M., (1990) “Index Arbitrage and Volatility”, Financial Analysts Journal 46,
No.4.

Mohanty, N., (2011) “Cost of Trading in Stock Exchanges: a Perspective”, NSE

Working Paper, November, National Stock Exchange, Mumbai, India.

Narain, “After effects of global financial crisis on Indian derivatives market”,

University of Delhi, Faculty of Management Studies, August 2011.

Pomeranets, A. and Weaver, D., (2011) “Security Transaction Taxes and Market
Quality”, Bank of Canada Working Paper 2011-26, November.

Pellizzari, P., and Westerhoff, F., (2007) “Some Effects of Transaction Taxes Under
Different Microstructures,” University of Technology Sydney Quantitative Finance Research
Centre Paper 212.

Pore, L., (2012) “Analysis on Securities Transaction Tax (STT)”, SEBI Bulletin,
March, 201-208.

Roll, R. (1989), “Price Volatility, international market links, and their implications
for regulatory policies”, Journal of Financial Services Research, Issue 2-3, Vol. 3, 211-246,
December 1989.

Schwert, G.W., and Seguin, P.J., (1993) “Security Transaction Taxes: An Overview
of Costs, Benefits and Unresolved Questions”, Financial Analysts Journal49, No.5, 27-35.
Sinha, P. and Mathur, K., (2012a) “Evolution of security transaction tax in India”,
Munich Personal RePEc Archive, June 30, Faculty of Management Studies, Delhi University.

Sinha, P. and Mathur, K., (2012b) “Securities transaction tax and the stock market-
an Indian experience”, Munich Personal RePEc Archive, November 20, Faculty of
Management Studies, Delhi University.

22
 Song, F., and J. Zhang, 2005, “Securities Transaction Tax and Market Volatility,”
Economic Journal 115, 1103–20.

Stiglitz, J.E., (1989) “Using Tax Policy to Curb Speculative Short-Term Trading”,
Journal of Financial Services Research 3: 101-115.

Umlauf, S.R., (1993) “Transaction Taxes and the Behaviour of the Swedish Stock
Market”, Journal of Financial Economics 33, no. 2, 227-40.

Wang, G.H.K. and Yau, J., (2012) “Would a Financial Transaction Tax Affect
Financial Market Activity? Insights from Futures Markets”, Policy Analysis, No. 702, July 9.

World Development Indicators, 2014, World Bank.

23
Appendix 1

Panel A Turnover and Volume in the Cash Segment of BSE and NSE (2000-12)

i) Combined Turnover of the Cash Segment of BSE and NSE (Rs. Cr) ii) Combined Traded Quantity in the Cash Segment of BSE and NSE (Lakh)

iii) Traded Quantity of BSE & NSE (Lakh) iv) Turnover of BSE and NSE (Rs. Cr)

Source: SEBI. Note: The dashed line represents NSE. The smooth curves represent trend lines. Source: SEBI. Note: The dashed line represents NSE. The smooth curves represent trend lines.

24
Panel B Turnover and Volume in the F&O segment of BSE and NSE (2001-12)

i) Turnover for Futures (BSE and NSE combined) (Rs. Cr) ii) No. of contracts Futures (BSE and NSE combined)

iii) Turnover for Options (BSE and NSE combined) (Rs. Cr) iv) No. of contracts Options (BSE and NSE combined

Source: Handbook of Statistics on Indian Securities Market 2012, April 2013

25
Appendix 2

Table A: Trading Frequency at BSE and NSE (Percent of traded companies to listed companies)

180

160

140

120

100

80 BSE
60 NSE
40

20

Source: Handbook of Statistics 2012, SEBI 2013.

Table B: Approximate cost of cash based delivery transactions valued at Rs 1 lakh (as of 2013) 35

Cost Percent of Cost Percent of

(incl. Total Cost (exc. Total Cost
brokerage) (incl. brokerage) Brokerage) (excl. brokerage)
User Charges 323 62.11 23 12.57
of which:
Brokerage (at the rate of 30 bps) 300 57.68 0 0
Exchange Transaction Charges 3 0.58 3 1.64
DP charges 20 3.85 20 10.93
Statutory levies 147.08 28.28 110 60.11
of which:
STT 100 19.23 100 54.64
Service tax on brokerage 37.08 7.13 0 0
Stamp duty 10 1.92 10 5.46
SEBI turnover fee 0 0 0 0
Impact cost 50 9.61 50 27.32
Total 520.08 100.00 183 100.00
Source: Author’s calculations

35
The methodology in Table 2 has been borrowed from Mohanty (2011). Figures for brokerage cost, exchange
transaction charges and SEBI turnover fee have been taken from Table 4, pg.206, SEBI Bulletin, March 2012.
Figures for service tax and impact cost have been taken from NSE’s website. Monthly Impact cost of CNX Nifty
from Jan-Oct 2013 has been averaged to arrive at the figure provided in Table 2.

26
Table C: Nifty Futures Trading in SGX and NSE

Year SGX Nifty Futures NSE Nifty Futures

Volume Open Interest Volume Open Interest

2008 36677 260165 741874 755599

2009 17128 140024 686597 637245
2010 38791 196297 509495 637420
2011 54685 252321 498515 534748
2012 57553 286562 324041 497241
2013 56981 306673 247886 392140
CAGR till 2012 11.92 2.45 -18.70 -9.93
Source: Bloomberg Note: Figures for 2013 are till the month of July

27
 LATEST ICRIER’S WORKING PAPERS

NO. TITLE Author YEAR

272 IMPACT OF TRANSACTION TAXES SAON RAY MARCH 2014

ON COMMODITY DERIVATIVES NEHA MALIK
TRADING IN INDIA
271 FEEDSTOCK FOR THE SAON RAY FEBRUARY
PETROCHEMICAL INDUSTRY AMRITA GOLDAR 2014
SWATI SALUJA
270 USING IPRS TO PROTECT NICHES? SUPARNA JANUARY
EVIDENCE FROM THE INDIAN KARMAKAR MEENU 2014
TEXTILE AND APPAREL INDUSTRY TEWARI
269 MONSOON 2013: ESTIMATING THE Ashok Gulati December 2013
IMPACT ON AGRICULTURE Shweta Saini
Surbhi Jain
268 REMOTENESS AND UNBALANCED SAMARJIT DAS SEPTEMBER
GROWTH: CHETAN GHATE 2013
UNDERSTANDING DIVERGENCE PETER E.
ACROSS ROBERTSON
INDIAN DISTRICTS
267 NORMALIZING INDIA PAKISTAN NISHA TANEJA SEPTEMBER
TRADE MISHITA MEHRA 2013
PRITHVIJIT
MUKHERJEE
SAMRIDHI BIMAL
ISHA DAYAL
266 RECESSION AND CHILD LABOR: SAHANA ROY APRIL 2013
A THEORETICAL ANALYSIS CHOWDHURY

265 IMPACT OF MACRO-ECONOMIC SAPTARSHI MARCH

ENVIRONMENT ON PURKAYASTHA 2013
DIVERSIFICATIONPERFORMANCE
RELATIONSHIP:
A CROSS COUNTRY STUDY OF
INDIA AND JAPAN
264 FACTOR INCOME TAXATION, MONISANKAR MARCH 2013
GROWTH, AND INVESTMENT BISHNU, CHETAN
SPECIFIC TECHNOLOGICAL CHANGE GHATE AND
PAWAN
GOPALAKRISHNAN

28
 About ICRIER

Established in August 1981, ICRIER is an autonomous, policy-oriented, not-for-profit, economic

policy think tank. ICRIER's main focus is to enhance the knowledge content of policy making by
undertaking analytical research that is targeted at informing India's policy makers and also at
improving the interface with the global economy. ICRIER's office is located in the institutional
complex of India Habitat Centre, New Delhi.

ICRIER's Board of Governors includes leading academicians, policymakers, and representatives

from the private sector. Dr. Isher Ahluwalia is ICRIER's chairperson. Dr. Rajat Kathuria is
Director and Chief Executive.

ICRIER conducts thematic research in the following seven thrust areas:

Macro-economic Management in an Open Economy

Trade, Openness, Restructuring and Competitiveness

Financial Sector Liberalisation and Regulation

WTO-related Issues

Regional Economic Co-operation with Focus on South Asia

Strategic Aspects of India's International Economic Relations

Environment and Climate Change

To effectively disseminate research findings, ICRIER organises workshops, seminars and

conferences to bring together academicians, policymakers, representatives from industry and
media to create a more informed understanding on issues of major policy interest. ICRIER
routinely invites distinguished scholars and policymakers from around the world to deliver
public lectures and give seminars on economic themes of interest to contemporary India.

29