A Report On The Indian Exchange Traded Funds (ETF) Industry
A Report On The Indian Exchange Traded Funds (ETF) Industry
A Report On The Indian Exchange Traded Funds (ETF) Industry
Mission
“To enhance member value by providing professional and career
development opportunities, promoting highest standards of ethics &
professional conduct and being a thought leader of the investment
profession for the ultimate benefit of society”
V. Recommendations11
a. Improve Investor Awareness11
b. Improve ETF Ecosystem12
c. Remove Stamp Duty on ETFs13
d. Continue Using ETFs as Vehicles for Divestments13
VI. Conclusion:13
Bibliography17
1
Although we make a few comments on fixed income ETFs, we focus a large part of the report on equities.
2
In India, some of the strategies are difficult absent a robust short-selling market.
The chart shows the Nifty TR index since 2001. The major events around ETFs are represented as callouts.
3
Surprisingly, most statistics including SEBI’s handbook of statistics continue to categorize equity ETFs under “other ETFs”,
a relic of an era where Gold ETFs were dominant.
Chart 2a.2: Indian ETF Assets Under Management Across Gold and Equities
The next major fillip came in the middle of 2015, when Employees Provident Fund Organization
(EPFO) took tentative steps towards investing in stock markets via an ETF route, starting
with a modest 5% of incremental flows, which were subsequently hiked to 15% of incremental
flows in 2017.
ETFs are regulated under SEBI’s mutual fund regulations 1996. However, specific diversification
norms for index funds and ETFs were circulated in January 2019. These norms were modest,
mandating a minimum 10 stocks in an ETF, restricting the individual stock weight to 25% (or
35% for a thematic ETF) and top 3 weight to 65%. There were other rules aimed at liquidity.
Lastly, India launched the first corporate bond ETF in December 2019, following a series of
policy moves, including the cabinet approval, SEBI’s norms for debt ETFs, and the reserve
bank tweaking its norms to allow debt ETFs to be eligible as collateral for repo transactions.
For a description of how ETFs evolved globally, please refer to Appendix A.
Chart 2b.1: Active vs Passive Equity funds - Assets under Management and Net
flows
US: AUM Across Active and US: Net Fund Flows Across
Passive Funds Active and Passive Funds
(Total=$8.5tn) ($bn)
300
200
100
0
49.9% 2017 2018 2019-YTD
-100
50.1% -200
-300
In primary market, ETF shares are created and redeemed in pre-specified number of units
(basket) by Authorized participants, which are firms that have agreements with the ETF
manager or distributor to create and redeem ETF shares at the NAV price at the end of each
day. If market prices deviate sufficiently away from the net asset value of the underlying basket,
the AP can buy (sell) the “mispriced” ETF in the open market and redeem (create) it at NAV
with the issuer. With this creation and redemption mechanism, ETFs operate like mutual
funds, with the difference that only AP could create and redeem a pre-specified basket at NAV.
Lastly, in addition to creation and redemption mechanism, APs have a range of tools to ensure
ETF prices are in line with the NAV. For example, APs could go long (short) the mispriced
ETF and short (long) other ETFs tracking the same index; APs could also create arbitrage
trades using futures and options tracking the index, orshort the underlying basket of stocks /
ETFs.
Like stocks, ETFs are traded in secondary markets with bid and offer. The factors that
determine the width of a bid-ask spread are the amount of ongoing order flow, the amount
of competition among market makers for that ETF, and the actual costs and risks associated
with an AP doing the creation/redemption process, and the expected profit margin. Although
it works the same way in India, given the low order flows and continuous two-way quotes,
liquidity is a challenging issue we’ll discuss later.
There are minor differences in ETF market structure between India and rest of the world.
Some of them are beneficial; for example, ETFs generate capital gains during rebalancing or
creation / redemption process. ETF sponsors globally are liable to pay capital gains taxes; in
India such capital gains are not taxed. However, other market structure differences, such as
direct redemption feature, may’ve unintended consequences. We will explore these issues in
Section IV.
With the background on ETF evolution as well as how they work, we’ll turn our attention to
the tailwinds and headwinds impacting the Indian ETF industry.
n Policy moves in recent years have provided a fillip to the passive industry. The moves were
aimed at making ETFs a favoured vehicle for asset sales for the government, providing a
truer picture of active fund performance vis-à-vis benchmark, providing better disclosures
on fees, as well as incentivizing fee-based advice.
n Government’s choice of ETF as investment vehicle for divestments for Central Public
Sector Enterprises (CPSE), as well as choice of investment by EPFO since 2015 have
increased awareness of retail investors in ETFs. Any possible future asset sales by the
government would also incentivize policymakers to develop ETF ecosystem further.
n In late 2017, SEBI came out with a set of 10 equity categories with precise definitions,
and mandated fund houses to have only one scheme per category (SEBI, 2017). The
categorization reduced the scope of mutual funds to stray outside the stated mandate - for
example, a large cap active fund benchmarked to Nifty investing a meaningful portion
of its assets in small- and mid-cap stocks, and potentially showing an outperformance.
Limiting the fund houses to only one scheme per category reduced the noise, and
potential for fund houses to cherry-pick the best schemes. The policy changes forced
greater transparency into the active fund industry, and a more level-playing field with the
passive industry.
n In 2018, SEBI asked fund houses to benchmark returns of equity schemes against a total
return index (TRI) instead of price return index. Until then, the mutual funds used to
benchmark their total return (including dividends and after fees) with Nifty price index
(excluding the impact of dividends4), providing a misleading picture of active fund
performance.
n Since SEBI came out with the investment advisor regulations in 2013, there has been
increasing awareness about fee-based model vis-à-vis traditional distribution model.
While nascent, fee-based Registered Investment Advisors (RIAs) who are held to a higher
standard of care compared to distributors, should help steer investors towards low cost and
passive funds over the long term.
4
For reference, the dividend yield of Nifty Index was 1.31% at the end of September 2019.
1. Equity ETF: Minimum 10 stocks; Max Single stock exposure capped at 25% weight;
Weight of top 3 stocks capped at 65%.
2. Debt ETF: Minimum 8 issues; Single exposure issuer capped at 15%; Only investment
grade.
In theory, defining such outer boundary for portfolio construction by regulations helps
spur product innovation. However, in practice, the resultant portfolios that adhere to these
norms could still be very concentrated. For example, CPSE ETFs comprise of 10 stocks
dominated by energy and materials sectors, with the top 4 stocks accounting for nearly
80% of portfolio weight. The allocation to CPSE ETF by Employee Provident Fund
Organization (EPFO), which has otherwise been conservative with its equity allocation, is
perceived as problematic.
IV.b. Liquidity
The liquidity of most of the Indian ETFs is modest. While US ETF flows are driven by both
advisor as well as institutional flows, over 90% of Indian ETFs are owned by institutions,
mainly the Employee Provident Fund Organization (EPFO).
The dominance of one-way flows results in poor liquidity. The most liquid ETF in India,
NiftyBees had less than one-hundredth of the daily traded value of Reliance Industries, a large
liquid stock. In contrast the most traded US ETF, SPDR S&P 500 ETF had 3.5 times the
daily trading volume of Apple5. While ETF liquidity is not a sole function of trading volumes,
5
Based on our calculations between Jun 2019 and Sep 2019.
the popularity of individual stocks vis-à-vis ETFs speaks volumes about the lack of market
maturity in India. The Median of the daily Bid-Ask Spread on the largest ETF is as high as
17 bps, which that of individual large cap stocks is less than 5 bps. In most geographies it works
the other way around.
To assess popularity, we analysed google search trends on mutual funds vs ETFs. The chart
below shows that globally ETFs are popular compared to mutual funds; for a recent period,
the number of searches on ETFs were 4 times those of mutual funds. In India however, mutual
funds are far more popular than ETFs, although relative interest have narrowed in recent
times. Interestingly, the spikes in interest have coincided with the government divestments in
public sector enterprises, a trend that is likely to continue.
The liquidity problem manifests in two ways. Despite the efficiencies offered by the ETF
structure – in-kind transactions and fairness in allocating costs of trading, Indian ETFs are
perceived as poorer alternatives to index funds tracking the same index. To compare the two
structures, we looked at the share-class returns of ETFs and index funds tracking the popular
50-stock S&P CNX Nifty index for a set of large asset management firms. To preserve their
anonymity, we’ve denoted them as Firm1, Firm2, …Firm5. On average, the ETF share-class
returns were about 50 bps higher. However, if liquidity costs such as the bid-ask spreads and
the impact costs of trading are high enough, it could potentially remove the performance
advantage offered by the ETF structure.
The liquidity problem with the ETF market also manifests in how well the ETF prices tracks
NAVs. To test this, we looked at 4 liquid Nifty ETFs and compared their price returns with
the index returns. In an ideal world, the median return difference would be the expense ratios
(the typical expense ratio of Nifty ETFs is about 6 bps) and tracking error would be 0. The
calculated standard deviations of daily returns were in the range of 18-30 bps.
However, standard deviation doesn’t tell investors much about what kind of return they can
expect, whether the fund is over- or under-performing its index, or how frequent outliers are
for different holding periods (Hougan, Hill, & Nadig, 2015). A better way to look at tracking
error is to look at return differences over the investment horizon, say 12-months. By this
measure the variation at 25th and 75th percentiles were much larger, in some cases over 50 bps6.
In other words, an investor with a one-year horizon might out- or under-perform the index by
over 50 bps, a quarter of the time.
Our analyses focus only on Nifty, a popular index. ETFs tracking broader market (such as
Nifty 100 or BSE 200) or smart beta indexes are expected to be more illiquid.
It may be argued that the liquidity issue in ETFs are salient because ETFs are transparent;
the liquidity costs in mutual funds are internalized within the structure in the form of return
impact. While there is merit to the argument, some of our comparisons of stock liquidity vs
ETF liquidity, as well as ETF prices vs. NAVs makes it clear that the lack of liquidity has a
meaningful impact on investor experience.
6
Calculations based on returns data between Sep 2016 – Sep 2019. 12-month return differences were calculated on a rolling
daily basis.
specific ETF segments, put their own capital at risk and intervene when prices move away from
fair values due to market demand for ETFs. Market makers perform a much more tactical role
of providing daily liquidity. In fact, the European Systemic Risk Board points to a conflict of
interest in the two roles during times of market stress, since arbitrage trades required to bring
the ETF price in line with the prices of the constituent securities may impose losses on APs
that can subsequently drain their capital and thus limit their liquidity provision, potentially
creating a negative spiral (Pagano, Serrano, & Zechner, 2019).
In any case, there aren’t many APs / market makers to support an ETF. A typical US ETF has
over 34 Authorized participants, with 5 active APs at any time. Even small ETFs (<$27M)
have 2 active APs (Antoniewicz & Heinrichs, 2015). Although large Indian ETFs may have
about 5 APs, few are active at any time and reasonably large limit-orders may take hours to
clear absent a vibrant secondary market.
Given thin volumes, the market making costs are (unsurprisingly) high. Whenever market
makers receive any substantial trade from investors it is typically one sized and they don’t have
offsetting orders. Hence, they have to go to the ETF manager for liquidity in lot increments.
The major costs that the market maker experiences in this regard are:
n Cost of hedging: Market maker provides ETF units on Day T+0 whereas the units from
the fund house are received on day T+1 in case of a buy trade.
n Cost of funds: The market maker has to provide funds to the fund house for unit creation
on T+0 while the funds from the investor are received only on T+2
n Cost of carry: Any fractional units of lots that are not accepted by the investor has to be
carried by the market maker because the liquidity in secondary market is so low.
We interviewed 2 out of the 3 largest market makers in the Indian markets and they mentioned
that anecdotally the charges for these come to about 10 bps for each leg of transaction over and
above normal transaction charges.
Indian Securities lending and borrowing (SLB) market is not very active. Securities lending
is an important enabler for short-selling and market-making programs, which in turn aids
liquidity in ETFs. It also has the potential to improve the performance of the ETF or
profitability of the fund sponsor, depending on how much of the revenue to passed through to
shareholders. A Morningstar study noted that ETF sponsors could routinely offset between 5%
and 50% of management fees through securities lending program (Morningstar, 2018). While
SEBI has allowed SLB in liquid ETFs at least since 2012, the practices vary. An analysis of
Scheme Information Documents (SID) reveal that some ETF sponsors do not engage in SLB
altogether, while others allow upto 20%. However, with the SEBI’s 2018 circular mandating
physical settlement for stock derivatives in phased manner, the securities lending market is
expected to become more active (SEBI, 2018).
Lastly, lack of robust hedging instruments such as futures and options also hinder ETF
liquidity. Liquidity providers such as ETF arbitrage desks routinely trade in ETFs and futures
when prices deviate too far from the fair value. While contracts are traded on the popular Nifty
index, the mid-cap Nifty Next 50, an index which is tracked by several ETFs lacks a futures
contract.
Fund houses, recognizing the liquidity problem permit direct creation and redemption. The
creation mechanism is available to large investors in multiples of ETF creation units, the value
of which could be as low as Rs. 1 million for some ETFs; the redemption is made contingent
on large discount to NAV, absence of quotes for consecutive days, or lack of secondary market
volume. While these measures help large investors in the fund, they also act as a backstop,
preventing market participants who might otherwise trade. They also lessen the distinction
between an ETF and an index fund.
IV.d. Distribution
One of the biggest impediments to ETF adoption relates to lack of incentives for ETF
distribution. Firstly, ETFs in India (like it is elsewhere) do not pay trail fees. Fee-only
registered investment advisors are a recent phenomenon, but their reach pales in comparison to
distributors. It is rare for retail investors to pay for advice when making investments in ETFs.
Given the vast disparity between the expense ratios of mutual funds and ETFs – median asset
weighted expense ratio in equity funds were 1.93% according to Morningstar 2019 Investor
Experience Study, while the cheapest ETFs carried an expense ratio of 5 bps (excluding
liquidity costs) – there is little incentive for mutual funds to promote investments in ETFs,
even in the face of underperformance in active funds. That said, there has been increasing
interest from high net worth individuals and family offices for investments in ETFs and other
passive investments. Wealth platforms have increased access to ETFs in response.
The second impediment relates to the logistical frictions in trading ETFs. An investor in mutual
fund needs to open a separate brokerage account, and till recently, it was not straightforward
for advisors to trade in ETFs on client’s behalf 7. While interest in index mutual funds (devoid
of these frictions) haven’t really increased either, such frictions have been occasionally cited by
market participants as a hassle.
V. Recommendations
The CFA Society India believes Indian investors would benefit from knowledge and access
to passive low-cost investment options. They would also benefit from a level-playing field
where they could evaluate the costs and benefits of investing in active funds vis-à-vis passive
funds, which as we’ve seen from the report is not always the case. Our recommendations cover
investor awareness, ETF ecosystem, and distribution.
7
Changes to SEBI (Investment Advisor Regulations) 2013 allowed implementation services through direct schemes/products
in the securities markets to investment advisors for the convenience of investors (SEBI, 2020).
VI. Conclusion:
Indian ETF industry has grown and matured considerably in the past 18 years, in terms
of assets, product launches, or adoption by institutional and high net-worth investors. The
industry has also benefited from tailwinds from the market and regulations.
Yet in other respects – investor awareness, liquidity, and market structures, the industry is still
nascent. Given the advantages offered by ETFs (low-cost, transparency, and liquidity), there is
considerable scope for deepening the market, increase awareness, and improve allocations in
retail portfolios.
The chart shows the S&P 500 index since inception (1926). The bars represent recession periods as
defined by the National Bureau of Economic Research. The major events in the evolution of ETFs are
represented as callouts.
8
This section is adapted from the CFA guide to Exchange Traded Funds (Hougan, Hill, & Nadig, 2015).
9
Mutual funds in the US are sometimes called the 40-Act funds.
The first index mutual fund was launched by John Bogle of the Vanguard Group, became
available in 1975. The modern ETF is, in their investment processes and organization, simply
an extension of index-based mutual funds. But ETFs also happen to be more tax efficient, have
lower cost than index funds, and be available on an exchange.
ETFs trace their roots back to the concept of “program trading,” a computer-based innovation
in the 1980s that allowed investors to purchase or sell all the shares of a major index (such as
the S&P 500) through a single trade order defined as the list of index stock tickers and shares
in each. Large program trading was a novel idea then, and was attributed as one of causes for
the crash of 1987, when computer programs started liquidating stocks in response to certain
stop loss targets and creating a domino effect.
Despite the setback, over the years several attempts were made to package these trades into
a single product. The idea for the first US ETF based on S&P 500 the SPY was born at
the American Stock Exchange in the early 1990s. State street global Advisors and American
Exchange created a structure that pioneered many of the key features of every ETF on the
market today: an exchange-traded access to a major market index that relied on an ongoing
creation/ redemption mechanism to keep the ETF’s market price tracking closely to fair value
throughout the day. SPY ended its first year with $475 million in assets under management
(AUM) and today is one of the largest ETFs in the world.
Over the next two decades, the industry saw several innovations. The international ETF series
launched by Morgan Stanley Capital International (MSCI) in partnership with Barclays
Global Investors were revolutionary, because it was the first-time investors discovered the
power of ETFs to offer price discovery, as ETFs were continuously traded even when the stock
exchanges of the underlying markets were closed.
ETFs became a popular vehicle of choice during the late 1990s, when there was tremendous
interest in technology stocks. QQQ, an ETF launched in 1999, garnered $18.7 billion in assets
in the first year of trading; by comparison, the total industry assets were only $15.7 billion the
previous year.
Powershares launched two smart beta ETFs tracking quant indexes in 2003, with an aim to
outperform the market. Since then, the share of smart beta ETFs as a proportion of overall
ETF assets have grown and accounted for over 20% of US ETF assets in 2018.
After the credit crisis of 2009, investors wanted access to products that protect them during
stock market crashes. Fixed income ETFs, minimum volatility ETFs, and short and leveraged
ETFs have all become popular over the last decade.
Bibliography
1. Antoniewicz, R., & Heinrichs, J. (2015). The Role and Activities of Authorized Participants
of Exchange-Traded Funds. Investment Company Institute. Retrieved from https://
www.ici.org/pdf/ppr_15_aps_etfs.pdf
2. ASX. (2020). Market Making. Retrieved from ASX: https://www.asx.com.au/products/
etf/market-making.htm
3. Australian Stock Exchange. (2020, 03 4). Market Making - ASX. Retrieved from
Australian Stock Exchange Web site: https://www.asx.com.au/products/etf/market-
making.htm
4. Hougan, J. M., Hill, D., & Nadig, M. (2015). A Comprehensive Guide to Exchange-
Traded Funds (ETFs). CFA Insitute Research Foundation. Retrieved from https://
www.cfainstitute.org/en/research/foundation/2015/a-comprehensive-guide-to-
exchange-traded-funds-etfs
5. Morningstar. (2018). Securities Lending - An Examination of the Risks and Reward.
Morningstar. Retrieved from https://www.morningstar.com/lp/securities-lending-
risks-rewards
6. Pagano, M., Serrano, A. S., & Zechner, J. (2019). Can ETFs contribute to systemic
risk? European Systemic Risk Board. Retrieved from https://www.esrb.europa.
eu/pub/pdf/asc/esrb.asc190617_9_canetfscontributesystemicrisk~983ea11870.
en.pdf ?c4e72c2dab45f2b74a5bc883a3e8009d
7. SEBI. (2017, October 6). Categorization and Rationalization of Mutual Fund
Schemes. Retrieved from SEBI: https://www.sebi.gov.in/legal/circulars/oct-2017/
categorization-and-rationalization-of-mutual-fund-schemes_36199.html
8. SEBI. (2018, December 31). Physical settlement of stock derivatives. Retrieved from SEBI:
https://www.sebi.gov.in/legal/circulars/dec-2018/physical-settlement-of-stock-
derivatives_41482.html
9. SEBI. (2020, February). Board meeting. Retrieved from SEBI: https://www.sebi.gov.in/
sebi_data/attachdocs/feb-2020/1581940465942.pdf
WEBSITE: www.cfasociety.org/india
EMAIL: [email protected]
PHONE: +91 022 61798952
Facebook: @CFASocietyIndia
Twitter: @CFASocietyIndia
Instagram: @cfa_society_india
Linkedin: @cfasocietyindia
CFA Society India is a brand licensed by CFA Institute to Indian Association of Investment Professionals