Features and Risks of ETF/ETN Investing through Non-Discretionary Bank Trusts Kwon, Min Kyeong Recently, banks have dramatically increased the purchases of ETFs and ETNs through non-discretionary trusts on behalf of their customers. Among ETFs and ETNs bought by banks, the volume of leveraged and derivativesbased products, in particular, has been growing rapidly. A leveraged ETF or ETN is a very risky product because its return can be highly volatile. Even when adding automatic termination agreement to mitigate the risk, the leveraged product is still more volatile than the underlying index. A short strangle strategy, the most widely used among derivatives-based products expects low volatility but its return could fall as much as the underlying index. More importantly, it could incur losses on even a bull market rally, which sets itself apart from conventional investment products. If customers are not sufficiently aware of such risks, customer damages and disputes could arise. Accordingly, proactive efforts seem necessary to prevent such events. FSS consumer warning about non-discretionary bank trusts In March this year, the Financial Supervisory Service (FSS), Korea s financial regulator, issued a consumer warning (No. 2018-1) regarding non-discretionary bank trusts. 1) In the midst of a surge in the sales of high-risk non-discretionary bank trusts used as a route to buy leveraged or inverse exchange traded funds (ETFs), this warning was to inform investors about the risks All opinions expressed in this paper represent the author s personal views and thus should not be interpreted as Korea Capital Market Institute s official position. Ph.D., Research Fellow, Fund & Pension Department, Tel: 82-2-3771-0681, E-mail: mkkwon@kcmi.re.kr 1) FSS, 29 March 2018, Consumer warning about the risks of investing in high-risk ETFs through non-discretionary bank trusts, press release. 1 July 10, 2018
of such products, in order to prevent harm to consumers. It is the first time that the FSS has warned consumers about a particular financial product since its consumer warning system was put in place in 2012. According to the FSS, the volume of ETFs that banks purchased via nondiscretionary trusts on behalf of their clients has skyrocketed from KRW 1.5 trillion in 2015 to KRW 8.0 trillion in 2017. The problem here is that leveraged and inverse products accounted for a significant portion of the ETFs. In 2015, banks bought KRW 269.4 billion worth of leveraged and inverse ETFs through non-discretionary trusts. In 2017, the volume of leveraged and inverse products they purchased stood at nearly KRW 4.1 trillion, a 15-fold increase in just two years. In addition, the monthly average sales of ETFs through non-discretionary bank trusts between January and February 2018 amounted to KRW 637.9 billion, far surpassing the 2017 monthly average sales of KRW 344.9 billion. As such, the trend has been accelerating. The same phenomenon may take place with regard to exchange traded notes (ETNs) as well. Recently, banks appear to have increased their efforts to sell ETNs through the trust channel, which are quite similar to ETFs. Non-discretionary ETN trusts, which were rolled out to the market for the first time in 2016, have been overshadowed by non-discretionary ETF trusts, receiving relatively little attention. Lately, however, this investment vehicle is growing rapidly with the popularity of ETNs whose underlying assets are derivatives. Among ETNs just like ETFs, there are also products offering leveraged and inverse exposure. Not only that, there are highrisk products based on derivatives such as options, which are used to track the performance of their benchmarks. Accordingly, not only ETF trusts but also ETN trusts place consumers at risk of suffering significant losses. This article explores the key features and potential risks of ETFs and ETNs that banks have bought through their non-discretionary trust accounts on behalf of their clients. Features of ETF and ETN investing through non-discretionary trusts in banks The analysis of 2015-2Q2018 data on ETFs and ETNs purchased by banks on the domestic stock exchange reveals the following three distinctive features. 2) 2) The data here do not necessarily serve the purpose of the analysis in this article because every ETF or ETN transaction at banks has not taken place only through non-discretionary trusts, and banks have not always bought ETFs and ETNs on the exchange. However, given that not many ETFs and ETNs have been traded by banks in their own accounts or other trust accounts, it can be assumed that most ETF and ETN transactions have been made through non-discretionary trusts. In addition, the objective of this analysis is not to accurately estimate the volumes of ETFs and ETNs but to capture the key features and risks of the products. Hence, the data would be sufficient to achieve the goal of the analysis without including over-the-counter ETF and ETN transactions. 2
Firstly, whereas the volume of ETFs bought by banks for their clients has soared but is staggering lately, the volume of ETNs has been growing steeply. As shown in Figure 1, the purchase of ETFs has continued to rise, reaching a peak of KRW 6.1 trillion in the first quarter of 2018. But it has clipped back significantly in the second quarter, which would most likely have been affected by stock market corrections and the FSS warning. By contrast, the purchase of ETNs began increasing notably in the fourth quarter of 2017 and has continued to rise rapidly to date. Figure 1. Purchases of ETFs and ETNs by banks in value terms Note: The cut-off date for the 2Q2018 numbers is 16 June 2018. Source: FnGuide Secondly, the purchase of ETFs showed high concentration in leveraged products. And the top picks among ETFs have changed, depending on the ups and downs of underlying indices. For example, KOSPI 200 or KOSPI 150 -linked leveraged ETFs gained high popularity for the whole year of 2017, accounting for about half of the ETF volume bought by banks. 3) Meantime, while the KOSPI index rose substantially during the first three quarters of 2017, KOSPI 200 leveraged ETFs soared in popularity. On the other hand, while the KOSDAQ market enjoyed a strong rally in the fourth quarter of 2017, KOSDAQ 150 leveraged ETFs became more popular. Over time, the top ETF picks have changed. Thirdly, the purchase of ETNs by banks exhibited very high concentration in derivatives-based products. Especially, one particular product among derivatives-based ETNs represented 89% of the total ETNs bought by banks. This product employs a short strangle which limits the potential 3) By contrast, leveraged ETFs based on KOSPI 200 or KOSPI 150 accounted for about 30% of the stock market s total trading value. 3 July 10, 2018
profit to the premiums received for selling the options if the underlying stock market index experiences little volatility and stays within a certain range (for example, 5% above and below the initial value of the underlying) on expiration date. The short strangle strategy is structured for a high probability of earning a limited profit with the risk of significant losses in the event of large movement in the underlying. Potential risks of ETF and ETN investing through non-discretionary trusts This section examines the risks of leveraged or derivatives-based exchange traded products banks purchased through their non-discretionary trusts. For starters, leveraged products tend to have far more volatile returns than the underlying index. For example, simulation results in Table 1 below show that while the monthly return volatility for the underlying index (KOSPI 200, 1x) was 6.4% from January 2000 to May 2018, that for leveraged ETFs/ETNs (KOSPI 200, 2x) was 12.9%. Due to the high volatility, customers sometimes use automatic termination agreement when trading leveraged products using non-discretionary trusts. Under the agreement, when a specified return threshold (for instance, 2% to 3%) is reached, automatic termination is triggered (meaning that the product is sold automatically). For a leveraged ETF or ETN, the return often reaches 2% to 3% because of its large volatility. The results in Table 1 indicate that 44% of entire leveraged ETFs and ETNs incurred losses, but their loss potential was reduced to 25% when the automatic termination contract was added. It can be understood that the automatic termination agreement mitigated the risk of loss to some extent, compared to leveraged products without that feature. Nevertheless, it is too early to conclude that a leveraged ETF or ETN with the automatic termination feature is safer investment than the underlying index (1x), because it could still suffer large losses due to the impact of leverage when the underlying index declines. In practice, the monthly return volatility for leveraged products with the automatic termination agreement added was 7.6%, higher than that for the underlying index (6.4%). For derivatives-based ETFs or ETNs, the profit and loss structure varies from one product to another. To understand the risks of derivatives-based products, take a short strangle strategy as an example. 4) There is a short strangle with the KOSPI 200 as underlying, which generates profits when the initial value of the underlying index remains within the range of ± 5%. Its return volatility 4) This strategy is constructed by selling a put whose strike price is less than 5% of the underlying index and simultaneously selling a call whose strike price is higher than 5% of the same index. 4
is not high because the maximum monthly profit is limited to the option premiums (for example, 50bp). As illustrated in Table 1, the simulation results reveal that the monthly return volatility for short strangle strategies was 3.1%, far below that for the underlying index (6.4%). 5) Even in that case, however, the risk of loss does not disappear. 38% of short strangles exhibited that the underlying index s monthly returns stayed outside the band of ± 5%. It means that the short strangle strategy is still at high risk of loss. 6) The maximum monthly loss for the short strangle was 17.6%, which was just as high as that for the underlying index (21.0%). Importantly, investors need to note that short strangles differ by nature from conventional investment vehicles, considering that they can result in losses when the stock market moves strongly in either direction. Table 1. ETF/ETN risks by product type (based on monthly returns) Underlying Index (1x) Leveraged Product (2x) Leveraged Product (2x, automatic termination) Short Strangle Volatility 6.4% 12.9% 7.6% 3.1% Loss Probability 43% 44% 25% 33% Bottom 10% -7.1% -14.5% -9.3% -5.7% Bottom 5% -10.1% -19.4% -14.5% -7.7% Minimum Value -21.0% -41.4% -41.4% -17.6% Note: 1) Based on monthly returns for KOSPI 200 from January 2000 to May 2018. 2) For leverage products with the automatic termination agreement added, it is assumed that the product is sold at its closing price when the daily closing price of the underlying asset exceeds 3% of the closing price at the end of the previous month. 3) For short strangles, 5% out-of-money options are considered, and the option premium is 50bp. 4) Actual returns are likely to be different, depending on various factors such as rollover costs, derivatives expiry, dividends, option premiums, and trustee fees. Meantime, the structural commonality between the automatic termination and the short strangle, as seen in Figure 2, is to reduce both the maximum target profit and loss potential. This structure, just like term deposits, is more appealing to bank customers who are familiar with fixed interest rate products. As discussed earlier, however, it can result in substantial losses, albeit with lower probability, which means that the products still pose risks. 5) For simulation purposes, the option premium is fixed at 50bp, assuming a simple environment without complex conditions and limitations. Actual return distribution may be different depending on realistic factors such as option premiums, rollover costs and maturity. 6) Yet, even if the value of the underlying stays outside the band of ±5%, loss does not necessarily occur. Depending on option premiums, the loss threshold may change. As KOPSI 200 s volatility has subdued greatly these days, the probability of short strangle losses has declined significantly. For instance, the monthly return of their underlying index drifted beyond the band of ±5% for five months out of 60 months (8% of the recent 5-year period). 5 July 10, 2018
Figure 2. Monthly return distribution of ETFs/ETNs by product type Note: 1) The horizontal axis depicts time, while the vertical axis shows the monthly returns. 2)The same assumptions as those for Table 1 apply. Implications Obviously, non-discretionary trusts have a good function in the sense that through them, ETFs and ETNs are introduced to customers with limited stock market participation. ETFs and ETNs are widely popular among individual and institutional investors thanks to low costs and transparent operation. However, one disadvantage is a lack of access to the products for investors who are not familiar with direct trading. Unlike publicly offered funds, investors in ETFs and ETNs pay no sales charges, giving little incentives for financial institutions to sell the products. These disadvantages could be addressed by non-discretionary trusts which enable customers to indirectly trade ETFs or ETNs through financial institutions. When considering the features of leveraged or derivatives-based ETFs and ETNs sold through non-discretionary bank trusts, it is necessary to make sure that customers understand the risks of the products properly. Because the majority of the customers have no experience in trading stocks or derivatives in the market, 7) it would not be easy for them to understand the complex profit and loss structure of leveraged or derivative-linked products. If they are not sufficiently aware of the risks, customer damages and disputes could arise. Even now, proactive efforts seem necessary to prevent such events. 7) ETFs or ETNs are investment products listed on the exchange, which provide little incentives for customers who are familiar with direct trading to use non-discretionary bank trusts with trustee fees chargeable. 6