ETF Research Buy and Adjust : Capturing a Structural Factor with PPLC Research compiled by Michael Venuto, CIO The first US-listed ETF targeting the S&P 500 Index began trading in 1993. Today the US ETF market is over $3 trillion, and over 185 ETFs representing almost $800 billion in assets are dedicated to US large caps, most in variations of S&P 500 strategies. These large-cap ETFs are by now quite varied in approach, targeting core, style boxes, factors, sectors, ESG, as well as leveraged, tactical and active approaches. Many newer ETFs are also designed to provide enhanced or smart beta intended for buy-and-hold exposure to US large caps an irony considering that the original S&P 500 ETF was designed with short-term traders and hedgers in mind. Smart beta tilts employed in these ETFs are usually based on the past and present characteristics of the holdings. Using these smart ETFs requires more than just picking the proper enhancement and then reaping the benefits. Investors must assume or at least expect that the factor being targeted will continue to materialize in the tilts or weightings. The challenge is that markets can be inefficient in the short run, therefore negating, or at least delaying, the benefits of many smart beta ETFs. It s probably fair to say that many, if not most, investors lack the patience to wait for the markets to return to efficiencies where a particular factor can produce a superior risk-adjusted return. What Should Investors Do? The expansion and the evolution from a pure liquid tradable beta vehicle to more than 180 varieties of US large-cap ETFs begs the question: How can ETFs best serve investors? Should they use ETFs as the intraday trading or hedging vehicles envisioned in 1993? Or should they hew to more of a Jack Bogle approach and embrace long-term buy-and-hold passive indexing? In this paper, we're proposing that it may be neither. We ll explore another permutation of investing. Somewhere between day trading and buy and hold is what we re calling a buy and adjust approach that we believe can capture both the trading benefits and structural enhancements of large-cap US ETF investing. 1
The Case for Large Caps Before examining closely what a large-cap buy and adjust ETF is and how it works, a few words about large-cap stocks are in order. Large caps, which we define as public companies with average market cap values above $10 billion, are hugely important in portfolio construction. That s because big companies have robust and resilient balance sheets and are typically among the most stable equity investments. They often market well-known brands, operate multiple wellmanaged business units and, not least, enjoy dependable pricing power in their target markets. All these attributes make them less vulnerable to market volatility and the ravages of an economic downturn than, say, small-cap stocks. Also unlike small-cap stocks, large-cap stocks shoot off attractive dividends that frequently rise over time as the typically reliable growth of larger companies translates into higher revenue and profit that investors can reap, all of which adds to the stability of their returns. Even so, large-cap stocks are sometimes dismissed as overly conservative or even boring. But they shouldn t be. Large-cap stocks represent a superb low-risk payoff for investors, and that s why we believe they should have a core presence in every investment portfolio. The Buy and Adjust Innovation Conversely, a buy and adjust strategy can be designed around a very different form of enhancement that s based on the ETF s structure rather than on shifts in a fund s holdings. The concept we re highlighting here is enhanced precise beta through the use of light leverage more specifically exposure that s designed to deliver 1.25 times the returns of the S&P 500 Index before fees and expenses, and rebalanced daily. Because the enhancement in this case is based on structure, the strategy s behavior, correlation and returns relative to plain-vanilla S&P 500 exposure are consistent, predictable (relative to the benchmark) and reasonably easy to model. With this in mind, let s explore the structure of the fund in question; the Portfolio + S&P 500 ETF (PPLC). The S&P 500 1.25 Hypothetical Leverage Returns represent a model based on mathematical principals, which cannot be altered and does not represent any given index or portfolio, in an effort to illustrate the effects of daily leveraged goals on a broad-based index. The model applies the leverage goal, 125%, to the daily return streams of the underlying index and is rebalanced each day, additionally, fees have been applied to the model to reflect the cost of gaining leverage exposure. Actual results may differ from the hypothetical returns presented. Small variances in the actual daily prices can have compounded impact on long term returns. Past performance is not indicative of future results. We ll also examine how deploying the PPLC model with the buy and adjust strategy can be used to target precise enhanced beta, equitize cash, express opinions on volatility or trend, and provide more predictable relative returns than holdings-based smart beta large-cap ETFs. 2
The Mechanics PPLC, the Portfolio + S&P 500 ETF, seeks daily investment results, before fees and expenses, of 125% of the performance of the S&P 500 Index. In other words, this is a lightly leveraged ETF that seeks a return that is 125% the return of its benchmark index for a single day, though there is no guarantee it will achieve this objective. This 1.25X exposure is quite different from the returns provided buy ETFs with 2X and 3X exposure. Specifically, even though the position size requires adjustment to account for possible decay of principal, the compounding risk of holding a lightly levered ETF with 1.25X exposure to its index is substantially less than a 2X or 3X ETF. Not All Leverage Is Created Equal Additionally, the 1.25 leverage in the Portfolio + S&P 500 ETF (PPLC) is less than what is commonly seen in the universe of closed-end funds (CEFs). A total of 134 CEFs focused on equities are on the market, and they employ leverage of up to 40% or 1.4, though the current average is about 22% or 1.22. But unlike ETFs, CEFs typically trade at prices that are significantly different from the net asset value (NAV) of their underlying holdings, which is a primary reason many Financial Advisors shy away from using them. The reason CEFs trade at premiums or discounts to their NAVs is because they can t easily redeem or create new shares. In the ETF format, PPLC does not have this limitation. In other words, PPLC combines the systematic use of CEF-like leverage with the intraday creation and redemption of shares that s at the heart of the ETF structure, thereby minimizing tracking error from NAV that advisors abhor. The bar chart below shows the hypothetical returns of the PPLC model relative to the standard S&P 500 Index in a variety of time frames. 3
The unique output is consistency. Since the enhancement in the PPLC model is structural, the fund s hypothetical price movements are relatively consistent and predictable. Again, that s unlike many of the traditional smart beta indexes that correlate very differently to the S&P 500, within market cycle, as the table below illustrates: 1 Year through 12-31-17 Return Excess Return R2 Tracking Error Std Dev Beta PPLC Model 27.15 5.46 74.00 3.88 6.71 1.46 S&P 500 Value Index 15.36-6.33 62.83 3.70 5.95 1.19 S&P 500 Growth Index 27.44 5.75 55.59 3.04 4.43 0.84 S&P 500 Momentum Index 28.27 6.57 50.26 3.95 5.61 1.01 MSCI USA Minimum Volatility Index 16.54-5.15 74.86 2.46 4.86 1.07 S&P 500 Equal Weighted Index 18.90-2.79 72.00 2.51 4.74 1.02 S&P 500 Index 21.69 0.00 100.00 0.00 3.93 1.00 3 Years through 12-31-17 PPLC Model 13.70 2.40 97.66 3.34 12.89 1.27 S&P 500 Value Index 9.47-1.83 89.19 3.44 10.39 0.98 S&P 500 Growth Index 12.86 1.56 91.70 3.07 10.72 1.02 S&P 500 Momentum Index 12.70 1.40 76.68 5.09 10.06 0.90 MSCI USA Minimum Volatility Index 9.08-2.22 68.05 5.67 8.35 0.67 S&P 500 Equal Weighted Index 10.11-1.19 93.90 2.57 10.42 1.01 S&P 500 Index 11.30 0.00 100.00 0.00 10.05 1.00 5 Years through 12-31-17 PPLC Model 19.39 3.73 98.42 2.94 12.08 1.27 S&P 500 Value Index 14.24-1.42 90.82 2.99 9.88 0.99 S&P 500 Growth Index 17.00 1.34 92.63 2.69 9.92 1.01 S&P 500 Momentum Index 15.91 0.25 79.29 4.50 9.74 0.91 MSCI USA Minimum Volatility Index 12.55-3.11 69.37 5.26 8.40 0.74 S&P 500 Equal Weighted Index 15.79 0.13 94.47 2.41 10.14 1.04 S&P 500 Index 15.66 9.47 100.00 0.00 9.47 1.00 10 Years through 12-31-17 PPLC Model 9.55 1.15 99.67 4.08 19.01 1.26 S&P 500 Value Index 6.80-1.60 95.54 3.57 16.32 1.06 S&P 500 Growth Index 9.99 1.58 95.43 3.22 14.63 0.95 S&P 500 Momentum Index 8.55 0.15 81.68 6.57 14.84 0.89 MSCI USA Minimum Volatility Index 7.19-1.22 84.52 6.23 11.91 0.73 S&P 500 Equal Weighted Index 10.15 1.75 95.09 4.54 17.74 1.15 S&P 500 Index 8.40 0.00 100.00 0.00 15.04 1.00 Source: Morningstar Direct 4
Another factor, High Beta is commonly used to express a bullish opinion on US Large Cap stocks. The actual decays of this strategy have historically caused substantial underperformance relative to the PPLC model. Which again highlights the precision of relative performance provided by PPLC model and the benefits of a structural factor as outlined in the charts below: Beyond the excess returns versus the S&P 500 High Beta Index, the PPLC model exhibits substantially less volatility and with almost half the Beta. 10 years ending 12-31-2017 Return Std Dev Downside Deviation Alpha Beta Sharpe Ratio Up Period Percent Down Period Percent PPLC Model 9.55 19.01 9.58 5.44 0.63 0.48 64.17 35.83 S&P 500 High Beta Index 4.87 27.84 0.00 0.00 1.00 0.16 55.83 44.17 Allocation Approaches What s potentially important in the table above is that the ability to predict the PPLC model s beta to the S&P 500 allows investors to be more precise with the amount of leverage they employ to augment their returns. With that type of precision in mind, the chart below illustrates how combining various weights of plain-vanilla S&P exposure represented by the SPDR S&P 500 ETF ( SPY ) with PPLC can be used to pinpoint exact levels of leverage: 5
For investors looking to maintain 10% in cash but who also want to maintain full market exposure, a 60/40 allocation (60% SPY/40% PPLC) can produce that return when the allocation is properly maintained and adjusted hence the use of our term buy and adjust. Comparing PPLC To Margin or Options Exposure Alternatively, investors holding positions in inversely correlated alternative investments such as managed futures or gold and who are looking to cover the performance drag in an upwardly moving market can precisely target the amount of leverage or S&P beta they require through buying and adjusting a position in PPLC. Additionally, targeting precise leverage using a vehicle such as PPLC is logistically superior to other alternatives in that it is not restricted by margin or options requirements, which are the traditional ways investors integrate light leverage into their portfolios. So how does PPLC compare to traditional forms of leverage, including buying shares on margin or using options? To begin, unlike the cost of PPLC, margin rates and options premiums are constantly changing based on interest rates, market volatility and other macroeconomic factors. In other words, the cost of achieving leverage through margin or options isn t predictable. Setting aside the trading cost of adjusting and maintaining 25% leverage using PPLC, we wanted to illustrate the differences in margin costs on different trading platforms and across a spectrum of different rate environments. 6
Platform US Margin Loan Rates Comparison $25K $300K $1.5M $3.5M Toroso Investments, LLC Equity Trades E Trade 9.25% 7.75% 6.25% 6.25% $6.95 Fidelity 8.07% 6.82% 4.25% 4.25% $4.95 Interactive Brokers 2.66% 2.32% 2.02% 1.78% $2.27 Scottrade 8.25% 7.00% 6.00% 6.00% $6.95 Schwab 8.07% 6.82% 6.25% 6.00% $4.95 TD Ameritrade 9.00% 7.50% 6.75% 6.75% $6.95 Vanguard 7.25% 6.00% 5.25% 5.25% For illustration only. Source: Interactive Brokers $7 - $20 Depends on the number of trades Commission Rates Comparison Option Trades $6.95 base +$0.75 per $4.95 base +$0.65 per No Base Cost +$0.75 per $6.95 base +$0.70 per $4.95 base +$0.65 per $6.95 base +$0.75 per $7.00 base +$1.00 per Futures Trades Exchange & Regulatory Fees + $1.50 N/A $1.75 N/A N/A Exchange & Regulatory Fees + $2.25 PPLC has a current net expense ratio of 0.34% (including acquired fund fees). The annual cost of a $100,000 investment in PPLC is $340.00. However, the investor has exposure to $125,000 of the S&P 500 Index and using the estimated expense ratio of SPY at 0.09% as a proxy, this cost would be $112.50. Therefore, the additional cost for the extra leveraged $25,000 is $227.50; or an implied margin loan rate of 0.91% (excluding the commission rates for trades shown in the above table which increases the cost of providing the leverage outside of the ETF). From this analysis, the investor can obtain leverage inside the ETF at very favorable terms (below all comparisons in the table above) in the current interest rate environment PPLC Costs Are Predictable Looking at the table below, even at a margin rate of 2%, setting a position at a $100,000 investment with 125% leverage would cost at least 61 bps or 27 bps more than PPLC including the acquired fund fees. Additionally, PPLC resets daily to 125% exposure, whereas maintaining similar exposure with margin or options would require constant trading and all the friction costs associated with that trading. N/A 7
Margin Cost Toroso Investments, LLC Annual Costs Savings PPLC 34 bps $ 340.00 2.0% 61 bps $ 612.50 27 bps 3.0% 86 bps $ 862.50 52 bps 4.0% 111 bps $ 1,112.50 77 bps 5.0% 136 bps $ 1,362.50 102 bps 6.0% 161 bps $ 1,612.50 127 bps 7.0% 186 bps $ 1,862.50 152 bps 8.0% 211 bps $ 2,112.50 177 bps 9.0% 236 bps $ 2,362.50 202 bps 10.0% 261 bps $ 2,612.50 227 bps The predictable cost and relative exposure provided by the PPLC model combined with the limited activity of a buy and adjust strategy appear to us to be superior to traditional forms of leverage. PPLC has an annual expense ratio of 25 basis points (bps) plus the acquired fund fees (estimated to be 9bps in the PPLC Fact Sheet), which is well below the 47 bps average of the other 184 US large-cap ETFs. That said, the four traditional market-cap ETFs control the bulk of the assets allocated to US large caps ($500 billion), and expense ratios on those market-cap choices average about 9 bps. However, excluding these four market-cap large-cap ETFs, the average expense ratio jumps to 88 bps. So, clearly paying 25 bps to acquire systematic and predictable leverage to the S&P seems reasonable relative to other large-cap smart beta ETFs. Defining Buy and Adjust So, what does buy and adjust really mean? Above, we explored how a structural factor such as the 1.25x leverage built into the PPLC model produced predictable price behavior relative to other leverage options available to investors. Specifically, we illustrated how the relative correlation and beta using this fund structure remains consistent throughout market cycles. But what is the downside to leverage and, crucially, how is 1.25x leverage different from double or triple exposure ETFs? The risk of the daily-reset leverage designed into such 2x, 3x and 1.25x ETFs is negative compounding. In practical terms, that means that market volatility can cause the principal invested in a leveraged ETF to decay. The decay in a 2X or 3X ETF can be substantial at times. That s why such funds are explicitly offered as short-term trading vehicles. But unlike in lightly levered ETFs, the correlation and beta of 2X and 3x levered ETFs can be difficult to predict in periods of high volatility. The chart below describes the environments where leverage benefits from compounding and, importantly, where leverage can result in potential decay of principal. 8
Periods of LOW Volatility Periods of MODERATE Volatility Periods of HIGH Volatility Market is Trending UP Improved Performance Generally Outweighs Cost of Decay Improved Performance Generally Outweighs Cost of Decay Decay May Overtake Performance No Discernable Trend to Market Slight Decay May Occur Slight Decay May Occur Decay More Likely Market is Trending DOWN Improved Performance Generally Outweighs Cost of Decay Negligible Decay Improved Performance Generally Outweighs Cost of Decay The takeaway from the table above is that exposure to daily reset leverage must be adjusted at intervals that are consistent with the market environment. Framed in the simplest terms, the greater the leverage, the more adjustments are required. That means that understanding how a particular environment affects exposure can provide indications of how best to make those adjustments. Hypothetical Buy and Adjust Example Examining this volatility/trend table gives us an opportunity to revisit the equitization of cash example highlighted above. In that example, we had a 60/40 mix of pure S&P 500 exposure combined with PPLC to create 110% exposure to the S&P while also allowing for a 10% position in cash. So, let s assume we ve just experienced six months of a rising market characterized by low volatility. Our exposure to PPLC would have grown more than our exposure to the S&P due to the positive effects of compounding. Thus, after those six months, our mix would look more like a 55/45 blend of the two funds and would be providing more leverage than we initially intended. Such a circumstance represents a key time to adjust. No less, the opposite can also be true, which is to say that in a trendless and high-volatility environment, decay of principal in PPLC would likely require additional allocation of PPLC to maintain the 110% exposure. Crucially, a lightly levered product such as PPLC, just like any smart beta ETF, can t be held blindly forever. Even when the fund delivers exactly what it promises, adjustments to the 9
position size are needed to compensate for changes in the market trend, volatility and overall position evolution. This is the essence of a buy and adjust strategy using the predictable relative returns of lightly levered ETFs. Conclusion When the first ETF began trading in 1993, it was designed to be traded as a more efficient exposure than futures. But the smart-beta ETFs that have followed in the years since have been designed with buy-and-hold investors in mind. The reality, is most of these smart beta ETFs are simply providing factor tilts, which can diverge quite a bit from their expected relative performance for extended time periods. Which ironically makes them potentially poor buy-and-hold investments. Conversely, the PPLC model provides structural and predictable relative performance to largecap US equities in a low-cost vehicle that can be deployed in a buy and adjust manner that s designed to solve multiple investing challenges. When it is all said and done, probably the most compelling aspect of this structurally smart beta ETF strategy in the large-cap core category has been; the strong performance relative to the S&P 500 Index, the consistency and precision of the relative performance, the lower cost (25bps excluding acquired fund fees), and the relative performance and consistency versus many factor tilts especially the high beta factor 10
Disclaimer Toroso Investments, LLC ( Toroso ) is an investment adviser registered with the United States Securities & Exchange Commission ( SEC ). Registration with the SEC does not imply any certain level of skill or training. A copy of Toroso s current written disclosure statement (i.e., Form ADV, Part 2A) that discusses Toroso s advisory services and conflicts of interest is available at http://www.adviserinfo.sec.gov by searching for CRD Number 164201. This research report was commissioned by Direxion Investments; the sponsor of the Direxion Daily S&P 500 Bull 1.25X Shares (PPLC), a leveraged exchange-traded fund ( ETF ) that seeks a return that is 125% the return of its benchmark index before fees and expenses for a single day (Note, the fund should not be expected to provide 1.25 times the return of the benchmark s cumulative return for periods greater than a day). Consequently, Toroso has a conflict of interest in compiling this research and in generating the opinions expressed in this report. Toroso, Michael Venuto and David Dziekanski may have positions in securities that they research and follow, including the ETFs discussed in this report and may engage in buying or selling securities contrary to any opinions expressed in this report. Additionally, some of the securities discussed may be or may have been held in advisory client accounts of Toroso. The compensation that Messrs. Venuto and Dziekanski receive from Toroso may be enhanced or otherwise increased as a result of Toroso s profitability, but neither received direct payments for compiling this research. This research report is distributed for informational and educational purposes only. It is not intended to constitute legal, tax, accounting or investment advice. Nothing in this research report constitutes an offer to sell or a solicitation of an offer to buy any security or service and any securities discussed are presented for illustration purposes only. It should not be assumed that any securities discussed herein were or will prove to be profitable or that an investment in any of the securities discussed will result in investment performance equal to that presented. Furthermore, investments or strategies discussed may not be suitable for all investors and nothing herein should be considered a recommendation to purchase or sell any particular security. ETFs are sold by prospectus. Investors should carefully consider the investment objectives, risks, charges and expenses of ETFs, which is disclosed along with other important information in the prospectus. Investors should make their own investment decisions based on their specific investment objectives and financial circumstances and are encouraged to seek professional advice before making any decisions. The market price of an ETF will fluctuate in response to changes in the underlying investments' market prices and can fluctuate due to market supply/demand dynamics. Consequently, ETFs can trade at a discount or premium to their net asset value. Additionally, certain ETFs are more volatile and less liquid than others and thus present greater risks of loss of capital. Toroso has compiled its research from sources that it believes to be reliable, but cannot guarantee that the information presented is accurate or that it is a complete statement of all material factors. Any opinions expressed in this research report are the opinions of Toroso and Messrs. Venuto and Dziekanski and do not reflect the opinions of any other affiliate. Employees and/or affiliates of Toroso may, at times, release written or oral commentary, technical analysis or trading strategies that differ from the opinions expressed in Toroso s research reports. Furthermore, all opinions are current only as of December 31, 2017, do not take into account the particular investment objectives, financial situation or needs of individual investors, and are subject to change without notice. Toroso does not have any obligation to provide revised opinions in the event of changed circumstances. All investment strategies and investments involve risk of loss and nothing within this research report should be construed as a guarantee of any specific outcome or profit. Securities discussed in this research report were selected for presentation because they serve as relevant examples of the respective points being made. 11