Return Stacking is a topic we wrote about recently (Part 1). That post briefly covered some of the basics including:
It also introduced the line-up of Return Stacked ETFs shown in Table 1.
| ETF | Asset Class 1 | Asset Class 2 | Net Assets |
|---|---|---|---|
| RSBT | US Bonds | Managed Futures | $50M |
| RSSB | Global Equities | US Treasuries | $69M |
| RSST | US Equities | Managed Futures | $67M |
As we pointed out last time, these ETFs have limited return history. Therefore, we specify proxy ETFs to model the return streams of the two sub-asset classes harvested by each Stack Returned fund. Table 2 shows how we will map them to liquid securities currently available to retail investors.
| Asset Class | ETF Proxy | Description | Inception |
|---|---|---|---|
| Global Equities | VT | Vanguard Total World Stock Index Fund | 2009 |
| Managed Futures | WDTI | WisdomTree Managed Futures Strategy | 2011 |
| US Bonds | AGG | iShares Core US Aggregate Bond | 2004 |
| US Equities | SPY | SPDR S&P 500 | 1994 |
| US Treasuries | IEF | iShares 10 Year Treasury | 2002 |
We have made liberal use of a simple OLS framework to analyze which asset classes have the most potential to add orthogonal returns to an existing portfolio. This post is going to use the same framework but this time we focus on RSBT and (by proxy) the WDTI and AGG ETFs.
As a quick refresher, we reintroduce the model here. It is a multivariate regression given by the following equation:
$$ r_t - r_f = \alpha + \beta_1\left(r_{1t} - r_f\right) + \beta_2\left(r_{2t} - r_f\right) + \epsilon_t $$ where $$r_t := \text{diversifying asset return}$$ $$r_{1t} := \text{asset class 1 return}$$ $$r_{2t} := \text{asset class 2 return}$$ $$r_f := \text{risk free rate}$$ $$\alpha_t := \text{intercept}$$ $$\beta_1 := \text{asset class 1 exposure}$$ $$\beta_2 := \text{asset class 2 exposure}$$
We focus on the intercept and asset class exposures (the alphas and the betas). In order to measure how much improvement a given ETF adds to RSBT we define a simple score whose purpose is to measure how much additional alpha is added per unit of beta. The naive score we have chosen takes the form
$$\text{Score} = \frac{\alpha}{|\beta_1| + |\beta_2|}$$
This ratio rewards assests with higher alpha and penalizes those with betas that deviate from zero.
Tables 3 through Table 9 show an assortment of ETFs we have chosen to consider. Each table shows the start and end dates that correspond to the cross-section of data available for WDTI, AGG, and each asset class ETF.
As shown in Table 3, DBC provides the best balance of alpha per unit of beta in this limited set of commodities ETFs.
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| DBC | DB Index | 5.81 | -0.39 | 2.20 | 2.25 | 2011 | 2023 |
| GSG | GSCI | 5.55 | -0.57 | 2.95 | 1.58 | 2011 | 2023 |
| USO | Oil | 7.64 | -1.07 | 6.12 | 1.06 | 2011 | 2023 |
| GLD | Gold | -0.01 | 1.30 | -0.37 | -0.00 | 2011 | 2023 |
Consistent with our previous post DBC, GSG, and USO all exhibit high exposure to WDTI which should not be too surprising.
Table 4 shows that HYG and JNK (both high yield ETFs) are additive.
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| HYG | iBoxx $ High Yield | 6.55 | 0.13 | 0.47 | 10.84 | 2011 | 2023 |
| JNK | Bloomberg Barclays High Yield | 6.61 | 0.11 | 0.70 | 8.11 | 2011 | 2023 |
| LQD | iBoxx $ Investment Grade | 1.53 | 1.42 | -0.05 | 1.03 | 2011 | 2023 |
LQD is investment grade debt and has high exposure to AGG as we would expect.
The currency ETFs in Table 5 show that UUP (bullish dollar fund) is additive.
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| UUP | US Dollar Fund | 2.79 | -0.31 | 0.58 | 3.15 | 2011 | 2023 |
| FXE | Euro Currency | -2.23 | 0.22 | 0.31 | -4.24 | 2011 | 2023 |
Exposure to the Euro provides negative alpha.
Table 6 says that REITs were a sensible addition for this time period.
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| VNQ | Real Estate Index | 11.27 | 0.63 | -0.19 | 13.67 | 2011 | 2023 |
| IYR | US Real Estate | 10.49 | 0.62 | -0.55 | 8.91 | 2011 | 2023 |
| RWX | International Real Estate | 5.74 | -0.01 | -0.82 | 6.94 | 2011 | 2023 |
Table 7 says there is little (to no) reason to add rates as they are most likely subsumed by AGG.
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| SHV | Short Treasury | 0.40 | 0.01 | 0.29 | 1.33 | 2011 | 2023 |
| SGOV | 0-3 Month Treasury | 0.77 | 0.00 | 0.74 | 1.04 | 2020 | 2023 |
| SHY | 1-3 Year Treasury | -0.08 | 0.19 | 0.09 | -0.29 | 2011 | 2023 |
| IEI | 3-7 Year Treasury | -0.26 | 0.75 | 0.03 | -0.33 | 2011 | 2023 |
| IEF | 7-10 Year Treasury | -0.90 | 1.46 | -0.26 | -0.52 | 2011 | 2023 |
| TLH | 10-20 Year Treasury | -1.50 | 2.11 | -0.52 | -0.57 | 2011 | 2023 |
| TLT | 20+ Year Treasury | -2.73 | 3.25 | -0.98 | -0.65 | 2011 | 2023 |
We should expect the outsized scores shown in Table 8 since we expect stocks to have limited exposure to WDTI (trend) and AGG (bonds).
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| XLP | Consumer Staples | 13.43 | -0.07 | 0.20 | 49.88 | 2011 | 2023 |
| XLY | Consumer Discretionary | 21.09 | -0.46 | 0.52 | 21.42 | 2011 | 2023 |
| XLK | Technology | 22.56 | -0.37 | -0.94 | 17.11 | 2011 | 2023 |
| XLV | Health Care | 16.99 | -0.45 | -0.71 | 14.66 | 2011 | 2023 |
| XLB | Materials | 16.23 | -0.77 | -0.66 | 11.35 | 2011 | 2023 |
| XLI | Industrials | 18.41 | -0.90 | -0.77 | 11.05 | 2011 | 2023 |
| XLRE | Real Estate | 10.90 | 0.98 | -0.32 | 8.39 | 2015 | 2023 |
| XLE | Energy | 15.70 | -1.14 | -0.80 | 8.09 | 2011 | 2023 |
| XLC | Communication | 14.24 | 0.44 | 1.40 | 7.73 | 2018 | 2023 |
| XLF | Finance | 19.11 | -1.53 | -1.15 | 7.13 | 2011 | 2023 |
| XLU | Utilities | 9.70 | 0.79 | -1.73 | 3.85 | 2011 | 2023 |
For this period of data Table 8 indicates that Consumer Staples and Consumer Discretionary would be sensible additions to RSBT.
If this is true for stock sectors then we should expect it to be directionally similar for stock indices. Table 9 shows that this is the case.
| ETF | Description | Alpha | AGG Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| QQQ | Nasdaq 100 | 22.61 | -0.38 | -0.20 | 39.60 | 2011 | 2023 |
| IWF | Russell 1000 Growth | 19.58 | -0.37 | -0.23 | 32.75 | 2011 | 2023 |
| SPY | SP500 | 17.89 | -0.57 | -0.31 | 20.34 | 2011 | 2023 |
| IWB | Russell 1000 | 17.77 | -0.56 | -0.38 | 18.93 | 2011 | 2023 |
| IWV | Russell 3000 | 17.76 | -0.59 | -0.39 | 18.13 | 2011 | 2023 |
| VT | Vanguard Total World | 13.74 | -0.53 | -0.42 | 14.42 | 2011 | 2023 |
| IWM | Russell 2000 | 17.29 | -0.81 | -0.56 | 12.66 | 2011 | 2023 |
| IWD | Russell 1000 Value | 15.76 | -0.76 | -0.54 | 12.14 | 2011 | 2023 |
| SCZ | MSCI EAFE Small Cap | 10.99 | -0.45 | -0.53 | 11.24 | 2011 | 2023 |
| EFA | MSCI EAFE | 10.36 | -0.59 | -0.53 | 9.27 | 2011 | 2023 |
| EEM | MSCI Emerging | 5.70 | -0.32 | -1.42 | 3.27 | 2011 | 2023 |
The Nasdaq and the Russell Growth indices clearly add value. The statistics for these two ETFs (QQQ nad IWF) are more or less in line with those for XLK which makes intuitive sense.
It is plausible that the superior scores exhibited by QQQ and IWF are artifacts of their recent out performance. Since all of the equity indices appear additive, it might be more prudent to add a broader based index like VT since it is more diversified. At the very least, investors preferring exposure to US based indices should consider SPY.
This simplistic analysis suggests that the ETFs in Table 10 are the most additive to RSBT:
| ETF | Description | Alpha | SPY Beta | WDTI Beta | Score | Start | End |
|---|---|---|---|---|---|---|---|
| XLP | Consumer Staples | 13.43 | -0.07 | 0.20 | 49.88 | 2011 | 2023 |
| QQQ | Nasdaq 100 | 22.61 | -0.38 | -0.20 | 39.60 | 2011 | 2023 |
| SPY* | SP500 | 17.89 | -0.57 | -0.31 | 20.34 | 2011 | 2023 |
| VT* | Vanguard Total World | 13.74 | -0.53 | -0.42 | 14.42 | 2011 | 2023 |
| VNQ | Real Estate Index | 11.27 | 0.63 | -0.19 | 13.67 | 2011 | 2023 |
| HYG | iBoxx $ High Yield | 6.55 | 0.13 | 0.47 | 10.84 | 2011 | 2023 |
| UUP | US Dollar Fund | 2.79 | -0.31 | 0.58 | 3.15 | 2011 | 2023 |
| DBC | DB Index | 5.81 | -0.39 | 2.20 | 2.25 | 2011 | 2023 |
| SHV | Short Treasury | 0.40 | 0.01 | 0.29 | 1.33 | 2011 | 2023 |
A mixture of ETFs additive to RSBT should include securities with long exposures to stocks, real estate, high yield bonds, and the dollar.
While SPY and VT were not the most additive ETFs in the stock index category we have added them to this table for comparison. We stand by our statement that, for long-term diversification, it makes more sense to add broader based indices to RSBT.
High yield bonds could add a stock-like return stream. Therefore, investors should investigate how additive these are after adding a stock index like SPY or VT.
As we have said in the past, the empirical observation that holding a security that is bullish the dollar could be transitory. We need to think more deeply about why this not simply an empirical artifact.
Keep in mind that we are only looking at a limited set of ETFs and for a limited amount of data (2011-2023).
We will consider the last Return Stacked ETF (RSSB) in subsequent posts.