A Superior All Weather Strategy – AWF Building Blocks (Part 2 of 2)

Alphalytics All Weather Fund Context

The resilience of the balanced portfolio may be largely attributed to a critical non-correlation: In the last 20 years, U.S. equities and U.S. Treasury Bond prices were negatively correlated (most of the time). Simply put, equities and bonds moved largely in opposite directions and served as a good hedge against each other.

A look at the correlation between the S&P 500 and 10-Year Treasury bond yield (when bond yield goes up, bond prices go down) shows that equity is positively correlated to bond yield since 2000.

(Bond prices and bond yield have an inverse relationship. Hence, a positive correlation between equity and bond yield suggests that prices of both asset classes move in the opposite direction; thus equity and bonds provide a good hedge against each other since 2000.)

Source: ACM Research

Looking further back, we see that the relationship between these two asset classes has not always been a positive correlation. Prior to 2000, during periods of rising oil prices and high inflation, equity prices and bond yields were slightly negatively correlated.

Source: ACM Research

Taking this long-range statistical insight, together with 2020 data from the Coronavirus market crash (both equities and bonds dipped in tandem which meant bonds were a lousy hedge), we can advance a research observation:

Because the correlation between equities and bonds is not static, there is no guarantee that the 20-year-old conventional wisdom of holding a balanced portfolio will continue to provide hedged returns.

Hence, Alphalytics research team’s challenge: Is designing a resilient, weatherproof portfolio that thrives for the next 20-year period possible?

____________________________________________________________

Anti-correlation > excess returns

Austin-based Artemis Capital published a research paper about the macro shifts in the long-range economic and monetary environments. One observation in the paper is that anti-correlation is an effective defensive component for a long-term, resilient portfolio. To recap and paraphrase this insight:

Suppose an investor is given the option to buy 2 out of 3 possible assets choices: Assets A, B, and C.

The first two assets (A and B) are highly correlated and both generate positive returns: 15 percent and 13 percent respectively.

Asset C, on the other hand, produces a slightly negative yield of -5 percent but is countertrend to assets A and B. In other words, asset C makes most substantiate gains during periods when the other two assets (A and B) suffer drawdowns.

Which two assets produce the best portfolio? Counterintuitively, combining assets A and C produce a portfolio that generates superior, risk-adjusted returns (10 percent returns and -5 percent drawdown) even when one of the assets (C) generates a negative yield.

Given the much higher return-to-drawdown ratio of Portfolio A + C, one can simply apply leverage to the A + C portfolio to achieve higher returns with lower risk than either Portfolio A + B or holding any individual asset.

Source: ACM Research

Though simple yet effective, many investors (and Portfolio Managers) have yet to fully appreciate or understand the immense value that a defensive asset brings to a long-term portfolio.

As summarized by the Artemis research team, anti-correlation is more valuable than excess returns.

Put simply, because the market is unpredictable, anti-correlation is key to negating nasty surprises and protecting wealth.

Now, the “million-dollar” investment research questions:

• How to construct a A + C portfolio that provides superior, all weather value to investors?;
• Which asset class or trading model is a best-fit defensive holding (asset C)?; and
• How to apply an optimal degree of leverage to a A + C portfolio?

____________________________________________________________

Identifying anti-correlated components (A and C)

To build the A and C components in AWF, we researched a wide range of asset classes and trading models, including Long Volatility.

(Long Volatility trading refers to investment strategies designed to profit during periods of market turbulence or downturn while generating flat or modest negative returns during normal market conditions. Think of Long Volatility as insurance and a counterweight to the broad equity market.)

Here, we highlight only the essence of our research findings. Specifically our analysis of the S&P 500 (rolling 90-day, daily return correlation) relative to four asset classes: bond yield, gold, U.S. dollar, and volatility (VIX as a proxy for Long Volatility).

Our research shows that:

a) Bond yield (inverse to bond price) is positively correlated with equity after the year 2000 – indicating that bonds provide good diversification value to an equity portfolio. However, prior to 2000, that is not the case.

b) Gold has a low correlation with equity in all periods. Post-2008 Great Financial Crisis (GFC), though the correlation between gold and equity turned positive from slightly negative, the correlation remained low – suggesting that gold has consistent diversification value.

c) The U.S. dollar has a relatively low correlation with equity in general. As opposed to gold, the correlation between the U.S. dollar and equity has turned positive to negative after the 2008 GFC.

d) VIX has a consistent, strong negative correlation with equity in all periods – which is not a surprise. Our research reinforces the case for including a Long Volatility strategy. i.e. Long Volatility is an exemplary, defensive asset C.

Source: ACM Research

Next, we study the “cross correlations” between these four non-equity assets.

Rationale: If there is a significant positive correlation between two or more of these asset classes, putting them together in a portfolio will likely diminish any diversification benefits.

Below correlation matrices of these four asset classes indicate that “cross correlations” are not a concern. (For example, gold has either negative or low correlation with various asset classes in all periods.)

Hence, it is safe and not counterproductive to utilize these assets as A and C in our strategy.

Source: ACM Research

____________________________________________________________

Smart beta construction: A + C block in AWF

Having proofed the diversification value of the above-mentioned assets, we propose a Simple All-Weather Portfolio that is rebalanced monthly.

Simple All-Weather Portfolio = A (equities, bonds, gold, U.S. dollar) + C (Long Volatility)

The portfolio construction:

• S&P 500: SPY ETF (SPY) at 20% allocation;
• Nasdaq 100: QQQ ETF (QQQ) at 20%;
• U.S. Treasury Bonds: iShares 7-10 Yr ETF (IEF) at 20%;
• U.S. Treasury Bonds: iShares 20+ Yr ETF (TLT) at 19%;
• Gold: SPDR Gold (GLD) at 5%;
• Currency: US Dollar ETF (UUP) at 15%; and
• Long Volatility: VIX ETN (VXX) at 1%.

This Simple All-Weather Portfolio (SAP) serves as the smart beta block in AWF. (We shall introduce an alpha block shortly.)

Backtest of SAP yields a 2.30 reward-to-risk ratio (CAGR/Max Monthly Drawdown) that is much higher than the 0.62 of the S&P 500. In other words, for each unit of drawdown or value at risk, SAP offers 2.30 times annualised returns.

SAP’s Sharpe Ratio is also significantly higher compared to S&P 500: 1.66 compared to 0.87.

Given that anti-correlation is more valuable than excess return, the high reward-to-risk ratio (2.30) and high Sharpe Ratio (1.66) of SAP allow us to apply leverage safely to achieve our desired annual return target.¹ See the below backtest results of SAP:

Source: ACM Research

Though the SAP portfolio provides a positive return each year (2010 to 2020), it does not meet our desired portfolio criterion of consistent, long-run outperformance of the S&P 500. (SAP portfolio only generated double-digit annual returns 6 out of 11 years.)

To further boost our smart beta SAP portfolio, we shall add an alpha overlay.

____________________________________________________________

Final ingredient: Alpha overlay

The alpha overlay is a set of in-house developed systematic strategies from six time-tested investment styles (adaptive asset allocation, long/short, momentum and reversal, statistical arbitrage, trend following, and volatility trading).

These strategies are implemented using highly liquid ETFs/ETNs and are similar to the flagship AQM strategy.

In sum, AWF invests in 80 percent smart beta and 60 percent alpha. (The fund applies 0.4x leverage or trades 140 percent of its AUM.)

Source: ACM Research

AWF = SAP plus leverage (equities, bonds, gold, currency, Long Volatility) + Alpha component

Backtest of AWF, inclusive of transaction costs, demonstrates a reward-to-risk ratio (CAGR/Max Monthly Drawdown) that is much higher compared to that of the S&P 500.

AWF offers a reward-to-risk ratio of 2.8 (20.0/7.2) while the equity market only produces a ratio of 0.68 (13.5/20.0). In other words, for each unit of drawdown or value at risk, AWF offers 2.8 times annualised returns (gross) – four times greater risk-adjusted returns than the S&P 500.

The Sharpe Ratio of AWF is also significantly higher compared to S&P 500: 1.80 compared to 1.01 which indicates that AWF is a lot less volatile relative to the market.

Furthermore, AWF seeks to generate positive returns year on year – even in 2011, 2015, and 2018 when the equity market is either flat or negative.

Source: ACM Research

Investors who are risk-averse prefer bonds while those willing to bear more risk, in exchange for higher returns, prefer equities.

AWF aims for best-of-both-worlds value creation:

Higher returns than the S&P 500 at a comparatively low risk like a bond portfolio (AGG).

Source: ACM Research

Furthermore, a historical drawdown analysis of AWF’s performance during each major market downturn (since 2010) shows excellent protection against large market corrections.

Source: ACM Research

____________________________________________________________

Alphalytics All Weather Fund: The best-of-both-worlds solution

In the short term, the equity market is unpredictable.

What has been relatively predictable is that, over the long run, the equity market continues to generate long-term outperformance despite world wars, the Depression, and the global financial crisis.

The predominant trade-off to equity’s long-run outperformance is increased risk; this risk has, for the past 20 years, been hedged by bonds. But the negative correlation between equity and bond is no sure thing and, based on ACM research, potentially a more blunt, less useful hedging instrument moving forward.

AWF grew out of our research effort to smoothen our investors’ journey: How can our investors consistently capture the outperformance of equities and also reap the stability of bonds?

We set a high bar for AWF and the primary benefits to investors are:

1. Better – seeking higher – returns than equities;
2. Similar stability to bonds; and
3. Positive returns each year (with low correlation to all major asset classes).

The resulting AWF all weather strategy consists of a smart beta component with an alpha overlay.

Smart beta is insurance to weather all market conditions – think of this as bonds on steroids. Components in the smart beta block are anti-correlated assets (equity, bonds, gold, currency) and a Long Volatility trading model.

The alpha overlay is a basket of time-tested strategies designed to capture excess returns – similar to the high-performance Alphalytics Quant Multi-Strategy (AQM) model.

In sum, for investors seeking capital preservation and equity-like performance, compared to most all weather funds, investors switch to Alphalytics All Weather Fund because of better performance and greater stability.

____________________________________________________________

Email hello @ alphalyticscm dot com to learn more.

The above results are not an indicator of future results. Please reference our disclaimer.