A “Protective” Alternative to the 60/40 Portfolio

By Justin Carbonneau (@JJCARBONNEAU)

As this Forbes article points out, the 60/40 portfolio is “one of the most dominant investment approaches of our time”. This asset allocation is incredibly simple and it’s worked, but with U.S. stocks up nearly 350% off the 2009 lows and interest rates at historical lows, some investors are wondering if the 60/40 stock/bond portfolio has the same risk reduction and return benefits as it has in the past. This article presents a conservative and time-tested strategy that could act as an alternative to the 60/40 portfolio and appeal to those investors who are seeking more a dynamic, flexible and active investment approach.

The Past, Present and Future of the 60/40 Portfolio

Over the past 92 years, a 60/40 stock/bond portfolio allocation has averaged a return of 8.63% a year according to Vanguard.  That’s a very respectable return and equates to more than a doubling of one’s money in a little over 8 years.

The returns in the last ten years have been even better. Through the end of October, the Vanguard Balanced Index Fund, which tracks a 60/40 mix, compounded at an annual rate of 9.65% over the past decade – which is a double of one’s money in about 7.5 years.

That’s the good news.  

The possible bad news is that returns over the next decade don’t look so compelling. Morgan Stanley recently put out a report projecting the returns for a vanilla 60/40 stock/bond allocation to be just shy of 3% a year over the next decade. What’s more, Research Affiliates estimates US large caps stocks will return about 2.6% per year in the next decade and US Intermediate Treasury Bonds 2% over the same time period (see returns here).

Projecting returns can be a tough business, but extrapolating historical results out into the future blindly without regard to the existing environment could also be bad. If returns are in fact far lower than in the past, many investors allocated to 60/40 balanced funds are likely to be disappointed.

In this article, I present an interesting alternative to a standard 60/40 allocation for investors seeking something different, unique and less susceptible to large declines, bear markets or risk from high inflationary environments. To be clear, this is not investment advice, and I’m not knocking the 60/40 portfolio, but the consideration of alternative approaches could be good for some investors.

Protective Asset Allocation

In 2016, a research paper titled “Protective Asset Allocation (PAA): A Simple Momentum-Based Alternative for Term Deposits” written by Wouter Keller and Jan Willem Keuning outlined and tested the an investment methodology that combines asset allocation with momentum investing concepts.

The authors write:

In this paper we describe a simple dual-momentum model (called Protective Asset Allocation or PAA) which a vigorous crash protection … [PAA] is a “tactical variation on the traditional 60/40 stock/bond portfolio where the optimal stock/bond mix is determined by multi-market breadth using dual momentum.”

Momentum, or dual momentum in this case, is a lynchpin in the Protective Asset Allocation (PAA) strategy.

Keller and Keuning explain the two types of momentum criteria – absolute momentum and relative momentum. Given momentum is a centerpiece of this PAA approach, it’s important to define these upfront.

The first type of momentum is called absolute momentum. Absolute momentum involves judging a security’s momentum using its own history. A good example of absolute momentum is using the 200-day moving average, which is one of the more common indicators used. With absolute momentum you can use any lookback period you’d like – the 50 day, 200 day, 250 day – but the key idea here is when the asset or asset class falls below the moving average you go to cash or reduce risk, and when it closes above you re-enter the position.

The second type of momentum is called relative momentum. With relative momentum, you are comparing different investments based on their relative performance over a period (say the last 12 months) and selecting those assets with the highest relative returns. So, as an example, if you wanted to build a 10-stock portfolio consisting of those names with the highest relative momentum you would screen the universe based on your relative momentum factor and buy the stocks in the top 10. Relative strength is the most common relative momentum indicator.

The Asset Classes

Before the actual momentum criteria is applied, we need to define the universe. In the paper, the authors identify 12 distinct asset classes ranging from U.S. and International equities, multiple bond categories and assets like gold, REITs and commodities.

S&P 500
the Russell 2000
the NASDAQ 100
European Equities
Japanese Equities
Emerging market equities
Long-term treasury bonds
High yield bonds
Corporate bonds
Commodities
Gold
Real estate

Crash Protection

One of the unique features of the model, which Keller and Keuning write is the “main innovation of our paper”, is that the portfolio can move to a “crash protection” position by moving part or the entire portfolio to a bond or cash position if a certain number of the assets in the list above display negative momentum. 

The idea here is that as the assets lose momentum and decline, there is a correlation between them and this weakness across multiple assets acts as a “multi-market breadth indicator”. They call this part of the strategy “Protective Momentum” since it is using momentum to protect the overall portfolio from major market declines and bear markets.

The Strategy

The authors ran various iterations and optimizations of the model over periods of time. They tested multiple momentum lookback periods (3-12 months), different size portfolios in terms of the number of positions (1-6 holdings) and even various levels of protective momentum used in the crash protection component (low, medium and high).

Below is a summary of the final set of criteria used in the PAA model:

  • # of Asset Classes When Invested: Six
  • Momentum Indicator: 12-Month Momentum Lookback
  • Position Weighting: Equally Weighted
  • Rebalancing Frequency: Monthly
  • Crash Protection: Six or more of the assets need to have negative momentum – at this point model moves to 100% crash protection positioning

Protection + Compounding is Powerful

The table below shows the combined in and out-of-sample results of PAA from 1980-2015 and it shows the returns (R), volatility (V), max drawdowns (D), Win Rates – Win0 and Win5, Sharpe ratio (SR), and risk adjusted returns (MAR), which is return (R) divided by max drawdown (D). The last column shows a $100 investment in each of these models, along with the S&P 500 and a 60/40 allocation since 1970. Not only are the returns of PAA higher than a stock-only or a 60/40, but the control of downside risk is also substantially better.

Full Sample period (Dec 1970 – Dec 2015)

Validea’s Implementation

Validea has introduced the PAA strategy in our ETF model portfolio tool (click here to view the tool) and we were able to test the strategy as far back as December 2006. The test period is important since we get the full bear market of 2008/2009 and the subsequent recovery. Within that recovery, we have years like 2011, 2015 and 2018 where stocks pulled back to near bear market levels. What we didn’t see in this period, is a rising or high inflation environment, and that would certainly have an impact on the fixed income portion of the PAA allocation.

A few things stand out:

  • First is the PAA model beats a 60/40 allocation by about 2%/year over the twelve years in our testing. Much of that outperformance comes from controlling downside risk. In 2008, for example, when the S&P 500 was down 38% and a 60/40 portfolio fell over 12%, the PAA model produced an exceptional return of +14.7%.
  • The second observation is in transition years, like 2009, when the market went from the depths of the bear market into the beginning stage of a bull market, it can be difficult for strategies that incorporate momentum since the positive momentum has to materialize before re-entering. As a result, the model trailed the 60/40 allocation by a wide margin.
  • The model produced positive results in all years except one, and that one year the portfolio was only down 0.1% (so effectively flat). These returns don’t take into consideration trading costs or taxes, but the model, at least in our testing period, demonstrates the ability to produce positive returns time after time.
PortfolioInception DateReturn60/40 Portfolio+/- 60/40BetaStd Dev.Max DrawdownSharpe Ratio
Protective Asset Allocation12/29/20068.10%6.20%1.90%0.111.40%17.90%0.54
YearPAA60/40
200718.00%6.20%
200814.70%-12.10%
2009-0.10%12.60%
201020.00%10.00%
20110.40%6.20%
20124.80%8.60%
20139.00%10.50%
20145.20%8.90%
20151.40%0.80%
20166.60%6.40%
201716.20%10.50%
20183.10%-1.30%
2019 YTD8.10%15.50%

Conclusion

For many investors, a 60/40 balanced allocation acts an anchor for their portfolios. Rather than making changes, this straightforward and sensible approach keeps investors on track and removes the temptation for tweaking their investments to look for something better. However, many investors need their investments to grow and if the 60/40 allocation fails to deliver, it could be detrimental for some. To that end, educating yourself on alternative approaches that are supported by performance and historical evidence, both from inside and outside the academic world, can be valuable. Protective Asset Allocation is one such approach that is worthy of consideration.  

Photo: Copyright: 123rf.com / lightwise


Justin J. Carbonneau is Partner at Validea Capital Management and Validea.com. You can follow Justin on Twitter @jjcarbonneau.