Long/Short Investments
2
All posts from Long/Short Investments
Long/Short Investments in L/S Equity: Valuation and Ideas,

How Simple Portfolio Construction Can Beat The Market

I believe in diversified portfolios as a matter of principle, but feel it’s especially important when going from the mid- to late-stage of the business cycle as we are now.

Portfolios high in risk assets, such as equities and high-yield, will perform the best 85% of the time. And it is true that the single biggest creator of wealth in the private sector is concentrated equity holdings – and it’s also nonetheless the biggest destroyer of it. Sometimes one might effectively luck out and “get rich” by being in the right companies, but the standard deviation in outcomes will be high.

In general, a high concentration of wealth in risk assets often generates an escalator ride up and an elevator ride down. Stocks generally lose 30%-35% of their value in a bull market and even worse in an economic depression-like scenario (50%-90%).

So to create “alpha” – or risk-adjusted excess returns beyond a benchmark (typically the S&P 500) – you need to pick securities wisely, time the market, and/or improve the portfolio’s risk adjustment. The former two are difficult, but the latter is more doable and not difficult. It revolves around balancing risk and then, if necessary, using moderate leverage to generate the returns desired.

We simply need assets that will do well in all forms of environments – high and low growth and high and low inflation (relative to embedded expectations).

Stocks do best in a risk-adjusted sense when growth and inflation come above expectations. High-yield and REITs do best when growth is above expectations and inflation below. Inflation-hedged assets – e.g., gold, floating-rate bonds – perform best when growth underperforms and inflation outperforms. Safe bonds, such as US Treasuries, do best in more doom-and-gloom scenarios where both growth and inflation are low.

Accordingly, allocating assets equitably such that the portfolio can perform well in all environments – and not just when stocks are going up – is a viable strategy.

For instance, take the following portfolio:

  • Long-Duration US Treasuries (TLT) – 27.5%
  • Long-Duration US TIPS (TIP) – 25.0%
  • Technology Common Stocks (VGT) – 19.5%
  • High-Yield Bonds (JNK) – 15.0%
  • US REITs (VNQ) – 5.0%
  • Gold (GLD) – 5.0%
  • Nustar Energy (NS) – 3.0%

All are ETFs aside from Nustar Energy, which provides diversification into MLPs, income-generating securities tied to the energy commodity markets. I chose an individual security as most MLP ETFs are relatively new and don’t backtest through the recession, which would limit the ability to see how this portfolio would likely perform in a down-cycle relative to a benchmark.

VGT is a tech ETF that has 40%+ of its weight concentrated in FAMG (FB, AAPL, MSFT, GOOG/L). Overall, 63% of the fund is concentrated in the following 15 names:

JNK is diversified across numerous HY bonds, none of which are weighted at more than 0.68% (though multiple securities from the same company are common). HY credit is included from S, WDC, FDC, FTR, VRX, HCA, DLTR, TMUS, among the most notable names.

VNQ invests in US REITs and approximately 43% of the fund is invested in the top 15 securities.

Comparison vs. the S&P 500

Going back ten years, on an unleveraged basis, this portfolio performed about on par with the S&P 500, but at substantially less risk.

  • The portfolio has just 52% of the volatility, which isn’t too bad considering we’re using just seven different holdings.
  • The reward-to-risk ratio is improved to 1.70x that of the S&P 500 using the Sharpe ratio and 1.84x using the Sortino ratio.
  • The maximum drawdown is just 31% of that of the S&P 500.
  • If this portfolio was leveraged up to match the volatility of the S&P 500 it would have returned 14.7% annualized, including the great recession, compared to just 7.8% for the S&P 500.

Conclusion

Portfolio construction can certainly get more complicated than this, but improving one’s reward-to-risk ratio can be as easy as simply including more assets that lack correlation to each other (or at least to a low extent) to avoid biasing the portfolio’s optimal setting to a particular growth or inflation environment.