Using the Ivy-20 asset class portfolio as a model, what is the performance of this portfolio when equal allocations are assigned to each ETF? How does that performance compare with a risk-adjust asset allocation model? Does the timeframe make a difference as to which approach to take when constructing a portfolio? To answer these questions, I ran Quantext Portfolio Planning analysis on the Ivy-20 Portfolio to find answers.
A few ETFs included in the Ivy-20 did not have five years of data. This required condensing the period of analysis to 50 months vs. the preferred time of 60 months. I wanted to bridge the last bear market as risk-adjust portfolios are expected to outperform capitalization asset allocation style portfolios in down markets such as we experienced in 2008 and early 2009.
Capitalization Asset Allocation: In the first screen shot, equal percentages are allocated to each ETF as advocated in Mebane T. Faber and Eric W. Richardson's book, The Ivy Portfolio. While future projections are interesting, in this analysis, we want to focus on the Historical Data section of the slide. Using capitalization asset allocation, this portfolio performed a little below the S&P 500 (without dividends) with similar standard deviation values. Note that this data begins near the middle of the 2008 bear market.
Risk-Adjusted Asset Allocation: When the various ETFs are risk-adjusted, the portfolio performs a little better than the S&P 500 with much lower standard deviation. This is what we expect to see, as BND (NYSEARCA:BND) and TIP (NYSEARCA:TIP) did not decline as much as the equity ETFs. Downside protection is achieved when lower volatile ETFs are assigned a larger percentage of the portfolio.
What happens when we move out of the bear market, period?
Thirty-Six Months Capitalization Asset Allocation (NYSE:CAA) Model: Over the past three years, we have experienced a bull market, and it is reflected in the following analysis. With a 9.7% annualized return, the Ivy-20 was not able to match the 11.9% (no dividends) performance of the S&P 500, but it does outperform the risk-adjusted portfolio shown in the fourth screen shot. This is exactly what one would assume, as the more volatile ETFs are assigned a higher percentage of the portfolio.
Risk-Adjust Portfolio (36-months): Adjusting the ETFs so each carries approximately the same amount of risk, return drops a little when compared to the CAA model, but with much lower risk. The Return/Risk ratio is superior for the risk-adjusted portfolios.
The question facing the investor or money manager is which approach to take when setting up a portfolio? Or should one set up a Tactical Asset Allocation model such as the ITA Risk Reduction model described on this site? The ITARR model is a slight variation of the Faber-Richardson model explained in their book. The Faber-Richardson timing model is designed to let equities run in bull markets, but go to cash or risk protected investment in bear markets.