The Ivy Portfolio strategy has been popularized by the book "The Ivy Portfolio" and its authors Mebane Faber and Eric Richardson. We have been tracking and working with the strategy since late 2011 and believe it is a worthy strategy for investors to consider.
The strategy has its own unique twists compared to a traditional buy-and-hold portfolio, which might make it unappealing for some investors. For those investors willing to hold cash, realize losses at times, and work on a total return basis - then this can be an attractive proposition. It is a straightforward and elegant strategy that is not mired in financial nuances to achieve its goals. And from our vantage point it achieves those goals utilizing the principle "Keep it simple, stupid" (or KISS for short).
The Ivy Portfolio strategy is a total return strategy based on the combination of a diversified portfolio and a risk management system. The diversified portfolio is based on multiple asset classes including stocks, bonds, and inflation-linked assets. The risk management system is a rules based system that uses a trend system to determine whether to hold or sell an investment. The combination of diversification and trend following lowers the volatility of the portfolio and minimizes large declines in the portfolio.
The strategy can be tailored to specific investor requirements making it a highly flexible portfolio strategy. The level of diversification, the defining of the length of the trend, and the rebalancing schedule are all flexible and determined by investor preferences. Do note that the total return focus is largely due to the fact that as any asset can be bought or sold in this system, planning for a consistent stream of income (dividends or interest income) is problematic.
The goal of the portfolio strategy is to deliver a respectable total return over an entire market cycle compared to alternative investment strategies. We use two simple portfolio measurements to measure the success of the portfolio strategy. The Sharpe ratio measures the attractiveness of the strategy on a risk-adjusted return basis, which factors the excess return over a risk free return and the realized volatility to generate that excess return. Secondly we measure the downside protection by calculating the portfolio's historic maximum peak-to-trough declines or maximum drawdown (MaxDD).
Risk Management Rules
The unique feature of the strategy is the risk management system. The rules can be altered but should follow the general framework:
- At a set periodic interval (typically monthly) compare the price of each investment (either held directly or allocated but not currently held) to its moving average.
- For each investment that is owned, if the price is above the average then continue holding. If the price is below the average then redeem and hold cash.
- For each investment that is held in cash (allocated but previously below trend), if the price is above the average then purchase. If the price remains below the trend, continue holding cash.
These rules are about as simple as it can get and clearly adheres to the KISS principle.
Sample Portfolio Review
The Ivy Portfolio strategy is ultimately a framework that can be tailored to each investor's preferences. While each individual portfolio can have its own twists, we will follow the KISS theme and review a straightforward 6 asset class, diversified portfolio.
In honor of the Ivy Portfolio namesake, our portfolio construction will come from David Swensen, Chief Investment Officer at Yale University, as described in his book "Unconventional Success: A Fundamental Approach to Personal Investment" (and found on page 84).
Figure 1: Swensen 6 Asset Allocation Model Portfolio
Our preference is to utilize exchange traded funds (ETFs) where possible. When an ETF is not available due to limited trading history, we utilize mutual funds.
The rules based system is based on end of month pricing and the 10-month moving average (10-MMA). [Investors can choose any length trend from 7 months to 12 months. Because the 10-MMA is similar to the 200-Day MA, which is widely followed, utilizing a different length trend such as the 9-MMA might avoid some of the overlap and/or whipsaws that can happen in choppy or non-trending markets.]
Figure 2 highlights the monthly closing price (adjusted for dividends) for the Domestic Stock Fund (VTI) and its 10-month moving average. When the monthly closing price (blue line) drops below the 10-MMA (red line) then it is sold and converted to cash.
Figure 2: Domestic Stock Market Fund
Figure 3 highlights a theoretical performance of the strategy based and compared to a traditional balanced fund. We use Vanguard's well-regarded Balanced Fund (VBINX) for our proxy. The model portfolio includes approximately 1% in annual friction, which is arbitrary but should accommodate some costs associated with commissions and/or other fees. (Taxes are not considered in this exercise.) Lastly, the portfolio is rebalanced quarterly.
Figure 3: Model Return of Swensen 6 Asset Class Portfolio
The next figure illustrates how the risk management system works over time. This portfolio strategy is currently at 34% in cash. The average cash percentage held over the past 8-year period has been ~22%. In general as the markets rise, the portfolio becomes fully invested. As volatility increases and markets trend down, the portfolio shifts to cash. Due to the lagging nature of the risk management system, choppy or non-trending environments will cause underperformance relative to buy-and-hold strategies.
Figure 4: Portfolio Mix of Investments and Cash
Looking at monthly returns, the worst month over the past ~8 years was in October 2008 at a negative 6.5%, with the month of May in 2010 taking second place at -5.3%. The monthly return data shows a positive skew with the bulk of returns in the 0%-3% range.
Figure 5: Monthly Returns
The Swensen 6 asset class portfolio performance has been on par with a balanced fund return over the past 8 years. Based on the goals of the portfolio - the strategy has been a success. First, the Sharpe ratio of 0.50 is higher than the balanced fund Sharpe ratio of 0.36. (The Sharpe ratio is the compounded return net of the risk free rate divided by the annualized volatility.) Given the returns are similar - the higher Sharpe ratio is largely driven by the 30% lower realized volatility of the fund. We used the compounded annual return of the iShares Barclay's 1-3 Year Treasury Bond (SHY) as a proxy for the risk free rate.
Secondly, the maximum drawdown (MaxDD) of the portfolio was significantly less than the balanced fund at 9.8%, which occurred from October 2007 to October 2008. The balanced fund, which was assumed held throughout the 2007-2009 market down turn, had a maximum decline of 32.6%. While not shown in the data - the minimization of the drawdown has important behavioral finance implications and removes a fair amount of fear during market duress and subsequent panic selling.
Figure 6: Swensen 6 Asset Class Portfolio Risk and Return Metrics
As the equity markets rally higher and money continues to flow into stock mutual funds and ETFs - the appeal of risk-managed portfolio strategies diminishes. This is a natural phenomenon, especially in a low volatile investment climate for equities, as recent performance is king. Unfortunately, the attractiveness of any risk management strategies is typically seen in hindsight.
However, the Ivy Portfolio remains an attractive strategy for long-term investors looking for a full market cycle investment strategy that mitigates monthly volatility and downside risk. Its KISS design principle is straightforward and does not require any heavy lifting to implement. If modeled performance of the Swensen 6 asset class portfolio is any indication, the strategy should deliver on its goals of attractive risk-adjusted returns and mitigation of large declines in the portfolio.
Additional disclosure: Clients of Smith Patrick Financial Advisors own VNQ at the time of this writing. This article is written for informational purposes and we believe investors should perform their own due diligence before making investment decisions.