In last week’s post I talked about a new fund – QSPIX from AQR. It is a long/short fund that seeks to capture four style premiums across multiple asset classes and countries. Since the return of this fund is derived from sources different from traditional stock and bond returns, adding this fund to our existing portfolios has the potential to reduce volatility while keeping the same, or higher, returns.

Below is a chart of the new fund, QSPIX, from 1/1990 through 9/2013. These returns are compiled from back testing the strategy. The returns are then discounted by conservative estimates of transaction costs and the fund’s expense ratio. This adjusted data, in my opinion, gives us the best insight about what to expect going forward.

The comparison between QSPIX and the Russell 3000 Index (an index closely resembling the total US stock market) and the Barclays US Aggregate Bond Index (an index closely resembling the total US bond market) is shown below.

Part 2-1
Just by looking at the chart you can see that the 3 lines don’t correlate very well and move up and down at different times. The bond index, not surprisingly, is much a smoother line than the other two, but it has a lower return. The stock index and QSPIX have higher returns, but are much more volatile. As you can see in the late 90’s QSPIX wouldn’t have performed very well, but recovered nicely while the tech bubble burst. Also note the small loss in 2008 which highlights the funds market neutrality. Academically speaking, QSPIX has a correlation to the Russell 3000 Index of 0.009 and a correlation to the Barclays US Aggregate Bond Index of 0.128, essentially uncorrelated on both accounts.

Building Portfolios

To keep things simple we will stick with US stocks and US bonds. A simple portfolio of stocks and bonds is one consisting of 50% allocated to the Russell 3000 Index and 50% allocated to the Barclays US Aggregate Bond Index (rebalanced annually). Below is a chart of this portfolio from 1/1990 through 9/2013.

Part 2-2
Part 2-3
As you can see this portfolio provided a pretty steady stream of returns since 1990. However, as I mentioned in last week’s post, this portfolio’s return is completely dominated by the return of the stock market. In fact about 93% of its return can be explained by its stock allocation. There is strong evidence that suggests the expected return of stocks and bonds in the future is less than their historical returns.

Now let’s see what happens when we add a third asset to the portfolio. This portfolio consists of a 34% allocation to the Russell 3000 Index, a 33% allocation to QSPIX and a 33% allocation to the Barclays US Aggregate Bond Index (rebalanced annually).

Part 2-4
This new portfolio had a better return with less volatility, as we would expect when adding an uncorrelated asset to the mix. Below is a table showing the statistics of the two portfolios.

Part 2-5
We were able to increase the return of the portfolio by 3.6% while reducing volatility by 18%!
One more step to add efficiency to the portfolio is to tilt the stock allocation towards other style premiums. Here we will replace the 34% allocation to the Russell 3000 with the AQR US Large Cap Core fund which captures the momentum, value and profitability premiums as well as the entire equity risk premium. Below is a graph of this new fund and our basic 50/50 portfolio.

Part 2-6

Part 2-7
By taking this two-pronged approach to capturing additional premiums (through adding an uncorrelated return stream with QSPIX and enhancing the equity portion of the portfolio by capturing risk premiums in addition to the equity premium), we were able to increase our return 15.7% and decrease our volatility by 12.5%. Additionally we were able to cut our reliance on the stock market dominating our risk in the portfolio. With our new portfolio, about 55% of its return can be explained by stocks compared to 93% for the basic 50/50 portfolio. While this characteristic is great, it also makes it very difficult to own.

To illustrate this point let’s look at a chart of our new portfolio compared to a common benchmark we hear in the news the S&P 500 (includes reinvesting dividends).

Part 2-8
As you can see our new portfolio is vastly superior, providing consistent returns over a long period of time, higher than that of the S&P 500. This is especially important for people nearing or in retirement. Taking withdrawals after large losses severely impacts the long term wealth of your portfolio. It wasn’t so easy for our hypothetical investor holding this portfolio though. Let’s zoom in on the 5 year period of 1/1995-12/1999.

Part 2-9
Our new portfolio “only” returned 15.11% per year while the S&P 500 returned 28.56% per year. The NASDAQ, which was in the news even more during that time, had much higher returns. Would you have had the discipline to stick to this portfolio and not get caught up in the hoopla of tech stocks? This is known in finance as tracking error risk, or the risk an investor will abandon their portfolio after it “fails” to keep up with a benchmark, even if the comparison is inappropriate.

Of course tracking error goes both ways. Let’s look at the 13.75-year period from 1/2000 through 9/2013.

Part 2-10
Tracking error looks pretty good to us in this scenario with our new portfolio returning 8.01% per year versus 2.91% per year for the S&P 500. Another advantageous time for this portfolio was during the crash of 2008-2009 where the S&P 500 lost over 50% of its value from peak-to-trough versus less than a 20% loss for our new portfolio.

The Take Away

As you can see adding this new mutual fund has the potential to give our portfolios a new stream of returns completely separate from traditional stock and bond returns. This can increase returns while reducing volatility.

In strong years of the stock market, like 2013, a well-diversified portfolio is all but guaranteed to underperform equity benchmarks. The best way to separate yourself from most investors is to ignore the short-term ups and downs of the stock market and stick to your well-developed, long-term plan. Don’t confuse strategy with outcome, especially over short periods of time. If you plan on adding an allocation to a fund like QSPIX make sure you do thorough research (reading these last two blog posts doesn’t count as “thorough”!). Only after fully understanding these funds will you have the wherewithal to stick with these funds during periods of stress, particularly when those periods coincide with strong equity years.