yawning

Olstein All-Cap Value? Nothing Special Once We Take A Look Under the Hood

yawning

The active fund industry is notorious for deceiving investors. We mean deceitful in terms of not giving investors the full-picture. To give an analogy, if we told you that we had magical powers that allowed us to make objects fall out of the sky and charged a fee to prove it, there might be a couple of people that would pay to see the show. We could simply throw an apple up in the air and say “abra cadabra” and have it come back down to earth. While most people would say, “well that is just gravity, nothing magical at all,” we would still get paid for showing you nothing special. The same can be applied to the performance of certain “all-star” funds. Once we account for the risks that we expect to be compensated for, the “magic” seems to disappear. 

The Olstein All Cap Value Fund is a pretty blatant example. We are going to show you how the media may overhype a certain fund’s performance, but when carefully measured, that excess performance is nothing more than what we would expect given the risk exposures of the fund. A recent article was published on CNBC entitled “Olstein: Index Media Misleading Investors.” They go on to give the conclusion that, “See investors! We can outperform the market,” citing their Olstein All Cap Value Fund’s performance from 10/1995 to 04/2015. Specifically, they claim, “Since its inception at year end 1995, Olstein’s All Cap Value Fund has outperformed the S&P 500 for 12 years...”

An investor could have gained access to a slightly higher return and about the same risk by investing 35% in a large value index fund and 65% in a small value index fund at a cost of only 0.35% per year, as opposed paying the Olstein All Cap Value Fund C Shares 2.28% per year.

Expense Ratio Comparison
Fund Name Weight Exp Ratio
DFA US Large Cap Value I 35% 0.27
DFA US Targeted Value I 65% 0.40
Weighted Net Expense Ratio 0.35
 
Olstein All Cap Value C -- 2.28

As you can see in the scatter plot below, the combination of just two Dimensional Funds had a return of 11.93%, while the Olstein Fund had a return of 10.98%. Both had a standard deviation of 18.7% over the 19 years and 8 months, which is a time period based on the 10/1/1995 inception date of the Olstein fund. You can also see that the S&P 500 had a risk measurement of 15.26% standard deviation compared to 18.7% for Olstein. This is one indication that the S&P 500 is not the best benchmark for the Olstein fund.

We often state the the missing link in investment analysis is proper benchmarking. As an example, CNBC compared the Olstein “All-Cap Value" fund, which is made up of value companies from the large, mid, small, and micro cap companies to the large cap blend asset class of the S&P 500 index.

To give a visual representation of the types of companies this fund has been investing in, the chart below illustrates the “style drift,” or the variation in the asset classes that the fund manager has exposed the fund to over time. As you can see, the proper benchmark for this fund has been constantly changing and one benchmark index does not capture the risks taken. Early in the fund’s life the manager was speculating on the entire spectrum of market capitalizations, which is consistent with the fund’s "All Cap" name, as well as sticking to value stocks across both large and small companies. But in 2003, there were almost no large cap value stocks and instead there were both small cap growth and large cap growth stocks. In the middle of 2005, the fund had significant allocations to both large cap value and small cap value, which carry different expected returns. Recently, the fund had almost no small cap or large cap value and mainly consisted of large cap growth stocks.

Investors should be concerned with the fund manager’s ability to be in the right asset class at the right time because the rotation of asset class winners has been very difficult to forecast.

Next, we are going to look at the annual variance in the funds’ alpha overtime relative to the Russell 1000 Index (A large cap blend benchmark assigned to the fund by Morningstar). The Alpha Chart below shows the relative under and over-performance relative to the Morningstar assigned benchmark since inception. It is clear to see that most of the excess return earned by the manager was from 1999 to 2001, with 8 of the last 13 years showing negative alpha. The volatility of the annual alpha is quantified by the 10.06% standard deviation of alpha. But the final nail in the coffin of active managers is the t-stat, which considers both the average and the standard deviation and indicates the existence of skill or luck. If the t-stat is less than 2, luck is the most likely explanation of the alpha. Olstein's t-stat was 1.17.

Now that we have a visual representations of how the fund manager has changed the fund’s style overtime and how alpha varied over time, let’s take a look at what is called performance attribution. Going back to our previous analogy, let’s see how much of the performance is “magic” and how much is due to gravity. We can run a what is called a multiple regression analysis of the Olstein All Cap Value Fund’s monthly returns since inception against the known risk factors (which can be bought with index funds) and their associated premiums. Premiums are the excess returns that investors are expected to be compensated for in different markets; namely, the small-cap premium (about 3%) and the value premium (about 5%). What we are trying to find out is how much excess return did the fund manager add (magic) on top of what investors were expected to be compensated for just by being in certain asset classes of the market (gravity). We will first show you the outcome of the multiple regression and then explain all of the variables.

Three Factor Regression
Fund Alpha
(Expected Excess Return over the 3 factors)
Market Exposure Small Cap Exposure Value Exposure R2
Olstein All Cap Value 0.07%
(with a t-stat of 0.5)
1.07 0.24 0.22 0.85

 

Here is an explanation of the variables:

  • Market Exposure is one of our gravity variables. It measures how the Olstein fund moves with the entire stock market. A coefficient of 1.07 means that it moves 107% relative to the overall market. So if the market goes up by 10%, the fund is expected to go up by 10.7%. If the overall market goes down by 10%, then the fund is expected to lose -10.7%. 
  • Small Cap Exposure is another gravity variable. We are expected to be compensated for investing in small capitalization companies versus large capitalization companies. You can find more information in Step 8 of our 12-Step Program for Active Investors. A positive number (0.24) for this variable means that the fund has primarily invested in, or tilted towards smaller capitalization stocks over period of the analysis. The premium for small cap exposure has been about 3% per year, so the excess expected return over the long term for this exposure would be 0.24 x 3% = 0.72% annualized.
  • Value Exposure is the last of our gravity variables. We are expected to be compensated for investing in companies with high book value-to-market ratios (value) versus companies with low book value-to-market ratios (growth) over the long term. A positive number (0.22) in this field indicates that the fund has primarily invested in value stocks over time, but as seen above, the fund has switched back and forth between value and growth at various times. A value tilt of 0.22 times the value premium of 5% equals an expected excess annualized return of 1.1%.
  • R2 is a measure of how much this overall regression captures the variation fund’s performance overtime. An R2 of 0.85% means that the market premium, the small cap premium, and the value premium explain 85% of the overall variation in the Olstein All Cap Value Fund’s returns, which is very high. Other random variations can come from the outcomes of the manager’s asset class timing as well as individual stock selection.
  • Alpha is the metric we care about the most. It measures the fund manager’s “magic.” This number indicates how much value the manager added above what we would expect to be compensated for from our gravity variables. As you can see, the estimate of the alpha is 0.07%, which means that the fund manager added 0.07% per year in return above what was explained by our gravity variables. Unfortunately, this estimate is not statistically significant with a T-Stat of 0.50. This indicates that we cannot be confident that the fund manager added any value to the fund’s overall performance and the excess return would be something we would expect from chance alone. It is no wonder that SEC requires warnings to investors that past performance is not an indication of future performance.

At this point, we might have lost a few of our readers, but the implications of this are extremely important since there is much at stake. We have shown that although CNBC may have highlighted a fund that seems to have outperformed the entire market, it really hasn’t done anything all that special. The reason for its performance versus that of the S&P 500 is because it exposed investors to risk factors with known risk premiums that investors are expected to be compensated for. The fund focused mainly on smaller companies and companies with high book value-to-market ratios, otherwise known as value stocks. We have known for decades that these stocks had a higher expected return than the overall market because they are riskier. No magic involved here.

While many managers will continue to make claims that their actively managed fund may justify a higher fee than a combination of lower cost index funds, we really have yet to find anything special once we have opened up the hood and taken a deeper look.