Analysis of First Eagle Fund of America
August 24, 2014
Analysis of MainStay ICAP International Fund
Our focus today is the First Eagle Fund of America (FEFAX; class A shares), profiled in a Barron’s story a few weeks ago. This $3.2 billion fund has a maximum sales charge of 5.00% and a total expense ratio of 1.42%. (With a 2.17% total expense ratio, class C shares are not attractive; class I shares require a minimum $1 million initial investment; class Y shares are closed to new investors.) According to the article, the fund’s performance has been quite remarkable:
Over the 15 years through Aug. 7, it’s up an average of 8.69% a year—versus 4.54% for the Standard & Poor’s 500. More impressive, perhaps, is the fund’s relative performance in down markets; since inception its losses have been about three-fourths that of its peers, according to Morningstar.
The primary benchmark of the First Eagle Fund of America is the S&P 500® index, one of a few accessible implementations of which is the SPDR® S&P 500® ETF (SPY). However, this may not be the best benchmark for the fund, which has a strong mid-cap tilt (currently 42.5% of holdings):
The fund invests in midsize to large companies; recent holdings ranged from the $1.1 billion biopharmaceutical firm Halozyme Therapeutics (HALO) to the $184 billion drug maker Pfizer (PFE).
In the simplest Alpholio™ analysis, both the membership and weights of ETFs in the reference portfolio are constant throughout an evaluation period. Such an analysis spanning almost 10 years to July 31, 2014 shows that the reference ETF with a dominant weight of 31.2% was the iShares Morningstar Mid-Cap Growth (JKH). The equity ETFs with the next highest weights in this analysis were the PowerShares Dynamic Market Portfolio (PWC; 12.4%) and the iShares Morningstar Mid-Cap ETF (JKG), both of which are mid-cap. Clearly, a blended large-cap index is not the best reference for the fund.
The fund’s website shows the following table of indices which, presumably, could be used as alternative benchmarks:
Given the fund’s mid-cap inclination, the Russell Midcap® Value index stands out as the most appropriate one. A practical embodiment of this index is the iShares Russell Mid-Cap Value ETF (IWS). Alpholio™’s calculations show that the fund returned more than the ETF in only 47.6% of all 12-month rolling periods since the end of 2004. In contrast, the same figure against the SPDR® S&P 500® ETF (SPY) was a much higher 75.2%.
Despite being fairly concentrated (currently 39 positions with top ten encompassing almost 46% of holdings), the fund exhibited a significantly smaller volatility than all of its alternative benchmarks:
In a more elaborate Alpholio™ analysis, the membership of ETFs in the reference portfolio is constant, while ETF weights can fluctuate to better match the composition of the fund over time. Here is a chart of the resulting cumulative RealAlpha™ for the fund:
From early 2005 through 2009, the fund generate a modest amount of RealAlpha™. The cumulative RealAlpha™ picked up in 2010 and from 2012 onwards. Overall, the fund delivered about 3.3% of annualized discounted RealAlpha™. The lag RealAlpha™ curve was, for the most part, below that of the regular RealAlpha™, which suggests that not all new investment ideas for the fund worked out as well as intended. At about 14.1%, the standard deviation of the fund was about 1% higher than that of the reference portfolio. This indicates that superior returns came at a price of increased volatility. At about 0.81, the RealBeta™ for the fund was quite modest.
The following chart illustrates the changes of ETF weights in the reference portfolio for the fund over the same analysis period:
The fund’s top equivalent equity positions were in the iShares Morningstar Mid-Cap ETF (JKG; average weight of 17.7%), Vanguard Health Care ETF (VHT; 15.0%), iShares Morningstar Mid-Cap Growth ETF (JKH; 11.2%), Vanguard Materials ETF (VAW; 10.9%), and PowerShares Dynamic Market Portfolio (PWC; 6.9%). An equivalent position in the iShares 1-3 Year Treasury Bond ETF (SHY; average weight of 17.5%) approximates fixed-income investments of the fund. Finally, the Other component in the chart collectively represents five other ETFs with smaller average weights.
In sum, the First Eagle Fund of America delivered solid risk-adjusted returns over the past 10 years. While the fund’s volatility (standard deviation) was lower than that of its featured single-index benchmarks, it was higher than that of its reference ETF portfolio.
To learn more about the First Eagle Fund of America and other mutual funds, please register on our website.
August 16, 2014
Analysis of Vulcan Value Partners Fund
In today’s issue, Barron’s profiles the MainStay ICAP International Fund (ICEUX, class I shares; ICEVX, class A shares). According to the article
Over the long haul, the fund has a stellar record, beating 87% of its peers over 10 years, and 76% over 15. Senser has helped lead the fund since its 1997 inception.
First, there is no mention of the portfolio manager who left the fund in early July 2014, after an almost 27-year long tenure. Hence the effort to convey a sense of management continuation by using the “has helped lead” phrase.
Second, the article uses class I shares as a basis for its evaluation of the fund. This share class requires a minimum $5 million initial investment, which is impractical for most individual investors. Clearly, such a high threshold is designed to convert a former accessible no-load share class to an “institutional” share class:
Effective 9/1/06, ICAP International Fund was renamed MainStay ICAP International Fund. At that time, the Fund’s existing no-load shares were redesignated Class I shares.
These shares have a lower expense ratio than the class A shares (0.95% vs. 1.27%) and are not encumbered by the initial sales charge of up to 5.5%. Consequently, the performance of class I shares has been much better than that of the class A shares:
Although class C shares of the fund have an investment minimum of only $1,000, they carry a prohibitively high expense ratio of 2.14%. For the purpose of further analysis, we will use class A shares. (However, with a minimum initial investment of $25,000, class A shares are not individual-investor-friendly either.) This share class was introduced in September 2006. Therefore, as of this writing 10-year performance statistics are not available.
The primary benchmark for the fund is the MSCI EAFE index, whose practical embodiment is the iShares MSCI EAFE ETF (EFA). The ETF outperformed the fund in terms of both annualized returns and the Sharpe ratio over the three- and five-year periods through July 2014.
Alpholio™ calculated that since inception class A shares of the fund returned more than the ETF in about 57% of all rolling 12-month periods. The median outperformance was 0.36% but the mean a negative 0.02%. However, given the fund’s recent focus on Japanese stocks, a static index like the MSCI EAFE may not be the best benchmark.
Let’s take a look at the risk-adjusted performance of the MainStay ICAP International fund using Alpholio™’s methodology. Here is the cumulative RealAlpha™ chart for the fund:
Since 2006, the fund’s cumulative RealAlpha™ trended flat to lower on average. As a result, the annualized discounted regular RealAlpha™ for the fund was a negative 0.46% and the lag one a negative 0.03% (to learn about the differences between the regular and lag RealAlpha™, please consult our FAQ). At 19.6%, the fund’s standard deviation was about 1.75% higher than that of its reference ETF portfolio.
The following chart illustrates the composition of the fund’s reference ETF portfolio in the same analysis period:
The fund had top equivalent positions in the iShares Europe ETF (IEV; average weight of 23.4%), iShares iBoxx $ Investment Grade Corporate Bond ETF (LQD; 16.6%), iShares MSCI Germany ETF (EWG; 15.9%), iShares MSCI Japan ETF (EWJ; 14.4%), iShares MSCI United Kingdom ETF (EWU; 7.0%), and SPDR® EURO STOXX 50® ETF (FEZ; 5.8%). The Other component in the chart collectively represents six additional ETFs with smaller average weights.
Overall, since inception class A shares of the MainStay ICAP International fund delivered an unimpressive performance when measured on a fully-risk adjusted basis. A substantial front sales charge coupled with a steep initial investment do not add to the fund’s appeal. A recent departure of a long-time manager also casts doubt on the future performance of the fund.
To learn more about the MainStay ICAP International and other mutual funds, please register on our website.
August 13, 2014
A recent article in Barron’s profiles the Vulcan Value Partners Fund (VVPLX). This $1.1 billion no-load fund, started at the end of 2009, has a total expense ratio of 1.18%. With just 32 holdings as of the end of June 2014, the fund is non-diversified. According to the article
…Vulcan Value Partners fund (ticker: VVPLX) is up an average of 22% a year over the last three years, putting it in the top 1% of large growth funds.
The primary prospectus benchmark for the fund is the Russell 1000® Value index. The secondary benchmark is the S&P 500® index. A practical implementation of the primary benchmark is the iShares Russell 1000 Value ETF (IWD). Since its inception, the fund outperformed this ETF in about 69% of all rolling 12-month periods.
However, neither index is a good reference for the Vulcan Value Partners fund, which has a growth tilt. In Alpholio™’s simplest analysis, both membership and weights of ETFs in the reference portfolio are fixed throughout the entire analysis period. In that reference portfolio, two out of the top three ETFs are technology-oriented: iShares Global Tech ETF (IXN; weight of 17.3%) and Vanguard Information Technology ETF (VGT; 12.6%).
In a more refined Alpholio™ analysis, the ETF membership in the reference portfolio is fixed, but weights can change over time. Using this approach, the following chart shows the cumulative RealAlpha™ for the Vulcan Value Partners fund since its inception:
From early 2010 through mid-2011, the cumulative RealAlpha™ for the fund trended lower. Subsequently, the cumulative RealAlpha™ rebounded and increased by about 20% through the end of 2013. The lag RealAlpha™ curve was below the regular one, which indicates that not all new investment ideas worked out as well as intended (for a detailed explanation of the regular and lag RealAlpha™, please consult the FAQ). Overall, the fund generated about 2.2% of the regular and 1% of the lag annualized discounted RealAlpha™. At about 14.1%, the fund’s standard deviation was about 0.5% higher than that of the reference ETF portfolio.
The following chart illustrates changes in the reference ETF portfolio composition over the same analysis period:
The fund’s top equivalent ETF positions were in the iShares S&P 100 ETF (OEF; average weight of 19.7%), Vanguard Information Technology ETF (VGT; 17.4%), Vanguard Consumer Staples ETF (VDC; 15%), iShares Russell 1000 Value ETF (IWD; 13.4%), Vanguard Consumer Discretionary ETF (VCR; 13.2%), and iShares MSCI United Kingdom ETF (EWU; 7.7%). The Other component of the chart collectively represents six other ETFs with smaller average weights.
The final chart demonstrates a hypothetical buy-sell signal for the fund derived from the smoothed cumulative RealAlpha™ curve shown previously:
An investor following this signal would have avoided a period of the fund’s underperformance from the fourth quarter of 2010 through the third quarter of 2011, and capture the benefits of subsequent outperformance.
In the past three years, the Vulcan Value Partners fund exhibited good risk-adjusted performance. However, in its first year as well as so far in 2014, the fund failed to outperform its reference ETF portfolio. Although the fund’s historical standard deviation was similar to that of the ETF implementing its primary benchmark (IWD), the fund’s concentrated portfolio (top ten holdings constitute over 44% of total assets) may result in a higher volatility in the future.
To learn more about the Vulcan Value Partners and other mutual funds, please register on our website.