Analysis of Parnassus Core Equity Fund
July 31, 2017
Analysis of Columbia Seligman Communications and Information Fund
A recent story in The New York Times focused on Parnassus Investments and its Core Equity Fund (PRBLX; Investor Class shares). This $15.7 billion large-cap no-load fund has a reasonable 0.87% net expense ratio and 23% turnover. According to the article
Value-oriented investors who screen out companies that don’t meet strict social standards, [the fund managers], over the last year, generated a respectable 14 percent return in their core equity fund where they have large stakes in Apple and Google. But the positions are not nearly enough to keep pace with the 18 percent return of the Standard & Poor’s 500-stock index, within which six of the 10 top components are now technology stocks.
Over the longer term, however, the Parnassus results are better. For 10 years, the core equity fund handily beats its benchmark — 9 percent compared with 7 percent, a record that outpaces 98 percent of the competition.
This post is a follow-up to our previous coverage of this fund.
Let’s start with rolling returns. The fund’s primary prospectus benchmark is the S&P 500® Index. One of the long-lived implementations of this index is the SPDR® S&P 500® ETF (SPY). From January 2000 through June 2017, the fund returned more than the ETF in approximately 62% of all rolling 36-month periods, 56% of 24-month periods and 54% of 12-month periods. However, the dispersion of outcomes was quite wide, as shown in the following chart and statistics:
While a rolling returns analysis provides useful information about relative performance over typical holding periods, it does not take the fund’s exposures or risk into account. To more accurately adjust for the latter, let’s employ the simplest variant of Alpholio™ patented methodology. In this approach, a custom reference ETF portfolio is built for each analyzed fund to most closely track the fund’s returns. The portfolio has a fixed ETF membership (with a configurable limit) and weights, thus facilitating an easy implementation.
The following chart with associated statistics shows the cumulative RealAlpha™ for Parnassus Core Equity over the ten years through June 2017 (to learn more about this and other performance measures, please visit our FAQ):
Compared to the reference portfolio of up to six ETFs, the fund added a fair amount of value over this analysis period and did so with a RealBeta™ well below one.
The following chart with statistics depicts the constant composition of the reference ETF portfolio over the same evaluation period:
The fund had equivalent positions in the iShares MSCI KLD 400 Social ETF (DSI), Vanguard Consumer Staples ETF (VDC), PowerShares BuyBack Achievers™ Portfolio (PKW), iShares Morningstar Large-Cap Growth ETF (JKE), First Trust Water ETF (FIW), and iShares U.S. Energy ETF (IYE).
Now let’s take a look at the fund’s performance over the last five years. Here is the resulting cumulative RealAlpha™ chart with related statistics:
Since mid-2015, the fund lost all of the cumulative RealAlpha™ it previously generated in this analysis period. Also, despite lower volatility (measured by the standard deviation) the RealBeta™ of the fund was higher than that over the broader evaluation period.
The following chart with statistics illustrates the static composition of the reference ETF portfolio over five years:
The fund had equivalent positions in the PowerShares S&P 500® Quality Portfolio (SPHQ), iShares MSCI USA ESG Select ETF (SUSA; formerly KLD), SPDR® S&P® Dividend ETF (SDY), iShares U.S. Industrials ETF (IYJ), Technology Select Sector SPDR® Fund (XLK), and Consumer Staples Select Sector SPDR® Fund (XLP).
The final chart with conventional statistics shows the total return of the fund and two of the reference ETFs:
Over the five-year period, the performance of the two ETFs, and especially DSI, converged with that of the fund.
In sum, while the Parnassus Core Equity Fund has a decent long-term record, its recent performance has been similar to that of index-based environmental, social and governance (ESG) products with lower expense ratios. With approximately 40 positions, the fund’s portfolio is fairly concentrated – top ten holdings currently constitute almost 39% of the total. In three of the last four calendar years, the fund had significant distributions, which made it less suitable for taxable accounts.
To learn more about the Parnassus Core Equity and other mutual funds, please register on our website.
October 1, 2015
Analysis of Alger Spectra Fund
A recent profile in Barron’s features the Columbia Seligman Communications and Information Fund (SLMCX, Class A shares). This $3.7 billion fund has a maximum 5.75% sales charge, 1.36% expense ratio and 61% turnover. According to the article
Over 10 years, the fund has returned 10.7%, beating 87% of its science-and-technology fund peers, according to Morningstar, and the broader market’s 7%. Its one-year performance is also impressive, returning 8.7% annually and beating 83% of its peers, but the fund’s performance can be very volatile over short periods.
The prospectus benchmark for the fund is the S&P North American Technology Sector Index. One of the long-lived and accessible implementations of this index is the iShares North American Tech ETF (IGM). Alpholio™’s calculations show that over the 10 years through August 2015, the fund returned more than the ETF in about 49% of all rolling 36-month periods, 55% of 24-month periods and 56% of 12-month periods. Over the last five years, these figures were 8%, 24% and 33%, respectively. This indicates that the more recent performance of the fund was likely not as good as the longer-term one.
Let’s take a closer look at the performance of Columbia Seligman Communications and Information by applying a variant of Alpholio™’s patented methodology. To more accurately adjust for the fund’s holdings and risk over time, the methodology constructs a reference ETF portfolio with fixed membership but variable weights. Here is a chart of the resulting cumulative RealAlpha™ for the fund:
Over the last five years, the fund produced a negative 1.8% of the regular and negative 1.2% of the lag annualized discounted cumulative RealAlpha™ (to learn more about RealAlpha™, please visit our FAQ). At 15.2%, the fund’s standard deviation was approximately 1.3% higher than that of the reference ETF portfolio. The fund’s RealBeta™ of over 1.23 underscores the fund’s volatility.
The following chart illustrates changes of ETF weights in the reference portfolio for the fund over the same analysis period:
The fund had top equivalent positions in the Vanguard Information Technology ETF (VGT; average weight of 53.8%), iShares North American Tech-Software ETF (IGV; 15.1%), iShares North American Tech-Multimedia Networking ETF (IGN; 13.1%), iShares S&P Mid-Cap 400 Value ETF (IJJ; 7.5%), iShares Morningstar Small-Cap ETF (JKJ; 4.3%), and iShares Nasdaq Biotechnology ETF (IBB; 3.2%). The Other component in the chart collectively represents two additional ETFs with smaller average weights.
Over the last five years, the Columbia Seligman Communications and Information Fund delivered an unimpressive performance on a truly risk-adjusted basis. The fund could have been substituted, with better return/risk characteristics, by a portfolio of a small number of the technology sector and small/mid-cap ETFs. In addition, the fund returned a few percentage points less per year than its stated benchmark, especially after accounting for its high front load. Finally, the fund’s periodic substantial long- and short-term capital gain distributions (e.g. collectively almost 13% of the NAV in 2014) made it less suitable for taxable accounts.
To learn more about the Columbia Seligman Communications and Information and other mutual funds, please register on our website.
February 21, 2015
Analysis of Morgan Stanley Institutional Growth Fund
A recent piece in Barron’s profiles the manager of the Alger Spectra (SPECX, Class A shares) and Alger Capital Appreciation (ACAAX, Class A shares) funds. According to the article
Over the 10 years ended in December, when Kelly celebrated a decade at the helm of both, they [the funds] ranked among the 10 best-performing stock funds in the U.S. Of both funds, Morningstar writes, “Manager Patrick Kelly is this fund’s most valuable asset.”
In this post, we will focus on the Alger Spectra fund since its return over the past ten years has been higher than its sibling’s. In addition, for our analysis we will use Class A shares of the fund (with the maximum front-end sales charge of 5.25% but a lower nominal expense ratio of 1.52%), as opposed to the Class C shares cited by the article (with a nominal expense ratio of 2.28%).
The prospectus benchmark for the fund is the Russell 3000® Growth index. A practical implementation of this index is the iShares Core U.S. Growth ETF (IUSG). The first full month of the current fund manager’s tenure was October 2004. According to Alpholio™’s calculations, from then through January 2015, the fund returned more than the ETF in approximately 80% of all rolling 12-month intervals, 85% of 24-month intervals and 92% of 36-month intervals. The median cumulative outperformance per interval was about 18.3%, 33.6% and 47.7%, respectively. However, these figures do not account for the fund’s risk.
Other calculations by Alpholio™ indicate that in over the same analysis period, Alger Spectra had an annualized standard deviation (a measure of volatility) of about 16.9%, beta of 1.05, Sharpe ratio of 0.79, and maximum drawdown of 49.7%. This compares favorably to 15.1%, 0.98, 0.55 and 48.4%, respectively, for its benchmark ETF as, by the Sharpe ratio measure, the fund’s risk-adjusted performance was clearly superior to the ETF’s.
Let’s take a closer look at the fund’s performance using a variant of Alpholio™’s methodology in which the membership of the reference ETF portfolio is fixed but individual ETF weights may vary over the analysis period. Here is a chart of the resulting cumulative RealAlpha™ for the fund:
Since late 2004, the fund generated an annualized discounted cumulative RealAlpha™ of approximately 4.7%, an impressive feat. (To learn more about RealAlpha™, please visit our FAQ.) At 17%, the fund’s standard deviation was only 0.3% higher than that of its reference ETF portfolio. The fund’s RealBeta™ was around 1.09.
The following chart shows how weights of ETFs in the reference portfolio varied over time:
The fund had top equivalent positions in the iShares Russell Mid-Cap Growth ETF (IWP; average weight of 23.8%), iShares Morningstar Large-Cap Growth ETF (JKE; 15.3%), PowerShares QQQ™ ETF (QQQ; 13.5%), iShares Morningstar Mid-Cap Growth ETF (JKH; 12.4%), Vanguard Information Technology ETF (VGT; 10.1%), and iShares North American Natural Resources ETF (IGE; 6.5%). The Other component in the chart collectively represents six additional ETFs with smaller average weights. Clearly, the fund exhibited large- and mid-cap growth characteristics.
Over the last ten years under current management, the Alger Spectra Fund delivered impressive risk-adjusted results. Currently, the fund has a significant exposure to information technology (31.2% of assets) and biotech (healthcare accounts for 20.8% of assets). In the top-ten holdings, four technology stocks account for 15.4% of the fund’s assets. Concentration in such high-risk sectors and industries may be detrimental, should their recently positive momentum subside.
It also should be noted that, according to the article,
Spectra can sell up to 10% of its assets short.
Some of the bets on falling equity prices may backfire, thus increasing the fund’s volatility.
The fund’s 150% turnover contributed to its hefty distributions, which in the past two years ranged from 5.8% to a whopping 13.7% of the net asset value (NAV). Thus, the fund may be better suited to tax-deferred accounts.
To learn more about the Alger Spectra and other mutual funds, please register on our website.
November 16, 2014
Analysis of Buffalo Discovery Fund
A recent profile in Barron’s features the Morgan Stanley Institutional Growth Fund (MSEGX, Class A retail shares). This $3.5 billion fund has a maximum 5.25% front load, 0.96% expense ratio and 31% turnover. The fund can invest up to 25% of assets in foreign securities. According to the article
The fund is up an average of 10.3% a year over the past decade, better than 94% if its large growth peers.
The primary benchmark for the Morgan Stanley Institutional Growth is the Russell 1000® Growth index. An accessible implementation of this index is the iShares Russell 1000 Growth ETF (IWF). Alpholio™’s calculations show that since mid-2004 (when the current management took over), the fund outperformed the ETF in about 61% of all rolling 12-month periods.
On an annualized basis, the fund returned more than the ETF in the past three-, five- and ten-year periods. However, the fund’s Sharpe ratio was below that of the ETF in the first two periods and just slightly higher in the third one. This was due to the fund’s higher standard deviation (volatility) compared to that of the ETF. (For example, in 2008 the fund lost over 50%, while the ETF about 38%.) As a result, the stated index may not be the most applicable benchmark for the fund.
Let’s take a look at the fund’s performance through the lens of Alpholio™’s methodology, which more accurately adjusts for risk. Here is a chart of the cumulative RealAlpha™ for the fund:
In the past ten years, the general trend of cumulative RealAlpha™ for the fund has been negative. As a result, the fund generated about negative 3.6% and negative 2.6% of annualized discounted regular and lag RealAlpha™, respectively (to learn more about the regular and lag RealAlpha™, please consult our FAQ). The fund’s annualized standard deviation was about 19.2%, approximately one percent higher than that of its reference ETF portfolio. The RealBeta™ of about 1.19 underscored the elevated volatility of the fund.
The following chart shows changes of ETF weights in the reference portfolio for the fund in the same analysis period:
The fund had top equivalent positions in the PowerShares QQQ™ ETF (QQQ; average weight of about 48.8%), iShares Morningstar Mid-Cap Growth ETF (JKH; 29.8%), Vanguard Materials ETF (VAW; 10.3%), iShares MSCI Hong Kong ETF (EWH; 5.7%), SPDR® Dow Jones® REIT ETF (RWR; 3.9%), and iShares MSCI Taiwan ETF (EWT; 1.5%). Since collectively these six ETFs were sufficient to explain the returns of the fund, the Other component in the chart was nil.
The above analysis clearly revealed that the fund’s investments were dominated by the technology sector as well as mid-cap securities. The former was responsible for the increased standard deviation of the fund. Eight of the fund’s top-ten holdings (which together accounted for 48% of assets) are traded on Nasdaq and are members of the index underlying the QQQ ETF. Therefore, the Nasdaq-100 Index®, and its QQQ implementation, would be a better benchmark for the fund. Alpholio™’s calculations indicate that the fund returned more than that ETF in only about 56% of all rolling 12-month periods since mid-2004.
Over the past ten years, the Morgan Stanley Institutional Growth Fund exhibited an unimpressive performance when truly adjusted for risk. The recovery in the fund’s cumulative RealAlpha™ that began in the second quarter of 2013 turned out to be short-lived. Investors should also be mindful of the fund’s heavy orientation toward technology stocks, which makes it more a sector rather than core large-cap growth holding in the overall portfolio. The fund could easily be substituted with a dynamic combination of just six ETFs with superior return and risk results. Finally, the fund’s substantial sales charge does not add to its appeal.
To learn more about the Morgan Stanley Institutional Growth and other mutual funds, please register on our website.
March 2, 2014
Value in Technology
A recent story in Barron’s profiles the Buffalo Discovery Fund (BUFTX). This $642 million no-load fund, formerly known as Buffalo Science and Technology, sports a reasonable expense ratio of 1.01% and portfolio turnover of 53%.
According to the article, the fund’s manager
Carlsen attributes his fund’s success—it has returned 28% annually over the past five years, beating 93% of other mid-cap growth funds, and 11% over the past 10, beating 84%—to his go-anywhere approach. Though the portfolio mainly consists of midsize companies (with an average market value of $7.3 billion, smaller than the category’s average of $8.4 billion), Buffalo Discovery defines itself as an all-sizes fund specializing in innovation.
The primary benchmark for Buffalo Discovery is the Russell 3000® Growth index. The fund returned more than the practical implementation of its benchmark, the iShares Russell 3000 Growth ETF (IWZ), in all but two of the last ten years. However, this may not be a proper index benchmark for the fund, which has 41% in technology and 26% in technology-oriented healthcare stocks, according to the article. Indeed, according to the fund’s description, it
Invests in companies which create value through the commercial application of innovative products, services, or intellectual property [and is] not benchmark-driven.
The majority of the fund’s holdings are classified as mid-cap, which is why Morningstar applies a Mid-Cap Growth category to the fund.
Let’s assess the fund’s performance using Alpholio™’s methodology. Here is the cumulative RealAlpha™ chart for the fund:
From early 2005 through 2011, the cumulative lag RealAlpha™ for the fund was roughly flat. In 2012, the fund generated some RealAlpha™ only to lose it all by the end of that year. The fund’s RealAlpha™ strongly rebounded in the second half of 2013. As a result, in the overall analysis period, the fund generated only a slightly positive (fraction of a percentage point) annualized lag RealAlpha™.
The chart also shows that the lag cumulative RealAlpha™ was generally lower than the regular one. This is an indication that in many sub-periods, new investment ideas did not produce desired outcomes. To put it another way, investors would have been better off by sticking to the lag reference portfolio. (To learn more about the relationship between the regular and lag RealAlpha™, please visit our FAQ.)
The following chart depicts ETF weights in the reference portfolio for the fund in the same analysis period:
The fund’s top equivalent positions were in the following ETFs:
The Other component in the above chart includes six additional ETFs, such as the iShares Nasdaq Biotechnology ETF (IBB) or iShares North American Tech-Software ETF (IGV). Combined, all these equivalent positions prove that the fund was predominantly of a technology and healthcare, and not just general mid-cap growth, nature at its core.
In conclusion, the above analysis demonstrated that Buffalo Discovery could be effectively substituted by a relatively small number of technology and healthcare ETFs, which formed a portfolio of a similar volatility. While the fund has lately generated positive RealAlpha™, history shows that this outperformance streak may not continue for a long time.
To get more information about Buffalo Discovery and other mutual funds, please register on our website.
October 26, 2013
According to an InvestmentNews article, Oakmark Funds, managed by the value investment firm Harris Associates, are increasingly holding technology stocks traditionally preferred by growth strategies:
The Oakmark Select [OAKLX] Fund has a 24% weighting to technology, and the flagship $11 billion Oakmark Fund (OAKMX) has a 19% weighting.
The reason is that prices of traditional value equities have increased to the point where the technology sector is more attractive. To illustrate that, here are some statistics from a recent edition of S&P’s The Outlook:
At 12.9, the price to estimated 2014 earnings ratio of the information technology sector is lower than that of classic value sectors, such as consumer staples (15.9) or utilities (14.7). In addition, the 13.1% year-to-date return of the sector trails that of the overall S&P 500® (19.1%). Finally, the price-to-earnings-growth (PEG) ratio of 1.0 for the sector matches that of consumer discretionary for the lowest value of all sectors. In a low-interest-rate environment maintained by the Fed, investors in search of dividend income have pushed the PEG of the consumer staples sector to 1.7 and telecom services to 1.6.
While the emphasis on technology stocks may improve the funds’ performance, as shown in a previous Alpholio™ post, past selection of securities in OAKMX did not lead to spectacular results when measured on a truly risk-adjusted basis. Therefore, investors should view the latest attempts with caution.