One of the fundamental questions that investors face is whether to own shares of individual firms or stick to funds. The argument for funds is simple: less company-specific risk. Owning a fund entails additional costs, of course, in the form of the expense ratio as well as other costs such as trading expenses. As an income investor, this question is often particularly difficult. First and foremost, companies with high yield may be in the throes of some form of distress. Prices have fallen but distributions do not yet reflect the market realities. On the other hand, individual companies often have markedly higher distributions than funds. In this article, I will compare an individual MLP, Kinder Morgan Energy Partners (KMP) to an MLP fund (AMLP).
The Alerian MLP ETF has a 6% yield, a five-star rating from Morningstar, and a trailing three-year annualized return of 10%. AMLP does not yet have five years of history. The expense ratio for this fund is 0.85% and the fund has $7.4B in assets. The average market cap of AMLP holdings is $12.7B. AMLP has a total of 25 holdings, and KMP is the second-largest holdings with an allocation of 9.2%.
Kinder Morgan Energy Partners has a 6.7% yield and a market cap of $35B. KMP has annualized three-year return of 10.7% and annualized five-year return of 17.8%.
|Ticker||Yield||3-Year Volatility||Projected Volatility|
KMP has a Beta (with respect to the S&P 500) of 62% vs. 51% for AMLP. This means that the additional volatility of KMP is partly market risk (higher Beta) and partly the company-specific risk (higher volatility beyond what can be explained with Beta).
There are options traded on both AMLP and KMP. When looking at risk levels, it is useful to compare the implied volatilities of put options. Put options are options that cover the probability of a downward move in prices. As such, the prices of put options reflect the market's consensus view of risk. There are standard techniques for backing out an implied volatility from options prices. When I obtained the implied volatilities for both AMLP and KMP from e*Trade, they are almost identical for both near-term options (expiring in Jan 2014) and longer-dated options expiring in Jan 2015. In other words, the option implied volatilities suggest that AMLP and KMP are equally risky, as opposed to the historical data which suggests that KMP is markedly riskier than AMLP. Morningstar's implied volatility estimates generally agree, although there are some differences between the implied volatilities from e*Trade and Morningstar. This is to be expected because the options are relatively illiquid (large bid-ask spreads and low volume) and the two providers use different methods for backing out the implied volatilities.
The incremental additional yield provided by KMP (0.7%) does not appear to justify the additional risk. The yield on KMP is 11.6% higher than that from AMLP but the historical volatility of KMP is 67% higher than that of AMLP. The option implied volatility values suggest that the risk levels of KMP and AMLP are comparable on a going-forward basis, however.
The lower volatility in AMLP is undoubtedly largely due to diversification benefits. The question for investors is whether the higher risk in KMP corresponds to higher potential growth than AMLP's portfolio. If this is not the case, the additional volatility in KMP may simply be idiosyncratic risk that provides no additional growth potential. The long-term total return of AMLP is certainly hampered by the funds 0.85% expense ratio.
Another factor to consider is the correlation between returns on KMP and AMLP as compared to an equity energy index (IGE) and a commodity index (DBC or DJP). Returns on KMP and AMLP have 46% and 49% correlations to returns on DJP over the past three years (respectively) and 54% and 63% correlations to IGE (respectively). These correlations, using three years of trailing data, are close.
Unless you have a strong conviction that KMP will outperform its asset class, the additional yield provided by owning KMP vs. owning AMLP does not justify the additional risk. If you have a broadly-diversified portfolio of individual MLPs, then the point is moot of course. Investors who hold KMP simply to gain access to MLP yields may be better served simply by owning AMLP. On the basis of yield alone, we would need to see a KMP trading at a lower price / higher yield to justify the additional volatility.