Master Limited Partnerships (MLPs) are back in vogue as an asset class that can offer both superior yields and portfolio diversification. You can purchase MLPs as individual units (purchased like a stock from your broker) or via Exchange Traded Funds (ETFs), Exchange Traded Notes (ETNs), or Closed End Funds (CEFs). There are many articles on Seeking Alpha discussing the merits of individual MLPs so I will not cover them here. Instead, in this article I will focus on the risk versus reward of MLP ETF/ETNs (I will extend the analysis to CEFs in a subsequent article).
But before I begin the analysis, I believe a quick tutorial is in order since there are some unique aspects of MLPs. This will be a high level overview as I am not qualified to cover all the tax related nuances of MLP investing.
The rules for modern day MLPs stem from the Revenue Act of 1987 which provided tax advantages to partnerships that earn at least 90% of their income from "qualified" sources, primarily energy and natural resource activities. MLPs do not pay any income tax as long as they make quarterly distributions of at least 90% of their earnings to shareholders (called unit holders in the partnership structure).
Most MLPs operate in the "midstream" portion of the energy production cycle which involves the storing, transporting, or processing of energy (as opposed to "upstream" exploration or "downstream" retail sales). These MLPs typically have huge infrastructure expenditures, such as pipelines and processing plants. In terms of taxes, these infrastructure investments can be depreciated each year. On average, about 80% of the distribution from MLPs consists of cash that is not net profits but is due to depreciation. The part of the distribution tied to depreciation is considered "return of capital" by the IRS and is not taxed until the units are sold. Thus, MLPs not only offer high yield but about 80% of that yield is tax deferred. Sounds too good to be true? Let's look at some of the downsides to understand why MLPs may not be "right" for everyone.
The first downside is the complexity in filing your tax returns. You will receive a K-1 partnership form for each MLP. The K-1 form is vastly different and much more complicated than the simple 1099 form for stock dividends. Although MLPs try to make the filing a K-1 as easy as possible, it is still complex and confusing. To make matters worse, K-1 forms are not required to be issued until March 15, much later than the January 31 date for 1099 forms. Thus, you will need to delay your tax filing until you receive all your K-1s.
Regulated Investment Companies, such as open ended mutual funds, are prohibited from having more than 25% of their assets invested in MLPs. Mutual funds also need to send out 1099s by 31 January but may not receive the K-1 forms until much later. Mutual funds would therefore need to keep amending their 1099 forms, which would be a hassle for their investors. For these reasons, most mutual funds do not invest in MLPs. Note that there are some actively managed "mutual funds" that have structured themselves as C corporations so that they can invest in MLPs. These funds typically charge a sales load and I have not included them in my analysis.
Another issue that may limit your ability to invest in MLPs is related to holding MLPs in an IRA. If your IRA investments generate more than $1,000 in Unrelated Business Taxable Income (UBTI), then this income becomes taxable even though it is in the IRA. Unfortunately, MLPs have the potential of generating UBTI, which could create a complex tax situation. Most experts assess the probability of generating significant UBTI as small, but most also urge individuals, to be on the safe side, not to purchase MLPs in an IRA accounts. In addition, because IRAs are already tax deferred, you do not receive any additional benefits by receiving depreciation income from the MLP.
Due to these complexities, investors have sought ways to receive the benefits of MLPs without the tax headaches and IRA restrictions. These objectives can be achieved by investing in ETFs, ETNs and CEFs rather than individual MLPs. There are currently 5 ETFs and 10 ETNs (compared with 22 CEFs that will be discussed in a subsequent article). However, the majority of these ETFs/ETNs were launched in 2011 or later and do not have a long track record. I will limit my analysis to the ETF/ETNs shown in Table 1, which were launched on or before 10/28/2010 (about a 2.5 year look-back period).
Table 1: List of MLP ETF/ETNs with at least a 2.5 year history
The indexes being tracked by these ETF/ETNs are:
- Alerian MLP (NYSEARCA:AMJ) is a cap-weighted index of the 50 most prominent energy MLPs.
- Alerian MLP Infrastructure is a cap-weighted index of 25 energy MLPs that are engaged primarily in the transportation, storage and processing of energy commodities.
- Cushing 30 MLP Index (NYSEARCA:MLPN) is an equal-weight index of 30 midstream energy MLPs.
- Wells Fargo MLP Index (NYSEARCA:MLPW) is a cap-weighted index consisting of 65 energy MLPs.
- Alerian Natural Gas MLP Index (NYSEARCA:MLPG) is an equal-weight composite of 20 natural gas infrastructure MLPs.
In Figure 1, I plotted the annualized rate of return in excess of the risk free rate (called Excess Mu on the charts) versus the volatility of these ETF/ETNs. SPY is also plotted for reference. The Smartfolio 3 program (smartfolio.com) was used to generate this chart.
Figure 1: Risk-Reward Plot for MLP ETF/ETNs
This plot reveals some interesting attributes of MLP ETF/ETNs.
AMLP is the only ETF of the group and has a significantly smaller rate of return than the ETNs. As we have noted previously, open ended funds can have only 25% of their assets invested in MLPs. To circumvent this restriction, AMLP is structured as a corporation and therefore has to pay corporate income tax. This, plus the other expenses incurred, causes AMLP to lag about 5% behind the benchmark index. On the plus side, AMLP is less volatile than the ETNs.
To avoid setting up a corporation, some issuers have used the ETN structure rather than forming an ETF. In ETNs, the issuing company promises to provide the buyer with the same return as he would have received from the index (less management fees) but the ETN does not actually own the underlying MLPs. Thus the 25% restriction does not apply. However, since there is no underlying collateral other than the credit of the issuing company, ETNs have credit risk. All the issuing companies (JP Morgan, UBS, and Credit Suisse) have excellent credit ratings so the probability of a default is very small but is not zero. If the issuing company was to experience a credit downgrade, then it is likely that the ETN would lose value.
Of the ETNs being analyzed, all have risk-reward profiles better than the SPY, that is, they all have a better rate of return than SPY coupled with a lower risk. Among the ETNs, MLPI had the best risk-reward over the last 2.5 years. AMJ has the second best profile.
An increased level of caution should be exercised if you plan to purchase AMJ. In June, 2012 JPMorgan decided to cap the number of shares that were issued for AMJ. This was very unusual for an ETN and effectively changed the ETN into a closed-end product. With the number of shares fixed, the price is based on supply and demand and the ETN can sell at a premium. At the current time, AMJ is selling at about a 4% premium. This increases the uncertainty since the premium could potentially collapse, causing losses that would be unrelated to the underlying benchmark.
Figure 2 provides the correlation matrix associated with these assets. MLPs provide a reasonable amount of diversification with respect to the S&P 500 (correlation between 40% and 60%). Among the ETF/ETNs, all have relatively high correlations except for MLPW which is only moderately correlated with the other MLP funds.
Figure 2: Correlation among MLP ETF/ETNs
It should also be noted that these ETF/ETNs were not launched until after the end of the 2008 bear market. However, by looking at the performance of the underlying indexes, it can be inferred that these MLPs would have been hard hit, losing about 47% from the 2007 peak to the 2009 low. This was a large loss but still better than the performance of the S&P 500. Like the S&P 500, the MLP indexes are now making all time new highs.
In this analysis I have equated risk with historical volatility but I realize that there are other measures of risk. One unknown quantity is the political risk associated with MLPs. What Congress gives, Congress can take away. Washington is continually searching for more sources of tax revenue, so keep a close eye on the politicians. However, in the absence of a "black swan" event, I would expect that MLPs will continue to an excellent asset class to boost your portfolio returns.