Over the last decade, Master Limited Partnerships (MLPs) have grown from a market cap of $30 billion to over $350 billion. After a rough final quarter of 2012, MLPs have been on fire, with the Alerian index gaining a whopping 19% so far this year.
In a previous Seeking Alpha article, I explained the MLP structure and evaluated the risk-to-reward attributes of MLP Exchange Traded Notes (ETNs) and Exchange Traded Funds (ETFs). In this article, I will extend the analysis to Closed End Funds (CEFs).
Several MLP CEFs have been created recently and do not have a long track record. Therefore, I will limit this analysis to the 13 CEFs that were launched before 2011. These CEFs are summarized in Table 1.
Table 1: CEFs with at least 2.5 years of historical data
One of the main benefits of MLP CEFs is the simplification of tax reporting. CEFs, just like ETFs and ETNs, provide investors with 1099 tax forms rather than the more complex K-1 partnership forms. In addition to taxes, MLPs CEFs have several other unique characteristics that are discussed below.
For various reasons, some CEFs are structured as Registered Investment Companies (RICs). The good news is that RICs do not pay income taxes but instead pass on the tax liabilities to their shareholders. This is the structure used by most open-end mutual funds. The bad news is that RICs are prohibited from having more than 25% of their assets invested in MLPs. Thus, MLP CEFs that are structured as RICs are not pure MLP plays; they usually invest in subsidiaries and affiliates of MLPs, as well as bonds and other energy equities. The first three of the CEFs listed in Table 1 are structured as RICs (based on data from www.morningstar.com).
A more common structure for MLP CEFs is the C-Corporation. These corporations can invest exclusively in MLPs but must pay corporate income tax. This may not be as detrimental as it sounds since the CEFs have ways to minimize these taxes. When a MLP pays out a distribution, the lion share comes from depreciation allowances and is treated by the IRS as "return of capital". This "return of capital" serves to reduce the basis associated with the MLP purchase and taxes are not paid until the MLP is sold. So when a CEF receives a distribution from one of its constituent MLPs, the fund is able to write down the basis of the investment for the portion of the distribution. Over the years, the CEF often builds up relatively large amounts of deferred tax liability that will not be realized until the MLPs are sold. By IRS rules, these liabilities reduce the Net Asset Value (NAV). Astute investors recognize that the NAV has been "artificially" lowered, so these investors may be willing to pay a premium over the reported NAV.
I normally do not like to purchase CEFs with a large premium. However, for C-Corp MLP CEFs, some of the premium may be related to the decreased basis of MLPs rather than based purely on supply and demand. It is difficult to know exactly what percentage of the premium is tied to each phenomenon. Therefore, for C-Corp CEFs, I tend to compare the current premium to the "average premium" rather than using the absolute magnitude of the premium/discount. Table 1 provides the current premium and the 52 week average premium (based on data from www.cefconnect.com).
To get a feel for how the MLPs performed over a complete market cycle, I analyzed CEFs data that began on 9 October, 2007 (start of the bear market) to the present. 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 the 9 CEFs that have been existence over the whole period. The Smartfolio 3 program (www.smartfolio.com) was used to generate this chart. Note that these plots are associated with the market price, not the NAV of the CEFs.
Figure 1: MLP CEFs risk versus reward over a market cycle
Over the market cycle, all the MLP CEFs, except SRV, had exceptional performance when compared with SPY (the ETF that tracks the S&P 500). All the CEFs had returns greater than the SPY but at the price of more volatility.
To assess if the reward was worth the increased risk, I calculated the Sharpe Ratio for each CEF. The Sharpe Ratio is a metric, developed by Nobel laureate William Sharpe that measures risk-adjusted performance. It is calculated as the ratio of the excess return over the volatility (the reward-to-risk ratio if you measure "risk" by the volatility). It is a good way to compare peers to assess if higher returns are due to superior investment performance or from taking additional risk. Reviewing Figure 1, it is easy to see that all the CEFs except SRV had a higher Sharpe Ratio indicating that the reward of investing in these CEFs was worth the risk.
SRV is a special case. It currently has an extraordinarily high distribution (over 11%), which is likely why it also sells at such a high premium (average of 22% but the premium has been as high as 37%). Most MLP CEFs have been able to increase their NAV over time but not SRV. The NAV of SRV decreased from the inception price of $19.06 (in August 2007) to about $3.79 in 2008 and although the price has increased recently, it has never fully recovered. The current NAV is only about $7.40. To be honest, I am not sure why SRV has had such poor performance. I suspect it is due to management issues that are not directly to the MLP space but I do not have enough data to be definitive. However, the selection criteria I use for my personal investment portfolio favors a good reward-to-risk ratio over a high distribution, so I would not invest in SRV.
If I shorten the look back period to 2.5 years, another 4 MLP CEFs (CEM, KMF, NTG, and TPZ) can be added to the analysis. The Risk-Reward plot for the last 2.5 years is shown in Figure 2 (note that the volatility scale has been compressed compared to Figure 1).
Figure 2: MLP CEFs risk versus reward for a 2.5 year period
Over this 2.5 year period, equities have been in a strong bull market so the risk-reward ratio associated with SPY has substantially improved (higher gain with lower risk). Generally, the MLPs have also done well but not all have kept pace with the SPY.
Reviewing Figure 2 we see that for the 5 best performers (KED, FMO, KYN, KMF, and CEM), one is a RIC and the other four are C-corporations. You may also notice that KYN, FMO, and KED have had a better Sharpe Ratio than the SPY over both the market cycle and the past 2.5 years. Of these CEFs, both KYN and FMO are selling at premiums significantly above their average premium. Only KED's premium is near the 52 week average.
In addition to relatively good risk-reward characteristics, MLP CEFs provide a reasonable amount of diversification with respect to the S&P 500 (correlations around 50%), Somewhat surprisingly, the CEFs are only moderately correlated among themselves, with correlations in the 50% to 60% range.
To complete my analysis, I plotted the ETFs/ETNs discussed in my previous article on the same CEF risk-reward plot. The composite picture is shown in Figure 3, with the ETFs/ETNs plotted with the larger "orange" dots. On the Figure, I also plotted a red line that represents the Sharpe Ratio of the SPY. If an asset is above the line, it has a higher Sharpe Ratio than the S&P 500. Conversely, if an asset is below the line, the reward-to-risk is worse than the SPY.
Figure 3: MLP ETF, ETNs, CEFs risk versus reward (2.5 years)
With the exception of SRV, over the past 2.5 years, all the MLP funds have clustered around the S&P 500 Sharpe Ratio line, with some having better reward-to-risk than the SPY and some worse. Unfortunately, there is no clear evidence as to whether C-Corps are better than RICs in terms of total return or risk. There also does not appear to be a clear trend as to whether CEFs are better than ETNs.
The bottom line is that MLP funds should be considered as a component of your portfolio. As an asset class, they offer diversification and have the potential to boost your portfolio returns. As with all investments, you must perform your due diligence and be selective about which funds to purchase. In my opinion, you should not necessarily choose the highest distribution but should carefully weigh premium and risk-to-reward.