Why MLP ETFs Require Careful Consideration

A small but fervent base of income-oriented investors loves publicly traded master limited partnerships. To capitalize on the growing interest in the product class, a couple fund providers, ALPS and Van Eck, have decided to introduce exchange-traded funds consisting solely of MLPs. In some regards, throwing MLPs into a fund provides benefits: Company-specific risks are minimized through sector diversification, the process of tax filing is considerably easier, and one does not have to worry about a tax deferred account suddenly becoming taxable (details are provided below in linked articles). However, the trade-offs include lower investor returns due to double-taxation and the potential for considerable tracking error from the benchmark. While none of these factors makes MLP ETFs a bad investment choice, potential investors should choose carefully if they want to own MLPs in their portfolios.

Master limited partnerships, particularly those specializing in transporting and storing energy commodities, have been phenomenally strong performers by the standards of income investors--or any standards, for that matter. Not only do virtually all MLPs provide current yields substantially higher than the market in general, or even long-term Treasuries, but the best-run partnerships have been able to increase their unitholder distributions faster than any comparable class of high-income securities.

Alas, the tax characteristics of MLPs render them difficult, if not impossible, for many investors to own. These securities present serious challenges to investors working with tax-deferred accounts, traditional mutual funds find them exceedingly cumbersome to own, and even those working with ordinary taxable accounts must accept a good deal of complex paperwork at tax time. My colleague Christine Benz recently pointed out the administrative challenges of investing in MLP securities, and these new ETFs mitigate some of those problems--but these fund solutions present their own new set of challenges.

The first set of solutions aimed at addressing the challenges of owning MLPs in a fund structure were three exchange-traded notes profiled by Patricia Oey here.  J.P. Morgan Chase (JPM) offers the Alerian MLP Index ETN (AMJ),  UBS (UBS) has launched E-TRACS Alerian MLP Infrastructure Index ETN , and Credit Suisse has just issued Cushing 30 MLP Index . Aside for relatively minor differences in the way each ETN's underlying index is constructed, they're basically the same--in fact, they all charge the same management fee of 0.85% per year. The upside to these funds is that they provide broad MLP exposure, with minimal tracking error, in a relatively tax-efficient vehicle that could be held in either a taxable or tax-deferred account without substantial tax ramifications. The downside to ETNs is that they are unsecured debt obligations that sport some credit risks if the issuing bank runs into financial trouble.

For better or worse, some people find the tax efficiency trade-off of ETNs insufficient to overcome the credit risk, so providers have brought forth MLP funds in the ETF wrapper. The upside is that the ETFs actually own partnership units, so the credit risk is gone. However, there are considerable challenges to owning partnership units in a registered investment company vehicle. First and foremost, some of the tax-advantages of owning MLPs are compromised. The RIC is going to incur its own level of corporate taxation, which is higher than personal income tax rates. Granted, most MLPs do not generate positive reportable taxable income at this time, but they may do so in the future. Furthermore, the C-corp structure now subjects some of the MLP's distributions to double-taxation, because a small portion of the distributions that are now passed on to investors will be taxed as dividends rather than as a return of capital. The upside is that investors do not have to worry about triggering Unrelated Business Taxable Income clauses in their tax-deferred accounts, but they are paying a price to do so. Finally, the ETF structure will likely lead to much more pronounced tracking error than the ETN structure.

ETNs generally have much less tracking error than ETFs in any asset class, but we anticipate that the deviations will be much more pronounced when the underlying holdings of the ETF are partnership units for two primary reasons. First, the MLP asset class is much less liquid than most equity classes, so replicating the index itself will present logistical challenges. This will likely result in ETF share "creations" that allow for both cash and securities within the creation basket. While the market is liquid enough to accommodate the fund structure, doing so will likely result in some price disruptions on the underlying securities. This alone will cause some tracking error. Second, the larger drag will stem from the structural differences between partnership "units" and equity "shares."

When you own an equity "share," typically the only taxable events an investor needs to worry about are 1) reporting the gains or losses when you sell the shares, or 2) reporting your dividend income. Partnership units are treated differently because you have to report your proportion of the companies net income on your tax return. As an example, if you own 1% of a partnership that earns $100,000, you must report $1,000 of income--even if the partnership retains all of those earnings and you do not sell your units. However, the income you receive from that partnership is tax advantaged relative to a dividend. If you receive a $1,000 "distribution" from the partnership, you may not have to report any of that as income. It is often considered a "return of capital," and the tax consequences will not manifest until you sell your units. A return of capital acts as a reduction in your cost basis in your units, which will eventually lead to a larger long-term capital gain when you sell your units. The upside is that long-term capital gain rates are both deferred over time and come at lower rates.

The consequence of this tax deferment will manifest into both tracking error and leverage in the registered investment company structure. In order to account for the growing deferred tax liability, the fund company will retain a portion of the distributions that would otherwise be passed directly on to unit holders. The quirk of the leverage is that it will both grow over time and lead to more volatility in the fund than that of the index itself. However, the upside of this leverage is that it comes in the form of a zero-interest loan from Uncle Sam. In this regard, we would neither say that the leverage is neither a pro or a con: it is merely a consequence that will result in return magnification. The directional movements will be highly correlated with the underlying index, but they will be magnified.

Perhaps the real upside of the pending MLP ETFs is that they will allow institutions to gain asset class exposure to MLPs without the administrative burden. While the solution will come with some costs, I believe that at least some folks will find utility in owning the funds. There is an explicit cost in this case to avoid the credit risk of ETNs and the tax-filing headaches of owning individual MLP issues.