In my last column, I highlighted the trend of “at-the-market” equity offerings or equity distribution agreements within the MLP space. Several more ATM deals have been announced since that column, including one from Enterprise Products Partners LP (EPD) that included 10 banks involved, as opposed to what we’ve seen previously for MLPs, with just one or two agents.
This week, I want to get into a niche within a niche: variable distribution CVI), which has been a target of Carl Icahn’s for months now. Second reason to care about variable distribution MLPs is that its constituency has more than doubled in the last three years (from two to five) and will likely double again in the next few years.. So, why would I write an article on a five-company subset of the already obscure MLP sector? First, a few of those five generate more news and buzz than most other MLPs, particularly CVR Partners, subsidiary of CVR Energy, Inc (
What Makes Variable Distribution MLPs Different?
Variable distribution MLPs have MLP-qualifying income under the current tax code, and are in many ways similar to traditional MLPs. They are partnerships and are considered flow-through entities for tax purposes, and investors in them receive annual K-1s from the partnership.
The major difference is variable distribution MLPs do not have a minimum quarterly distribution like traditional MLPs, which generally pay out a steady distribution that usually only goes up or stays flat. Variable distribution MLPs pay out quarterly distributions that will vary with the strength of the cash flows of the business in that quarter. In other words, the distributions go both up and down. Other slight differences are highlighted below.
I wrote generally above, because there are exceptions. Some MLPs have volatile cash flows, and as mentioned above, mitigate that volatility in some way. Other exceptions: Terra Nitrogen Company, L.P. (TNH) has IDRs, and CVR Partners, LP (UAN) has at least filed to do a follow-on equity offering in the past.
Variable distribution MLPs have yet to do a follow-on equity offering. Stability of unit count is something traditional MLPs do not generally offer. Most traditional MLPs issue primary equity in public offerings at least once every two years or so. Some, like MarkWest Energy Partners, L.P. (MWE) and Energy Transfer Partners LP (ETP), have issued equity several times each year the last few years. Equity offerings are not necessarily bad, because it usually means the MLP is buying or building something that has a value greater than the cost of issuing the new units has on the current unitholders. But in this one aspect of the structure, units outstanding, variable distribution MLPs generally offer more stability.
Current Variable Distribution MLPs
As shown in the chart below, variable distribution MLPs generally are smaller, but have produced better returns on average for investors than the MLP Index since the beginning of 2010, mostly tied to strength in the fundamentals of the nitrogen business. Also of note, 3 of the 5 variable distribution MLPs currently trading have gone public since beginning of 2011.
With the exception of Dorchester Minerals LP (DMLP) below, all of the variable distribution MLPs are single asset partnerships right now, meaning they each own a single large plant in one location. They have not been acquisitive to date, and none have issued primary equity following their respective IPOs. Follow-on equity offerings are challenging, because without the forward projection included in the prospectus, it can be hard to price an equity offering.
Terra Nitrogen (TNH), which went public in March 1992, is the oldest variable distribution MLP and one of the oldest MLPs that still trades. TNH is also the exception to the rules listed above. TNH has a minimum quarterly distribution and has incentive distribution rights that entitle its general partner to up to 50% of incremental distributions. As a result, in 2011, 41.7% of total cash distributions at TNH went to its general partner, which is the second highest percentage paid to a general partner of any MLP, behind only Kinder Morgan Energy Partners (KMP) at 45.3%.
Dorchester Minerals is the second oldest variable distribution MLP, and has been public since a spin off in 2003.
Why do Companies go Public with the Variable Distribution Model?
If you have a company that produces MLP qualifying income, but perhaps doesn’t have cash flow that’s suitable for a traditional MLP (i.e. the cash flow is cyclical or otherwise unpredictable), you have a few options if you want to take that company public in a flow-through vehicle. You can structure it as a traditional MLP, and manage cash flow volatility with one or a combination of the following:
- Carry higher distribution coverage (like Alliance Resource Partners LP (ARLP) has done as an operating coal MLP)
- Secure long term contracts or fixed fee contracts
- Purchase hedges
Each of these methods can smooth cash flow, but each can potentially lead the MLP to leave substantial cash flow on the table. MLP management teams are generally willing to forego that variability in hopes of ensuring steady stock price, because a steady unit price means ready access to capital for the MLP.
If your company consists of a single large asset, and you are not so concerned with raising additional equity capital, you can forego the above complications by going public as a variable distribution MLP.
Over the long term, the quarterly cash distribution variability gets smoothed out and variable distribution MLPs are capable of producing long term total returns that rival those of the traditional MLP set. See below for a simple comparison of how TNH has performed relative to one of the oldest and largest MLPs, Kinder Morgan Energy Partners, and relative to the Alerian MLP Index, since 1996.
TNH has produced total returns that have outpaced the MLP Index, but those returns haven’t been quite as steady or as high as bellwether traditional MLP KMP.
Future Variable Distribution MLPs
There are business lines in MLPs today that might be better suited for the variable distribution structure. A research analyst I spoke with indicated that NuStar Energy L.P.’s (NS) asphalt refining business would be much better suited to be in a variable distribution structure, rather than within NS’s other operations, where it currently adds uncertainty to NS’s cash flows.
Relative to regular MLPs, there is a discount associated with going public as a variable distribution as compared with the valuations of regular MLPs. For example the last 3 variable distribution MLPs went public with average yields of 11.9%, compared with an average of 8.1% for the last 10 traditional MLP IPOs (even lower for midstream MLPs, 8.1% includes E&P MLPs). However, that discount is at least somewhat due to the assets of the current variable distribution MLPs.
Going forward, we may see refining assets, oilfield services assets or E&P assets put into variable distribution MLP structures and taken public. For those asset types, sponsors still get a valuation advantage of going public in a variable distribution MLP wrapper still far exceeds the valuation they would receive by going public as a corporation, based on comparable trading multiples. It is likely that future variable distribution MLPs will attempt to be more active in terms of acquisitions and equity offerings.
As with any public company, traditional and variable distribution MLPs included, the performance of the stock depends on much more than its corporate structure. Fundamentals of the sector (commodity prices, demand, etc.), management team, and strength of assets (location, contract mix, customer mix) all matter much more than structure. But, it’s important for investors to understand the differences highlighted above, so you are not taken by surprise when a variable distribution MLP announces a distribution much lower than a prior quarter. Given the strength of the nitrogen business, we have not seen how the market reacts to a down cycle for this latest batch of variable distribution MLPs.