MLP general partners have outperformed consistently in recent years. This week, we have seen MLP general partners separate themselves further. Notably, from Friday through Tuesday, Kinder Morgan, Inc. (KMI) was up more than 6%, and Atlas Energy, L.P. (ATLS) was up 4.1%. The value of the general partner interest (and the incentive distribution rights, or “IDRs”, in particular) seems to be continually undervalued, and can even be hidden in the shadow of a larger company that has other assets, as was the case with Sunoco (SUN), El Paso (EPB), Williams (WPZ), etc. before those companies spun off other parts of their businesses.
Only in the last six years or so have investors had the opportunity to invest alongside management and benefit from IDR economics directly. But for decades now, the magic of IDRs have made billionaires out of MLP founders like Dan Duncan, Rich Kinder, and Joe Craft.
What accounts for GP holding companies consistently outperforming? Why are they persistently undervalued despite the track record? Through the traditional MLP yield lens, general partner holding companies look expensive at yields below 3% for some, but saying that point of view discounts the massive growth of these entities that stems from their unique derivative assets, which allow for distribution/dividend growth at multiples of their subsidiary MLPs’ distributions growth rates.
Current GP Holding Companies
There are currently 7 pure play publicly-traded general partner holding companies (only counting those that have incentive distribution rights): Alliance Holdings GP (AHGP), Energy Transfer Equity (ETE), Nustar GP Holdings (NSH), Targa Resources (TRGP), Crosstex Energy (XTXI), ATLS, and KMI. KMI and ETE are currently executing mergers that will change them from pure play GPs to operating companies with GP interests, shortening the list to just five GP holdings companies soon.
There are also a few other public companies (notably Williams Companies and Sunoco) that hold GP interests in MLPs that have split off other parts of their businesses in the last 12 months to become closer to pure play general partners holding companies. They may at some point be included in this discussion. Three out of the seven GP holding companies above are structured as corporations (KMI, TRGP and XTXI). The others are structured as partnerships.
How GPs Grow
MLP general partners grow in two ways:
- From growth in distributions of the underlying MLP
- Each incremental increase in distributions must include the marginal percentage split to the IDRs holder
- From growth in the number of L.P. units outstanding
- IDRs are percentage overrides on total cash distributed, and they are never diluted.
- If total cash distributed increases because you have more units that receive distribution, then the IDR holder would be entitled to the same percentage, but more cash flow, even if the MLP does not increase its distribution per unit.
Understanding these two levers helps shed light on how valuable these general partner interests can be, particular in drop down scenarios, like with MLPs WPZ and EPB, which are situations that engineer a golden combination of repeated equity offerings and rapid distribution growth.
Example: Targa Resources Corp
To help illustrate how the GP math works, I will use the example of Targa Resources Partners LP (NGLS) and its public general partner Targa Resources Corp (TRGP). TRGP is a relatively simple and representative MLP general partner (i.e. not as confusing as Energy Transfer Equity, which holds IDRs for 2 MLPs, and more leveraged to growth than NuStar Holdings, which only has a top IDR tier of 25%).
In this example, I will illustrate how TRGP’s cash flow and dividends would grow if NGLS raised its distribution at an assumed rate for the next 5 years. I will then show how that cash flow changes if NGLS were to issue $250 million equity each year. After that, I’ll review how sensitive TRGP dividend growth is to NGLS distribution growth rates and to amount of equity issued annually.
- TRGP owns a 2% GP interest and 100% of the IDRs in NGLS.
- TRGP holds 11.6 million L.P. units of NGLS.
- Holding company cash interest expense: constant at $4.0 million
- Holding company G&A expenses: constant $8.3 million
- Holding company cash taxes: constant percentage of total cash distributions received minus G&A expenses and interest expense (currently 11.4%)
- Cash distributions per L.P. unit at NGLS: starting at current distribution annualized, growing at 10% next year, and 8% thereafter
- Management at NGLS has provided guidance of 10% to 15% for 2012.
In calculating the distribution to the GP interest, you have to run the distribution through a spreadsheet that calculates how much cash goes to each IDR tier. In the business, this spreadsheet is called a waterfall. Building your own waterfall is a rite of passage, and key to understanding how the cash flows, so I won’t detail how that is done or offer up my own waterfall to you, for your own good.
Under the assumptions above, the chart below is the resulting cash flow that TRGP would receive and could pay out in dividends.
In order to grow distribution at the above pace, NGLS will need to issue equity to fund its ongoing growth projects and any future acquisitions. The numbers above assume that NGLS units are held constant. The numbers below assume that NGLS issues $250 million of equity annually, with the issue price assumed to be the distribution that year divided by the current yield of 5.71%. It should be noted that since the middle of 2009, NGLS has issued more than $1.0 billion of equity, a rate of more than $400 million per year.
GP holding company dividend growth rates are most sensitive to distribution growth of the underlying MLP. In the chart below, a 2% increase in annual NGLS distribution growth from 8% to 10% results in a 3.5% increase in the annual dividend growth rate for TRGP shareholders. Equity issuance has a substantial impact as well. In this example, over the course of five years, NGLS issuing $250 million of equity per year results in 5% more in annual dividend growth for TRGP shareholders than issuing no equity at all.
There are a variety of ways to value GP holding companies. Comparing cash flow multiples among peers is useful. For multiples you can use the cash flow of holding company value compared with holding company cash flow, or you can look at the value of just the GP and IDRs (i.e. backing out L.P. units owned) compared with the GP and IDR cash flow.
Looking at yield comparisons to peers and to subsidiary MLPs is useful. You can look at the difference in yield between the subsidiary MLP and the GP holding company. If the GP is yielding more than the MLP (which happened in 2008 and 2009), and you believe the MLP distribution is not in jeopardy, you can bet on math and go long the GP holding company and short the MLP and play the pending correction. Another useful method is to look at the future yield of the GP holding company compared with the future yield of the subsidiary MLP, and then determine how many years before the faster growing GP will have a higher future yield than the GP holding company. Sooner is better obviously.
Dividend discount models or holding company discounted cash flow methods are also useful. Whatever method you use, I would suggest you take a look at the general partner holding companies. It has certainly been well worth the brain cells in the past.
Disclosure: The information in this article is not meant to be financial advice, we are not your financial advisor and I am posting my comments for informational purposes only. Long NGLS and KMI in client accounts. Long KMI and TRGP in personal accounts.