In the first article in this series, I demonstrated that besides dividend/distribution yield, no matter what valuation metric is used, MLPs appear overvalued. MLPs seem particularly overvalued when compared to dividend yielding peers such as Dividend Aristocrats, Utilities and Telecom.
As I mentioned in the first article, comparative valuation multiples cannot be looked at in isolation. There are various factors that can mitigate the apparent overvaluation of MLPs: JP Morgan Alerian MLP Index ETN (AMJ), Credit Suisse Cushing 30 MLP Index ETN (MLPN). On the other hand, there are other factors that end to make the overvaluation of MLPs look even worse.
In the next two installments I will examine factors that, all other things being equal, would justify MLPs trading at a premium to other C-Corps on a P/E, P/CF and/or an EV/EBITDA basis. In other words, these factors would tend to mitigate the apparent overvaluation of MLPs.
Consistent and Reliable Income Stream
For many investors, the primary reason to invest in MLPs are their relatively steady and reliable cash distributions. Many investors need or desire a steady income stream and are willing to pay a premium to obtain this.
However, how much of a valuation premium is justified on this basis?
The precise answer to this question may depend on an investor’s age, goals, psychological profile as well as other factors. However, setting these particularities aside, it is possible to analyze this issue in general terms, particularly if looked at from a long-term perspective.
The first observation that should be made is that financial theory, as well as common sense, suggests that investors should be indifferent regarding whether income is received via dividends or capital gains. Cash is cash. It makes no difference what its source is.
Take two companies. Company A distributes all available cash flow annually as a distribution. Company B pays 50% of its available cash as a dividend and reinvests the balance to build up the asset base and fund future growth. Assume for a moment that the market appreciation of both companies annually reflects their accumulation of assets and increase in earnings capacity. This means that while Company A will not experience capital appreciation in real terms, Company B will generate capital appreciation in proportion to its reinvestment. Thus, the total return (dividends + capital appreciation) yielded by both companies will be exactly the same.
Assuming that the tax treatment of capital gains and dividends is the same, it should make no difference to investors whether they derive income from dividend distributions or from partial realization of capital gains every year.
A basic objection to this example is that in practice, capital appreciation is not steady. However, under most assumptions, over long periods of time, the total returns of Companies A & B should be equal. Some years the capital appreciation of Company B will exceed the growth of assets and earnings power, and in other years it will lag or be negative. In the long run, and under most circumstances, this will not matter. The good years compensate for the bad, and over time, investors in Company B will be able to derive the same income from taking partial capital gains as they will from receiving cash dividends or distributions.
Notwithstanding this fact, some investors that desire a steady income stream feel great psychological distress at the thought that they are “dipping into capital,” even if that feeling is illusory. For this reason, many investors will prefer investment in Company A rather than Company B. And many investors will express this preference by being willing to pay a valuation premium for company A.
How high of a premium is justified? In the long run, in a relatively efficient market, the premium that can be sustained on the sole basis of the consistency of distributions is quite low.
Ironically, in the long run, the most disciplined and longest-term investors will prefer the greater total returns offered by stocks in higher growth segments to the high cash distributions offered in sectors such as MLPs.
For example, let us imagine for a moment that Company B is a C-Corp whose projected long-run total return (dividends plus capital appreciation) is higher than that of company A, an MLP. However, due to the consistency of its cash distributions, Company A actually trades at a premium to Company B in terms of multiples of earnings and cash flow.
This is currently the case of MLPs. The expected total return of the average stock in the S&P 500 as well as comparable sectors such as utilities and telecoms is currently higher than that offered by MLPs. As reported in the last installment, the total 5Y annual return (current yield + 5Y CF growth) currently forecasted for MLPs is 10.7%. The comparable numbers for Dividend Aristocrats, Utilities and Telecoms are 12.8%, 12.7% and 15.7%, respectively.
In a relatively efficient market, the large difference in valuation between the total MLPs and stocks in other sectors will be arbitraged away until future expected returns are equalized. This will occur through two mechanisms. First, on the margin, investors that are less sensitive to the dividend distribution issue, will tend to sell company MLP units in favor of C-Corps with better total return prospects. Second, various market participants will tend to employ strategies that involve shorting MLPs and going long various C-Corps given the latter’s lower valuations and higher expected total returns. In this case, the valuation premium that could be sustained by MLPs would equal the costs associated with implementing these sorts of paired investment strategies (margin interest and etc.)
In practice, the market is not entirely efficient in this sense. For example, it is relatively difficult to short MLPs. And while it would be possible to achieve the same results through various derivatives transactions, the cost of such operations is relatively high and the number of market participants that can execute them is relatively low.
Thus, it is clear that there are some inefficiencies in capital markets that can sustain some level of MLP overvaluation. However, investors should not exaggerate the magnitude of these inefficiencies – particularly when taking a long-term view. It would be foolhardy to suppose that the gross valuation inefficiencies of the sort that currently exist between MLPs and other C-Corps can persist indefinitely. The valuation gap will tend to narrow over time.
In sum, in the long term, there is simply no difference between capital gains income and distribution income. And while some level of premium for consistency of distributions can be sustained, total return is the determining factor in valuation in the long run. In the fullness of time, the large valuation premiums for stocks will not be maintained when, all things being equal, those stocks produce equal total returns.
Certainly, all things are never equal, and in the next installment we will examine various factors such as tax considerations, limited business risk and the cost of capital that, in addition to steady cash distributions, tend to favor the valuation of MLP’s.
But for now, it is important to understand that taken in isolation, large valuation premiums for MLPs such as Kinder Morgan (KMP), Magellan Midstream Partners (MMP), Buckeye Partners (BPL) and Cheniere Energy Partners (CQP) are neither warranted nor will they be sustained on the sole basis of their steady cash distributions.
Thus, even taking into account the value of the reliability of MLP cash distributions, MLPs still seem very overvalued. Other factors must be sought if the apparent overvaluation is to be justified.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.