In a number of my recent posts, as well as articles published elsewhere, I have noted that the challenge for income investors is creating a portfolio with substantial yield and acceptable risk that also limits exposure to rising interest rates. There are a number of ways to achieve this objective. In this article, I will present an energy-focused portfolio that meets these aims.
MLPs are an attractive asset class for meeting these goals. They have been remarkably interest rate neutral this year, as well as providing high levels of income. More generally, MLPs tend to have a positive response to rising bond yields. Utilities have historically provided solid income streams but they tend to exhibit negative correlations to rising yields. This is well known - nothing new. In addition to MLPs and utilities, some of the major oil and gas companies have attractive yields. Total (TOT) and Royal Dutch (RDS.A) (RDS.B) are notable examples. A central appeal of utilities, oil majors and MLPs is that their business models are straightforward. They provide for the production, refining, delivery and storage of energy commodities. Their earnings streams are somewhat responsive to economic cycles, but investors are far less dependent on estimates of earnings growth than is typical with most other asset classes. All three of these sectors are worthy of consideration, but the challenge is how best to combine them.
Building the Ideal Energy Allocation for Income
Imagine that we wanted to set out the ideal characteristics of an energy-centric income portfolio. We want high income, but we also want reasonable levels of risk, as well as a positive correlation between portfolio performance and bond yield. If interest rates rise, we don't want to get stuck with a basket of leveraged firms that see their earnings collapse as the cost of capital rises. I would also posit that it is desirable to have a portfolio with a meaningful allocation to highly-liquid near-cash equivalents that could be used to purchase more shares of the attractive firms if their prices should fall. Along with cash equivalents (BIL), I have allowed allocations to corporate bonds, high-yield bonds, and an aggregate bond index.
After some research, I have come up with a portfolio that meets these criteria:
|iShares Corporate Bond||LQD||13%||3.8%|
|iShares High Yield Bond||HYG||12%||6.2%|
|Energy Transfer Partners||ETP||6%||6.9%|
|Boardwalk Pipeline Partners||BWP||6%||8.7%|
|Calumet Specialty Products||CLMT||6%||10.5%|
|Kinder Morgan Energy Partners||KMP||6%||6.8%|
|iShares Aggregate Bond||AGG||5%||2.3%|
|Cheniere Energy Partners||CQP||2%||6.1%|
|Royal Dutch Shell||RDS.B||2%||5.1%|
Over the past five years, this portfolio has had a volatility of 9.4% (as compared to 15.8% for the S&P 500). My Monte Carlo simulation suggests that this portfolio's volatility will continue to be about two-thirds of that of the S&P500 - about the same level of volatility as a portfolio that is 65% allocated to the S&P 500 (SPY) and 35% allocated to an aggregate bond index (AGG).
This portfolio has a yield of 5.5% and a +20.4% correlation to the 10-year Treasury yield. For a portfolio that is comprised of energy and utility stocks and MLPs, this is a high yield for the level of risk and exposure to interest rates.
As I have noted elsewhere, utilities are not terribly attractive as a sector due to their relatively high valuations (and therefore modest yields) and their negative response to rising rates. Equity energy indexes tend to have a positive response to rising rates, but broad energy indexes have fairly low yields, too. The Vanguard Energy ETF (VDE) has a yield of 1.6% and the iShares Global Energy ETF (IXC) has a yield of 2.3%.
It should be noted that most of the attractive properties of this portfolio can largely be achieved simply by investing in an MLP fund such as AMLP. AMLP was launched in September 2010, so its has slightly more than a three-year track record. Because AMLP does not have five years of data yet, we will compare the characteristics of AMLP and the model portfolio over the past three years. AMLP has a robust +29% correlation to 10-year Treasury yield and a 6.2% yield. The trailing 3-year volatility of AMLP is 9.3% vs. 8.3% for the model portfolio. AMLP and the model portfolio have very similar yields relative to their risk levels.
One major distinction between AMLP and the model portfolio is that the model portfolio is 44% allocated to bonds. The bond allocation, with 14% in T-Bills, provides a significant amount of "dry powder" in the event of a market decline and a subsequent increase in yields across energy firms. In addition, we have obtained some exposure to utilities and energy majors with the model portfolio.
For some investors, simply buying AMLP or a similar fund will be the best alternative due to its simplicity and the fact that you file a 1099 at tax time and don't have to worry about the K-1 form required for individual MLPs. For those investors who prefer individual stocks and MLPs and who prefer to obtain their income from the energy sector more broadly, the model portfolio could be more attractive.