If you want to do an adequate amount of due diligence before a MLP (Master Limited Partnership) purchase, you will listen to the most recent conference call, read the last company presentation materials, and read a full analyst update. That takes at least two to just over four hours to do those tasks for a single investment. Given the number of MLP options, doing that basic research over the full sector would take months of "spare time."
I believe that if you are going to do a decent amount of due diligence before the purchase, you need to successfully let the data weed out the MLPs that you would not want to investigate because they offer lower total returns. A good weeding job should result in time savings that can be reinvested in a better due diligence job on the smaller group of investments you place on your menu of options.
This article examines the total returns - or unit price appreciation plus unreinvested distributions - for the large cap midstream MLPs since the beginning of 2010. This article will highlight some key metric attributes that led to those returns. I make an attempt to put a story to those numbers. This article is low on specific information on specific companies. This article is numbers intensive. This is a "weeding" article.
This article will have a focus on Boardwalk Pipeline Partners (BWP), Enbridge Energy Partners (EEP), Enterprise Products Partners (EPD), Energy Transfer Partners (ETP), Magellan Midstream Partners (MMP) and Plains All American Pipeline (PAA). I will tell the heavily abridged story of the numbers breaking those six MLPs into two sets of three. In an effort to show that the ratios used in this article to examine MLPs are meaningful, I will spend some time discussing those same ratios using several consumer staple stocks as examples.
The purpose of this article is to get an average investor comfortable with using some of the key MLP ratios in explaining past performance. It will also serve to get readers ready to use those same metrics in forward projections they should make as part of their purchase decision. This article is intentionally repetitive due to having an acclimation purpose. If you become acclimated before my complete telling of the examples, please skip to the conclusion labeled "What are the lessons in all these numbers?"
When I look at the total returns and the distribution growth since 2010, I see a strong (but imperfect) correlation between higher distribution growth and higher total returns. There is more to the story than that. So let's start with some data and start to tell that story.
Price Changes and Total Returns Since the Beginning of 2012, 2011 and 2010
|Large Capped MLPs|
Historical Distribution/DCF Ratios
|DCF / Unit||Q1 Distribution||Dist/DCF Ratios|
A story of some consistency: EPD, EEP and ETP
EPD's total return was 126%, while EEP had 43% and ETP had 40% total returns. Take a few seconds to go back to the end of 2009 to see the stats. EPD had a yield of 7.04%; EEP at 7.38%; and ETP at 7.95%. The ends of 2009 distribution/DCF ratios were 81% for EPD; 93% for EEP; and 101% for ETP. The consensus CAGR projections at the end of 2009: EPD at 5.8% with OK distribution growth inertia; EEP at 2.4% in line with its anemic inertia; ETP at an erroneous 4.2% with zero distribution growth inertia. For a tad bit more of yield, the investors in EEP and ETP will give up a huge amount of total return.
Go back to the end of 2010 to see the stats. The yields: EPD at 5.60%; EEP at 6.59%; and ETP at 6.90%. The consensus CAGR projections: EPD at 6.2%; EEP at 2.5%; and ETP at a falling 3.3% with zero inertia. The end of 2010 distribution/DCF ratios were 71% for EPD; 87% for EEP; and 105% for ETP.
Go to the end of 2011 to see the stats. The end of 2011 yields. EPD at 5.28%; EEP at 6.42%; and ETP at 7.80%. The consensus CAGR projections: EPD at 6.0%; EEP at 3.5%; and ETP at a realistic 1.5% with zero inertia. The end of 2011 distribution/DCF ratios were 66% for EPD; 93% for EEP; and 102% for ETP. As can be seen by the yields, the "story" was already partially priced into the market by the end of 2011. Was it too late to make a change at the beginning of 2012? No. Total returns since the beginning of 2011 were 43.17% for EDP; 3.61% for EEP; and 21.23% for ETP.
A story of significant change: BWP, MMP and PAA
Since the beginning of 2010, MMP has had a total return of 181%, PAA at 137% and BWP at 28%. Go back to the end of 2009 to see the yields. MMP at 6.55%; PAA at 6.96%; and BWP at 6.80%. For a tad bit more of yield, the investors in BWP gave up a huge amount of total return.
The 2009 distribution/DCF ratio was 94% for MMP; 87% for PAA; and 115% for BWP.
The consensus CAGR projections at the end of 2009: MMP at 5.0%; PAA at 5.3%; BWP was giving excuses for their DCF under performance that the analyst were buying and had a 5.0% projection with adequate distribution growth inertia. So let's say the story was not clear as the decade began. But as the decade continued, the story changed. Fracking for gas became fracking for oil. And declining oil producing became increasing oil production.
Go back to the end of 2010 to see the stats. The yields: MMP at 5.27%; PAA at 6.06%; and BWP at 6.62%. LTM distribution growth: MMP at 4.93%; PAA at 3.32%; and BWP at 4.04%. The consensus CAGR projections: MMP at 5.0%; PAA at 4.3%; and BWP at 4.5%. The 2010 distribution/DCF ratios: 78% for MMP; 90% for PAA; and 93% for BWP. BWP still had adequate distribution growth inertia.
Go to the end of 2011 to see the stats. The yields: MMP at 4.65%; PAA at 5.42%; and BWP at 7.63%. LTM distribution growth: MMP at 7.38%; PAA at 4.68%; and BWP at 2.43%. The consensus CAGR projections: MMP at 7.0%; PAA at 5.0%; and BWP at 1.5%. 2011 distribution/DCF ratios: 75% for MMP; 73% for PAA; and 107% for BWP.
As can be seen by the yields, the "story" was already partially priced into the market by the end of 2011. Was it too late to make a change at the beginning of 2012? No. Total returns since the beginning of 2011 were 67.86% for MMP; 60.37% for PAA; and 23.61% for BWP.
The tax deferral attribute of MLPs is recaptured at their sale. This makes existing holders of underperforming MLPs over-reluctant to sell. And I believe this selling reluctance makes the underperformers overvalued most of the time. This is bad news for the old money in existing underperforming MLPs. This is potentially good news for new money and new investors if they can correctly perceive the over valuations and avoid the lure of the higher yields.
Let's look at MLP sector returns since 2010 that one can harvest from the major MLP ETNs (exchange traded notes) - and compare that to the returns on the S&P 500 index ETF SPY.
Price Changes and Total Returns Since the Beginning of 2012, 2011 and 2010
Given that the MLP ETNs are cap weighted indexes, for the ETNs to have outperformed, the large cap midstream MLPs would also need to outperform. Why did midstream MLPs outperform the S&P 500 since the beginning of 2010? It was mainly shifts in valuation in a large minority of large-cap midstream stocks - along with shifts in valuations for most G&P stocks.
High "shift" MLPs:
- EPD went from a 7.04% yield to its current 4.29% yield - a 704/429 or 64.1% shift.
- MMP went from a 6.55% yield to its current 3.69% yield - a 655/369 or 77.5% shift.
- PAA went from a 6.96% yield to its current 4.13% yield - a 696/413 or 68.5% shift.
Low "shift" MLPs:
- BWP went from a 6.59% yield to its current 6.87% yield - a 659/687 or -4.1% shift.
- EEP went from a 7.38% yield to its current 6.98% yield - a 738/698 or 5.7% shift.
- ETP went from a 7.95% yield to its current 7.12% yield - a 795/712 or 11.6% shift.
The case that MLPs' valuation shifts also came with changes in the CAGR projection might be strengthened if I could also show evidence that the same link can be found in other sectors. Since I have current stats on the consumer staple stocks that I follow, that is the sector I will use to provide that evidence.
Price Changes and Total Returns Since the Beginning of 2012, 2011 and 2010
|Procter & Gamble||PG||78.34||66.71||17.43||22.50||64.33||21.78||30.23||60.63||29.21||41.28||36.93|
For the consumer staple stocks that doubled or came close to that:
- Flowers (FLO) went from a 2.95% yield to its current 1.90% yield - a 295/190 or 55.3% shift.
- Hormel (HRL) went from a 1.98% yield to its current 1.74% yield - a 198/174 or 13.8% shift.
- Hershey (HSY) went from a 3.33% yield to its current 1.87% yield - a 333/187 or 78.1% shift.
- McCormick (MKC) went from a 2.64% yield to its current 1.95% yield - a 264/195 or 35.4% shift.
Other large-cap consumer staple stocks and their shifts:
- Colgate-Palmolive (CL) went from a 2.14% yield to its current 2.35% yield - a 214/235 or -8.9% shift.
- Coca-Cola (KO) went from a 2.88% yield to its current 2.76% yield - a 288/276 or 4.3% shift.
- General Mills (GIS) went from a 2.65% yield to its current 2.69% yield - a 265/269 or -1.5% shift.
- Kellogg (K) went from a 2.82% yield to its current 2.69% yield - a 282/269 or 4.8% shift.
- Kimberly-Clark (KMB) went from a 3.77% yield to its current 3.33% yield - a 377/333 or 13.2% shift.
- PepsiCo (PEP) went from a 2.96% yield to its current 2.81% yield - a 296/281 or 5.3% shift.
- Procter & Gamble (PG) went from a 2.90% yield to its current 3.08% yield - a 290/308 or -5.8% shift.
- J. M. Smucker (SJM) went from a 2.27% yield to its current 2.02% yield - a 227/202 or 12.4% shift.
I will now provide spreadsheets to show if there have been shifts in dividend growth, EPS growth, or changes in the dividend/EPS ratio.
Dividend Growth and Dividend/EPS Ratios 7-05-13
Stats use the Q1 Distribution for each year
|Dividend Growth over prior year||Div/EPS Ratios|
Sector average dividend growth has slightly fallen the last two years. Average growth the last 9 years has been 10.5%. HRL has had double-digit dividend growth the last four years - and has beaten sector average in 6 of the 9 years. FLO's pace of dividend growth shrank in 2013, but the dividend/EPS ratio signals a return to superior growth. HSY's dividend growth has grown since 2010. MRK has had inconsistent dividend growth.
The sector average Dividend/EPS ratio has moved from low 40s to mid 40s. HRL has dividend/EPS ratios that have consistently been around 30. FLO has a dividend/EPS ratio that has moved from 2 years in the 50s to the mid 40s. HSY has dividend/EPS ratios that have moved from high and inconsistent to 2 years in the mid 40s. MKC has dividend/EPS ratios that has inconsistently been in the mid to low 40s.
EPS Growth & Price/Earnings Ratios 07-05-13
SJM started fiscal 2014 on calendar Q1-13
|EPS / Share||% EPS Growth||Price/EPS||13 EPS Range|
The sector average EPS growth has been very inconsistent and frequently low single digit. HRL has had EPS growth in the mid teens. FLO has had EPS growth inconsistently in teens. HSY has had EPS growth consistently in the teens. MRK has had average and inconsistent EPS growth.
For the two stocks with negative shifts: The dividend/EPS ratios are signaling slower dividend growth in the immediate future for both CL and PG. PG's dividend growth has slowed. PG's EPS growth has been below sector average the last three years.
There is also a headline effect (or acquisition news) happening in some of these price movements since 2010. HRL buys Skippy - and its price jumps. FLO buys the rights to make Twinkies - and its price jumps. And some of the price movement of HSY relates to its prior purchase of Cadbury.
The evidence is not overwhelming, but the consumer staple stocks that have had the larger valuation shifts have also had shifts in their CAGR predictive ratios. And shifts in dividend CAGR projections merit those valuation shifts. The same metric reason for the valuation shifts in market beating MLPs is imperfectly echoed in the market beating stocks in the consumer staple sector - but my key message is that the story is still echoed.
I suspect that many investors will have the impression that reading these ratios is too much like reading tea leaves - that it is more voodoo than science. I could do the same exercise with BDCs, regional banks and REITs. But I hope I have been sufficiently redundant already. I see more science than voodoo in these tea leaves.
What are the lessons in all these numbers?
(1) My Internet search skills are reasonably good. While I can find academic research on the correlation between yields and dividend growth, I have not found anything on the correlation between the dividend/EPS ratio and dividend growth. The correlation may be nebulous and inconsistent, but the correlation appears very logical and real to me.
(2) I am also not finding any academic research on the accuracy of analyst dividend growth projections - or the use of those projections in building an outperforming portfolio. I am surprised that this predictive metric fails to generate much retail investor conversation. This may be due to an unawareness of its existence. At the same time, the metric can be found in most analyst updates. It can even be found in Value Line updates. It is my impression that it is a "don't make a new investment without it" kind of number.
(3) There is bad news in MLP CAGR projections and current yields. Going by those metrics, MLPs are fully valued. The MLPs that have strongly outperformed have had huge valuation or yield shifts as they became correctly priced given their mostly evolving CAGR projections. The big valuation shifts are not going to keep on happening for those same stocks. And most (or maybe all) of the stocks lacking those big valuation shifts failed to merit valuation shifts in the recent past - and do not look to merit them in the short term or mid-term future.
(4) So what should the 2013 class of MLP investors do? And what should the earlier classes of MLP investors be doing with their new money or re-deployed money? New money and new investors need to keep using the same recipe - and look for MLPs where forward CAGR projections are not in line with the current price implied (or yield implied) CAGR projections. If your due diligence fails to find those opportunities, then do not expect to get the same kind of total returns that have come to stocks like EPD, MMP and PAA.
(5) This retired investor can live with stocks that offer 4% partially tax differed yields and 6% annual distribution growth in the core midstream large-cap investment grade holdings that make up the majority of my MLP portfolio. I also own two general partners with larger CAGR projections - and those investments should result in larger total returns. I own four G&P (gathering and processing) MLPs, which offer higher yields and higher distribution CAGRs. But those higher G&P total return expectations also come with higher risks. The earnings from the G&Ps are more commodity sensitive. They tend to have smaller geographical footprints - and that lack of geographical diversification adds to their risk. There is a metric basis for having the expectation that the large cap MLPs may fade to being market tying investments. But it is a bit hard to just tie the market - or the S&P 500. If it was easy, the institutional pros would be doing it. So my advice is for you to be happy with being average in that core portion of your MLP portfolio.
EPD, PAA and MMP have current trends of even stronger distribution growth than 6%. Their cost of both equity and debt capital are low. In a capital intensive industry, that is a major advantage. Their cash flows are relatively dependable. They should be core holdings in any MLP portfolio. The only question an investor should have is when to buy - not whether to buy. With rising yields on the ten-year treasury, the large cap midstream MLPs have yet to have much of a reaction to that change. At the current time, I would be very slow in deploying new money into this sector.