A Ranking Protocol For The MLP Space

There are unique aspects of Master Limited Partnerships ("MLP") that warrant a more specialized approach when it comes to comparing various MLP investment choices:

• As a Limited Partner ("LP") unitholder you need to remain cognizant of the amount of cash flow that is being paid to the General Partner ("GP") under the Incentive Distribution Right ("IDR") structures.
• The amount of the LP distribution that is effectively tax deferred varies across the space.
• The Distribution Coverage Ratio ("DCR" = Trailing Twelve Month ("TTM") Distributable Cash Flow divided by TTM Limited Partner and General Partner Distributions) is a relative indicator of the safety of the distribution (i.e. if it is less than 1.0x that is a red flag).
• The DCR is also a relative indicator of potential forward distribution growth (all else being equal an MLP with a consistently higher DCR can more comfortably grow its distributions).

One approach to using MLP specific analysis and ratios involves the following:

3YR FYAT: The 3 Year Forward Yield After Tax ("3YR FYAT") is calculated by projecting the forward distributions after tax for each MLP taking into account the estimated growth rate of their distributions and the estimated portion of the distributions that will be tax deferred and for simplicity sake assuming the highest tax bracket (35% for now, I do know that will change unless Congress extends it before year end, but for now the assumption remains flat at 35%, also the potential end of those tax cuts is comparatively less detrimental to MLP investors when compared to the larger relative increase in C-Corp cash dividend tax rates). The sum of the projected distribution stream after estimated taxes is then divided by the current LP unit price.

Valuation Coverage Ratio ("VCR"): The VCR compound ratio is calculated by taking the Total Enterprise Value ("TEV" = total equity market cap plus total debt less cash) divided by Adjusted EBITDA (TTM EBITDA less the TTM General Partner distributions = "ADJ EBITDA"), and then dividing that result (the "TEV Multiple") by the Distribution Coverage Ratio, so VCR = (TEV/ADJ EBITDA)/DCR.

3YR FYAT/VCR Ratio: This compound ratio is calculated by taking the 3YR FYAT times 100 and dividing that by the VCR ratio.

Example Calculations

(Click to enlarge)

(Click to enlarge)

(Click to enlarge)

What VCR Means

The VCR essentially measures the cost in valuation for an MLP's distribution coverage. A lower TEV Multiple equates to a better/more favorable valuation and generally speaking the higher the DCR the better. A lower numerator is better and a higher denominator is better so, just like a golf score, the lowest relative VCR within an MLP peer group is the "best" and the highest is the "worst." And in a big picture sense the ratio is somewhat comparable across the MLP universe, keeping in mind that the resulting VCR number is most impacted by changes in the DCR and it is most comparable within an MLP peer group. Here's how the Valuation Coverage Ratios look for the Natural Gas & NGL Pipeline segment, this peer group includes Boardwalk Pipeline Partners, L.P. (BWP), El Paso Pipeline Partners, L.P. (EPB), Enterprise Products Partners L.P. (EPD), Energy Transfer Partners L.P. (ETP), Kinder Morgan Energy Partners L.P. (KMP), Kinder Morgan Management LLC (KMR), ONEOK Partners, L.P. (OKS), Spectra Energy Partners, L.P. (SEP), TC Pipelines L.P. (TCP) and Williams Partners L.P. (WPZ):

(Click to enlarge)

So looking at the chart you can see that EPD has the lowest/best VCR and ETP and BWP are the two highest/worst. Both ETP and BWP have DCRs less than 1.0x which effectively increases their VCR making them look relatively worse compared to the rest of their peer group. EPD has a very high DCR (2.25x) which is why they have the lowest VCR (EPD's VCR is benefiting from approximately $1.9 billion of proceeds from asset sales and related transactions, if you back that benefit out EPD's VCR is ~11.3).

What 3YR FYAT/VCR Means

The theory for this next compound ratio is that it takes into account the projected after tax yield and adjusts for the effective LP enterprise valuation cost for the MLP's distribution coverage ratio. Again as a compound ratio the absolute levels are less important and they should primarily be viewed on a comparative basis within an MLP segment peer group. A higher 3YR FYAT/VCR ratio is better and could be driven by the now 3 key inputs:

1. An MLP could have a very high projected after tax yield, possibly driven by a very high projected distribution growth rate and/or differing estimated tax deferrals on the distributions.
2. An MLP could be trading at a comparatively low Total Enterprise Value to ADJ EBITDA multiple, so MLPs without a high IDR burden have an advantage since ADJ EBITDA is after the GP distributions.
3. An MLP could have a very high Distribution Coverage Ratio.

So now in the 3YR FYAT/VCR ratio rankings an MLP that does not have the highest forward yield and is expensive on an TEV/ADJ EBITDA basis could still be better ranked than its peers if it has a very high Distribution Coverage Ratio (intuitively that seems reasonable, the market would place a higher valuation on a safer expected distribution stream). Here are the current 3YR FYAT/VCR ratios for the Natural Gas & NGL Pipeline segment:

(Click to enlarge)

On this metric, TCP is ranked first, so relatively speaking within the peer group TCP's forward yield is a pretty good relative value after taking into account its TEV/ADJ EBITDA multiple (the third lowest in the group) and its DCR (at 1.41x it is the fourth highest in the group, or third highest if you adjust EPD's). EPD is the second highest on this metric which is still being driven mainly by their very high DCR, but again if you back that $1.9 billion asset sales proceeds benefit out EPD's 3YR FYAT/VCR is 1.27, which is more in line with KMP's 3YR FYAT/VCR of 1.29. BWP is again at the bottom driven mainly by its DCR being less than 1.0x.

Limitations

Anything that involves projecting future numbers (i.e. three years of after tax distributions) is always limited as the future is always uncertain so any projected metrics should be viewed with some healthy level of skepticism. The estimated tax deferral percentage is also a rough estimate and can change over time depending on the specific MLP's capital expenditures, acquisitions, etc. Additionally, the growth assumptions are subject to review and changes based on any new data that comes out about a given MLP. The 3YR FYAT/VCR is a compound ratio so it is not a cut and dry scale for comparison versus say Debt/EBITDA, i.e. a 3YR FYAT/VCR ratio of 6.0 is not necessarily twice as good as a 3YR FYAT/VCR ratio of 3.0, whereas Debt/EBITDA of 6.0x does mean that it is twice as high as a company with Debt/EBITDA of 3.0x.

Conclusion

The 3YR FYAT/VCR ratio is one MLP specific metric that can be used as a tool to rank, compare and determine the "best" projected yielding investment within an MLP segment peer group.

Natural Gas & NGL Pipeline Segment Peer Data

(Click to enlarge)

Disclosure: I am long KMR, KMI. I currently have an open limit order to add to KMR at lower prices. I may add to any current positions and may start new positions in any of the tickers shown over the next 72 hours.

Comments

[galicianova]

Actually, i did not own MLP's before the end of last year so i will have to see what all these tax complications mean. but the 25% tax is off any cream send to me. I own KMI which was often explained on SA as preferable to KMR and KMP. is it? but 1 mlp will not solve my quest for mlp asset. so i still would appreciate your present favorite short list (i can purchase ETN but there other riskes, right?)
as for 25% it is not with holding it is chopping off for i will never get it back so let's call the devil by its hoofs. my option is not to play the US market but then at home we have the same 25% so whose coffins my money fills is secondary for me. money triumphs patriotism. Phil, many thanks for your kindness!

[Philip Trinder]

galicianova,

Yes excellent point on ETNs, if you purchase any ETN you are exposed to the credit risk of the entity who created the ETN.

OK so you are long 5 MLPs that will send you K-1s (well in the U.S. they would but again I have no idea how it works outside the U.S.) and an additional 2 MLP related names that send a 1099-DIV.

Here's how I view the group by the MLP Protocol segments:

1 General Partner (KMI), whose underlying MLPs are in the Natural Gas & NGL Pipeline segment, I know Kinder Morgan et al is more diversified than that, including oil and gas production, but their businesses are majority related to the Nat Gas & NGL Pipeline segment

1 Terminaling & Storage MLP (GEL)

1 Natural Gas & NGL Pipeline MLP (EPD)

4 Upstream MLPs (LNCO, VNR, QRE, BBEP)

The larger investment grade MLPs are what I view as the "baseline" risk level in the space so KMI and EPD for example. Everything outside that group means you are taking on higher amounts of risk so those will typically have higher yields BUT higher risk and more volatility when overall market prices go down (see the period from July 2007 through early 2009). From the discussion we've had thus far it seems like the larger investment grade names that are the "baseline" risk level are where you should be focusing your attention. And then perhaps you hold LNCO as a "riskier" name. Linn Energy (LINE and LNCO) is the largest upstream MLP, in fact its Enterprise Value is roughly close to all of the other upstream MLPs EVs added together, so I view it as the lower relative risk choice in that segment.

The current top ranked MLP in the Natural Gas & NGL Pipeline segment by 3YR FYAT / VCR is EPD (you own this one already and its definitely one of the best names, also has a handy DRIP program where you can reinvest distributions at a 5% discount, although again I'm unsure how that would be impacted for a foreign holder).
The current top ranked MLP in the Crude & Refined Product Pipeline segment by 3YR FYAT / VCR is SXL (so may be worth some investigating).

Then of course the KMR and EEQ names are worth looking at given their unique stock dividend paying structures. I'll be interested to hear what happens from your perspective at tax time with the MLPs that send a K-1 and hope that it is not a total pain for you.

Cheers,
Phil

[galicianova]

Phil, thanks for the prompt reply. SO far i did no get any K-1. any dividend i get is shaved by 25%. i know about a number of ets and cef's, but their performance does not look very impressive compared to KMI, LNCO or EPD.which i own, i also have VNR,QRE, BBEP and GEL, but it looks as though it gets out of hand. thus my quest for some structure. but if you can point to a CEF that does charges a reasonable fee and or ETF i will go for simplicity any time. (a side question: is it only 1% which separates them from invidual choices?) many thanks!!

[Philip Trinder]

galicianova,

Have you owned any of the MLPs before 12/31/11?

Given your international tax status it seems like there are two unique MLP related structures that might work well for you: KMR and EEQ. They both pay dividends in the form of additional shares so not cash, which should mean you will only pay the 25% withholding tax when you sell some of the shares. I prefer KMR over EEQ given its relative discount to KMP but I do own both. I have an article about them on Seeking Alpha here: http://seekingalpha.co.... If their link doesn't work the article is called "The 2 Most Tax Efficient Investments in the MLP Space." Personally that 25% dividend withholding tax would drive me crazy.

Cheers,
Phil

[galicianova]

Philip, assuming that one wants to put a given percentage of his investments, with total more important that present income, with upstream midstream and down stream represented, what would at this time the list be?? of course, i don;t want 20 names, this would be a closet index. A more manageable number will be appreciated! - many, many thanks for sharing you take on the subject!!

[Philip Trinder]

galicianova,

Hmm that's an interesting question, so the main objective is total return instead of income?

1. How many names for each segment do you want? Or how many total names/tickers?
2. Are you comfortable with variable rate MLPs that will have their distributions go up and down depending upon underlying cash flow performance of the assets?
3. How do you define midstream? Is that just pipelines or does it include everything in the broadest definition of "midstream", i.e. Gathering & Processing, Terminaling & Storage, etc.?
4. Would the hypothetical investment be made in a taxable account or a non-taxable account (IRA, Keough, etc.)?
5. Is it intended to be a "low touch" portfolio, i.e. buy and hold with less adjustments over time? If so is the time frame something like ten years or longer?
6. If the portfolio goes down 50% in value are you the type of investor to buy more or would you have exited all of the positions earlier as the prices were coming down?

I think if you give me your answers to those questions that might help me narrow it down some with the caveat that I am NOT a Registered Investment Advisor so any response from me would just be my personal opinion at that moment in time based on the analytics that I pull together for my subscription service.

Cheers,
Phil

[galicianova]

Philip, thanks for the reply. of course, the decision for everything i do is mine. With that out of the way let me provide few parameters to help you help me. I'm 66 so his determines he horizon. i dont need
at his point and with all probability wont need it as a monthly income
i consider MLP space as an asset class to invest in. so the total outcome is what matters as well asset safety/volatility. I suppose tha i aim at investment which can be left for at least a month without watching, 3 months even better. do i want to be best investor?- NO!- good, is enough for me. for excellence i'm looking elsewhere. i will readily give 1% of profit for few percents of safety. i want a certain spread among different mlp\s to get a variety, diversification and risk spread.otherwise i don't care whether it is oil, coal or mushrooms. i'm not salmon so i dont do up stream down stream or side stream. anyone will serve me well
However, i would like to keep the # of mlp's manageable. will 10 do it? fine can 7 do it?- even better.

and now to the other parameters. i live 7 time zones away from new york. so my US tax affairs are simple i pay 25% on US dividents and capital appreciation. it is bad but simple..

Philip, if after all that info you are still in the mood to advise, i will be most grateful. as you can see, i'm trying to construct out of zillion data points some coherent and actionable strategy!

[Philip Trinder]

galicianova,

Ah, very interesting, I had assumed that you were a U.S. based investor so thank you for pointing that out (sort of a bias from my perspective not to have asked that question).

Do you currently have any direct MLP investments that generate K-1s? If so how do you handle the K-1s that you currently receive?

My initial impression is that it might make more sense to focus on MLP related investments that DO NOT generate a K-1.

If you would "readily give 1% of profit for [a] few percents of safety" have you looked at the growing number of MLP related ETFs and mutual funds choices that are now available?

Cheers,
Phil

[papaone]

I am concerned that a big change in the tax pass through benefits may be on Obama's radar if he is reelected. Of course, if the MLP's loose the pass through, there goes the party. I suspect any analysis of the total taxes paid if MLP's had to pay taxes Vs. the taxes paid by the investors would show a huge gain in revenue for the government. What are you views?

[Philip Trinder]

papaone,

Thanks for reading and commenting. My personal view is that if you look at the entire energy related MLP space and assumed all the LPs became taxed as C-Corps it wouldn't generate that much revenue for the government. The total market cap of the ones I follow is roughly ~$380 billion (which includes the General Partners many of which are C-Corps and some Marine MLPs that have elected to be taxed as C-Corps) compared to the market cap of just ExxonMobil of ~$415 billion. So there may be other areas for the government to more easily pursue bigger tax dollars but you just never know what will happen when they get backed into a $16 Trillion plus corner.

And you are right that tax change/government risk is the biggest overriding concern for the MLP space and occasionally gives me nightmares...

Cheers,
Phil

[sportscliche]

Realize that there is recent talk of moving "green energy" projects under the MLP umbrella. If this were to happen, I'd think it would add a layer of safety to their tax-favored status. Of course, the idiots in Congress might decide that ONLY wind farms and the like would qualify and separate out anything to do with filthy fossil fuels.

The general population is becoming increasingly aware of what's happening in the various shale plays, fracking, etc. The continued development of these resources will result in lower fuel prices and potential independence from OPEC and all the associated instability. MLPs have been a big part of that story, so why kill the golden goose? Of course, one can never over-estimate the stupidity of our elected officials.

[Philip Trinder]

Very true. The IRS also recently gave a private letter ruling that expanded the definition of "qualifying income" to include facilities that process NGLs into olefins, so an ethylene cracker. This asset is more widely held by chemical companies and would have cash flow volatility similar to what PGH has shown in its two reported quarters. The research I read was estimating that it was WPZ that had requested the ruling due to their pending acquisition of Geismar.

Personally I hope that's the case and that the large number of chemical companies with ethylene cracker assets don't decide to start forming a bunch of MLPs. The bigger and wider and louder the MLP party gets, the more concerned I get that the cops are going to show up and shut it down...

[gfmn2000]

Brucejfern,

I like your thinking. Which MLP's that focus on organic growth do you like?

[Brucejfern]

Another metric that should get more attention IMO is MLPs that are focused on organic growth versus acquisition growth.

I take as an example KMP that plans on growing their distribution 7.6% over the next three years. However, they just barely cover dcf based on those growth rates. In reality, KMP's growth rate should be closer to EPD who maintains a very healthy distribution to dcf ratio.

In an era of super low interest rates perhaps it doesn't matter whether the growth is organic or acquisition related but if we ever get out of this economic mess and interest rates start rising again my thesis is owning the MLPs emphasizing organic growth should translate into superior distribution growth versus acquisition growth oriented MLPs that may struggle more in an era of higher interest rates.

[Philip Trinder]

An excellent point, do you view possible drop down transactions from a Parent/GP as more organic growth or do you view them as similar risk as third party acquisitions? As MLPs have grown, the larger they become the more they need to focus on organic growth throughout the family of entities (so including the GP). Whereas smaller MLPs that have a large Parent or GP with a large amount of assets than can be "dropped down" into the MLP tend to trade at better/lower yields because of the high expected growth rate from the GP/Sponsor selling assets into the MLP (i.e. a very clear growth path versus a small MLP that only has third party acquisitions and internal organic growth).

Also quick point on EPD, it's trailing Distributable Cash Flow as of June 30, 2012 is benefiting from around $1.9 billion of asset sales and related transactions (if you use the company's numbers directly), so their DCR is currently higher than "normal." EPD also has more commodity exposure in comparison to KMP so it should typically carry a higher DCR to provide some cushion for adverse commodity price moves (on page 6 of EPD's June 29 Company presentation they estimate that 80% of 2013 Gross Operating Margin will be Fee-Based, so roughly 20% has some possible impact from commodity prices http://bit.ly/NNLu8b).

[LIVING PROOF]

Thanks, very helpful. Like others I will have to go through this again at a much slower pace before I master the valuations you have introduced. I have been in EPD, ETP and BWP. For various reasons I'm only in BWP at this time. I liked it for the location of their facilities and their pursuit of additional storage capacity. I guess those are things that don't enter into your numerical analysis. Hopefully I can put both to use in future decisions. Any thoughts about that? Thanks again.

[Philip Trinder]

Thanks Living Proof and thanks for reading. Well the projected growth assumptions of distributions does attempt to take into account the ability of the MLPs to grow based on their current asset bases, growth initiatives, recent acquisitions, planned dropped down acquisitions, track record of growing distributions, etc., etc.

One other key aspect is the Distribution Coverage Ratio (calculated on a trailing twelve month basis) when that number is below 1.0x I typically view the MLP as 0% projected distribution growth until the DCR moves back above 1.0x. In the Peer Data Table you can see that any DCR below 1.0x is highlighted in red (so BWP and ETP), which is also probably why they are trading at the 2 highest Current Yields (i.e. the market wants a higher yield for those two since the future growth prospects of the distributions are perceived to be lower than the other MLPs in the peer group).

[rickevantodd]

Your article is excellent and quite thorough. I respect that you took the time to answer the commenters of your article. I will read it a few times just to comprehend all you discussed. I am quite long in the MLP space, unfortunately, not early enough.

[Philip Trinder]

Thanks rickevan and thanks for reading.

[Philip Trinder]

I just noticed that there is a slight glitch in the TCP example thanks to a person who just emailed who was trying to recreate the spreadsheet. The "Estimated Taxes" line item for the Quarterly Distributions is incorrectly using a 90% tax deferral rate instead of the assumed 80% tax deferral rate for TCP so that line item is actually showing half of what the assumptions should calculate as far as taxes. However the "Sum of Estimated Taxes" for TCP of $0.6648 is correct and is using the 80% tax deferral assumption so the 3YR FYAT for TCP is also correct. Apologies for the slight error in the example calculations.

[Harbert24]

Excellent (and new to me) way of analyzing MLP's. I have been in these since I bought OKS in 2002 and over the years have added most of the higher ranked of the MLP's in your analysis. I am a bit lazy so I would like an easy way to periodically analyze the total return of my MLP's without the laborious work of looking up the historical distributions and their timings (I just never kept the data in a spreadsheet.) Is there a data source that I could tap into? Thanks.

[Philip Trinder]

Thank you and thanks for reading. If you need the data for distribution history for particular MLPs that you own the best data source will be the website for the respective MLPs themselves, i.e. you can find OKS' distribution history here: http://bit.ly/LFmbSv. Also it looks like they have not split adjusted the historical data for the distributions prior to the 2 for 1 split in July 2011.

[rick o]

Loved the info. I'm in about 20 mid stream MLP. Heavy into your top picks. I also have a lot of PAA and MMP. What is your opinion of those? Thanks.

[Philip Trinder]

I like both of them, PAA's management presentations are some of the best and most thorough I've seen. As of 7/20 closing prices the top ranked MLP in the Crude & Refined Products Pipeline peer group using the 3YR FYAT/VCR metric was actually SXL driven by its high DCR (although we are in the middle of everyone announcing their 2Q12 distributions and then 2Q financials so everyone is sort of moving around right now).

[milkchaser]

PAA has been on fire recently (maybe a bad metaphor for a pipeline company).

[Philip Trinder]

LOL, yep "on fire" and a pipeline name don't generally play well together in the same sentence!

[gcmagone]

Are there "basket" funds like Cushing's SRV available? I like their huge distribution yield.

[mr091468]

It's ROC. You're getting your own money back and the NAV reflects this by massive decreases. Compare to TYG which has dividends from income generated by the CEF.

http://yhoo.it/MOSmzB

[Philip Trinder]

There are a growing number of MLP focused funds, CEFs, ETFs, ETNs, etc. out there, which is great because they are attracting more capital to invest in MLPs and also since MLPs are difficult to short the MLP focused ETFs etc. can be shorted as portfolio hedges for MLP investors (i.e. I have a tiny put spread on AMJ which I expect to expire worthless). I also have a position in SRV as well (to keep an eye on what they are doing), Jerry Swank (founder of Swank Capital who runs SRV among other MLP focused funds) has also been around the MLP space for a long time and knows it extremely well.

[gfmn2000]

Hello Philip,

Thanks for the information.
I have held EPD and MMP since 2006 and earlier this year bought WPZ and TCP. So your article seems to confirm that I made good choices? I must admit though that I am not really comfortable in selecting MPL's. I plan to study your article later when I have more time. This may help in future investments. Thanks.

[Philip Trinder]

Congrats on being in EPD and MMP since 2006 especially if you've been reinvesting the EPD distributions under their DRIP. The ranking metrics shown in the article constantly change based primarily on underlying MLP unit prices and also on distribution growth assumptions (just something to keep in mind), but yes presently TCP and WPZ are more favorably ranked. For not being that comfortable in selecting MLPs I am highly impressed with your 4 names!

Also while the space has expanded substantially, generally speaking the lower risk names are the larger cap names found in the Natural Gas & NGL Pipeline segment above and also the Crude & Refined Products Pipeline segment: BPL, EEP, EEQ, HEP, MMP, NS (this one seems a little riskier given its earnings preannouncement on July 6 and also not investment grade), PAA, and SXL.

Thanks for reading and commenting.

[galicianova]

if overall epd and oks appear the be the best why, may i ask,
you prefer kmr and kmi?

[Philip Trinder]

An excellent question, by the second chart the 2 best ranked in this particular segment would be TCP and OKS (assuming you adjust EPD). I have substantial (well substantial to me at least) investments all across the MLP space including privately held companies so currently in the publicly traded MLPs I tend to focus more on the riskier/higher yielding names in other segments, i.e. Gathering & Processing, Upstream, Variable Rate, Marine, the soon to be formed Downstream & Marketing segment, etc. I'm currently long CLMT as an example of one that will be included in the D&M peer group. I hold both KMR and EEQ because of their unique structures (seekingalpha.com/arti... and own KMI since it is the General Partner of the Kinder Morgan group of companies. All three of those (KMR, EEQ, KMI) are now all 100% held in my IRA account. As a younger energy banker, I also worked on the Kinder Morgan relationship starting back in the mid 90s so I have a lot of personal conviction in the strength of Rich Kinder and his management team.

If I were to deploy more capital in taxable accounts in the Natural Gas & NGL Pipeline space right now (beyond KMR) I would look very closely at TCP, OKS, and WPZ based on the metrics discussed in the article (well and also EPD because of their good distribution reinvestment plan).

Thank you very much for reading and commenting.

And generally to everyone please fell free to fire away with any questions and challenges to my theories/logic, I have found that vigorous discussion around concepts can help enhance and expand an analysis, so I am open to anyone’s feedback on my MLP specific metrics (especially since I gradually came up with the 3YR FYAT and VCR concepts myself so I don’t think many others consciously look at things this way).

[sportscliche]

I'm curious about private MLPs and how one gets access to them. I'm pretty sure this is not going to be available to the average investor. The only vehicle I'm aware of is the Tortoise fund TTO, although it recently converted from a BDC to an REIT.

I agree that projecting distribution growth is a difficult and risky proposition. I've saved analyst reports from a few years back that gushed about the distribution prospects at ETP, CPNO, KSP, and USS, among others. Things didn't quite go as planned...

[Philip Trinder]

There are other MLP focused CEFs, ETNs, ETFs etc. that have investments in private companies so I think there may be a few ways to get access, for example Kayne Anderson has the fund ticker KED which is 25% invested in private companies: http://bit.ly/MRf0Xd (no position). Also keep in mind that the 25% calculation is based on their in house valuations of the private companies (my gut feel is that these valuations err on the side of being more conservative, I would do that if I were running the fund) so if/when one of them eventually goes public there should be a step-up in the valuation once it is out and trading or if the company gets sold in an M&A deal the fund's partial ownership position will experience a gain. Kayne Anderson knows the MLP space extremely well.