In my March article on Ferrellgas (FGP), I outlined how the company's $2.00 per share distribution has only been sustained by taking on massive amounts of debt, near-continuous issuance of new shares, and the acquisition of hundreds of companies, masking the true decline rate of the business.
The fact that FGP over-distributes is not in question. The company itself acknowledges this fact in its more recent quarterly and annual filings:
We will get to a more comprehensive discussion of "Distributable Cash Flow" later on. For now, simply note that these more recent disclosures appear in the "Liquidity and Capital Resources" section of the 10-Qs and 10-Ks.
Investors are certainly left with the impression that this is a meaningful disclosure relating to the company's liquidity. However, even though the numbers are alarming, a further analysis reveals that the concept of "Distributable Cash Flow" or "DCF" is primarily relevant only to management compensation. In fact, as can be seen in the most recent 10-K, apart from the brief comment in the liquidity section, all of the discussion and disclosure around DCF appears in the management compensation section -- a small portion of which is shown below:
In case there was any doubt that the concept of "Distributable Cash Flow" is really constructed for the purposes of management compensation, a read of the 2010 10-K confirms the term isn't used. Instead of DCF, that filing refers to something called "Incentive Operating Cash Flow" or "OCF." You can see below that the DCF has simply been swapped out for OCF:
I note that in the 2010 10-K, there is no reference to "OCF" in the Liquidity and Capital Resource section of the filing. I suppose changing the word "Operating" to "Distributable" somehow makes it suddenly relevant to liquidity.
However, what would be the point of a simple name change when you can also change the calculations? A comparison of the formula changes is below:
Of course, if one looks all the way back to 2006, the concepts of either OCF or DCF and a "Non-Equity Incentive Plan" don't appear anywhere. Back in those simpler times, we only had salary, bonus, and option grants:
One can speculate that perhaps the Compensation Committee finally figured out that offering stock options on a stock that is worse than flat for 19 years, has amassed a crippling debt load, and enjoys a debt/equity ratio that is infinite due to the negative book value isn't a terribly attractive incentive. Bring on the cash.
In the compensation section of the 10-K, FGP has this to say about DCF:
"Management believes the presentation of this measure is relevant and useful because it allows investors to view the partnership's performance in a manner similar to the method management uses, adjusted for items management believes makes it easier to compare its results with other companies that have different financing and capital structures. This method of calculating DCF may not be consistent with that of other companies and should be viewed in conjunction with measurements that are computed in accordance with GAAP."
Interesting. Hopefully investors don't pay too much attention to GAAP lest they notice the cumulative GAAP earnings over the past 3 years of minus 27 cents (vs. "distributions" of $6.00). But I digress.
What "other companies" is management referring to here?
"Distributable Cash Flow" is a term used extensively in the MLP sector. According to the National Association of Publicly Traded Partners (of which Ferrellgas is a member), their simple definition of DCF is as follows:
A critical and underappreciated point is that, while an MLP, Ferrellgas is a very different animal from almost every other MLP. A Master Limited Partnership is a legal and financial structure, not a business -- this is a very important point. Most MLPs are, in fact, long lived midstream assets -- i.e., pipelines and storage facilities for oil and natural gas. The charts below from NAPTP's "Master Limited Partnerships 101" presentation confirm this:
You can see from the above charts that propane distribution is a miniscule part of the MLP sector. In fact, the NAPTP specifically states the retail sales of propane is an outlier in the MLP sector:
As further evidence that FGP is an outlier, the NAPTP notes:
As noted in my prior article, FGP hasn't increased its distribution once in its 19 years as a public company. If FGP views its "business" as "growing" the share count and its debt load, I suppose they at least have succeeded in that respect.
"One of these things is not like the others,
One of these things just doesn't belong,
Can you tell which thing is not like the others
By the time I finish my song?"
- Sesame Street
To drive the point home that FGP is not really comparable to most MLPs, below is a comparison of the Property, Plant and Equipment from the 10-Ks of FGP and a "normal" MLP -- Kinder Morgan Energy Partners:
Perhaps the point can better be made visually:
The NAPTP conveniently provides a good (if slightly dated) primer from Credit Suisse on the MLP industry. I note this nugget from the report:
It is little wonder that Ferrellgas management would have you compare its "Distributable Cash Flow" to those of other MLPs and use similar calculations to those entities -- the "other companies." Given that Suburban Propane (SPH), FGP's primary public comparable does not use the term either, it seems clear that it wants to be lumped in with other MLPs. Unfortunately, such comparisons make little sense, in my opinion, for reasons that will become clear if they are not already.
As my prior article demonstrated, the propane business is in secular decline. I don't need to repeat what I wrote in March, where I put conservative estimate of the decline rate in the 2% range on the top line, but with fixed operating costs and margins under long term pressure, I conservatively used a 3% organic decline rate for FGP's core business.
In MLP terms, that would be like 1,300 miles of Kinder Morgan's 44,000 miles of pipeline being carried off by aliens every year. If Kinder Morgan then told you that the money it spends to "grow" the pipeline back to 44,000 miles is really "growth capital expenditures," would that make sense to you?
Of course not. However, this is what Ferrellgas would have investors believe in allocating most of their capital expenditures into a bucket they call "growth capital expenditures." Conveniently, these costs are left out of the company's "Distributable Cash Flow" calculation (and thus perhaps even more conveniently management's incentive cash compensation calculations).
Also excluded from the DCF calculation is "acquisition capital expenditures." As noted in my March article, FGP has acquired a company once every 7 weeks on average for the past 25 years.
The table below shows the total capital expenditure by category for the past 5 years, as well as the 5-year average:
Below is FGP's Gross Profit and Operating Income (Excluding Stock Compensation) over the same period:
We are going to be generous and assume that over these past 5 years that gross profit and operating income is flat. Thus, the total capex in the table above was required to sustain this income.
Which brings us to the general thrust of today's article on "Distributable Cash Flow." In analyzing a traditional MLP such as a pipeline, the generic formula of Net Income + Depreciation - Maintenance Capex is a reasonable proxy for SUSTAINABLE cash flow. If you own x miles of pipeline and keep it in a state of good repair (and often with a regulated tolling structure), it's easy to understand how you can keep earning those cash flows over the long term.
Ferrellgas on the other hand, with demonstrated organic erosion in its business, must "grow" and "acquire" just to maintain its "pipeline" (continuing with the analogy).
In simple terms, in my opinion, FGP's true "maintenance" capital expenditure should include all 3 of the above categories (at least over the past 5 years, when no outsized acquisitions were undertaken) when trying to ascertain SUSTAINABLE Distributable Cash Flow. This would make for a more meaningful comparison to other MLPs whose DCF figure is much closer to sustainable DCF than that of Ferrellgas.
The difference is enormous when looked at in that light. Under FGP management's definition of "maintenance capex," it has spent an average of $18.6 million over the past 5 years. Under my (far more defensible in my opinion) definition of sustainable DCF, the number is actually over $65 million on average.
Pulling it all together, using the 5-year average financial performance as a (generous) proxy for the current and forward state of affairs, and using the snappy little definition of DCF given by the NAPTP, we arrive at the following:
Based on the above reasonable and conservative assumptions, I estimate that FGP's sustainable distributable cash flow is only about 35% of the current distribution.
Is it any wonder why the company's book value is negative, the debt load is enormous, and the share count has continued to balloon?
I note that recently management has released guidance of "Adjusted EBITDA" of $245-$260 million for FY 2013. I could write a whole other article on "Adjusted" EBITDA (maybe I will!) which excludes a host of items (such as stock compensation, which is a direct transfer of shareholder wealth), "loss on disposal of assets and other," and a several other real and recurring items, but let's ignore all that for now and assume a "clean" EBITDA figure of $250M. Interest eats up approximately $100M. If true "maintenance" capital expenditures are in the $65M range (I think I have made the case here), that leaves (250-100-65) about $85 million left over to pay a current distribution level of $158 million.
"How did you go bankrupt?
Two ways. Gradually, then suddenly."
- Ernest Hemingway, The Sun Also Rises
Additional disclosure: This article reflects my personal views only. I have a short position in FGP stock. All data and calculations presented are accurate to the best of my knowledge but have not been vetted, checked, proofread, or independently verified. This article should not be relied upon for any purpose other than for entertainment. I welcome comments and or corrections.