NGL Energy Partners LP (“NGL”)

I recommend a short position in NGL Energy Partners LP (“NGL”).

NGL is a diversified midstream MLP. Since IPO’ing in May 2011 as a pure play retail propane business, NGL has grown Adjusted EBITDA from $24mm to almost $300mm (LTM) by completing over 35 acquisitions. NGL’s textbook play has been to utilize cheap debt and equity to buy low quality businesses at low multiples of cash flow creating immediate cash flow / unit accretion. This allows them to increase their distribution / unit rapidly and “justifies” a low distribution yield / high multiple of cash flow which in turn allows them to repeat the process all over again. Unfortunately, there is a reason these businesses were sold at low multiples. Of its current EBITDA ~40% is fee-based and even less is contracted. Although NGL management has yet to admit it, its businesses will face severe headwinds over the next couple of years due to the recent industry slowdown as well as management missteps. Despite a roughly 30% in unit price fall over the past couple of months, NGL’s current TEV at 14x FY2016 (CY2Q15 – CY1Q16) EBITDA (13x if you strip out TLP value and cash flow) does not reflect the poor quality and outlook of its businesses.

Already, deteriorating business results have increased NGL’s leverage ratio. This, along with its decreased unit price, will severely hurt NGL’s ability to make accretive acquisitions, their historical driver of growth. Furthermore, whether NGL goes forward with the development of the Grand Mesa pipeline project (discussed further below) or not, they will be forced to make a large equity raise sometime 1H CY2015 to decrease leverage, immediately stalling distribution growth and potentially mandating a distribution cut. As business results deteriorate further and NGL’s distribution growth comes to a halt, the market will recognize their businesses for what they are and rerate their cash flow multiple accordingly.

NGL has business five segments:

1. Crude Oil Logistics (Designated growth segment)

2. Water Solutions (Designated growth segment)

3. Liquids

4. Retail Propane

5. Refined Products and Renewables (Includes Transmontaigne)

Crude Oil Logistics (10% of LTM EBITDA)

Consisting mainly of two large acquisitions, High Sierra and Gavilon, the Crude Oil Logistics segment is the most obvious problem child given current industry dynamics. There are two sizable contracted, fee based assets in this segment, NGL’s Cushing storage and a 50% stake in Glass Mountain pipeline, but most of its earning power is driven by crude oil marketing. At its most basic, crude oil marketing consists of buying barrels at the wellhead from producers and transporting/selling those barrels to a market (either a refinery or another marketer). Barrels are bought and sold back to back so there is no price risk and contracts are typically renewed monthly. Marketers make money by selling barrels for slightly more than their purchase price plus transportation costs. This can be a very lucrative business when location differentials are wide, as we saw during 2011-2013. It was during this period of historically high profitability that NGL bought High Sierra (2012) and Gavilon (YE 2013). In CY 2014 as transportation infrastructure caught up with production, differentials collapsed along with crude oil marketer’s profits. Take the Gavilon acquisition as a case study. To say it was a disaster is an understatement. In their acquisition presentation, NGL projected $120mm of run rate EBITDA (7.5x implied multiple), 90% of which would be included in the Crude Oil Logistics segment implying incremental Crude Oil Logistics EBITDA of $108mm. The table below shows NGL’s Crude Oil Logistics results for the four quarters immediately prior to the Gavilon transaction and four quarters immediately after the Gavilon transaction.

¹Includes my estimate of CY4Q14 (FY3Q15) results

Instead of increasing by over $100mm as projected, in the four quarters immediately following the Gavilon transaction, segment cash flow actually declined by $5mm. And the most recent four quarters have been even worse. Furthermore, all of these results were before the recent downturn in oil drilling activity. As production growth slows/stops, competition for barrels at the wellhead will become more and more intense, decreasing marketer’s margins even further. Unsurprisingly, NGL recently filed an 8-K (2/3/15) announcing that David Kehoe, their current head of Crude Oil Logistics, will be stepping down from that role and will continue working for NGL only in his capacity as head of Refined Products and Renewables.

 One mitigating factor to all of this is the recently developed contango market in the NYMEX curve. NGL will now be able to use its heretofore unutilized Cushing crude oil storage profitably once again. According to their CY2Q14 (FY 1Q15) transcript, they have 3mm barrels of uncontracted storage capacity. Assuming they can realize a net $0.30/bbl margin/month on the contango play that’s an additional $11mm/year of cash flow. This will not be enough to counteract the changing tide. The current crude oil environment is one that rewards operating and commercial expertise. NGL has not demonstrated that they possess either of these qualities. My cash flow projections assume that roughly half of the contango opportunity fills in further deterioration of the crude business and the other half is true incremental EBITDA. The Grand Mesa pipeline project is a potential bright spot in this segment; however, it is not a panacea. I will discuss it further below in its own section. 

 Water Solutions (30% of LTM EBITDA)

                Water is a byproduct of oil and gas production and must be disposed of. For a fee, NGL will pump your water production into disposal wells. As this is a fee based business, NGL has been trying to grow it over the past few years. The Water Solutions segment now accounts for the majority of NGL’s fee based EBITDA. Unfortunately, only 50% of NGL’s current water volumes are contracted. This is important. There are two types of water production. There is flowback water and produced water. Produced water comes out of the reservoir with the petroleum products and declines with the production of the well. Flowback water is just that. It is the water that was injected to fracture the well flowing back up out of the well. Flowback water accounts for a very large portion of total water production. In NGL’s case, management estimates that flowback is somewhere <50% of their total volume. The problem with this is that flowback water falls off a cliff in the first few months after the well comes online. In other words, as the rig count drops, flowback water drops. Since only 50% of NGL’s water volumes are contracted and some large % less than 50% of their volume is flowback water, NGL’s volumes and therefore revenues will take a material hit as the rig count declines. I cannot estimate how large this hit will be; however, the rig count has already dropped 25% since NGL’s last reported quarterly results. If current prices hold for even a few more months, that drop will almost certainly hit 50% implying up to a 25% drop in Water Solutions revenue from the previous run rate. On their last conference call (11/11/14) NGL management guided to a 50% increase in Water Solutions EBITDA in FY2016 (CY2Q15 – CY1Q16) over FY2015 (CY2Q14 – CY1Q15). Given the business drivers discussed above, any planned growth / acquisition capital spent in this segment will be filling a large and growing hole.  My cash flow projections assume a 10% decrease in Water Volumes and revenue through FY2016 (CY 2Q15 – CY 1Q16), after which they increase again to FY2014 levels.

Liquids Segment and Retail Propane Segment (55% of LTM EBITDA)

I do not have any special insight into these two businesses. The Liquids business is an NGL marketing and wholesale propane business. Cold weather and price volatility are good for this segment. Cold weather drives volume and volatility drives margins. The Retail Propane segment also needs a cold winter to drive volume. Lower propane prices are good for this business as it is easier to charge a higher margin to customers. These two segments had an outstanding FY2014 (CY2Q13 – CY1Q14) when propane prices at Conway blew out to >$4.00/gal and the polar vortex drove increased volumes. There is absolutely no chance that the combined results of these two businesses can be greater than their results that year.  Management has declared the crude oil and water segments to be the planned beneficiaries of the vast majority of future growth capital, and true to their word very little capital has been spent in either Liquids or Retail Propane since FY2014. To be conservative, my model projects FY2014 results as the new baseline. However, my simple understanding leads me to believe that probably overstates EBITDA by $20mm at least.

Refined Products / Renewables (10% of LTM EBITDA)

This segment consists of the distributions NGL receives from its GP interest and 20% LP interest in TLP as well as the fuels business that came along with the Gavilon acquisition. The Gavilon business has had sporadic results over the past four quarters. My forecast assumes the Gavilon business achieves the results outlined in NGL’s Gavilon acquisition presentation (~$12mm/yr EBITDA). Current quarterly distributions from TLP are running around $4mm. In a previous VIC post on TLP, I outlined why I do not think that TLP will achieve sustained distribution growth. In fact, TLP announced on 1/4/2015 that they held their distribution constant for CY4Q14. To be conservative, my model assumes that they are able to achieve 1% q/q LP distribution growth for CY1Q15-thereafter. 

Grand Mesa Pipeline Project

                The Grand Mesa pipeline will be a 130kb/d pipeline from the DJ Basin to Cushing. This is the type of project that MLPs aspire to: long term contracts, fee based revenues, and a growing volume basin. There is just one problem. The basin is not growing. 8 months ago many forecasters were calling for the DJ Basin to double production from 240kb/d to 480 kb/d over the next five years. However, since that time (even before the price drop) basin production has stalled at 240kb/d.  Although NGL has announced a successful open season and are moving forward with the project, it seems unlikely that they will flow much more than the minimum volume in the near term. My model assumes the project does get built, comes online CY3Q16 and flows at 50% capacity through FY2018 (CY2Q17 – CY1Q18). Construction cost is estimated to be $750mm for an all in EBITDA multiple of 11x. There is likely upside to this project after that assuming oil prices recover and the basin resumes growth.

Catalysts: Distribution Coverage / Leverage / Equity Issuance

                NGL has backed themselves into a corner. They have very poor prospects without building Grand Mesa; however, going forward with Grand Mesa will force them to issue large chunks of equity at unattractive valuations diluting any benefit to existing unit holders. At the end of CY 3Q14, they had $2.2Bn of total debt. By the end of CY YE2014, this figure had increased to $2.5Bn due to the $310mm buyout of their JV partner’s 50% stake in Grand Mesa. Excluding $1.0Bn that is held on the Working Capital revolver and adding all allowed Pro Forma adjustments to their LTM EBITDA produces a seemingly reasonable leverage ratio of 3.5x (this includes the Pro Forma EBITDA adjustment for Grand Mesa). However, given the massive amount of capital needed to build Grand Mesa, NGL will exceed the max leverage ratio allowed in their credit agreement (4.25x) by the end of CY 3Q15 unless they issue equity before then. As NGL’s current revolver interest rate is roughly 2% and their cash cost of capital on their units is over 10%, this is massively dilutive and will force them to stop growing their distribution, and perhaps even cut it. Their distribution coverage ratio for FY 2014 was barely 1:1 in a much better industry environment.  Since then, Crude and Logistics segment imploded, the oil and gas industry has collapsed and they have continued to grow their distribution. Management has already begun lowering distribution growth expectations. In a January 2015 presentation NGL decreased their distribution growth guidance from 10% to “6% - 8% depending on unit price”.  I project that after they issue equity to finance the first tranche of capital for Grand Mesa, they will cut their distribution growth to 0%. Even assuming no additional distribution growth, I project their FY2016 coverage to be .72x. As Grand Mesa comes online CY 3Q16 (FY 2Q17), coverage (still assuming no distribution growth) inches back up to .91x, and finally hits 1:1 in FY 2018. It is worth noting that all of these coverage ratios are based on NGL’s generous definition of EBITDA. If we back out equity comp and only include the actual distributions received from TLP, the coverage ratio dips to .50x in 2016 and only recovers to .80x in 2018. These cash flow shortfalls will have to be financed with debt exacerbating their leverage issue. See Exhibit 1 for cash flow and distribution projections. Also, the projections assume unidentified growth capital spend of $30mm/quarter at 10x EBITDA throughout the projection periods (increasing to $50mm/quarter after Grand Mesa comes online) financed with as much 2% revolver debt as possible within the leverage limit. Their current ROIC is <7%.

Of course all of this begs the question: Why build Grand Mesa? Believe it or not, the future is probably worse if they do not build Grand Mesa. If they cancel the project, you have to strip out the Pro Forma EBITDA adjustment for the project and their leverage ratio immediately jumps to 4.4x forcing an abrupt equity raise to pay down debt. As there are no future cash flows coming to cover the cost of capital, their distribution coverage ratio drops well below 1:1 and stays there until their existing business results improve. As their unit price would get crushed, they would be unable to “grow” their way out of trouble with acquisitions or additional capital projects. This situation would almost certainly necessitate a distribution cut.

Valuation

There are not many publicly traded comps for crude oil / NGL marketing businesses and definitely not for water disposal businesses. If you look at their average announced acquisition multiples then a fair multiple for this business would be between 10x and 8x next year’s EBITDA. Given the small % of contracted and fee based cash flows, these multiples seem reasonable. If you assume TLP’s common units are fairly valued at market and apply a 20x multiple to the TLP GP distribution, then the remaining businesses at NGL are trading at 13x FY2016 EBITDA (assuming NGL GP valued pari passu with LP). Applying a 10x to 8x multiple on those businesses implies a fair value 40%-70% below the current unit price of $29.51. I believe that the announcement of a large equity issuance to pay down debt and/or a further decrease in distribution growth will be the near to medium term catalyst.

NGL will release FY3Q15 (CY4Q14) earnings Monday, February, 9 after market close. They will hold a conference call the following morning. I expect quarterly results to disappoint, and if management is realistic about near to medium term prospects, guidance will disappoint as well. However, based on previous earnings calls, realism may be too much to ask for. But a positive market reaction this week will only create a more attractive entry point for a 3 – 6 month short.