|Shares Out. (in M):||245||P/E||0.0x||0.0x|
|Market Cap (in M):||4,900||P/FCF||0.0x||0.0x|
|Net Debt (in M):||-200||EBIT||0||0|
I am recommending a long position in Cheniere Energy (LNG). The stock is a 2-3 year double to triple ($40-60 target) with limited downside and a strong event path over the next 3-6 months which will drive a re-rating of the stock.
Cheniere has been a hotly debated stock over the years on VIC and much of the details of the company and management have been covered in these discussions so I will try not to be repetitive and will instead focus on what is new and why this is still compelling despite the 40% move it has had over the last 3 months.
Cheniere Energy Inc. (LNG) has a $4.9b market cap ($20.10/share on 245mm fully-diluted shares), no debt and ~$215mm in cash. LNG owns 4 key assets:
- ~60% fully-diluted LP stake in Cheniere Energy Partners (CQP), a publicly-traded MLP
- 100% GP stake in CQP
- 100% stake in the Creole Trail Pipeline (recently valued at $480mm by Blackstone)
- 100% stake of a proposed natural gas liquefaction facility located in Corpus Christi, TX (currently in the permitting process)
CQP is current constructing the first natural gas export terminal in the US at Sabine Pass, LA. As of now Sabine Pass is the only permitted export facility. The plant is comprised of four liquefaction trains totaling 18.0mm tpa of nameplate capacity (4.5mm tpa per train). Trains 1-2 have been financed and are under construction. Trains 3-4 will be financed in Q1 and will break ground next summer. The plants are 90% contracted with 20-year take-or-pay contracts will European and Asian utilities (British Gas, Fenosa, Korea Gas, GAIL). The remaining 10% can be sold on a merchant basis by Cheniere. We have spoken to numerous industry contacts and the feedback we have gotten is that the Sabine Pass contracts are rock solid (i.e. favorable to Cheniere). The four trains are being built by Bechtel (the leading E&C company for LNG construction with 45% global share in liquefaction trains) under lump-sum turnkey contracts. Total construction costs will be ~$10.5-$11.0b on a fully built out basis ($7.8b to Bechtel, $1.1b balance of plant/owners costs/contingency, remainder capitalized interest). The first train comes online in mid-2015 with another train approximately every 6-9 months thereafter (4th train done by mid-2017).
Financing for trains 1-2 was completed last summer. CQP raised a total of $5.6b - $3.6b in bank debt at L+350 and $2.0b in preferred stock from Blackstone (75%) and LNG (25%). Please note that the per share distributable cash numbers below include all dilution from the Blackstone preferred and management compensation associated with trains 1-4.
Once constructed the facilities will do ~$2.4b in contracted EBITDA (excluding any merchant contribution) and CQP will generate distributable cash of ~$3.00/share. At a 7.5% dividend yield this implies a $40/share price for CQP. When you take that back to LNG (through its LP stake and GP splits) LNG will generate $1.0b ($4.00/share) in pre-tax cash flow (no taxes until 2022). Merchant contribution would be incremental to this but at current (wide and likely unsustainable) spread between US natural gas and global LNG prices would yield another ~$3.00-4.00/share in cash flow. The appropriate comp group for LNG will be the publicly-traded GP owners (c-corporations) of MLPs: WMB, KMI, OKE, ETE, XTXI are the key ones. These stocks trade at 3-4% dividend yields but arguably will have higher growth than LNG (assuming LNG/CQP can’t grow beyond trains 1-4 at Sabine). If we assume a much more conservative 7% yield, we get a $57-$85 target in 2016 (vs. current stock at $20) on $4.00-$6.00/share of distributable cash at LNG. As a reminder this is on Trains 1-4 at Sabine Pass only.
Incremental upside optionality which would take these price targets even higher:
- Sabine Pass expansion: Cheniere has signed a 20-year take-or-pay contract with Total for a Train 5 at Sabine. If they get a second contract then Sabine 5 will likely be built (assuming it can be permitted). They are also marketing a Train 6.
- Corpus Christi: In the permitting process
- Financing for trains 3-4: this is launching at the end of this month. The credit markets currently are white hot as we all know and our discussions with project finance lenders indicate this is shaping up to be an easy deal. The company is targeting $4.5b in bank debt and ~$0.6b in CQP equity. We have modeled $1.0b in CQP at $20.00/share in the numbers above.
- Contract for the remaining capacity at the proposed Train 5
- Contract for Corpus Christi capacity
- Once the financing is completed, the there are two primary risks: cost overruns and capital allocation.
|Subject||Why low rating|
|Entry||01/19/2013 03:24 PM|
Not sure why people are giving this a low rating. I'm interested to know why. Thanks
|Subject||RE: Gas Sources?|
|Entry||01/21/2013 07:52 PM|
Yarak - they really have no gas price or basis exposure, it is taken by the customer (with one small exception detailed below). 90% of Sabine Pass trains 1-4 is contracted on a 20-year take-or-pay basis at a fixed $/mcf amount ($2.25-$3.00/mcf based on the customer). That amount is paid regardless of whether or not the LNG is actually dispatched. If the customer chooses to actually exercise their capacity rights and dispatch the LNG, they are charged Henry Hub plus 15% (i.e if Henry Hub is $5.00/mcf they charge the customer $5.75). Of the 15%, around 3-4% is margin for CQP and 11-12% is costs of liquefaction/transport/storage. The only gas sensitivity in the model is that if the economics of gas exports are upside down and customers don't export, then CQP doesn't get that 3-4% margin. But they still get the $2.25-$3.00/mcf capacity payment.
Sabine Pass is tied in directly to Henry Hub via the Creole Trail pipeline they own (already built for the regas terminal, only needs to be reversed). So there is no basis risk.
|Entry||01/22/2013 07:34 PM|
In addition to the matters Biffins raised, there are a few additional issues / common misconceptions surrounding the name that may serve as short catalysts in the next 3-12mo. They can be illustrated using some of the assumptions found in the write-up:
These restrictions may be less onerous now that Cheniere seems to be moving away from bank financing for phase 1 and back to the public markets. At least the decision to rely on note issuance may help with the transparency of the cost of capital. Given that note issuance to finance phase 1 was just announced, it may be a stretch to refer to financing in the past tense.
Common belief: Trains 1-4 are built “under lump-sum turnkey contracts”
Reality: (Moody’s notes on the 1st lien bank loan concerning Trains 1-2)
“$739 million of provisional sums that are not currently fixed in the EPC contract, sizeable
owner costs, the project's location in a region susceptible to hurricanes and reliance on
$180 MM of Train 1 revenues to complete construction that reduces the benefit of the
$300 million in budgeted contingency“
Common belief: “The appropriate comp group for LNG will be the publicly-traded GP owners (c-corporations) of MLPs: WMB, KMI, OKE, ETE, XTXI are the key ones.”
Reality: The sources and credit risk of the revenue stream of LNG differ quite a bit from those of the diversified US mid-stream MLPs: ~40% of phase 1 revenues are backed by a Spanish utility (Gas Natural Fenosa). While LNG may fit in the same investment bucket based on technical / market indexing criteria, it may display different trading characteristics if investors perceive a marginally higher risk of receiving 40% of the negotiated payments for T1,2 (20% of the total) in pesetas.
Common belief: Upside from further trains and arbitrage profits.
Reality: Insufficient authorized export volumes for arbitrage trading post train 4.
Even on the valuation front focus on theoretical FCF to equity in seven years misses the massive valuation the enterprise already receives: The aggregate US pipeline space trades ~14.5x current EV/EBITDA (at 10-yr heights) implying ~7% cost of capital. The 10-yr average is ~11.5x, or ~9% cost of capital. These are generally established presently cash flowing assets with none of the above credit or project risks. Meanwhile, Cheniere corporate family trades >=10x 2019 (projected contracted) EBITDA – far richer than the space of established players considering these are cash flows 6-7 years from now (depending on the discounting endpoints one chooses) and with at least 1x turn of EBITDA downside to the peer valuations (assuming peer cost of capital).
|Subject||RE: RE: RE: Catalysts|
|Entry||01/23/2013 07:06 PM|
Thank you for a thoughtful question about the big-picture of the comepetetive landscape. While not relevant to the short-term performance of the shares, the answers may demonstrate why LNG is a structural short at current levels even absent project, financing, and execution hair:
Assets that provide alternatives to potential acquirers are both foreign and domestic. For simplicity, let’s analyze two outcomes for US export authorization:
Scenario 1: US LNG exports are broadly authorized. (This appears to be the most likely outcome given the current trends)
There are about two dozen proposed US LNG export projects adding up to >20Bcf/d of liquefaction capacity. Most of these are conversions similar to the Sabine pass with roughly similar economics and value proposition. Many of these are also already affiliated with majors and large independents (since it took a large balance sheet to build an import project to begin with). The entire global LNG market was about 33Bcf/d in 2012. IHS CERA projects that if all the proposed US projects are completed, US (far from the low cost source of supply as discussed lower) will be ~40% of the projected non-US 2020 liquefaction capacity. Given the massive oversupply of 2018-2020 export volumes that the above US export figures would imply, IHS CERA projects only ~5Bcf/d of US liquefaction capacity coming online by 2020. Even if their 2020 Global LNG demand is significantly off, it seems safe to say that a typical US LNG import facility will not be converted to export and there is no scarcity value for these assets. We can drill into the drivers of value of the export facilities that can support exports at slightly lower cost than the competition and will thus be built and hopefully remain economic, but it seems pretty clear that there is no scarcity value for US import facilities that can be re-directed to export and that in general the value of these should not exceed replacement costs.
Note that in this scenario Sabine Pass receives the authorization to export additional volumes beyond those already contracted, but there is no material arbitrage profit from post-2018 US LNG exports. In this case the cost of supply from Sabine Pass is not materially economically differentiated from the competing US projects which are problematic relative to the ROW (discussed below).
Scenario 2: Sabine Pass is the only authorized US export project.
In this scenario Sabine Pass competes with greenfield Canadian and ROW greenfield and brownfield projects. While Sabine Pass has lower capital costs than ROW greenfield projects, such capital costs are generally higher than those of ROW brownfield projects. (Liquefaction projects are typically built with an eye to many stages of subsequent expansion leveraging sunk costs.) Assuming mid-cycle NA HH price of $4.00/Mcf, Sabine Pass specifically and NA LNG in general are some of the highest-cost sources of supply to the Atlantic and Pacific LNG markets. You can consult any of a number of industry research report on the full-cycle cost of new LNG supply to these market from various sources. While differing in the details, the rough numbers should be along the lines of: Middle East brownfields are expansions of the existing Mammoth LNG projects which benefit from the existing Marine infrastructure and scale of the original construction with subsequent expansion in mind. These offer lower capital costs than Sabine Pass (think massive subsidies, lower regulation, energy prices, and legions of cheap foreign workers) as well as near-0 feedstock cost. They are also some of the easiest to bring to market (relatively) quickly. These can supply LNG to the Pacific at a cost of <$4.00/MMBtu. Middle East greenfield projects are in the middle of the pack with higher capital costs but few other costs and with a total cost of supply of ~$5.00/MMBtu (basically all capital and a bit of transportation). In general, <$8/MMbtu one finds all of the ROW brownfield projects as well as African and some Pacific greenfield projects (think of all those Tcfs of gas that IOC supposedly found in PNG which has 0 alternative value so only development and capital costs). Sabine Pass probably can’t compete with the cost of new supply from most of the above even at $3.00/Mcf HH in perpetuity.
In this scenario Sabine Pass at >=$8.00/MMbtu is one of the higher-cost suppliers. (There is a reason LNG does not illustrate competitive economics of the project.) Sabine Pass could warrant some scarcity value given the geopolitical diversification it offers, tough much of the competing supply is from reasonably politically safe sources also. Even with scarcity of US diversification options in mind it is challenging to see an acquisition at ~2x the build cost for a relatively high-cost project attached to a feedstock marketplace that retains some political uncertainty of its own. Note that in this scenario Sabine Pass in not authorized to export any volumes beyond those already contracted – an acquisition may thus be practically impossible.
|Subject||RE: RE: RE: RE: Catalysts|
|Entry||01/24/2013 07:07 AM|
Guys thanks for the pushback/thoughts. dr123 i hear you on the overbuild risk but keep this in mind:
- LNG's plants are fully contracted for 20 years w/take-or-pay deals. So regardless of what happens w/the issues you raise, LNG is generating a $4.00/share dividend (20% dividend yield) on the contracted business alone. Merchant value, incremental terminals, etc are just upside to that. At current merchant spreads the dividend yield would be closer to 30-35%.
Also I think you overstate the overbuild risk - projects are too large and only move forward w/contracts in this area, there is no merchant LNG spec building. The LNG market is pretty much a contracted market where merchant volumes result from secondary trading of cargos amongst the off-takers. So regardless of the number of projects proposed, only those that can convince customers and get contracts will go forward and at some point customers will stop signing contracts. Cheniere is well past that point though - they are fully contract on Sabine 1-4 and now Sabine 5 (not in the numbers above btw).
|Subject||RE: RE: RE: RE: RE: Catalysts|
|Entry||01/25/2013 12:49 PM|
The discussion of the relative attractiveness of alternative projects vs. the Sabine Pass was meant to illustrate the lack of economic rationale for an asset transaction at a material premium to replacement cost that rookie was inquiring about.
While the competitive landscape in theory does not affect the signed contracts, it matters if one wants to make a case for any value from expansion optionality and arbitrage profits. In the absence of any scarcity of US-based and lower-cost brownfield projects shopping around their 2018+ LNG volumes, discussions over the take-or-pay margins seem to be drifting to the logical floor of the level needed to provide a return on invested capital but no windfall. In practice, given the apparently less than forthright communication of LNG’s current project and financing arrangements, one would not be surprised if the take-or-pay deals offered similar escape mechanisms for the counter-parties as the EPC and the proposed financing arrangements for Trains 1 and 2.
In the perfect world where everything goes according to plan, I see LNG generating ~$3/sh in 2019 CF. Note again that this is using fairly heroic assumptions that Biffins would rightfully scoff at. The delta with your cost of capital and dilution assumptions is discussed on earlier threads so I will not go into that here. Trading at ~7x 2019 CF is quite expensive, even ignoring the leverage and assuming one is comfortable with projecting anything 6 years into the future. In practice, such approach breaks down since it ignores the cost of capital and the distance of the cash flows: With ~$40Bn in annual FCF and assuming 0-growth AAPL is trading ~3.5x (2019) earnings per such approach. I would submit there is about as much certainty about AAPL’s revenue path over the period as about LNG’s hair. Once the TVM is considered, $3.00 in 2019 CF is worth ~$1.00 on 1/2013 given the ~15% discount rate on Blackstone’s (senior and thoroughly protected) equity.
|Subject||Responses to dr123 and biffins|
|Entry||01/25/2013 12:55 PM|
Rebuttal to the points you raised (in no particular order):
"Trains 1-2 haven't been financed" - i think this is incorrect. Unless I"m reading it incorrectly the language you cite from the credit agreement simply requires that equity fund first (creating the collateral equity for lenders) before debt is drawn. This is standard stuff.
"Trains 1-4 aren't fixed price" - yes they largely are. There are exposures as you cite but they are on a very small % of the project. Ask anyone at Blackstone or anyone who has done due diligence on the projects and the EPC contract from the lending site - the costs are largely fixed.
"Can't arb volumes no export authorization" - this is arguable but possibly correct, however even if it is correct, getting a FTA authorization requires 3-4 weeks and a five page letter (the DOE issues them constantly, XOM just got one recently). Non-FTAs are where the bottleneck is but Chenere won't need non-FTA to place those 2mm tpa. Cheniere hasn't directly resolved this discrepancy w/the DOE formally yet b/c the focus has been on getting the projects underway, financed, under construction and didn't want to risk any delay to this b/c that will create 95% of the value here. However once construction for 3-4 is underway this will be formally resolved. And getting FTA authorization for the incremental 2.0mm tpa is a given
"Trains 1-2 will come in overbudget b/c cost is much lower than any other projects in the world so it must be understated" - this reflects a fundamental misunderstanding of the project. Other LNG projects are built in remote areas w/no infrastructure and are fully vertically integrated (i.e. upstream onshore or offshore wells, upstream oil/gas processing facilities, pipelines to the liquefaction facility, a power plant to power the liquefaction facility, full accomodations for all the workers who have to be flown in and out and paid very expensive wages, construction of storage takes and a deep water port. And these projects are happening in extremely remote areas (western australia, papua new guinea, east africa). The Cheniere project simply isn't comparable - you're just building a liquefaction plant and basically nothing more. No upstream solution (it's already tied into henry hub and pipeline is built, it just needs to be reversed), no storage or deepwater port (already built). They have grid power.
The cash flows aren't really sensitive to whether volumes are flowing through the liquefaction plant or not - our math is ~7-8% swing to cash flow
|Subject||RE: RE: RE: RE: RE: RE: Catalysts|
|Entry||01/25/2013 01:04 PM|
What's the relevance of replacement cost here? You have signed contracts, this isn't a merchant plant. If I have a widget factory and I was the first to build it when widgets were scarce and I contracted it out for 20 years at $5/widget but market is actually now $1/widget b/c the price has fallen to reflect replacement cost economics how are you going to value me?
Again on expansion optionality, it's all about signed contracts. 4 contracts that are currently signed yield $4.00/share in distribution at LNG with no arbitration profits. They have received a signed contract for Train 5 at Sabine so (assuming they get a permit to build it) that is real optionality b/c it's a signed contract. That train 5 isn't in the $4.00/share dividend.
I don't see how you get to $3/share in 2019 cash flow. I also don't see how you can possibly compare AAPL's future earnings to LNG's. LNG is contracted with literally no debate about what they will be making on those contracts. Who knows if AAPL can continue earning 50-60% gross margins on a consumer electronics sale where until now they've had no real competition.
Also using blackstone cost of equity is a mistake - blackstone moved first and got a good deal. equity being raised now is being raised in the market at substantially less attractive terms (witness CQP equity deal a few weeks back at $24 vs. Blackstone at $15).
Why do you think LNG hasn't been forthright in communications about their financing arranagements? They've disclosed terms of everything as they've reached deals in SEC filings and have never obfuscated any of the details to me. Obviously they haven't told anyone what they are going to do in advance because 1) they haven't known 2) it would be illegal to do so.
|Subject||RE: RE: RE: RE: RE: RE: RE: Catalysts|
|Entry||01/28/2013 08:06 PM|
We are in vigorous agreement that replacement costs are not relevant to the economic value of the signed contracts. The discussion was relevant to “buy vs. build” questions that rookie raised. The discussion is also relevant to the future arbitrage economics, the value of any optionality from future trains, and finally the (probably remote) optionality that the counterparties try to get out of the $5/widget contracts if/when the forward price declines to $1 (to build on your analogy).
While there is no debate about the revenue associated with the signed Sabine Pass contracts, there seems to be a fair bit of uncertainty about the pro-forma EV at the time such contracts begin generating revenue. The end result is a pretty wide range of plausible per-share CF outcomes in 7 years. The AAPL analogy was (a poor) attempt to illustrate how the uncertainty in _per-share_ metrics yields suspect results in valuation on estimated distant future profitability.
The various 2019 profitability scenarios were beaten to death on the 11/26/2011 and 12/21/2011 LNG write-ups and Snarfy’s model (with adjustments discussed on the above threads) can be used to see the range of (relatively optimistic) assumptions that yield $3/sh in 2019 CF. It is difficult to comment on your numbers without a model/access to the calculations of the expected 2019 capital structure. From a cursory glance your $11.0b ceiling for total cost seems quite a bit off from my floor of ~$12.0bn.
It is possible that the Blackstone deal is less relevant in the current environment. So one can look at your peers (WMB, OKE, etc.), keeping in mind their lower risk (known capital structure, domestic exposure, and lower project risk) as well as the trailing and projected growth (supported by the generally perceived secular tailwinds for the NA mid-stream infrastructure in the coming years). These generally trade 9-10x NTM CF and if one builds out a 7-year model (an uncertain exercise but no less complex than estimating Cheniere’s future cap structure) around a similar 3-4x 2019 CF that the cost of Blackstone equity implies for Cheniere.
For discussion on the clarity of LNG’s disclosures I will again references the past threads that are too extensive the be re-hashed here. Selective disclosure of allegedly successful financing in various news snippets while management was selling shares outside of plans (and presumably were not in possession of material non-public information regarding the state of financing is one of my personal favorites). Restrictions on financing in the text I quoted go beyond the standard requirements for equity raise. Hence, CQP is still currently in the process of raising debt for trains 1 and 2.
The bears seem to agree that the costs are _largely_ fixed and seem to concur that the project will be _largely_ on budget. This is quite different from claiming that the costs _are_ fixed. The problem is that the model has powerful feedback loops so small slippages in cost and schedule build up to have non-linear impact on the value of equity.
Sabine Pass is comparable to brownfield projects in areas with similar/lower cost structures to the USGC when upstream costs are excluded or are negligible (say, the Middle East). Such brownfields have existing storage, marine, and pipeline infrastructure as well as lower energy and potentially lower labor costs and lower regulatory burden. They also don’t have the weird headaches of pipeline reversals or contending with existing regasification volumes/pipeline flows which remain cloudy. No one is arguing that say AU projects are directly comparable but even those can be reconciled by looking at their expansion/brownfield cost components only with the upstream costs excluded and adjustments made for higher labor costs. But why would one expect an expansion of an existing export facility in Qatar (at a cost of ~$1K/tpa) built with a runway for multiple stages of expansion to be materially more costly than a conversion of an import facility in the USGC?
We have not discussed the cost estimates with Blackstone folks but we have discussed them with one of the technical shops retained by the company and funders. Their estimate for Sabine Pass construction costs is only slightly higher than the company’s (at ~$600/tpa, excluding any financing costs) As we worked through the assumptions and the comparable projects used we could not quite get to the shop’s numbers. In our discussion the commentary went along the lines of: “Well, we do not think they will hit our estimate and we can’t get really get to our estimate ourselves but we can’t publish anything much higher and call Cheniere out since they are a client.”
|Entry||01/30/2013 09:51 AM|
CQP just issued $1.5b of debt at 5.6%. Original deal size was $1.0b and it was upsized to $1.5b, there was ~$5b in indications.
Assuming the bank group rolls the incremental capacity into the trains 3-4 financing (which is the plan and current indications from the banks), that means they've now raised ~$2.5b of the total $4.5b. Banking group from the $3.6b facility had originally indicated ~$1b in incremental exposure they could take for trains 3-4. The $1.5b bond deal increases that to $2.5b.
This leaves $2.0b to raise. Given the oversubscription I think it is likely they do another secured bond deal and/or some combination of secured/unsecured bonds when they go to market next month for trains 3-4 financing and that likely completes the financing. Given the current state of the financing markets i think this materially reduces the dilution risk going forward.
|Subject||RE: RE: Debt deal|
|Entry||02/07/2013 01:23 PM|
Last 2 years fully diluted average shares in LNG per quarter.
Diluted Weighted Avg Shares
|Subject||RE: RE: Author Exit Recommendation|
|Entry||09/12/2014 09:01 AM|
In truth it's not driven by any specific risk or concern but rather a view that between $85-$100 pretty much fully values Sabine and Corpus on a present value basis in an MLP structure and the debate at this point is really about cap rates and how aggressive you want to be there. I'd be curious to get your thoughts on how you are thinking about upside -- I'll concede I might be too conservative in stepping aside at $85 but ultimately the upside here is bounded - it's a discrete set of two assets, we largely know the economics from those assets and the cash flows don't grow, so its a question of what that stream of cash flow is worth by the end of the decade when these projects all come on.
|Subject||Re: Re: any thoughts lately?|
|Entry||02/03/2016 07:25 PM|
I've done a fair amt of work on the CQP entity, esp the SPL box where the bonds trade in the mid 80s to low 90s, YTM of ~8% and current btwn 6.5% to 7%. I think these bonds are money good there's a good chance these get refi'd at the CQP HoldCo level (similar to recent proposed deal) in ~2 yrs, once the tolling arrangements come in, making YTM closer to 10%.
As for LNG, I'm still having a hard time getting bullish on the ultimate equity given I don't have a view on LNG prices & dont' believe the current tolling agreements are enough to give me confort all the way to the LNG level.