Golar LNG Limited (GLNG) Q2 2025 Earnings Report, Transcript and Summary

Welcome to Golar's Q2 2025 Earnings Results Presentation. My name is Karl Fredrik Staubo, I'm the CEO of Golar LNG. And today, I'm accompanied by our CFO, Eduardo Maranhao. Before we get into the presentation, please note the forward-looking statements on Slide 2. As normal, we start on Slide 3 and an overview of Golar today. Golar is a focused FLNG company. We own three units where two is on the water and 1 under conversion. The key event of the quarter is securing the 20-year charter for the Hilli as well as signing definitive agreements and consequent final investment decision for the Mark II FLNG. On the back of securing 20-year charters for our existing fleet, we're increasing focus on adding additional FLNG units as we see strengthening demand for FLNG solutions to monetize stranded associated and reinjected gas opportunities. During the quarter, we increased our cash position through a $575 million convertible bond issuance, and we now have a cash position of just under $900 million. With a fully delivered net debt-to-EBITDA ratio of around 3.4x, $17 billion of EBITDA backlog before commodity upside and 20-year cash flow visibility, we have capacity to fund incremental FLNG growth units from financing proceeds. Turning to Slide 4 and a look at our EBITDA backlog and cash flow visibility. On the back of the Q2 contracting events, our existing fleet is now fully contracted for 20 years. The highlights, as mentioned, was fulfillment of all the CPs and FID of the Hilli as well as the definitive agreements for the Mark II in May as well as FID now on August 6. The Mark II FLNG charter remains subject to regulatory approvals and customary conditions precedent. We expect all of these CPs to be met within 2025, and we are progressing according to schedule to meet them. The backlog stands at $17 billion before inflationary adjustments and commodity exposure through the SESA contracts that can together add significant further upside to the backlog. Turning to Slide 5 and an overview of the global FLNG fleet. Golar remains the market leader measured by liquefaction capacity. The global fleet consists of 7 units on the water and 4 units under construction. We expect to see increased newbuilding activity from the existing FLNG owners combined adding up to 5 units of announced FLNG project plans if they all progress according to schedule. We also see increased interest from gas resource owners increasingly interested in FLNG solutions to monetize gas. This development is a testimony to the attractive costs, proven operational track record, and flexibility that FLNGs offer versus traditional land-based liquefaction solutions. Golar remains the only proven provider of FLNG as a service. The other companies as presented on this slide, only monetize gas that they control. A contemplated fourth FLNG unit by Golar will therefore, likely be the only available open FLNG capacity with delivery within this decade. We believe we can drive best value in ordering our next unit before locking in an associated charter. This is similar to the successful strategy executed when ordering the Hilli and later the Mark II FLNG. Once we do secure a potential long-term contract for our next unit, we plan on continuing our growth ambitions with the fifth unit. We will remain with our strategy to only have one open FLNG at the time, and we expect to be able to fund all visible growth plans with proceeds from debt financing activities of the existing fleet and the overall balance sheet flexibility. We plan on continuing our growth trajectory as long as we can obtain accretive and solid economic returns to our shareholders. Turning to Slide 7 on quarterly highlights and developments. The second quarter was a very active and very successful quarter for Golar with several key milestones met. As alluded to a few times already, all the CPs and FID for Hilli redeployment charter in Argentina was met. Simultaneously, we signed a definitive agreement for the Mark II and later the FID now on August 6. In June, we issued $575 million of convertible bonds and used around $103 million of the proceeds to buy back 2.5 million shares or just under 2.5% of our company. Eduardo will detail some of the benefits of the CB later in the presentation. In July, we issued our annual ESG report containing detailed information about our operations and the benefit to the countries we operate in. It's worth a read for those interested in development that we provide to the countries we operate in. During the quarter, we saw the exit of two of our existing Board members, Georgina Sousa and Thorleif Egeli. We would like to thank them for their long- term contribution to Golar's Board. We also welcomed three new Board members, Benoît Fouchardiere, Mi Hong Yoon, and Stephen Schaefer. We're excited to have the addition of these three members and look forward to their contribution to Golar's continued success. Consistent with our communicated strategy of adding accretive FLNG growth once the existing fleet has been derisked and committed on long-term charters, we have increased our efforts for our fourth unit. We have signed agreements with our target shipyards to confirm EPC price and delivery schedule for the next unit, and we expect to decide on design, size and shipyard in the coming months. Turning to Slide 8 for an update on Hilli. Again, the key milestone for Hilli during the quarter was the conclusion of our 20-year redeployment charter in Argentina, which adds $5.7 billion in EBITDA backlog before commodity upside. Under her existing contract, Hilli continued her perfect economic uptime and have now delivered 137 cargoes since contract start-up or exported more than 9.5 million tons of LNG from Cameroon. We're in the final phase of entering into a redeployment, upgrade and life extension contract for the shipyard scope in between the units contract in Cameroon and before starting up the 20-year contract in Argentina. We expect this yard contract to be concluded within the third quarter of this year, and the contract is in line with our budgeted redeployment scope. Turning to Slide 9 on Gimi. The key milestone for Gimi during the quarter was reaching commercial operations date in June. This is a massive milestone for Golar and our partners, BP, Kosmos, Petrosen, and SMH and concludes a long-awaited project start-up. The unit is now conducting an appraisal period where we will fine-tune the FLNG to maximize liquefaction capacity. We're satisfied with the vessel performance to date, and we're now in the progress of offloading cargo #8 since starting operations. The contemplated $1.2 billion Chinese sale leaseback refinancing facility has taken longer than approved by our stakeholders. The facility is fully concluded between Golar and our Chinese counterparts. The delays caused are caused by the stakeholders outside of our control. However, with commercial operations date now concluded, we will work to develop alternative refinancing structures and believe that we can obtain facilities with improved terms versus the $1.2 billion Chinese sale leaseback facility. The picture on the bottom left is from the first cargo celebration on the GTA field on May 22, and the event was attended by the Presidents of Mauritania and Senegal as well as the CEOs and senior management of all the stakeholding companies. This marks the first phase of the GTA project, and the partners will now turn attention to the next phases of the project with planned production increases that may involve additional FLNG units. Turning to Slide 10 and the Mark II FLNG. The second quarter, as alluded to, was also an eventful quarter for the Mark II FLNG ordered in September last year. We signed a definitive agreement for a 20-year charter on May 2 to operate alongside Hilli for SESA in Argentina. On August 6, we obtained the final investment decision for the charter. And as explained, we're expecting the remaining CPs to be fulfilled within '25. The conversion itself is progressing according to schedule, now almost 19% complete. The unit is scheduled to be delivered during Q4 '27. And upon completion, the FLNG will sail to Argentina to start our 20-year operations during 2028. As you can see from the pictures, the donor vessel, the LNG carrier, Fuji, has now been cut in half and skidded on to land. The picture to the bottom right is from the fabrication of the sponsons that will be added to the width of the vessel. We have funded approximately $800 million of the $2.2 billion conversion budget fully with equity to date. On the back of securing a long-term charter, we're progressing financing initiatives for the unit and expect to add asset level debt within 2026. The proceeds from such financing will be used to fully finance the delivery and an equity component into Golar's next FLNG. As part of the yard contracts, we also hold an option for a second Mark II FLNG with an estimated construction time of about 38 months from ordering. Turning to Slide 11, where we provide an overview of the key commercial terms of our existing FLNG contracts and some of the risk mitigations and embedded upsides in the contract. Golar does not control where attractive gas reserves are located in the world, but we can structure our contracts to mitigate risks and maximize return for our stakeholders. Some of the means to do so is highlighted on this slide. All of our FLNG contracts are on English law, paid in U.S. dollars, and paid offshore. We work with strong charter counterparties that effectively create a buffer between us and the host governments in the geographies where we operate. In the case of Hilli in Cameroon, that buffer is Perenco. For Gimi offshore Mauritania and Senegal, it's BP. And for our Argentina contracts, it's SESA, which is a consortium comprising Pan American Energy, YPF, Pampa Energia, Harbour Energy and Golar LNG. Some of our contracts have meaningful upside through commodity links. For the case of Hilli in Cameroon, Golar earns $3.1 million in excess cash earnings for every dollar Brent is above $60. We also make $3.7 million of annual free cash flow for every incremental dollar per MMBtu of the TTF gas price with no ceiling. In the case of Gimi, we do not have a commodity-linked upside, but we have a utilization-linked upside if we achieve commercial utilization north of 90% of nameplate. In the case of the CEA contracts for Hilli, Mark II and our ownership in Southern Energy, we make approximately $100 million in excess annual earnings for every dollar offtake is above $8 per MMBtu. The total downside is around $28. So we have a $28 million downside of less than $7.5 per FOB, and we have $100 million upside. So it's a highly skewed risk reward and one that we think could provide significant excess value to Golar. Turning to Slide 12 and the progress made on our FLNG growth units. We have three key vessel designs, and we differentiate them by Mark I, II, and III. The difference is effectively the liquefaction capacity, so you could categorize them as small, medium, and large. Mark I can produce up to 2.7 million tons per annum, Mark II, 3.5 million tons, and the Mark III, 5.4 million tons. Both Mark I and II would be vessel conversions and Mark III would be a new build. Given the slightly different approach to construction, we would also select shipyards accordingly. Both Hilli and Gimi is built at Seatrium Shipyard in Singapore. And if we are to proceed with the Mark I, that's the likely shipyard selection. We are currently working with Seatrium to confirm an updated EPC price and delivery schedule. The Mark II FLNG is currently under construction at CIMC Raffles. And as already explained, we hold an option to do a second unit at CIMC Raffles, and we already know the design and the price as we have just ordered this unit. For the Mark III, we see a slightly longer conversion time. We see around 38 months for the first 2 and 48 for the Mark III, given that you start from scratch and not with the conversion vessel. We've also signed agreements with Samsung Heavy Industries to develop updated pricing and delivery schedule if we are to order such a vessel within this year. As explained earlier in the presentation, we expect to get the final price and delivery schedule in the coming months and thereafter make a design decision for where we will continue with our fourth unit. We do see increasing long lead times for critical components, and we, therefore, plan on securing slot reservations within Q3 2025. During the quarter, we have also completed inspections for potential donor vessels should we continue with the Mark I and Mark II FLNG conversion. If we do secure charter of a contemplated fourth unit, we expect to rapidly thereafter order a fifth unit, which is why we are progressing yard discussions with all three yards at the moment. Turning to Slide 13 and highlighting some of the cost benefits of FLNG versus land-based solutions. As you can see from the far left, there is plenty of proven gas reserves in the world that are currently not being produced. If it's not being produced and it's proven, you tend to get them cheap. The second graph is the cost of liquefaction. As we are building a generic standardized design in the far East, we're able to obtain a far better CapEx per ton compared to land-based solution mainly built in the Middle East and in the U.S. We still expect new FLNG orders to have a CapEx of around $600 per ton fully delivered versus approximately double that price for greenfield developments in the U.S. The last cost component of gas production is the shipping distance to end users from where the gas is produced to where it will be consumed. Subject to which field we are discussing, we see significant shipping advantages -- shipping distance advantages versus the key global exporter being the U.S. Hence, if you have a business with three cost drivers, the cost of gas, the cost of liquefaction, and the shipping distance and you're cheaper on all three, we believe you have a good and sustainable business over time. Turning to Slide 14 and elaborating on this point. When we entered this year, the global LNG production was around 430 million tons, where around 23% or 100 million tons is supplied by the U.S. Global LNG supply is set to grow by around 40% between now and 2030. However, the absolute majority of that volume is expected to come out of the U.S., increasing its global market share from 23% to 37% of the global market. This is interesting because the U.S. is already the marginal producer of LNG globally. Hence, if you more than double the volume, you increase domestic demand for natural gas on the back of Trump's policies on America First and increased data center build-out in the U.S., we see upward pressure on Henry Hub. Cost inflation of land-based liquefaction plants are suggesting increased tolling fees and shipping distances are obviously unchanged. However, if 2 of the 3 cost drivers of the incremental producer of LNG globally see upward pressure, we see that as a strong support to international LNG prices. And if you then can obtain cheaper source gas, cheaper liquefaction and shorter shipping distance, we believe we have significant upside potential in our commodity exposure in our Southern Energy contracts. Turning to Page 15 and a slide first introduced by our Chairman on our Q1 call. How big is the FLNG opportunity? Well, the first ever FLNG to be delivered was Golar's Hilli delivered in 2018. The industry was skeptic to introduce complex liquefaction technology in a maritime environment. Since that time, we now have seven vessels on the water, four newbuildings under construction, and at least another five vessels planned by the same group of existing owners. We believe this trajectory is similar to what we saw happen in the FPSO industry in the 1980s. The first FPSO globally was delivered in '85. We believe we're now in sort of the early 90s compared to the FPSO industry. And the FPSO industry now stands at more than 250 units. Basically, the cost competitiveness, flexibility, and time of construction is what is increasingly driving recognition of FLNG technology and build-out of what we think is an industry in strong growth. This is why we feel very comfortable with continuing to add units and we'll continue to do so for as long as we are the only service provider and can secure economics similar to those secured on our existing contracts. I'll now hand the call over to Eduardo to take us through the group results and key developments on the financial highlights during the quarter.

Thank you, Karl, and good morning, everyone. I'm pleased to provide an overview of Golar's financial performance for the second quarter of 2025. Moving to Slide 17. Let's review some of the key financial highlights of this quarter. We achieved total operating revenues of $75 million with FLNG tariffs reaching $82 million in the quarter, a significant increase from the previous quarter as a result of Gimi COD, which took place on June 12. Going forward, we will always refer to FLNG tariffs to illustrate the total revenues generated from FLNG, Hilli and Gimi, including realized gains from TTF and Brent-linked fees as well as all revenues under the Gimi contract. Total EBITDA reached $49 million in Q2, positively impacted by Gimi COD. Total EBITDA for the last 12 months ended in Q2 stood at $208 million. This quarter, we report a net income of $31 million. This figure is inclusive of certain non-cash items such as adjustments in the value of embedded TTF and Brent derivatives within the Hilli contract as well as changes in interest rate swaps, but also a $30 million gain recognized on day 1 after the start-up of Gimi's contract. We deemed the Gimi contract as a sales-type lease. And now that we have reached COD, we derecognized the capitalized amounts under assets under development in our balance sheet to a net investment as a sales type lease. Please refer to the appendix of this presentation for a detailed walk-through of the accounting treatment of Gimi contract. In June, we raised $575 million of convertible bonds and repurchased 2.5 million shares for a total consideration of just under $103 million. Following that, our liquidity position has been significantly strengthened with just under $900 million of cash on hand at the quarter end. I will talk more about this transaction in the next slides. Further to that, our net debt position at the end of the quarter stood at just under $1.2 billion. Lastly, we're pleased to declare a dividend of $0.25 per share this quarter with a record date of August 26 and payment scheduled for around September 4. Now moving to Slide 18. Here, we can see the breakdown of our Q2 EBITDA and the contributions coming from Hilli and Gimi, resulting in total FLNG tariffs for the quarter of $82 million. Hilli revenues are split between Base, Brent, and TTF linked fees, as you can see on the slide, while Gimi revenues going forward will be recognized between a capital element as well as an operating element, as you can see on the graph below. Total EBITDA from FLNG operations alone was $57 million after certain project development expenses associated with our growth projects, which Karl explained before. When we consider our corporate costs and other overhead expenses, our total reported EBITDA in the quarter ended at $49 million. Now moving to Slide 19. In June, we raised $575 million of convertible bonds. As part of this transaction, we also bought back 2.5 million shares for just under $103 million at an average share price of $41.09 per share. The notes were priced with a fixed coupon of 2.75% and a 40% premium. One important point to note is the effect of the buyback. When considering them, the notes are net dilutive to our share count prior to the notes offering only when our share price exceeds $76.7 at maturity in December 2030 before adjusting any dividends paid in the period. This represents a premium of close to 87% above the reference price of $41.9. After the share repurchase, we received total net proceeds of approximately $464 million, giving us extra flexibility to fund our growth developments with a fourth or a fifth unit, as explained by Karl. Turning now to Slide 20. Looking back during the past 4.5 years, we can see our historical record of shareholder returns. I wanted to highlight a few points. We did a total of $787 million of dividends, share, and asset buybacks over this period versus an operating cash flow after debt service of around $500 million in the same period. During that time, we bought back over 9.3 million shares at an average price of $125, bringing the total share count to 102.3 million shares at the end of Q2. We have also spent over $560 million in dividends and asset buybacks during the same period. Going forward, we plan to use our liquidity release from debt financing proceeds to be allocated to fund accretive FLNG growth while using our free cash flow generation to significantly increase our shareholder returns. So let's take a closer look at that on Slide 21. When looking at the time when our three vessels are fully delivered in operation and in operation, we expect our fully contracted EBITDA to grow by more than 4x our current last 12 months figure. This huge increase is even before further commodity upside coming from Hilli, Mark II, and our stake in SESA. As you can see on the slide, in 2028, when Mark II comes online, our contracted free cash flow generation could be in excess of $600 million or approximately $6 per share before further commodity upside. Further free cash flow generation under the SESA charters can be estimated at approximately $100 million per year or $1 per share for every dollar per million BTU increase in LNG FOB prices above $8. On a fully delivered basis, our estimated net debt at that point in time would be at 3.4x to adjusted EBITDA, all fully backed by long-term 20-year charters. This will provide us with ample financing capacity to support even more additional FLNG growth units. Now moving to Slide 22, and as explained by Karl before, following the confirmation of our contracts in Argentina with the FIDs of Hilli and Mark II, we can now see the full breakdown of our total firm EBITDA backlog of more than $17 billion over the next 20 years. One point to highlight is that this number is still before further commodity upside and further inflation adjustments that will happen over time. Moving to the next slide, we can see how changes to LNG prices could positively affect our contract with SESA. We have a limited downside with a very significant upside. For example, assuming FOB LNG prices of $10 per million BTU, our estimated EBITDA could be in excess of $1 billion per year, as you can see on the graph. This number could even grow further. So for example, $15 per million BTU, this number would grow to $1.5 billion of annual EBITDA. Just to illustrate the potential scenario and as a reference, if all of our units were in operation in 2022 and assuming average LNG prices during that time, we would be earning close to $3.5 billion on that particular year. This really shows the scale and significant upside potential of the embedded kickers in these contracts. That now concludes my update. I'll hand the call back over to Karl.

Thank you, Eduardo. Moving to the final section of the prepared remarks and the summary. On Slide 25, we show our key '25 milestones and the focus for the remainder of the year. This year, we have obtained the Gimi commercial operations state, and we're now offloading cargo #8. We've achieved FID and conclusion of all conditions precedent for the 20-year charter of Southern Energy for the Hilli following her existing contract in Cameroon. We've also obtained a 20-year charter and the FID for the Mark II FLNG. As Eduardo has explained in detail, we issued $575 million of convertible bonds and repurchased 2.3% of our company. And we have sold the remaining non-core assets, including our shareholding in Avenir Shipping and our last LNG carrier, the Golar Arctic. For the remainder of the year, we plan on optimizing our FLNG financing on the back of the 20-year charters, conclude the regulatory conditions precedent on the 20-year contract for the Mark II and add further FLNG growth units as explained in detail. Turning to Slide 26 to summarize some of the key developments following a very strong quarter. This quarter, we've added $13.7 billion of EBITDA backlog, now giving our total backlog at around $17 billion. Our adjusted -- fully delivered adjusted EBITDA is set to grow 4x versus the last 12 months EBITDA before commodity upside. This provides a pathway to increase our dividend distributions to shareholders by more than -- or significantly more than 4x as fully delivered adjusted EBITDA is far greater than the debt service increase. Our balance sheet with a fully delivered net debt to EBITDA standing at 3.4x provides capacity to fund additional FLNG growth while maintaining an attractive and increasing free cash flow to equity. Our focus is now on ordering an accretive fourth FLNG unit, and that's what we will turn our attention to. That concludes the prepared remarks, and we would like to turn the call over to the operator for any questions.

[Operator Instructions] We are now going to proceed with our first question. And the questions come from the line of Chris Robertson from Deutsche Bank.

Karl, I just wanted to touch based on the new board members that you mentioned, the three. Just what types of skills or networks and connections that they bring to the table for Golar and if they are driving any commercial discussions?

Yes. Chris, sure. So Mi Hong Yoon is effectively a replacement of Georgina Sousa and as our Bermuda representative. She comes with a strong accounting background and knows the company well. Then Thorleif Egeli departed and to some extent, can be seen as replaced by Stephen Schaefer. Stephen is the Chairman of Talen Energy in the U.S. He's got extensive financing background and has worked his entire career in large energy infrastructure assets, including natural gas, and gas-fired power plants. So we expect him to add significant capacities, both on financing and other industry connections. Lastly, we have the introduction of Benoît. Benoît is the former CEO of Perenco, and Perenco is the single largest shareholder of Golar, owning just shy of 10% and are also the charterer of the Hilli today in Cameroon. And Perenco is also one of the world's largest independent producers of oil and gas, and they have several fields around the world that could see FLNG deployment. So we expect Benoît's background to be highly relevant for us, in particular, when it comes to identification of attractive gas fields and upstream know-how.

My second question, I think you mentioned earlier as it relates to the Gimi that there's contemplation around maybe a second asset there. As it relates to discussions, let's say, a Mark III was contemplated that was large enough to encompass any kind of growth or expansion of GTA. Could you then swap out the Gimi and have that be open for employment elsewhere? Or would that be in addition to?

I think right now, all options are being evaluated. If a Mark III would be introduced to the project, it could either swap out the Gimi or be in addition to. That depends on the rest of the infrastructure. The way we understand it is that it's a relatively low cost to double the throughput capacity of the FPSO. Hence, it should be highly accretive to double the capacity from today's 2.5 to 5 MTPA. Arguably, one of the simpler ways of doing that is to order Mark III and that we take -- we swap it out with Gimi. Subject to final commercial arrangements, that's a deal that we would welcome because we see very strong demand for that type of size elsewhere. So that's absolutely not something we would rule out, but it could also be in addition to, but then there's other constraints on the upstream that needs to be unlocked.

We are now going to proceed with our next question. And the next question come from the line of Frode Morkedal from Clarksons Securities.

I just wanted to pick your brain on valuation. Of course, you have a huge backlog. And I feel like as an analyst, a lot of the discussion I have with investors just end up in talking about the valuation of that backlog, right? So particularly what's the appropriate discount rate, which, of course, is down to assessing the risk, right? So I just wanted to hear what kind of framework you think is useful when you value that backlog?

Sure. So in terms of understanding the valuation, which I guess is the question in the backlog here, we obviously have a fixed set of EBITDA, which comprise of Hilli, Mark II, and Gimi. And if you combine them all the net of G&A, it's around $800 million per year before CPI adjustment and commodity upside. The right discount factor on those, we will leave it to you to decide. But what we historically have seen that some of our peers, including Cheniere and Venture Global and that type of companies trade at, we see that probably north of 10x, but the market can decide what the right multiple factor is. Where we think the biggest sort of misunderstanding or arguably mispricing is understanding the upside of our commodity exposure. We believe the uniqueness of that exposure is that we have fixed the breakeven of which we get a profit share from at around $8 per MMBtu for 20 years. If you look at the developments of where Henry Hub is likely to develop on the back of significant increase for Henry Hub gas, both domestically and for exports, as explained earlier on, we see upward pressure on Henry Hub. And in line with what Cheniere, Venture Global, and other U.S. liquefaction providers have explained on each call they have is that the tolling fees for U.S. exports is on its way up. And according to Venture Global, it needs to go north of $4 to justify any exports. If those two are true, the $8 fixed profit share mechanism we have, we think is unique in the market. You can't obtain it anywhere else, and we think it's worth a lot to have that 20 years out. In addition, we also don't think that the market price our platform as the only proven service provider of FLNG in the world and certainly not growth. And I don't think we want to enter into a conversation today of exactly what we think it's worth, but we do think as much as if we bought 2.3% of the company at $41.09 as late as June. So I think that speaks to where we think it is. And I think over the last 4.5 years, we bought back 9.3 million shares, and we have bought back assets where we've been able to buy them back more accretively than buying our own share, which is effectively buying back Hilli from NFE and Hilli from minority interest from Black & Veatch and Seatrium. So I think overall, we are putting money where our mouth is, and we continue to do so as late as June.

Yes. That's great. Good color. I guess what you're saying that, of course, is -- you think this stock has higher intrinsic value than the current market price, right? So I guess -- and you've been buying back stock, as you said. But just like generally speaking, how important is it for you to close that gap, so to speak? Would you just prefer to let the cash flow speaks for themselves and market prices adjust over time? Or are you actually consider actively boosting that value further, right?

So speaking as management, what we can do is to run the company to the best of our ability. That's what we try to do every single day getting into this office. It is also our job to see that, that value is reflected in the stock market or in the value of the business. As our Chairman who joined our quarterly call for the first time in a very long time last quarter alluded to is that if the market fails to recognize that value, the Board will seek other alternatives to crystallize the value. I'll let him stand for that statement, but I saw him reconfirming those intentions in newspapers as late as July this year. So I think as management, our work is to do the -- run the business to the best of our ability. And if there are price discrepancies between private versus public money, we will look to identify it. But for us, the key effort on a daily basis is to run the business on the best of our ability. And we think there is a lot of attractive growth that can be executed. I think we showed that this latest ordering our third unit in September and fixing it in May, and we intend to do a similar type of structure for unit #4.

We are now going to proceed with our next, and the next questions come from the line of Alexander Bidwell from Webber Research & Advisory.

Could you provide a little bit more color on some of the commercial prospects for a fourth unit? What sort of demand are you seeing around contract start-up? And are you seeing any gravitation towards a particular FLNG spec?

Yes. So it's down to -- when it comes to start-up, I think the construction time that we have 3 years or more or less 3 years for Mark I and II is sufficient to provide the required upstream infrastructure. So we do not see the upstream infrastructure being the gating item. Most of the opportunities we are discussing in West Africa is likely Mark I or II. And then some of the projects we are discussing in South America and in the Middle East could be a Mark II, so that's sort of broadly where we see it. Naturally, just given size and throughput, there are more commercial opportunities for smaller sizes. But as long as you have a few people in the running on the larger size, you can still drive competitive tension, and that's what we're focused on. In terms of number, the smaller is where there's more and larger, you still need to create -- have enough for competitive attention.

All right. Appreciate the color. And then turning over to long lead items. You'd mentioned that some of the time lines have been increasing. Just curious, is there any cross compatibility between long lead items for the Mark I, II, and III? Could you say order certain long lead items and then, I guess, change which direction you go in terms of asset?

That's a very good question, and the answer is yes. It is compatible. So basically, we use Black & Veatch topside technology for all the three designs. It's just a matter of how much of the equipment you put on. The key long lead items that has been dragged out in time is gas turbines on the back of very strong demand from the aviation industry and data centers. So those are the ones we're likely to put the slot reservation on first, and that's interchangeable between the designs. It's just the number of turbines you need is different. So the whole idea is to do the slot reservation and decide on the number of turbines later.

We are now going to proceed with our next question. And the questions come from the line of Sherif Elmaghrabi from BTIG.

First, I just want to follow up on Chris' question at the beginning. Last week, one of the GTA partners mentioned the possibility of debottlenecking Gimi to increase production. Can you shed some light on what that would entail?

Yes. So as I explained in the prepared remarks, we're currently in the appraisal period. We're fine-tuning the operations of the FLNG. And if you want to get very technical, it has to do with the air inlet chilling. It has to do with the turbo expander that we can tweak and some other of the equipment that we are currently fine-tuning to boost production. That's really the key technical items. There are also certain improvements that can be extracted from the interaction between the FPSO and the FLNG. So throughput of a liquefaction plant is highly linked to ambient temperature and gas quality. So if you can through air inlet chillers, impact the ambient temperatures and through interaction between the FPSO and the FLNG, tweak the gas composition, you can boost throughput.

That's good color. And then -- for the range between a Mark I and a Mark III, I appreciate you giving us kind of a dollar -- rough dollar per ton given pricing is still a bit of an unknown. But are there economies of scale associated with going for a large -- like a Mark III versus a Mark I?

In terms of CapEx per ton?

Right, yes, in terms of CapEx per ton.

CapEx per ton, I would actually say less so because you save a lot of time and money starting with the conversion unit versus building from scratch. However, in terms of operating costs, there is significant economies of scale because you don't need twice as many people to run a Mark III as a Mark I, you probably need the same. So on a unit economic point of view, it's more efficient, both in terms of OpEx and demerge and offloading.

We are now going to proceed with our next question. And the questions come from the line of Liam Burke from B. Riley Securities.

Karl, I mean, you stepped up the Mark I to the Mark IV -- III rather, and you essentially doubled your capacity. Do you anticipate further capacity growth on next-generation FLNGs? Or is this just a function of the offshore reserves?

We do not -- no, you're right. We don't see a very -- we don't expect to go for like a 10 MTPA unit, if that's the question, because then you start to get into far too few fields. One thing is that you need large reserves. The other thing you need is be able to service the throughput. So I'm not an upstream expert, but some wells, even if you have a massive reserve, if you empty it too quickly, you can have pressure losses and you can destroy the recoverability of the entire reserve. So it becomes very sensitive if you start to go beyond this size.

Okay. At the higher capacity levels, I mean, the FLNGs generally -- the original ones generated mid-to-high teens returns on invested capital. When you get to twice the capacity, how do you see those returns increasing?

If you look at what we did in Argentina with the Mark II, we're able to do 20-year contracts at around -- that one is 5.5x CapEx to EBITDA before inflation adjustment and before commodity upside, where OpEx and asset level tax is passed through to the offtakers. So I think if we are able to do 20-year business at around 5x with commodity exposure and CPI adjustments, we think that's a good business for 20-year contracts. And that's sort of similar to what we would target on any new project.

We are now going to take our next question. And the questions come from the line of Craig Shere from Tuohy Brothers Investment Research.

So it seems your focus is on projects with commodity kickers. So in that respect, ultimately, is Gimi something that you might contemplate divesting? And to the degree BP needs to double the size of that project, would they be amenable to the next phase having similar commodity kickers as what you're looking at in Argentina and elsewhere?

To the first part of the question, we've explained this on some previous calls, but the Gimi contract is -- or was very important to prove the concept. Hilli was the first FLNG to ever be delivered. It's well known in the maritime industry that BP has the highest operational standards in the industry. And if a Golar FLNG can qualify as sufficiently safe and reliable to operate for BP, then it's a proof of concept. So for us, the BP contract is attractive in itself that it's mainly a proof of concept, and it has opened the doors for significant further FLNG deployments. To sort of put some more color into that, Pan American Energy, which is the largest counterpart in SESA is owned 50% by BP. So the fact that they have seen what both our assets and the company can deliver in Mauritania and Senegal has given them confidence to continue to expand the working relationship. We have no current plans of divesting that asset. We think the way the structure of the contract is, you can obviously leverage it differently and still get very attractive equity returns. As for growth on the GTA, that's subject to commercial negotiation, but we do expect any incremental liquefaction capacity to come at a higher tariff than that of Gimi.

And not only a higher tariff, but a commodity link like everything else you're focused on?

Yes. So when we do contracts, it's always a trade-off between the fixed component versus commodity. So if people don't want commodity, you need to pay up on the fixed. It's a trade-off.

Got you. Okay. And last question, I just want to get clear on the order of things that you've laid out. So you're spending balance sheet cash on long lead items ahead of the next FID and contracting, mostly you mentioned about the turbines. But you won't FID a fourth FLNG shipyard commitment until you have a contract in hand. But after you have that, you'll go on balance sheet for a fifth FLNG project. Do I have that right?

Yes. So we do long leads now. Then when we get sufficiently comfortable, we order unit #4. Once unit #4 is locked in a 20-year or a long-term contract, we then go and order unit #5. But we will constantly ensure that we have ample financial flexibility to do it. We will never ever put this company in a growth position that challenges the balance sheet.

We have no further questions at this time. I will now hand back to you, Mr. Karl Fredrik Staubo, for closing remarks.

That concludes Golar's Q2 earnings call. Thank you all for dialing in, and have a good day.