January 31, 2025
Things I Learned This Week in the Business
Very Little. Upon reflection, other than the continued meltdown of MSNBC and others about everything Trump says or does, not much else percolated to the top. Crude oil is at $73, with the 2025 futures strip trending down to $68 by the end of the year. Sure, sure, the strip is always wrong, but it is all companies and investors can use to judge cash flows and valuations. “Bob thinks it will be $80” has rarely gotten an $80 price into many budgets. It was $76 last year. So, this generally means for oil drilling and activity that capital will be less available, capital spending will be less, revenues will be lower and margins will erode, at least slightly. No disaster. Only down slightly, like low single digits, right? That is both the consensus and the guidance. All these price estimates could be well off… in either direction. Trump wants cheap oil; we don’t. China doesn’t want as much as we thought it would. There is shut-in oil that could produce between 1 and 3 million barrels a day. Bulls will say 1, bears will say 3. Either way, it is a lot of oil that can come onto the market, and we haven’t even discussed Iran or Russia. So, for now, for the U.S., assume oil prices will be down ~8%, which would depress activity somewhere in that neighborhood.
Natural Gas is Throwing a Party. So, last year natural gas averaged $2.21 MMBtu, down from $2.51 in 2023, and was the lowest average annual price ever recorded when adjusted for inflation. Great for consumers, not so good for natural gas producers. Drilling activity was down 20% last year. But welcome to 2025, where the current futures strip is $3.60. That shows to be up about 63% in 2025. And if you can’t make money at $3.60 gas, maybe you should change businesses. With Mr. Trump back in the White House, we would expect more pipeline approvals and more transmission capacity out of the Marcellus and Utica. Those two basins should see a big resurgence over the next couple of years, and the Haynesville should be the first to move in terms of drilling activity to boost production, even after a 50% decline in rig count over the last year. Midstream should continue to do well with both higher volumes and higher prices, but before getting too bullish, we have a great deal of natural gas and the proven capability of over-producing it.
The U.S. rig count and frac spread count indexed to the beginning of 2021. We are up over 50% since Covid, but we have been gently rolling over for the past two years.
ChatSmith. With the current fervor over AI, I decided to bring out the big guns and ask the Wizard. I asked, “What is the outlook for international oil and gas activity in 2025 and the price assumptions behind it?” I have no fear of being replaced. AI’s response below:
The outlook for international oil and gas activity in 2025 will depend on various factors, including geopolitical developments, technological advancements, environmental policies and market dynamics. Here are some key trends and assumptions that may shape the industry:
Energy Transition: The push for renewable energy and sustainability will likely continue to impact traditional oil and gas operations. Companies may shift focus towards cleaner technologies and diversify their energy portfolios.
Demand Recovery: After fluctuations due to the pandemic, oil and gas demand could stabilize, driven by economic recovery in emerging markets and ongoing demand in transportation and industrial sectors.
OPEC+ Decisions: Production decisions made by OPEC+ will significantly influence oil prices. Continued cooperation or potential disagreements within the group can lead to market volatility.
Technological Innovations: Advancements in extraction and production technology may reduce costs and increase efficiency, impacting supply levels and prices.
Supply Chain Challenges: Ongoing supply chain disruptions, whether due to geopolitical tensions or pandemic-related issues, may affect production levels, leading to price fluctuations.
Regulatory Policies: Stricter environmental regulations and carbon pricing initiatives in various countries could alter production strategies and influence investment trends in the sector.
Price Assumptions
Crude Oil Prices: Prices may remain volatile, fluctuating between $60 to $100 per barrel depending on supply-demand balance, geopolitical tensions and economic recovery.
Natural Gas Prices: Likely to experience upward pressure due to increased demand for LNG, with prices possibly ranging from $3 to $6 per MMBtu.
Investment: May decline slightly as more capital is redirected towards renewables and green technologies, although traditional oil and gas projects will still attract significant investment.
Overall, the outlook for oil and gas in 2025 suggests a transitional phase where the industry adapts to changing market conditions and increased scrutiny on environmental impacts.
Cross-Pollination. Engine No. 1, the activist investment group that has several seats on Exxon’s board, and Chevron are partnering in a new company to develop scalable, reliable power solutions for data centers running on U.S. natural gas. The JV is working with GE Vernova to build a 4-gigawatt plant by the end of 2027, enough to power 3–3.5 million U.S. homes, according to their documentation. “The company will directly address the need for affordable, reliable energy to meet the significant demand for electricity to power U.S. data centers.” A Lawrence Berkeley National Laboratory report finds that data centers consumed about 4.4% of total U.S. electricity in 2023 and are expected to consume about 6.7%–12% of total U.S. electricity by 2028.
Pressure What? Liberty Energy, formerly an oilfield services, pressure pumping and completion services business, is most definitely transitioning, now being identified as something different. Seriously, I take nothing away from Liberty. In addition to supplying the new Secretary of Energy, last year the company had a 17% adjusted pretax ROCE and a $21 cash return on invested capital. Not only was it good, it was highlighted. The company distributed $175 million with the repurchase of 3.8% of shares and quarterly cash dividends. The company clearly identified two strategic priorities. Focusing on continuing technological innovation within completions and expanding into the power generation market. It is doing this through a couple of acquisitions and a recent collaboration with DC Grid to expand into commercial fleets, electric vehicle hubs and data centers. One analyst report mentioned the power expansion before addressing pressure pumping at all. The Q&A demonstrated it as well. Remember, a fellow pressure pumping company, ProPetro, announced a new power division and pushed the stock up almost 15% on the announcement. We have been asking companies to think outside the box and find “transition” businesses to combine with their existing knowledge and expertise. Now we are seeing it done. Excellent.
Sand Who? Atlas has added another leg to its stool by purchasing a distributed power company to complement its sand business. It is understandable. Power is hot. We need more power, and the demand for power is going up and accelerating for the foreseeable future. I remember when all the OFS companies wanted to be classified as “industrial” to improve valuations. Now it is power. Atlas is buying Moser Energy Systems for $220 million, closing this quarter. The acquisition will combine Atlas's completion platform with Moser's distributed power platform, and who doesn’t like distributed power?!? It makes sense, as it did with ProPetro and Liberty. It does call into question future expectations of completions businesses. Pressure pumping is oversupplied, and so is sand, but we need more power. Now the only question is, who is next?
It Never Ends. First, it wasn’t climate change. Two years of wet weather followed by one year of drought is not considered climate change, but weather. And the idea that climate change causes forest fires makes about as much sense as saying they cause volcanoes. But that doesn’t stop the state of California. A senator has proposed SV222. The Affordable Insurance and Climate Recovery Act. Basically, the victims of the California wildfires are saying it is the fault of the oil companies, and so they are going to sue the oil and gas companies that operate in California for the cost of recovery from the wildfires. All the oil and gas companies would be liable for damage caused by climate change, specifically, climate change-related disasters in California under this legislation. This could only make sense in California. The senator said their bill will provide a major infusion of financial support to stabilize the insurance market and contain the high cost of insurance for Californians. So, all the insurance companies have left California high and dry because their requirements and bureaucracy are now somehow the fault of a company making gasoline in the formulation required by the government. It’s damp and wet outside, which must be the result of global warming! Exxon, please send me a check. Want to take the under on that?
Two Steps Forward… The largest undeveloped oil and gas field in the UK is Rosebank, and since winning government approval in 2023, the project had moved ahead. Until now… A UK court has ruled that the approval is unlawful, and the project is stopped because the companies developing the field had not taken into account the carbon emissions created by using the oil and gas. No stopping the users, who burn it and cause the emission release. Just the ones who are trying to make their business as clean as possible while providing one of the most essential materials of our society. One activist said, “This means that Rosebank cannot go ahead without accounting for its enormous climate harm.” The UK produces 1.2% of global emissions and already has some of the highest utility costs in Europe.
Headlines.
“Panama Launches Charm Offensive in First Bilateral Meeting With Trump Administration.”
“No Plan B for Broken German Economy.”
Go Big. Oracle and Softbank announced a joint venture, “Stargate,” to invest up to $500 billion for building out AI infrastructure, and Microsoft plans to spend $80 billion on AI data centers in this year alone. Meta will spend about $65 million on AI projects this year and build a data center “so large that it would cover a significant part of Manhattan.”
Money Money Money. Enverus Intelligence Research published its summary of 2024 upstream M&A activity, reporting $105 billion in U.S. upstream deals closed in 2024, the third highest total tracked by Enverus, only trailing behind a record-setting $192 billion in 2023 and just under the $108 billion booked in 2014.
PPHB – U.S. Energy Market Update Highlights:
Commodity Prices: WTI crude oil is currently $72.73 per barrel down ~2.6% week-over-week) and natural gas is $3.05 per MMBtu (down ~11.7% week-over-week).
Crude Oil Production: U.S. crude oil production is currently ~13.2 MM BOPD (up ~1.8% year-over-year).
Crude Oil Inventories: U.S. crude oil inventories decreased by ~1.0 million barrels week-over-week vs. an estimated decrease of ~2.2 million barrels.
Frac Spread Count: There are currently 186 frac spreads operating in the U.S. (a decrease of 5 spreads week-over-week).
Onshore Drilling Rig Count: There are currently 560 drilling rigs operating in the U.S. (a decrease of 4 rigs week-over-week).
Now the Winner Can Emerge. The U.S. Department of Transportation’s Pipeline and Hazardous Materials Safety Administration finalized a rule to enhance the safety of natural gas pipelines through “updated leak detection and repair requirements.” In government numbers, it is expected to have up to $1.5 billion in annual net benefits to the public and eliminate up to 500,000 metric tons of methane emissions from approximately 2.8 million miles of gas transmission, distribution and gathering pipeline facilities, 398 underground natural gas storage facilities and 173 liquefied natural gas facilities across the U.S.
There are a number of competing technologies and methods in this sector, with none breaking out as dominant, yet. The rule requires pipeline operators to establish advanced leak detection programs aimed at detecting and repairing all gas leaks by:
Increasing the frequency of required surveys and requiring the use of commercially available advanced leak detection technology, such as aerial or vehicle surveys, handheld detection devices and continuous monitoring systems.
Reducing the volume of gas released due to unintentional emissions such as leaks and equipment failures and revising the reporting minimum threshold to detect smaller leaks sooner.
Minimizing releases caused by equipment venting or blowdowns, associated with pipeline maintenance, repair and construction and encouraging operators to consider cost-effective equipment that can capture methane and other gases for later use.
Options. So, Trump's executive order says that all federal workers have to come back to the office five days a week. The videos on YouTube of weeping workers were amazing. It's estimated that 30% of the federal workforce would rather quit than come back to work full-time. Others have said this is a brilliant way to cut government employment and slim down government spending. It's probably both. But now it's been sweetened, and I would even call this an incentive. If you refuse to return to the office, you will be let go, but the government will pay you for the next eight months, and you have until February 6 to decide. Either come back to work right now or quit and get eight months’ pay. And I misspoke; they don't get fired. They were given a choice to return full-time or resign. White House officials said that the office requirement will total 5%–10% of the cost incurred by federal employees quitting; as noted above, that number would be considerably higher. The Senator from Virginia is saying that if you accept that offer and resign, “he'll stiff you just like he stiffs the contractors.” That was in quotes. The White House is saying that this “deferred resignation” will be available for all full-time federal employees, except military personnel, post office employees and positions related to immigration enforcement and national security. All you have to do is send an email to human resources and put “Resign” in the subject line. They do make it easy.
It Won’t Start. Following the news of the last several weeks of Mercedes and Volkswagen taking multi-billion-dollar losses in the quarter due to the slowdowns in China, General Motors has joined the party. The $4 billion restructuring of its Chinese operation resulted in the company losing almost $3 billion in the quarter. There are stories this week of small towns in Germany having dramatic problems because of the layoffs by companies like Volkswagen. As we've written before, and we mention in a headline this week, Germany, the world's third largest manufacturer, is having significant economic problems, and is in the second year of a recession, in part driven by high energy costs due to the years of reliance on cheap Russian gas.
We Didn’t See This Coming. The recent market volatility has been primarily due to the release of the Chinese AI model called DeepSeek. Nvidia lost $600 billion of its market cap in one day. The tech sector, which has driven the NASDAQ to record after record, appears to be surprised, and investors are very worried that the perceived monopolies are being disrupted. Just like in energy, the lowest-cost producers always seem to prevail. I have the sneaking suspicion that these AI chips are going to be priced below those of Nvidia, even if slightly. All of this leads to the point that there appears to be another player in the AI universe that will take, at this point, an undetermined amount of market share. Why do we care? It seems that the DeepSeek chips use significantly less power than NVIDIA and other U.S. developers. One of the things driving our industry has been the demand for power for the data centers that drive AI. If any significant share of the market consists of chips that use far less power, less power will be needed. Seems simple enough. It also means that the forward forecast for natural gas-powered demand may be more optimistic than reality.
Taking Fair Advantage. Sustainable financing. When the sustainable lending craze first hit, the money was almost free. If you were “green,” we would fund or support you. Rates moved up, and everyone became more sustainable. And rightly so, some companies are smart enough to describe their business in the best possible light, especially if there is cheap money involved. So, congratulations to Select. Select Water Solutions is “a leading provider of sustainable water and technology solutions to the energy industry.” Select closed a new five-year senior secured sustainability-linked credit facility. CEO John Schmitz said that the five-year term of the facility reinforces financial stability and enhances flexibility to “expand our large-scale water infrastructure networks, advance our industry-leading water recycling and automation technologies, and drive sustainable and safe solutions.”
Okay, a Primer. There are chips and algorithms. The chips have improved significantly, capable of parallel computing (doing multiple tasks at once) and they are able to process tremendous volumes of data. They are pieces of equipment. The algorithms written to optimize the ability of the chip are incredibly complex. And because of that, just as there are different opinions on how to solve a problem, the programming approaches and methods are different. When we talk about AI doing or saying something, that is a collection of algorithms, making up a “model” that runs on a computer chip. Nvidia makes the best AI-optimized chips. This collection of algorithms that form the model is what we used to just call “software.” These collections of algorithms, running on an AI-designed chips, are models and they require massive data inputs. This is called “training.” Just as in any geoscience model, the more data the better, and these algorithms collect, calculate and learn. Some types of AI algorithms can be taught to learn on their own, access new data and refine their “intelligence.” The range of situations that can be impacted or affected, from policy to returns, is stunning.
Supervised learning algorithms. These work by taking in clearly labeled data while being trained and using that data to learn and grow. They use the labeled data to predict outcomes for other data. Categories of supervised learning include:
Decision Tree: One of the most common supervised learning algorithms, decision trees get their name because of their tree-like structure (even though the tree is inverted). The “roots” of the tree are the training datasets, and they lead to specific nodes which denote a test attribute. Nodes often lead to other nodes, and a node that doesn’t lead onward is called a “leaf.” Decision trees classify all the data into decision nodes. They use a selection criteria called attribute selection measures (“ASM”), which takes into account various measures (some examples would be entropy, gain ratio, information gain, etc.).
Random Forest: The random forest algorithm is actually a broad collection of different decision trees, leading to its name. The random forest builds different decision trees and connects them to generate more accurate results.
Support Vector Machines: The support vector machine (“SVM”) algorithm works by plotting each piece of data on a chart (in “N” dimensional space, where N = the number of datapoints). Then, the algorithm classifies the datapoints by finding the hyper-place that separates each class. There can be more than one hyper-place.
Naive Bayes: The reason this algorithm is called “Naive Bayes” is that it’s based on Bayes’ Theorem and it also relies heavily on a large assumption; that the presence of one particular feature is unrelated to the presence of other features in the same class. That major assumption is the “naive” aspect of the name.
Linear regression: A supervised learning AI algorithm used for regression modeling. It’s mostly used for discovering the relationship between data points, predictions and forecasting. Much like SVM, it works by plotting pieces of data on a chart with the X-axis as the independent variable and the Y-axis as the dependent variable. The data points are then plotted out in a linear fashion to determine their relationship and forecast possible future data.
Logistic regression: A logistic regression algorithm usually uses a binary value, 0 or 1, to estimate values from a set of independent variables. The output of logistic regressions is either 1 or 0, yes or no. An example of this would be a spam filter in email. The filter uses logistic regression to mark whether an incoming email is spam (0) or not (1).
Unsupervised Learning Algorithms: Unsupervised learning algorithms are given data that isn’t labeled. Unsupervised learning algorithms use unlabeled data to create models and evaluate the relationships between different data points to give more insight to the data.
Real Open AI. A Chinese AI model was released, that, in benchmark tests, did as well as U.S. models in reasoning, math and other subjects. But it has two secret weapons. DeepSeek doesn’t use the latest AI-designed chips, instead utilizing chips that are less sophisticated and powerful than those of Nvidia and others. Their AI model was built using novel methods and technologies in training. In a recent test, the DeepSeek model using “regular” chips performed as well as the AI models using Nvidia chips. Ergo, Nvidia stock plummets. But it is a clear negative for all of us in the energy and power business. These “regular” chips use far less energy than the Nvidia chips, driving down power forecasts. As a result of the lower costs, DeepSeek is very price competitive with other model providers. They also released the open-source code for the model so others can write additional code and capabilities. While it won’t matter today or tomorrow, the longer-term power demand forecasts may be optimistic.
Shape of Things to Come. Private equity firm, Blackstone Energy Transition Partners, has agreed to acquire Potomac Energy Center, a 774 megawatt natural gas plant, from Ares Management for $1 billion. It is located in Virginia, and sits near 130+ data centers. That number is expected to grow. The Northern Virginia region currently represents approximately 25% of U.S. data center capacity, but Texas is gaining. Transitioning to power. Smart move.
Damn! $48 billion expansion. Not new construction. An expansion. And after spending all that money, only now do they start producing. One field. A big one I'll grant you, but one field, nonetheless. That accelerates the development of technology. Chevron, the operator of Kazakhstan’s Tengiz oil field said it achieved first oil from its newly built production and processing facilities. At 1 million barrels per day, Tengiz is one of the world's deepest and most complex fields due to the high levels of sulfur and the harsh weather conditions. This is a very expensive game to play.
Any and all comments, arguments and rebuttals are welcome!
In addition to my association with PPHB, I serve on three private company boards. Merit Advisors is a property valuation company and I have long been a fan of optimizing how a business is run, not just the tools we make. Merit is in the business of savings companies’ money, actual cash, by doing a much more in-depth and realistic view of equipment and reserve valuations and I am very impressed with their work. I am also on the advisory board of Preng & Associates, a leading executive search boutique that specializes in all things related to Energy & Power. Nova is a gas compression company run by a very dynamic CEO with a very strong board and ownership.
I serve on the Advisory board of the Energy Workforce & Technology Council (formerly PESA), the National Ocean Industries Association (NOIA), and the Maguire Energy Institute at SMU my alma mater.
jim
214-755-3914 | james.wicklund@pphb.com
Leveraging deep industry knowledge and experience, since its formation in 2003, PPHB has advised on more than 180 transactions exceeding $11 Billion in total value. PPHB advises in mergers & acquisitions, both sell-side and buy-side, raises institutional private equity and debt and offers debt and restructuring advisory services. The firm provides clients with proven investment banking partners, committed to the industry, and committed to success.
