Baker Hughes Sees No Energy Project Slowdown Amid High Rates

Have you noticed how every rate-hike cycle is supposed to freeze the heavy stuff first? The big plants. The multi-year pipelines. The liquefaction trains that take years and a small fortune to stand up. That is the usual script. This time, at least in energy, the script is not landing the way textbooks promised.

I have been watching project chatter for a while, and the mood coming out of the gas world is oddly calm. Not reckless. Calm. The message from Baker Hughes leadership is straightforward: higher borrowing costs have not, so far, choked investment in major energy projects. Offtake agreements still matter more than the latest move in policy rates. Energy demand is not politely waiting for cheaper money.

Why Energy Projects Keep Moving When Money Gets Expensive

Financing is not irrelevant. Anyone who has sat through a project committee meeting knows that. Sponsors still stress the model. Banks still ask ugly questions. Equity still wants a story that survives a bad year. But bankability, as the company framed it, rests first on who has already agreed to buy the molecules and on how durable the demand outlook looks from here.

That is a different hierarchy than the one retail investors often assume. We talk about rates as if they were a master switch. Flip them up and the cranes stop. In practice, a contracted LNG cargo or a multi-year power offtake can outweigh a couple of hundred basis points. I find that both obvious and easy to forget when headlines scream about the cost of capital.

We haven’t seen a slowdown, and the bankability is really based on the offtake agreements that are in place, as well as the outlook of energy demand.

– Industry executive commentary

Population growth does not pause for a central bank meeting. Industrial activity does not either. And now there is a third pillar that did not dominate every energy conversation a decade ago: the electricity appetite of data centers tied to artificial intelligence. That linkage is no longer a side note. It is becoming part of the core demand case.

The Quiet Logic Behind Offtake First, Rates Second

Think of a large energy project as a marriage between geology, steel, and contracts. The geology has to work. The steel has to arrive. The contracts have to survive stress. Rates sit in the third bucket. They change the shape of returns. They rarely erase a project that already has credible buyers locked in for years.

That is why the phrase bankability keeps coming back. It is not poetry. It is a reminder that lenders care about cash that shows up on a schedule. A floating rate can be hedged. A missing customer cannot. In my experience, the market underprices that distinction until a project actually closes.

• Contracted offtake reduces revenue uncertainty more than a modest rate move increases interest expense.
• Long-cycle energy assets are planned on decade views, not on one policy meeting.
• High prices can, paradoxically, pull investment forward rather than freeze it.
• Grid bottlenecks push some buyers toward on-site or behind-the-meter power solutions.

None of that means every proposed plant gets built. Far from it. Weak sponsors still fail. Poor locations still fail. Political risk still fails deals at the last minute. The claim is narrower and more interesting: the rate shock alone has not produced the slowdown many expected in this corner of the industrial economy.

AI Buildouts And The New Hunger For Power

If you work adjacent to markets, you have heard the data-center story so many times it can start to sound like a slogan. Then you look at interconnection queues and local grid constraints and realize the slogan has teeth. Computing clusters do not sip power. They gulp it. Cooling systems gulp water too, which is a separate political headache, but the electricity piece is the one that keeps colliding with gas and generation equipment.

Baker Hughes does not expect that buildout to cool off quickly. The company is adding capacity to meet demand rather than waiting for a pause that may not arrive on schedule. That stance will look either farsighted or stubborn in two years. Right now it matches what utilities, developers, and hyperscale buyers keep repeating in private rooms: the bottleneck is power, not ambition.

In parts of Southeast Asia, the grid cannot always keep up. So operators look at distributed generation and behind-the-meter setups. That is not a romantic energy transition slide. It is a practical workaround. If the wires cannot deliver, you generate closer to the load. Equipment makers that already live in turbines, compression, and related kit sit in a useful spot when that workaround becomes a plan.

I keep coming back to a simple question. If AI demand is even half as durable as the industry claims, who supplies the electrons in the ugly years before new nuclear, new transmission, and new storage catch up? Gas keeps answering that question, whether or not that answer is fashionable.

Natural Gas As Destination Fuel, Not A Temporary Guest

The old talking point called gas a bridge. Bridges are temporary. You cross them and forget them. The newer line from this management team is sharper: gas is a destination fuel in an energy demand decade. That is a values statement as much as a forecast. It will annoy people who want a faster exit from hydrocarbons. It will comfort people who look at baseload math and get nervous.

As you look at natural gas, it’s not a transition fuel; it’s a destination fuel. We’re in an energy demand decade, and gas is central to it.

You do not have to love that framing to take the industrial implication seriously. If gas remains central, then liquefaction, pipelines, compression, and field services stay relevant longer than a neat 2030 slide deck implied. Backlog becomes a kind of vote. The company has pointed to more than $37 billion sitting in backlog, with demand tied to gas infrastructure, data-center power, and LNG.

Backlog is not destiny. Projects slip. Customers delay. Scope changes. Still, a number that large is not a vibe. It is a queue of work that someone already signed. When people ask whether higher rates killed the cycle, that queue is one of the better answers available.

High Prices, Geopolitics, And The Strange Incentive To Build

There is another layer that makes this moment messy. Conflict risk in the Middle East has shaken energy flows and pushed oil back into psychologically loud territory above $100 a barrel at points. That kind of move feeds inflation anxiety and, yes, borrowing-cost anxiety. It also tightens nerves around gas shipping routes, including traffic through the Strait of Hormuz and the reliability of Qatari LNG.

Here is the part that sounds backward until you have watched a few cycles. High prices can stimulate the very investment that later brings supply. Pain today becomes capacity tomorrow. Management called the LNG supply build full steam ahead in that spirit: look past the short term, accept that elevated prices fund the next wave.

Does that mean a glut is guaranteed? The company line is more measured. Prices are expected to stay roughly range-bound, and a coming wave of LNG is not seen as a recipe for a long, crushing surplus. The longer-run estimate tossed into the conversation is blunt: installed LNG capacity may need to approach 900 million tons per annum by 2035 if demand keeps compounding the way bulls expect.

I am not going to pretend that number is sacred. Demand forecasts in energy have a talent for humiliation. But the direction of travel is the useful bit. If you believe the world still needs a lot more flexible molecules, then cancelling projects because rates are annoying looks like the wrong kind of discipline.

What “No Slowdown” Does Not Mean

It does not mean costs are trivial. Steel, labor, vessels, and specialized equipment can still blow up a budget. It does not mean every geography is open for business. Permitting can freeze a file faster than a rate hike. It does not mean oil and gas stocks automatically win from here. Equity prices discount a lot of things that project engineers never see on a Gantt chart.

It also does not mean environmental pressure vanished. Water use around data centers is already a local fight in dry regions. Methane rules keep tightening in some markets. Community opposition can stall a pipeline even when the economics look clean on a spreadsheet. A honest reading has to hold those frictions next to the demand story, not underneath it.

1. Separate contracted projects from hopeful press releases.
2. Watch offtake quality, not just nameplate capacity announcements.
3. Track grid interconnection delays as closely as commodity prices.
4. Treat geopolitical premia as both a risk and a reason some buyers want more diversified supply.
5. Remember that equipment backlog can lead reported earnings by a long stretch.

Perhaps the most interesting aspect is how unemotional the corporate tone has been. No victory lap. No denial that financing matters. Just a stubborn observation: the slowdown has not shown up in the order book the way the rate narrative suggested it would.

How Investors Might Read An Energy Demand Decade

If you hold energy services or infrastructure names, the useful question is not “Are rates high?” The useful question is “Who still has a customer?” Companies leveraged to gas midstream, LNG equipment, and flexible generation sit closer to that customer than a generic oil beta trade.

There is a valuation trap here, though. Markets can pay up for an AI-adjacent story and then punish any quarter that looks merely solid. Energy investors have lived through enough false dawns that skepticism is rational. I have found that the cleaner approach is to map backlog composition. How much is truly tied to LNG? How much is maintenance? How much is speculative capacity that could slip if a single offtaker blinks?

Another angle is regional. Importing countries with weak domestic grids and fast data-center ambitions may keep signing gas deals even when global headlines feel chaotic. Exporting countries with cheap feedstock and existing shipping access still have a structural edge. The map matters more than the average price on a screen.

A simple filter I keep coming back to:
  1. Is there a signed buyer?
  2. Can the power actually reach the load?
  3. Does the project still clear after higher rates?
  4. What happens if prices mean-revert for two years?

If a project fails test four, it was never as robust as the ribbon-cutting photos implied. If it passes all four, higher rates are a bruise, not a verdict.

LNG Supply Growth Without The Cartoon Glut

Every LNG upcycle invites the same cartoon. A wall of supply arrives. Prices collapse. Developers swear they will never overbuild again. Then five years later someone builds anyway. The current corporate view tries to dodge that cartoon by pairing supply growth with a fat demand number out toward 2035.

Is that optimistic? Sure. Asia’s industrial recovery could disappoint. Efficiency gains in computing could shave some power intensity. Policy could lean harder into alternatives. All of that is fair. What is less fair is treating any new train as automatic oversupply. Markets can absorb a lot of LNG when winter is harsh, when coal plants retire, or when a shipping chokepoint gets scary.

Range-bound pricing is the unglamorous middle path. Not a boom that makes every project a lottery ticket. Not a bust that strands half the fleet. Just a band wide enough to keep decent projects alive and narrow enough to punish the sloppy ones. That is a market adults can plan around, even if it makes for dull television.

Behind The Meter: When The Grid Becomes The Bottleneck

Public grids were not designed for sudden clusters of ultra-dense computing. Interconnection studies drag. Transformers have lead times that would make a shipyard blush. Local opposition shows up at hearings. In that environment, a data-center developer does not always wait. The developer looks at on-site generation, private wires, and hybrid setups that keep the campus alive while the utility catches up.

That shift is easy to mock as a rich-company workaround. It is also a real addressable market for firms that sell generation and related equipment. I do not think every campus will become a mini power station. I do think enough of them will try that the equipment channel stays busier than a pure “wait for the grid” thesis allows.

Water is the sleeper issue. Communities that will tolerate a new substation may still revolt over cooling demand. That fight will shape where campuses go next. It may even shape which generation technologies look politically easier in a given county. Gas is not automatically the winner of that politics. Reliability still gives it a seat at the table.

A Human Read On Corporate Confidence

Executives always sound confident on stage. That is part of the job. What felt different here was the refusal to over-claim a slowdown that had not arrived. “We continue to monitor it” is not a slogan. It is an admission that the rate channel could still bite later. I prefer that to the usual chest-thumping.

There is also a cultural tell. Calling gas a destination fuel is a choice to pick a side in a long argument. Some readers will bounce off that sentence. Fine. The operational point survives the branding: this management team is not running the company as if gas were already a sunset product. Capital allocation, hiring, and manufacturing capacity tend to follow that belief whether you like the slogan or not.

If you want a personal take, here it is. I would rather underwrite a boring contracted gas project than a dazzling uncontracted dream that only works in a perfect rate cut. Boring pays the notes. Dreams look good in decks.

Risks That Could Still Break The Story

A sharper recession could cut industrial gas demand faster than data centers can grow. A sudden peace premium could unwind some of the geopolitical tightness that supports prices and urgency. A wave of project cost inflation could eat the returns that offtake was supposed to protect. And yes, if rates stay restrictive long enough, even good projects can get recut or postponed.

Technology risk sits in the background too. If power usage per unit of compute falls faster than total compute rises, the electricity scare eases. That would be a gift to grids and a headache for anyone who underwrote a straight-line explosion in generation kit. I would not build an entire thesis on that rescue arriving on time. I also would not ignore it.

Policy is the wild card that does not fit a tidy table. Export rules, sanctions, shipping insurance, and domestic content requirements can rearrange the winner list without changing the physics of demand. Investors who treat energy as a pure commodity chart miss that layer until it slaps them.

What To Watch Over The Next Several Quarters

Watch final investment decisions, not keynote adjectives. Watch whether LNG buyers keep signing long contracts or drift toward shorter, more optional structures. Watch equipment lead times. If those stretch, demand is still colliding with finite workshops. If they collapse, the boom is cooling even if the speeches stay hot.

Watch the mix inside backlog. Growth in gas infrastructure and data-center generation tells a different story than a pile of low-margin aftermarket work. Both can be healthy. Only one supports the “energy demand decade” pitch with real force.

And watch local politics around power and water. A handful of high-profile campus fights can change siting patterns faster than a global forecast. Energy is local more often than the macro slides admit.

The Through-Line Worth Keeping

Higher rates were supposed to be the wet blanket. Geopolitics was supposed to be the chaos agent. AI was supposed to be the wild new buyer. Put those three in one room and you do not get a freeze. You get a strange kind of continuity. Projects with real offtake keep marching. Gas stays in the middle of the power conversation. LNG capacity still has a long shopping list if the demand case holds.

That continuity can break. Most cycles do, usually at the moment the consensus gets comfortable. For now, the industrial signal is clearer than the rate narrative: energy projects have not stalled on cue. The people building the kit are still busy. The people buying the electrons and the molecules are still raising their hands.

If you came here looking for a neat moral, I do not have one. I have a working observation. When demand is structural and contracts are real, the cost of money can hurt without stopping the work. That is a less exciting sentence than a crash call. It may also be the one that matches the next few years of project reality.

Keep an eye on the offtake sheet. Keep an eye on the grid. Keep an eye on whether that 2035 LNG capacity ambition still looks necessary after the next scare. The rest is noise until a crane actually stops.