I’ve been listening to a lot of earnings calls lately to tease out power market related implications for data centers. Here’s a compilation:
Deutz group acquired Blue Star Energy in the US to capture the genset market growth story with Data Centers:
regardless which party and which candidate will win the election later this year. So this relocation is in the end, a result of several government subsidies, particularly in the high-tech sector, if we mentioned battery production, semiconductors.
And for those sectors, the requirement, the necessity for backup gensets for business-critical tasks is another growth driver for decentralized electricity supply. And connected to that investments, including public investments in digital infrastructure think about data centers. Again, they need to higher demand for backup solutions, including also the various laws in the U.S., pushing really the investments in airports, water infrastructure and transit.
Let's move to Blue Star, in particular, and Blue Star in particular is one of the top 10 genset producers in the North American market. What are they producing? Producing and obviously selling diesel and gas-powered gensets capacity from 20 to 2,000 kilowatts. Currently, the company is employing a bit more than 100 employees, and we expect the revenue in '24 of above USD 100 million.
Energy Transfer LP knows that demand is going up, just by how much isn’t clear:
And so we're looking at playing a pipeline to a large chip manufacturer in Texas and as well as that we will -- we're believers like everybody else, there is the data centers and especially around AI, it's going to happen. Whether that means over the next 5 or 8 years, it's going to go drive by 3 Bcf demand of gas-generated electricity or 8 Bcf. We don't know. We just know it's going up. So in combination with population growth, as I mentioned earlier, industry growth, ammonia growth, all the AI data centers, et cetera, power plant growth. We're talking to probably separate different power plants at least on fairly significant natural gas-fired generation expansions in Texas, a handful of Oklahoma as well. So it's just that common theme that Tom and I keep talking about in this call is that the demand for natural gas is going to do nothing but go up for many years to come, and we're excited that we have the assets that we believe will benefit the most from those opportunities
Core Scientific believes that DC demand will double by 2030. So that will bring ~25GWs to 50GWs in total by 2030. Incremental of 25GWs:
Yes. No, it's a great question. I think Morgan Stanley put out a very good report related to the opportunity that bitcoin miners actually have today given the fact that just on electrical equipment alone, it's at least 36 months lead time for traditional data center. So just having access to the power is a significant advantage and it's actually a much higher value to traditional data centers and really the valuations that we're seeing bitcoin mining infrastructure traded today. I wouldn't rule that out, traditional data centers are definitely trying to find ways to bring power online more quickly.
What we're seeing across a number of reports is that data center capacity is going to double over the course of the next 6 years. So I think that's something that we're still in the early stages. I would imagine that companies throughout the industry are having those types of conversations. From our perspective, we're focused on executing this because we believe we can drive a significant amount of short-term and long-term value for our shareholders.
Suburban Propane believes that the voluntary market for RNG will be a lot bigger over time, even though LCFS has had a lot of volatility in general:
So I think for the longer term, we do see -- again, back to the concept of we're in the early innings, we do see the markets for RNG developing in a way where RNG demand is going to expand beyond just transportation I think as more and more of the demand for electricity grows from the adoption of electric vehicles and expanded AI with the need for enhanced computing and data centers, I think you're going to see a significant uptick in demand for RNG to be a good source of power generation for the electric grid because I think the reality of powering with pure renewable sources from wind and solar is never going to keep up with the amount of demand increase on the electric grid. So I think we see voluntary markets for RNG demanding or developing over the course of the next 5 to 10 years as the reality of the transition towards electrification sets in that we need additional cleaner energy alternatives. And that's, I think, the opportunities of the future that we see.
EQT believes almost 10 BCF of gas demand per day:
Along with the material cost structure advantage, the combination of EQT and Equitrans will also create an integrated well to watch solution that will help enable and power growing demand associated with the data center and artificial intelligence booms that are burgeoning across the Southeast and Mid-Atlantic regions of the United States.
Our base case view suggests the proliferation of data centers along with growth in other electricity-intensive markets such as electric vehicles to drive an incremental 10 Bcf per day of natural gas demand by 2030. While there is a plausible upside case that could take this number up to 18 Bcf per day. This means growth in the power generation segment that exceed LNG exports as a bullish demand catalyst for the natural gas market this decade. And this structural baseload demand growth story resides at the doorstep of our asset base.
GE Verona:
ow the general economics for us with aeroderivatives are the equipment margins are more healthy than they are on a heavy-duty gas turbine project because admittedly, the services annuity stream and the operating profile of those aeroderivative units may not be as high or as consistent that we can count on because they're really there more as a support to those wind or solar farms that get developed. So when we underwrite gas projects on a heavy-duty project where it's something like a coal to gas switching, a larger proportion of the economics are on the services. With aeroderivatives, a larger proportion of the economics for us are with the day 1 equipment revenue.
DTE Energy:
In terms of growth, very good question. What we see there is if some of this growth accelerates, we will need to build -- and the growth that we're looking at, like data centers, for example, are 24/7 operations. That will need generation that cannot be -- and power that cannot be interrupted. And so we will need 24/7 dispatchable -- more 24/7 dispatchable generation and renewable resources.
So again, it will be a mix. It does have the potential for accelerating our next IRP should such demand present itself. In terms of impact to customers, we expect that those investments will pay for themselves and actually provide a benefit to customer bills to some degree. So that's our going-in position and we see that as very possible. It will also add value for our investors because those investments will be in rate base, and they'll attract the typical returns. So I think it's a win-win all around
Alliance Resource Partners:
While I could talk for hours about those examples, let me highlight 3 in particular. First, according to the grid -- the Clean Grid Initiatives/Grid Strategies report, the era of flat power demand is over. According to 2023 FERC filings, $630 billion of near-term investment in large load has driven the 5-year outlook for nationwide peak demand to 852 gigawatts a from just 835 gigawatts in the year ago report. That's a doubling of the 5-year growth projection from 2.6% to 4.7% in just a 12-month window.
n response to these rules, the National Mining Association called out EPA for, one, refusing to account for irrefutable evidence that electricity demand is soaring; two, disregarding validated warnings from grid experts related to coal plant closures; and three, ignoring the basic fact that there is no adequate replacement ready to replace the sorely needed dispatchable generating capacity coal is providing.
America's Power also issued a statement last week in response to the EPA's new Clean Power Plan 2.0 that describe the rule as "an extreme and unlawful overreach that endangers America supply of dependable and affordable electricity." They followed by saying, "The new clean power plan, is the same kind of overreach that caused the U.S. Supreme Court to reject EPA's first clean power plant in 2022."
Eaton sees tremendous growth:
Moving to Slide 9, you'll see an updated view on the data center market. Last fall, in our Q3 2023 earnings call. We highlighted the data center market and shared our view that we expected the market to grow at a 16% compounded growth rate between 2022 and 2025. We want to provide an update as we've seen continued momentum in this market, driven by the rise of AI, big data and certainly edge computing. As expected, the biggest increase is coming from the very strong demand for AI data centers which is reflected both in our orders and in our negotiation pipeline. Here, orders on a trailing 12-month basis have more than doubled, and our negotiations in the U.S. have increased by more than 4x. We now think the overall market grows at a 25% compounded growth rate between 2022 and 2025. And as you know, we have a strong position in the data center market and the data center/IT channel accounted for 14% of our revenue last year.
Ormat is expecting to price higher than expected for renewing assets:
We are negotiating today with data centers in the areas where we operate in Nevada and California. Mainly in Nevada, we are negotiating with utilities. There's a big question, who is willing to pay more because at the end of the day, utility company is regulated so they have much more strength over the long term. We're signing a 20-year PPA, 25 years PPA. So we need to balance between it. At the end of the day, if I will ask you if I don't know, Microsoft, Amazon, Google will be for 30 years.
You would say probably yes. But what happens if in 10 years from today, they can find much, much, much cheaper electricity. How will they manage the contract? Will they come like what you said and ask us to renegotiate it or not renegotiate it? It's a question. A utility company usually is regulated. That's less of a likelihood. So the question is on the price. Today, we see similar pricing. But if we are able to get better pricing for data centers, we'll definitely sign with data centers.
Wartsila isn’t sure about how much impact there is from the DC market:
I cannot link them immediately to data centers. We are, of course, supplying with many of the power plants, power to the grid. And in the grid, there might be a data center. So I would say it's more indirect than direct, in my answering. I'm not specifically now in this call aware of any direct power plant support, I mean power plants supporting directly an AI data center and what the running hours of that is. But I cannot say, but I would assume that they are running pretty much baseload actually, typically.
Next Era:
This is one of the most important slides that I'll cover today. And this shows different generation sources on a nearly firm basis. And you can see it isn't even close. Renewables are, by far, the lowest cost generation resource when paired together with battery storage. While the relative price ranges of different technology have increased due to inflation, new renewables have been less impacted and are favored by a wide margin. Renewables face few regulatory constraints, unlike other forms of new power generation. But not only are they low cost, they're fast to deploy and they're clean. And they also benefit from declining equipment costs. There's no surprise that demand for new renewables is high as reflected in a tripling of REC prices over the last 3 years.
Connecting gas plants takes long as well:
Here are some of the challenges, new build gas generation has to overcome. First, they also have to get in a long line transmission queues. Nine out of every 10 megawatts in the transmission queue today are renewables. They're expensive, gas turbines are expensive. Take a look at the right-hand side of the slide, I think it says it all. Long lead times, get in line. It takes 3 to 4 years to even get your hands on a gas turbine. And they face considerable regulatory uncertainty, particularly in light of the new EPA rules that just came out. And look, even if we all have a view that those regulations ultimately won't succeed from a litigation perspective by around '26, 2027, it's still going to take years to play out. It's going to create a lot of uncertainty as that unfolds. And it won't stop there. There'll be the next thing behind it, the next thing and the next thing
Source: Seeking Alpha