Hungry, Hungry, Hyperscalers with Scott Harlan

Greg Dowling (00:05):

Welcome to the FEG Insight Bridge. This is Greg Dowling, Head of Research and CIO at FEG. This show spans global markets and institutional investments through conversations with some of the world's leading investments, economic and philanthropic minds, to provide insight on how institutional investors can survive and even thrive in the world of markets and finance. In our latest FEG Insight bridge titled Hungry, Hungry Hyperscalers, we discuss the digital world's insatiable need for energy, and we are very excited to host Scott Harlan, a founding partner of Rockland Capital, to help us with this discussion. How will the AI transition be powered? How do we combine the growing need for energy with a desire for more renewables? Does this impact grid reliability? Can batteries plug the gap? Hear all of this and more. This is sure to be a very powerful podcast. Scott, welcome to the FEG Insight Bridge.

Scott Harlan (01:05):

Thank you very much for having me. I appreciate it, Greg.

Greg Dowling (01:08):

Alright, would you mind introducing yourself and Rockland Capital?

Scott Harlan (01:11):

Sure. So I'm one of two co-managing partners for Rockland Capital. Rockland is a private equity fund that specializes in the UK and the U.S. power industry. Most of our investments have been in existing power generation assets, which are stressed due to operational or commercial challenges or distress due to poor market conditions. Could be poor management or excess leverage. We're actively seeking out inefficiencies in the plants we purchase. We apply a commercial and a technical or operational discipline to cleaning up the inefficiencies and riding the ship. One distinguishing feature of Rockland is that our asset management team, which has the depth of experience, it really isn't matched by any others who invest in power. This team allows us to vet the operational and technical issues that we find in power plants so that we can understand what we're getting into and they manage the turnaround if that's part of the investment thesis.

Scott Harlan (02:09):

So I do think this team truly gives us access to a portion of the market that other investors shy away from. We formed Rockland back in 2003 as a pledge fund. Some would say a fundless sponsor. We started investing through committed private equity funds in 2009. We're currently investing our fourth flagship fund called Rockland Power Partners, not terribly creative on the name. I think the overall theme here is that power, you know, it's always been deemed a critical element in any economy, providing essential life sustaining services and being critical resource in the economic development puzzle. But really today the power industry is taking on such greater significance because of its role in decarbonizing the economy, meeting climate change mitigation goals, driving us into the digital and artificial intelligence age. So you know, this digital economy, it's just so much more power intensive than anything we've seen in the past and concerns about how the availability of reliable power will be a limiting factor on the digital economy that is intensifying today. So it's a pretty interesting industry. We're experts in this industry and we've been doing this for a long time now.

Greg Dowling (03:17):

Well, that's great and we're going to kind of dive into each of those little points that you made. But before we go there, I think this is correct. You were a meter reader and now you're a principal in a private equity firm. Let's talk a little bit about your career journey.

Scott Harlan (03:30):

Okay, sure. Yeah. I truly started at the bottom of the industry. Meter reader, just happened to be my first summer job after my freshman year in college. I was studying mechanical engineering in college and became interested in power plant engineering. So my education and career moved in that direction. After working as a performance engineer and construction engineer and power plants, I moved into the commodity and the finance side of the business. I moved my family to Houston, helped startup an electric power marketing and trading business for Koch Industries. Getting that perspective from a trade floor was really helpful. Some partners and I identified a power plant restructuring business opportunity that really arose through the intersection of all of these skills, the operations experience that I had, the commodity market experience I had. And you know, through the course of getting my MBA, the finance experience I had, and ultimately this restructuring business is what evolved into a successful private equity business investing in the power industry. So that's the quick story.

Greg Dowling (04:32):

Well, I love it. I've always seen the meter readers out there. I don't know if they, they don't read meters now, right?

Scott Harlan (04:38):

It's hardly a job anymore. Yeah, there are very few meter readers out there. Most of it is done electronically. You can drive down the road and read meters, you know, remotely. It's a job that really doesn't exist that much anymore.

Greg Dowling (04:50):

It's like the horse and buggy industry.

Scott Harlan (04:52):

That's it. Yeah.

Greg Dowling (04:53):

So you do that, you become this private equity guy. But during that whole period of time, you've hit all these different roles within the power generation industry. What's it like? What are the challenges of operating a power plant?

Scott Harlan (05:07):

Let me preface it by saying, first of all, the industry really was dominated by vertically integrated utilities. When I started my career. There's been a gradual shift, you know, where power plant ownership has moved from the utilities over to both private and public, non-utility owners. And so that regulated business model really led to lethargic management and operating practices, given that profits don't depend on efficiency, but rather by how much capital you deploy. So on the operating side, the biggest challenges we've had are usually when we inherit a power plant from one of these utilities, there's usually a fairly big culture shift required to get operators to focus on things that drive performance and profitability today. Given that the grid is really increasingly being supplied by these inflexible and intermittent renewable generation, you know, this has increased the burden on the older dispatchable resources that we tend to buy.

Scott Harlan (06:00):

And so these plants need to start on shorter notice. They need to ramp up and down more quickly. They need to fill the gaps when the wind isn't blowing or the clouds move in front of the sun. Modifying operating practices for plants that historically have just kind of turned on and run, and enhancing the equipment to increase the flexibility to start reliability of existing generation is one of the ways that we add value. And it requires a mindset shift with the operators and that takes time and attention from our asset management staff. So I think that's generally the greatest challenge is a people challenge. It's a mind shift that you implement with the plants that you're purchasing.

Greg Dowling (06:37):

Alright, I'm going to ask the maybe the billion dollar question. Can we meet our current expected energy demand given the age of our power generation facilities?

Scott Harlan (06:48):

That is the billion dollar question and it may become a trillion dollar question actually. I think the quick answer is that it's going to be a major challenge, but I think we're up to it. I don't think the people understand the magnitude of the challenge though. So, you know, new power generation is needed under any scenario. And the most economical way to meet the future needs is to also extend the life of existing generation while simultaneously building this new generation. The problems really exacerbated because most of the power plants that are retiring today are dispatchable thermal plants. While most of the plants that are being built are intermittent renewable projects, and this is really happening for two different reasons. The first is economics, and the second is environmental policy. So on the economic side, you know, it can be costly to run older plants, coal plants in particular.

Scott Harlan (07:37):

And some of the older nuclear plants have really high fixed operating costs and they don't make enough money in the energy market to cover those costs. So it's easier to extend the life of those older gas plants. Nuclear plants are increasingly receiving subsidies in some states to stay afloat. But except for one notable exception, actually in the state of Ohio coal plants, they're not attracting these subsidies for environmental reasons. So coal plants are shutting down for economic reasons. On the environmental policy side, there's obviously broad government and corporate efforts to reduce carbon footprint. Solar and wind are really, you know, fairly low cost means of producing electrical energy, considering that they have such low operating costs and they don't have any fuel costs. The problem is that they aren't controllable and they only produce energy when the sun's shining and the wind is blowing. So really what's needed is a balanced mix of resources to meet the growing need for power in a way that meets some economic and environmental goals as well.

Scott Harlan (08:40):

So I think there's four different things that we need to do. Number one, we need to build new renewables that's been happening. These are the most effective generators of energy, but they don't provide reliability services. Number two, I think we need to build new battery storage projects and that provides short duration storage. Those, so these storage projects they'll charge when the sun is up during the day and they discharge in a couple of hours after the sun goes down. But while power demand is still at a fairly elevated level. The third thing is we need to build new dispatchable thermal generation to provide reliability and backup services to the grid when the sun isn't shining and the wind isn't blowing. And then the last thing, and this is an area where Rockland focuses mostly, is extending the lives of existing thermal generation and repurposing it to be more responsive to a grid that's experiencing greater price and demand volatility that's caused by those intermittent renewable resources. You know, I did take a long time to get through that, but I think those four things are needed. There's not just a magic bullet, it's really truly all of the above that's needed.

Greg Dowling (09:43):

So let me ask you this question. Who pays for all of that?

Scott Harlan (09:46):

Yeah, well ultimately it's consumers that have to pay for all of it. But the investments I think increasingly are going to be coming from the private sector. In the past, when I say the past, like going back 20-30 years, this all came from regulated utilities that had cost-based regulation. So when they made an investment, their rate payers, their customers were on the hook for it. Today, when more of these investments are being made by non-regulated entities, either private or public independent power companies, the investment is being made by those companies. But the risk of of loss is also borne by those companies. So I think it's overall a better construct today because capital is being deployed, I think in a more effective way, rather than regulators and utility managers making those decisions and knowing that they can always jam any cost to their customers. The risks are being taken by the private sector and I think that's where most of the money is going to be coming from.

Greg Dowling (10:44):

And I am a true believer that the private sector can do things more effectively, efficiently than the public sector. Right. And they are bearing the risk of their capital. I guess what I worry about though is the pushback is there is a profit motive and if you're trying to make profit, is there a potential backlash if rates through, and there's different layers of this, but if it kind of leaks through and goes up ultimately to higher consumer prices, no consumer's going to want to know that, hey, the reason our prices are so high is a private equity firm is making a lot of money.

Scott Harlan (11:23):

No, that's never good. But it's also never good when they're having to pay for useless assets. And that's the problem on the utility side. And I'm going to give you an example because I think that, I think this is an immediate example that's happening this year, and it has to do with nuclear power. So nuclear power is on the surface, you know, a wonderful resource, but it's also not cost effective. And so this year, the first nuclear plant came on in a long time, and it's this Vogel plant in Georgia, the Georgia Power bill, it came in seven years behind schedule. I think the original budget was $14 billion and overall they invested $35 billion to build it. The overall capital cost there is in excess of $15,000 per kilowatt of installed capacity. To put that in perspective, to build a solar project, I could do that for about $1,200 per kilowatt.

Scott Harlan (12:16):

I can build a gas peaker for $800 per kilowatt. I can build a very efficient gas combined cycle for $1,500 a kilowatt. All of these are less than 10% of the cost of building the last nuclear plant. Now the customers in Georgia are going to have to foot the bill for all of the mistakes for number one, the decision to build the nuclear plant. Number two, most of the excess costs that they incurred to me. So I think it goes both ways. I think the deregulation of the industry where more and more the power generation has been owned by the private sector, all the data suggests that our power costs are lower today because of this. Unfortunately, it doesn't always get communicated properly to consumers. So that's the challenge that we have.

Greg Dowling (13:02):

Yeah, that's our current status and we're going to have problems there. But what about this digital age? I mean, boy, these data centers, they consume lots and lots of power. How are we gonna do this?

Scott Harlan (13:14):

Yeah, they do. So in the long run I do think we're going to be able to satisfy this, but I think that access to power is actually going to be throttling the growth a little bit on this relative to what we're reading about. So the problem is not the ultimate demand, but the pace at which these data centers are coming onto the grid. These data centers are going to locate where public policy and resource availability allows the power infrastructure to grow more quickly. Ohio frankly is one of those places. Texas is another place. The southeast is really stepping up now as well. I think it's helpful to think about the digital age and the different types of data centers that are coming on. So I guess before we get into this, let me emphasize this. Really three different types of data centers. There's the traditional data centers. These have been around for a long time.

Scott Harlan (14:01):

It's the largest of these three different groups. You know, the world's just processing a lot more information, right? And so this has been growing for decades now. The trait that they have is they need a hundred percent reliability. It's very, very expensive. When the power goes out to one of these traditional data centers, the economic costs are staggering. The second one is crypto loads. Obviously this has been coming onto the grid at a fever's pace over the last five to 10 years. The downside of crypto power load is that lots of physical resources, fuel that kind of stuff are going into a currency, which typically hasn't been the case in the past. The upside though for the power grid is that this load is interruptable. They don't need perfect reliability. So it doesn't really require additional capacity to be built to satisfy the peak needs of the power grid.

Scott Harlan (14:46):

It consumes energy and fuel, like I said, but it doesn't require infrastructure additions beyond the substation to connect to the grid. And then these AI data centers, that's the last thing. I mean, they're the new kids on the block. Many applications, including lots of just basic search features, are now using AI technology to accomplish the task more effectively. Look at all the changes just in the last couple of months, like Google and Apple where they're incorporating like now you do a Google search and it says this is your AI answer, right? It's fundamentally changing it. And that's been a big driver.

Greg Dowling (15:17):

And let's focus more on that AI side. Can you give us any expectation or how do we frame that energy need that the AI needs versus just a normal data center?

Scott Harlan (15:28):

In general, what I've read is that a rack at a dedicated AI data center uses about seven times the power of the same rack at a traditional data center. If you want to think more on absolute terms, haven't seen this for a couple of years, but I know a couple of years ago, overall, you know, 400 terawatt hours of electricity was consumed by traditional data centers and hardly any by ai. There was another a hundred terawatt hours that was concerned by crypto mining. Generally, you know, what we've seen is that the first two of those are going to increase by about 25% over the next four years. But with AI added to it, the overall data center load is going to almost double over that four year period. So it is all the growth is really being driven by the AI portion of it. And I guess the last thing I'll say again, to put it in perspective, in 2022, I think about 10% of the overall U.S. consumption of power went to data centers. I think by the time we get to 2030, that's probably going to be more like 15%. Wow. So it's a significant part of the overall demand.

Greg Dowling (16:35):

And just kind of a jargon check, you said terawatt and that loosely translates into lots, right? A lot. A lot of lot.

Scott Harlan (16:42):

Right. Right. It's a million megawatt hours or a billion kilowatt hours. So kilowatt hours is what you get on your electric bill. So a terawatt hour is a billion of those.

Greg Dowling (16:54):

Could you imagine being a meter reader and going up to a house, if people like you spent seven terawatts.

Scott Harlan (17:02):

I'm used to reading kilowatt hours, not terawatt hours. Yes.

Greg Dowling (17:06):

So how are we going to power these? Is it going to be like a combination of renewables with some natural gas and does that drive where you locate these data centers?

Scott Harlan (17:14):

Yes, it does. And it's not just the power intensity. There's various different considerations that the data centers have that might be different from typical loads. So one is just speed to connect right now, this is a quickly evolving industry, you know, the digital economy, like everyone's scrambling to be the first right and to be the biggest, right? So speed is really important and these interconnection queues are slowing that down. It's very difficult to build a greenfield data center, right? Secondly is reliability needs. I mentioned both the AI and the traditional data centers need very, very high reliability. Third thing is economics. And this is critical. Fourth thing is a lot of these hyperscalers are ESG sensitive and they're looking for low carbon power sources, right? You know, let's focus on crypto first. This is interruptible. Frankly, the Texas market is ideal for a crypto load because Texas is allowing renewables to come in that drops the price of energy.

Scott Harlan (18:09):

They're very price sensitive, but at the same time it's a very volatile market, right? And so they can interrupt their load. And we've looked at this in Texas, the actual power price. If you clip the top 5% price wise, the top 5% of the hours in the year, your average price is going to be less than half. What it would be is you needed power 100% of the time. So crypto is just looking for volatility and low prices and they don't care about reliability. But for these other data centers, both the traditional and the ai, they need this absolute perfect reliability. And they have ESG pressures, you know, if they want zero carbon, nuclear is really their best option since it's a continuous, unlike renewables. And since it has zero carbon. And lately the owners of nuclear plants, two of them that come to mind are Constellation Energy and Talent.

Scott Harlan (19:00):

They've been striking deals with the Googles and Amazons out there to locate their data centers right at the new plant. And then they carve out a portion of the capacity to dedicate it to the facility, to the data center. And this can work for an existing nuclear plant. But like I said, a new nuclear plant is really completely unfeasible from an economic point of view. So I don't see new nuclear being built to do this, but I think just cannibalizing the existing new plants for this purpose really isn't sustainable because the grid needs those as well. Right? So that's the ideal. I think ultimately though, they need a strong power interconnection. And I think ultimately the same thing that is true for the whole grid is going to apply to the data centers. It's going to be a mix of renewables of battery storage and then gas fire generation to kind of back those things up when the wind doesn't blow and the sun doesn't shine for a period of time. I think it's going to be all of the above and a really strong grid interconnection.

Greg Dowling (19:56):

It's interesting you mentioned the utilities and the lead up to us recording this podcast. You know, we were talking back and forth and you had sent a pretty interesting article about something in our own backyard we're, headquartered in Cincinnati, Ohio, but this was more Columbus, Ohio, where there are a lot of different data centers. Intel's building a plan up there and AEP was saying, Hey, I just want to make sure you're going to use all this power that you're demanding and it's kind of like a take or pay type of demand that they're making. So maybe you can talk about how the utilities kind of intersect with this.

Scott Harlan (20:31):

Yeah, I mean the thing in Ohio that we're trying to avoid frankly, is having the utilities, you know, with their regulated model be the ones that provide all of the services. And you can see just in that article, they're demanding to be indemnified, effectively, you know, by the people that are building. You know, that generally doesn't happen in the private sector and people are more willing to take the risk of, you know, the demand going away and having to reposition the plant to sell it to someone else and that kind of stuff. You know, I'm not so sure that if I'm a data center that I want to go to a utility that historically is not the most effective at building things and then they're demanding that I've got to do a 20 year take or pay contract, right? So to me that's not the most attractive thing. And if I were Google, I'd be calling Rockland Capital instead of AEP to build the power plant. So it does show the concern that the load may not be a resilient change, you know, to the digital economy. And maybe these data centers aren't going to stick around forever. I don't know that we're moving away from the digital economy anytime soon. So I think I'm probably less concerned about that than AEP is.

Greg Dowling (21:38):

So you mentioned batteries. Has battery technology changed enough or is it going to change enough in the future? You mentioned like two hours. I mean that's about the most it can do. Is it cost effective for two hours?

Scott Harlan (21:50):

Yeah, it is. Two to four hours is sort of the sweet spot for batteries. Like I said, you know, they can charge during the day that way you know, when the sun is out and then, you know, the evenings are increasingly a difficult time for the grid operators because demand stays high and the sun goes down. So I think that's what batteries are going to do very well. They're going to help to manage the ramp up in the morning as the sun comes up and the ramp down in the evening as it goes down. What they're not gonna do well is deal with wind droughts, for example. I mean, wind droughts is a new meteorological term that you hear people talking about where the wind just doesn't blow for a week or two. Right? It's particularly bad in northwestern Europe. The North Sea sometimes just goes calm, right?

Scott Harlan (22:31):

And they have a ton of wind capacity out there. It's the reason that we're investing in gas fired peaks in the UK because there's gonna be periods of weeks, you know, where, where the wind doesn't blow. And that's what they're relying on there. The other thing is, I don't think the battery technology is going to be good, even for overnight storage. The sun is going to continue to go down every day below the horizon. It's not going to shine at night. If you get a time where the wind isn't blowing and obviously the sun is down for, you know, on average 12 hours a day, you need something to produce energy at that time and it's not going to be a battery anytime soon. I don't think so. So I really do think the economic, the batteries are critical because they can turn on instantly and turn off instantly. They can switch between charge and discharge just on the snap of a finger. But I think they're going to be critical for that two to four hour period and gas is going to be needed to fill the gaps better longer than that.

Greg Dowling (23:23):

Very good. Well let's talk a little politics here.

Scott Harlan (23:26):

Okay. I always love to talk about politics.

Greg Dowling (23:29):

Does it matter who wins in November? Does will that change our energy policy?

Scott Harlan (23:36):

I'm going to say quickly, not as much as you think, particularly at the federal level. I do think the caveat there is at the federal level, I think it can matter more at the state level. Unsurprisingly, nobody is running on a plank of grid reliability. Not exactly the sexy topic for either Biden or Trump to talk about meaningful issues are really more about environmental regulation and permitting reform. You know, the basic structure that we have today of a competitive wholesale market where anyone can own power generation is not gonna change if Democrats or Republicans are in office. I do think the federal government can have an outsized impact on environmental regulation. So generally you would expect, you know, a less burdensome environmental controls from the Republicans than the Democrats that could affect the life of coal plants on the margin. But the bottom line is that coal plants are shutting down for real economic reasons more than due to the environmental regulations.

Scott Harlan (24:33):

You know, the next thing is natural gas plants. They are being targeted by the Biden administration right now for greater carbon dioxide controls. There were regulations that the EPA just came out with that for any new gas plants that have high capacity factors, so anything other than peakers, you have to capture 90% of the carbon dioxide coming off of those. It's economically really not feasible and Republicans would stop that I think, but it's likely that the courts are going to stop it anyway. Right? So I don't know that the winner of the election is going to have that material of an impact there. I think surprising to many tax credits for renewables seem to be the one thing that Republicans and Democrats can agree on. Democrats for environmental reasons, Republicans because it reduces corporate tax collections. So I do think that tax credits are going to continue to drive the renewable buildout.

Scott Harlan (25:23):

That's something that's not going to change depending on who's elected. I think one of the bigger issues could be defer the president appoints for commissioners, much like they appoint judges. You know, however, unlike judges for commissioners have defined short terms. And so the makeup of FERC changes more quickly after an election than the make of the federal judiciary. Democratic FERC commissioners sometimes have an environmental bent and Republican commissioners tend to lean hard towards states' rights over federal protection of interstate commerce. You know, both of those can have an impact. I think that's probably where the presidential election has the biggest impact on our business. Like I said before, though, I think state issues really are the ones that keep me up at night. Governors and state legislators tend to ignore issues, you know, affecting free and fair functioning of markets and they can do things that tilt the balance towards whatever their preferred resource is. So I think that's where the greater risk is.

Greg Dowling (26:16):

That's great. It's kind of the inside baseball, it's like we hear about this top level, but it's, it's kind of below the presidential seat where the changes really impact you.

Scott Harlan (26:27):

All politics are local, right?

Greg Dowling (26:28):

That's right. All politics are local. That's exactly right. Well let's talk some dollars and cents. So what's the investment case? How do you make money in this? Is that kind of a picks and shovels approach or are you looking for those sexy properties or generation? Tell us about what you're doing.

Scott Harlan (26:44):

It's more the former, the picks and shovels. We tend to not go after the latest and greatest. You know, we've talked a lot about the massive growth of renewables and batteries that we expect. If you look at the forecast right now, I think between now and 2050, the Energy Information Administration is forecasting about an 800 gigawatt growth in solar and maybe more like 150, you know, or 200 gigawatt growth of gas plants. So much more solar coming onto the grid. You would think that that would mean we should go after that. But that's not necessarily where we're putting most of our capital. What we're doing on the renewable side, really focusing on the development portion of this, but we're not putting our capital into building and being the long-term owners of the renewables. A lot of the times the renewables have long-term contracts. It tends to trade more like bonds, but I don't think they have the risk of a bond.

Scott Harlan (27:38):

So I don't think that owning the renewables is the best risk adjusted return. So where we make our money on the renewable side is developing projects, getting interconnection agreements in place, getting land options secured, permits secured, and then we'll turn around and sell a construction ready project to someone that has a lower cost of capital. Where we're putting most of our money is into existing operating gas plants that are currently providing reliability services or can be repositioned somehow to provide reliability services to a grid that needs those services more. A lot of times we're leaning heavily on our asset management team to make power plants more flexible so that they're a better tool for a grid that depends more on those intermittent renewables. And that's where we see generally the better returns. And that's where 80-90% of our capital is going.

Greg Dowling (28:26):

Well, let's take those two examples that you had. So one is kind of the shovel ready, got to get the property ready for it. Why aren't the utilities taking that risk? They're obviously okay taking some risk, but not that risk. So are you being properly compensated for, you know, getting the permitting and everything else done and why can't they do it?

Scott Harlan (28:45):

They're just not that nimble. I mean that's the advantage that we have over them is that I think we're better at assessing risk and moving quickly, you know, in response to different risk stimuli. So do we get properly compensated for, generally we underwrite the development efforts to a three to four times multiple on invested capital. It is higher risk than just buying an operating power plant. But I think we can easily, and our history has been that we easily get three to four times multiples and utilities just, they're not structured to deal with any kind of losses. So the way we get a three to four times multiple, they're going to a be a few projects that we put money into where very quickly we say, oh, there's a binary risk here that I, you know, there's a high risk, we're not going to be passed that one and so we're going to cut our losses and stop.

Scott Harlan (29:32):

But then there could be another one where we make a 10 times multiple on it. And then, so the whole portfolio, we're making a four x on utilities can't deal with any losses. They can't deal with it, right? Because they might be disallowed by the regulators and development is not something that they do well. So I think it's our nimbleness, our ability to manage risk and to recognize risk and to make those decisions quickly. 

Greg Dowling (30:05):

That makes sense. And then on the existing power plants, you're trying to make them more efficient, you said it's maybe better operators, different mindset. What else? I mean, are you doing conversions? Are you doing like a coal, a gas conversion? What's the risk there?

Scott Harlan (30:20):

We've tried to do coal to gas conversions. You know, we had one project in New Jersey and believe it or not, the Sierra Club made it very difficult for us to do it. So we ended up continuing to operate as a coal plan and then ultimately shut it down. This is before we were managing private equity capital. I do think the interconnections at a coal plant I think are very valuable. I believe that today it's easier to use those interconnections to connect a renewable project or a battery project. I think in the future you're going to see more and more gas plants co-located renewable projects are interconnected. I think that that's a thesis that we have now is that a lot of these older plants that we're buying, they'll ultimately be repurposed for something else that may be a low carbon dispatchable gen, or it may be a renewable project or a battery project. That coal plant that I mentioned, it is now going be the landing point for an offshore wind project off the New Jersey coast. The transmission line is going to come underwater, come up right there at the power plant, and that's going to be the next incarnation of that plant.

Greg Dowling (31:23):

That's great that you can actually reuse these big giant plants. It's probably good for the community, although I guess probably a lot less jobs I would think going from one to the other.

Scott Harlan (31:32):

It's less jobs, but from an overall capital productivity purpose, you know, the power grid is designed to have a lot of power coming in at these discrete points, right? And so in order to avoid a lot of, you know, system upgrades, if you can have the renewable projects coming in at the same point, it's much more effective for the power grid.

Greg Dowling (31:51):

So you're making these investments, you mentioned that you might sell to a utility. Where else do you exit to? Is it infrastructure funds? How do you ultimately, because you only make money once you sell it?

Scott Harlan (32:04):

We rarely sell to utilities. Normally we've done a few things after we buy it, right? We've de-risked the operations frequently, and then we really focus on commodity risk too, and trying to de-risk the commodity side. So to the extent that we take a project that has all of these operating challenges and remove those and de-risk the commodity side, then an infrastructure fund is primed to take it. And so there's been a number of times where we've just exited to infrastructure funds. The other thing we'll do, a lot of our competitors in this industry, you know, on the private side, are much larger than us. We're truly the smallest of the private equity funds that's investing in power. And so a lot of times we'll buy these individual plants kind of in one-off transaction, aggregate portfolios of similarly situated assets together, build a larger portfolio. We did this in Illinois for example. We're doing this in the UK where we build portfolios of projects through multiple acquisitions and then sold them to we, we haven't sold the UK assets yet, but in Illinois we sold to Vision Ridge who was, you know, trying to make, you know, they were a larger fund than us and they're trying to make a bigger splash. And so that's frequently something that we'll do. Is that aggregation play to a larger private equity player.

Greg Dowling (33:17):

Do you have to be private? Can I just buy utilities? It seems like outside of Nvidia there's been a few pretty amazing utility stocks this year. Is that a way to play this?

Scott Harlan (33:27):

You know, there are, like, there's Vistra, NRG, Calpine used to be public, but Energy Capital partners took them private. They do play in the same space that we play in. So there are similarities, but generally they also have a retail business. They're selling to individual consumers and to industrials. And so it is different. It's not a pure generation play the way we are right now. The correlation between NVIDIA and NRG on one hand and Vistra on the other hand is stunning. Over the last year you just look at their stocks and their trading together. So a lot of people I think are feeling like there's a direct line between the digital infrastructure and our infrastructure. I'm not so sure that I agree with that. Absolutely. Like I said before, I do think that the power system is going to throttle the growth on the data center growth, but I don't see the same lockstep, you know, growth opportunities and power that you might see with Nvidia and some of the large tech firms. I do think one of our benefits is our size. We're much more nimble than a lot of the public counterparts in the industry. You know, we're not having to meet the quarterly expectations of Wall Street, which I think is a huge advantage.

Greg Dowling (34:37):

So let's talk about some risks here. What about higher interest rates on private equity? Does that impact your business?

Scott Harlan (34:44):

Not really. I mean I didn't address this when you asked me kind of how we create value, but one of the things that we do is use very little debt on our projects. And that's because we want to have, you know, a lot of control over how we affect the turnaround. When we get into a plan, there could be regulatory risk. So to your question about risk, when we're stepping into something where there's regulatory or market risk, having a low debt burden helps to ride through the ebbs and flows of ill-advised policy. And so like in our fund four portfolio, we have less than 15% debt to total cap. Now the area where in the power industry you are exposed to interest rates, notwithstanding the fact that we don't use a lot of debt is if you're in the renewable sector. So renewables typically have long-term contracts and so they tend to be much more fixed cash flow investments. And so I said before they look a little bit like a bond and they trade like a bond in that environment of higher interest rates. Those investments should theoretically go down in value and if interest rates go down, they should go up in value. So they trade like utilities and like bonds in that fashion. But I really don't think that Rockland strategy has that much exposure to interest rates.

Greg Dowling (35:58):

Is there any risk that this power demand story growth story is overstated and this doesn't come to fruition?

Scott Harlan (36:07):

First of all, I don't think that Rockland's strategy is as dependent upon demand growth as it is on changes in the supply mix. That's where we've always made our money, is by trying to project what's going to happen on the supply side. Now the growth and demand is obviously creating a tailwind for the industry. And so to the extent that we're investing in power plants, you know, for end to 20 cents on the dollar compared to building a new plant, then we're going to do well when the industry is having to build new plants, right? So there's going to be an uplift. But I actually think the demand story is robust. It's not just data centers. If you think about it, everything about the energy transition economy wide is relying on the power sector. In order to decarbonize the transportation sector, you have to use electric vehicles in order to decarbonize buildings.

Scott Harlan (36:57):

You're replacing gas and oil heat with electric heat in order to capture carbon dioxide out of industrial facilities. It's very power intensive. And even the hydrogen, I mean you're hearing a lot about using hydrogen now to make green hydrogen electricity is effectively a feedstock for it, right? You put a current of renewable power through water, you split the molecule into hydrogen and oxygen. You know, all of these things are driving growth. There's also the onshoring of manufacturing. There's public policy now that says we're defending too much on China for critical chip manufacturing and that kind of stuff. So you see a lot of onshoring and manufacturing that's particularly having an impact in the Midwest and the southeast right now in the increasing demand, I think it's resilient, but it's not the end of the world if we go back to a 1% a year growth and power demand because our strategy is really more focused on what's the supply mix going to be in the future.

Greg Dowling (37:49):

Kinda wrapping it up and bringing this full circle, were you ever chased by a dog as a meter reader?

Scott Harlan (37:56):

Ah, yes. As a matter of fact, I have scars to prove it. I would carry a flashlight in my back pocket and that's what I used to bang the dogs over the head if I felt like I was threatened. And so, yeah, I was bit by a dog and I had another dog that I thought was going to tear me up, but the owner, frankly was there and he screamed at me and he said, don't move. I just froze. And thankfully the dog was all over, but he didn't bite me.

Greg Dowling (38:26):

See, I felt like I had to come back to that question. I felt like there was something there.

Scott Harlan (38:29):

It all comes back to meter readers. My biggest nemesis as a meter were fleas. Honestly, in the day you had to go down into the basement of houses in the city and a lot of these houses had fleas. So we also had to carry Deep Woods Off with us. There were all kinds of perils as a meter reader.

Greg Dowling (38:46):

Fleas, dogs. Wow. That's great. Hey Scott, thank you for this very energetic podcast.

Scott Harlan (38:54):

Okay, good. I'm glad it was energetic.

Greg Dowling (39:01):

If you are interested in more information on FEG, check out our website@www.feg.com. And don't forget to subscribe to our communications. You don't miss the next episode. Please keep in mind that this information is intended to be general education that needs to be framed with the unique risk and return objectives of each client. Therefore, nobody should consider these to be FEG recommendations. This podcast was prepared by FEG. Neither the information nor any opinion expressed in this podcast constitutes an offer or an invitation to make an offer to buy or sell any securities. The views and opinions expressed by guest speakers are solely their own and do not necessarily represent the views or opinions of their firm or of FEG.