Old Wind or Solar Farm: Run it Ragged, or Rebuild?

Discussion is based on a couple of recent pieces he did for Bloomberg Green as part of his spark Lines series. One is peaker gas plants may have peaked after all, and the other is aging renewable energy assets are future options. You can get these reports on the Bloomberg terminal under an I Bullard, or you can actually subscribe on Bloomberg dot com as a reminder being if it does not

provide investment or strategy advice. And you can hear the fullest claimer at the end of the show him Mark Taylor, and you're listening to switch on to ben f podcast. Nat, thanks for joining us, Thanks for having me here. Mark, It's nice to be back. You've done a couple of reports recently on renewable reliability. Can you kind of explain where those came from and why we're talking about this today.

The history of renewable energy being integrated into grids is that for every few percent more renewable energy that gets integrated into the grid, people sort of tear their hair out and run around like the houses on fire, thinking that reliability is going to go through the floor and we're never going to be able to integrate more than the current percentage of renewable energy in the grid. You know, first it was five percent, that was ten minutes twenty.

You can look at grids in say Germany for instance, that are able to handle the integrate more than renewable same in Denmark as well. When you get to some other grids, like in the United States, you start to have some real challenges and thinking about what are we gonna do with this much renewable energy in our grid? How is it going to impact reliability? How is a grid planner, how is the developer, how is an asset owner? Do I need to be thinking about that and what

do I need to do? And California is a great sort of test bed for this as as you would imagine. One thing is big state in the big mark it too. It has a lot of renewable energy. And three, it has quite a lot of wind and solar that tend to arrive on the grid at specific times. Right. Wind tends to have its peaks in the early morning and in the late evening, whereas solar tends to have its

peak obviously in the middle of the day. And what those what those things I'll do together is they make the grid more variable in terms of the generation that's coming in. They also mean that you have to still plan for reliability in a time of greater variability, especially as California looks out towards a hundred percent renewable energy at some point zero carbon electricity. Then you have to you have to start thinking, how is a grid full of these variable assets going to play well with all

the other children? Right? How is solar or wind going to maintain the integrity and quality of the grid as you reach very high penetrations. And so California ran the study on a term that I would imagine, honestly, not that many people are familiar with, which is the effective load carry in capable, which is how much of a plant's total theoretical capacity of any kind can you be

counted on to provide when the grid needs it most? So, if you've got a hundred megawat wind or a hundred megawat soilar plant at the time when the grid needs it most, how much of that energy can you count on? And this is where things start to get a little bit interested before we get into that more deeply, can we put something to bed the grids in Germany and Denmark? Are they fundamentally different than what we find in California or the US? Or is it more of a dogma

you know that more renewables equals less reliable grid. It's a bit of both. But but on the technical side, the grids are just better integrated, like they're better built interact. But it very plainly and very candidate. And also they're

very different different they're very different types of systems. I mean, California is sort to buy its nature, long and linear in many cases, and they're still designed around central station generation historically, right, you know, the grids of the US are designed around a small number of large and then then eventually very large bulk generation plants way far out of town, and then a large number of small assets, probably closer into load centers, but they're not designed for

hundreds of thousands or millions of generators like in the really distributed fashion, or even for that matter, they're not really designed for I don't know, potentially thousands of utility scale. It's the relatively small solar projects that have variability in their output, even though that variability is pretty well known at this point. Okay, so let's go into that variability and reliability. So the effective load carrying capacity or capability,

what is that for wind and solar. The study which looked out to the end of this decade in California shows that the effective load carrying capability for wind it's been nineteen cent next year, meaning if you have a hundred megawat project, you could ring eye on nineteen megawats and that when the grid needs it most, and it's it's fairly stable measure out to the end of this decade. If you add four hours of energy storage to it, it performs better, but it doesn't do like amazingly well.

It's definitely very far from like the kind of load carrying capability you would get out of a natural gas plant. Just to clarify, four hour storage is a battery, right, that's about well, they say say four hours of energy storage in the study, so we we I think it's safe to assume in most cases that that is a battery, but that could be a gigantic kinetic energy storage system, that could be pump hydros And about that, the study

very deliberately does not specified batteries. However, let's be honest, most people that they're going to be developing four hours of storage right now are probably developing it using a lithi ion storage battery. Now, let's talk about solar, which

is not wind solar. Even the big industrial scale installations are pretty much useless from an effective load carrying capability in two it's it gets even worse as you go further into the future, because you've got you have more of you have more in the grid, you have more sort of collected variability. Reliability of each individual assets is

less significant over the breadth of this entire grid. But where it gets interesting and where it gets sort of fundamental, I think, is that if you add storage to that four hours of storage to get doesn't say that it's a battery, but not just assume that it's a battery, you get to an effective load care and capability of which i'll call it pretty close to a hundred in

this case. So basically, at the time when the grid is most going to need power from solar, if you have a battery attached to it for four hours of storage, you can pretty much treat it as if it was any kind of other, any kind of other high quality asset on the grid. And there's a little bit of decline in that overtime, but really not very much, like it goes down slightly to about nine. So that must make solar plus storage, you know, something we've seen kind

of emerging right now. It's kind of a no brainer for the future. I think it may. It may. It's the case very clear if there is any sort of concern about how an asset is going to interact with the grid, it makes it very clear that storage is probably, if not essential, then at least very very important. You would probably be leaving benefit on the table to not be doing storage in a in a situation like this.

You know, and if you compare the effective mode carrying capability in tracking PV with four hours of storage to tracking PD without storage, one point seven percent. I think you'll probably find that there is sufficient reason from a developer's perspective to go ahead and build that storage, and and to make it even even perhaps more compelling, you

may find regulatory imperative to do so. It may be a requirement that if you were to be building, you know, more than fifty megawatts of somewhere in any good installation, you have to build with it four hours of energy storage, or you have to in one way another contract for the same thing. It doesn't have to be on site. But developers being developers, it's probably pretty easy to bundle rather than to do separate parallel tracks for development. And

guess you're a fewer place storage developer. So what does this mean for build? Maybe I'm just not thinking about this right, but would you build more capacity than you than you think you would need or would you right size it to supply the grid for when it needs it most? I mean, I think you're you're gonna build

my math and this super reductive. You're gonna build basically what you get paid for it, right, if you're gonna if you can get paid, if you're gonna get paid for reliability, and you build the storage to do that if you're gonna get paid to shift your generation curve right to times when it's more valuable than you might do that as well. I mean, there there would be many dueling imperatives. One would be the imperative to make

more money by selling more lectrons at certain times. The other imperative would be preserved whatever contract you've established by increasing your load carrying capability. We tend to not see these conversations happening, you know, I live were at real time, but they definitely are happening, Like these are conversations that developers and utilities are having, that grid operators are having with asset owners, and so I think that we need to sort of view them in a fairly supple fashion,

which is a good thing. I mean, I think that you can build for it out four hours of storage for multiple sets of benefits. You could do it to make more money at certain times, or you could do it because you have to either Either way, I think there will be imperatives to do so. So who's taking the lead on this. Are there developers that are saying, hey, give me storage, or their storage companies that are approaching developers. How's the corpor dynamic playing out. It's a good question.

We see a bit of both. I think that the most common right now is that you have a large soilar developer that has essentially embedded storage as part of its development process and proposition. So you know, we also see some large puer play storage developers who are also

developers of of large steel generation assets. So it's kind of difficult to say which, you know, is it a separate development track, are they separate groups, or are they just sort of making this assumption that they're the kind of thing that you can integrate if you want or debt.

I think it will be very curious to see how in future you have an interplay between the off site sort of like third party four our storage that could be contracted alongside the generation asset, a solar generation asset. That will come down to wonky bits of technical specification I think, or requirements written by regulators. So at the end of the day, are we saying that wind and solar are reliable that you can factor in you know, a high percentage of renewables and deer mix. So I

think we're getting we're getting to this point. I mean we we in a sense look at the lead presented to us by the regulatory bodies in the In this sense, I think that like studies that have been commissioned by the grid operators and by the regulators are important guidance to us because that internal shape what developers are able to do, which in turn will help shape the grid.

I think on an economic basis, we certainly see this being possible, like we already you see the case that there's an overlap between the levelies cost of energy from four hours of battery storage and from a gas speaker plant, depending on where you are in the world, and that certainly is the kind of compelling market signal that makes people want to go out and start developing like this. But greater reliability, just I think we're greater effective mode

carrying capability. Think of that as a proxy for greater reliability. Here enables people that have confidence in in in planning for greater penetration to renewable energy. Having having witnessed as you have as well more than a decade of these development cycles. They go from seeming crazy outlandish and definitely pushing the boundaries of what's technically capable to business as usual within a couple of years. So you're in washing in DC. Tell me about how this can or can't

play out in an area that's not California. Well, it's all about it's all about what what's the imperative at the moment in time. So I think if you were building, for instance, song or in Nebraska, you may not particularly have this kind of requirement early on in the in the life cycle of greening the grid. At the same time, you have to look very closely at what the nature

of that individual grid is. Like if if the rest of the generation assets do not ramp very quickly, you know, if they're mostly plants that have a snower ramping rate and comes of how much power they can add permitted to the grid, then maybe you do end up sort of ahead of time having some kind of some kind of storage assets there to enhance the good carrying capability

because you need it earlier. I also just think that it's it's the sort of useful guidance as you think else sort of like, if this is the trend that has begun in California, will probably propagate elsewhere. You know, if this is if this is the sort of thing that begins, it begins in one of the world's biggest, highest renewable penetration, multi resource power markets that's very professionally run,

that has a way of propagating elsewhere. So California also, I mean, I guess we have to talk about the elephant the room, right, California also has a lot of gas that that serves this role is to supply power and when the grid needs it most. How does that

play out into the future. So I think this is this is very worth watching, you know, on on a pure economics basis, right now, we definitely anticipate the world's development shift in gas assets goes from combined cycle plants meeting basically baseload and to some extent shoulder generation too, mostly speakers because you get so much cheap, variable renewable energy in the grid that that that basically sets your

merit order. But there are the times of day and the times of year in which you probably need to quickly ramp up generation and the best way to do that historically has been with an open cycle or peak

or gas plant. But if it turns out that a renewables have a much higher load carrying capability sol with storage typically than previously anticipated, then that means that you probably have fewer of those moments uh in in in the grid operation space wherein you definitely need to turn something on with a few minutes notice, especially because the battery can respond pretty much instantaneously. And two, it doesn't mean that you're necessarily having your revenue go away from

doing operations with say peaker cast plan. But I think it's our view that you you are going to compress the operational sphere of that plant from hundreds of hours a year or two maybe dozens of hours a year. I believe that that's going to be a challenging development proposition. Why any hours per year? Well, that's that's another fair question is at what point is this not viewed as necessary at all? And I think again you're looking at you have to think about the where the decision making

is distributed. So the development the wind and solar and predicting the solar development decision making is highly distributed amongst hundreds of different players. There's relatively smaller number of players doing the same on gas, right, And so what I tend to think is that like you'll have a kind of you'll have a kind of high mind approach and sol we're moving much more quickly than you will developing thermal assets. So what does this mean for for solar

and wind developers? What are some of their options going forward to put more power under the grid. So I think the most basic one if you're a solar developer is that you will be operating under the assumption that you're probably going to add some quantum of energy storage to your projects in the future. I think that it would be probably a challenging proposition to say we don't plan on doing this at all anywhere in our future development life cycle. I think the second option is to

is to start considering putting storage into existing assets. And that's something that we haven't really seen quite yet because most assets are not really yet at the point in their operational lives where they might where they might be ready for that. Right, But we generally think about the

inverter the inverter replacement cycle being about ten years. We really don't have that many ten year old assets yet, at least in wind and so where in wind in the United States, we have some wind projects that are my age, but they're not big and there aren't that

many of them. We have solar projects that are ten years old, but not yet at the point in their operational life would probably make sense to replace them yet completely, but that's probably coming and I think that is where you'll find developers essentially viewing their in existing asset as an embedded option on doing something different in the future. So let's stick into that a bit. So just to clarify, you're not saying adds storage or battery to an old asset.

You're saying effectively repower or tear down and rebuild an existing asset. Is that right, I'm saying a bit of both. There's definitely the possibility for the former. I'm going to go to my hundred megawatt project in the Central Valley of California and add I'm gonna add a four hundred meg one hour battery. Do it? Okay, that's one possibility.

But the other is to look across that site essentially look at the package of permitted land and interconnection agreement and off take agreement a solar resource, and look at the grid of the future, what's it going to be in five ten years time and say, you know what, I could probably make more money by ripping up what I have here and building something different. I'm probably going

to generate more power. I also have a load that I can shape better with a battery, may be able to enhance my load care and capability by adding a battery. So maybe it's time to sort of re envision this, this asset that I have. This is an idea that

I've been I've been pushing. I think it's fair to say, pretty pretty hard in the last couple of months, is that we should probably start thinking about assets as they approach the end of their life, well before they approach the end of their life, so to speak, we should start thinking what would I do with this right now? Knowing what I knowing the tech knowlogy at hand, and knowing the market, how does a ten year old solar project or a fifteen year old or twenty year old

wind project look like? What else can I do with this? You mentioned this in your latest op ed that you said with an old wind project, you just let it run like an old used car. Is that right? Until it start to fall apart and then you think about replacing it. There's certainly, certainly that's been that's been the case. You you work it hard, right like you you probably you're maximizing your internal rate of return out of the

project for the sake of your investors. But there may come a time when sort of running at ragged is no longer the best way to generate a return. Right, you may find that you're able to make more money on a returns basis, on a cash flow basis by doing something different. And certainly there are reasons, you know, in the next couple of years to be to be repowering a wind project for financial reasons like getting the

production tax credit again and things like that. But also just think that's be clear that the technology is just much much better. Hey, can you explain some of that.

You mentioned some of the numbers there that were just kind of jaw dropping, certainly, so like like if you if you look at a project in another project in California, there was a wind farm not far from Palm Springs, that is that is in the process of being of being repowered, and it's set to be replacing four hundred and sixty turbines and in order to doing like from like replacement on an energy capacity basis to get the thirty megawatts about but that comes from these four hundred

and sixty turbines. The new to the developer was looking at building at most eleven new turbines or sixty four eleven swap uh. And that's because well, look, the new ones are bigger, much much bigger. You couldn't put four hundred and sixty new turbines on the same plot as Warrington sixty old turbines. There's always other issues you have to look at, like how much more generation would you

be allowed to integrate? Like would you be about to enhance your your your connection to the transmission grid and therefore put in a lot more power, which case you might build more. But this just gives an indication that you know, the new technology is better. Definitely you will generate probably more power from these new assets, um though you may not be able to get more peak power

out of it given the great connection issues. But there's also other stuff, I mean, like they're just going to perform better, Like a fourty year old winter turbine generator has probably got zero semiconductors in it, whereas a new one was going to have dozens of sensors. It's going to have all kinds of sophisticated telemetry and communication capabilities. It's going to be far easier to maintain in a predictive maintenance kind of fashion. Uh, it's it's going to

be a power quality. It's probably less and likely to surge power at times. And so you know, you're, you're, you're, you start to view this wind turbine project as an embedded in an embedded option. As I said before, like the asset isn't the forgency turbines, it's their grid connection, the site, the set of options that you can then do with that within your your planning and permission capability. So what are some of the other options that you

might have besides repowering. Besides repowering, Well, you could just run it. You can run it into the ground. I've certainly seen older solar projects in California that are that are really run quite ragged right there. There's there's there's

little reason to make incremental improvements over time. You can do something else complete with the site, though, I would have to say, and I'm sure you're familiar with this from from not only driving in the West and seeing these wind sites, but also for that that are going to geo thermal sites, is whatever was within human capability to identify as a good site four decades ago is a great site to do stuff. So you would probably not go to the San Gorgonio Pass with this wind

project is and decide to make it a dissolber. It's like, if it was within the rather rustic capabilities of developers forty five years ago to identify as a good site, then by today's standards, it would be tremendous. So you probably would want to keep doing the same thing. I mean, I don't know, Mark, I would actually throw this question back to you from a from a geothermal old hands perspective, like, how good are the resources in uh forty to fifty

year old geothermal site now? Fantastic? Exactly? So you would you you would be hard pressed to do any kind of substitution. I think you're overwhelmingly likely to do the same kind of asset there. But I will add one wrinkle of that. If you're gonna replace four ins in the sixty turbines with eleven, you are going to have a lot of area on site that you could fill up with other stuff matteries for instance, And you wouldn't even have to necessarily have it be a battery integrated

to that project. It could be you know, it could be done by a different developer, It could be done by the same developer not be attached to this asset. You could also fill up a lot of that land with solar, again very California partlots, but it's considered disturbed land, meaning it's already been messed with essentially, So given that, why not when I treat some of that asset as

space to fill in with with other technologies. Again, I think that we to to an extent, actually we may almost over analyze the technicals here rather than view this the way that the real estate developer. It's a site, What am I going to do with it? You know? What is what is today's taste to dictate is for real estate is the equivalent for the creative, Like what

does tomorrow's power market necessitate? And you mentioned in the outbed that you know, it might not be a power generation asset that you would use with the extra space that you have, but you might put a demand center there, so like a server farm or something that would eat up a lot of electricity. There is also that, right, So what if I've got the ability to build more power and the grid wants to take I should probably start looking for somebody who wants that power at my

marginal cost of production. So yeah, maybe it's a server farm. It could be cryptocurrency mining. I think, and I know that you and I have talked about this before in other in other context, if not here on the podcast. If you give people very very cheap electricity, they will figure out something to do with it. Like right, I think again, we we may perhaps over analyze, well, you know,

this industry uses this kind of cheap electricity. The classic case would be, well, you you you do aluminum smell thing in Iceland because you want to make it clean, and there's a lot of cheap geothermal available there. But I think we should turn it around and just say, if you provide electricity of this type of this quality,

what will people do with it? So rather than saying, oh, this industry can't use capacity factor variable wind power, even if it is only I don't know, fifteen dollars a Mega one hour, and think instead about what industries would do. Something who exists that would say take fifteen dollar Omega one hour wind anytime they can get it. A classic example that we've been looking at quite a bit would

be hydrogen production using an electrolyzer. You may not want to run an entirety of a mission critical data center on it knowing that there's a variability. But if you're talking about running an electrocalizer where you have potentially a much higher sort of appetite for variability than great, you know, we need to start thinking about this as a feature rather than a bug like variability and ultra low cost or the ultralow cost offsets the variability to make a

price point. That's a feature for somebody, and we should keep an open mind able what that might be. Okay, So now you've got me excited about going to find an old solar wind project and terry down and billing it new, or finding something to do with the power. So yesterday we came out with our Portfolio Hunters Report where we talk about operational assets that are changing hands. I would expect that as time goes on, we'll see more and more of these older assets exchanging hands, given

that companies to see the optionality in them. Is that what we're seeing or or what do you see? So I agree with your assessment about the future. That is not what we're seeing at the moment. So last year there were just over ten giga wats of operational PV projects that changed hands new buyers, which is down thirty from and the investment volume sold was about thirteen million

dollars was down thirty seven. So definitely in the last year we didn't we didn't see that, but a lot of these things we had, like having of the market the United States which had which had previously been very, very huge, this is still chunky. It's still early days considering how much somewhere has been installed in later years. Right. What we're also seeing that was that the stages of acquisition are becoming a lot and as apparent. This is

actually very interesting to see. There's there's historically been a kind of classical path from man and cowboy hat wandering around a desert to slightly bigger developer to more institutionally capitalized developer to eventually an institutional investor owning the asset. We're starting to see some things actually be a little bit more flexible there. So the the kind of distinction between the primary and the secondary markets is becoming less apparent.

You get meeting that, you get investors who are seeking assets at earlier stages in their life. Um that can that can improve margins and it can reduce competition. They can also allow you to retain some flexibility when you're

signing an eventual power purchase agreement. That's interesting. So they're seeking young assets rather than old assets for the optional and they're seeking the stable return right which which is which is which is well, they're seeking young younger, and they're seeking assets earlier in the development cycle uh to be acquired, meaning that they have generally speaking more confidence in the market or potentially more confidence you're only stage

developer themselves. We're also seeing that yields are are coming in, so the target deals have decreased, which is generally speaking a good sign, and also again assigned it more institutional capital. That's that means that you've got lead stage, made stage money willing to play from lower from lower yields than say that that aforementioned person in cowboy hat and boots walking around in the desert. Not thanks for coming in,

Thanks for having me, Mark, It's always a pleasure. Bloomberginia is a service provided by Bloomberg Finance LP and its affiliates. This recording does not constitute, nor it should it be construed as investment advice, investment recommendations, or recommendation as to an investment or other strategy Bloombergunia should not be considered as information sufficient upon which to base an investment decision.

Neither Bloomberg Finance LP nor any of its affiliates makes any representation or warranty as to the accuracy or completeness of the information and contained in this recording, and any liability as a result of this recording is expressly disclaimed