This is Dana Perkins and you're listening to Switched on the b and EF podcast. Earlier this year, we spoke with my colleague sun Jeet and Meredith from BNEF about one of the most overlooked areas of the energy sector, that's grids, and during the conversations, Suenjeit briefly mentioned a piece of equipment that is essential for the grid to function. Those are transformers. Transformers come in all shapes and sizes, as we're going to find out in today's episode, and
they serve a wide variety of purposes. However, large power transformers are lpts for short, because after all, everyone loves an acronym, are integral for regulating the supply of electricity that runs through the grid. And with gigawatts of renewable energy additions set to be added to the US energy makeup, as well as even more units needing replacing at their end of life, well we're just going to simply need
more transformers. To put this in perspective, under BNF's Economic Transition scenario, which is part of our annual New Energy Outlook exercise, we foresee the US day demand for large power transformers to reach over thirteen hundred units by the end of the decade. That's up by forty seven percent when compared with twenty twenty two. So what can be done to ensure the US grid has the required amount
of transformers to reliably function. To take a closer look at transformers, on today's show, I get to speak with a member of beanuf's modeling team, Ava Gonzalez Isla. Together, we discuss the makeup of lpts and in particular the very specific type of steel that is required for their manufacturing, which is so important that the Department of Energy in
the US rates it as a near critical material. We also discuss the lack of domestic production capacity in the US to manufacture large power transformers and which countries are making up the shortfall, And we go through the tariffs that have hampered the domestic production capabilities in the US and whether they have the ability to meet the demand for transformers moving forward or will continue to rely on imports.
As always, if you like the show, if you subscribe, you'll receive an update when future shows are published, and if you give us a review then you'll make us more discoverable by others. But right now, let's jump straight into my conversation about transformers with Ava. Ava, thank you very much for joining today.
We thank you for having me. I mean, been listening to a lot of shows, so it's nice to be here.
Ava used to.
Help produce Switched On, so has picked many an episode that people have listened to. But now we have you here as a guest, and actually I want to start today by talking a little bit about you specifically. But let's talk actually first about the sort of work that you do for BNOF because we have a lot of different roles in people who write different types of research here. But you have a very strong numerical background and that you've done modeling for us, but then also write long
form research for us. Talk a little bit about yourself and the expertise that you bring.
I joined the moderning tame on. What we usually do is we build Excel tools.
And we publish them online so that clients can see how we are calculating our net present values and a lot of stuff. They can challenge our sanctions, they can change the default values that we provide. So yeah, we provide advances of very useful tools on the web, and I've also built some models for other teams and then written a long report on the back of that.
So how did it come about that we started focusing on transformers in a serious way? How did this land on your desk?
Clients have been asking people in the US for all our forecasts of solar and if we were taken into consideration transformer bottlenecks, and we didn't know the answer. So they were like, we should probably answer this question, and that's what I That's what I did.
So you've researched transformers and you know about transformers, because let's rewind the clock in your life a little bit. You actually have experience working in a transformer production facility, which is not the typical thing from a being a f analyst. So it's nice to bring the real into the research. Tell us a little bit about what that experience was like.
Sure, So I worked in the assembly line in a facility in the middle of Spain close to Almagro, where I'm from, for six weeks, just assembling very small transformers, so these ones would go inside swimming pools or maybe inside lifts. Yeah, most people that are listening don't know about there's transformers everywhere on electronics and also on the power system, as we'll say later.
So it was very nice to work in a factory.
So the transformers that we're here to talk about today are very large transformers, yes, specifically used within the grid. So tell us a little bit about the transformers we're specifically going to talk about in the context of the US, and we'll get to the supply constraints. What are the transformers, what are they called, and actually, really importantly, why are they such a critical part of clean energy rollout?
So transformers are pretty much everywhere on the grid because the most efficient way to transport power is at high voltages. So the transformer is just a pecific way that would convert the power from low voltage too high so that we can transport it during great distances with very low power losses. So it's about system efficiency and low losses basically. And you have different voltages in the line, so every time you change voltage, you need a transformer. So yeah,
they are pretty much everywhere. In the report, I focus off on the very large ones, as you said, so the ones that will go between a power plant and the high voltage transmission line or at the substations that would convert the voltage down to more manageable voltages for consumption.
Everything is made of something, certainly, that's why b and EF covers commodities. It is the beginning of everything you end up seeing in the transition, and this is no exception. So what are some of the critical materials that are required in order to create a transformer?
Yeah, that's a great question. So transformer is basically made of commodities. The most important party is electrical still, which is part of the core, and I guess you'll want to dig into that later because it was listed at one of the nearer critical materials by the US Department of Energy in July. But it also has a lot of copper and steel and insulation materials. So yeah, the price of the transformer is massively influenced by the price of these commodities, obviously.
So let's have a bit of a vocabulary lesson then here, because within steel, there are three different types that go into this transformer. Yeah, so this is the goes or as it goes.
Yeah, Yeah, I call it goes.
The nose and the amorphous steel.
And the amorphous stel.
Yeah.
But I think it would be really well helpful for me and hopefully for our listeners. To better understand what these three different types of steel that make up the overall steel requirements for transformers. So let's start with the goes the goes what is a stand for? And yeah, where does it come from? And let's go into that.
Yeah, so goes On knows they're both flat rolled product, which means they're produced us seeing like hot and cold processes that make it a role like aluminium foil, but obviously not aluminium foil, because this is something different is still and it has a small percentage of silicon. And then with the ghost, there's an additional process called a kneeling to orient the particles so that the magnetic field
inside the transformer can be can be better transmitted. Basically, so the ghost and from grain oriented electrical steel, and it refers to these particles that are oriented and that can transport the magnetic field so that the transformer works better. In the nose, it is non oriented electrical steel, so the particles are not oriented obviously, and they don't work as good in the transformers because the losses are like three to five times bigger because of this lack of
particle orientation. So yeah, all the ghosts demand is basically transformers, and because the large ones, there's such a big part of it, that is the core that like, yeah, large power transformers basically make the demand for ghosts. Nose is also is used in all types of motors, electric motors and moving parts, so there's a big demand rising due to IVY uptake in general, but it's also used in power generators and yeah, any moving motor basically.
And then there's a MorphOS steel. And I know what the word of morphus means, but what is in a MorphOS steel?
So amorphos steel is a very niche, niche material actually, but it has some properties that can be used in very small transformers. And in fact, the US is thinking about a standard that would instead of using goos or high permeability goes for some of the transformers, they want to use amorpho steel because it's slightly more efficient than the ghost. But that of course means that then the
transformer designs have to change. We think that in the US there's like twelve million metric tons produced per year of gooss. Well there's like forty five thousand produce of amorphous steel. So obviously, well, if you want a lot of transformers to use these materials, there's a big need for scale.
Also, let's talk about among these, which one is in or do we think will be in the future. In the scarces supply where we're seeing a real gap in terms of ramping up supply for this increased demand. Is it goes, the nose, Morphy steel are all of the above.
The reason why we talk about nose is because the process, the fabrication process is super similar with ghosts, so you could upgrade a nose factory into goes and make it useable in transformers. You cannot use noses in transformers as it is because it has bigger losses and nobody wants bigger losses in the power system. That's why we have transformers in the first place. So the good thing about goes is that in the flatterol product you can then stack them and produce.
Huge transformers with that.
While I'm morpho, still you cannot stack it, so that's why you're using it for a very small transformer. There's different classes of goals as well, and we want the super high quality, high permeability ones because that's the one that is going to have lower losses, and there's been a high supply issue of goals in the US for a very long time because they only produce yeah, two
million tons a year or something like that. So there used to be two domestic players of goals in the US, and then the US decided to set up a new standard so that the goals would be of higher permeability, and that led to the closing of one facility and then more imports of this high quality material instead of producing domestically. To be fair, producing it is very expensive and very hard, and there's a very high failure rate and low yielding. So it's an expertise that not many
companies have. But there's very little domestic capacity in the US.
So I would love to know where in the world transformers are manufactured and really where they're coming from. Is this increasingly a domestic market in the US, or is it offshore or near shore?
Well, we do think that domestic manufacturers in the US are increasing their capacity slowly because there's in a surge in domestic demand, so they're coming more and more from the US, But historically they've run at fourty percent capacity, and the import penetration rate for these particular transformers round it, so mostly from abroad.
Today, the biggest exporters.
To the US are Mexico and Canada, and that's because labor can be cheaper. There's been less uncertainties about policies, trade restrictions, and import standards. There's been, honestly, so much going on since twenty ten to today that manufacturers felt that they had more of an incertentive to invest in Mexico and to invest domestically.
And then historically they haven't come from any of these places. They had been imported from quite far away. What has historically been the transformer market. And really when did we see it change to being a North American market servicing North America versus imported on.
A ship from far away.
Sure, so there was a big part of large power transformers that were imported from South Korea, and the US started some investigations on these, which led to at least humanae power transformers may building up in the States, And right after they approved the anti dumping order, humandised production just substituted the South Korean imports with their local production.
So in that sense, I guess that order worked, but that doesn't mean that it made it easier for local players, because, as I said, capacity was still forty percent, and imports were still very high. And Yeah, in fact, some plants closed in twenty seventeen so well after this anti dumping order. So it's just been a tough place and with many changes year after year.
So historically transformers were coming from South Korea to the United States to meet the existing demand, which we do see and will come to this which we do see increasing in the future. Yeah, and we're seeing some changes in the last few years, not just in this industry, but in other parts of the energy value chain of production facilities moving nearer or near shoring. So in this circumstance Mexico or also Canada, with that production moving nearby,
what has that done? Has that increased decreased not only costs but also the availability of supply because you mentioned that this is actually a really tough business to be in, that there's a lot of waste, it's highly technical, and that there are some concerns around long term demand. Has it made it more expensive and has it made it more difficult to attain the transformers with the supply chains having moved a little closer to Mexico and Canada, has that been good or bad for the US?
Well? Since Mexico and US are trade freely. For the US, importing should then be a national security risk or anything like that.
However, for the ghost part of it, which is.
You know, fifty percent of the cost, a lot of the weight, Mexico and Canada do not have domestic production facilities, which means they're importing all these metal from other places. And forty six percent of the tones that Mexico imported in twenty twenty two so last year were from China, and nine percent of the tones that Knada imported of goes last year were also from China. So this is
obviously not ideal for the US. If they are really concerned about supply issues with China, the priority should be to ensure there's domestic capacity.
Can you also talk a little bit about the import tarraff that was brought into place during the Trump administration, specifically with four steel.
Yeah, Section two to two of the US Trade Expansion Act. It's a tariff of twenty five percent on steel that was applied during the Trump administration, and that applied to a lot of products still and of course because ghost.
Is a part of STEL, it also applied to ghosts.
There's a small loophole in this supply chain because what goes into transformers is basically goes but just cut with some laser machines. So it's very specific. But is the raw material. It's the same, nothing is done to it, just cut. So when these startiffs came in place, we saw a drop in importance of gosts of the raw material in a flatterol shape and an increase of just the raw material cut because it was on there. Another hts growth for imports so they wouldn't suffer the twenty
five percent tariffs and steel. So that puts some gutters in the US out of business or they force them to move to Mexico or Canada because goos is like sixty percent of the price of the lamination that goes into a transformer, and if you apply a twenty five percent of that, you're just hurting the local businesses.
Let's talk a little bit about the future demand for these very large transformers that are used for the grid. Yeah, and before we actually go into some of what our models are telling us we think the demand could be in the future, can we talk a little bit about the different things we think are impacting demand. What are the different variables that you've had to look at to try and essentially quantify what that growth will look like.
As I said earlier in the episode Transformers, I put next two generations places so that you can increase the voltage, next to demand places so that you can lower it down. New generation projects such as renewable projects, which you know are part of net zero targets, and all these things are going to be a surge in transformer demand. They all have to have them, yeah, exactly, the need they need a transformer, all of them. And then the US
greed is starting to get at old. So according to the US Department of Energy, the average say, is of a transformer is already around forty years, which is the
recommended replacement AIDS according to many manufacturers. So that means they have a lot of transformers on the grid already that are getting old and they have to replace them because if they fail and they cannot transport the grid, that's going to be critical for you know, the affected areas for either the power that is generating or the demand that is connected to that particular transformer and then system reinformer. So the bigger grid gets, the more interconnected
you need it to be. So we made some assumptions on okay, if we have to raise the demand, the generation plans by these match and the demands in terms about these moths, how many reinforcements we need to make, and how many of them will contain transformers.
So there's this convergence between wanting to create security of supply, continuing to keep the existing infrastructure up to date so that it works properly. And what we see is increased demand for new projects coming online. So what does that mean then for the future demand of transformers and where do we see it going.
So most of the demands, according to our calculations, are from replacements of an aging assets and system reinforcements. There's a big part, well you know this better than me. NA has these two scenarios, the economic transition scenario and the net zero scenario, So we basically calculated the demand under both and an economic transition scenario, most of the
demand for transformers comes from replacements and system reinformers. In the net serio scenario, the proportion is slightly higher for new connected projects, which would make sense because we need a lot of renewals to get to net zero fast.
Really, simply put there isn't enough supply to reach future demand. So one would think then if you believe that there is the invisible hand of capitalism, the companies would want to go into the space and this would then be profitable. Why are there not companies jumping at the opportunity. Surely there is some growth, but why do we think there isn't enough growth in the transformer production space to meet future demand.
Yeah, that's a really good question and one I've asked myself when I was doing this research. Making a transformer takes a very long time. It takes from five to twelve months if you don't have any waiting time or an equus or anything. What we are seeing now is that the waiting time that their manufacturers are giving is
around from two to four years, sometimes five years. Obviously, project developers are seeing this and they are getting their orders in advance to make sure, you know, they can connect their solar firm or their wind.
Firem on time and these things.
Obviously, when transformer makers get some orders for like in five years time, are skeptical of like is the demand really rising or is it not?
Are they just trying to secure what they need because their backlogs in the existing system, so they're having difficulty also looking at future demand.
Yeah, exactly.
They are extremely cautious, and they see that they have a big back logo for others, and they are willing to increase their capacity from the historical forty percent to maybe one hundred slowly. They are keen to release their bottleleggs, but they not all of them, are keen to build a new transformer manufacturing plant and to just romp up production on a larger scale.
Because this is the review that once we replace the existing infrastructure, which is a known quantity, and then add the additional supply that is forecasted, then it will sort of die down and essentially there won't be as much demand until the next group of transformers gets to the end of their useful life, and that could be forty years from now. So is the review that this could be a short term spike in demand, and therefore, if you're in this business, it may not be worth.
It if you're asks develop suppliers, transformer suppliers.
That's what they will tell you, because.
I've asked them, I've talked to some of them, Yeah, and they are really skeptical. They I've told you about how tough business was. I've told you they were running a forty percent capacity. They don't want to make a huge investment for nothing, which I think is fair enough. If I were them, I would also want to be super sure that if I'm building a new factory, it's going to.
Be called presumably quite expensive.
Exactly it's going to be utilized for more than a couple of years. There's a couple of transformer manufacturers that have decided to expand operations though, so some of them are ready to invest, but most of them are proceeding very cautiously ramping up their current capacity, sorting out some of the bottlenecks, maybe bringing some automation because also these plans,
some of them are very old. We're building the nineteen fifties, nineteen sixties, so they are king to ramp up, but they don't necessarily see the need of building a new plant until they are absolutely sure that the demand is happening.
I mean, this is really interesting because the grid and different parts of the grid keep cropping up in the shows that we end up having here at BNEF and on Switched on, and we're going to be talking about both Onshore and offshore wind in the US. And one of the things that's happened with wind, particularly offshore, but also on shore, there's the timelines to get these things connected to the grid have gotten longer and longer and longer.
And you had mentioned that the delivery timelines for transformers has also extended. Has that played into the timelines that you're seeing for the developers and has the move from a transformer taking historically eighteen months to arrive to your project now taking four years. Has that been a part of that bottleneck.
So what we've seen so far is that as long as utilities placed them or they're in advance that areansformer will be on time, and for now for years, it's not a significant change on the project lifetime. If there are improvements in how the US approaches what the US and other markets actually approach their great connections and the waiting times are shortened on that end, maybe transformers could become a bottleneck, but as of today, not really.
So we've established that some of these companies are domestically in the US or in nearby countries Canada and Mexico, who are the biggest players in the transformer market.
You want know many of them because these companies, yeah, they're either they're all private companies, either part of a large conglomerate, or they've just been doing this thing for a very long time. The biggest players are they'llta start, Hitachi Energy, these ones you might know, Hiko Hendi Power Transformers, and then we have Niagara Power Transformers, Pennsylvania Transformers, Project Ge, Virginia Transformers, and Wig Transformers.
So as you get down that list, they sound quite rich. So some of these are not huge companies owned by multinationals. They're servicing a specific market and the US is probably a good part of their demand and that's where they're selling to.
Yeah, so Pennsylvania Transformers for example, or yeah, and Virginia.
I think Virginia has a plant in India as well, and a couple of plants in Mexico. But yeah, Virginia and Pennsylvania, and I think one more. They're family owned. Kitasi Energy is owned by Hitashi, and Hyumdai Power Transformers I think is owned by Hyundai Industry. Other companies that in my list are family owned or owned by their employers, so they're not all of them are part of alert international conglomerate.
Basically, some of them are quite local.
Which then you can also see why they're approaching risk differently because the capital that they have available to them is different. They have to borrow in a different way. So when it comes to policy, if I'm a policy maker, I am sitting here thinking, ah, that we must, we
need to address this. So if companies are not incentivized by the economic condition to increase supply, have there been policy interventions in order to try and put more transformers out there for some of the projects that many of them have been spurred on by the policies available in Inflation Reduction Act.
Yeah, So you might think, I mean we talked earlier about the invisible hand, and you might think that the IRA is there to help the US supply saints. And the answer is yes, the IRA is there to help the US supply saints. However, the money is more localized on very upstream sectors that are usually imported from China, So there's not really that much money available for greed
on transformers from the IRA. And well, you might think the IRA is going to increase the solar generation, is going to increase the EVS, that's going to raise the demand, and then the visible hand, local manufacturers will just raise their capacity, but we said they are so risk.
Covers, it may not happen, yeah, or it.
Might happen but later, which means that, yeah, if the US doesn't raise their capacity, imports have made eighty percent of the demand, So you know imports will have to keep up because demand is rising according to our.
Outlook, So there either needs to be something that stimulates imports or leads to more domestic supply or needless to say, there will be long delays and we may not see the degree of retrofitting or replacing existing infrastructure. Now that's
actually something that is not a brand new story. I think about the timelines for actually very well monitored facilities like nuclear facilities getting pushed out with the right certifications, or dams in the US, or even bridges, and you do have this aging infrastructure that has been pushed back
it's originally expected date. Do you anticipate that that will happen with transformers and that they will essentially get more time because the kit isn't going to necessarily be available right away.
So we do know that utilities are refurbishing some transformers to make sure you know they have a transformer on time for a particularly you could also, I mean this is bad for power losses.
You could also use.
Two smaller transformers that take less time to manufacture instead of one, but that would increase your losses. So if you needed, if you absolutely needed that transformer, there's other ways than getting in brand new.
So clearly I'm not in the market for one, but I want to know how much does a transformer cost? Not just in dollar terms, but I want to better understand whether or not this is a significant cost to the utilities themselves. And this is something that you know their sweating.
So the large power transformers, they're very big. They weigh one hundred metric tones. There's a lot of material in there. If you look at the import data, the average price of a large power transformer in twenty twenty two was over a million dollars.
And how many would you? Would you have multiple on the same location?
Usually not, so there isn't Also why they are so expensive and so niche is that you would only use the design it's design once, so they're not very standard. Every time you place another. You have to design the transformer then make it, and you'll never Yeah, it's very rare that you would use the same design twice, so if the US is important seven hundred transformers, they'll all be different.
But let's use this very nice round number of roughly a million, million, million, saying it's actually a little bit over, but we'll just go with a million. How many units are we expecting to see now and in the future.
So we estimate that the demand for transformers in twenty twenty two was nine hundred and twenty units, so.
Nine hundred and twenty million.
Yeah, around one nine hundred and twenty million dollars. And we expect the demand in our economic transition scenario to be one thousand, three hundred and fifty seven units. And if you look at the net zero scenario is over two thousand units by twenty thirty.
And do we expect there to be cost the clients? Are We feeling pretty confidently that this is pretty much what they're going to cost, at least in the medium term.
I mean, we have been manufactured, not we, but money.
People people have been manufacturing have been manufacturing transformers for a very long time, and it's pretty much a refined product already, so there's no learning rates coming for this. These are going to continue to be expensive and important pieces of kit, and it has that nice round one million number that we can put a dollar value against
the future units that are required. Well, Eva, thank you very much for sharing your views and your research on where we see transformer demand going in the future and honestly just what transformers are and why they're so critically important. Thank you for being here.
Thank you for having me.
Bloomberg ne Ef is a service provided by Bloomberg Finance LP and its affiliates. This recording does not constitute, nor should it be construed as investment advice, investment recommendations, or a rectcommendation as to an investment or other strategy. Bloomberg ne Ef 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 contained in this recording, and any liability as a result of this recording is expressly disclaimed.