The device throttling the world’s electrified future: Bottlenecks Series
33:16 Share

This is an iHeart Podcast.

Welcome to Xero. I'm Akshat Rati. This week, bottlenecks.

Today, we're going to start a new series, which we are calling Bottlenecks. And the first episode is about why we are working on this series for Bloomberg Green and about the first story that I wrote for it. To help me set out the series, Xero's producer, Oscar Boyd, will be asking me the questions instead. Welcome, Oscar. Hello, Akshat. Thank you for lending me your chair for today. So for as long as I've known you, you've really nerded out over electrification and its role in decarbonizing everything around us. I

I remember back to one of the first ever episodes we made of Zero together, and that was you trying to explain to me the inner workings of a battery at some atomic level. And I remember first day of the job thinking,

I've bitten off way, way, way more than I can chew here. And I think I can safely say that your interest in electrification has only grown as it's become more and more apparent that electrification is the way forward for so much of the cleantech story. And now you and other colleagues at Bloomberg Green have started this series called Botanx. Do you want to start just by explaining what the series is about?

So it's not like it's a new thing. Electrify everything has been a mantra that's been used by many climate people for quite a few years. And there's a really good reason for why. Because when you use electricity and you use electrical devices, you're typically very efficient at converting energy into productive outcome. Whether that's moving your car, heating your home, powering a data center, which is only something you can do with electricity, but you can do it very efficiently.

And you have to decarbonize electricity. So if you can electrify the economy and have all these renewable energy sources, zero carbon sources that are becoming cheaper and more widely acceptable happen simultaneously, you start to decarbonize the world. And that's a really, really crucial way in which we can solve majority of the climate problem. And we aren't doing both of those things.

But break it down to me. Maybe we can start first with the decarbonizing of the electricity sector. How are we doing in regards to that? So this is something we've talked about on the show so many times. In 2024, the world spent $2.1 trillion on clean energy and the energy transition. The vast majority of that money went into decarbonizing electricity. That's building huge solar plants and wind farms,

But also strengthening the grid, because if you don't do that, we can get things like the Spanish blackout. And sure, the politics have shifted a little bit in 2025 with Donald Trump in the White House. And so we might see a slowdown in places like the US where policies will be rolled back or even in places like the UK where an offshore wind farm was cancelled because it was too expensive to build.

But overall, we do know that decarbonizing electricity is on a momentum that will continue. It's not the same on electrifying everything. And when we're saying electrify everything, that's really like the objects that we use in our everyday life, the objects industry use as well. Things like electrification.

electric cars to replace fossil fuel engines, things like electric arc furnaces to replace blast furnaces, which would typically be powered by coal, things like that. That's what we're talking about when we say electrify everything. Yeah, very much so. But also things that aren't typically on people's horizons. So electrifying heating,

because a lot of energy that is used not just in homes through heat pumps, but in industry to run industrial processes requires a very hot steam, for example. And you could now build electric heaters that are essentially very big heat pumps to do that. Or you could start to electrify planes. Initially, planes going over short distances, but eventually even longer distances could

could be electrified. So there's a whole host of things that we do know we can move away from burning stuff to using clean electrons. On the electrify everything front, you've said that electrical devices are more efficient, they're better at converting energy into the final output you actually want.

And normally efficiency ultimately eventually leads to things being cheaper. And if they're cheaper, you think people would actually want these things. Also, you get the nice side effect of no pollution. So surely case made, job done. Why aren't these things taking off? Exactly. This is something I've scratched my head over so many times. There's this fantastic ad.

ad that was put out by one of the electrical appliance makers, I think sometime in the last century, where you see people going about an office or a home using devices which we use with electricity, but instead having like a little internal combustion engine. So when you're making coffee, you actually kick on a little engine that's burning fuel, that's producing pollution, and you're getting your coffee on the side. You're doing the same thing for your printer. You're doing the same thing to run your computer.

It's bizarre, right? We want electricity because it's just a convenient way of using energy. You're not having to look after a two-stroke diesel engine every time you want to use your printer. Right. Not having enough mechanics to fix your things. But you're right. If that is the case, we want convenience and these devices are efficient, which means they ought to be saving money. Why aren't we electrifying everything at pace?

And the answer is complicated. On the one hand, electricity itself is a much more complicated source of energy to manage. It's difficult to understand how it moves and how it fails when it does, right? We've had these two blackouts recently, one at Heathrow, one in Spain, and we still don't know weeks on.

What exactly happened in those places? So it's complicated. It's difficult to store. We know that energy storage is such a crucial aspect of the energy transition that hasn't been solved. And pricing of electricity, sadly, still remains a real challenge.

So many places like here in the UK, electricity is far too expensive to actually make the efficiency also make it cheaper. So all these combinations, which are also bottlenecks, need to be sorted to make the economic case for electrifying everything really land.

Okay, so you've worked out there's all these bottlenecks, but you're not the first person to discover these, surely? So we know that bottlenecks to electrifying, there are some real physical challenges that we talked about, but also just bureaucratic challenges, right? Getting permissions to build solar plants or transmission lines is getting harder and harder in most countries, not just in Western rich countries.

We know that building these electrical infrastructure requires upfront capital, which isn't always available, especially when interest rates are very high or the stock market is doing funny things as it has done over the past few months. Those are things that people have identified as bottlenecks that need to be resolved. They're complex solutions to them. But there are all these other little ones, or perhaps oddities,

Little seen ones that are also holding back electrification. Things like making a particular device just in the sheer numbers that we need to build them today, but there's not enough manufacturing capacity or training the people that we need to be able to go out and build this infrastructure. And so this series is about the less obvious bottlenecks to electrifying everything.

And I suppose there's also quite a lot of inertia in the system, right? For quite a few decades, you know, electricity has been basically plateaued in most countries. It's not been seen as this exciting sector of the economy that we need more money and time and skills into. So things haven't been built for a while. That's particularly true in Western economies, in the US, in Europe. In the US, electricity demand has been pretty much flat, maybe slightly rising over the past three decades. In Europe, it's actually been declining.

But places like India and China, which have seen 5, 10% annual electricity growth, those guys face very different bottlenecks. Their challenges aren't the same ones as the Europeans and the Americans face. And now, with the rise of these devices, from electric cars to heat pumps to data centers...

and this movement to electrify everything, finally, US and Europe are seeing electricity demand rise quickly. And that is also creating bottlenecks. So let's move us on to the subject of today's episode, our first episode of the Bottleneck Series on Xero, where we'll be discussing an article that you recently wrote and published. Do you want to introduce the subject for us? So I started looking at these things called transformers. They're devices on the grid that are crucial to make the grid work.

And I'd been working on figuring out just why are they bottlenecks? What is holding them back? Why are people in the industry worried about it, but nobody in the public knows? And then about a week before I was supposed to publish the article about this bottleneck,

There was a serious outage in London near the Heathrow Airport that brought down the airport for nearly 24 hours, caused £60 million worth of loss for a single device that blew up, which was a transformer. I had my notifications blow up because there were all these people I was talking to about transformers and they were saying, hey, have you seen this? Hey, have you seen this?

And it's a coincidence that I wasn't expecting, but it kind of makes the point. Transformers are a bottleneck to maintaining electricity, not just growing the grid. So Transformers was suddenly huge international news because, of course, Heathrow is the UK's largest airport. It's one of the largest airports in Europe. I think a thousand flights were cancelled that day and at least 100,000 passengers were left stranded there.

And suddenly everyone was thinking about Transformers, as you say, this incredibly important bit of electrical infrastructure. But before we continue, before we get too deep into this, when you say Transformers, the first thing that pops into my mind is Optimus Prime doing his battle against the Decepticons. Everyone, everyone! That would make for a very different discussion to the one we're about to have.

So when you're talking about transformers, what exactly do you mean? Can you paint a little bit of a picture of the size, the scale, the thing that we are talking about? Yeah, let me take even one more step back because the reason we use transformers is that we are all using alternating current.

This is AC. This is AC. And it is something we don't think about on a day-to-day basis. But about a century and a bit ago, there was actually a war of currents between the famous inventor Thomas Edison and a less famous but pretty important inventor, George Westinghouse.

And they were going head to head. It was actually Edison who wanted DC, direct current. And it was Westinghouse who was pushing AC. We live in Westinghouse's world, even though Edison is the more famous one.

And the reason that AC went out is because at that time, the technology to move power over long distances could only be pulled off using alternating current. And the reason that could happen is because of transformers. Transformers are these devices that convert the voltage that an electricity travels at.

Either they increase it or they decrease it. Now, voltage is a very weird concept to get your head around, but the best way to think about it is to think of a waterfall. The height of the waterfall is the voltage at which the electricity is traveling, and the amount of water falling through it is the current that is going through the cable at the moment.

And so current and voltage are very important characteristics of electricity. They are inversely proportional, which means if you increase the voltage, you can decrease the current or vice versa. And when you're traveling long distances, you actually want electricity to travel with low currents. So there's low resistance. So there's low losses. But at high voltage, that's what is possible. And transformers are the objects that make that possible.

And the reason this is important is because the way the electricity system works is that at different points in the electricity system, you want different voltages because different voltages allow those specific parts of the system to work at their optimum. So at home, in the UK at least, we want 230 volts higher than that and all your computers, your iPhones, etc. would probably fry.

Bingo. At power plants, like a gas power plant, you're producing electricity at a few thousand volts, at a solar plant at a few hundred volts.

But that's in a way too low to actually move at long distances. So right at the power plant, there's a big transformer. It's called a step-up transformer that takes it from a few hundred or a thousand volts to maybe even hundreds of thousands of volts. And this is to send it then through the cables that we think of as transmission lines. Those big pylons that you see when you're traveling outside on a highway.

they'll take that power hundreds of kilometers away at that high voltage while losing very little of the energy that electricity is carrying at that moment. Let's just pause there for a second. If that electricity going through the cables was at a much lower voltage and a higher current, what would happen? So that is the problem that Edison faced. If you had direct current at that time, there was no way to increase the voltage of direct current to a very high level.

you could only actually take it a few miles out. All your power will be lost just in the transporting of that electricity to that distance. AC overcomes that problem and makes it possible, not just hundreds of kilometers,

but thousands of kilometers. But these days we have HVDC cables, so high voltage direct current cables that run a thousand kilometers under the sea. They exist. Do they not overcome this problem? Yes, they do. And in fact, we have come around to what Thomas Edison wanted to happen. And in fact, HVDC is even more efficient than

than AC. But I think we should save that for another discussion because that gets into a weird territory of technical points about why exactly the technology to make high voltage direct current took so long to be developed. Okay, so today we'll park HVDC cables to the side and stick on transformers.

So you said there were step-up transformers, which increase the voltage, and then you've got step-down transformers. So what does that do? A step-down transformer does what it says on the tin. It lowers the voltage because after electricity has travelled hundreds of kilometres and

at hundreds of thousands of volts, it needs to be brought down to the level that it could be used in electrical devices on a day-to-day basis, which is typically 100, 200 volts, depending on where you are in the world. Tell me, what does a transformer look like? Oh, it's like weird shaped.

and all size-shaped. So it could be as small as a trash can-sized device, which is hung up on a pole outside your home. Many people see it, never realize it's a transformer. Or it could be the size of a shipping container and so heavy that you need not just one truck, but several trucks and several hundred tires to be able to carry such a device. So transformers look...

weird metal boxes with weird things sticking out of them of all kinds of sizes. And I can guarantee you if you have walked in a city, any city in the world, you have at one point passed a transformer without realizing that there was one. We know in numbers, at least in the US, there are something like 80 million small transformers, which means one for every four or five people in America.

There are also transformers of crazy number of varieties. So one estimate from the US Department of Energy was there 80,000 models of transformers just in the US.

And I really couldn't wrap my head around it. So the nearest comparison I could find is that is as much as every unique car model ever manufactured. I mean, that's huge. 80,000 different models of Transformers. Why are there so many different varieties? I would expect, you know, if they're necessary everywhere, you would just say, OK, we need a big one, a medium one and a small one. Let's just mass produce those and we're done with it. Yeah, exactly. And this is the...

difficulty of electricity. What happens is that the grid behaves in different ways, depending on the size and the location, where power is moving from, where it's being generated, where it's being consumed. And so what ended up happening was to manage the grid, electrical engineers came up with those specific transformers needed at that node in the grid.

And it worked like magic. And so they just kept building these customized transformers for these specific places. So yes, there is some standardization on the small transformers, but on the large transformers, there's just whatever is needed at that point in the grid.

And it is really becoming a problem because if you need to have a custom built transformer, you need to have a custom maker of that transformer somewhere taking your order and working on your specific needs.

There is a movement that is happening around the industry to try and standardize the transformer, but the electricity industry, which is the other thing I learned reporting this story, is just really very conservative. It's actually not like the oil and gas industry, which takes on these risks and tries to shale or tries to figure out how to compress things in a more efficient way, has these large companies who want to throw money at R&D. It

electricity companies, for the most part, are either state-owned or heavily regulated by the government and are just wanting to make sure things work and there are no blackouts and they are really conservative. So if something's working like a custom-made transformer, let's keep going that way, please. How bad in terms of the timeline to get a transformer delivered has this shortage become? So you could have got a large power transformer, which is custom-made,

In a year's time, 15 months, 18 months, that's now doubled or tripled in some cases. So three to five years for a large power transformer is pretty common these days. And that's a real challenge for anybody planning a power plant. A solar power plant at most takes six months, 12 months to put together. But if a transformer can't be secured for three or five years, the energy transition gets delayed.

We'll be back with more of my conversation with Akshat Rathi after this short break. And hey, if you're enjoying this episode, please rate and review Xero on Apple Podcasts and Spotify. Your feedback really matters to the show and helps new listeners discover it. Thank you.

How can you grow your business from idea to industry leader? Bring your vision to life with smart business buying tools and technology from Amazon Business. From fast, free shipping to in-depth buying insights and automated purchase approvals, they deliver everything you need to achieve your goals. It's not easy to stand out from the crowd. Simplify how you stock up to get ahead. Go to AmazonBusiness.com for support.

You say they're everywhere. They're crucial to maintaining our entire electricity system. And now we have this huge demand for electricity. Surely that's prime conditions for there to be a boom in the manufacture of transformers. Yes. So when I asked the question to industry experts, they said the difficulty has been that the transformer industry

has gone through a boom and bust cycle that has made them really conservative. So companies like GE Vanova, Hitachi Energy, these are big, giant, multi-billion dollar companies in the world that make all kinds of infrastructure. But

They saw the transformer business not be as profitable as they had expected. So around the late 2000s, there was a growth in electricity demand and transformer makers, just like now, saw that they needed to boost their manufacturing to be able to keep up with that demand and to be able to make more profits. And so they started investing in this manufacturing capacity. And of course,

then the financial crisis came and all that demand that was expected disappeared and those investments turned sour and they made a bunch of losses. So this time around, when they are seeing this electricity demand rise,

They're saying, wait a minute, maybe it's there, maybe it's not there in the future. We aren't ready yet to make the investments needed unless, of course, governments want to subsidize that. And as part of your reporting, you actually went to one of these transformer factories in the town of Stafford, which is just north of Birmingham in England.

What was it like in this factory? It's the only transformer factory left in the UK. So I was lucky at least there's one I could go to. It was a stunning place. So Stafford has a history of actually making electrical infrastructure for well over 100 years. Huge factory. I mean, you go inside and I've been inside factories, but like this, the scale of this one was just enormous.

Floor to ceiling, probably seven, eight floors, all empty space because you need to have equipment that's moving around. When we ship a large transformer on one of these transporters, we actually have police escort

All the way to Ellesmere Port. Because of the weight of them. Yeah, typically 300 tonnes, something like that, 350 tonnes. The transformers that they made at this factory are very large power transformers. They're used typically to take electricity that is coming from offshore wind farms and then brought onshore. So they're handling electricity at hundreds of thousands of volts each.

And making them is really interesting because it goes through this process that requires engineering precision for each component in the device. And we had the operations manager at GE, Varnova, Eduardo Villar, tell us it's not like making a bottle. There are thousands of parts in a transformer. How many of these big transformers are they making a year? Are we talking hundreds? Are we talking thousands? No, it's a factory that is getting expanded right now.

and is going to increase its capacity by 50%. That's good. And by the end of it, it's going to be able to make about three dozen transformers a year. So 36. That's right. So I want to try and ground this a little bit in, I guess, people's more everyday experience of transformers. You know, we have these challenges going on behind the scenes, but obviously most of us don't think about transformers and how they're used. So what are some of the real impacts of this shortage of transformers?

So Heathrow actually brought this home really well, right? If you get a transformer blowing up and you get the spare one as it happened in Heathrow's case, also going down at the same time, well, you get a blackout. That is an extreme scenario.

In most cases, what a shortage of transformers is doing is just slowing down the building of electrical infrastructure. And that's increasing costs. So your hyperscalers, as they are known these days, the data center companies or the tech companies that want data centers, just want data centers and they want it today.

But if there are no transformers, they can't get the power coming at the right voltages that they need. And they're going to have to wait, which means they're going to lose out on business or they're going to lose out to a competitor. Or they just have to pay through the nose to kick somebody out in another waiting list and get that transformer at whatever price they can. It happened earlier this year when Microsoft signed a contract to get a nuclear power plant in the US restarted.

That nuclear power plant is a giant one. It requires a giant transformer. And the company that is going to restart that plant was

was willing to pay $100 million... $100 million. ...for a transformer. So more than the cost of Heathrow going down for a day. Yes, and the cost of a typical transformer, the highest that I could ask industry expert, has been about $2 million, $3 million. So somebody is willing to pay 30 times as much just to make sure that they have the right device in the right place. And one of the things I think that was really interesting in your article is you talked about the lack of spares in the industry. So it's not just for people...

who are building new projects, it's that when things go down now, there isn't the infrastructure to replace the transformers that have broken. And this has a really interesting intersection with climate as we're seeing more extreme weather events. So in your piece, you talked about the impacts of Hurricane Helene when it hit eastern Tennessee last year and how that actually took out a bunch of transformers and they've had a really tough time replacing those. Yes, we know that extreme weather events are having all sorts of impacts on infrastructure and

And electrical infrastructure is, of course, everywhere and very exposed. We had another example that we found in Kentucky where a cyclone took down the electrical infrastructure and the transformers along with it and replacing them just took a lot longer. So the business park in the Tennessee case couldn't get power for

eight months and the houses that needed to be built in Kentucky could not get power for many months afterwards. And utilities are starting to manage that somehow by essentially hoarding transformers, having many more spares than would have been typically needed had there been not this level of extreme weather event.

Or if you can't hoard a transformer, you go begging, as we learn from utilities that some of them just go to their neighbors and be like, well, we need it now. You have some spares. Can we get them now? We'll fill your banks later. And sometimes the neighborly utility is friendly enough to give you one. So you said it's taking five years now to get a large transformer. But I assume a lot of that data was collected online.

pre-global trade war, what has been the effect of tariffs on transformers? Yes, the data is pre-tariff wars. We do know one thing, that tariffs are already making transformers more expensive because a lot of the transformers, especially in the US, 80% of the transformers are imported. And those contracts on imports are linked to the commodity price of steel because a

As soon as the price of steel goes up because you've got these tariffs, the price of a transformer goes up, the price of what the utility is paying is going up, and thus the price of electricity is going up. All sorts of downstream impacts.

Experts told me that they expect that if the tariff wars continue, we'll get a longer timeline on transformers. But it's also a dynamic system. Maybe it forces people to actually finally make the decision on building bigger factories like G.Vernova is doing here in the UK. And you mentioned steel there. That's also another problem, sourcing this very specific type of steel that's used in transformers, which is called grain-oriented electric steel. Go's?

Yeah, it's a very funny acronym and there are quite a few of those in the electrical world. But steel in general is what transformers need. Obviously the steel on the outside, that is what most people see, that's pretty standard steel. You can get it wherever you want, but the steel that goes inside is a very specific kind. So yes, transformers are all different sizes from trash cans to shipping containers, but the components inside the transformer, each of them are pretty standard.

So you have a core which is made out of this grain-oriented steel and there is a global shortage of enough people making that type of steel. There are two sets of copper cables that are attached on this core. So you need copper, you need steel, and you need them to be built in a very specific way for the need of the grid.

So I imagine one way we could get around the shortage would be, hopefully, technological improvements. Are we seeing many developments in the transformer space when it

it comes to improving their efficiency or the amount of steel you need or copper or anything? So the core technology for transformers is a hundred years old. There have been small developments that have made it more efficient over the years, but there have been all things on the edges. There hasn't been any fundamental invention that has become commercial on transformers. That is not to say those don't exist. There are all these ideas where you could get a

transformer that could be trucked to a site when it's needed and it's not custom built for that location but it could be adjusted to meet the needs of that specific spot or there are these new types of transformers called solid state transformers which are much more compact that also do not need to be custom built it could be programmed to work for the voltages that are needed on the grid and

But those are all things that are technologies that haven't yet been commercialized because the industry that uses these transformers are very conservative about trying new technology. But again, this kind of shortage is exactly the point that'll cause enough stress on the system and get people moving to perhaps try some of these technologies out. That's what Benjamin Boucher, a senior analyst for Wood Mackenzie told me, that it's going to be a long battle.

Technology is not going to come to save us in the shortages that we are facing today. It's just going to take years, probably five years before we see any of this bottleneck actually be resolved. But these are crucial parts of our energy infrastructure.

And people are not going to be happy if there are widespread blackouts across their cities, their countries. Isn't it in the interest of governments to step in to get things rolling here? Yeah, and the industry has been trying to get governments to step up. Not all places are willing. As we know, government budgets are stretched. They're cutting down aid budgets in Western economies.

But a couple of countries that we found in our reporting have shown how this could be done. So Germany is underwriting many of the transformer purchases by telling the makers of transformers that, look, increase your manufacturing capacity.

We'll make sure if this utility here in Germany isn't buying it, we, the government, will buy your transformer. So that certainty gives them the ability to actually plan out. The other one is Canada, which is also supporting just manufacturing sites. That's allowed Hitachi Energy to actually expand a Canadian factory while it had to shut down one in America. So governments can come into the fray

But reporting the story, it's just that transformers are these boring devices nobody thinks about unless somebody raises the alarm. And I just don't think that alarm has risen high enough at the level that governments all around the world are waking up to what is a real, real challenge. And those are Western countries. I want to bring into the room China because China is making leaps and bounds in its electrification industry.

Are they facing the same problems or have they factored in the need for transformers into their electricity boom? So this is the story of places which have seen 5 to 10% electricity growth on an annual basis that they don't see the same level of disruption or difficulty that Western economies are seeing. So in India's case, yes, there has been a little increase in the delivery times for transformers. It's nothing like the tripling that we saw in the US, for example. In

In China, there is no shortage. It's the country that is making enough of them to export to other countries. And if Western economies want to catch up with China on the race to build AI, on the race to electrify their auto sector, they're going to have to step up on the transformer game. Akshat, thank you very much. Thank you. And thank you for listening to Xero.

This is the first episode in the bottleneck series. Next, we'll look at the shortage of skilled workers. Look out for that episode in your feed. And now for the sound of the week. That's the sound of the inrush current into a 40 megavolt ampere transformer. Or is it the sound of Optimus Prime summoning the Autobots? You decide.

If you like this episode, please take a moment to rate and review the show on Apple Podcasts or Spotify. Share this episode with a friend or with someone whose kid is obsessed with Transformers. This episode was hosted and produced by Oscar Boyd. Bloomberg's head of podcast is Sage Bowman and head of talk is Brendan Newnham. Our theme music is composed by Wonderly. Special thanks to Jessica Beck, Somersadi, Moses Andim and Siobhan Wagner.

Thanks also to all the reporters and editors from Bloomberg News who contributed to the feature story. Noreen Malik, Tiffany Choi, Olivia Rudgaard, Mark Chediak, Dan Moutar, Somnath Bhatt, Jody Mexen, Emily Buzo, Aaron Rotkoff, and Amanda Coulson-Hurley. I'm Akshat Rati, back soon.