Interview with Sunpower chief Tom Werner

Emerging from the solar shake-out

TRISTAN EDIS: We’ve just gone through a very difficult time for the solar sector, red ink everywhere, and it looks like we’re now emerging out of that huge overcapacity situation. What do you see going forward in terms of the shape of the sector and who are the players going to be out of this shakeout? What’s going to happen with prices? Are we going to see prices just go up and we have to forget this beautiful time when prices plummeted? Or are we going to to see prices go down again hopefully in the not too distant future?

TOM WERNER: So, the answer to your question as I see it is things turned around in early ’13 and that was largely driven by Japan and China, so you had a huge amount of demand come out from Japan and China. You also had a rationalising of supply which was prior to that, including our company.And then in ’13 those two things came together, costs came down, so you get the scale of China and Japan allowing you to continue to keep costs down and then you can start to reach grid parity. So, the answer to your question is costs will continue to come down, not driven by panels exclusively, but also driven by a balance of the system. And so, the good news is the cost of energy will continue to come down and you’ll see things like what we did in Chile where you have a merchant power plant where we’re selling through the grid at market rates.

Competing against the Chinese through a shift closer to the customer and energy management solutions

TE: I suppose the other thing is though what we’ve seen is China has emerged as an incredibly dominant competitor. They’ve blown everyone else sort of out of the water with just the ambition of the scale which has then given them really low costs in terms of module manufacture and I remember a few years back, REC Solar, and you guys and Kyocera etc. were the leaders. But now the Chinese are pushing production costs of fifty cents per watt. How the hell do you compete against those guys? How is Sunpower going to survive against guys that can mobilise such a huge amount of capital it seems at such a low cost?

TW: So, I think the key is to not stand directly in front of the competitor and compete. It’s to change the game. So, you know, two or three years ago we went into developing utility scale generation projects ourselves. We vertically integrated towards the customer and now we’re going to take that same channel strategy and we’re going to add to that services. Where things are going is in our view is solar providers will add storage and energy management and we’ll control both load and generation in time, allow our customers to match the two and optimise. So, for you the customer I’m not selling you a solar panel, I’m not selling you solar energy, I’m selling you an energy solution. So, we’re going to change the nature of the competition. And note, while we sit here, we’re expanding capacity. So, you know, we’ll be upwards almost two gigawatts a year, 1.7 gigawatts a year, so we’re very much in the game on a panel [production] capacity and panel cost. The other point is we’re technologically differentiated, so we keep piling money into R and D, so that the panel differentiation persists. And then change the nature of the strategic clients’ game.

TE: So, you guys have always been the leaders in terms of efficiency of your cells relative to everyone else. Is it a question of lower cost per unit yield? Because you guys could go and do services, let’s say. You do that really well, get rid of your panel business, just do that and let the Chinese do panels. 

TW: Yeah. The idea is that we’ll make a differentiated panel that gives us low cost solar generation, so cost per kilowatt hour, not cost per watt, and that’s the foundation to build the other services off of. So, the Trojan horse, the lead-in entry is solar energy. That’s why we can be a solution provider, because there are a hell of a lot of companies that want to be a solution provider. And what we’ll do is we’ll build that on top of what we offer our solar customer. So, we’re not going to go to the market and say whether or not you have solar we can do these services for you, it’s going to be on top of your solar.

TE: But I mean you could look like a… You could become a solar panel retailer, let’s say. You go in and you get the conversation with the customer about them wanting to install solar and say, ‘hey look what else I’ve got here’, and then you dump the manufacturing of panels. Is it that you guys have got some certain things up your sleeve that you reckon you’ll be able to manufacture a solar panel that delivers a…

TW: Well, there’s a time I think where it won’t make sense to make solar panels. I just don’t know when that is. And of course you have to predict that before it happens, so you have to have sufficient differentiation to justify the vertical integration. And because, as you pointed out earlier, it’s very capital intensive and, you know, we’re five to 10, maybe 15 years away where either we’ll licence our technology or we’ll use other people’s product as well, but it’s not imminent. 

Battery storage and energy management services

TE: With this energy management services concept, I was chatting with the head of the US National Renewable Energy Laboratory (NREL) the other day. He’s sceptical about batteries getting there in terms of the benchmarks he has in mind about what it’s going to take for it to be an attractive option, but he might not be in touch with the market. I hear other people say to me ‘oh no look we are very close in terms of price per kWh differentials between peak and off-peak rates or the price of export feed-in tariffs vs grid imports with products in the marketplace now . So, I’m hearing different stories from different people.

TW: Ask NREL for their crystalline silicon efficiency roadmap and I think we’re shipping what they’re projecting to happen in 2020… 

TE: Okay [laughing] that’s true! 

TW: It certainly was true a couple of years ago. We know the NREL guys well. We have people that work with them and they’re very smart people. The guy on the panel I was on too [at the time of the interview Werner had just been on a panel discussion at the Energy Networks Association conference with executives from energy network businesses], he was sort of like there needs to be a breakthrough, and I don’t agree. I think this is a scale game and this is a grinding down the supply chaingame, and you need to get the scale. Tesla you know…..One thing I don’t see here is… I haven’t seen a single electric car and I haven’t even heard anybody even say the phrase electric car. Whereas in California of course it’s prevalent. So, you will probably see the scaling of lithium ion batteries. That scaling, we’ll see, but should get the cost down … But I couldn’t agree more today, batteries are ridiculously expensive, two to three times [what’s economic] right. But there’s too much market pull. And it just makes sense too. Look at the structure of the way you receive electricity today. It’s just going to blow up, right. You’re going to have more control. You’ve going to use… generate where you use it. It just makes sense. There’s also a McKinsey report that explains this.

The Silicon Valley electricity revolution

TE: One last thing with this market pull theme. I get the feeling that Silicon Valley sees this problem differently to the rest of the world. It’s kind of like we’ve done this before, we’ve blown away incumbents before, we know software, we know how to build things with scale and they’re used to costs going down exponentially. Is sort of like you guys look at this and think we’re going to do it differently … it’s not like building a coal fired power station. We can get exponential reductions, we can pull it off?

TW: Yeah. I reckon that’s a great characterisation. To me it just makes sense that the idea of building these gigantic power plants with transmission lines at… and I think we’re ten years away from going, ‘remember when we used to do that?’ So, the transmission lines, what value do they add? You’re burning electricity. So if you can generate where you use, it’s just a much better model. And it turns out it’s just very hard to store an electron, but it’s not like you need long term storage either; you’re just moving around a few hours. And then the other side is this demand side, so if you can optimise a combination of demand and generation, that’s the key. So, you have more than one lever to pull here… And the lever of [appliance] automation is going to be cool. This isn’t going to be on an app where you’re having to say okay run my air conditioning now. We want to automate this. So, we want you to say here’s how I want to live and here’s what I’m willing to pay and it just happens. So, the algorithms are done for you. 

TE: This is also what I heard from NREL where they can see the US is damn good at big data, we’re learning how to analyse bucket loads of data, do it quickly, we’re going to come up with these algorithms that are looking around for the energy demand patterns and then they can do little tiny changes across lots of appliances and homes which then add up to big shifts of demand in time.

TW: Oh, for sure. There are so many cool start-ups that are taking smart meter data and can analyse the signature and tell you what you’re running in your house because you can get smart… People are saying fifteen minutes. You can get fifteen second readouts on your meter and you can then profile the house and, you know, when the microwave is running, the hot tub, etc, etc. And these are start-ups in America. There are some in Europeans as well. There are probably some here. I just haven’t been here long enough. So, the data analytics are pretty much there.