 in the entire world uses in a year. As the world races to avert catastrophic climate change, the sun offers by far the most abundant source of clean energy. And by harnessing it, countries can also power economic growth, expand electricity access, and reduce energy imports. Over the last decade, solar has gone from expensive novelty to the cheapest, fastest growing power source on earth, but here's the thing, we could squander its potential if we don't plan for the future. Today, solar supplies just 2% of the world's electricity. And even though it might look like solar power could keep on growing exponentially, its rise could very well hit a ceiling, flattening out in coming decades, far before it unseats dominant fossil fuels. In Europe, we're already seeing the slowdown. And if that happens on a global scale, the solar revolution could sputter out. Preventing that is the point of my new book, Taming the Sun. I argue that three types of innovation are needed to unlock solar's full potential. The first innovation is financial. Right now, solar needs to attract trillions of dollars to fuel its rise. But so far, the world's most deep-pocketed investors have largely sat on the sidelines. So the solar industry needs to take a page out of the playbook from the fossil fuel, automobile, and mortgage industries and bundle together solar projects. So big institutional investors feel comfortable buying and trading them. I'm pretty confident the industry will figure that one out. But just as soon as solar gets over that funding speed bump, it could run into a much more serious obstacle known as value deflation. See, all this investment will help the industry produce and deploy more solar panels, driving down the cost of building a new solar project. That's the good news. But the bad news is that the value of the electricity produced by solar will plunge even faster. As more solar panels come online, they'll flood the grid with power in the middle of the day, but shut off when the sun sets. Even though customers will need power during dinner time, the next solar panel will just feed them more lunchtime power. That's not very valuable. Soon, the value will fall below the cost. So it won't make economic sense to install any more solar panels. That will halt the momentum of solar's rise. Overcoming this barrier starts with technological innovation. Breakthroughs in solar technology could cause the cost of solar to plunge, enabling more solar to be deployed economically. Next generation technologies, such as perovskites, already exist in laboratories. They could transform today's heavy, rigid, and frankly ugly solar panels into lightweight, flexible, and colorful coatings that tomorrow could cover cityscapes around the world. Additionally, developing advanced solar thermal plants could convert the sun's energy into heat and use that heat to generate power 24-7, rather than just at lunchtime. And one day, artificial leaf technology could even harness sunlight to make portable fuels, finally making oil obsolete. Still, even with these two types of innovation, solar will need a third to limit the decline of its value as more of its deployed. And that would be systemic innovation, which includes things like continence fanning power grids that link sun-drenched deserts to power-hungry cities, energy markets that pay for energy storage and flexible generators to smooth out the volatile swings of solar power, and smart software that can turn electric vehicles into mobile batteries to resupply the grid once the sun goes down. These innovative energy systems would preserve solar's value by making sure that solar power can be used no matter when it's produced or how it fluctuates, promoting all three kinds of innovation will require urgent investments by governments all around the world. And that needs to start right now. If we wait until solar runs out of steam, it'll be too late to get it back on track. But if we get this right, the 21st century will finally be the one in which humankind secures cheap, clean, and virtually limitless energy, all by taming the sun. Thank you. Well, thanks, everybody. That was the talk. We can all go back to class. I'm going to have Aaron Sivarum from the Council on... The year has been a success, man. I had a great joke, too. You were doing great. Just sit down. I thought it was there. Thanks again. All right, so that joke felt flat. Look, I am so happy to be here. This is kind of like coming home. I see all kinds of friendly faces in the audience. My whole family and a half are here. I think I see my uncle there, my mom here, my dad there. I don't know why you guys distributed yourselves. This is like... I will talk about DG later on. I want to say thank you to John for having me. John, I'm a fan of your work. Your work is in my syllabus. I want to say thank you also to... I see Sally here. I see Chris. I see Arun. I'm so grateful to serve on the pre-court and Woods Councils. Thank you for having me. Arun, just about a year ago, you were hosting a study group session for the book here. And I got on my plane back to D.C. with this thick sheaf of papers from Arun with annotations on every page. And I went, oh, man, I got to rewrite this whole thing. So thank you. I made it through and I think the book is better for it. My mental health, though. I don't know. I wanted to talk to everybody about some of the data and graphs behind the insights that you just saw on the video because I hear Stanford brings a pretty wonky crowd. So let me walk you through what this book is about, how it's structured and what the data and modeling is that underpins some of my conclusions. And then I'd love to chat some more with you and I hear that we'll be able to do some questions afterwards and I'm also signing books after. So feel free to stop by and check out the book if you're interested. So the book Taving the Sun is divided into four sections. The first section called Playing the Long Game kinda sets the stage. It sets up this paradox. Solar has come a very great distance just in the last decade and yet I argue it has even further to go. The next three sections of the book will walk you through those three kinds of innovation that I mentioned, financial innovation, technological innovation and systemic innovation. You know, I start by talking about the abundant potential of solar energy. These two cubes give you a sense of that potential. More energy hits the earth from the sun every hour than the world uses in an entire year. The problem though as you all well know is that despite this abundant potential, solar is particularly inconvenient. In contrast to energy dense fossil fuels, solar is diffuse, it is intermittent and volatile and for 3,000 years humanity has attempted to harness the sun's energy with quite minimal success. But now we appear to be at a tipping point and countries around the world are now seeing solar as a viable way to increase the proportion of clean, secure and affordable energy in their mix. I used to see an example of India because India is really going all in for solar. My first blog post as a fellow at the Council on Foreign Relations in 2015 was why India's solar target is sheer idiocy. As it turns out, India's solar target turned out to be a stroke of genius. In 2014, Prime Minister Modi entered office and said India's not gonna target the 20 gigawatts of solar that we had originally targeted under the previous administration. We're gonna multiply that by five. We're gonna aim for 100 gigawatts. All of us said there's no way you're gonna meet this target. By the way, I just found my undergraduate advisor in the audience, Bruce Clemens. Professor, thanks for coming. Now, Professor Clemens, as you guys may know, is well known for his wordplay and so I'm going to try and sneak in a few puns that hopefully you won't be the only one that gets. So India decides it's going to dramatically ramp up its solar power and you can see that ramp. Right now, in 2018, we are less than 20% of the way toward India's target. India actually has not kept pace toward that steep ramp up. But here's the thing. Even if India does achieve this 100 gigawatt target, which everyone thought was crazy, well it's still only going to be powering about 10% of its electricity mix with solar energy. For solar to achieve all of the things that India wants and Modi has called it the ultimate energy solution to increase energy access, energy security, energy affordability, well it's gonna have to grow even more than that and unfortunately, India's power grids probably aren't equipped to handle this influx of extremely intermittent energy. So this is just one example of a case where the hopes and expectations might clash with the realities and why I fear that if solar stalls in the coming decades, we could see a lot of expectations get dashed. You know, today's euphoria over solar is reminiscent of the euphoria in the 1970s over nuclear. Even before that, you heard folks say nuclear will soon be too cheap to meter. It was the great hope for clean, cheap and abundant energy for all and yet nuclear energy hit this peak sometime in the 1990s, I think 1996, it peaked never achieving 20% of global electricity supply and has declined ever since. You may say nuclear is very different from solar so even though today's solar accounts for 2% of global electricity, right where nuclear was in the 1970s, we should not be comparing the two. But I have the sneaking suspicion that there is actually an underlying similarity between nuclear and solar. That underlying similarity between the stories is a phenomenon called technology lock-in. That phenomenon is something where a first generation clean energy technology, in nuclear it was light water reactors and solar, it's silicon solar panels, that first generation technology can actually prevent the emergence of a next generation technology, locking it out through its dominance. It happened in nuclear because we don't have advanced nuclear designs commercially deployed at great scale because light water reactors account for greater than 90% of all nuclear deployment. It's possibly the case that nuclear may never recover. In the case of solar, I desperately want to avoid this phenomenon of technology lock-in. Now, in the last several decades solar has made dramatic strides and I don't wanna discount the progress we've made. So let me tell you a little bit about how far we've come. This is that popular NREL graph that probably half the room has seen. I'm showing you the lines for various different solar materials and how the efficiency of a single solar cell has improved over time. The dominant technology, silicon, is that blue line and you see it's done well and even increased a little bit in the most recent years. But by and large, in recent decades, silicon has stayed put. It's getting close to its theoretical efficiency ceiling. That red line though, I'm gonna single out Thomas over there, he's the blonde Dutchman. Thomas is one of the leading researchers on Perovskite, a next generation solar technology that's increased faster than any other emerging solar technology. Actually mature or emerging technology. And let me tell you, I get to pick on Thomas cause we went to grad school together and he picked on me throughout grad school. Now, in addition to technology progress, we've seen a remarkable manufacturing story. China has driven the cost of solar down through manufacturing scale. Today China accounts for over 70% of solar production and increasingly China's the largest market for solar deployment in the world. China now accounts for around 50% of all solar deployment. So thanks to Chinese government policy, we've seen this boom both in solar production and then solar deployment. It hasn't always had salutary effects and we can talk in the question and answer session about whether US companies or German companies went bankrupt as a result of Chinese government largesse toward its companies. But the fact remains that China and Silicon have won the day and that has been a good thing for the near term deployment of cheap and abundant solar power. You can see the trend in costs around the world. Folks are signing long term 20 year contracts for solar power at prices that were once unheard of. If you told me a decade ago that we'd see two cents per kilowatt hour signed around the world, Latin America, the Middle East, I would tell you you are crazy and yet that's what's happened. These costs are very real, even if they are for delivery of projects in a couple years. It doesn't change the fact that solar is now the cheapest source of electricity in many parts of the world. And that's a good thing. As a result, folks project this hockey stick-like trajectory for solar's growth. You're seeing, for example, projections in India, China and the Middle East accounting for half of this growth. By 2040, Bloomberg project solar could hit 15% of global electricity and I target 33% by mid-century. But here's the problem. This hockey stick-like trajectory isn't gonna happen on current course and speed. At least that's my prediction. That's my warning. And in the book, I argued that unless we proactively invest in those three kinds of innovation, financial, technological and systemic, this trajectory could turn into an S-curve and that S-curve could be catastrophic. If we get to that S-curve, it will already be too late. We will have let solar stall, let a clean energy transition stall and not really done anything to prepare the groundwork to avoid it. I told you a little bit about value deflation. Now let me give you some numbers. These are three simulations of three different electricity grids around the world that a colleague at GTM and I compiled for Nature Energy. You can see that across Texas, Germany and California, at 15% of the electricity coming from solar in a grid, the value of that electricity plummets by 50%. At 30% of electricity in a particular market, the value plummets by 70%. And I told you that we want solar to get to 33% globally. If we're gonna get to that proportion, well, solar's gonna have to get a lot cheaper, a lot faster because its value is going to erode very quickly as it produces more and more lunchtime power, even as dinnertime power is what's in demand. Now, a lot of folks, as I've given talks like this around the country, a lot of folks have said, look, why are you so hell-bent on solar being the primary driver of decarbonization? Until last week, I had to tell them, you know, this and that reason and like I've done this modeling and whatever, now I have somebody else to talk about. Just last week, Shell released their Sky scenario. Shell, this is an oil company, guys. Shell said, hey, here's a scenario that realistically can achieve a below two degrees climate trajectory. And in that trajectory, Shell predicts that solar will have to achieve 36% of the electricity mixed by 2050 and 62% of final energy demand by 2100. That's not just electricity, that is global total final energy consumption for humanity. If I and an oil company can agree on something and it happens to be the primacy of solar toward meeting our climate goals, maybe we're on to something. Now, this graph should more visually demonstrate to you why value deflation happens. This is our state, California. I get to say it's our state, right? Even though I live over in that awful place, Washington. In our state, this is a last March, you saw solar produce 50% of the electricity right around noon time. And as a result, power prices actually went negative. They were paying you to shut off your power plants. And then power prices spike in the evening when the sun sets and wipes out solar output. This is a very visual representation of why it is the case that solar's value declines as more of it's deployed. In California, we've got over 10% of our supply coming from solar. What happens when we need to get to 30, 33%? This picture or the familiar duck curve that many of you have seen will only be exacerbated. All right, that's setting the stage. Now let's talk about the hopeful stuff. You know, I talked with Gwen and Sunny from Aurora Solar before this and we talked a lot about all the problem solar faces. And at the very end, Gwen goes, come on, man. Like, can you just give us some hope? So I'd love to give you some hope. Let me talk about the three kinds of innovation and drill into them just a little bit more than what you saw in the video. The first kind of innovation is financial innovation. You know, as I mentioned, solar could run into this capital crunch, a shortfall of investment capital. Bloomberg projects in that top graph that between now and 2040, we could see a shortfall of two and a half trillion dollars between what we need to keep that hockey stick trajectory going and what current investors are probably willing to provide. Now, those current investors aren't the most deep-pocketed investors in the world. Those are the institutional investors and they've largely sat on the sidelines. So the bottom graph breaks out how we could meet that shortfall if those institutional investors got in the game. It's like an EA sports promo, get in the game. No one? The guys who played like FIFA 2004 laughing. So in that bottom graph, there's several ways that in addition to the current sources of investment, you could see institutional investors provide both debt and equity financing. Here's one. You could provide equity financing if institutional investors had vehicles that they could buy and trade on public equity markets, on stock markets. Now, yield codes were the first incarnation of such a vehicle. The logic here is pretty good. You pool together a lot of solar projects, you put them into a vehicle, it's a diversified portfolio and you enable people to buy and trade them so they no longer have liquidity risk. They also don't have due diligence constraints because they're trading a diversified portfolio. This is great. The problem though was in the design of the initial yield code vehicles. They were designed in such a way that there were some conflicts of interest between the parent developers and the child yield codes, conflicts of interest over governance, for example. And there were also structured in such a way that these yield codes had to greedily continue to grow in order to keep their stock prices rising and pay out the dividends that they had promised investors. These were all mistakes in what should have been the design of a very boring instrument. What institutional investors need is a boring instrument to package together boring solar projects. That's their strength. Boring solar projects pay out revenue year after year for 20 years. They're very low risk because there's very little you have to do to maintain the projects and they're uncorrelated with market volatility. That's what we should be aiming for and I believe and I write in my book that there could be a second generation of yield codes. The Europeans are already on this that are much more conservative and frankly boring. Here's another cool way that we can source capital, especially for distributed solar assets. Remember the financial crisis when we packaged and sliced and diced mortgage securities? Let's do the same thing for solar, no last there. No, don't do the same thing for solar, but do something similar, do it responsibly. Asset-backed securities are actually a popular investment destination and they are used responsibly in a large number of asset classes. I think that they could be used in solar and already in 2017 we saw over a billion dollars of solar securitization. Now some of the most exciting stuff, you know I just flew through those investment slides because they're kind of boring. Some of the most exciting stuff though is in the developing world. You know over a billion people still lack access to electricity, reliable electricity. They're concentrated in sub-Saharan Africa and South Asia as you can see from this graph. And new business model innovations are enabling folks to get a taste of electricity for the first time. This is the pay-as-you-go model. It's a business model innovation that was pioneered in East Africa, now is being rolled out in South Asia and is enabling folks to purchase a solar panel and battery installation with basically no money down because the startups are able to source their capital. Well actually right now they're sourcing their capital for venture capitalists, but down the road they'll be able to source their capital from financial markets to build those upfront systems. Consumers then will just pay using mobile money. Mobile money is taken off, for example, in East Africa, organizations like Mpesa. And using that mobile money, you can actually pay off your entire solar installation and own a valuable asset. Now you've got the solar system. It's got solar panels, it's got a battery, it's got some appliances, and that system can be used as collateral for you to take out a loan. So not only are you bringing electricity to consumers for the first time, you're actually economically empowering them to have credit. This is a big deal. Now it's not just business model innovation. On top of this you've got the falling cost of hardware. And I want to say that not only are solar panels and batteries falling in cost, but so are the energy efficient appliances that enable these systems to pencil economically. Look at the cost declines there, 26% in the cost of energy efficient DC appliances. What can governments do? I haven't even talked about governments and I work at a public policy institution. Come on. Governments can absolutely help here. You know, in many cases I urge governments to just kind of get out of the way and allow financial markets to do what they do best. But in cases like this in the developing world, it's actually important for a government like Nigeria's if they want to meet their energy access goal to coordinate efforts between the central government and organizations that are nimbly trying to bring off great electricity systems to villagers. Because if the government doesn't tell these off-grid system developers that they're coming in and then the government extends the grid, they can put those developers out of business and chill the investment climate. So here's an example of how Nigeria has demarcated different areas of the country as optimal for a particular type of electrification. Some areas are optimal for extending the electricity grid. Some areas are optimal for developing micro grids or individual solar home systems. Ultimately, the grid should extend and link up with those micro grids. But in the near term, if you're gonna make a very aggressive energy access target, it's a good idea to coordinate. All right. The part you've all been waiting for. Cool technologies. Because that's what we do here, right? We reinvent solar. I was very lucky to work in Oxford along with Thomas in the lab of a real visionary, Dr. Henry Snaith. Many of you may know him. And I will be perfectly honest. I did not recognize the transformative potential of the Paravs Guide early on. So Thomas makes fun of me because I was at the time making nanowire solar cells. And I said, you know what? Nanowires are a great idea because they will streamline charge, transport, they'll turn it into electron highways and we will collect charges really efficiently. And the Paravs Guide was discovered by a fellow grad student and I said, perfect. I'm gonna put the Paravs Guide on top of my nanowires. Completely missing the fact that one of the cool things about the Paravs Guide is it's very good at charge transport. I no longer need the nanowires. And so, you know, I made one of these devices with nanowires and the Paravs Guide on top and I said to my professor, look, this device performs at 3% power conversion efficiency. That's three times better than I was doing yesterday. And my professor says, I know, if you took the nanowires away, it would be 10%. Anyway, I said that preemptively because I knew you would raise your hand and ask that question in the questions here. So this is a cell that one of Thomas's and my colleague Sam Stranks made over at MIT. It demonstrates that Paravs Guides could actually be flexible. You know, this material gives you a lot of versatility. You can make it flexible, you can make it semi-transparent, you can make it colorful, and on top of that, you can make it super efficient. Still, it is the case that these Paravs Guide solar cells, if they go up against the silicon behemoth incumbent, are probably going to lose because there are a lot of examples. There is this, you know, graveyard littered with companies that tried to go up against silicon, tried to go up against Chinese behemoths. So that's why Oxford Photovol takes, which is the company that Henry Sneith has spun off, is targeting tandem devices. They're trying to take Paravs Guides, which are a great technology in their own right and putting them right on top of silicon so that if you have a silicon panel with the Paravs Guide coating, you have a performance boost. Paravs Guide, as you can see in the bottom left-hand side of that slide, is helping to harness a different part of the solar spectrum, the high-frequency part of the spectrum, than silicon is. That tandem piggyback approach might be a way to gain some commercial scale of manufacturing Paravs Guides. And then down the road, you might then be able to make Paravs Guide only devices, even Paravs Guide on Paravs Guide tandems. Here's a rough sketch of how good such devices could get. On the left-hand side, you see two other emerging technologies, quantum dots and organic cells. Paravs Guides, as you can see, have managed to achieve a 22.7% cell efficiency. Folks are improving their stability, they're improving their performance characteristics, they're making them bigger. Silicon, of course, is still better than Paravs Guide, but perhaps not for long. And the silicon Paravs Guide tandem might very soon surpass silicon alone. Down the road, though, I'm excited for the prospect of Paravs Guide only tandem devices. Thomas is working on some of them. I think you actually own the world record for them, right? He nods modestly, yep. Down the road, we could actually see Paravs Guide only devices achieving a 35% efficiency. So let me recap. Paravs Guides could be dirt cheap because you could literally print them. They could be highly efficient, they could be flexible, semi-transparent, and lightweight. They could revolutionize the way that solar is deployed around the world. Now, I don't only want to talk about photovoltaics, I also want to talk about concentrated solar power. See, concentrated solar power, for those of you not in the field, this is mirrors concentrating the sun's rays to generate heat. Concentrated solar power comes with built-in storage. Now that you have a hot fluid, you can store it. You can store it in the form of molten salts. You can generate electricity in the middle of the night. And that's gonna be increasingly valuable, especially as solar value deflation starts to take root. So even though CSP, concentrated solar power, has seen a market lull in recent years as photovoltaics is raced ahead, I foresee that as photovoltaics penetration increases, we're gonna see more demand for CSP. And you can see even more demand if technological innovation for CSP actually increases the cost down to that 2020 target or even lower. And the way to do that is to increase the temperature to which the mirrors heat up that solar energy because then you can store the energy more efficiently and you can also generate power more efficiently by using different thermodynamic cycles. For example, the Brayton cycle with supercritical carbon dioxide instead of steam. Finally, I'm really excited about artificial leaf technology. Now this stuff is kind of way out there, all right? This will be deployed probably not in the next decade but beyond that. But it is an exciting prospect to think that instead of oil refineries, we could have solar refineries in the future. We could have this case where artificial leaf technology harnesses sunlight to split water generating hydrogen fuel which could be used as a transport fuel. That's what you see in the right-hand side, the top right-hand side of that diagram. Or it could be combined with carbon dioxide. For example, a waste product from a smokestack and used for a range of industries. Fertilizer, plastics, pharmaceuticals, et cetera. So solar is not just an electricity source. It is a source for powering our entire economy. That's really important. That's what underpinned Shell's prediction that if we're going to meet the two degrees decarbonization scenario, we're going to need solar to do double duty. It's going to have to produce electricity. It's also going to have to produce fuels. All right, let's put it all together. This is the final part of the book. By the way, I will note that this is only a small subset of the fun charts and graphs in the book. So I'm not giving it all away. So the final chapter, the final part, is about systemic innovation. I argue that there's a whole battery of approaches to integrating solar power and a battery is only one of them. And Dr. Creason got it. Dr. Creason, I want to thank you and Mrs. Creason for joining tonight. You guys have been so formative throughout my career. Thank you. And also thank you for laughing at that. No one else laughed at that joke. Come on, guys. This is one way that we can integrate solar energy into our power systems by expanding the size of our grid and also by making our grids much smarter at the micro scale. Now, neither of these is the intuitive solution of just pairing batteries with solar. But again, I argue that there's a whole panoply of ways to make our power systems more flexible. Larger grids are great because you can aggregate supply and demand over a larger area, which enables you to smooth out volatility. Smarter grids are great because you can marshal lots of different demand side resources. Now, to get the best of both worlds, I created this ugly busy graphic, which attempts to say that a hybrid grid, which has long distance, high voltage DC transmission at the macro scale, but also networked micro grids at the micro scale that are super smart, well, that might be the way to get the best of all worlds, to integrate the largest amount of renewable electricity possible. Now, I do want to talk a little bit about batteries because the point I always get is, look, man, batteries are falling in cost. Why can't we just rely on them to store solar energy and counteract value deflation? Well, batteries are indeed falling in cost. But there are a whole lot of energy storage that are not just batteries. This graph shows you some of them. And if we only relied on batteries, here's what would happen. Colleagues at MIT and I have simulated the Texas Power Grid, and we determined that if you only used batteries, and we assumed really cheap batteries, $150 per kilowatt hour fully installed, and really cheap solar, $0.25 per watt, four times cheaper than today's solar, well, if you had all this, solar still wouldn't be a majority of your electricity in the lowest cost grid configuration. That's because batteries still aren't very good at storing electricity for longer than a few hours. And solar is intermittent over the course of days, weeks, months, seasons. There are a whole lot of storage requirements, and there are a whole lot of technologies to achieve that storage. Batteries are just one of them. Lithium ion batteries in particular are just one of them. I'll end because I've got 32 seconds. I will end with the last policy recommendation. Look, I think that governments really need to focus on technological and systemic innovation. And I think that energy innovation spending is a crucial driver of enabling the United States to be a leader in some of these new technologies. That's why I was thrilled last week when President Trump miraculously signed the omnibus spending bill that will increase funding for solar energy research and development by 16%, and increase overall funding for energy innovation and continue to fund the organization, ARPA-E, that Arun founded. It is a phenomenal organization. We need to keep doing that because if we don't, as you can see from this figure, China is going to eclipse us, no pun intended. Even Bruce isn't laughing. He's disowned me as a advisee. Going to eclipse the United States in terms of its leading role in energy innovation. So, Arun, sorry to embarrass you, but you embarrassed me. Thank you so much for saying these really nice words. I hope you guys take a chance to check out the book, and thanks so much for listening. It's really great to be back. Thanks for reading, and thanks for leaving time for questions. I didn't use the PV magazine title for you, The Hamilton of Solar Energy, because I didn't actually know what the heck that meant. Now I think I know what it means. Any questions we usually like to start with students first? Student questions? There's a student with a question. Sir. Do you anticipate perovskites being able to escape a lot for long-term use? I mean, you really should answer that question, but I'm going to try. Perovskites have been demonstrated in laboratory settings to be stable for over a thousand hours of testing. They've also actually been demonstrated in real-world settings to be stable on that time scale of months. I actually don't think that there will be a stability problem if perovskites are encapsulated in the same way as silicon solar cells are encapsulated in a way that really prevents moisture ingress, moisture attacks perovskites particularly badly. I do think that it'll be important down the road if you want to make perovskite-only products to figure out how to really protect the material if you're encapsulating it inside polymers. You really need to protect it from, for example, moisture without the benefit of the heavy encapsulation materials that are used in silicon solar panels. Thomas, did I miss anything? The thousand-hour tests are accelerated tests. So typically if you pass those, then that indicates you can have a 10, 20-year lifeline. Exactly. So thousand hours is a stress test. So it indicates that you'll survive for even longer than a thousand hours. You've been light-soaked, for example. Oftentimes you may be under high humidity, high temperature, et cetera. Thanks for the question. Great. Over there. Thank you for that. Thank you. I would love to know what the question is. No. But I'm acting surprised. I'd love to know what the implications are of perovskite technology on the installation component of the cell, both in commercial, residential, or grid scale. I think that's a great question. And to back up a step, a lot of people have asked me, why on earth are you supporting innovation in solar materials? Look, the solar panel itself accounts for a minority of the cost of the solar installation. So if solar panels are a dwindling cost component, it may not be a good idea to continue innovating in that part of the system. You should innovate in the rest of the system, known as the balance of systems. But it is the case that the more efficient your solar cell or solar panel, the lower the rest of your costs are. Because the more efficient your solar panel is, the less your solar panel costs. But also the less land you need, the less labor you need, the less equipment you need to mount those fewer panels, et cetera. So I think that more efficient silicon perovskite tandems immediately give you a benefit, not just on the module cost, but on the installation cost. Beyond that, I actually think that perovskite only or other flexible, lightweight materials allow you to have dramatically different system economics. For example, you could have building integrated PV that honestly just has a completely different cost structure than today's silicon solar panels. You could be, rather than installing ground-mounted solar panels, you could be carpeting deserts with coatings. You might even replace them every couple of years because they're so cheap. You might even throw away a whole lot of the power and turn them into rampable power plants during the day. There are all kinds of opportunities that open up when you have dirt cheap materials, which is why I tend to reject the notion that we should compare today's system costs with what tomorrow's technologies can offer us. Thanks for the question. Come on, let's go right there. I'll go back and move on. Go ahead. Yes. Hi, everyone. I'm curious about some of the underlying causes of technology lock-in. Why it happens with some technology that's not with others and what you think is necessary to prevent it from happening with solar panel production? So I think technology lock-in, first of all, is not a solar-specific phenomenon. It's not even a nuclear-specific phenomenon or energy-specific. It happens across fields. A common cited example of technology lock-in is the QWERTY keyboard. They say that they invented the QWERTY layout so that typists didn't type too fast and jam the typewriters. But now we're stuck with it because everybody uses it. So technology lock-in has many different causes. One cause in the QWERTY keyboard case is network effects. The more that a particular technology is used, the more valuable it becomes, making it difficult for another technology. Raise your hands if you guys use Dvorak keyboards. Not a single one. See, there you go. Locked out. Network effects are one driver. Another driver is economic scale. If a technology is able to achieve economies of scale, that is massive production scale that enables your variable cost to swamp, your fixed costs, or learning, you've achieved scale and so you've gotten really good at optimizing your production process. Well, that's another way you can have lock-in. A final way you can have lock-in is through what's called regulatory lock-in. If your public policy regulations are tailored, for example, to light-water nuclear reactors, it's really hard for that next nuclear reactor, say, a small modular reactor to break in. Because that small modular reactor may well require very different emergency planning zone sizes, for example, very different safeguards. And yet it has to conform to the same regulations that were tailored for light-water reactors. So these are all different causes of lock-in. But lock-in is a well-known phenomenon, and I worry that it's taking root in solar. Did I answer your question? I guess. Yeah. Straight back up the other side. Yeah. So thanks very much for your talk. I was wondering, in your book, do you address at all a structure or governance of an ideal public-private partnership for the US or Europe to contest with China? Yes. The China case is a very interesting public policy case. Look, it is definitely the case that China underbid, or rather, that Chinese manufacturers produced panels and sold them below cost on global markets. By the way, I can tell you that because Dr. Reichelstein sitting there just gave me a look. But it's your research, I think, that shows that the price of solar panels was below an economically sustainable level, especially in the period during which Chinese manufacturers increased their hold on global markets. Was that an accurate statement? Define economically sustainable cost, and whether pricing below there is an offense, so to speak. Oh, I haven't called it an offense. But I am saying that that's what happened. Now, whether it's an offense, you and I can definitely debate. But this is a case where, as a result of Chinese manufacturers producing a lot of solar panels, and they had the help of extensive loans from China Development Bank, for example, as a result of this, a lot of US firms went out of business. Now, the policy response from the United States has been three sets of tariffs. Two under President Obama, one under President Trump. And I don't actually believe that any of these sets of tariffs are going to solve the problem. I don't think that we're going to have competitive manufacturing here at the United States as a result of any of these tariffs. And frankly, we'll probably be poking ourselves in the eye because we will destroy jobs in the installation of solar. We might invite retaliation from China. You're already seeing some Chinese retaliation. And I should think that one perverse effect has been discouraging innovation. There are companies like Sunpower that depend on revenue from imported panels, but also conduct R&D here in the United States that aren't going to be able to conduct that R&D, because that's the first thing that goes to not affect the quarterly earnings bottom line. So many reasons for why tariffs are not the right idea. Your question, though, was, so what do we do? Is there a way, for example, for there to be some international cooperation where in cases, not just of solar, but clean energy products broadly, batteries, for example, can there be an early warning system or a body that monitors firms to determine whether they're pricing below cost or a special safeguards agreement that is an addition to or a side agreement to the World Trade Organization saying, for this list of environmental goods, perhaps, we'll have a particular protocol for handling disputes in a expeditious fashion. I don't actually know what the right solution is, but I can certainly tell you that the tariffs were definitely the wrong solution. Let's go here and then back up, yeah. Do you think the massive ramp up in electric vehicle technology is going to provide a way of evening out this electric demand curve to solar PV? Is it going to be soon enough? And is that addressed in the forecast? It could really go either way. This is one of those cases where we don't know which way this is going to go. Electric vehicles are coming, everybody, and they're going to stress grids out. And without proactive planning, electric vehicles could sharply increase demand, for example, on congested distribution circuits, require costly infrastructure upgrades so that electric utilities can serve those customers. And it's not clear at all that electric vehicles are going to make it easier rather than harder to integrate solar energy. They may just make demand more peaky at the wrong times of the day. However, I think you're onto something really profound, which is there is this new source of demand coming on the grid, electric vehicles. And if we harness them right, they could turn into fleets of mobile batteries to be used in service of balancing supply and demand. In order for that to happen, I think you need two things. First, you need intelligent infrastructure. You need a system of charging stations that is digitally connected, both through communication networks and managed through intelligent protocols so that you can tell exactly where all the electric vehicles are charging and when they are charging. Second, I think you need price signals. If utilities send price signals that are very granular at a geographic and temporal domain to these electric vehicles saying, now is a good time to charge. There is a glut of solar energy on the grid. And their incentive eyes to do so. Otherwise, they're going to go ahead and charge when it's convenient for the driver, which may very well not be convenient for the grid. That, I think, could compound the problems we have with an increasing share of intermittent solar energy, not solve them. OK, last question. For decreasing the footprint of solar panels and for increasing efficiency in chips and panels. Yeah, so if you want to decrease, I assume you're talking about the physical footprint, not the environmental footprint. Yeah, so the best way to decrease the footprint is to increase the efficiency. And that's what some of these technologies like perovskites promise to do. Look, if you are able to achieve a 30% efficient solar panel or coating in the future, you'll correspondingly reduce the land area needed for solar compared with today's 20% or less than 20% efficient commercial products. That could be very important. Yeah, I mean, in that particular case, you go from 20% to 30%. I think you'd decrease your land area requirement by 50%. But what's why that's important is that even today, when solar is just 2% of the world's electricity mix and growing, we're already seeing land scarcity issues. For example, in India, the initial auctions were super easy to achieve because developers went ahead and found parcels of land. They won their auctions. They provided contracts. And they built their projects. But now we're in a regime where developers are struggling to secure parcels of land. There are eminent domain issues. They need to consolidate land to build these large projects. And so not only do they have to design and permit a particular installation, they've got to go search scour for land. I think land will become an increasingly important constraint. And that's why efficiency will become even more important. Thanks for the question. Great. Thanks, Farron. I guess I had one question. When do you turn 35? Six years. Thanks again. Before we wrap up here, I want to announce that Varun has graciously agreed to do a little meet-and-greet book signing with cookies provided by Mary right outside the auditorium here. And with that, let's give Varun one last thank you. Thank you so much. Thank you.