 To follow up on past, there'll be lots of light-related puns regarding shining light and so on in my remarks. So let me begin by first thanking the Institute for the very kind invitation. It was very nice indeed. And from my perspective in particular, I think it's great to have this opportunity to address this topic in the Irish context now as we enter into perhaps a period of nasancy for solar in Ireland. In the United States over the past five years, we have seen a tremendous transition. Solar once was certainly a very niche element, I mean one with a lot of potential but ultimately not a lot of on-the-ground deployment. I think today where solar is now the largest, the most rapidly growing form of new installation. And one that is incredibly beginning to enter into an economically competitive envelope that we didn't even imagine was possible even five years ago. With that said, this transition towards a more solar heavy future in the States is already proving to be relatively complicated. There are some very specific characteristics of our markets that are not necessarily relevant here. But in my remarks now I'm going to provide you with some context from the US perspective, which I think hopefully will seed a discussion that we can all have over the next hour or so. Let me begin by first emphasizing why solar is so important. So without doubt it is the most important of our renewable energy pathways for the world. If we wish to decarbonize the global energy system significantly, solar is going to play a very, very important role. And fortunately it's a very large resource so it's well capable of doing that from a technical perspective. Additionally it's also a very well distributed resource. So here's just a map of the global solar resource and I won't emphasize my slides too much because I know it may be difficult to see them. But basically the point is that there's lots of sun, typically where there are lots of people on the planet, except in Northern Europe which is slightly ironic. But crucially going forward where demand growth for electricity in particular is centered. So Asia, India, Africa for example. These are regions that have excellent solar resources and that's going to help these parts of the world transition their energy systems in a more low carbon sense than has been possible in the more developed world. Now the one technical slide isn't very technical but for any of you not particularly familiar with the technical aspects of solar, there are two flavors of solar energy. Photovoltaics is the one that I'm sure most people are familiar with so that's the panel. I should be more precise. Two forms of two flavors of solar for electricity generation. Hot water is a separate issue. But for electricity generation you have photovoltaics and that's certainly one that's most common. 97% of global systems are photovoltaic based. And that's one by the way that will be relevant in Ireland. The second option is what we call concentrated solar power. And CSP is different in the sense that rather than a direct light to electricity pathway it involves light to heat and then heat to electricity. That's extremely advantageous because you can store the heat which means that CSP represents a dispatchable renewable energy technology which is really what we'd love to have ultimately. The drawback unfortunately is that it doesn't work well where you've got clouds. So hence my comment that this is not relevant in Ireland. In fact it's relevant only in a very small set of regions in the world. The southwest of the United States, the Australian desert, North Africa and a few places like that. So coming back to my original point that we've seen this tremendous transition, this rapid growth in solar, solar arriving at the big table. The numbers say it all actually. This year globally we've installed about 60 gigawatts of solar PV and we're accelerating year on year. So this is the first year actually that we will install more solar globally than wind. And I think that's a point that's going to continue. Solar is really the global renewable ultimately. A few points though that are important. Deployment to date has been centered or at least concentrated heavily in Europe. So Germany in particular was in the vanguard of solar deployment. They had a very strong support mechanism in place. That is now waning of course. But what we're seeing is acceleration in three markets in particular, three important markets. So China of course is important in every context. Japan and the United States. And those three markets are going to be three very important vectors for long term growth for solar. In the United States, as I said, more than half of all the new generation installed last year was actually solar PV. And we're going to continue to see that accelerating for the time being. Now, solar PV is different than almost every other energy technology for one particular reason. Its technical efficiency at converting light in this instance to electricity does not scale with size of system. So one solar panel is just as technically efficient as a thousand solar panels in a utility scale facility. That's not true for almost any other energy technology, even concentrated solar power. In fact, it's certainly not true for concentrated solar power. And this is what's particularly interesting about PV. It means that we can deploy solar PV at many different scales. So at residential rooftops, commercial rooftops, right up to large ground mounted facilities. And this flexibility is of course to be welcomed, but it brings additional challenges in terms of designing the system. And particularly in terms of evolving a system that has been designed for central stations in order to fully accommodate this. And it's going to be quite complicated actually in many settings. We're already seeing that in the United States. The US is particularly important in terms of exploring this space because we have deployment, large scale deployment at all of these scales. So last year, 60% of our capacity was deployed in very large ground mounted facilities, many of them in the southwest desert. The balance, 40%, was deployed evenly between large rooftops, so factory rooftops and so on, and residential rooftops. And there is an interesting juxtaposition there because what you have is a small number of very large systems bringing a huge amount of capacity. But a huge number of very small systems bringing much more general awareness of the technology. In fact, this month we will have installed the one millionth solar PV facility in the United States. And almost all of that are on residential rooftops. Now making all of this possible, and this is not just in the US but of course globally, has been a dramatic reduction in the cost of PV technology. So it's interesting even coming from the technical perspective because this has actually altered entirely our perspective on what matters. Five or six years ago, focus for solar research, the lion's share, was on the panels themselves, the technology, the conversion technology. And the reason for that was that in terms of an overall systems economics, the panel was the most expensive part by a long shot. Today the panel for a typical residential rooftop system in the United States makes up less than 20% of the systems cost. In fact, the labor component, how much you're paying the guy putting it on the roof is now a much more important determinant of the systems economics. And this is actually turning a paradigm on its head somewhat. How do we go about altering our focus in research in terms of developing a technology and a balance of system that allows us to move the economics in the right direction? Scale in Chinese manufacturing basically is what moved the needle with respect to the panel's costs. And wild competition and a whole host of other issues with respect to trade subsidies and so on, which are contentious and complicated. But nevertheless, the Chinese cost reduction has done us all a favor because now we have a much, much lower cost technology out there. But now going forward, we're going to face the challenge of how to move the balance of system costs lower. And as I said, that's going to involve reducing costs of racking and unwiring and all sorts of things. But more importantly, it's going to involve reducing the cost of actual kind of soft labor financing and so on. And all of this now must be brought together effectively for going to continue the cost reductions that we've seen. So quickly on Solar's competitiveness today, just a snapshot. So in the United States today, a utility can go out to the market and can have an issue an RFP for some solar power at large scale. And the developers that we have in the US market at least today will come back and they will tell you, we can sell you solar power for $40 per megawatt hour, $0.04 per kilowatt hour or less. That's what they can deliver. Now that is net of some subsidy, certainly. But only five or six years ago, that would have been much closer to $200 per megawatt hour. And that's what's really moving the needle, okay? This past month, construction has begun on the first merchant solar facility in the United States. So a plant that is now being built which will compete in the local wholesale market in Texas without the assurance of any backstops. No feed-in tariffs and so on. Certainly there are a few subsidies that they're taking advantage of. But what is most interesting is that the financiers in particular now view solar as a vector for growing power generation in a much more kind of standard sense. It's now part of the wholesale market and can be competitive without long-term guarantees. I won't belabor this, but really all I'm trying to do here is illustrate how what we call the levelized cost of electricity from solar looks compared to some alternatives. That's gas in the United States. And the message really is that in certain parts of the U.S. today, so the southwest in Texas, anywhere in the south, where we have an excellent solar resource, ground-mounted systems are beginning to be competitive with more traditional assets. Some of them even without any subsidy, right? And that's a very, very important. By contrast, and this is just to demonstrate the complexity of the kind of the solar space and why scale matters, rooftop systems or systems that are installed in regions like Massachusetts where I'm from where our solar resources significantly inferior are much more expensive. And this then leads to a complex discussion that must be had. You can have very cheap solar from large facilities in the desert, but that's a long way away. It has significant environmental impacts, certainly, and a whole host of other issues. Or you can have rooftop solar, which is near to the demand and provides a flexibility and so on, but it doesn't have the scale you need to really move the needle in terms of targets, for example. This is going to play out, but it's complex. And this complexity, the fact that's jumping around a lot, means that developing a regulatory paradigm, developing a subsidy paradigm that's fit for purpose is also complicated and needs to be quite bespoke. Now, briefly on some challenges ahead, and there are some real challenges. And in the U.S. context, and this isn't a context that's relevant everywhere, but in the U.S. context, this graphic here actually illustrates probably one of the most important challenges, certainly for large scale solar. So today, and in the previous slide, I just mentioned that solar is becoming increasingly competitive. It might even be competitive without subsidy, but that is at what we call the margin. So in our wholesale power markets, for example, when solar is just providing a little bit of power on the market at zero marginal cost, I might add, the clearing price for many of our markets, the Texas market, for example, will be high enough such that that solar asset is going to be able to pay for itself quite happily. But what happens when we begin to move enough solar onto the system to matter, right? To matter if you're worried about climate change? Well, in our paradigm where we have wholesale markets, what happens is that solar begins to produce a lot of power right in the middle of the day, obviously enough if you don't have storage. And right in the middle of the day, by the way, happens to be when our power prices are highest today because demand is highest in most of our daytime peaking systems, but suddenly you have a lot of solar, which is zero marginal cost. So that means you don't need to run your more expensive peaking unit or your gas turbine, which is setting a higher marginal price. You begin to actually depress the overall price, so power becomes cheaper for everybody. That's why that red line begins to fall as the penetration of solar increases. It falls marginally because, of course, your solar plants only generate during the middle of the day. But for solar asset owners, the price really collapses. So today as a solar asset owner on a system, I get this premium product. I'm delivering a premium product because my solar farm generates power when demand is highest and when prices are highest. But if everybody's just delivering solar power and it has a zero marginal price, which today is market structures, the price collapses, and I'm no longer achieving the rents I need in order to pay for my plant. So there's a negative feedback effect that's potentially possible with this in terms of incentivizing deployment. We're going to have to contend with this. There are two ways to do that. One is to bring storage to the table. And once you have storage, you can actually move that power around the system. Storage costs. And in addition, we might have to rethink our market structures quite significantly because the reduction in the wholesale price that takes place during this period, it doesn't just impact you, the solar asset owner. It also impacts other asset owners. And in doing that, you're reducing reliability on the system in certain instances. It's impacting the equity value of many of the other players and so on. Now, this is a simulation. The real Texas system today just moving more solar onto it. But this has happened in Germany. This has really happened, this exact dynamic. Basically in Germany, 2006, 2007, 10 years ago, they had a very peaky wholesale power price. Middle of the day demand was relatively high. That's when their utilities were running more expensive units. That's when they were paying down the capital on those systems and so on. The Germans introduced significant volumes of PV since then. And every year that has actually pushed down that average price during the middle of the day. And that has been causing significant challenges for many of the utilities. Now, whether or not you wish to feel sympathy for the utilities, that's entirely separate. But the fact that we can't keep units on is an issue. You see, for example, in the States, actually wind is causing this dynamic in certain regions. It's actually shutting down some of our nuclear facilities. And again, nuclear is a complex topic, but that's zero carbon generation that we're losing to the system now. So it's quite a complex setup. The German feed-in tariff meant that there was no feedback on this dynamic until they reduced it. We don't have that paradigm in the States, so we're looking at this issue as perhaps a throttling effect that we're going to have to deal with for deployment of solar in many of our markets. Now, very briefly, just to speak about how we deal with this. Well, ultimately, we deal with this with technology. That's very helpful. We'll deal with this with technology. What we need to do is we need to innovate, and we're doing that at MIT in many, many places around the world, looking to actually develop solar systems that can be deployed at much lower costs than even today's low costs. And in doing so, then, you can continue to compete even in these markets. There are real options available for doing that. Thin-film technologies are going to be particularly important. They provide a flexible film that can be manufactured in a very high-throughput, roll-to-roll framework. That means it's extremely cheap. You don't need the high-temperature, low-vacuum settings that are needed for crystalline technologies. You can make them from abundant materials, which is crucial at scale. But today they're immature. It's going to take a long time to bring them to the market. So we're going to have to work with what we have, which is quite good for the next decade or longer before we get to this new technical paradigm. And so my final section before I sit down, a little bit on the solar business model and subsidy structures. So low costs, that's been crucial. A good resource has been crucial. Societal demand for solar, interestingly, has been increasingly important in the US broadly. Now, one reason that it has been increasingly important is that there has been some innovative business model development on the residential side, which has meant that more people are actually seeing solar panels on their neighbours' roofs and deciding, I would also like a solar panel because he's quite sensible, so that must be a good idea. And these business models have enabled that. And they've enabled that by lowering the cost of acquisition, by helping make the subsidy monetisation more straightforward, and so on. It has, however, led to some challenges. And in particular today, in the residential market in the States, we don't have enough competition and we don't have enough information. So people are actually paying too much for their residential solar systems. There's a concept known as value pricing. Today, our residential solar installers value price their systems. They sell you a system relative to what the ESP equivalent will offer you power for. And people, as it turns out, from a behavioural economics perspective, like about a 15% discount. If it's 15% cheaper, I'll have it. It seems to work. And of course, the installers have figured this out. And they price at about that level, even if they're able to take a lot more cost out. That cost is for them. And that's great. But ultimately, we would like to see more information in consumers' hands so that they could actually push down their energy costs even further. By doing that, that would help in terms of supporting the deployment of things like batteries, which are ultimately going to be required. As Amon mentioned, the German context, Germany has much lower residential costs than we have in the States today. And part of that is because the German market, consumer market, is much more aware of what it actually costs to put a system on the roof. By contrast, our utility space is extremely competitive, really good. And we have seen that the way we've structured our subsidy doesn't incentivise kind of holding prices up. It actually helps drive the competition among developers. And that competition among developers has meant that we're now approaching a situation where utility-scale solar facilities are being contracted at about $1.25 per watt. Really, really very attractive numbers. So finally, to that point of subsidy. Ireland is now entering into, through the white paper and so on, clearly a consultative process at least with respect to how to support solar or how to bring solar to the market. We have had many experiments in the States at various levels of government with respect to this and many cautionary tales can be told. First, let's talk about our federal level. So we have a federal investment tax credit for solar. We have a federal production tax credit for wind. So why does that matter? Well, an investment tax credit does exactly what it says in the tin. It provides you with a tax credit based on the cost of the system. With solar, different places, different system scales, very different costs for the system, even though that system produces exactly the same low carbon product. And what that means ultimately is that today, depending on where you are and depending on the type of system you have, you will be receiving anything from $30 or $40 per megawatt hour to more than $100 per megawatt hour of subsidy for exactly the same megawatt hour of low carbon energy. And that's a silly system. So the United States really ought to transition to federal subsidies that at least incentivize generation, so production, because that's what matters if you're worried about carbon. We're not going to do that though. We have extended this subsidy paradigm till 2021 and then it will expire. But at that point, we'll probably be good enough for solar to win even without subsidy. But for the less, I think it's worthy of pointing out it is a silly and inefficient and expensive system. Furthermore, the tax credit element means that it's not easily monetized for individuals and even for large solar developers, which requires us to go to what we call the tax equity market, which involves the big banks, giving you $0.70 for each dollar worth of credit to monetize it, which means you're leaking taxpayer money directly into that private sector with no particular benefit from a solar perspective. We have state-by-state level issues with respect to renewable targets. Many of which are quite silly. Massachusetts, for example, has a solar-specific target in our renewable portfolio standard, which requires the solar energy be generated in Massachusetts. The objective of our renewable portfolio target is, of course, to reduce carbon emissions. But requiring us to generate in Massachusetts is very expensive. It'd be quite simple for our and their markets in place in the States for our Massachusetts utilities to just contract with a facility in Arizona where you could get the exact same energy for half the price from the rate payer perspective. But the complexities of also wanting to be able to engender and develop and nurture a local solar industry and so on plays a role, and I think that's important to appreciate. The other big issue, perhaps the biggest issue of all on my final slide is on what we call net metering. So in the United States today, if I put a solar system on my roof, I will receive a credit for every kilowatt-hour it generates that is equivalent to my total rate for electricity that I pay the utility. Now, that's a big subsidy to me. Why is that a big subsidy to me? Because in reality, my bill per kilowatt-hour from the utility is made up of, half of it is, in my instance, half is for energy and half is for the wires. But by net metering and allowing, providing the energy that I produced with a rate that's double that, the utility is in fact paying me for using their wires. And it's a double pay me for the utility. They have lost the revenue from my use of the wires and they're then paying me back to use the wire. This is very contentious now among the utilities in the United States because many of them agree to this 10 years ago. This was administratively very simple and nobody had a rooftop solar system. Now we have a million rooftop solar systems and that is eroding the revenue base for the utilities very considerably. And they have a legitimate issue here because that erosion is reducing the availability of funds to fund the system. And if we don't change that, that's ultimately going to lead to a welfare transfer. And poorer people, for example, who do not have solar systems on their roof would have to pay higher rates to the utility. We're beginning to see traction in terms of changes on a state-by-state basis because this is a state-level issue. And it's proving very, very controversial because it decimates ultimately the rooftop solar business, today's business model. So a lot of tensions and there's no clear path forward yet. But we will continue with it I'm sure and some pathway will be carved from there. So let me conclude by bringing in another pun and say that the future of solar is very bright indeed. But really thoughtfulness on several fronts is needed to ensure that this very significant potential can be realised. So first, a long-term approach to technology development is crucial. So a lot of people are saying, well look, the panels are very cheap now. Let's double down on getting guys faster, climbing ladders and we've solved the balance of system problem. But this dynamic in the wholesale market and solar actually eroding its own competitiveness requires us to actually have lower cost technologies, significantly lower cost. So we need to continue that long-term effort to bring these new technologies to the market. We need to prepare the system for these new types of resources. So we have a system designed for centralised units and we might have a lot of centralised solar. It's somewhat intermittent obviously, but we're also going to have a lot of distributed solar. And we're going to have to prepare the system for that in terms of bringing storage in particular to bear on the system. And then finally, and this is perhaps important going forward for Ireland, subsidy and deployment support for solar development really needs to be developed very thoughtfully and wherever possible it should be designed to be as efficient and as dynamic as possible. So the solar industry costs have moved very quickly. It's important that the support mechanisms can reflect that in order to kind of minimise any efficiencies that may exist. And with that, I thank you all very much and I look forward to the conversation.