CBS Faculty Live: The Global Rise of Clean Energy with Professor Conor Walsh

Okay, good evening. Thank you for joining us. My name is Elisa Douglas, director of Alumni Engagement Programs at Columbia Business School. Welcome to our CBS Faculty Live Virtual Talk. This virtual series brings our CBS thought leaders to you no matter where you are located. In just a few short moments, we will hear from Professor Connor Walsh on the global rise of clean energy. This talk will discuss how power markets around the world are being disrupted by improvements in renewable energy. So over the next hour, professor Walsh will present. Then we will open up for q and a. Just housekeeping, please use the q and a function to submit your questions. Connor Walsh is an assistant professor of economics at CBS. His research focuses on macroeconomics and economic growth with an emphasis on how cities energy use and technological change shape economic outcomes. Professor Walsh's work connects closely to real world challenges, exploring topics like urbanization, sustainable growth, and the impact of local policy on businesses and regional economies. Prior to Columbia, he was a postdoctoral associate at Princeton University and a visiting scholar at the Federal Reserve Bank in Minneapolis. He earned his PhD in economics from Yale University. Now without further ado, I'd like to hand it over to Professor Walsh. - Thanks, Elise. Thanks so much. Well, it's great to be with you all to hopefully share what I think is some fairly positive news. We see movements in fossil fuel markets and the oil price really grabbing headlines in the last week or two, and I think much more slowly, something occurring in the background is the rise of clean energy that gets much less attention, but I think is at least equally as important. Where we are in the US at the moment is over the last decade we've seen fairly rapid growth in the share of electricity being generated from wind and solar, new renewable technologies. Beginning off a very low base from a decade ago, there was very little utility scale solar anywhere in the continental US a decade ago, and the penetration of wind was around sort of 5% of total generation. Now we're up more than 16% and indeed we ended out the year at 17%. For the US as a whole. To put this in context, wind and solar together have now overtaken coal as a source of major generation on the electricity grid. Coal share has been falling fairly steadily over the same time with some, some variability. And coal share is now down to about 15% of total electricity generation. So renewables have come a long way in the last decade from nothing to where they are now. And I think, you know, all signs point to the market being poised for further growth and there is a, a long way to go, particularly to challenge the dominance of natural gas on the US grid in particular. So natural gas is something like 40% of electricity generation in the US and has played a big role in driving coal off the grid just because of cheaper prices for liquified natural gas in particular over the last year. So, so that's been generation slowly increasing or I would say rapidly increasing for wind and solar. Almost all of that is coming from changes of the flows of new generation. So new capital being stall on the grid, that's almost entirely solar, wind and batteries. So if you look at all the new power plants being built in the US, the vast, vast majority of those are in new renewable technologies. So it starts off with a lot of wind about 10 years ago and then transitions into a big burst of solar over the last five years and just very recently in the last two years, we've seen a massive increase in investment in grid scale batteries to soak up the intermittency, the obvious intermittency of solar in particular, bringing it from the middle of the day into kind of the evening peak when it's most valuable. Coal and gas plants as a group have been exiting the grid. So that's this black line here. So fossil fuels are, are transitioning out. The grid coal, as I said, had been shrinking rather rapidly.

Gas capacity was still being added to the grid over the last 10 years, but in the last two or three years, that's more or less dried up and there's really nothing else happening outside of these technologies. Very little happening in hydro. So hydro is a very mature technology with a lot of the most promising sites have already been built out. And nuclear for reasons I'll get into, is mainly constrained on a regulatory basis. And there's just not a lot of new capacity being built, not for want of trying in nuclear. So it's not moving the grid in a significant way. Now this phenomenon is happening all over the world and is really being driven by rapid cost declines for these technologies. So here I've got some total installed project costs for solar and wind across a bunch of major industrialized economies, China, France, Germany, India, Italy, Japan, and the like. You can see solar costs, these are unsubsidized costs. So the cost to build, say a kilowatt of capacity have been falling very fast across these economies over the last 10, 15 years at a rate of about minus 15% in nominal terms every year. So that, you know, that's, that's very rapid rates of let's say technical progress and production progress. Wind has been falling at a slower rate, something like minus 4% a year, but it started off a much lower base than solar did. And now we're around the point where wind and solar together are roughly cost competitive with each other and on a levelized cost of energy basis, which just means the raw unsubsidized bulk cost of supplying electricity without thinking about intermittency and transmission. These two technologies together are far cheaper than any other new build technology. And you know, we are seeing this just in the flows of generation capacity. As with the US the vast majority of new capacity being built around the world now is renewables. So in 2024 we hit about 90% share of new generation capacity being renewables and that's mainly solar and wind. And in 2025 the data's still coming in, but it looks like that's gonna tick over 95%. So that's these blue bars here, it's just been this kind of hockey stick. Let's say the increase in the non-renewable is the gray bars that was fairly steady in the mid two thousands but has been declining gently ever since. So really outside of China and India, investment in new coal capacity has basically dried up. And even China and India have RA have massively scaled back their coal expansion plans as they've switched over to building lots and lots of renewables as I'll get to in a second. But you can now see this in the aggregate data for the world as a whole. So this is from Ember energy. If you look at the generation mix for the world as a whole, you know, there wasn't really a lot of solar and wind on the global grid up till sort of 2010. And now that's just been surging for both solar and wind. Hydro as a share of world generation has been somewhat steady. I would say it's still being built out in some countries, but again it's, it's a pretty mature technology and there's not a lot of scope for expansion nuclear as well. There's not a lot of new nuclear being built across the world, again, for regulatory reasons. And so as a share of global generation of electricity, you know, the size of the fleet stays stable, but as a share of the global total, it's declining really. Where we're seeing something, you know, almost crazy is in China, China's annual additions for solar and wind to sort of bogle the mind in terms of scale. So in 2024 they had about 260 gigawatts of new installation of solar power. And we're still waiting on the data for 2025, but it looks like 2025 might have crossed 300 gigawatts. Just to give you an idea of what that means, China just announced plans or last year announced plans to dam the Brahma RA river, which is a river that flows off the Tibetan plateau and down into India it would be by far the largest hydropower dam in the world. It would be something like 60 gigawatts of capacity and that's three times the size of the three Gorges dam

which is already the largest dam in the world. But building that would take roughly ten eight to 10 years at best estimates. And in one year we got, you know, 260 gigawatts versus the, the size of this, this proposed dam and wind as well. I mean wind is nowhere near as big as solar in China in terms of new capacity. But 2025 looks like we might have crossed the a hundred gigawatt mark there too. So this is just really enormous build outs of renewable technologies and as a result it looks like a total emissions for China emissions of carbon dioxide might have peaked last year and started to decline as they're building so much renewable energy, they're actually outpacing demand growth. So demand growth for electricity growths fast in China just 'cause GDP growth and industrial demand is growing fast anywhere from five to 10% a year. Renewables are now accounting for that total demand increase and even more driving out existing fossil fuels. So it'll be a while, it'll be a few years before we can confirm whether that's the case, but it's quite optimistic for the world's challenge in meeting our total emission reduction targets if China has peaked their emissions. Yeah, I thought this was kind of an interesting slide. That's just because, because China and China is building out so much capacity to meet soaring electricity, to demand its percentage that it's meeting from wind and solar is, is roughly the same as the US if you kind of line them up. They're roughly the same in their electricity transition journey. And you can see the seasonal spikes here roughly line up for the US and China now. Yeah, I think you know, a lot of the story here is solar. Solar in particular has just been a juggernaut the last 10 years, but that's not really something that happened in the last 10 years. It's been an inflection point that's been building on progress that's been going for decades now, which is that solar panels have been getting exponentially cheaper over time. So this is what this graph shows. This graph shows the mod module price for solar panels in dollars per kilowatt going back to the 1980s and in the 1980s, you know, solar panel, the solar, solar powers basically uneconomic everywhere except in space. So if you wanted to put solar panels on your roof in the 1980s, it would've been an a cost of millions of dollars. And now depending on where you are in the world, you can get it for five to $10,000 per roof. And so that kind of exponential decline, you know, is about as fast a price decline as we've seen for any technology that the US government tracks in its investment statistics outside of Moore's Law. So Moore's law being the doubling of transistor counts on, on computer chips every two years. This price decline is, I would say of a comparable speed, but a little slower than Moore's law, but still enough to kind of fundamentally change the economics of power generation if, if it runs long enough and there's every indication that it's continuing year on year to drive big declines in the price of solar power. Now some of that in the early years was research and development, but really over the last 20 years it's been massive industrial expansion and economies of scale, particularly in China, just producing these things in large numbers and driving the unit costs down relentlessly through industrial competition. That scale up for solar is now faster than any power technology in history for the globe. So if you look at how long it took or how many terawatt hours per year is generated by these different technologies after they pass some baseline level, you can see, you know, hydro, coal and gas, you know, they expanded relatively quickly. Nuclear actually expanded much more quickly as it was being built out in the sixties and seventies. But solar now, you know, it's been sort of 10 years since we passed this minimum generation level and it's, it's scaling up faster than any technology we've ever seen. And I think, you know, it's fair to say that no one really saw this coming. This is one of my favorite charts in all of economics. It's from the Economist magazine, but it's really building off some work that's been around on, on Twitter for a while.

And what it shows is the capacity at a global level of new solar installations every year. So 2024 might have been around 700 gigawatts of solar power installed. You can see this hockey stick or this exponential graph, the kind of exponential technical change. And the yellow lines are the predictions of the International Energy Agency for how much solar was going to be installed over the future. So you can see in sort of 2010, 2015, they're predicting that there's whatever's being installed there is going to stay at those installation levels for the, you know, coming decades. And every time the installations increase year on year, they kind of update their ch their spreadsheet and just project that that's going to continue. And they're just proved wrong every single year. They just completely underestimate the scale of this exponential technical change. And even in the recent years when this has been pointed out to them, they've still been unable to grasp just how quickly this stuff is getting deployed and will continue to get deployed into the future. To give you an idea of where we might have been in 2025, that's probably something like 800 or 900 gigawatts. Now every technology is not an exponential forever, it's gonna be an S-curve at some point, but the right point to predict inflection of the scur is not when the second derivative is positive. So I think we've still got quite a way to run in a chart like this. Now obviously the main problem with renewables that has limited their rollout in the past has been there in intermittency. So solar power is obviously not there at night. Wind is intermittent in a different way. It can be intermittent at a scale of sort of weeks or months and have big variable increases or decreases in output for solar. Lithium ion batteries are now solving that intermittency problem in key grids across the world. So this is the fuel mix for California on just a random day last year. You can see in the middle of the day the California grid is just generating a huge glut of solar power that basically disappears around six o'clock when the sun goes down. And then you can see this purple shaded area here is the contribution from batteries, grid scale batteries. Those batteries are charging in the middle of the day and soaking up excess solar power and then they're delivering it onto the grid after six o'clock to sort of nine o'clock at night, right when the power is most valuable and prices tend to spike 'cause that's when everyone gets home from work turns on, the TV cooks, cooks dinner and the like. Now to tell you, you know, this purple shaded area for batteries did not exist two years ago. There was just no grid scale batteries of any size delivering onto the grid. So this is really, really sudden progress coming from, I think the intersection of two things. One is this massive penetration of solar that needs somewhere to go from the middle of the day where we don't need it so much to the night. And two batteries just getting really, really cheap at the same time. So this is the planned over the over 2025 and 2026 new generation in the United States. And so yellow is solar, so there's a huge glut of solar coming on online all over the country in different markets. Blue is gas, so you can see some significant gas plants, but really in totality compared to the, the renewables that are coming, not that significant green is wind. So there are some big offshore wind plants in the northeast that are, you know, having let's say some troubles with the current administration and then black is batteries. So this really came outta nowhere again like two or three years ago you would not have seen something like this for projected battery installations. And really it's concentrated in two key markets. One is California where they have a lot of solar power in the middle of the day. The other is Texas. And Texas is really scaling up energy production partly just to meet rapidly growing demand from population and industrial needs like AI data centers. And at the same time that they're scaling up solar and wind, they're investing huge amounts in just utility scale battery storage. And you know, I find the contrast kind of interesting because California is a very regulated market in many ways.

Texas is just much easier to build new generation, the permitting requirements nowhere near as serious as they are in California. And you know, Texas for sure, the state government is not particularly interested in let's say the economics of climate change. And even given that they're building mainly renewable technologies revealing I think the, the advantages, the cost advantages of battery and solar. Yeah, with batteries in the US we basically saw a doubling the last two years each year on a cumulative, let's say utility scale battery power capacity. And again, that's coming from battery prices falling. I think in a similar way that solar modules fell. I mean lithium ion batteries have made huge progress technical progress as an input into consumer electronics. But building them out at grid scale really begins I think with Tesla producing them at scale for electric cars and then diversifying into producing them for grid scale applications. And then in the last two years, Chinese companies scaling up massively and producing are, are grid scale batteries. The other thing that I think is important to mention about renewable technologies that really drives home the contrast is the time to build. So we are now at this point in the US where we're seeing demand overall grid demand for electricity start to rise again after a period of flat lining for essentially 20 years demand really didn't expand for the grid as a whole since about, you know, 2005 now with AI data centers just popping up everywhere, total grid demand is rising. Again, though it's anyone's guess how much it's going to rise. And when you look at projected needs to meet this growing demand, you really run into a couple of problems for fossil fuels. One is cost, like I said, new build coal and new build gas just kind of uneconomic compared to renewables now. But the second thing is time to build. It's just much, much quicker to build, let's say hybrid solar installations, which is solar with batteries and wind plants, you're looking at timelines of two, three years versus sort of 5, 6, 10 years for new coal and new gas. Nuclear, yeah, unfortunately for nuclear it's, it's even longer. And again, that's not about the physical constraints of the technology. That's just really immense regulatory burdens that stand in the way of nuclear getting built out to meet these very quick demands for power. You know, there are some efforts to try and streamline the regulatory process, but this is such a slow moving beast that I think anyone that tells you that nuclear is going to service, that AI data centers is they may have a bridge to sell you. It just doesn't look feasible from where I stand now, I don't wanna say that everything's rosy in the world, particularly in the US we have kind of a, a fairly large problem with the interconnection queue and that's just the idea that there's so much new capacity trying to come onto the grid that it's, it's posing problems from the standpoint of grid integrators and grid operators who need to model the power flows of any new generation asset. And when many are trying to come onto the grid at the same time, that becomes a very complicated endeavor. And so ensuring safety and quality of power supply, you know, becomes really a challenge. And as a result there's just a lot of solar, wind and storage stuck in the queue for approval across the different wholesale markets of the US that's just taking longer and longer to get approved because there's so much of it. And so there are some reforms looking at shortening these times and improving the modeling. But so far I would say not a lot of progress has been made there. And so now as all this new renewable capacity comes online, I think it's worth asking, you know, what is this going to do to power markets? And I think it, it really challenges traditional ways of thinking about generation in particular because both solar and wind are essentially zero marginal cost technologies. So you know, the, the cost of supplying to the grid when the sun is shining or when the wind is blowing is essentially zero from the standpoint of the operators. These technologies, unlike fossil fuels don't burn anything. They don't have fuel inputs and so they tend to win bids

to bid into generation and as a result, as they start to come into the grid and bigger numbers, they put downward pressure on generation costs at the grid level, which should eventually flow through to industrial power prices and residential power prices. So we did a modeling exercise for the brooking pa Brookings papers on economic activity, try to think through what would be the impact in the long run of high renewable penetration on power prices. And to get that right, you kind of really have to project out what's actually gonna happen to the capital costs because if they have zero marginal costs and no fuel costs, it's really the capital costs that determine the ultimate impact on power prices. So we sort of started from today and said, well if you really want firmed solar power, which was sort of eight hours of storage so you can supply well into the night right now, that's gonna cost you about $2,000 per kilowatt with, with about half of that coming from the battery, the panels now because the panels have gotten so cheap, actually a relatively minor component of setting up a, a hybrid solar farm. But those battery costs are projected to radically decrease in the coming decade, particularly with expansion in EV fleets worldwide. We're just gonna get a lot more scale in production of batteries. So we're probably gonna see something upwards of, you know, a 75% reduction in the all in capital costs of building firm solar in the next 15 years. And you can use that to think through what should happen to power prices in the, in the medium to long run. So if you think that building one of these whole hybrid solar power factory of installations, let's say needs to justify its upfront capital costs and these things tend to last for 30 to 40 years, if you have some estimate of what the required rate of return is and what the required discount rate is, you can back out what the long run power prices would have to be to justify the installation costs being something like $500 per kilowatt. Now this because potential is very different across space, so you might get three times something like, you know, let's say twice as much electricity of a solar out of a solar panel that you put up in Arizona compared to New York. This bound on power prices doesn't have to hold everywhere, but it does give you an upper bound on what could feasibly charge be charged in a long run equilibrium across different parts of the grid. And so what we found is that, you know, once we get out to 2040, you're looking at power prices per megawatt hour of something like 15 to $20 in a big region of the country where sunlight is abundant and land is cheap. And even in kind of the industrial, the Midwest and the northeast of let's say upstate New York, you are looking at sort of $30 per megawatt hour would be the ceiling on what, on what could be charged without stimulating more and more entry of these solar plants. And of course these red bands here are places where you're gonna get much higher bounds just because land costs are extremely expensive. But for most of the country, yeah, sort of 10 20 to $30 per megawatt hour in let's say 2020 $5, that res represents a pretty significant drop on wholesale prices from where they were at the time that we did the modeling. So in the west of the country, you are looking at anywhere of a 60 to 80% drop in wholesale power costs that would be required to just offset this, this investment. And in the, let's say the Midwest, it's more conservative, it's sort of 20%, but even that should have significant benefits both for consumers and industrial users. And I think a, a big question we get asked when we do this kind of analysis is, well how much land is there available to build out this renewable capacity? Obviously solar and wind to a lesser extent are much more land intensive than fossil fuel power stations or in particular nuclear. Now in the aggregate this is clearly not a constraint. So you know, we calculated that you could meet the entire generation needs of the US with about 1% to 2% of the total land of the us. Now that's not nothing that's, you know, a fairly significant chunk of land, but for reference it's about five times the land that we use for golf courses.

So certainly not unimaginable more locally. How much land is available? Well if you look at all the different, let's say cities, metropolitan areas, counties, commuting zones of the us almost all of them can meet essentially all demand locally with solar, solar and wind within a hundred miles. There's a bunch of places, large cities, New York in particular where that becomes more challenging and that you'll have to build out transmission to lower cost generation areas. But you know, a big chunk of the US can source cheap, cheap renewable electricity locally that should put downward pressure on wholesale prices. It's an open question how much that feeds through to residential prices, the prices to the end consumer. 'cause there you've really seen a big increase from di what's known as distribution costs. That's the cost of not transmitting a long distance, high voltage lines, but really going from substations to the end consumer at the household level that's been getting more expensive over time. And part of it is the way regulated mono distribution monopolies work is often they're, you know, allowed to charge a regulated margin on their asset base. And so this gives incentives to sort of overbuild distribution networks. What we're seeing now, I think we're in the very early stages of on the residential side, is seeing price pressures from rooftop solar completely separate from utility scale solar. You know, this is a graph of using Google's rooftop potential data, which is kind of amazing for every house in the US you can get how much it can generate in solar power. Basically every city in the US could get almost all its current needs just from the roofs, the available roof space that it has. Again, New York, big outlier, obviously Manhattan, you're not gonna be able to power with solar power, but much of the rest of the US there's enough roof space sitting there to get local power needs. And as this, the penetration of solar onto local roof space starts to increase, that should create competition for these distribution monopolies and put downward pressure on retail prices separately from the, the wholesale price effect. You know, the US is, is really well, well behind a couple of frontier economies on this, on this metric. So you know, one where this is, this transition is just much more well advanced is Australia. You know, Australia has similar amounts of abundant sunshine to much of the US but in the end, much lower costs for things like permitting and soft costs and installation labor. And as a result something like one in three houses in Australia now has solar panels on the roof. So you can just drive around any random suburb and you'll guarantee you'll see significant amounts of rooftop solar. You know, this is where the us this is, you know, hopefully where the US is heading, if we can get some of these permit permitting costs down and really take advantage of those cheaper panel prices. I mean, what you see on the Australian grid is kind of remarkable. This is again, a random day in May last year and what you see is this, this darker yellow band is utility scale solar sort of meeting significant chunk of midday demand. The lighter yellow band is just what's coming from the rooftops. So now you're getting this flood of just rooftops solar generating in the middle of the day. And as batteries start to become cheaper at the residential level as well, we should be able to soak up some of that middle day supply and supply in, into the evening peak where it's, where it's most valuable. Now what does this mean for just general economic growth? Well that's part of the research agenda I've been pushing is thinking carefully about what cheaper power does to things like employment wages, industrial locations. And you get fairly heterogeneous effects when you think about general wage growth coming from, let's just say cheaper power prices across, across the country. And there'll be many places where just cheaper power alone is gonna drive significant wage growth for, for local economies. So put Salt Lake City in there, you know, salt Lake City if power prices come down the way that you would expect given the falling capital costs for renewables. So Salt Lake City can get something like 5% wage growth just from that alone in the background of, of great technical change that's occurring in other industries. So we think this is, this is quite positive.

The real question, the open question is, you know, can we give the rest of the country that, let's say the Midwest, the northeast access to the abundant generation opportunities that are available in the west of the country. This is a map of simplified representation of the grid of the US showing the transmission linkages between different areas. So the red lines are more transmission capacity and you can see kind of the east of the country is very, very well connected. There's a lot of power flows flowing between areas in the east. There's not a lot of connection capacity between the west of the United States and the eastern part. There's, you know, the US has really three separate grids in many set senses. They're called the three interconnections. There's the eastern interconnection, the western interconnection, and then Texas just does its own thing. There is some lines that move across these different grids, but they're relatively limited in terms of capacity and, and it's an open question how much we need to build to really take advantage of, of these cheaper, cheaper generation areas. You know, we did some more, some modeling and said that, you know, if you could take advantage of that cheap generation and give everyone in the country or every area in the country access to those cheap solar resources, you would generate something like $300 billion in additional GDP every single year. So no doubt it's expensive to build out transmission lines. We know something about how expensive it is, but you know, really any sort of capital project can be justified at a national level if it's generating $300 billion a year. So we think that's quite positive. And then, you know, what can you do really with ludicrously cheap solar power? You know, left this slide blank because I think it's a bit of an open question we don't really know as power prices come down to the level that's implied by these capital costs, you know, we start to open up not just avenues for, for like data centers and ai, but things like cheap desalination, hydrogen, all kinds of things that are uneconomic at current power prices suddenly become possible when you have just abundant renewable energy. So, you know, this is kind of exciting and I think, I think is worth thinking about. And yeah, it's, it's kind of an open question so I'm getting towards the end now. I just wanna bring it back to the, the volatility we've seen in fossil fuel markets just in the last couple of weeks. You know, the US really has two distinct phases of growth in its history. You know, the first is really driven by the expansion of fossil fuels first coal and then oil. And from essentially the civil war, up until the 1970s fossil fuel prices were fairly, let's say predictable. Over time there was some volatility but not a great deal and they were very cheap and this drove massive economic growth in the US and mass industrialization coming from these cheap sources of energy. Post 1970, that era of cheap fossil fuels essentially ended and we saw a big increase first with the arable oil embargo. And then in subsequent subsequently in the two thousands, big increase in the average prices of energy, but also in their volatility. And we've seen that in just the last week or two with get, with oil going from $70 a barrel to a hundred dollars a barrel and maybe even higher, this mass volatility causes significant inflationary effects in the US and around the globe and exposes industrial production to big swings in the cost of energy, which you know, is a big crimp on growth if we can replace this kind of volatility with abundant renewable energy. You know, we did a, you know, calculations on what that means for economic growth in the US and we came up with some kind of interesting figures that just removing that volatility from fossil fuels could generate something upwards of an extra half a percentage point in growth a year, which doesn't sound like much on a year to year basis, but if you extrapolate that forward over decades, you are looking at a, a resumption in the, the miracle years of economic growth from say the fifties and sixties. So I think the, the trends are exciting, the possibilities are exciting and when you think about energy, I just want you to encourage you to think that it's not all doom and gloom that, you know, oil grabs the headlines

but in the background we're seeing this consistent trend of increasing renewable penetration and there should be cheaper power coming with it over the coming decades. So maybe I'll just stop there and take any questions that you have. - Amazing. Thank you so much Professor Walsh. Okay, we have some questions. How should business leaders think about balancing sustainability goals with cost pressures and shareholder expectations? - Yeah, so this is an interesting one. I think they're not as in conflict as they might seem and definitely the trade off has improved over time. So if you think about sourcing renewable energy for your needs, if you're a manufacturing plant or a, or let's say an AI data center, you know, the dominant reason to do that is no longer sustainability. It's no longer about driving, let's say corporate social responsibility or driving reductions in CO2, now it's about sourcing just the cheapest available power that's not fossil fuels anymore. So that trade off, at least for power generation is relaxing over time. Now, you know, in various other areas of the economy, de decarbonization in industrials, there is still a trade off between green goals, let's say and, and raw costs for shareholders. But there too, we're seeing a lot of innovation. So I think these trade-offs are becoming less stark over time. - Thank you. We have a question from Azi Ahmed, thanks for joining Azi. Azi writes wondering what the lifespan is for grid scale batteries and what happens to them at the end of their life. Yeah. Is that captured in the uni unit cost benefit economics? - Yeah, so it's a great question. So most of these grid scale batteries are rated for enough cycles for about 10 years currently, and then they, they do need to be recycled and repurposed. So I think this has people worried, but you know, honestly for as for recycling challenge, it's nowhere near as much as we recycle daily for steel and aluminum in the US economy. And if you think about the material or the waste externalities of this battery recycling, they're going to be significant, but they're gonna be orders of magnitude smaller in environmental damage from the amount that we just burn every single day of oil and coal. So I see that more as kind of a transitional pressure that we will work, need to work out a way to deal with, but is definitely not beyond our abilities to do so. - Okay. We have a question here from Steve. To what extent does the evidence show whether or not solar and wind are positioned to meet the rapidly growing demand from data centers? At this moment? It appears that developers are arguing that some mix of natural gas and nuclear will be needed to meet demand. - Yeah, I think you're, I think you're right in the very short term. So if we are talking about if you wanna meet firmed power supply 24 7 power supply for an AI data center this year, you won't be able to do it economically with solar and batteries alone. You'll probably need some mix of co-located gas and renewables now. Yeah. So that, that is, you know, I think the short term answer is gas is going to be needed over the next five to 10 years to meet the demands of these hyperscalers. No question nuclear, I think I'm much more skeptical on nuclear, just as we said, there have to be big regulatory cha changes at the state and federal level to get nuclear even afoot in the door. You know, we've seen, we've seen big tech companies purchase and restart mothballed nuclear plants. That's a very different proposition to building new nuclear from scratch, which if we've see if the rest of the world is anything to go by is an endeavor of 10, 20 years to get one off the ground. Yeah. If you are, if you are building an AI data center next year, it's not gonna come from new nuclear. There's no way. - Our next question is from Travis, given that the United States has become a net exporter of oil and continues to benefit from global oil market dynamics

and dollar denominated energy trade, do you think US policymakers face any conflicting incentives when advancing poly policies that accelerate the transition toward renewable energy? - Absolutely, and I th I think you've nailed it implicit in the question. Now a global exporter of oil is somewhat different to incentives for power generation. Oil doesn't really exist more than microscopic percentages as generating electricity on the US grid, so they're somewhat separate, but it is true that they, they seem to be conflated in policymakers minds that there's fossil fuels and then there's renewables and this administration certainly is doing as much as they can to promote production of fossil fuels and export of fossil fuels. But yeah, I think the, the market has already spoken on the generation side. If you let the free, you know, the free market say as in Texas run, you don't get a lot of new fossil fuel generation, the cost is just too high. You know, there's separate conversations we can have about oil dominance in, in transport because that's a much more contentious area, I think whether EVs are going to take off or not. But on the power generation side, I think it's, it's already solved. There should be no incentive problems for the US to move towards renewables just on a cost basis alone. - We have a question from Amelia. Amelia, thank you for joining. Will the Lee batteries be something that homeowners can purchase to use along with their sono solar panels? - Yeah. - And if so, are there regulations against these batteries in light of recent battery fire seen with e-bikes? - Yeah, so, so that's, that's starting to come online now in the us Again, it's nowhere near as advanced as the rest of the world. I mean, Tesla makes home batteries that you can purchase today that are, you know, extremely safe in terms of fires. I think this is something where the perception of the consumer hasn't quite caught up with the latest generation of technologies. Now they're still relatively expensive in the us Part of that again is permitting and soft costs. But as we get, let's say more production out of China in home batteries that's coming in line this year and next year, that should put downward pressure on those prices to the point where they start to become economic at the residential level for ordinary homeowners. So I think again, that's going to change the game coming soon. Oh, you're muted Elisa. - Sorry about that. We think I've got this down by now. Okay, so we have a question from Mark. Given the increasing economic efficiency of solar wind, how useful, beneficial are financial incentives for consumers? - Yeah, so yeah, this is an interesting one. I mean, my personal view now is that we are past the point where significant subsidies are needed for these technologies. So of course under the Biden administration we had the inflation reduction act, which put in place significant subsidies for the generation and production of these technologies. And in my view, they were way, way too generous given how cheap these technologies already are. That's gone now. The current administration has, has essentially gutted most of the inflation reduction act and we are not seeing significant drying up in the pipeline of these technologies. So, you know, I don't think subsidies are necessary on the utility side, on the household side, you know, to get costs down. There could be an argument for continued, let's say state level subsidies for rooftop solar and batteries maybe to take pressure off the distribution and transmission grid. But I think they're, they're not at the point where they're needed to drive adoption anymore. Things have just changed so rapidly in the, in the last 10 years that they would be more the cherry on the top and potentially not very cost effective. - Okay. Do you see, I'm sorry, my computer has just frozen and now I can't see. Okay, here we go. Do you see clean energy as a force for reducing global inequality or could it widen existing gaps? - You know, it's interesting. Yeah. Clean energy in particular is a very democratic

decentralized form of energy. It's just such a different way of generating that. Like I was saying in the talk, I think different mental models are needed to think about it. Let me give you an example. Take Pakistan. So Pakistan in the last two years has imported about 40 gigawatts of solar panels entirely from the household sector. And the reason this happened was because, you know, Pakistan had extremely inefficient, centralized generation with high prices that households just couldn't meet by connecting to the grid. And solar panels got cheap enough coming in from China close by that households have been able to bypass the grid entirely and just set up panels and very cheap, let's say lead acid batteries at the household level, giving them a way of kind of getting power, let's say power over their own power supply in a way that doesn't depend on centralized government, which in this case in Pakistan's just very inefficient at supplying energy to desperately people who are desperately energy poor. You know, decentralized solar is a technology that by its very nature leads to distributed generation. So I see it as a force for, for compacting, perhaps not inequality, but, but alleviating let's say energy scarcity in a way that doesn't require big centralized investments in places like Pakistan, but also Sub-Saharan Africa in particular. - Well, thank you so much Professor Walsh. Thank you. This was really insightful, a big thank you to our attendees for taking the time to be with us during this talk. A quick plug, our next CBS faculty live virtual talk will be held on March 31st at 12:00 PM Eastern with Professor Brett house on understanding the global economic context. He will discuss how we equip business leaders to understand emerging developments and long-term trends in the global economy and interpret what they mean for their organizations. So please join in again soon. Thank you again Professor Walsh. I have hope and I really appreciate - You should have hope things aren't as bad as they seem. - Thank you so much everyone. - Thank you.