 So good afternoon and welcome to today's Energy Seminar. It's my great pleasure to introduce Hal Harvey, who we were just reminiscing about this with a few of you, including Gil Masters, was a undergraduate in ME design here at Stanford, and then a civil and, I don't know if it's even called Civil and Environmental Engineering, and maybe it was called Civil Engineering. Gil, where are you? After that, and worked in a program then known as infrastructure planning here, I think, which is mostly energy planning as far as I remember. I actually knew a lot of house classmates, but not Hal. So I met him later. He's now the CEO of a group called Energy Innovation. I was thinking about kind of his background. It's kind of interesting that we all know these very big important institutions like the Energy Foundation, the Climate Works Foundation, the Environment Program at the Hewlett Foundation, and lo and behold, unknown to many of you probably as Hal was kind of the founder and CEO of all of the above, on his way to his current job at Energy Innovation, which of course in the post-perest period is becoming more and more active and more and more progressive, I might say, for based on our earlier, our discussion earlier today. So with that, I'll let Hal talk about what he's doing and how he is still an optimist. I saw those videos a couple of years ago, an optimist on climate change. I assume you still are an optimist, sir. Institutionally, if not practically. Okay. With that, Hal, I'll take it over. Thank you. Thank you, John. It's great to be here. There's obviously lots one could talk about in the climate world. What I want to talk about is essentially triage. What are the minimum number of things you have to do in order to win? The subset of that being how to design energy policy. There's a great deal of energy policy in the world, and we've studied a lot of it and found that much of it is what I would call decorative policy. It just doesn't matter that much. There's about a dozen policies that make an enormous difference, but it turns out you have to not only choose which policies to pursue, but you have to design them very carefully. The literature on policy design in the energy space is thin, very thin indeed. Let me start out with some obvious slides about why policy. The world spends about $5 trillion a year on energy, and another $6 trillion or more dollars a year on the infrastructure that sets future energy patterns. So cash flow is not the problem. When you hear about all the battles in the UNFCCC process about assembling a climate abatement fund, what they're trying to do with all these heroic meetings is come up to $100 billion a year, which is about 1 percent of our current spend. So if you get the right policy, you either land money on the tar sands in Canada or on utility-scale solar fields, for example. Obviously, those are different energy realms, but the point is clear. National security, do we import most of our fuel or do we generate it closer to home? Transportation is obviously another set of choices. Do we try to fill as much of the continent or I should say the world as possible with automobiles, or do we build systems like this bus rapid transit system in Guangzhou that handles many more people with much less frustration? Let me just give a quick illustration about this. This is bus rapid transit, many of you have heard of it. This was built in nine months. It has the speed of a subway and the capacity of a subway system, and it costs 5 percent as much as a subway system. It requires really intensive, pretty brilliant engineering and systems optimization, but it doesn't require heroics. What they do is obviously dedicate lanes for buses only. They create stations instead of stops, and the stations are the same level as the stop. You pay to get into the station rather than is paying to get into the bus. The buses are double in trouble long in some of these systems, so the bus rolls up, the door is open, just like a subway people get on and off, it goes. Then they have transponders to turn the lights, green. This is not heroic stuff, but it dramatically changes the options for people in a city, and that's down to good policy. Most people in the energy space fall in love. They fall in love with a certain technology. They love nuclear, or they love solar, or they love coal, and they also fall in love with certain types of policy. For example, a carbon tax will solve everything, or a carbon tax, a carbon cap is an evil scheme, or what have you. I think falling in love makes you a bad judge. I should know. So what we've tried to do is organize policies into three types. Economic signals, fairly straightforward. Do you subsidize fossil fuels or not? Do you have a carbon tax or not? Performance standards have been the big winner by far. How much energy does your refrigerator consume to keep your food cold? It turns out that the state of California has a lot to say about that. It's a performance standard, a renewable portfolio standard is another performance standard. It says that electric utilities will have a certain minimum fraction of their energy from renewable sources. So performance standards have been the standout winner here, and then of course research and development. This drives the cost of everything down. These three types of policies encompass pretty much the whole lot, although there's a few other things that are important, and have really important interactions. If you have strong performance standards, for example, you reduce the price shock of a carbon cap. If you have a super efficient house because of a good building code, which is a performance standard, then a carbon tax doesn't affect you that much. Pretty straightforward. Conversely, if there's a carbon tax, then the performance standard seems like a more rational choice. Right now we're battling each other sometimes. We have strong fuel efficiency standards for cars and super cheap gas. So you have this policy battle instead of policy reinforcement. Economic signals also support R&D. If the price on carbon is higher, firms are going to invest more in R&D. And conversely, if you have good R&D, the carbon tax doesn't cost you very much. And then finally, R&D reduces the cost of performance standards and performance standards create markets for R&D. And if you don't believe that, go look at what's happening at Volkswagen right now when they try to become very quickly the electric vehicle producer of the world. They kind of screwed up on one of these performance standards. So they're investing very heavily in R&D right now. And it works that way. So this is a really good topology to help think through the opportunities in energy policy. There's a lot more to it though, and that's where I want to go. Each of these has an advantage. Economists love economic signals because they simultaneously help your purchase decision and your consumption decisions. So if you have a very strong gasoline tax, you're gonna buy a more fuel efficient car and you're gonna be a little more thoughtful about when and how often you use it. A good economic signal doesn't require a lot of government intervention. Although people will argue a carbon tax is simple and a carbon cap is complicated. And I implore you, if you believe that a carbon tax is simple, go look at the US tax code, which is about 10,000 pages long. So a tax can be horribly designed or well designed and so can a cap. And we have plenty of experience with that, the Waxman-Markey bill, which almost became law was 1,200 pages long and that's probably the wrong length of a bill. So economic signals have their adherence and they can do a lot for you. There are some areas by the way where economic signals are essentially beside the point. The people who design your home are not paying the utility bill for it. And the people who built your home are not paying the utility bill for it. So what's their incentive to wrap the ducks really well or put in super efficient windows? What's that? Demand, but the market signals are very attenuated. So when you wanna buy a house, you're looking for a school in the neighborhood. You want a quiet street, maybe like granite marble tops or a fireplace over here. Even if you care a lot about energy, it's gonna be number 10 on your list for most people. And even if it's number one on your list, good luck figuring it out. How are you gonna compare this house to that house? You can't really do it very easily. So the market signal simply fails in buildings. Turns out it fails also in transportation. And here's a quick thought experiment to that end. Carbon tax is proposed in the US or typically around 25 bucks a ton rising with inflation. That's 25 cents a gallon, easy calculation. So 25 cents a gallon is in the noise compared to the variability of gasoline prices anyway. If you really wanna find out how well carbon tax works in the transportation sector, pretend that the taxes in Europe on fuel are carbon taxes because they're about 600 bucks a ton. We're not about to do that. And yet they still have to have fuel efficiency standards to drive performance changes. So there are realms in which the carbon tax is essentially beside the point. There are other realms where it's a very good deal. And then there are some that are just plain confusing like utility dispatch models because taxes tend to be passed throughs in a highly regulated utility system. All right, I think I've covered all this stuff. Performance standards have been, as I said at the beginning, the big, big winner. And we have them all around us and we take them for granted. When economists tell me they don't like performance standards because there's something arbitrary about them, I ask them, in the last hotel they stayed at, did they get a slightly cheaper rate because they didn't have fire suppression in the roof? Or were they willing to pay a little extra for having rebar in the foundation so it didn't fall down during an earthquake? So the point is we have performance standards all the time. When you buy meat in a supermarket, you expect it not to be poisoned. That's a performance standard. And it turns out that you can have performance standards for lighting or refrigerators or air compressors or anything else. A well-designed performance standard, asterisk, asterisk, I'm gonna get to that in a minute, is a dramatic way to affect performance. Their principle disadvantage is what I said earlier. They're at war with price signals. So if you only have performance standards, you create demand, demand crashes supply, and so you don't get where you wanna go. The last one is R&D, and this too has some adherence, that if you have enough innovation, all the world's problems will be solved. And I think Bill Gates is the emblematic superstar for R&D. I'm a strong proponent of R&D, and in fact, energy is profoundly a technological business. If you look at all the incredible advances we've had recently in solar and wind, light emitting diodes and everything else, it's because of new innovation. But R&D doesn't happen in a vacuum. People are not gonna invest mightily in R&D unless there's a damn reason to do so. And you can't land on the fluctuating prices of commodities as a sufficient rationale to create the R&D for, I'm kind of talking like the FedEx guy, right? Just a little. If you wanna understand how much we invest in R&D, public and private energy R&D together accounts for considerably less than 1% of our energy budget. We spend more on potato chips in America than we do on energy R&D right now. We spend much more on all salty snacks put together than we do on R&D. So we're foreclosing the future if we fail to spend money on R&D. Now obviously if you have strong performance signals or strong economic signals, it encourages the private sector companies to invest in R&D. So R&D you can think of as a push, pushing new technologies into the market and the other two are pulls to pull new technologies into the market. What I wanna turn now to is for all these policies, we analyzed which policies are the most effective in the world. Which ones drive the fastest carbon reductions, the greatest technological innovation? And what do they have in common? And what are the screw ups that people make along the way? And you don't have to read these all because I'm gonna walk through them one by one. We came up with six that are kind of killer apps if you do them right. And this is important stuff. It's not well understood stuff. You will not find, there's not much literature on this. So let me go after them one after the other. The first one seems fairly obvious which is you should have regulatory certainty. If regulations pop around year to year, it's really hard to build a business model. This is a US windmill installations year by year. So the production tax credit which is an economic signal for wind was authorized every other year and then it would expire. And so every year, a bunch of windmills got built and then next year none. And then a bunch and then none and so forth. They transcended this problem for a few years and then they got back into this habit. So imagine running a windmill company where you have to rush like crazy to build all your windmills in a 12 month period and then the next year you have to lay everybody off. It's just insanity and yet that's how we do it. Now the only rational explanation I've heard for this is if you're fundraising and you're a US senator you have a shot at the kitty every two years. But this is insane. By the way regulatory certainty happens across the board. If you wanna put up a solar, a utility scale solar field in Southern California in the desert and you don't know whether it's likely you'll get your permits or not, you will spend tens of millions of dollars in many years figuring that question out. So if instead sites were pre-zoned, green you're gonna get your permits in 90 days, red forget about it, it'll never happen and yellow we're gonna go to war which is today everything. Then you could save a lot of money and prevent a lot of ecological harm. So regulatory certainty is an elixir for reducing costs for clean energy. The next one is long-term time horizons and this one is probably fairly obvious but if you wanna build a super efficient car you have to invest money in R&D, aerodynamics, engines, transmissions, tires, electrical loads, light weighting, all kinds of options. You have to do market research, figure out what kind of car you wanna build. You have to design the damn thing. You have to build a factory to produce it. You have to market it and then you have to extract value from it for half a decade to a decade, right? So you stretch that out, that's 15 year cycle. And if you don't know whether the fuel efficiency standard, what it's gonna be in 15 years you're shooting in the dark which is an incredible waste of money and it's an incredibly bad way to prosecute your policy. So you're not really in the policy game if you can't create long-term targets. This is an illustration of California's AB 32. It's got a 2050 target but until last year it only had a 2020 law, 2020 is pretty soon. So what was happening is people quit buying carbon permits because they didn't know what was gonna happen after 2020. So you can see the regulatory uncertainty manifests in the economic signal for that. We got locked into law a 2030 target now with a 2050 indicator. So that's a great thing but we need to think out a ways longer. By the way, it's like cloning money if you create regulatory certainty plus long-term targets because you've lined up everybody's business plans to what your social goal is. If you fail with those two, you're jerking them around. You're not doing yourself any favors. The third one I like to push is that you should have technology, neutral and price finding, these are actually separate items but if you wanna abate a ton of carbon you can do it with solar or wind or energy efficiency or geothermal, there's a bunch of ways to do it. And if you think you can sit in an office today and predict what's the right choice for everywhere in the country, three years or five years from now then you're fooling yourself. That's kind of the Soviet style of planning didn't work out so well. So you should be technology neutral to the extent possible, big asterisk here. There's a number of technologies for various reasons. They're outside the realm of competition but they're highly valuable to society that you wanna drive in in which case you cannot be technology neutral. Technology neutral competition is not gonna get you advanced nuclear power or carbon capture and sequestration or advanced biofuels. It will give you the best choice between the efficient light bulb and solar and wind. By the way, it wouldn't have worked with solar six years ago, but it certainly does now. Price finding is the other one. Whenever you specify the price for something it will either be too high or too low. It will never be right because we live in a dynamic world with dynamic technologies. This shows the levelized price of solar dropping by 80% in about seven years. So Germany bravely offered more than 20, I think 23 cents a kilowatt hour for their feed-in tariff in the early days. Now what people don't understand is they stair-stepped it down at least six times since then. So they've had kind of a price falling rather than a price finding policy, but last year the cheapest solar in the world put in or this last year was under three cents per kilowatt hour in Mexico and Dubai, two and a half cents per kilowatt hour. That's the cheapest power ever been supplied on Earth in history. And yet Japan was offering 23 cents a kilowatt hour and its feed-in tariff for solar. So it was not a price-finding policy. So Japan was paying about seven times the market rate for solar, which means they were buying one seventh as much as they should have been buying. And that's no way to reduce carbon dioxide emissions. So policy should be technology-finding for most technologies and price-finding as well. This is one that is incredibly important and I'll tell you two quick stories about this. When Jerry Brown was the youngest governor in California's history, he got passed a building code called Title 24 and Gil Masters made me go to Sacramento and watch the hearings for that. So Title 24 is, in my mind, one of the most brilliant pieces of energy policy ever written, but one of the things that's got in it is it gets tighter every three years without going back to the legislature, without a political battle. And the way they did that is they divided the state into 16 climate zones and they created a merit rule which roughly goes any technology that pays for itself in seven years gets rolled into the next set of standards. So if you make triple glazed, low emissivity, argon gas filled windows that in the Sierras pay for themselves in energy savings in seven years, their performance spec rolls into the code. So now Jerry Brown is the oldest governor in California's history and the code's gotten tighter every year, including the Duke Major years, the Wilson years. They used the political bandwidth once and they created, there's amazing improvements where we've cut energy use by much more than half compared to pre-code buildings. And by the way, they're now talking zero net energy buildings. The city of Santa Monica has now mandated that after 2020, all new residences have to be zero net energy. Zero is a really nice number. Should pay attention to that. Let me give you the counter example I'm jumping ahead, but for continuous improvement. When Gerald Ford was president, he doubled fuel efficiency in cars from 13 miles per gallon to 26, which was between 1974 and 1985. And then we went to sleep for more than 30 years and we kept them there. And then when Obama became president, which is a long time, they doubled again from 26 to 54 miles per gallon. So if instead Gerald Ford had said three or 4% improvement per year, that one line in his law, and we increased every year, we would have saved over a trillion dollars that we sent to countries that hate us. And we would have saved the auto industry because it would have been a long-term certainty with regulatory, long-term signal with regulatory certainty employing continuous improvement so automakers would have known to innovate in transmissions and engines. And instead, we innovated in cup holders. So we bankrupted the American auto industry. We released infinite amounts of CO2 into the air. We spent a trillion bucks of people that don't like us because we failed in that simple policy design principle. Continuous improvement is really powerful. And then back to the original point, it's not powerful least because of political bandwidth problems. Any serious leader, a governor, president, secretary can only do a few things in their tenure. So lock something in that's going to be the gift that keeps on giving. Reward performance, not investment. When I was here at Stanford, we had this standard offer for contracts, which and wind, sorry, scratch that, wind mills we paid according to how much they cost to build was the tax credit, not how much energy they produced. In China, they're required to build a certain number of gigawatts of wind a year. They're not required to hook them up. They don't have to spin. They just got to build them. And so China has the largest curtailment as much as 40% of windmills are not producing energy in some provinces because that wasn't the incentive. So it's a straightforward idea to reward performance, not investment. But when you're in love with tax credits and capital expenditures, and many politicians are in love with these two things, they tend to reward investment, not performance. So that's backwards. Two more. If you have plowed your way through carbon cap legislation here or in Europe, you begin to realize that there are better and worse points of regulation. So I always argue, go upstream. Go to the mind, mouth, and the well head, or the blending gate for petroleum. Find the pinch point where there's the fewest number of actors making the largest quantity of decisions and regulate that, tax that. The inverse is when you're trying to regulate millions of people and also trying to incent things that don't happen. So if you create a carbon cap and you want to give somebody credit for changing the light bulbs in the supermarket, you have a regulatory nightmare because you have to do an engineering estimate of what didn't happen compared to what did happen and you have to go to every damn supermarket. And yet that's how we do a lot of policy in this country. So I just say, go upstream. How am I doing on time? A little more? OK, OK. What I thought I'd do now is walk through the transportation sector and see what happens when you apply these policies one after the other, and then compare that to what the real world does in Japan, in Europe, in the United States, and in China. So if you want to clean up transportation, you can make vehicles more efficient. You can make fuel have lower carbon intensity. Or you can have people drive less because there's better options out there. And each of these are susceptible to each of these things. I'm not going to do the R&D part of this because I think it's self-evident. The standards for clean vehicles are vehicle fuel efficiency standards, CAFE standards, as they've been called. There's a lot of ways you can screw those up. But they're incredibly powerful if you get them even close to right. The standard in clean fuel, there's really only a few of these in the world. But the low carbon fuel standard in California is the best designed one, except for it being physically impossible, about which more in a minute if you want. But what they did is they basically said, and the oil company signed on to this, failing to do the math, they said, if a unit of gasoline today emits 100 units of carbon, next year it can only emit 99, and then 98, and then 97. It's a performance spec. It's technology neutral. It's got a long-term signal. It's just really, really hard to get there. But it meets a lot of the things. They just didn't pay enough attention, in my opinion, to the real-world constraints there. And then alternative mobility. This is urban planning. It's a kind of performance spec. People should have access to mobility without being in a car. That requires so many lane miles of bike path per kilometer, and decent sidewalks, and public transit within 500 meters of highly dense housing, and so forth. There's a small number of design specs or performance standards for urban planning. This is, by the way, also an understudied field that deliver the goods that really get you there. We've been doing a lot of work with the Chinese, both cities and development banks, on laying out these specs so that they can be used generically instead of one by the other. The fiscal signals are more straightforward. Fee bates are an interesting way to clean up vehicles. In Sweden and Norway, the purchase price of a car is proportional to the fuel it uses. So an S-class Mercedes in Sweden or Norway costs twice as much as it does in America. Because the tax is about 100%. There's some kind of goofed up fiscal policy in China where they set the tax according to the engine size, the displacement of the engine and leaders. And so as soon as you do that, everybody puts in a little engine and turbo charges it. So they didn't get it quite right. They didn't reward performance. They rewarded a different specification than performance. If you don't reward what you want, you won't get what you want. It's pretty straightforward. And of course, this happened. Now, every car in China has a turbocharger on it. So straightforward. With fuel, it's fuel pricing. And this is from the gas that they pay you to take in Venezuela and Iran to the gas with a very modest tax we have in America to the gas with a very high tax in Europe. And then, of course, an interesting realm with alternative mobility is starting to price congestion. This is a realm that I call car control after a topic that Ann and Paul Ehrlich pioneered, population control. The real population we need to control might be cars. There's over a billion of them on the planet now. There are a half a dozen fascinating ways to do this that actually blend some of these. But just to give you some examples, you're familiar with the congestion pricing in London where it costs 25 pounds per day to drive an SUV in downtown London. They call them Chelsea tractors. So the ratio of Bentley's to Volkswagen's has grown a lot. On the other hand, so has public transit and walking. In Shanghai, you have to buy a permit to buy a car. And it costs about $13,000 now. In Beijing, they went up this way, you have to win the lottery to buy a car. But the leaders in Beijing and Shanghai know there's way too many cars. So they're putting a cap on the amount added every year. You actually need to get into the subtraction business, which is a little bit harder. Tokyo, you have to prove you have a parking place before you can buy a car. That's an interesting performance standard. San Francisco, you know what they do here? They change the parking meters depending on the congestion, depending on how many of them are full. So it's very expensive to use a parking meter in San Francisco if most of the parking meters are in use already, which is what it should be. And my favorite of all is Copenhagen, where they monitor all the parking meters, and when they're 80% full, they turn all the lights on all the main roads coming into the city red. And they leave them red for a minute and 45 seconds, then they turn them green for 15 seconds, and then they go red again. They're just saying, we're full. It's a pretty good signal. OK. So how do we do against those principles? I've actually walked through some of these, so I'm going to skip this. Sorry, it's a long slide. OK. So let's look at fuel efficiency standards, how we've done it in the US and what we need to do. In the US, first of all, there was a truck loophole. Everybody remember that? If you buy a pickup truck, you get lower standards. And so everybody redefined their car to be a truck. And it was pretty easy to do that. You just change a few characteristics in the car, and it gets to be called a truck. Then they made them weight-based. And so one of your principle technological strategies for increasing fuel efficiency of cars is to de-mass them. And they basically said, don't do that. Don't do that, because your standard is going to be stricter. Weight's not the issue. Size is the issue we should care about. They put stair steps on the function instead of a continuous function, because it's easier math. I don't know why, but that makes no sense. You can expect everything to be crowded to the stair step, just like wrestling. Everybody wants to be exactly 130 pounds, not 131, because that's where the cutoff is to the next weight class. In wrestling. So if you flip it, and also they regulated fuel consumption, which is miles per gallon, which is intuitively interesting, except if you think about it a little bit, if you increase the fuel efficiency of a truck from 10 miles an hour to 10 miles per gallon to 20 miles per gallon, you save half the fuel. How much do you have to increase it to save the same amount of fuel after 20? To infinity, exactly. People think well to 30, or maybe they think double to 40, but no, you saved half the fuel. Now you want to save the other half. That's infinity, right? So miles per gallon is the inverse of what you care about, which is gallons consumed or carbons consumed. So flip it around. Make it grams of CO2 per mile. Grams of CO2 are after grams, not fuel, right? So this handles your carbon intensity of the fuel. Make that the x-axis be your footprint. Size of the car. If somebody needs a big car, they need a big car. If they want a big car, they want a big car, rather than mass. And then move that slow, and be very careful about setting the slope. If you set the slope wrong, you jam people up to one end or the other end of that thing. And then move the whole bar down at a predicted pace of say three or 4% a year. And all of a sudden, you get dramatically different results. A couple of the things you gotta do is have realistic test procedures. So we found out, European test procedures are spectacularly bad. I'll just give a couple examples. You can certify your car through any private sector actor in any one of the 28 countries in Europe. So that's kind of goofy. And they don't have public facilities to do this. You can create what they call a golden car. You take the seats out. You put in a little tiny gas tank. You fill the tires with liquid so they have less rolling resistance. You tape all the seams so there's less aerodynamic resistance. This is all legal. And VW sold 125th as many cars in America as they did in Europe, but they're paying all the fines over here because the enforcement is not a civil penalty over there. So they basically say, cheat as much as you want. And by the way, if you get caught, we're not gonna really bother you that much. Is that your policy? Clearly? You get the result you might think about. So these fuel efficiency standards for cars are an incredible policy if you have continuous improvement, if you reward the right things, if you test them properly, a few other things you gotta do. The US, by the way, is moving strongly in this direction. They've handed control of fuel efficiency standards. They don't call them that anymore. They're now carbon standards. They've taken it away from, they haven't taken it away from NHTSA, but they've superseded them with the Environmental Protection Agency regulating CO2. So you have two agencies from a physics perspective and identical mandate and from an operational perspective, totally different constituencies and totally different methods. It's a perfect setup for a conflict. I'm gonna wind up and I'm gonna skip this one. It's about designing rebates and go to the last, my last slide. I argued at the very beginning that there's a small number of policies that are incredibly important and a large number that are essentially decorative. And so this is my list. And every time I put this up, somebody raises their hand and says, well, you forgot this or you shouldn't include that. But my point is not that this is exactly the right list, but certainly it's mostly the right list. And it's a small number of policies that make dramatic differences. If you ask electric utilities to increase the amount of energy they produce from renewable sources by 3% a year, every year, you're gonna transform the utility system. If you increase building code efficiency, we already talked about that. It has a dramatic effect on that and so forth. If you try to build economic signals that don't directly conflict with your performance standards, that's a pretty nice thing to do as well. The reason this is important is because if you read energy bills, you don't find this. They're not measured according to their carbon abatement potential. They're generally not measured according to their economic incentive. They're not tested according to whether they drive a technology dynamic. They're just titles. They're integrated by stapler is what they are. If you look at an omnibus bill in America, it's a sad thing. And it's not that different in other places. We have the benefit of energy modeling in America, but most countries don't. So I'm gonna wind up with this. Let me just say one last thing because I know it'll come up with respect to our new president. So pursuing this requires political vigor and intelligence and commitment to do the right thing. Most of these are or can be done at the state level. And that's really important. California is allowed to set its own vehicle performance standard and is elected to do so. And so if we soften the new CAFE standards, that's fine. California is gonna do a Cal exit legally, has done so, and 10 other states with 50% of the country's population are already on board to follow them. Building codes are set state by state. California is allowed to set energy efficiency standards and other states are allowed to copy them and they do copy them. Urban design, clearly not a federal priority. Renewable portfolios, public utilities commissions, there are 50 of them in America. There's zero in Washington. There's the Federal Energy Regulatory Commission which probably can't muck us up too badly. Utilities scale energy efficiency programs, also state by state. We have a carbon cap in California. That didn't require the federal government. I'm not saying it's a good thing. We're gonna lose a lot of good stuff. But it's not as bleak for this purpose as one might expect. The last thing I'll say here is the wind, the technological winds and the economic winds are very definitely at our back. I'll just give a couple examples. We just sponsored a study that says in Colorado you can build a brand new wind farm, all capital costs, all installations costs for less than simply the operating cost of an existing coal plant. So that's a nice threshold to cross. And here's the other thing. Commodity prices are random walks. Oil, coal, natural gas. They go up and down over time. Technology prices tend to be one way vectors. And so when a technology price like solar or wind crosses a commodity price, especially a low commodity price, you win. So you can try to resuscitate coal but you have to revoke free market economics, which we will do for a while. But that's a very interesting tension. So it's not this way across the board. There's still some areas like liquid fuels for transportation where we have a lot of work to do. Electrification of transportation is happening at a pretty amazing speed, though. So I would say focus on the things that have to get done, the small number of policies that make a big difference. Try to make sure they're well designed and prosecute them with vigor at any form that allows you to get it done. And I'll leave it at that. So that was great. We have about 10 minutes for questions. I actually remembered in that talk what I really love about Halis. He's kind of the guy in the world that's at the point of the policy prices where the rubber meets the road. I also feel he's one of the best critiquers of the modeling analysis game which I participated in because he knows when to use them and when not to use them. And I think both are important for the further improvement of that discipline. Any questions? Students first in the back. Come on, you must have questions. There we go. Hi, a lot of these policies that you discussed require being extremely clear on the ultimate objective of the policy. And when it comes to actually passing policies that can often be a detriment to building a coalition that can actually get these policies passed. So passing functional policies now often be sort of at crosswinds with political feasibility. And I wondered if you could comment on how you see that those two intersecting and which of these six policies are most important to stick by and not bend in the way the politics. Well, they see politics as the art of compromise and it certainly is. I guess I'm of the school that trying to accommodate every interest ultimately significantly reduces your political chances rather than increases them. And maybe this is the Waxman-Markey lesson, right? That they had to make exceptions for Alaska and then the Midwestern states needed a little more biofuels and we should give credit for this and credit for that and credit for the other thing. And pretty soon you end up with a bill which literally no one can read or should read. And the antithesis, and of course these are slightly different conditions, was Hank Paulson getting a trillion dollars with a three-page bill to save the world, which worked. So those are extremes, but I actually think the answer is not to try to placate every interest group with every bill, but to look deeply at the interests of the most affected parties. Most utilities and most auto companies will accept strong regulations if they have long-term certainty, right? What they can't stand is being buffeted around. Uncertainty is a killer for a business plan. So I think keep it simple, make your objectives clear, test the benefits. If there are people who are resistant to it. This was interesting with building codes in California back in the 70s. They discovered the politics of this where you could divide the world into good builders and schlock builders, right? The people that were not insulating ducks didn't like the code and the people that were doing it anyway loved the code. So divide and conquer, go for it. All right, other questions? Sir. You mentioned technology neutral options. But if you do this with the different profiles and generation of solar and wind and straw and biomass, don't you have a risk as a country that you lose control of your end with that? So the tacit question and what you're asking is, how do we build a stable energy system if you don't choose a stable mix? And which is especially important for the grid, right? Because you need second by second balancing. There's a lot of, the old paradigm which I was schooled on here at Stanford, power systems planning, a two quarter graduate series class was that you build three classes of power plants, baseload power plants, shouldering and peakers. And then you dispatch them in economic order to keep the system. And the demand was the independent variable and supply was the dependent variable. So the new paradigm, which is emerging is that of system optimization where you look across a suite of resources including demand resource and you build and dispatch accordingly. And it turns out there are half a dozen ways to balance the system that don't have to do with just dispatching power plants. So demand response is an enormous, essentially untapped opportunity. A study in California found six gigawatts of demand response cheaper than any supply. And there was a bid where PGM, the biggest power plant in America, allowed supply and demand to bid for fast ramping services. So breaking down the market is really interesting. A very small, quantitatively, part of the market is fast ramping services. You need to turn it on and turn it off. You can create a little bid structure just for that and let all kinds of technologies in, right? The demand side bids were 80% cheaper than the supply side bids. At some point, they leveled out. Well, that's fine. That's not a problem. Geography helps. There's very little correlation between wind in the north and the south. There's predictable non-correlation between, well, it's correlated, but they're different between solar in Arizona and solar in California. Solar thermal operates differently than solar PV. So putting in a heterogeneous mix of supply side options, adding in demand side options, connecting large areas with wire, and then paying especially for, if you have a natural gas power plant and you turn it on and off quickly, very little carbon, but fantastic system balancing and much cheaper than batteries. I haven't even got the batteries yet. The point is, there's very cheap ways to do this without having to go through the engineering mentality where you design everything in advance. Sir. You can have batteries. What are your thoughts on sensible energy storage policy? So it would be really great to have cheap, compact, high power density, high energy density batteries. But batteries are really, really hard to make. The advances in batteries in the last few years have been unbelievably great, but it's still, we're still not there where we need to be. So according to what I said below, the electricity system really doesn't need batteries until you get to 80% renewables or better. And that's pretty important. Cars need batteries. And I would say the cost of batteries have dropped at a spectacular, the performance of batteries have increased at a much slower rate, but they're definitely getting better. But batteries to me are like R and D in general. You don't need to hold your breath and wait for it to happen. You can move right away with improvements, dramatic decarbonization of the system, and that will pull new technologies, including batteries into the market. Is that, answer your question at all? You look skeptical. Yeah, I like that. So, very interesting. I don't want to talk to you about the selection by Stapler, but you're going on the performance standards before this thing. I think that's being missed. That is more important than perhaps even your own emphasis. Because for instance, the wind farm that was built up in, you know, Montana or wherever that you talked about, its performance standard is extremely poor. And the performance standard of a mass of PV is extremely poor. So why are we being saddled with an environmental source of energy which consumes vast areas of land and dangerous species? Well, we have alternatives already that we know how to use. And I mean, for years, I mean, President Kennedy knew what to do. And Burt Richter signed a plan here in California in 2011, which the state includes. Our RPS standard is a joke. Other states have much better RPS standards. Would you be satisfied with the CO2 per kilowatt hour standard? I don't know what to be satisfied with. I wouldn't be satisfied with wasting the environment and resources on things that have very low power, density, and predictability. So, you know, the fissure line if you're pardoning pun and clean energy advocacy is between nuclear and solar wind, right? Geothermal is nuclear. So what I said was, don't fall in love with the technology. Set. We're toy and level science. Set. And by performance standard, I mean, ask yourself what does society want. We want zero carbon electricity. We want reliability and affordability. And we want, so it's environmental amenity, reliability, and affordability. Set SPACs, except BIDS. That's a performance standard. And you know, the Swiss have just voted to do that by keeping that open. OK, we have one, two, three, one, Katie, and four. OK, four. Quick. I'll be quick. Quick. No, my question was just about the, you mentioned that carbon tax being about $25 a ton with regards to fuel standards. But yet, the California cap and trade program seem to have failed in that respect, or it's gone down to below $15 a ton. So I was wondering if you could comment on that, why that is. Yeah, so carbon prices have been set everywhere from about $2 to about $600 a ton around the world. So there's heterogeneity there. California has a cap which sets a legal limit on carbon for most but not all sources. And it has a floor price of about $12 right now. So if they don't get any BIDS at $12, they don't sell permits. So the consequence of this, and so there's some interesting revelations here at least from my perspective. The first is, it's cheap to save carbon. We haven't tried that hard, and we're beating the targets. We're beating them in the, sorry, in the Northeast States. We're beating them in California. We're just crushing it in Europe. The price is so low in Europe it's $2 or $3 because it's really cheap to save carbon. From my perspective, that's a good news story. We should accelerate the rate that we reduce permits to take advantage of this. The California system and some other systems have callers where they floor on the price and a cap on the price raise. It's a de facto through banking borrowing they call it. I think a caller is a very smart thing but I would design a system so if you don't sell the permits you ratchet down the number that are issued. So it's a ratchet, one way ratchet. Okay over here. Yeah, so batteries at $100 per hour with two years, over $800 per hour, over $700 per hour, filled in over 800 people at once. I'm assuming, what is that due to your calculations? Electric vehicles are gonna blow away most people's doors in terms of the rate of penetration is what it would do. If utility scale storage I think will be slower. But that's a six. Same for six beds. But again, my point about arguing with these other alternatives is that you don't have to have batteries. If you have them, they're cheap and they're long, they last a long time too, that's fantastic. So I think the transition to a renewable, to a zero carbon grid is gonna be much faster, can be much faster than people realize. I talk sometimes about four zeros. Zero carbon grid, zero net energy buildings, zero waste industry, and zero emission vehicles. Every one of those has a technological vector that's going in the right way. Everyone has a suite of policies that drive them that way. Each of those can accelerate dramatically if you combine those three sets of policies in the right way. Okay, Katie. Clean energy under Trump. And so you really emphasize the degree which states that want to move have a lot of entry points. And I was wondering if you could maybe give your best pitch for the bipartisan answer to that. So across your list of 10, what are the greatest entry points in states that are red, as well as blue, and Congress, and federal agencies? Best things for states to do. So which of these are gonna be attractive? What are the red states and the blue states at the same time? So guess which state has the most wind power in America? Texas. By a huge amount, more than three times as much as California with just number two. I worked on that when the first renewable portfolio stand in America was signed by Governor George W. Bush. Right, and Pat, one of the biggest advocates of clean energy in America was Bush's chairman of FERC, Pat Wood. So what Texas did is they set up a free market for trading electrons and they built transmission lines for wind and they lived on that production tax credit for a little while and now wind is just blowing gas and coal and everything else out of the water. I'm supposed to meet with the governor of Wyoming in a couple of weeks. He wants to talk about coal. I wanna talk about wind. We'll see where it goes. But look, they think Wyoming is a coal colony. Everybody there works on coal or gas. It's really windy in Wyoming. They already have transmission lines that go everywhere. They have a bunch of guys that know how to build stuff with steel and concrete. Just switch gears. So I guess I would say that if you can show the jobs and if you can show the right kind of jobs in the right places, you're more than halfway there. There are 2.5 million advanced energy jobs in America. There are 40,000 coal miners in America. Final word. Yeah, so I'm reminded actually of a lot of things we used to do at Bell Labs because we have long term you know, good stuff, right? So it's all that really resonates, right? We had a mantra which was never scheduled breakthroughs. Right, guys I work for end of being President Bell Labs. So the question is, do you see any things that look like they're feeling like be breakthroughs? I mean, you know, or is it most of this, the plausible cost volume learning curve stuff for a while? So I think the breakthroughs are a different kind of breakthrough now. I think the breakthroughs are system optimization rather than device optimization. Enabled by big data and control paradigms that we didn't have when I was a kid. Certainly in the utility sector. I think there are breakthroughs in market design that can help a lot. I think there's a symbiosis between technology push and technology pull that we've never taken advantage of. So for me, the breakthrough is 5% annual or 7% annual or 12% annual improvement rates rather than two or three or four. I mean, thermodynamics is an unyielding science. It doesn't have a Moore's law but system optimization is just fine with that kind of stuff. Good, on that note, thanks again for a great seminar.