 the technology to get into zero. Thank you. So can you share the screen? Well, yes, so I'm here to speak about technology, but I'm terrible with it. So let's see if this works. Okay. All right. So now, can you see? Yes, we can see, but how about that? Is that better? Very good. Yeah. Wow. That's a miracle. Okay. Great. Well, thank you so much. This is a wonderful conference. I wish I were there and sorry to have missed the other talks, but grateful to have this opportunity. This is a paper that I'm just working on. I've basically started it recently. So it's a work in progress, but I wanted to share what it's about and where I want to go with it and what results I have so far. Basically, it has to do with how do you get the entire world to transform the system that has given us high life expectancy, a material standard of living, and much else, all based on or very much of it on fossil fuels and definitely on activities that add greenhouse gases to the atmosphere and also affect radiative forcing in other ways. So how do you do that? And the main thing I want to talk about is that the focus up to this point, and this is true both of the climate negotiations and of the academic literature that this paper would fall into, has been on reducing the emissions of countries at the level of the entire economy. For example, the Kyoto Protocol sets emission limits for individual countries. The Paris Agreement asked countries to make pledges for what they would do to reduce their aggregate emissions. And the approach I'm going to take here is very different. It's to focus on the emissions of sectors, not whole economies, and to look at this not from the point of view of what individual countries are doing, but from the perspective of what the entire world is doing. So the orientation is very different from what we've had up to this point. On policy measures, instruments, the focus up to now has been on targets and timetables. So all the climate treaties focused on that. And economists have advocated the use of economic instruments like cap and trade and carbon taxes as a means to achieve these targets and timetables, even to the point where the Kyoto Protocol was basically designed to incorporate a kind of a grand cap and trade approach. My approach is very different. It's going to be a focus on technology standards, fuel standards, things like that. These are approaches actually that economists normally run for the hills, if anyone mentions this approach. But in this context, I think it can be very valuable. And also strategic research and development, joint projects. I'm not going to get into everything, but there's quite a large portfolio. And the main thing, which I hope to get across, is that I think if we take a very different approach that we've taken in the past, we'll be more successful. The focus in the past has been on reducing emissions. For example, the Kyoto Protocol only wanted to reduce emissions by a small number of countries, 5% over a period of five years. So the idea was we're going to start small and then build up. And what I'm going to be looking at is how would you actually get to zero, which, as you all know, we need to get to if we want to stabilize the climate. And the critical insight behind this paper and this kind of thinking is that countries don't simply emit greenhouse gases. They emit these gases as a byproduct of activities that combine an energy source with a technology to provide services. The main point to understand is that we don't use the fuel independent of the technology. Those two things are intimately related. So think of petrol or gasoline and the internal combustion engine or jet fuel and jets, jet engines, coaking coal and the predominant technology we use for making steel and so forth. So that's true across the board. It's the combination of the fuel source and the technology. And we've been focusing on reducing emissions. And what I'm suggesting is that we should be focusing more on the technology or on the fuel itself. Okay. Now, if you look at the approach we've taken so far, how successful has it been? One policy that's often noted as being very successful is the British Columbia carbon tax. Well, here's an analysis that was done, not published yet, but I found which estimated that that tax reduced emissions and manufacturing 4%. And this is clearly not going to get us to zero. The EU emissions trading system, which is the most ambitious policy worldwide in an analysis was estimated to have reduced emissions 3.8% over an eight-year period. An analysis of the Kyoto Protocol, which again, aimed to get a subset of countries to reduce emissions, 5% found that actually it was successful at that. It got these countries to reduce emissions. But because of international trade and the fact that as some countries reduce emissions, forcing up costs for their own industry, production and therefore emissions outside those countries increased. And their analysis concluded that at best, the Kyoto Protocol had no effect on worldwide emissions, which is the only thing that matters. So the question is, the approach we've been taking, is that the right approach? What really counts, of course, is the big picture. So these are, I think, figures that are very important. On the left, you see the classic keeling curve for CO2 in the atmosphere. And I just drew in a vertical green line to indicate when the climate negotiations began. And I like to invite people to look at this picture and then ask if you think the climate negotiations have made any difference. Well, you can't really tell because we don't have the counterfactual, we don't have a world in which there were no climate negotiations. But I think the first impression you have at least is that, yeah, it doesn't look like it's made much of a difference bearing in mind that we want to see that curve flat. So I think there's a very strong case to be made that whatever we've been doing is not working. Looking forward, okay, we now have a Paris Agreement. And in this picture you see in the black here, this is real data for emissions for countries from 2000 to, I don't know, 2014 or so. And the orange bands are projections of business as usual emissions. And these little yellow columns here, these are the voluntary pledges made by countries in Paris. So what you basically see is that those voluntary pledges are a little lower than business as usual, but still mean the global emissions are going to be higher through 2030. And that's clearly not a way to bring net emissions to zero. And then the framework convention secretariat calculated what you would need to do to actually have even a 50% chance of meeting the two degree goal. And you have to get on this path here from 2030. And my question, looking at this picture is, what is it we're going to do in 2030 that we haven't done previously? Because we've been trying to negotiate agreements that change behavior for 30 years. And I don't think we have a clear answer to that question. Okay, Glasgow, okay, we just keep following this. One thing, this picture here at the top shows you emissions worldwide from pre-industrial periods. It starts from 1750 until today. It dips down here because of COVID. But it's been going up and people are expecting it to rebound in post COVID. So basically, you have many decades, centuries of investment in an infrastructure to supply energy primarily through fossil fuels. And what we're talking about is completely changing that whole infrastructure, that entire investment worldwide, when a lot of the world wants more energy, and when all the technologies we even need to do this don't even exist yet, and certainly haven't been scaled up to any kind of level. So we know all the properties of these technologies. And again, Glasgow, you got the same thing. This is history. This is only from 1990 to around 2015 or so. Rejections are that if with the pledges that exist, you'll end up about here. I should say we have 30 years of countries making pledges and not meeting them. Going into Glasgow, the pledges increased a little bit. But basically, we're definitely not on course for meeting the goals that countries said we must meet. Now, one provocative and valuable idea for how we could do better is from Bill Nordhaus, Yale University. And his idea is let's couple adoption of a carbon tax by a subgroup of countries. He calls them a club with trade measures so that those countries will impose tariffs against others that don't adopt the same carbon tax. And what his analysis shows is that so we have here the carbon tax levels. So 12 and a half dollars up to $100 on the horizontal scale. You have the number of countries on the vertical scale and each of these bars represents the participation level in a club for a given carbon tax and for different tariff levels starting from zero. That's over here 1% up to 10%. Basically, what you see in the picture is that the tariff can be pretty effective at a very low carbon tax. But once the carbon tax gets high, let's look around $50, this whole thing starts to fall apart. And what you're seeing is in no case can you have the whole world working together. But actually, you may end up with just a fraction of the world. Now, in his model for his assumptions, that's not the worst possibility because he's assuming there will be no retaliation. So in other words, a club imposes tariffs on other countries, other countries do not retaliate. And that's actually not remotely to be expected. It will definitely be pretty clear there would be retaliation. In fact, the entire system of international trade is predicated on reciprocity. And it turns out that once you take into account the possibility for retaliation, the case here is even weaker. But my main point is that even in his best case, this is not going to solve the problem. He has a newer paper out on PNAS looking that if you couple tariffs with technological progress, you can do much better. And that completely to be expected if you worked out the theory behind this whole thing. But in his model, the technological progress is exogenous. And what we all know is that if you have enough technological progress, if you drop the price of alternatives to fossil fuels to a very, very, very low level, the markets will actually distribute new technologies all on their own. So I think that that's valuable to understand that. But it's not clear how you would actually map out a strategy for what to do from all of this. Here's some work I've done with Esther Dandenberg experiments. This came out this year, where we take a model like what he has of countries cooperating on trade, trying to cooperate on climate change, and then asking how would they link? And what would the consequences of that linkage be? And in particular, we're looking at two particular scenarios for institutions. One is that countries can actually make this choice unilaterally. So a country could decide on its own that it wants to try to link and others have to respond to it. That's the default in the international system. And the other is that you need a critical mass of countries multilaterally to choose to link. And then any other countries have to respond to that choice. So the unilateral is in a sense without a treaty. And the multilateral would be a treaty specifically trying to link. Well, the unilateral, well, the first thing is the theory shows that the best case scenario is that you would have what's called a coordination game, the trade measures. So I'm going to show you pictures like this throughout the talk. So let me explain what this is. On the horizontal axis, you have the number of the other players that join in some choice of actions. So in this case, it would be adopting a high carbon tax. I'll just call it that. So this is what the number of others that are imposing this high carbon tax. And these lines here show you the payoffs that any one country, I'm assuming that all countries are symmetric or identical here, that can be changed, but you can't just depict that in the picture if countries are asymmetric. So this is the payoff a country gets if it sets a zero carbon tax, let's say, or a low one, even better, low one. And that's this top line. And then the bottom line is what that same country would get if it imposed a higher carbon tax. And on the horizontal axis, you can interpret this as being the number of others that set a high carbon tax. So what you find is that no matter what others do, each is always better off imposing the lower carbon tax, because the payoff is always higher, no matter what this value is down here. But all countries collectively are much better off when they all impose the higher carbon tax than when they don't. So that in a nutshell is the cooperation problem that in a nutshell is what the climate negotiations have been confronting. And they've been because they're focusing on things like, sorry, targets and timetables, they're confronting that problem directly. And what we have to do, I think, is approach it differently. Well, if you couple the targets and timetables approach with trade measures, then what happens is on the right side here, so now you're coupling the two. If you adopt the higher carbon tax and you're linking this to trade, that means you kind of trade with the rest of the world, or you at least have to have a tariff with respect to the rest of the world, and you get a reduction in the gains from trade here because I'm assuming there will be, or can be, and it turns out it would be optimal for the response of others to actually retaliate. So that country on its own is going to be really hurt. But as more countries join in this effort to limit emissions and couple this with the trade measure, the incentives to want to join increase. And eventually what happens is those curves cross. And that's what's very significant because then when I can't actually see it in my, okay, then on the far right side, if everyone else in this particular example, if everyone else links climate and trade, it actually will pay the last country to want to do that too. So what basically happens is now, so previously you had a national equilibrium, like in this original game here, where given what others are doing, the best thing for any country to do is to set the low carbon tax. And, but here you have two national equilibrium, you have the original one, and you have another one over here where every country has the higher carbon tax, and every country is imposing the tariff against non the countries that are outside, but everyone's in. So in equilibrium, tariffs are never imposed, but it's the credible threat to impose them that sustains this outcome, this preferable outcome as an equilibrium. So it's really quite important to see that that's a best case scenario for the kind of approach that Nordhaus is proposing. And if you take this approach and put it into a laboratory setting, so Nordhaus, again, he assumes there's no retaliation, which means he doesn't get this best case. He's getting a range of other cases that are even more favorable, but I don't think institutionally feasible. But if you, he's also solving for his model using computer algorithm, but there are actually deep behavioral questions here about whether countries would both understand what the game is actually about, and choose wisely, I'll use that term choose wisely. So Astrid and I go into a lab and asked how people would actually behave. And they have to start off playing the original game, which is the game countries been playing for 30 years. That's this voluntary contribution game right here. And you see this classic behavior group behavior, which is countries start off or the players start off cooperating, and then cooperation basically collapses very quickly. This is classic. It's been seen in hundreds of studies. What happens here after that first effort has failed is you only need one country to propose linking. And then what happens is that everyone links and does the right thing. It's quite astonishing. So a complete flipping behavior. And these here are, if you remember, if they're payoffs or contributions, but the picture will look much the same. So a complete flipping behavior. This is more or less what we've been seeing in the climate negotiations. This is what we want to see. And what we get in this particular picture is that much more favorable outcome. But it's not guaranteed. So what happens between the blue and the orange pointer that you use this linkage between? Yeah. So sorry, excellent question. Sorry, I'm skirting over this. And so thank you for that. Yeah. So what's happening is every the group, so the group, these are groups of five, they have to they have to play this game here at first. Then having observed how they played, and I'm being aware of what this game looks like, not this particular picture, but they are aware of what the paths would be in that other game. They get to choose which game they want to play. And you only for this scenario here, you only need one player to decide to want to link, and then they're playing this game. And what you're seeing in this picture is when given the chance to choose three out of five in this case, because that's the voting is in this green, these green bands, three out of five choose to switch to B. That means everyone has to play B because they have to deal with it, the trade measures. And when they see that they all basically, essentially adopt, they all contribute to the public good. So they do what you want them to do here in this situation here. And it's all sustained. You get this as a Nash equilibrium, it's all sustained by the threat that if you were to deviate, the others would impose this tariff against you. So that's the, sorry. Oh, thanks. Okay. All right. So that's the ideal, but it's not to be relied upon. Here's another case you start off in the kind of classic way failing in that direct contributions game. But when you actually have four out of five players say they want to switch to the B game, they don't actually behave correctly in that game. And they're unable to coordinate and it collapses. And then they try again, can't do it, try again, they can't do it. And then they're able to vote a last time and they decide to go back to A. And then they get, again, the classic occur. Now, the reason the payoffs are lower here than you could ever get with this A game is because you now have retaliation. So you're not only failing to address climate, but you're also reducing the gains from trade. So it's like a double loss. So that's the downside to linkage. It actually may fail. And up to this point, the climate negotiations, the idea of linkage has come up before, but countries have avoided adopting it. I think inevitably it's inevitable that if we keep trying and failing to address climate change, there will be very strong pressures to link with trade. But my point is that we can't rely on that linkage necessarily to be all that helpful. If you approach this multilaterally, what we show is that countries do better. So here's the classic. It starts this way. But now you need at least three, because the total five and you need a majority to actually want to switch and you get this ideal situation. So this looks very much like we saw up here. So the best cases are very similar. The worst case is not as bad as here, but you go back. What happens here is you never get three out of five wanting to switch. I'm not sure exactly why it looks like you're getting it here, but they don't because it's all in blue. So anyway, they're always choosing here. You don't get enough countries wanting to switch. The majority you need for this game is actually four. That's why. Anyway, the main point is that with multilateral, you can only switch if a lot of countries agree to switch. And it may be that they don't see the advantage in switching. So there's a lot of variation behind that. What is the meaning of the number at the top of each panel? Oh, nothing. Those are all the different groups we played with. So they're just, yeah, this has been done with several hundred people. Okay. Groups and treatments. Thanks. Okay. Now, if you look at the previous attempts to link, they've actually all been sector based. So the U.S. tried to pass legislation, because we're trying again today, tried to pass legislation in 2009, leading up to the Copenhagen talks for a very ambitious effort to reduce emissions in the United States, 80%. I mean, it's pretty impressive. It was passed the House. It didn't pass. It never was voted on by the Senate. But the main thing to know about this is that it incorporated trade measures with the focus on sectors. I think this is really significant. Also, the EU tried to do this once before in its aviation directive. Again, it did it unilaterally. Other countries, the U.S., China, India, Russia, others, all threatened retaliation. And in the end, Europe backed off on that. There's a new agreement. Actually, the probably the most significant thing mentioned in Glasgow was that Europe and the United States agreed to enter talks into an agreement on the use of trade measures in the steel and aluminium sectors. I could talk about that, but I'll let that go. But the main point I wanted to make is that the world is looking at sectors as the analysts are looking at whole economies. Okay, so there's something going on here that I think we have to learn from. And if you look at which treaty has been the most successful so far, it's pretty clear. It's the Montreal Protocol, which was never intending to address climate change. But the CFCs that were controlled by the Montreal Protocol are a greenhouse gas. And an analysis by a Dutch scientist shows that Montreal achieved four times as much as Kyoto ever aspired to achieve and Kyoto failed as we saw before. And there's a new treaty Cagalli Amendment that was negotiated about nine months after Paris. And it has the same model of coupling action on reducing emissions for this one greenhouse gas HFCs to trade measures. And in previous analysis I did, you can explain the success of Montreal because that linkage creates this coordination game where you get these two Nash equilibria. So the basic insight is that when you ask countries to coordinate over something, they can do it pretty well. But when you ask them to cooperate voluntarily, they don't do very well. So I think we've been asking countries to do the wrong thing. Now, if you look at other successes, one would be the price of photovoltaic solar panels, wind turbines, battery storage, these have all fallen. And the nice thing about this is that it's the opposite of the leakage problem as the price of these technologies falls that take up by other countries increases, like worldwide, right? There's some just natural economic forces pushing for that change. Now, why did these prices fall? Well, they're actually were deliberate policies, but they weren't undertaken to actually cause this fall and this change in global behavior. They were taken more for domestic reasons. Japan had a program and the United States had a program in the 1970s because of the energy crisis. Australia invested in this in the 2000s. Germany adopted these feed-in tariffs to promote solar photovoltaics. And of course, China ramped up production and achieved enormous economies of scale. So these actions by countries that have not been taken in a coordinated way have been successful in a way that the other approaches we've taken have not been. So the usual approach has the potential of being efficient, the term economists use, but it may not be under increasing returns, which is going to be the focus of the rest of the talk and cannot be enforced. So that's like a really critical point. The strategic approach that I'm going to be focusing on is going to look at what's the best global outcome that could possibly be self-enforcing. That will be for a sector, not for the whole economy. And how do we get there? And in particular, how can we use treaties to get there? So we're going to be looking at treaties in a very different way than we have in the past. Okay, so very roughly, this is going to be a very simple model, but I'll just go through it quickly. First, we want to start off with the usual approach. So I'm talking about this using a technology called A, which is the installed technology. So for example, with cars, it's the internal combustion engine. With aluminium production, it's using a carbon nanotec, so on and so forth. So for every kind of sector, again, you're always coupling the fuel with the technology. And these would be the emissions for that sector, for this country I. And this is kind of the baseline, if there's no policy, and up T sub I as a carbon tax, that will reduce emissions. Okay, so this is the basic description I have of how under technology a policy to reduce emissions would have that effect. If you just normalize that emissions to one with zero carbon tax, you get this for and you solve the that equation for the tax, you get this expression here. And then if you assume that the emission reductions are achieved cost effectively, then this will equal marginal cost. And then you can integrate marginal costs to arrive at total costs. You get this expression here. And all that really means is that if nothing is done to limit emissions costs are zero. If you push things to the maximum, so you're increasing your reducing emissions as far as possible, the cost would converge to this value beta. Okay, now using so what are the payoffs, what do countries get by making those choices? Well, that depends on what they do individually, and what others do. This little gamma, this gamma is the social cost of carbon, which is the value of reduction in carbon going into the atmosphere to an individual country. This would be country eyes emissions. This would be, sorry, abatement rather. This would be the abatement by all other countries. And this would be the cost for two I of achieving that level of abatement. Okay, so this is the benefit to country eye of abatement by everyone, including itself minus the cost of its own abatement. And if countries make these choices independently of how much to abate or what carbon tax to set, you get an actually equilibrium in what the emission level will be. And what's key here is that it depends on the social cost of carbon for an individual country. And you can then plug that value back into the payoffs and you get this expression here. Okay. And one way to think about this, similar to what I described before, is imagine that you're comparing the national equilibrium choices that countries would make, which would be like a, think of it as a small carbon tax with what they would do if they cooperate fully, which would be the global social cost of carbon, which is the value that most economists in this area have tried to estimate. And what you find is that if countries set their own social cost of carbon, a carbon tax equal to the own social cost of carbon, and all do that, they get this kind of payoff here. As more others adopt the global social cost of carbon, which they've been trying to do, the payoff for countries that don't increase is, but the payoff to a country that switches from the Nash carbon tax to the global value would fall. They'd actually be worse off. And of course, the key thing is if all countries adopt to the global social cost of carbon for driving their policy, payoffs would be higher than in the Nash equilibrium. So the cooperation problem has been looked at, how do we get the world to go from here, which is the tendency of unilateralism to take the world up here? Okay. So that's using technology A. But now let's introduce technology B, which is a very different technology. And it's going to exhibit increase in returns in the same way that technology A has in the past. For example, you would get economies of scale and manufacturer. You're going to have economies of learning as as more output is produced over time. There may be network externalities, a variety of different aspects. There's a reason why technologies are so standardized worldwide. And this is what explains it. And the increase in returns explains why the existing incumbent technology is entrenched, like the internal combustion engine, and why it's harder for a new technology to penetrate the market, like a battery powered electric motors for cars. Okay. So this would be the payoff to a country of using technology B. And that will depend on the reward to get the payoff, I guess, from the countries that don't use. So now K is the number of others that choose B. So this is the number of others that stick with the old technology. This is their abatement. So country I gets this benefit per ton, as it were, of emissions for all those countries. Plus, it reduces, this B reduces emissions to zero by assumption here. And so if no other country acts to adopt B, the country would get lambda times one, because it reduces emissions to zero. But it also is going to benefit from the others that adopt B. And this is the cost to country I of adopting B, rather than A. And what you see is that this, by assumption, as more others adopt B, the cost to country I is going to fall. Okay. So that's where increasing returns are coming in. As more countries adopt the new technology, let's say switch to electric vehicles, the cost to each country of switching is going to fall. You can also add a term here for the incumbent technology to reflect this possibility that you start off where everyone's using the internal combustion engine. But as more and more countries switch to the electric motor, what could happen, powered by battery, what could happen is that the economics of the incumbent technology could start to fall. That would be particularly the case if there are network externalities. Now, this capital C in the payoff for technology B, that's the cost to, in a sense, the first country to switch or the one that's switching all by itself. And for automobiles, you can look at Norway. So Norway is a pretty ambitious policy to move to all electric. And estimates by economists put the equivalent marginal cost in terms of tons per CO2 at over 1,000 euro. And so that's very high. That's higher than the global social cost of carbon, almost certainly, and very costly. But what's underpinning this model is the idea that if a lot of countries started to do that, that value would come down. Scott, can I ask you a question? Yeah, please. This is very, very nice. This is a good aspect of knowledge that you're picking up. But suppose you had it as a two-period problem. In other words, is there likely to be a problem of a first entry? Because you'll be shouldering most of the costs of the chance of free riding over time. You see what I mean? Yes, I do. I haven't modeled that yet. Again, I've just kind of started this. So I'm having, more or less, what I'm looking at is one shot with a stage game. So I think more or less is an accurate description. Yeah, this kind of sequential choices. I haven't looked at that yet, but definitely the early movers, well, based on what I have right here, would pay a penalty. Now, what I'm going to be looking at though, Partha, is that you're going to get not one but a critical mass of intense early movers acting. But they're going to do it with a strategic purpose, which is to cause all the others to move after them. So that's where I'm getting this. What you were pointing out is the possibility that you can get these cascades. I will move to that, but I haven't gotten there yet. I think the cascading is going to be interesting when you look at when I started introducing asymmetries among countries. Because there may be some that really do have a strong incentive to act and that their action, even if it weren't strategic, may cause others to act subsequently. And then you want to look at what are the parameters that would stimulate that kind of cascading. Okay. So again, this is the one we started off looking at before, the Prism's Dilemma. And this is what it looks like. This is what it could look like when you're looking at switching from A to B. So this game here is the switching game. So this payoff here is the payoff a country gets by adopting A, the original technology. And on the horizontal axis, you have the number of others that are actually switching to B. And I've drawn two different curves for what the payoff to adopting B might look like. If you take this bottom one, so everyone's benefiting, whether you choose A or B, as more choose B, that these payoffs are all positive slope. But what happens is the payoff, there's a big penalty to switching, whoops, there's a big penalty to switching to B. And it may narrow a little bit, but it never quite closes. So you still end up with a Prism's Dilemma. This is the best outcome everyone switches, but it can't be sustained as the Nash equilibrium. And nor could the one for sticking with technology A, the best case there, but switching to B now is better than A. And yet you still won't necessarily do it. However, if the payoff curve happens to look like this, what's critical is that it crosses the payoff to sticking with A. And now you get this flip in the incentives. And once you get past this critical value of K hat, if K hat others switch to B, everyone else wants to switch to B. And then you end up with a situation that's a Nash equilibrium, it's also quite a stable situation. And that is by far in this picture, the highest payoff attainable. So if this is the picture we're looking at, and we're adopting incremental approaches, we're not going to get there. But if you took an approach that focused on switching to B, it's possible we'll get there, but not guaranteed, because the threshold in this picture is pretty high. And what that basically is going to mean is that the payoff dominant Nash equilibrium is quite high, but the risk dominant one will not be, will be over here. So it's not guaranteed that coordination on this better one would actually succeed. So how do you solve that? Well, this is what a treaty can do. So here's a very simple model. So this is that stage game I just mentioned. And so this is in three stages. In the first stage, each country is going to decide independently, this is sovereignty, whether it wants to join the treaty, yes or no. In the second stage, countries are going to decide collectively, whether they ought to switch. So in a sense, in the second stage, that's what the treaty, that's how it's being negotiated. And finally, the non parties get to decide again, this is sovereignty, whether they're going to switch. And you solve these games backwards. So what we can look at for the non parties in stage three, they will, they will want to switch, but only if there are enough others who are already switching, right? Oops. So what we know is that they will, they will want to switch if the number of others is at least k hat. Well, the parties in stage two can anticipate that this is the case. So what they're going to do is decide to switch only if participation in the treaty is at least k hat. So what that means is if fewer countries join, you're under no, even if you're in the treaty doesn't enter into force, and you're under no obligation to switch. But when you join to get past that entry into force of k hat, the reason that's you have a strong incentive to do that is you're going to cause all the other countries in the treaty now to switch. So you're going to get the benefit from their action. And in addition, you can foresee that the collective action by all these countries will cause the remaining countries to want to join. And so what I'm basically saying is a treaty can be designed in such a way as to coordinate behavior around the better Nash equilibrium. So you're solving the enforcement problem. And all the treaty needs to do is coordinate participation. The key thing here is to you have to think about this strategically, instead of asking countries just to limit their emissions, which is what we've been doing. That's this picture here, you've got to ask them to do something else. And that's something else has to have this feature that basically is a coordination game. So that is the basic the essence of the strategic thinking behind this approach. Okay, now you could have if you had something like network externalities in the technology, that payoff curve might actually fall. So in other words, as more countries adopt the new technology, the payoff to sticking with the old one might fall, even though you're benefiting from the additional emission reductions caused by the other countries. And, you know, that would be good in terms of lowering the threshold for entry into force. But now I've drawn a different payoff curve than we saw before. So what happens here is you again have a coordination game here. But the the full cooperative outcome in this B game, the switching game actually gives you a payoff that's lower than the potential in the A game. Okay, so it's better than this one, better than that one, which is the only alternative basically, that you're ever going to see an equilibrium. But if you're looking at this in an idealistic way, and you're playing the A game, you're thinking, I want to end up here. So why would I bother going over here? So again, it requires strategic thinking to actually see that by going for something that in a way is second best, you could be much better off than if you went for the ideal. And Astrid and I doing experiments have found that people really suffer from this, they have a difficulty that tendency is to want to go for the ideal. And I think that that explains why in the climate negotiations, countries try again, and again, and again for doing playing this game and aiming for an ideal when I think they would be better off playing the B game, even if it's less efficient in some, you know, abstract sense, the full cooperative outcome would yield a lower payoff for everyone in this game. But this one's attainable, it's an actually equilibrium, and this one is not. And that again introduces this behavioral difficulty. Some examples, by the way, can I have a five minute warning because I'm sure I'm going over, I always do, but. Yes. Yes. All right, so just let me know. Thank you. So just some examples to try to make this more concrete. Okay, in aviation. First off, you have an international organization called the International Civil Aviation Organization, or IKF, and countries meet there to decide on things like standards. There is a fuel standard. All airport international airports in the entire world have the same fuel so that planes can refuel everywhere. And what this approach I'm proposing would do would be to say, let's replace that fuel we use now, which releases greenhouse gases with an alternative. Probably the most reasonable one for aviation is a synthetic fuel. And you would make that by producing hydrogen using renewable energy, that's the green hydrogen. And you're going to couple that with CO2, you're moving from the air to produce a synthetic fuel that actually can run on the same engine as we use now. And you'd have to require that every airport adopt this new fuel standard. Okay. And then what you're doing is you're recycling the CO2. So emissions are essentially zero. And the reason you could enforce that is because of network externalities. So it would have to be the case that countries would say we're only going to fly to other countries that meet this standard. And then you'd be changing the entire, yeah, the standard worldwide to this new one. With shipping, it's not clear exactly what the best approach is going to be. It's a little less clear, I think, than aviation probably. But one that looks attractive is to use ammonia as a fuel. And you would make this with hydrogen as above with aviation. But couple it down with nitrogen that you would remove from the air. In this case, you'd have to change the whole infrastructure. You need new engines, new ships, new ports. So it would be a massive investment. But again, because of the network externalities, I think it would be very powerful that as more countries did this, more others would want to do this, partly because no one's going to bring their tankers into their ports if they can't load ammonia for the return journeys. Okay. Aluminium or aluminum, you could switch from a carbon anode to an inert anode, iron and steel. You could move from the basic system that's used today of a blast furnace and oxygen furnace to direct reduction, again, using hydrogen and the electric arc furnace. This would reduce emissions dramatically. And if the energy were produced using renewable sources or nuclear, that would bring emissions to zero. And then in automobiles, we seem to be going in this direction of a switch to electric cars. And it's interesting that the world switched from just the internal combustion engine coupled with leaded petrol to the internal combustion engine fitted with a catalytic converter and unleaded petrol. And it's really interesting that occurred worldwide. So that's a universal standard now. And actually, there was no particular treaty requiring this. And it still happened. And the reason it happened is because the increasing returns are so powerful in that sector. Okay. So how much time do I have left? So I think we are up over. But essentially, we are in the discussion time. So maybe you can say take a couple more minutes and get to the conclusion. So that's the discussion. Okay. All right, lovely. Thank you so much. I always go into this being very, you know, I think I can do those first slides really quickly. And then as soon as I see them, I realize no one's seen these before I have to explain them. So apologies for my timing. Now, one of the things is when we have the installed technology, and we know how that works, but we know something about the new technology B, but we don't know everything about it. And we don't know exactly how much it's going to cost if it were adopted by everyone. We don't know that exactly. So then the question is, okay, well, suppose these switching costs are uncertain, what happens? And obviously here you can make errors. So one way to think about this is those switching costs, if they were high, you'd end up with this kind of curve. And you would no longer, you would have a prison dilemma still. If they were low, you'd have this kind of, whoops, this kind of curve. And if you don't know the question is how would that change the game? And the main insight from this is that when now what a critical mass of countries decides to switch to change the behavior of the other countries, even though this is a one shot model, you can think of this as those countries, when they switch, they implement. And when they do that, they reveal information about what the costs are. And now the countries, the non parties to the treaty are responding both to the critical mass we saw before and this new information. And that new information could show that the initial beliefs were optimistic or pessimistic. So what this does is the possibility of errors. And in this literature on technology and increasing returns, there's a lot of discussion about how rational choices anywhere along a path could actually leave the entire system to end up in the wrong spot. And you get the same kind of result here, even though I don't have a dynamic formulation. You also can introduce, once you've introduced, so the model up to this point assumes that choices are irreversible. But if you relax that, and you allow countries to make to change their decisions, and you also introduce the possibility of spunk costs. So if you switch from A to B, and then you decide later, oops, that wasn't a really good move, I want to go back to a we may have lost a portion of your upfront investment in that switch to B. And one interesting thing that happens when you take this into account is that countries may find that they made an error. But if some costs are big enough, they're not going to want to revert back to A. And even though they wish they hadn't chosen B in the first place individually, collectively, they're actually better off given that they're stuck there and they can't switch back. This wouldn't affect their behavior early on. But it is a feature of this approach to treating negotiations. I also want to mention R&D. That's been another success. This is the fall in the price of solar voltaics. And there's strong evidence that R&D has played a role in bringing this cost down by one study estimated about 60% of the reduction the price of photovoltaics is due to R&D. And that raises the question about whether countries might want to adopt R&D for the purpose, not only of lowering their own cost, but of lowering everyone's cost so as to change their behavior. And here's pictures of what this might look like. So if you lower, if you adopt the R&D in some preliminary stage, then what, and that lowers costs, then what it does is it pushes up payoff to adopting B. It could, though, that don't transform the game, you're still going to have a prison's dilemma. So that's not going to help you. It could be that you would, but with a very high threshold, and maybe that R&D will give you a lower threshold, which is going to make it more likely that the switching will take place. You'll get that critical, get past that critical mass. But the most interesting situation is where without the R&D, you've got a prison's dilemma with the R&D, you create a coordination game. Because now that R&D has fundamentally changed the incentives for countries to act. Okay. All right. So and it's easy to show that countries would have an incentive to cooperate on R&D. And I would mention on the US something that you may not have heard about because I just happened to discover this almost by accident. But the Biden administration has policies for investing in critical technologies. So one is on hydrogen. One is on removing CO2 from the atmosphere. And the third is on long-term energy storage. And these actually are three really critical areas for transforming any system towards zero emissions. But the wonderful feature of this is that if this is successful, this is going to be helping the rest of the world also want to make this transition. And I find it striking that there has not been international cooperation on this because it's not obvious that any country independently has enough of an incentive to invest in this R&D. So to sum up, what I want to do in this paper is look at the choices of technologies for individual sectors, looking at how we can get them to be adopted globally. So the focus is on a global, in a sense, transformation and adoption of a technology. And to show that that could be more successful, for the reasons I showed here, than the alternative, which is the approach we've taken up to this point. And the essential need is to transform the game. We've been facing a prison dilemma. We've actually chosen to look at the problem that way. I think it's an unhelpful approach. And we need to try a different approach that transforms the game into a coordination game. Thank you. Thanks for your patience. The past rusher from Michigan who was picked number 45 for the ball. Okay, so questions. This is Joel. May I ask a question, please? Let's go ahead. Thank you very much, Scott. I admire your approach and support your effort to develop a coordination game. I do not claim to understand all the details of the economic models or the equations. I couldn't follow it at that speed. So that's my weakness. In Plato's Republic, Plato models the functioning of a state as the functioning of an individual writ large. And he models the executive function by the head, and another function by the arms, and another function by the feet. And he imposes a model on a state of an individual. I think a lot of political scientists have abandoned that model as not reflecting the internal complexities of real states. And I'm wondering about the extrapolation from an experiment with five people to a system of international states. I'm wondering whether the model of Plato is perhaps affecting the willingness to use results from individual players as a model of results from plays among nations. Would you help me understand if that's a problem or if it's not a problem, please? Okay, Joel, thanks so much. And very kind of you just say that you couldn't follow everything because of your own limitations, because I think I had a skirt over some things, and I apologize for that. But your question is very spot on. This is the issue in experiments called external validity, which means you're trying to study a system and you're basically abstracting it and you're performing experiments using people who are not the right people, they're not the decision makers on the climate game, for example. The context is not quite right because it's not an administration or government negotiating a treaty. It's an individual, it's very controlled. So the whole thing is we do theory, it's very abstract, and what we're looking at here would be an experiment that's also rather abstract and also you're not using the same people and you're not, the situation is quite different. So that's true. There are two responses to that. One is to say, well, because you can't do it as it is real, you shouldn't do it at all. And I don't think that's the right way to think about it because we have no alternative. We can't run repeated exercises with the entire world believing that the stakes are real. When the stakes are real, you only get to see it once, like you have one Paris meeting, that's all. So the question is, how do you interpret what's going on in a place like Paris? And what I'm looking at here, Joel, is how do you actually understand something that's never been done? How do you approach that? Countries have not tried to approach the problem by choosing a technology. So like in a lot of our research, I think the way to think about this is not that I've discovered truth, but rather that I am revealing insights. And I think these insights help explain why we do see what we see. And very critically, how if we try to do things differently, we would do better. I have to say a couple of this with a lot of other work I do and exploring history and so on. So you want to use all the tools available, but it's definitely the case, Joel, that anyone looking at this issue is having to confront the problem that we can only see what we can see. We can't do experiments on human behavior at this kind of scale. And so what we're looking at would be insights. And the last thing I would say is, you know, one thing we do know from experiments is that when people make choices in groups, in other words, what I have here are that each country as it were is represented by one individual. And that government, if you think of it that way, would be a monolith, which is not how the world works, obviously. And as you said, political science doesn't focus on that. So that's very true. But if you had a collection of people within, let's say, one decision is going to be made by, I'll call it player one, but player one consists of an entire group. Well, the literature shows that actually groups choosing in that way are more consistent with the theory than individuals. In other words, they're closer to expectations about pursuit of self interest than you actually see with individuals. So there's some work we can take around this. And in some areas, people have been able to do experiments with professionals to see if they behave differently from students. I have played games for decades with real negotiators and a lot of other people who are very real world engaged. And I find very consistent kind of behavior. But I can never say that this would explain exactly what's happening in a place like Paris or in Glasgow or the next city we move on to. So an excellent point. It's just a limitation of the approach and there's no, I don't think we can do better, but we can't do as well as we'd like to. Thank you. That's a very helpful response. Thank you. Scott, so do you need to get government in the loop in this agreement? So is that a way in which because they are sector wide agreement, so couldn't they, couldn't one think of designing situations where essentially this could be just a treaty between private entities? I'm sorry, treaty between private companies. I mean, well, treaties are among governments. Yeah, but what you can get, of course you can get country companies that trade associations and so on that coordinate. That's a little careful because some of this could violate antitrust. But basically, let me put your question a little differently. Maybe as you look, well, if the costs of the alternatives were sufficiently low, then you would get a transformation worldwide and it would all be achieved by the market. But we have a massive public good problem here, as Partha mentioned. And the reason we have governments is because they provide public goods. That's the essence of it. That's what Thomas Hobbs was saying. Scott, you still have the coordination problem, right, of being the market human having to go to the good equilibrium, so the market would not do it. Oh, well, that's okay. With the increase, I was thinking if you just lowered costs sufficiently so that each had a private incentive to move, even if no one else did, that would be a big reduction in that capital C I had. It would be possible. That would be a massive reduction. But if it were short of that, yes, there would have to be a coordination even in the private sector. That's right. Yeah, but this is an area where actually when you focus on sectors, you actually are talking about not about governments making these choices just on their own like you see in places like Paris, but actually you need to have a dialogue with the sector. So I've actually been looking into individual sectors. It's actually complicated when you look at how do you reduce emissions in iron and steel? How do you do it in aluminum? How do you do it in aviation or glass or cement or all sorts of different sectors? And so I think the way that the negotiations would be structured would be to combine the expertise in industries with governments because the industries want governments to create things like a level playing field and to address these issues like critical mass. Of course, they would be lobbying. So that's going to introduce another consideration and extra costs and so on. But the Montreal Protocol, it was a very successful treaty. One thing that made it successful was that and the industry representatives deciding what could be done and from that agreeing that certain goals were not only desirable but achievable. And we've not had discussions like that on climate. Yeah, I wanted to follow up on that question because it's very similar to what I had which is governments change probably more rapidly than CEOs of companies or the long-term interest of companies and governments therefore have I think much higher discount rates that is government officials do. So I was interested in not just the sector-based approach but the linkages between sectors because you and I have been involved with some discussions and for example the insurance and reinsurance companies are really concerned about climate change. So what if you thought not just about sectors but the linkages between sectors, the sort of business ecosystem and whether there aren't points of leverage there that ultimately can bring pressure on governments. Well there are a number of things in that comment Simon. I think about you know whether CEOs and governments, I think that's a hypothesis, interesting to test. Yeah, but certainly you know the insurance, the reinsurance industry for example has a persistent long-term concern, 10-20 year time zone on what climate change is going to mean for their risks. Yeah, I have to say I'm not terribly worried about them. I mean what they can do is just obviously raise premiums or decide not even. I didn't mean that. My experience with talking with people in risk management that have been in the insurance industry is that they want to bring pressure on companies to take action. Right, well they could do that through their policies. But anyway, leaving that aside I think there were a couple of issues there. One is the involvement of industries with the policy making. The other is the linkages among the different sectors. The one I was thinking of doing in this paper is the second one which is part of what you were saying. So for example one of those slides I had the word hydrogen appeared several times and when you look at hydrogen actually realize that it plays a critical role overall and even within individual sectors. But if you approach that problem just sector by sector I think you'll miss the importance of investing in a hydrogen infrastructure for the entire system. So my point is which I think is connected with what you were saying. If you look at everything piecemeal you also will be missing things. Right, if you look at the whole economy we're missing things. If you look at a piecemeal you may and the interconnections among different parts of that system I think are quite important. So that's one of the things that's on my list of to add to this paper. I hadn't I haven't at least so far in this paper modeled industry or private sector acting independently or at least to some extent independently of public sector. I could do that but the paper is already going to be relatively expansive. Thanks very much Stott for all this. You know it's also it's going to be really interesting you know if you were to do this how do you map out a kind of a grander strategy about you can connect all the individual pieces. I've never gotten there yet. Some of this is hard to know because you don't know how early moves are going to actually change technologies and so on and we know you know we can pretty much explain things looking backwards but right I mean history you know worked by people like you know historians like David Landis and others but when you're looking at the future you know how do you actually do that? How do you actually think about that? How do you model that? You know that that's that that's this kind of work is pointing in that direction. How do you actually think about all these interconnections and where you not only have the kind of things that we're seeing here but you've got the climate surprising us and you've got political institutions changing and sometimes violently and so you know modeling that entire system of interactions among states but also involving technologies engineering which is advancing like what determines the direction of its advance and the earth system itself and all of these are interacting and that's the thing we really care about but it's it's a very hard thing to grab a hold of even if you're to look at it historically but definitely when you're looking at it from today's perspective and you want to know what's it going to be like you know decades or a century from now or beyond okay so Scott thank you very much and thank you actually we reached our first time I would postpone discussion to tomorrow so tomorrow we started 9.45 with Pablo Marquez lecture and then we have a break and then discussion and then around so thank you all and see you tomorrow