 I am pleased to introduce our next speaker, Anais Acevedo, who will be speaking on the topic of behavior and environmental justice. Anais is a professor in the Department of Energy Resources Engineering at Stanford, and the title of her talk is Decarbonized or Green Energy Systems Does Not Equal Equitable Energy Systems. Over to you Anais. Thank you so much for the introduction, Sarah. There is an enormous importance on decarbonizing the energy system, but that doesn't necessarily equate with equitable energy systems. So while decarbonization in itself is a daunting and challenging task that we need to start yesterday, we need to take into account also other sustainability aspects tied to environmental justice. And this relate to how the transitions and solutions that we propose for decarbonization may help also improve or deteriorate issues like air quality and air pollution health consequences. The distribution of the economic burden associated with the deployment of green technologies, the access to the adoption of green technologies and jobs, and this just to name a few. So I'll start by highlighting some of the recent work tied to access to technology and how policies influence the outcomes of technology adoption. So Craig mentioned some of these perhaps smaller, more modular source of technologies for decarbonization. Solar is an early example for that. And what this picture is showing this plot is a comparison between the average, the median household income by county and the distribution of early and more recent adoption of solar PV. So in blue, we have the distribution of income by county in green and to the right-most side, the adoption of solar PV systems installed in 2006 and in reds and kind of the middle of this plot, the adoption of solar PV systems in 2014. And this plot makes the point that early adopters and surprisingly were mostly high income households benefiting from the system that was in place for solar subsidies. So the same amount of subsidies by household may not be the most equitable solution for the adoption of those technologies. And we can envision that other types of technologies would suffer from the same sort of issues thinking about home storage devices, vehicle electrification and so on and so forth. Now the other aspect that I would like to raise related to distributional effects is that while we do need energy currently in the system as we have it, we have the implications for climate change, but yet the other implication from the systems as they operate currently relying on fossil fuels, we have very serious health damages from air pollution. And so this constitutes a problem in terms of who suffers from those damages, when and where. So as we think about the usage of fossil fuels and the consequences of removing potentially GHGs and not air pollutants or removing both, we'll look at very different consequences. So notably GHGs will have a global dispersion and will stay in the atmosphere for a very long time, whereas a co-emission that potentially will occur depending on technology and choices would be the emissions of things like SO2 NOx and PM which will have health damaging consequences built in the form of primary PM 2.5 emitted as well as the secondary formation of PM 2.5 from the emission of SO2 NOx and ammonia. Now the issue is that those effects are going to be much more localized. There is still going to be a distribution that will need to be modeled. The effects, for example, from the stack of a co-par plant will have a large and long dispersion but are not going to remain for a long period of time in the atmosphere. And so how do we contrast those two issues namely under the lenses of technological change and adoption of new green technologies? Not only that, but as we think about different types of services that will be produced, whether it's electricity or mobility, the types of emissions currently associated with that will be fairly different and differently distributed in space and time too, even just across the criteria pollutants. So the consequences will depend on whether the emissions are from a very high stack of a co-par plant that you're using to charge a vehicle in India or an internal combustion gasoline powered vehicle that is running very densely populated areas in New Delhi. So I'll provide an example. This is to our knowledge, the first paper that looked at the issue in this detail where colleagues Manny D'Retain, Chris Tesson and Julian Marshall, we developed a quantification of the distributional effects from air pollution from the US electricity sector. And so as we think about the adoption of new technologies, we can contrast to what is being removed or displaced from the grid as it operates today. And why the environmental justice aspect? Well, we may think that yes, the grid is getting cleaner over time, but electricity generation even in the United States is still a significant contributor to air pollution namely associated with coal plants. Even if that has declined thanks to environmental regulations and the transition from coal to natural gas and to some extent to renewables. So when we look at the grid in the United States today, it still comes from somewhere between 10,000 and 52,000 per meter that's per year. There's uncertainty associated with the air quality model that is being used. And the demographic distribution resulting from the exposure to this pollution is largely unknown. So what we've done is to estimate the exposure and held impacts from both primary and secondary PM associated with electricity operations across the United States in each regional transmission organization or RTO for each state as well as by income and race. And so I'll jump to the results from this modeling. This plot shows on the vertical axis, the premature mortality per 100,000 people associated with the operation of electricity generating units in the United States. And there was a horizontal axis shows the self-reported race and ethnicity. And so across the United States will have people suffering an average of 5.3 per meter deaths per 100,000 people. And we see that black and the African-American people suffer more higher values than the overall population average followed by white non-natino with the other self-reported race and ethnicity suffering lower values of premature mortality. Now, importantly, this effect doesn't disappear by income. Yet again, in the vertical axis we have premature mortality per 100,000 people and the horizontal axis shows household income groups. And so we see that high income groups suffer less or are located in regions where they suffer less in terms of premature mortality from exposure to air pollution. And we see that across all income groups black African-Americans suffer from higher exposure than the average and then white non-latino and that this effect doesn't dissipate with income it continues to persist. Now, that's one dimension of distributional effects yet another one is how the new air pollution's consequence relate to where pollution is emitted and who suffers the burden of those emissions. So this first plot shows a premature mortality in the form of total annual deaths that occur within a state regardless of where the emissions occur. So the emissions may occur in a very distant state and this reports the values that the state suffers. So that's within state. And we see some states like Pennsylvania, Texas and Ohio suffering very large figures in terms of premature mortality. Now we decompose the symptoms of damages that are imposed by the emissions that occur between the state. So kind of self-induced damages on this plot B. And we see that, for example, Pennsylvania and Ohio have much lower values in this case meaning that the emissions that occur in those states contributes still for fairly high values much less than what is the observed total premature death toll in the state. Now, the other distributional effect consequences is looking at the damages to others. And so in this plot we show kind of the export of premature mortality caused by a state. And so we do see that places like Pennsylvania and Ohio have are contributing to a large number of premature deaths that are not occurring between the state boundaries but are occurring elsewhere. And finally, we have a net sort of effect and highlighting here that the state like New York where the premature deaths that are imposed by self or to others are extremely small see an enormous number of imported deaths. Basically, it is importing the pollution from other states and suffering its consequences on an issue that they have little control over. So key findings from this portion of the work is that there are very important distributional consequences associated with the greed as it is today that should inform the greed of the future as we think about heavy decarbonization of both the greed and other sectors. We find that the average exposures are the highest for black and African-American followed by non-Latino white. And the exposures for remaining groups are somehow lower. The disparities are observed for each income category and this indicates that the racial and ethnic differences in exposure hold even after and counting for differences in income. We also see key differences across states on whom instance who suffers the consequences from air pollution. And finally, for 36 states, most of the health impacts are attributable to emissions to other states, highlighting the need for consolidated action and coordination across states as we think about both air pollution and carbon policy, which leads me to the second piece that I would like to highlight. This piece in ESNT was co-outored with one of my former students, Brian Surgey, as well as several colleagues with expertise in air quality and in environmental economics. And so the issue here was Ken will look at the decarbonization of the US power sector under both the climate and health benefits lenses and also what does that mean for distributional effects related to health? So the challenge and motivation here is that climate policies often quantify as a site co-benefits improvements in air quality, but we haven't seen really detailed treatments of those two issues together in terms of understanding what would be the optimal strategies across reducing damages from air pollution and climate change and whether that changes a lot from looking at climate change alone. So we did that. We basically developed a capacity retirement and expansion expansion model for the US grid and looked at what would be the optimal investments in new generating capacity and there are different types of objectives and constraints. And the objectives were to either just meet climate change mitigation goal of reducing emissions by 30% or explicitly optimizing for minimization of the damages from both air pollution and climate change while at the same time still meeting that constraint of hitting a 30% emissions reductions. And we do several simplifying assumptions. We assume that the new capacity will need to be built in the same county where a plant is being retired. So we ignore issues related to transmission constraints and we'll look at just a few options and those options are natural gas as well as wind and solar. Where for wind and solar we assume that they do need to be paired up with storage to provide exactly the same service as the coal power plant that is being retired. So I'll jump directly on the results and I'll be happy to talk much more in detail about the modeling offline. The vertical axis provides the annual damages in billion dollars and we'll show those from climate change related damages and air pollution damages. The baseline is a system as it operates today. And so we see the climate change damages which we value at the social cost of carbon of $40 per ton of CO2 being around $70 billion. And the air quality related damages from premature mortality depend on the air quality model that you assume and the concentration response function. So here we are like already in these results the sensitivity associated with that given its importance. So if one uses the Harvard city study the health damages associated with air pollution are actually even higher than the climate change damages. If we assume the ACS study they are smaller but kind of in the same sort of broad order of magnitude as the climate change damages too. And the different symbols correspond to different reduced form air quality models that can be used to relate emissions to concentration and to damages of air pollution. Now this is the baseline alone. And now when we think about just climate only policy what are the results from such? So when we impose climate change reductions of 30% emissions of course the emissions do reduce by 30% and we see declining the associated damages. And we have a co-benefit in terms of the air pollution reduction that goes with it given that the model selects to remove a lot of the existing coal power plants and replace them with either natural gas or renewables plus storage. Now importantly if we explicitly include health damages and climate damages and try to minimize those together in the objective function we see a further reduction in health damages and their consequences. So the question that follows is is that more costly and how much more costly can that be? So in this next plot we show both the annual benefits as well as the mitigation costs from pursuing a climate only strategy versus explicitly incorporating health damages as well as the climate goals. And we see that the costs when doing so in terms of deployment of technologies increase by a little bit but we still see net benefits all across by incorporating those two things together. So we just suggest that policy makers could look at those policies together and even improve on the net benefits associated from those emissions reductions. Now this talk was motivated by equity and environmental justice. So one of the things that we pursued in this scenario is to understand how the health damages are reduced when we impose just a climate policy or a climate plus health damages policy by income quantile. And so the good news is that we see benefits across all income levels and we see those benefits being even larger if we were to explicitly incorporate health plus climate policies. And we see that the benefits are the largest for lower income quantiles. Now the same effect is not seen by minority quantiles. So this is the same plot as previously but in this case in a distribution from the minorities that exist in the county being assessed. And so we see that regions that have lower amounts of minorities represented are the ones that harvest the largest portion of the benefits. This changes of course are also gonna be very different from state to state. If we can consider just a climate policy versus a climate plus damages from air pollution policy. So in this plot, the orange bars correspond to climate only policies and the blue bars correspond to policies related to health plus climate. The bars correspond to new capacity that would need to be installed in a state to achieve the goals of the policy. Whereas on the right hand side in this axis and with the black diamonds we show the amount of capacity, the share capacity that would need to be replaced in a state. And so we see major differences across some of the states when we go from one scenario to another. So I'll end here and open it up for questions and hopefully this will spur a discussion. Thank you. Thank you so much Inez. Okay, so there's clearly an economic impact of poor air quality on things like health costs and worker productivity, things of that nature. Is this on the order of magnitude that policymakers take notice? It is indeed on the order of magnitude that policymakers should take notice. So the evidence of that would be the figure that I just showed where we computed just the annual damages for climate change and the annual damages in terms of premature mortality from health. And they're kind of in the same ballpark. And depending on what you assume for the social cost of carbon we could have a three hour long discussion on what that should be and implications of such. And the uncertainty associated with these air quality models but they are definitely in orders of magnitude that should be considered and should be considered together and that policymakers should indeed take notice. If anything in most instances this will provide also the outcome of larger net benefits to society from more strategic levels of carbon emissions adoptions at least in the very large instances where the criteria pollutants also are reduced. There are some questions where tensions arise and so I didn't show that but if you are thinking about electrifying in regions like in India currently where you'd be charging your vehicles with coal power grid where scrubbers and other air pollution control technologies are not installed, the effect could be actually increasing the damages rather than decreasing the damages altogether. And we have some work showing that I'll be happy to share with interested participants. Okay, do you see a role for changes in local air quality regulations at the state or federal level like the changes to the Clean Air Act or anything like that? Yes, that's of course tremendously hard and the Clean Air Act is already a phenomenal achievement that we have in place but under the lenses of state level policy I would hope that those discussions indeed emerge and materialize in actual policy that takes the two aspects into consideration together. My sense is that we'll be missing opportunities and actually producing some serious unintended consequences if we don't. And do you see any policy coming or emissions crossing state borders could force changes in other states? Indeed, that's where both the cooperation negotiations may be extremely challenging but given the magnitude of the impacts that we found for some of these states it's actually quite surprising to me that they are not pushing harder already right now. So the effects that we've seen in New York, for example extremely damaging for something that it's really not to their benefit. There is another layer that I didn't touch upon which is who is consuming and producing the electricity. So we just talked about imports and exports of consequences from air pollution but we're having greed that is really so interconnected that some of that electricity consumption may be related to those states that don't have units that are highly polluting but are still consuming the greed. So there is this really neat work by Jack the Schalender and Sally Benson that track the consumption based emissions of CO2 across greeds and pushing that further and connecting that with the air pollution consequences may be an important next step. Okay, thank you so much, Nate and I really appreciate that.