TRANSCRIPCIÓN 🔗
Ep. 53, Sarah Mills: Cities can't function without critical infrastructure. Wastewater treatment plants, power plants, and major roadways. They keep our water clean, our lights on, and our economies moving. But they also come with trade-offs. Historically, much of this infrastructure is left behind real environmental harm, and too often it was built in communities with the least power to push back. Neighborhoods with less wealth and less political influence. Now data centers are opening a new chapter in this much older debate about infrastructure sighting. And they're raising many of the same questions. Who bears the burden and who gets a say? That's why I'm joined today by Dr. Sarah Mills from the University of Michigan. Sarah has spent years studying how communities respond to renewable energy development where many of these same tensions have played out. Now, Michigan is seeing growing interest from data center developers, and communities are grappling with what these projects mean for their land, resources, and local economies. So how can communities become better negotiators when the next big development comes knocking? I'm Kate Gasparo and this is Building Better Cities. When I was in Michigan earlier this fall, I was just so caught off guard by how much it was part of the everyday lexicon. It had really become like this lightning rod for how communities were thinking about what they wanted to be in the future. And I think that that was a huge topic of concern for the midterm elections. And I'm curious, like, why have data centers become the issue? And what's driving such strong opposition, not just in the communities you work with, but across also urban areas that are being impacted by this type of infrastructure. Part of the reason it's becoming such a lightning rod is that it's coming so fast. it feels like there are data center conversations everywhere. I'll be honest that I didn't know what a data center looked like where they were cited a a year and a half ago. And now I have a guide about this, but like it just feels very fast to many people. Often there's secrecy associated with them and people are signing NDAs and that doesn't work well. When things are going fast, there's not time for community engagement. And I think ultimately people care about the place that they live. Especially recently, I think that people have felt like government is hard for them to reach and they can have lots of say at the local level. And so they often feel like this is their opportunity to take control of their life, show up and speak out against the data center in the you know, in the place that they care about because they feel like they haven't been included in that process. And, you know, we've seen viral clips about data centers impacting water quality, electricity costs, noise, farmland conservation, and then you're bringing up this critical element of transparency and speed to market. Which of these concerns is well founded in data are really happening in the places that you work and which are sort of being co-opted by different narratives around data centers and their importance or lack thereof for the communities that they're being set up in? That is a very big question. Yeah, yeah. We can break it down. Well, I think that all of those aspects, energy, water, noise. use of land, transparency and also what is this data center like what kind of data is it actually processing? All of those things are I think actually very reasonable questions for communities to ask. About how much water use a data center will use, it depends on what technology the data center developer chooses in your particular place. What impact will data center have on electricity rates or is it going to use clean power or not? All of that depends on the state policy framework for how utility rates are allocated and designed and also whether the state requires that utilities in get power from clean sources. But I think the other thing that's really tricky is that for so much of this infrastructure, the impacts really depend on scale. And I think the thing that's been really hard in this space is that We've had data centers among us for decades, for before we had the internet, right? It was the server room in the office or in the hospital. it's the scale matters and the scale of things brings along with it negative impacts, but it also brings with it significantly more positive impacts too, in terms of the investment per acre. And so, I think that it's really it's been hard. For communities to sort out like the different to grapple with this different scale. When we usually talk about building better cities, it necessarily includes infrastructure we don't always want in our backyards. There's power plants, wastewater treatment facilities, industrial operations, and to your point, increasingly renewable projects, which can be an eyesore, even if they are less smelly and less toxic than other of these infrastructures. Are data centers simply like the next chapter in infrastructure that needs to be cited to support city building and city development? So I think that there are similarities. And I think that this is and kind of putting data centers into the like necessary infrastructure is important context when we're thinking about data centers that host the cloud. And the digital services that we depend on. And again, we know that we're increasingly dependent on digital infrastructure. And so those tiny data centers that have been among us for decades, like we know we're gonna need to see more of them or bigger facilities. I think though, where there's a key difference in the data center development conversation s specifically surrounding AI. And there's a Deep skepticism and I think questions about whether that is necessary. not all people depend on AI because I don't think that there is unanimity that AI infrastructure is an essential service that is something that we really want. I think, and to the extent that AI
Load full transcript
data centers get combined along with cloud. computing like is something that is is hard to disentangle here. And from a land use perspective, honestly, it's really hard to disentangle. Wow. Yeah, the conflation between like what does AI mean, the necessity of cloud services and this as critical infrastructure, like it sort of gets baked into the infrastructure building and the land use needs of it itself. And I'm curious When I listed off the other types of infrastructure that cities need, communities need, they seem to be servicing the population that they're closest to. How much can be said about data centers servicing the population that they are adjacent to, that they are sided within? or is that just another differentiation between critical infrastructure? So I Both agree and disagree with this idea that a lot of that critical infrastructure is servicing what is next near. That's true for wastewater treatment facilities. That would be true for landfills to some extent, right? Even though some people do import trash. it for power plants, that is not necessarily true. And actually, that's one of the things that I hear about a lot from places that have always been importers of energy. From those fossil fuel plants that are on the edges of cities. In rural places, them saying like one or two wind turbines would serve all of our needs, and we're being asked to host 50 wind turbines. But I do think that this is where scale matters to a lot of people. And a lot of folks would say, I'm willing to service my own, you know, like to put up with the inconveniences to service myself. But I don't wanna be an I don't wanna be forced to. deal with somebody else's critical infrastructure. And so I think that there are some this is where people when they're thinking about data centers and that you can't necessarily pin down like whose data is being processed there and whether that is going to serve me. I think that that's that is something that comes into play. But I would put it more into the kind of scale and whether this is seen as something that we want to export to other places that we are a willing host for or not a willing host? You know, as we think historically about where critical infrastructure has been cited, and some of this infrastructure comes with its own environmental burden. So you pulled up the landfills, we've talked about wastewater treatment plants, power plants that are without clean energy. Have disproportionately been cited in communities with the least political and economic power and often disinvested in neighborhoods and communities of color. And we refer to this as sort of the environmental justice issues that a lot of communities are facing today. I'm curious if those inequities that we've seen in decades past with citing critical infrastructure are showing up in the same communities or If they're being community agnostic at this point when it comes to color, socioeconomic status, land use type, can you sort of talk about who this this issue of citing data centers is touching right now? Well, first of all, I'm glad that you're bringing this up. And I putting it in this context too, because I think that this is absolutely the next iteration of industrial development. And we know that industry hasn't always been kind to communities. haven't looked at the demographics. And there's actually a couple of students on campus that I know that are trying to understand where data center developers want to go. My sense is that they they want to go where there's power lines, there whether there's transmission, they want to go to places where there is backbone fiber. Both of those things happen across urban and rural places. Footprint-wise. For really large data centers, they do need to go where they can get a hundreds of acres, so it tends to be more rural. But also, all things considered, you're gonna go to places where land is cheap and property tax rates are low. And so sometimes that does look like disadvantaged communities. We know this from industry in the past, very often this infrastructure is cited in places that don't have the political clout to say no, or who are so dependent on getting new Economic new revenue that they can't afford to say no. This is what you do every day is talking with communities that are going to be and are impacted by critical infrastructure. So I'm curious when you are having discussions on data centers, how much are you relying back on the lessons you've learned with your work with renewable energy? What are some of the biggest challenges that these communities are facing in parallel? And what has an experience? With renewable energy taught you about why people support or oppose critical infrastructure. Some of the lessons that is important in renewable energy that is also true in data centers is that economics matter. So this is something that came from my dissertation. But the people who host energy infrastructure on their property, renewable energy infrastructure, are not doing it for climate reasons, largely. They are doing it for economic reasons. And we also know that when From the renewable energy space, when you include more people in direct financial compensation, you make for happier neighbors, you're able to spread that benefit more broadly. The other thing that we know from renewable energy space that I think also applies for data centers is that the fit of that infrastructure of renewable energy infrastructure in the landscape depends on why people live there. So in landscapes. That are productivist, where there's lots of farmers where people are living there to make money off their land, right? This is this wind farms, solar farms are just another opportunity for that. In places where people are living for view out of their patio door for aesthetic reasons. And we have a lot of rural environments like this, right? Where people buy a house out in the country to retire to, or this is where people have cottages. This is where that
clean energy infrastructure or that data center is a change to that landscape. The other thing that I'm just gonna hammer home because I think it's really important is that we know from the renewable energy space that bigger is not always better. Wind farms, solar farms, data centers are getting bigger because there's economies of scale. But we actually a couple of years ago, we did a survey, a nationwide survey of solar farm neighbors. So these are people living within three miles of a solar farm. And what we found is that among these solar farm neighbors, positive attitudes outweigh negative attitudes three to one when you look nationally. Wow. But solar farm neighbors, a whole bunch of those solar farms are like five or ten acres. When you only look at the solar farms that are over about a square mile, that's it that it's a hundred megawatts in energy terms. But a square mile is the land use. In that situation, negative attitudes outweigh positive attitudes twelve to one. So stale really matters. And as with so many things, like we're just biggering kind of the world. And I and I think that that is a similar reaction. The community members have a similar reaction to this kind of bigger mentality for this industrial infrastructure across the board. Hmm. When you have worked with communities that maybe have had to overcome that 12 to 1 ratio of those who don't like it to do like it, what ends up becoming successful in getting this infrastructure to be built? Or maybe success in these communities is that it doesn't get built and what does that look like. I guess I'm trying to ask like what does it look like to have a successful outcome? And success might mean different things to different people in these situations. Success definitely means different things to different people. And I think one of the things that I've tried to do is to elevate the conversation about what are the benefits. And who is benefiting here. And sometimes there's this perception that like compensating neighbors feels like a bribe. And sometimes, honestly, it is approached by a developer in that way. And on the other hand, if you see this infrastructure as impacting the neighbors, as changing the soundscape, as changing the landscape around the place that they live, I think there's an expectation that. We would in some way include them meaningfully in the project. sometimes that can be done financially. Sometimes that can actually be done by inviting them to have a say in kind of what the project looks like. Actually, an interesting example of this is the Detroit Solar Neighborhoods Initiative. Did I don't know if you followed this since you've left I have our great state. Inform if you could inform our listeners, I'm sure they would love to hear about this too. So this is the idea that Detroit effectively said to meet our to meet the city's energy needs, we could enter into a power purchase agreement to get power from a solar farm on the other side of the state, or we could invite neighborhoods where there's a a bunch of vacant properties to effectively sign up to host a small solar farm. There's l a lot of things that I think that were innovative about that. But one of the things that I thought was particularly unique that I don't see other solar developers doing is that once neighborhoods again they opted in, they were kind of selected, then there was an engineering study done to say like, yeah, you may be a a suitable candidate. The neighbors were given the choice to choose what the fence around the solar farm looks like. They were given the ability to say what they wanted planted under and around the solar panels. Do they want turf grass? Do they want prairie flowering pollinator habitat? Do they want agrivoltaics? So growing food under and around the solar panels? This idea of including neighbors. In what that thing looks like, I think is really important. And the fine an another aspect of Detroit Solar Neighborhoods initiative is that there was meaningful financial compensation for the neighbors around that project. That they were able to like investments in energy efficiency upgrades in their homes. Or if they, you know, were already super energy efficient, which most homes are not, solar panels on the roofs. And so I think. they shared in the benefits of this. And I think that that matters. I I don't want to say they're nicer and there are worser parts of impacts that come from this critical infrastructure, but I do want to call out the environmental issues of choosing the wrong technology when it comes to critical infrastructure and how You are starting to work with local governments about what they should know about data centers as critical infrastructure that we've learned from your work in renewables, that we've learned from past stories of citing environmentally detrimental infrastructure near communities that help empower them to ask the right questions, to demand something that'll be healthy for their constituents. Can you talk about maybe not the eyesore? of this infrastructure, but about the worser impacts of it when it comes to long-term degradation of resources that communities need. One of the things that comes with a data center often is on-site backup generation. And so this is diesel or natural gas generators, right? that has air pollution effects, right? Very localized in that place. It's also loud. Often in states, requirements that the data center developer get air quality permits associated with that. But having on site monitoring, being clear about the times of day that the generators are allowed to run, the extended period, all of that is something that I think tangibly can be done. The clear comparison though that I draw a lot, we've been talking about like what happens The data center is operational. But we also should be thinking about the end of life of projects. And this is something that is standard for renewable energy projects. The technical term is decommissioning,
thinking about taking away that infrastructure at the end. And with met most renewable energy projects, the renewable energy developer buys a decommissioning bond. It's effectively an insurance policy. That says if we're not here at the end of the life of this project, there's money that the local government can draw from to take the infrastructure away. And I think that that is one of the clearest parallels of something directly learned from renewable energy that can apply to data centers in thinking about again the environmental impacts, not just now, but long term. Because think about how many brown fields we have in this country that maybe we wouldn't have had if we Had required industries of the past to clean up after themselves when they're done. Wow, that's a really good point. I don't think we talk enough about decommissioning. And you know, living in Detroit, that is something that you saw every day is the lack of decommissioning responsibility and sort of the challenges that's left with communities for decades afterwards. Yeah. So you've talked a little bit about terms and requirements for decommissioning. for air quality monitoring, for water quality monitoring. but in terms of how you're weighing that as a local decision maker and the tax incentive slash the tax benefit that the developer is getting and giving to the community. Yeah. I'm a huge property tax nerd. like we are too. We've done a lot of property tax conversations on this podcast. So please An important calculus, and I'll and again, I'll start with the renewable energy space where I could talk about property taxes for hours. Is thinking about does the new development, like is it a net contributor to property taxes, or does it place more burdens on those the public infrastructure than what it is taking? For renewable energy, it is a net contributor. It is not sending kids to school. Once the renewable energy project is built, you like it's there's not a bunch of people that are using roads, like it's not generating a whole bunch of trash that you know need to have waste management take care of. And all things considered, like renewable energy projects are net contributors to local tax revenues. I would also say that like some of my research has found community members don't have any sense of what public infrastructure costs, what how much property taxes renewable energy developers are paying. And so some of these benefits that renewable energy projects bring to communities are invisible to residents. And this is why like actually thinking about direct compensation to neighbors is important because when it's paid to the local government, like it just especially if it kind of just goes into the general fund, it can be very invisible. And For a while it was just my work, but there's actually been a recent paper that came out of RMI that effectively said this very same thing that like economic benefits for renewables projects need to be visible to community members. This is among the reasons that a lot of people have thought about community benefits agreements. These are negotiated. You can actually write in what matters to your community. There may be a more public conversation about what the benefits are. rather than just going to local governments, general funds, this is tricky to community benefits agreements, particularly if they are with the local government, it can be tricky to tie that to a land use permit or land use approval. There's legal considerations, Nolan Dolan sheets. This is like federal pot like constitutional rights about takings, not just this is not A s in a specific state. And so sometimes though, local governments can require the a community benefits in exchange for something of value. So when there is public investment or there is a property tax abatement, a partial property tax abatement, for example, they can demand a community benefits agreement or can accept a community benefits agreement in exchange. And so I think The the benefit of that again is that it now becomes there might be more public conversation. You can more clearly tie benefits associated with the project to like you to to visible, tangible things in the community. And maybe target that revenue to the places that matter most. Mm-hmm. To your point though, like going back to data centers, like this is a conversation about like are the harms that are caused. does does this place more of a burden on the community? Like and so thinking about whether that the property taxes that the community might get in exchange kind of cover the potential negative impacts of again, those backup generators running. and po co posing problems for public health. Or again, I mean, usually we've got policies that Think about runoff and that we're protecting wetlands that we apply to lots like all kinds of infrastructure. But I think that that's a calculus that would need to be done. Can you walk us through if you're advising a city council that just received a proposal from a major data center? What would you tell them? Where do you start with them? What questions should they ask? What should they be negotiating for? Just give us a few of the breadcrumbs that you're leaving for them before they pick up your center's guide, what Michigan Local Government should know about data centers. That was a very shameless plug. Yes. I mean and an uncleverly titled guide. I would say if I were a city council person, and what our guide does say is I would I would ask first of all, on let's start with the financial benefits side, like what Are you able or willing to invest in us and our place? Right now, data center developers are trying to find a place, especially the big ones, are trying to find some place to say yes. They are in a race to get their project developed. And so they many of them right now are willing to pay an awful lot to get to a a yes land use permit, a kind of land use approval. And so I would say. I would ask, what are you able and willing to
invest in for our community? For all of the other impacts about decommissioning, sound, air quality monitoring, one of the things that I would say is get it in writing. I actually also say that local governments should ask for a property tax guarantee in writing because at least in the state of Michigan, and I don't know what this is like in other places, but sometimes taxpayers like to argue. About how much taxes they owe. And especially industries that have a lot of clout may argue that, you know, when you approved our project, we thought we would be paying you $20 million or $200 million over the course of our project. In fact, that's it's not worth quite as much now. And so we should be paying you $50 million. And I think that that's not fair. And so I We have an example of this of a solar farm in Michigan's upper peninsula, where the solar developer offered a property tax guarantee that said, We will pay you what we're telling you right now. And if actually we owe you more money because the tax tables change and we as the solar developer owe you more, we'll pay you more. But we will at minimum honor what we're telling you when you're giving us your land use approval. And I so I think getting things in writing is really important. To be honest. Yes. A city council coming member coming to me and asking me what they should do when the developer has shown up is too late. and so that's like human nature. That's when people start to think about planning. is like when they don't have time to plan. But I would say if there are city council people who have not been approached by a data center yet, now is the time to start having the conversation. about what you expect out of industry. Data center development is the newest iteration of industrial development. And even if a data center's not built, there'll be some other industrial development that is interested in your place. So thinking about what does it look like to be a good neighbor in your community? Like what do your community members love about your place? And also what would make it better. The what would make it better is great things to build into a community benefits agreements. Maybe I'll end with something like spicy. You had me here to talk about data centers. And I'm gonna say, to be honest, I think that there are way more proposals than there will be actually data actual real live data centers built. Right now, again, data center developers are just fight trying to find someplace to build. But I'm not going around telling people like, chill out. don't have a conversation about data centers. I have wanting to use this as a conversation to start the like or as a place to start a conversation about what we expect out of data centers or out of industry, what we expect out of land use in our places so that we can prepare for kind of the next thing. What this conversation sheds a light on is that data centers are just the latest chapter in a much longer history of critical infrastructure in our cities. So maybe the real question isn't simply whether communities should say yes or no to a data center. It's how we build a better relationship between developers and the places where they build. And how we make sure the communities hosting our critical infrastructure share in its benefits. If you enjoyed this conversation, please like, share, and subscribe to Building Better Cities. And if you have a moment, please leave us a review. It really helps make a difference in helping more people discover the show and join the conversation. Thanks for listening and until next time, let's keep building better cities together.