Everyone is talking about Data Centers. Recently, host Phil Coover mentioned that he was doing a podcast about data centers. “Do you think that this data center thing is a big bubble?” someone asked. “No,” he replied. “That implies that the demand will somehow go away, causing it to pop. This demand isn’t going away.” In order to break down what is really happening in Data Centers, Phil invited two McGuireWoods’ colleagues with decades of experience in data center real estate, incentives, and government affairs: Greg Riegle, co-leader of the Real Estate Transactions Group and head of its data center practice, and Chris Lloyd, senior vice president and director of Infrastructure and Economic Development at McGuireWoods Consulting. “I don't know that there's been a more – pick your adjective – ‘interesting,’ ‘dynamic,’ or ‘challenging’ time with respect to the data center industry,” Greg says. Tune in for the panelists’ insights about the data center epicenter in Virginia, factors that a community must take into account when inviting a data center (like water consumption), and communities that are navigating the opportunity, like DeKalb, Illinois.

Connect and Learn More

☑️ Greg Riegle | LinkedIn

☑️ Chris Lloyd | LinkedIn

☑️ Phil Coover | LinkedIn

☑️ McGuireWoods | LinkedIn | Facebook | Instagram | X

☑️ Subscribe Apple Podcasts | Spotify

This podcast was recorded and is being made available by McGuireWoods for informational purposes only. By accessing this podcast, you acknowledge that McGuireWoods makes no warranty, guarantee, or representation as to the accuracy or sufficiency of the information featured in the podcast. The views, information, or opinions expressed during this podcast series are solely those of the individuals involved and do not necessarily reflect those of McGuireWoods. This podcast should not be used as a substitute for competent legal advice from a licensed professional attorney in your state and should not be construed as an offer to make or consider any investment or course of action.

Real Estate for Breakfast, where McGuireWoods partner, Phil Coover, brings you essential conversations with leaders in commercial real estate, translating complex industry challenges into actionable insights.

Good morning. This is the Real Estate for Breakfast podcast. I'm your host, Philip Coover, a partner in McGuireWoods Chicago office in the Real Estate Transactions Group. Today we're having the data center episode. This is very exciting for me and I think for everybody because everybody wants to know about data centers, but it's hard to get in there and find out from people who truly know what's going on.

And today, I don't do internal podcasts of McGuireWoods folks too much, usually I have external guests, but today we have two guests who really know what they're talking about. We have Chris Lloyd, he's a senior VP and director of infrastructure and economic development on our McGuireWoods consulting side. And then we have Greg Riegle, who is the co-leader of our Real Estate Transactions Group and he leads up our data center practice, and very exciting to have them both on the show because it's rare to come across people that have actually done a lot of data center work, Chris on the consulting and incentive side and then Greg just doing the work as a lawyer helping people do the data center developments. Greg, we'll start with you. Thanks for coming on the show.

Thanks for having us. It's a pleasure to be here. To your point, Phil, I don't know that there's been a more, pick your adjective, interesting, dynamic, or challenging time with respect to the data center industry, and I thought it would be helpful to give a little bit of perspective on how we approach it as a firm and maybe a little context as to how we've evolved the practice of law to kind of meet the ever-evolving dynamics within the industry.

And as you were kind to allude to in your introduction, I've had the privilege, if you want to call it that, for being around this industry probably for the better part of 25 years, and if you know anything about the history of data center industry, it has very deep and strong roots in Northern Virginia, and not coincidentally, my office is in Tysons, Virginia, and for a long, long time, a lot of my work was centered on land use projects in the Northern Virginia markets, and then and now Northern Virginia still remains the biggest concentration of data center facilities really in the world. And if you ever fly in or out of Dulles Airport, particularly on the northbound departure or approach, look out the window and you will see it dramatically illustrated right before your eyes.

So we got into this industry, true to my roots, as a real estate lawyer doing real estate acquisitions, transactions, and land use for the first generation of these facilities in Northern Virginia, and it's been an interesting sort of dynamic in the way that the data center industry has evolved right alongside our law firm and that we've become much more geographically diverse in our reach. We now serve, I believe, 25 markets coast to coast and coincidentally in all of the big markets where the data center industry is active. Certainly strong presence in Virginia, very strong presence in Texas, California, throughout the Southeast, and of course the Midwest where you and Phil and I think much of your audience resides. And I think what's always distinguished McGuireWoods has been a focus on real estate and land use and parallel expertise and focus on regulated energy and government affairs through our affiliate McGuireWoods Consulting which is now celebrating its 25th anniversary or thereabouts, and I'll pass the baton to Chris in a moment to talk about the McGuireWoods Consulting perspective.

But if you pick up any newspaper or publication, you see that the issues in the data center industry are the ability to get local approvals and navigate the regulatory process, securing and [inaudible 00:03:47] power and general government relations. And it sort of fits the three legs of that stool that we have unique resources in the energy space, the real estate space, and the government affair space, and it's an industry that is facing just an onslaught of regulatory changes and evolutions from all directions and power and energy are sort of at the forefront of the narrative, and I will talk about all that as we go through the morning. But that's what's the firm perspective, and Chris, as we're often want to do, maybe you can round out the McGuireWoods Consulting perspective as well within that framework.

No, I appreciate that, and Philip, again, thanks for allowing us to be with you this morning. I echo much of what Greg said, and just as he noted with the roots that this firm has in Virginia which, as many people are aware, was really where the data center industry was born and has prospered. Virginia now has one of the greatest concentrations of data centers in the world. While Greg's right close to it in Northern Virginia and where it really came to be, I'm in Richmond which is the state capital, which has had to deal with the policy and the regulatory issues related to data centers.

Many of your listeners, Phillip, will know that Virginia was the first state to adopt the sales and use tax exemption for data centers that passed in 2010. So that equipment that is purchased, the servers and the racks and stacks and the HVAC equipment and all that is essential to the operation of a data center is exempt from the Virginia sales and use tax exemption. Virginia was the first state to do that, and now I think it's 37 other states have a similar exemption. There's only a few states that do not have the exemption and many of those that do not have the exemption don't have one because they don't have a sales tax. Virginia really started that trend.

We also adopted a very favorable legislation in Virginia that allows for local government to reduce the property tax rate on data center equipment, that they essentially can say the data center equipment is taxed differently and at a lower rate from other personal property. That too has been a significant incentive for the growth of the data center industry in Virginia which, as I noted, is one of the largest concentrations in the world. Really starting in Northern Virginia, but now you're starting to see as Northern Virginia is filling up and as some of the opposition, public opposition that we'll talk about later has started to get more active in Northern Virginia, you've seen as well as the data center industry has evolved. Instead of building on smaller plots in Ashburn and Data Center Alley, they're moving to large campuses, and the first place you started to see that happen in Virginia was they started to move south down the Interstate 95 towards the Richmond area. So we have a significant concentration here in the Richmond area and you're now starting to see them spread across the state.

As that has happened, some of the incentive issues and the policy issues have started to become front and center in our legislature. In fact, we have one of the shortest legislative sessions in the country. We meet only 60 days in even years and 45 days in odd years where legislators are dealing with 3,500 pieces of legislation and a budget which now approaches 100 billion dollars a year. And we ended our session in March without a budget and we are on a July 1 to June 30th budget cycle here in Virginia. So we have to have some compromise with regards to budget, and the sticking point is whether to continue the sales and use tax exemption in Virginia.

The sales use tax exemption in Virginia, quote unquote, "cost a billion dollars of lost revenue last year," according to some estimates in Virginia. But if you look at the economic impact of data centers and what they've brought in with regards to not only the direct impact of data centers on job creation, local property taxes, but also Virginia's becoming a major hub for the supply chain for data centers, the benefit from data centers far exceeds the, quote, unquote, "cost or lost revenue" coming from the sales and use tax exemption. But because the General Assembly was not able to reach a consensus about what to do with the data center exemption going forward that was set to expire in 2035, our Senate wants to end it at the end of this year with no grandfather provision, we do not still have a budget in Virginia.

That issue and what you see in Virginia, whether it comes regarding treatment of the sales use tax exemption, regarding regulatory issues, related data centers is now shaping the national debate. So having a front row seat here in Richmond to see how it's unfolding from the state policy level is a fascinating perspective to see how it's evolving and what you're seeing, the debates in other states, how they're being shaped by this.

That's super interesting to hear. As an Illinoian I like to hear when other states are grappling with budget issues, it makes me feel better, Illinois situations. I was always wondering, why do you think Virginia was such a leader in the data center? That sales and use tax exemption certainly. Is it due to the proximity to Washington DC or why do you think that Virginia is probably the center of data center?

No, that's a great question. So I think some of it is as if you look back at the history of the evolution of the worldwide web or the internet, obviously the internet was product of DARPA and a lot of the Department of Defense spending, and one of the major nodes for the internet was MAE-East which was one of the main computer nodes near Tyson's Corner. That then spawned the development of the early internet companies, U-Net and AOL, which many of people will remember from You've Got Mail, developed in the Dulles area, and you just had this then the building of the ecosystem regarding data centers, and that of course was in the mid to late '90s.

And you then started to see as the internet moved from just an email exchange to certainly obviously cloud storage, and then you had of course the smartphone evolution or revolution that started in 2007 with apps and the increasing move to digitalization of just about every transaction we have, whether it's financial services, travel, maps, everything we do. Virginia realized, hey, we've got these underlying assets here in Virginia because of the critical investments that were made by the private sector and the federal government, let's make sure we get our fair share of this, because at the time, Virginia, when you look at the cables that were connecting us globally, not many of them came into Virginia, but a lot of them, meaning when they crossed the Atlantic, they came into New York or they cross the Pacific, they were coming into California or Washington State or Oregon, but Virginia had this critical mass.

So the state's perspective was, let's see what we can do to incentivize. These investments and equipment are huge capital expenditures by the data center companies. If we give them this incentive, then maybe we can capture not only our fair share, but more than our fair share. And Virginia has seen the significant, not only job benefits, but also the fiscal benefits thereof.

I agree with all that, and you talk to different people, you hear different perspectives on why Virginia. Everything Chris said is right. I mean, there's been a certain amount of birds of a feather flocking together, if you will, and when the industry was born and was sort of dot-com boom 1.0 in Northern Virginia. You also hear interestingly that Virginia had a certain amount of appeal just because of a fairly moderate climate. We don't get hurricanes, we don't get earthquakes, and the capital investment in these facilities is stable in that regard. There was considered to be mostly pro business perspective both in state government and among many local governments. And I wouldn't underestimate the value or the importance of, frankly, reliable and well-priced power largely through Dominion Energy.

And it's interesting as the industry has moved outside of Virginia by necessity, you ask yourself, well, why is that happening, and it's an obvious point in terms of the rush for scale and size. In many respects, Virginia and Northern Virginia has a land supply problem. The data center facilities I did 20 years ago where a big one was 15 acres or 20 acres and we're now measuring them in hundreds of acre increments. So it's the combination, I think, just a collective growth and back to the power situation that the industry is going, frankly, where power is available and where there's a realistic chance of securing approvals which is exponentially more complicated than it used to be.

If you tell the data center story too, Chris touched on a lot of economic sorts of variables, but these facilities are still probably on a square foot basis or dollar for dollar the single biggest economic development tool that any locality is ever going to bring to their town, city, county, or township. We regularly see local revenues projected in the hundreds and millions of dollars on an annual basis for a big hyperscale campus. So you ask yourselves, why are localities venturing into this, and the answer is a pretty obvious one, if you can find the power, the scale, and the wherewithal to get it done.

So, Greg, how is it generating the revenue for the locality? In my mind, and Chris mentioned it brings jobs, I could see there being jobs to build it for sure, but tell me about how that revenue comes in for the locality.

For the localities too, remember you're getting revenue without traditional development impacts, very little traffic, not putting kids in schools, you're not putting people in parks, all of that. It's local taxation of the equipment and the facilities, which is why to Chris's point, the sales use tax exemption is a powerful tool, but local governments make that kind of money effectively taxing the equipment. And Chris, you live that side of the equation even more closely than I do and certainly feel free to elaborate, but that's where the story starts on the revenue side.

No, Philip, Greg's exactly right. I mean, you take Loudoun County, for example, which is where Ashburn and Data Center Alley is. Last year, Loudoun County got $900 millions worth of property taxes off the data center industry. You look at Henrico County which is here in the Richmond suburbs, they get hundreds of millions of dollars each year from the data center industry, and in fact have dedicated $60 million to an affordable housing fund. It's just the order of magnitude of what you can do.

There's a county here in Virginia, Mecklenburg County which is right on the North Carolina border, which is where Microsoft located a significant campus. This is a community that had relied on furniture, textiles, and tobacco for its entire life, and in the '90s, of course, the bottom fell out. So a fairly rural county, Microsoft came in, and that county now has, because of the money, the hundreds of millions of dollars that comes in because of the data center taxation there, the property taxation, that county now has the lowest property tax rate in Virginia at 38 cents per $100 of assessed value.

It has rebuilt its entire school system with cash in the past five years, and with impacting, I think what I heard the county administrator said, it's four-tenths of 1% of the landmass of the county. It's on a thousand acre tract of land. It's surrounded by trees, well off the road. Nobody even knows it's there. And so Greg was talking about as these campuses get bigger, they're going to these areas where to have power, but they don't have some of the issues that they've had nor the urban areas and they just sit there and spin off a lot of money.

And again, the only thing I will somewhat challenge, Philip, what you said is that data centers don't create a lot of jobs. And yeah, I guess if you look at it, a quarter million square foot data center will maybe only have 25 to 50 direct jobs that are associated with it, whereas a quarter million square foot warehouse might have several hundred people associated with it. So if you look at it jobs per square foot, it certainly is less than a distribution facility or a manufacturing facility, but this pay scale is very high. The spinoff from that in the contractor and the ecosystem from that is significant. And you get some of these projects that are of scale that are on several hundred acres and you may have several thousand people working at that campus which are making on average anywhere from 75 to $100,000 a year. Again, the economic benefit and spinoff that comes from that is very significant, without the traffic impacts, without the other impacts that you may see from more intensive development patterns.

Totally makes sense. I mean, you're taking in a lot of instances unused land or lightly used land for agricultural purposes where there's very little jobs and you're putting jobs on there and you're making it high tax, real property taxed land. Some of that historically agricultural land is taxed at a pretty low rate. So once you put that data center on top, that's going to really spike the property taxes. We're going to talk about DeKalb, Illinois in a little bit, but that's my hometown. That's where I'm from. So anything we can do, which is unique in that it's like coming up in the national conversation, and so I'm always looking for a way to revive my hometown. So we're going to talk about DeKalb in a little bit.

But Greg, what I want to ask you to kick off the conversation is you field calls all day, every day from people that want to develop data centers. And if anyone in the firm has a client or potential client that wants to do data center, we're like, "We got to get Greg on the line," helping people probably figure out if these are real projects that are really going to take off. When you get an initial call and you have someone that wants to develop a data center, what are you trying to evaluate in terms of the feasibility of the project and how you can help them?

It's a great question, and a lot of times it is a one-two punch. Chris and I spend a lot of time together on the front end stuff in jurisdictions all over the country answering that question. It is a different answer than it might've been 10 years ago. It really does start with the power in terms of have you figured that out, and truth be told, the power companies and the power providers are getting more sophisticated in their own analytics, and I think the process is more involved, the upfront costs are bigger. They're effectively trying to separate the players from the pretenders in which of these massive projects are really real and going to happen. But it builds on everything we've been saying. I think job one is, can you confirm the availability of energy, and that's followed by what's the climate on the ground locally from a political standpoint.

We approach it with a robust network of local expertise. And that last question is the ultimate local question and I'm not going to answer it from Tysons, Virginia, but we've got a well-built sort of infrastructure of local land use council, local political operatives that can give a read as to whether there is the wherewithal or the awareness. I think sometimes it's a function of are there existing data center facilities. History shows and experience shows there could be a point of saturation. They tend to be popular in the first one or two that go into a jurisdiction because of the revenue, and oftentimes there's an attempt to draw a boundary in terms of just how much of that scale a community might want for any number of reasons, fair or unfair, legitimate or unlegitimate.

Power, question two, is getting more complex. A few years ago, it would've just been evaluating timing and availability of traditional grid connections, but we spend a lot of time now looking at so-called interconnection issues whereby the data center campus might be generating its own power through gas turbines or things like that and effectively selling power back to the grid. The tip of that evolutionary spear is going to be nuclear considerations, for better or worse, which is a whole new realm of both regulation and construction dynamics and so forth. I think anybody that's active in this industry would tell you that it is going to be very challenging for traditional grid power to meet the needs of the industry at the pace it's evolving and growing. And certainly, Chris, you ought to comment as well in terms of what sort of first questions you're asking and what job one looks like when you get the same calls I get in terms of how to sort these opportunities out.

Greg covered the waterfront pretty well. I think then it's not only the objective factors, but some of these subjective factors, and that is is this a community that welcomes development or is it one where that the people don't want to see development, not only just data centers, but don't want to see development at all. Are the natural resources there? Because data centers do consume, obviously they're moving to a lesser of this, but they do consume water. You do have to get rights of way for utility lines to get to these facilities. And so is this a community that understands that industry does require this sort of infrastructure and is it willing to be a partner in developing that infrastructure?

So that kind of that community receptivity, the vision of the community to make sure that the windfall of money that it's going to get is not squandered, not only not is it squandered, but it can be unpredictable sometimes because valuations change. We've seen that happen in other places around the country where communities start to pledge the resources they're getting to ongoing operating costs and then the assessments fall one year for various and asunder reasons and the money isn't there. So how do you segregate that and make sure that the community still can pay its bills and doesn't become too dependent on it?

So that again, that governance issue, that community receptivity issue, then also just looking at is the permitting expertise there. These are very complicated facilities and if a community doesn't have the permitting expertise, they're just used to dealing with residential properties or agricultural facilities, are they going to be used to how do you permit something like this? Those kind of issues are very important as part of the analysis, in addition to some of those more tangible items that Greg was talking about.

You're taking, in a lot of instances, unused land or lightly used land for agricultural purposes where there's very little jobs, you're putting jobs on there and you're making it high taxed, real property taxed land. Some of that historically agricultural land is taxed at a pretty low rate. So once you put that data center on top, that's going to really spike the property taxes. We're going to talk about DeKalb, Illinois in a little bit, but that's my hometown. That's where I'm from. So anything we can do, which is unique in that it's like coming up in the national conversation, and so I'm always looking for a way to revive my hometown. So we're going to talk about DeKalb in a little bit.

But Greg, what I want to ask you to kick off the conversation is you field calls all day, every day from people that want to develop data centers. And if anyone in the firm has a client or potential client that wants to do data center, we're like, "We got to get Greg on the line." So you're always ferreting out helping people probably figure out if these are real projects that are really going to take off. When you get an initial call and you have someone that wants to develop a data center, what are you trying to evaluate in terms of the feasibility of the project and how you can help them?

It's a good and it's appropriate question that we've all said different ways, that is the first item of priority to sort out. To your question, I mean, the nuclear component takes on a couple of components. Many in the audience have probably seen or heard about nuclear power plants restarting to effectively serve the data center industry. And for those that are old enough to remember, Three Mile Island is being restarted for a number of reasons, but not the least of which is to power a series of Amazon web service data center campuses in that part of Pennsylvania.

And the whole notion of directly co-locating large data center campuses with the power source, in some instances, a nuclear power source, I think is clearly a trendline and that has a number of advantages in terms of eliminating the need for long distance transmission lines. There's a certain sort of logical adjacency of uses by putting the large data center campus at the power generation source if you have the land and the capacity to do that. And the regulated utilities are leaning into that fairly heavily, and we've done some interesting partnerships between regulated utilities and the industry to meet that need.

But beyond that phenomenon, and I would've told you that not long ago that SMRs may be something that we're talking about in 10 years from now. That future is now. There's a pilot project in Virginia that's between Dominion Energy and one of the big hyperscalers. There's some stuff in Texas that's coming to fruition. In its simplest form, an SMR, small modular reactor is kind of what the name implies. It's effectively a self-contained package nuclear power source that can be co-located on the data center campus and self-contained, doesn't generate noise, doesn't generate any sort of outward emissions, not like a traditional nuclear power source.

Raises an additional dynamic in the regulatory sphere in that, not surprisingly, the Nuclear Regulatory Commission has primary safety authority over really anything nuclear, and there was preemption in terms of state and local actors not getting involved in the nuclear safety aspects of those facilities, but you're layering on an additional approval process. To say that you've got a concurrent educational exercise with local governments and community stakeholders is probably a massive understatement in terms of what the realities are and aren't of a nuclear option for a data center.

We find in many cases those narratives are easier in communities that have some orientation to nuclear power which isn't a large number but there are some out there that have kind of grown up with that source of power, but goes back to just demand. And even if you're not in real estate or in the data center industry, you hear about the growth that AI is necessitating, and by turning to these non-traditional or non-grid sources, you're not consuming power from the grid. You're in some instances building grid resiliency by the ability of the data center campus to sell power back to the grid, and it takes away the argument that you're consuming energy that could either be devoted to something else or raising the cost to consumers and all those sorts of things. So I think you're going to see more of that. You also see a movement toward natural gas power in addition to traditional electrical grid power for all the reasons that I've been talking about. But Chris, I'll give you a swing at that if you want to add any.

To be clear, on the natural gas side, what you're seeing is not only the addition of concentrated generation, meaning traditional power plants, but you're seeing a number of places, particularly Texas, New Mexico, where data centers are developing their own natural gas fire generation onsite. You're seeing that. You're seeing a significant commitment to renewables. The hyperscalers have a commitment to renewables and that being part of the energy mix. They're combining that with batteries backup and storage because, again, the sun doesn't shine all day and the wind isn't always blowing. And the scale of these facilities, the amount of energy that they consume, I heard an example yesterday, I was at an economic development conference and someone was saying some of these large campuses, in order to have enough solar to power them, you'd need a solar field the size of 3,000 football fields. Well, that's just not going to be feasible in a lot of places to have that kind of a solar field available. So you have battery storage, you have natural gas backup, you have SMRs which are coming on board.

We've just had here in Virginia, it was announced recently that a commercial fusion plant is going to be developed here in Virginia. It will be the first commercial fusion plant in the world. They've signed a power purchase agreement with a data center offtaker. They just filed yesterday their application for interconnection with the PJM regional grid. So you're starting to see people look at these new technologies like what Greg was talking about, but it really is a energy story.

Thanks, Chris. That's super interesting to hear. I mean, I run into permitting challenges just on simple commercial developments for communities that aren't used to it. So I can only imagine on something with the scale of a data center that that would be more challenging, especially because they probably, a lot of them don't have any experience until the first one comes along. Let's talk some more about energy. I want to hear about SMRs. I know, Greg, you reference nuclear power. We were laughing the other day about as a land use attorney, if you ever thought you would spend time in your day talking about nuclear power. I'd love to hear about the SMRs, but I'd also love to hear just more about what's the success story in terms of how are they tapping into the traditional grid, just more on the energy front.

Yeah, I think you stated very well, Philip, in that the power is the biggest cost of these data centers for their ongoing operating costs. So they have every incentive to reduce the amount of power. Where does technology with regards to having more efficient chips or even moving to something like quantum computing, where does the line start to intersect between innovation versus the power versus the demand? Because you're right, I don't think the demand's going to fall off and I think it's only going to accelerate, particularly once you go to autonomous vehicles and other things which consume vast amounts of computing power which we haven't even scratched the surface of.

You knit all that together and what data center project, again, 15, 20 years ago, your diligence would've been primarily in terms of can you get the local government approvals together to get the thing approved, and in many states from a electric utility standpoint there's effectively an obligation to serve what is otherwise constructed, and that's less simple now. But now you've got this patchwork of regulation in terms of driving a data center solution. You've still got to sort out the local approvals wherever you're going to go, whether that's DeKalb County or Cook County or Fairfax County, Virginia, or whatever the case may be.

But then you've got the traditional grid power is primarily state regulated and then now you're moving into a frontier with potentially nuclear and other sources which are federally regulated. And people like Chris and I and our teams and the other lawyers at McGuireWoods have to sort out that layer of regulatory things that are all necessary to bring one of these to fruition, which is again, kind of separating the players and the pretenders. It's just not as simple as it would've been years ago to find the land and convince a locality that the upsides there and plug into the power. Those were the days, as the expression goes.

I totally agree with everything because I had someone yesterday at a cocktail party ask me, I was excited about doing the data center podcast and they said, "Do you think that this data center is a big bubble?" And I was like, "Well, no, because that implies that the demand will somehow go away and that the investment will no longer be needed." And he was referring to that, he's like, "Because we don't have enough energy." I'm like, "Not at all because the demand for the data centers and for that computing power is just growing." And so we got to figure out how to get the energy, but it's not going to burst, like all of the investment that's made into it will just somehow evaporate because that demand is multiplying with every day. And so it's more of a function of how are we going to get creative with generating the power that's going to be required, rather than a bubble, like the investment will just disappear.

The principal reason or need for water is you use your laptop or your computer, you notice it gets hot after a while. That's because the chips give off a lot of heat just because of the electronic, the connections that are going on in the chips. So water is used as part of the cooling process to make sure that the chips don't overheat because if they overheat then they're less efficient and effective. So water is used in the cooling process, whether it's an evaporative process or other things that help to cool the chips and keep the data center from, again, overheating. But what you're seeing is people are concerned about the amount of water that that process requires. And the old technologies did require a lot of water because these facilities, the bigger they get, the more heat they generate and that requires more and more cooling activity.

Again, after power, water was a big expense for these data centers, so they too are looking for ways to mitigate that. So you've seen a number of creative solutions, whether it's reused water, so non-potable water that's coming out of industrial applications, or water that's been treated but not to human consumption standards by wastewater treatment facilities that are using that to cool these facilities. You're seeing people move to closed loop systems so they're not just constantly pulling fresh water out of the system but instead have a closed loop systems. So they're reusing the water over and over and over again to do their cooling. You're seeing some interesting solutions with regards to putting data centers in cold weather climates, whether it's the northern part of the US or Iceland or something so you've got natural cooling that takes place. You've seen some interesting geothermal solutions that are out there.

Again, there are concerns out there by people regarding water use. A lot of that is based on an old understanding or a old technology that a lot of the data centers are moving away from. So I think that problem and again, the creative solutions that are out there, you're seeing even facilities moving away from water altogether where they're essentially immersing the chips in some sort of cooling agent. It's a special cooling agent. So you don't even need significant amounts of water for the facility. So again, I think you're going to see technology helping to take care of a lot of the concerns regarding too much water use.

Chris, we've talked a little bit about the water consumption. What are you seeing in terms of regulatory activity and kind of public sentiment towards the consumption of water? And maybe even start with how do the data centers use water. I understand it's for cooling, but maybe you could elaborate on that.

Illinois is certainly a big hub for data centers. When you think of the big states, Illinois, Texas, Georgia, and Ohio are really where you've seen the greatest concentration in the US with several other states, Arizona and Oregon out west there. And so it has historically been a very strong market for data centers. You're starting to see some challenges out there. You saw several weeks ago there was the town in Wisconsin which decided to not accept a data center and impose a moratorium. You've got Governor Pritzker talking about a statewide moratorium, data centers, and you're seeing that in Ohio, you're seeing that in other states. Heretofore, Illinois has been a real good market for data centers and particularly outside of just the Chicago metropolitan area.

Well, you talked about Virginia earlier just being good from a temperature climate standpoint. Is Illinois also pretty good? I mean, we don't have earthquakes or hurricanes either, and our biggest complaint is that it gets cold. The cold might help. If it's cold six months a year, that could help.

My sense is, I have heard it referred to as DTown, but my sense is DeKalb has been sort of leaning into the forefront of the regulatory stuff, you want to call it that. They've adopted some things, reduced water usage, encourage energy efficiency, and so forth. I think you can read as wanting to be open to the industry and the economic benefits, but do it in a way that is as responsible and stakeholder sensitive as possible. And you see some places that make it well known that they want no part of it. What I know of the DeKalb regulatory history is we may be willing to consider it, but we want it to be at the forefront of all the technology Chris described to try to ensure that these facilities can be compatible in the long term.

Yeah. So let's talk about DeKalb and a few that are in our notes there's some water, light onsite power proposals. Do you all know anything about what's going on, DTown?

Given all the controversy that's around data centers right now, that kind of regulatory clarity is actually welcomed by the data center companies. I think they would rather know the rules of the road coming in and find a community that has been very thoughtful in the way that it's approaching it from a zoning perspective, from a utility perspective, from a water use perspective, from a how are you going to get the power there perspective, and they've gone through a consensus building process instead of going into a community that may have no standards and thinking, "Oh, this one's going to be easy," and running into the buzzsaw of opposition. So as long as I think that a community has struck the appropriate balance between preserving and protecting what it holds sacred but still being open to the economic and fiscal benefits that come from these, again, the data center companies really welcome that because they appreciate that clarity and that speed to market that that kind of clarity delivers.

And it's a really good point. What are we looking for in practice? I think some of the tools you see that are a good indicator of what Chris was describing is many places both in the Midwest and the East Coast and other places have adopted so-called technology overlay districts which is in its simplest form a locality saying, "This is the spot where we think the industry can be compatible." And it's not a total panacea, but at least you're starting from a standpoint where the establishment of that technology zone went through some sort of public process and the stakeholders in the community had a say and everybody agreed presumably that this was where we're going to try to channel the industry, and that's a huge, huge help.

Again, it's not a free pass, but at least you're starting from a standpoint of where the community has tried to tell the industry they want to go, which for a place that wants to harness the economic upside of this industry and try to manage some of the outcry and the uprising that's happened, that's often a good way to do it, at least as a starting point.

And this, Philip, may be painful for people from Illinois to hear, but I would point to Michigan as a good example which I think from a state policy perspective actually approached it very well, and that is when they updated their sales and use tax exemption in 2024, they built into this statute. They worked with the data center industry on coming up with that fair and balanced approach to regulation. So they extended the sales and use tax exemption out into the future, but in return for that they asked that the data centers adopt certain sustainability provisions with regards to their building design, that they work towards using more renewable energy, and also work towards certain water conservation measures. And I think that's actually a very good example for a number of states to look at about how you balance these issues out.

Thank you for being sensitive to our Midwest rivalries the way you phrase that. Who is building these data centers? Who's out there? There's a lot of terminology when people talk data centers, but tell us about what a co-locator is, what a hyperscaler is. Who are the people, the companies, the types? You don't have to name names if you don't want to.

It's a good question, and I think let's start with the framework of your question and we can elaborate from there. But the hyperscalers, that end of the spectrum, those are the big users. The names are obvious. We can name them. I think they're top of mind for everybody. That's Meta, that's Amazon, that's Google, Oracle, yes. Total trade and scale, they have demand for those facilities independently and they are probably the biggest driver of industry growth at the moment. It's interesting from a real estate perspective that they're moving at a pace and scale where they often need local or regional or national merchant developers to be securing the sites, running the traps on the entitlement, wading through the power sorts of stories and everything. Notwithstanding the vast size and scale of those companies, they don't have the bandwidth to simultaneously be pursuing 30 sites in 20 different states. So there is definitely a role for the development community however you want to define that.

If you turn back the clock 10 years or more, you would've found the so-called co-locators being the dominant player in the industry, and those are companies like Digital Realty, Equinix, and others that were really there from the beginning and they were effectively building and managing the facilities and selling the space or renting the space, leasing the space, technology companies to law firms like ours, to banks, to insurance companies who need to store, process, manage data. And they still have a role in the industry, and frankly, the hyperscalers are a big customer of theirs as well. But a co-locator is effectively someone who is just leasing space to multiple tenants, considered a multiple tenant building effectively.

Those are probably still the dominant players. I mean, even further in time, you would've had banks and technology companies and so forth having independent data center facilities. Some of that still exists, but it's of a scale exponentially smaller than what we're talking about right now. But it's definitely the so-called hyperscalers driving the demand for obvious reasons and still a meaningful place for all the co-locators as well.

And then I would say that then the next tier you have a number of firms which are essentially real estate developers, but real estate developers of scale. So you have companies like Tract, some of the large industrial development firms, name names, but there are others that are out there that are assembling these large parcels of land that have relationships with either the hyperscalers or the co-lo companies. And so they, to Greg's point, since the market is moving so fast and so quickly, they'll go out and assemble land and essentially bank it for five or 10 years, get it ready for when the market needs that site, and then being able to bring it online with either a large co-location company or one of the hyperscalers. So they would be probably the third tier of entities that you're going to bump into in the marketplace.

That dynamic we both described does create some challenges because when we walk into a local government, the question always comes up, well, who's it going to be, and a lot of times you don't know that answer. I mean, you're talking lead time is measured in years to get these projects done. Look, it's a fair question. If you're sitting in the seat of being a mayor or a county commissioner or a township supervisor or whatever the case may be, you don't want to invest the political capital in something that's not going to be real.

I think it's something that is a challenge for the industry and a challenge for all of us to try to convince a locality that what they're investing their time and their political capital is something that's going to happen. And as we both said, the pace and the scale, it's not always known who the operator is going to be in many cases till after some of the approval process is done. And much like retailers, they're not interested in the site till you've got a position and got some of the approvals in place.

Thanks for filling us in, everybody. Gentlemen, we're almost up on time. I want to be respectful of your time. I could talk to you for hours about this. I feel like that's really only the tip of the iceberg and maybe that's how we are, maybe we do a part two here in a few months and see how things have developed. If anyone out there is looking to develop data centers, Chris and Greg are your guys. They're going to help you, guide you through it, get you to the right people, help you figure out the process. Do you have any parting thoughts on the future if you have your crystal ball out or do you have anything else you want to say to the people before we step off?

No, Philip, I appreciate that, and Greg, thank you. So the only thing that I would add is that I think the political climate regarding this is only going to get more bitter and divisive. So the idea that you're looking to develop a data center or be involved in one and think that it's going to fly under the radar, people are not going to understand or know what's going on or that they're just going to be bowled over by the fiscal benefits of these, those days are over. Most of it driven ironically by social media, including some foreign interests that are trying to undermine it, a lot of environmental interests that have global connections that are trying to undermine these processes. And so I think you're going to see these debates only intensify at the local level and the state level.

And so going in with your eyes wide open, going in with a strategy to be engaged and not try to do this through subterfuge, going in trying to educate a community with facts and telling the good stories that are associated with these projects, and going in and working with a community to do something that's for them and not just a one-size-fits-all. I think that the industry is going to have to do that if they want to be successful.

Very well said. Thank you both for your time and your expertise. Thanks for doing your part to educate the public on what's going on. We appreciate it here today. Thank you so much.

Thank you for joining us on this episode of Real Estate for Breakfast. To learn more about today's discussion, please email host Phil Coover at pcoover@mcguirewoods.com. We look forward to hearing from you. This series was recorded and is being made available by McGuireWoods for informational purposes only. By accessing this series, you acknowledge that McGuireWoods makes no warranty, guarantee, or representation as to the accuracy or sufficiency of the information featured in this installment. The views, information, or opinions expressed are solely those of the individuals involved and do not necessarily reflect those of McGuireWoods. This series should not be used as a substitute for competent legal advice from a licensed professional attorney in your state and should not be construed as an offer to make or consider any investment or course of action.