Ohio House and Senate Select Committees on Data Centers - 5-27-2026

Here.

Here.

Here.

Senator Blackshear.

Representative Claggett.

Representative Workman.

Representative Glassburn.

We pledge allegiance to the flag of the United States of America and to the republic for which it stands, one nation under God, indivisible, with liberty and justice for all.

Yes, thank you, Chairman, and thank you to everyone on this committee, their commitment and their leadership and understanding of the urgency of the situation. Very appreciative. A lot of work already done just to get us into this building together right now, and that's a statement on behalf of all of Ohio. As a review, the Joint Data Center Committee, the mission specifically is to ensure that Ohio citizens have accurate, relevant, and usable information concerning the economic, environmental, and security impacts of Ohio's data center development through discussion with experts, stakeholders, interested parties, and Ohio citizens. We have five specific tasks. One, determine data energy cost and availability impacts for other Ohio rate payers. Two, identify state and local community economics of data center development. Three, determine the environmental impacts of data center development. Four, identify public safety and national security considerations. And five, increase public awareness and sector knowledge. Energy is a prime mover of modern society. Energy is not simply one sector of our economy, it's the sector that enables every other endeavor of our society and human life. In our modern society, access to information and data has become equally important in providing better lives. Reliable energy and information resources are not only primary 21st century economic drivers, they're also a national security imperative. In summary, energy is the foundation of human progress. Data is the driver of human capability. We want every Ohioan to have access to these societal benefits in a positive way that best supports our communities in all aspects. The Joint Data Center Committee of Ohio will accelerate us towards that mission accomplishment.

Thank you, Co-Chair Chavez and Co-Chair Holmes. My name is Dan DiIorio. I am Vice President of State Policy for the Data Center Coalition. DCC is the National Membership Association for the Data Center Industry. My role here today is to provide a level-setting conversation for why data centers, why now? What is driving data center development? How do they factor into the economy of Ohio, our national security, and ultimately how do we address important issues around water and energy? I do have a presentation that I believe was distributed out to the members here, both electronically and a hard copy, and I'll be reading from that slide deck right now. First off, you can see our membership. We have over 40 members, which are leading data center owners and operators. You will see many familiar names on that list there. You'll also see names that have been long investing in the state of Ohio. With our owner-operator members, we do have as well large end-user members. So these are companies that lease large amounts of data center capacity. We have associate members who are pre-operational, building their first data center currently. But importantly, we also have our industry advisory council. These are partner companies within the business ecosystem that are working closely with the data center industry on important cooling systems, mechanical components, electrical components. including the company Vertiv, a homegrown Ohio company here that has just announced an expansion to help supply cooling systems for data centers. To begin, there are generally two types of data centers. You have a self-perform or an enterprise data center. This is a company that will build a data center to own and operate for its own use. Typically, these are cloud providers or companies like Meta, which are owning and operating a data center for their own products and services. You could also actually throw a company like Visa, the credit card company in there. They are a member of the ECC. They have two proprietary data centers that they own and operate to facilitate credit card transactions. In fact, their one in Northern Virginia does 31,000 credit card transactions a second. The second type of data center is a multi-tenant, a build-to-suit data center, or a co-location data center. This is a company that will build a data center and lease out space inside. They could lease out an entire building or a series of buildings to one tenant. They could have 20 tenants in a building, 200 tenants in a building. It varies. And those tenants can be local health care providers, local hospitals. Those tenants can be payment processing companies. They can be local banks and financial institutions, big box retailers, those cloud providers who need additional capacity. Basically anyone and everyone who needs access to digital infrastructure. It's important to remember why we have data centers and ultimately what has led to the evolution of moving to a data center model. Computers at home or we had server closets in the basements of buildings. Those closets were not particularly secure, nor were they well-cooled. But as we saw the migration to data centers throughout the late aughts through the 2010s, what we realized is that by aggregating our collective computing needs together, we can leverage economies of scale to achieve significant efficiencies. From 2010 to 2018 computing output rose 550 while energy consumption only rose 6 And while we are in a significant high growth period now that principle still holds true This is a far more efficient way of doing all of the computing processing and storage that we need to facilitate our daily lives and the 21st century economy. Data center demand has been on an exponential curve. According to McKinsey, data center demand in the U.S. is expected to reach 80 gigawatts by 2030, up from 24 gigawatts in 2024. I'm sorry, 25 gigawatts in 2024. So an over-tripling. This is being driven by consumers and businesses, all of us demanding the services, technologies, and devices that we rely on every day in our everyday lives. The U.S. is 40% of the global market. When you look to the graph on the right, what you'll see is variations in data center demand. We are constantly recalibrating and our companies are constantly recalibrating data center demand. Just as a utility or an energy grid does not want to overbuild or underbuild, same thing is true for the data center industry. We want to ensure that we right-size these investments, that we are building out appropriately for this infrastructure, for this steel in the ground that we all require. And so we are constantly looking to ensure that we are right-sizing that demand and reevaluating that demand. We are all driving data center demand. Consumers and businesses will generate twice as much data in the next five years as they did in the previous ten. The increasing digitization of the economy and the services that we rely on every day are helping to drive data center demand. Approximately two-thirds of the global population are online, with 45% of that coming since 2018. We still have a third of the global population that need to come online, and they're coming online very quickly. But to bring this down to the most relatable level, the average household has 21 connected devices. It's phones and laptops. It's watches, thermostats, ring doorbells, multiple smart TVs, garage doors, lights. My oven is connected to Wi-Fi. I can preheat it from here. And if you have teenagers, you can effectively double that number. All of this requires digital infrastructure. And so as consumers and businesses, we are increasingly using the Internet to not just provide the services in our homes that we like to rely on, but also to facilitate the 21st century economy. Whether it's remote work or other opportunities, monthly household broadband consumption is rising significantly. You'll see the graph here. We use the Internet a lot, and if you multiply that graph out by your immediate neighbors, your street, your neighborhood, your precinct, your city, your town, your county, and your state, As you can see, the need for connectivity for the Internet is driving the development of the digital infrastructure that supports that. But it's important to remember that data centers are not just AI. There is a consistent misperception that all we are doing now is building out the infrastructure for AI. AI is an important growth driver. There is no question about it. But as I've described, there are many products and services that are being facilitated by digital infrastructure. Telehealth appointments, electronic health care records, online banking transactions, any financial transaction, state and local governments, 911 geolocation services, online purchases, you name it. Everything that you do in your daily life requires a data center. As you can see here, there are various products and services that we've described. Cloud is the underpinning of everything. 95% of Fortune 500 companies operate on the cloud. There are business applications healthcare applications all the connected devices There e online learning opportunities autonomous vehicles if you ever had a chance to ride an autonomous vehicle The things that help us relax like streaming video But at its core it is innovation. It is innovation in the economy, innovation in the products and services that we rely on, and innovation that helps to drive the U.S. as the global leader in technology. And again, going back, you'll see on the graph here, AI is an important growth driver, but cloud remains the underpinning of why data centers are being built today. The vast majority of data centers being built in the United States are to support cloud infrastructure. AI, again, is an important growth driver, but it is not the be-all, end-all of data centers. It will continue to grow and to become an important part of overall compute. But right now it is currently less than a quarter of current compute, and while that will grow, cloud is still the primary driver of data center development. But importantly, as we look globally, data centers and digital infrastructure are a crucial component of maintaining the U.S.'s global economic competitiveness. We need to keep pace with other countries that are building both digital infrastructure and energy infrastructure at a rapid pace. The U.S. does it right, and we need to remain the world leader in advanced technologies. But importantly, data centers play an essential role in our national security. This is our data, and we want it stored here, safe and securely. As you can see on this image here, data centers are the underpinning of several different national security factors. We have military operations and cybersecurity, increasing cyber threats. Data centers play a crucial role in ensuring that our data remains secure. As I mentioned, 911, emergency services, disaster response, critical infrastructure like grid and water and telecom resilience, economic systems, banking, payments, financial security, government services, secure communications, space and satellite. Digital infrastructure is the backbone of our national security, and our national security depends on robust digital infrastructure so that we can house our data here and ensure that we can provide the robust domestic protection that digital infrastructure provides. Why do data centers cite where they do? There are a variety of factors for where data centers will choose to develop. The overarching factor is time to market. We are on an increasing exponential demand curve. All indications are that the industry is behind meeting that demand curve. So time to market is key. Within that, a variety of factors are evaluated by data center companies. You have access to reliable, clean, competitive energy. You have low risk of natural disaster and climate considerations. Land availability and costs. Access to fiber. Fiber is the network. We need to have ready access to fiber. Ownership and occupancy costs, a skilled construction and technology workforce, something Ohio has in spades. And most importantly, I think, a good tax and regulatory climate. Ohio is a great place to do business, and that is why data centers are locating here. But importantly, the data center markets are becoming increasingly competitive. On the graph here, you have your primary markets and your secondary markets, which are essentially your top 10, top 11. But increasingly, we are seeing absorption within the frontier markets or the developing tertiary markets. Data center development is getting increasingly competitive throughout the country. No longer is it concentrated in those top ten markets. Ohio is likely within those top ten. It depends on which real estate firm is evaluating it. But Ohio is a strong, growing market. data center market, but it is important for Ohio to remain competitive because we see states like Pennsylvania, we see states like Indiana, we see states like Michigan, which have significant development projects announced there. We also see a state like West Virginia, like Kentucky, also increasingly becoming competitive for data center development. What do data centers bring to Ohio communities? These are tremendous capital investments. A couple years ago, I would have said that they are $100 million. Right now, a billion dollars is most likely the floor. These are multi-billion dollars, steel-in-the-ground capital investments. With that comes significant tax revenue, especially at the local level, property taxes, but also just the general economic activity that data centers generate, whether it be labor income, whether it be other sales and use taxes, whether it be just overall capital expenditure taxes. These are high-wage jobs, good-paying jobs. They are low demand on services. Data centers do not require the amount of public services as many other developments do. In a place like Loudoun County, Virginia, we see for every $1 in public services that data centers utilize, they give $26 back in revenue to Loudoun County. This is substantial construction jobs and activity. Data centers work very closely with the building trades because we rely on a skilled workforce to build our facilities. These crews then got contracted for multiple sites on a parcel or often moving, I'm sorry, multiple buildings on a parcel and then moving from site to site within a given jurisdiction. These turn these construction jobs into longer-term jobs. What used to be probably about a two- to three-year job is now turning into a four-, five-, six-, seven-, eight-, nine-, ten-year job. This provides a lot of certainty, the ability for the building trades then to ensure that they can stay in the community, put roots down in the community, and have consistent, steady work. But data centers rely on those same skilled trades when they're up and operational. We need to ensure that these facilities run at an optimal level at all times. We depend on local electricians, local plumbers, local HVAC technicians to continuously maintain these facilities. And that comes with years of reliable support for local economies. They build and they strengthen the tech ecosystem. Ohio is now becoming the Silicon Valley of the Midwest. And this is data centers' play, and having ready digital infrastructure is a crucial part of maintaining that title as the Silicon Valley of the Midwest. And data centers are a catalyst for energy development. We've just come out of about 20 years of flat load growth. And now we are experiencing significant load growth on our system from data centers, manufacturing, electrification, and hydrogen production. We need to invest in our energy infrastructure. We need to invest to ensure that we have a 21st century grid. and data centers provide a crucial role in facilitating that development, and I'll get into that in just a little bit. As you can see on this slide, the U.S. data center industry supported 1 million direct jobs in 2024, 5.5 million in total employment, which is a 17% year-over-year increase, 525 billion in labor income. For every one job in a data center, 4.5 jobs are supported elsewhere in the economy, $1.7 trillion in cumulative impact on GDP, and $369 billion in federal, state, and local taxes. But looking at Ohio, 24,000 direct jobs in the industry in 2024, 106,000 total employment contributions, a 19% year-over-year increase from 2023, $8.2 billion contribution to labor income, a 24% increase from 2023. 14 contributed to Ohio GDP a 22 increase from 2023 In 2024 the industry contributed billion in taxes to state and local governments million to Ohio state government million to Ohio local governments and that is net of incentives Data centers, as I mentioned, are steel-in-the-ground infrastructure. You can see on this graph here, approximately $40 billion steel-in-the-ground has been put in as of July 2025. This is the first time the U.S. Census Bureau is tracking construction spending on data centers alone. But if you look at the overall construction trends, you'll see that from May 2023 to May 2024, data center construction experienced a 69% year-over-year increase. This far outpaces any other sector of construction. Manufacturing is the next closest at 20%, power at 11%, and so on and so forth. but traditional construction activities like commercial and office are actually in a negative. And on the following graph here from the New York Times, data centers and AI infrastructure spending has overtaken office construction in the United States, and that is poised to continue into the foreseeable future. As I mentioned, with that construction comes significant job creation and workforce development opportunities, not just on the skilled trades and construction side, but the ability to get a good paying job in a data center without having a traditional four-year degree. Data centers are heavily investing in workforce development programs at community colleges, technical colleges, and others. A several-month certification program can lead to a very high-paying job as a data center technician directly out of that program. These are unique opportunities that the data center industry is heavily investing in because we need the workforce of the future to operate these highly technical facilities. There has been a lot of conversations around data centers and water use. Data centers are amongst the most efficient water users within the economy. According to Bluefield Research, data centers used 39 billion gallons per year of water in 2025. This compares to the food and beverage industry, which used 533 billion gallons of water in 2025, the The semiconductor industry, which used 59 billion gallons of water per year in 2025, and then the 2,500 billion gallons of water per year that were lost to water leaks in municipal water systems on an annual basis in 2025. In the world's largest market, Virginia, 83% of data centers use as much, if not less, water than a large commercial office building, and data center water use is sustainable according to the Joint Legislative Audit and Research Committee in the state. In Texas, data centers are less than a half a percent of the state's overall water use in 2025. In Arizona, the top five market, data centers use less water than craft beer production and less water than golf courses. There is no one-size-fits-all method of cooling. Data center servers must be cooled. If they're not cooled, the temperature can rise in a server room as much as a degree a minute until ultimately those servers fail and with them the products and services that we rely on. There are several local factors that data centers consider in terms of the type of cooling technology that they use. Humidity, climate temperature, the availability of water including recycled, non-potable, and harvested rainwater sources. There are a variety of different cooling methods that data centers use. Air-based cooling, closed-loop cooling, evaporative cooling, free cooling in which outside ambient air is used during the coldest months to cool. A form of liquid cooling like immersion or direct chip, or a hybrid cooling strategy that uses some combination of these techniques. We have data centers that will generally use an air cooling system but be able to switch to a water cooling system during the hottest day of the year to ensure that they can reduce their peak load on the grid There are inherent trade between energy and water and cooling If you use more water for cooling, you will use less electricity. If you use more electricity for cooling, you will use less water. So to use the most efficient cooling method often depends on the evaluation of those factors and ensuring that you are using the most efficient technology available but the data clearly shows the data centers are amongst the most efficient water users in the economy data centers are a key driver in electricity load growth right now but I think there are various there are varying opinions on what data center load growth means for our grid you can see a variety of headlines here from throughout the country on how data centers are impacting utilities and grids. And what we see here is that data centers have the tremendous potential to put downward pressure on rates. We see just west here in Indiana Michigan Power, home to one of the first large load tariffs in electricity tariffs in the country, a tariff that the data center coalition was a party to. I&M just announced that they are lowering rates because of data center revenue coming in. Georgia Power, in one of the fastest growing markets in the country, announced a three-year rate freeze, lowering rates after that three-year freeze because of data center revenue coming in. DTE Energy in Michigan will not pursue a rate increase for the foreseeable future because of data center projects coming in to their territory. Data centers have an opportunity to spread to help be the large energy user to spread those fixed costs out and to put downward pressure on rates for others. A recent study from E3 helped to illustrate the current electricity, the current drivers of rising electricity rates and the role of data centers. This study found five key, had five key findings. One is that there is an unclear relationship between growth and rate hikes. There's no direct correlation between data center load growth and rising electricity rates. Data centers can lower customer rates because paying their full cost of infrastructure allows them to bear more of that burden on the necessary infrastructure upgrades for the system, and it generated net positive revenue for those utilities. The industry is paying its own way. The review found no evidence of a historic cost shift from data centers to other customers. The data center industry is fully committed to paying our full cost of service for electricity. We are an end user and a customer. We depend on utilities, regulators, and grid operators to ensure that when we get our bill it is our full cost of service, that there are no costs passed along to other ratepayers. But data centers have not rested on their laurels. They have been a party to large load tariff settlements like the I&M case and others throughout the country. We have also pursued innovative partnerships to help alleviate some of those short-term constraints on the grid. Co-location opportunities in which data centers can utilize existing or dormant resources on the grid. I think of META and the announcements with the two Ohio nuclear plants, the ability to upright their capacity, invest in that facility, and be a dedicated off-taker for those. They're also pursuing behind-the-meter solutions, including natural gas opportunities, something that you all helped to facilitate with House Bill 15 last year. Data centers are also increasingly building their own generation This is not a monolithic industry Not every data center company will have the ability to bring its own generation but those that are pursuing those opportunities And they doing this and being proactive to be a good partner on the grid and to help alleviate those short constraints The primary drivers of rising rates are also design choices, fuel cost volatility, generation retirements, inflation, increasing labor costs, grid modernization, wildfire mitigation, and other policy compliance. There are a variety of factors that are going in to increasing electricity costs. But it is fair to say within PJM, the data centers are approximately 50% of the capacity cost increases, while those other factors were driving that remaining 50%. And so on the last slide here, you will see that within PJM as well, tightening supply conditions have very much affected the capacity markets. Data centers, again, are investing in those opportunities, those innovative partnerships to help be a solution to that, and working with PJM, working with the governors of PJM states to advance solutions that will allow data centers to invest in the grid and to be that good partner within PJM to alleviate these supply constraints. I'm grateful for your time today. I hope that this information that I have presented is helpful within the context of these broader conversations. We are, most importantly, I think, an information resource. We want to be an information resource for you all. We want to be an information resource for your communities, for other stakeholders throughout the state of Ohio. But the data center industry is committed to investing in the state of Ohio for the long term, bringing jobs, bringing capital investment, bringing economic growth and economic opportunities, all while playing a crucial role in ensuring our national security and global economic competitiveness. Thank you. I'm happy to answer questions.

Thank you, Mr. Chairman. Thank you for being here today. here today. So you had mentioned that we are behind the demand curve. Can you tell us how far behind we are today and how long it would take to catch up to where we should be?

Representative Glassburn.

Co-chairs, representative, thank you for the question. Ohio has been a developing market for several years. But I think as we continue to see developments announced, we are likely still at the early stages. And I think from a national perspective, we are likely still at the early stages of data center build out as a whole. and so I think there is still significant development to go as we see continuous projects seek to develop in this

Co-chairs, Representative, I believe that many of those projects were in place or in development before the passage of House Bill 15, but what House Bill 15 has done has opened up opportunities opportunities for additional types of projects and the ability to bring generation into those projects. And so I do think it's more likely than not that most were in development before that, but now many of the projects moving forward are looking to utilize the new opportunities opened up by House Bill 15.

Go, Chair's Representative. I think the key thing right now is within the PJM market, working to open up new opportunities within the PJM market to alleviate many of those constraints. I think the best thing that Ohio can do now is provide certainty and predictability. Those new opportunities opened up by House Bill 15 are an excellent start. I think the next move there would be to move into PJM and help facilitate reforms to get new supply within the PJM market to alleviate those capacity constraints and the additional capacity costs from the recent auctions. And I think that certainty and predictability provide the best opportunity for sound development in the state. I think one thing that the industry has very much taken notice of and is engaging in heavily is stakeholder engagement. Ensuring that communities understand the projects, know what they are and aren't, and how these projects will be a good partner in those communities. And I think that is a key part of providing certainty, predictability, and stability moving forward.

I'll give you a breather after this one.

Okay.

Thank you.

Co-chairs, representative, the data center industry is among the top corporate clean energy procurers. We were responsible for half of all corporate clean energy procurement in 2024. from January through July of 2025 we were responsible for 84% of corporate clean energy procurement according to S&P Global. Data centers are helping to drive clean energy transition and deploying their own private capital to do so and so I think the data center industry is heavily investing in all forms of generation including clean and renewable generation to help facilitate supply of all types of generation onto the system Senator Reinecke.

Can we use old warehouses? Can we use office buildings that are empty? Co-Chair, Senator, thank you for the question. The data center industry is pursuing, I think, any and all opportunities. It doesn't have to just be greenfield sites. You have data centers utilizing existing industrial sites that provide significant ready-made infrastructure into that site. You have data centers repurposing things like the old Sears headquarters in Chicago. And so data centers are utilizing any and all opportunities available to them. When it comes to things like office space, it's going to depend on ultimately the structure of that building. The servers can get quite heavy, so you're not going to be able to go up as high as you might be in a typical downtown office structure. However, there are many data centers already in those buildings in places throughout the country. Many of those office buildings were fiber hubs or telecom hubs within downtowns. I can think of one in downtown Denver where that building itself was a hub for all that different connectivity, provided a really great opportunity to put a data center in there, and one of our members has. So while it's just not quite as large as many of the others that are being built today and being able to utilize those economies of scale for efficiencies. So it's any and all opportunities, and I think brownfields and repurposing of existing sites or existing buildings play a key part in that.

Thank you, co-chair. To my knowledge, these are all personal and private transactions. I'm not aware of a taking of land for a data center.

Yes, thank you, Co-Chair. And thank you. That's outstanding, brief, and for everyone. Again, that's online, so everyone can see this eventually. There's a lot of good information. But back to, I was hoping you could expand on something that you mentioned to Rep. Glassburn. One of the ideas that we're looking at at Ohio, when I think of data centers, is how they're distributed across. You really don't want to concentrate them for natural disasters or infrastructure protection. I mean, an idea we're trying to hold is that when a data center shows up in a rural area or an Appalachian area, the infrastructure that's required to support it, you bring a lot of things there, not just electricity, but maybe broadband, water, roads, all the basic connectivity to the rest of the state. Could you expand on that? Am I thinking about that the right way?

Senator Blackshear. Thank you, Chair, and thank you so much for your presentation. I like how you laid everything out. I just have a couple of questions for you. My first question is, are subsidies necessary to bring data centers to Ohio?

Thank you so much, Chair. Since being appointed to this committee, I've heard from a number of people from around the state, and some feel that the orange isn't worth the squeeze. They feel like data centers take so much from the communities, yet they don't put much back in. I guess this is actually a two-part question. My question is, when data centers are built and put in communities, is this something that's long-term? What you see is what you get in terms of are they here to stay? Is this a temporary entity in these communities? But also, too, what community investments are data centers providing to host communities?

Follow-up. Thank you so much, Chair. I was looking at your presentation, and you were talking about the water usage, and currently 39 billion gallons per year, and then you compared it to other industries. When you think about the food and beverage industry, that water is reusable. Semiconductor industry, that water is reusable as well. From what I've heard, when it comes to data centers, that water isn't reusable. but in your report you said that the report also found data center water usage in the state to be sustainable. Of course, with the increase of data centers, we're currently fifth in the nation. We're a little bit over 200 with probably more to come. Long term, obviously the water usage will increase. Do you truly feel like in the next 10, 15 years this will truly be sustainable?

Senator Wilkin.

Coach, Senator, thank you. It would be that tenant, so that specific tenant. itself within that building will own the computer equipment. I think that's one thing actually that is an important distinction for those multi-tenant, those co-location data centers. Oftentimes within the job count within that specific company it could look you know perhaps a little bit low but what that job count does not factor into is to those tenants because those tenants are going to have their own employees that are badging in and out of that building working on that equipment day in day out. So it is the tenants that own it and it's important partnership and investment in the data center itself when it comes to the sales tax exemption program as well so it's important to remember that that exemption applies to those local technology companies those local health care facilities local banks who have their own proprietary servers thank you follow up follow up so in cases like that the the data center is simply the building providing water fiber electric co-chairs senator yes

for the question you know I think it's it's especially on on energy infrastructure I think when we look across the economy and we look at infrastructure providers on energy on water you know we have significant infrastructure needs and so we need those providers to be investing in the infrastructure. With that, they need drivers on their system that will pay for it. And I think data centers play a crucial role in being that driver and paying for that infrastructure. We know that the industry has the ability to put downward pressure on rates by sharing more of that building for upgrading of the building and the upgrading of that infrastructure. And I think that's the crucial relationship here that provides tremendous opportunity to facilitate a a 21st century grid or water system, all while ensuring that the industry that is causing the cost is bearing the cost.

Go Chair, Senator. Yes. Data centers are fully committed to paying our full costs. So all the costs attributed to us, data centers are committed to paying.

Co-chair, Senator, thank you. The crucial thing is that that 50% of that cost should be directly assigned to the data center industry. None of that 50% of cost should be passed along to other rate payers, and that's the key is ensuring that. As I mentioned, we're an end user. We're a customer. When we get our bill we expect that bill to be our cost in full that nobody else is paying for that So that 50 of cost needs to be allocated to the data center industry Can I just follow up on that one Follow up

Co-chair Senator. Thank you. An effective large load tariff will ensure that costs are allocated appropriately, that risk is allocated appropriately, and that there is certainty on both ends for development timelines. And I say large load tariff because just as data centers are a source of load growth today. There can be other industries that are a source of load growth tomorrow. Long-term durable policies are end-use agnostic and they are committed to protecting ratepayers from large loads and ensuring costs are allocated appropriately today and into the future. So I think a well-structured large load tariff as we have seen effectively introduced in other states like Indiana provide that certainty.

Thank you, Chairman. I appreciate your presentation in regard to a macro view of the industry. The United States, through its citizenry over the last hundred and some years, has went through a number of what I'll call industrial revolutions. Significant changes to American life and city life and so forth. More recently, we had cellular towers that were a big thing 30 years ago and maybe 10 years ago. It was a significant public debate. So using that as an example, we are here today primarily because there is concern among the citizenry of things that they may understand or not understand. You addressed your presentation from the industry and giving what the facts and figures are for your understanding. But what I don't think you did address is if you were taking your hand and laying it on the issue, that is the citizenry's concern, how would you address that? What would you say to the American citizen, the citizen of Ohio that says you have concerns, you're going to be here later next week or so forth? What do we need to understand about that from the industry's perspective?

follow up

thank you Mr. Chair so I want to come back to Senator Reineke's original point about locating and siting in pre-existing areas that have been pre-developed we don't in my district, a large brownfield is becoming a brown stadium, but we can't do that over and over again in Ohio. It was a large Ford plant before that. We have dozens of abandoned automotive, steel, large industrial facilities that would, in theory, have the large infrastructure in place. you've talked about the time to market being kind of the main driver consideration. What is the barrier or is there a barrier to data centers going into those types of sites versus going to those open greenfield areas that Chairman Holmes talked about?

Follow-up? So following on that specifically, in House Bill 15 for energy generation, we created a quicker development timeframe for sites for energy generation that would go into brownfields and other sites, priority investment areas. If we were to try to do something along that line for data centers, is there a model in another state where other state has done that, or is there some kind of pathway where we might be able to change that calculus of where, again, an abandoned steel plant in my district or in Appalachia is more attractive than a greenfield

Thank you, Chair. Just one quick question. Is it true that data centers are having local officials sign NDAs, and if that is so, why?

So I'll follow up on that. Is there a timeline when that NDA can be waived so that the public can get some information before there's any kind of votes or anything like that at a local forum?

Thank you, Mr. Chairman. I'd like to follow up on Representative Claggett's concern and issue. I really like this presentation, but I have the same issue with my constituents, and I don't see this as a legislative issue as much as a constituent issue. So how is the industry, how do you explain what is the cloud? What is the cloud? What do you mean national What about the water usage thing? How can it be less than the water containment companies that are using that? So I guess and what do you mean 21 connected devices? So I think those are the issues that the public is not understanding that I'm this not really a question, but hoping that your industry can help us educate people so we understand what's going on with this issue.

Representative Workman. Thank you, co-chair. So I do hear frequently, and I think that there is a significant amount of fear related to what data centers bring in our future. And I frequently hear concerns about these being centers of mass surveillance, maybe not immediately, but into the future. Can you speak to that a bit?

Follow up? Thank you, Chair. So on that note, can you talk to me a little bit, and I'm not sure that it has been discussed yet, what the standard cybersecurity measures are for the transfer and storage of personal data? Chair, Representative, thank you for the question.

Please send that information to Chairman Holmes' office, and he'll disseminate it to the committee.

Senator Wilkin. Thank you again, Chairman. Just a quick question. You went through your data centers, the water use, and the cooling systems. Since we're specifically dealing with Ohio here, what is the most likely cooling system that's going to be used? I mean, I hear lots about the closed loop, but I'm assuming some of these are not going to work for one of the few states that has four seasons a day.

Well, seeing no further questions, I think you've weathered the storm. I'll ask one final one. Data centers, you mentioned credit card transactions. It made me think data centers are not new by any means. They are evolving. They are the topic du jour too. But it's been a long time since we've had to put our credit card and swipe it in that little machine. How long have data centers really been around?

Thank you for that. Seeing no further questions, we thank you for your testimony. Thank you very much. Next, we have testimony from PJ Interconnections, Executive Vice President of Government and Member Services, Awesome Huck.

You may begin when ready.

But with the supply-demand dynamics that we are seeing today, we are hopeful for a better result than that result. I'm not going to go through. The next slide discusses. So we've talked about who is PJM, talked a little bit about the current status of supply and demand. We'll now move into what are we doing to meet this moment in history of this intense growth. We'll talk about just a few things. our board in January made a decision to take on a number of different initiatives related to this massive uptick in demand and this major data center growth. I'm just going to sort of verbalize all of them, and we won't get through, we won't talk through the next five or six slides, but we'll just hit these, the highlights of each of these. So first, significant load forecasting improvements. So there's sort of the two curves that end up deciding what reliability and price look like. They're supply and demand. And we need to try and get both of those as sort of correct as possible when we run any marketplace to try and produce the most accurate price signal for our entire marketplace But we have embarked on trying to really sharpen our pen on the demand that finds its way into that demand curve. So we are going to require that any of these data centers, we call them large load additions, show financial commitments. We are going to tackle this issue of duplicative requests that may come in through one transmission owner who then feeds that into our demand curve and then also submitted to another transmission owner who then may feed that into our demand curve. We need to tackle sort of those duplicative requests. We're going to provide an opportunity for state commissions to review these submitted large load additions or data centers in their given state. No one knows their states like those who are physically based and local. And so we want to provide the commission with that ability to say, hey, we know that this was what was submitted to you, but we're not really seeing this one or these two progress to the level that has been or at least reported to you such that you should include it in your demand curve. The next item I'll talk about is, look, our desired framework, and I would just say the industry's desired framework, is for these data centers to bring their own generation. I think that this is an opportunity for them. It's an opportunity for the entire industry. But ultimately, we're going to need a lot more supply in order to service this demand. The data centers generally are the growth engine for that demand. And so what we have said and what our board has said is that the ideal scenario is if you bring your own generation. If you bring your own generation, we'll pair that with an opportunity for you to get through our generation interconnection queue processes quicker through a process we created called the expedited interconnection track. Here's the challenge, though. The challenge is that it takes a year or two to construct a data center, and it takes, to be generous, four to seven years to construct the corresponding supply. And so we are going to find ourselves in this period, and we are hopeful that we get a lot more supply, and we'll do everything we can to facilitate getting a lot more supply. We're going to find ourselves in this period where we've got a reserve margin. Remember that 185 and 150-some numbers? We've got a reserve margin, but that's going to continue to get chewed up. Now, as that reserve margin continues to get chewed up, when we are operating the power grid, trying to maintain reliability for 67 million consumers, there may be periods of strain on the grid where we are going to have to ask the data centers to effectively get off the grid and move to their backups so as to not have to shed likely residential load residential consumers So that is something that the board our board has adopted It is being discussed actually today in a stakeholder meeting at PJM Interconnection. This concept is called Connect and Manage. So that's going to be part of our framework on a going-forward basis. We need it for operational reliability, and we need it to protect consumers at the end of the day. Last thing I want to talk about is the work that we're doing on markets. Remember, PJM, its primary purpose is to preserve reliability. It does that through three core functions. Planning the system like an urban planner would. Operations, which is what happens in the control room. and then having competitive wholesale markets that spur investment. These are especially relevant for restructured states like Ohio, so states that have characterized as deregulated in the generation space. Now, last year the organization had quite the target on its back after a very high capacity market clearing price found its way to consumer bills. And understandably so. It was quite the jump from $28 a megawatt day to $270 a megawatt day. After sort of a litigation proceeding, the organization entered into a settlement to cap the market for two years and then extended that cap for an additional two years. So the marketplace that created, at least initially, the electric price jump for consumers has now been capped until mid-2030. So that's one thing that you should be aware of. We have taken measures to try and sort of stem what would have been, because actually the true cap of that market, just based on how the market has been approved by a regulator of the Federal Energy Regulatory Commission, would have been a couple hundred dollars per megawatt day higher than the cap that we ultimately agreed to. You're going to ask the question, okay, well, if you cap the market, how are you going to get new supply? And it's a very, very fair question. I want to talk about just two more things that PJM is doing in this era because we have to adapt, right? If this is a sort of generational shift in combination of demand and the overarching industry, we are also going to have to adapt, okay? So the market's capped. How are we going to obtain new supply? Well, the first thing is that our board agreed to run a one-time procurement to try and bring in as much new supply in one swath as possible. That procurement is going to take sort of, at least as proposed, has sort of two different components to it. One is a centralized procurement, kind of the same way that PJM generally procures power. and then the other is trying to facilitate, basically be a matchmaking service between data centers and generation developers that want to service those data centers. That's called the reliability backstop. We going to hopefully procure a significant chunk of power through the combination of centralized procurement and the bilateral framework The other thing that we are doing is the capacity market as it exists today it has to exist with our other markets, which are really the energy markets and ancillary services, to incentivize new supply. Those markets have to work together to incentivize new supply. The capacity market in particular, that marketplace that created the sort of big jump for consumers in their bills, we have questions around whether or not how that marketplace is constructed going forward, how that marketplace going as constructed, if that marketplace, excuse me, will in fact incentivize new supply for a couple reasons. The first reason is for a reason, frankly, that we've already talked about, which is we published a paper on this. It's called Powering Reliability Through Markets. I should have and I will push it to the committee members, both the House and Senate Energy Committees, because I think you'll find it to be a really interesting read, as well as all of you. because one of the issues outlined in that paper is around, hey, if prices are going to go up and we're going to experience the kind of, we'll characterize it as governmental intervention that we've experienced here, and it's every state, and it's the White House, frankly, that asked us to cap this market. We are going to have to think of a new framework if that is going to be the case, number one. And then number two, what we are seeing is the capacity market is designed as a one-year spot price that is three years forward looking. The deals being done between data center and developer are far different than that. These are long-term agreements, 10-, 15-year agreements, that are far different and far, frankly, less volatile than what the capacity market has historically produced. Capacity market's done great. It did great during the shale revolution. It's done its job for a particular period. But what we've done by putting this cap on the marketplace is we've bought ourselves some time to potentially reimagine what that marketplace should look like in conjunction with the other markets. So that work is starting up at PJM Interconnection as well. If we're experiencing a generational shift, we also have to adapt. We're now going to skip to the end, and the end is really about how electricity costs get generated effectively and then ultimately allocated to consumers. You heard my colleague from the data center coalition say, hey, we have agreed to pay. for necessary infrastructure, for market price increases. That commitment is solidified by many of the large hyperscalers in this ratepayer protection pledge, which is in slide 19 of my deck, the sort of primary components of the ratepayer protection pledge that they executed with the White House on March the 4th of this year. So there is a commitment to, in fact, pay. Okay. So if there's a commitment to pay, which is, which is excellent, the, the, The framework for how electricity costs get created and then ultimately allocated to individual consumer classes is something that I think we should all understand. Okay? So your typical bill, you know, all of your constituents' typical bills in Ohio will be broken down into four components. The generation of the watt, the transmission of the watt. Then that WOT gets to a distribution substation where then jurisdictionally it changes the framework, so the distribution of the WOT, and then what are called riders, which are single issue sort of reasons to allow for a utility to recover rates. Those four components, the way I've characterized this, and this is my characterization, is that the regulatory authority for cost creation for generation and transmission, which are generally wholesale costs, that is overseen by the Federal Energy Regulatory Commission. And then through our processes, we effectuate what the Federal Energy Regulatory Commission has told us to do. on distribution and riders, that is, in this instance, in this state, it's Public Utilities Commission of Ohio. Now, cost allocation to specific customer classes, that is done on the retail side. And there is no error between myself and Chair French and the rest of the commissioners. That is a very fine agency. They understand the task at hand, and we'll talk about some other measures that can be taken to deal with affordability, and they actually are the gold standard on one of those very important measures. Governor DeWine signed the Statement of Principles regarding PJM, specifically stating that he will work to allocate costs to the data centers themselves. So purely talking about electricity here, not talking about water, nothing else. Just purely talking about electricity here. So just there is, this gets complicated. And I will tell you that in the cost allocation of specific customer classes, we have to work with the PUCO because allocation of capacity costs to a specific customer class is not easy. We're doing the work ourselves internally on this, and I know that they're looking at it as well. But that's how the framework works, just for your edification on a going-forward basis. Last few slides, I published an op-ed in a periodical called Utility Dive. I pray that you are not reading Utility Dive by your bedside, but in case you are, publishing some recommendations on how to deal with affordability on the retail side. We have capped the marketplace that created the wholesale pricing challenge for consumers. But there's a lot of work that can be done on the retail side. This second item on slide 21, data center entry commitments, PCO and AEP's tariffs, gold standard. Many other states are trying to emulate that tariff. That deals with effectively the cost of distribution infrastructure as data centers are finding their way onto the system, and it deals with the stranded cost-related issues. So if the data center says, I'm out, so your consumers don't have to pick up the tab for the data center dropping out, that's an excellent piece of work that the commission has done. What we talked about in terms of cost allocation that the first item here retail cost allocation So allocating the costs of generation and transmission and distribution riders et cetera to whether it's a new customer class that is data centers or however the commission decides it's the best to handle this. I know that they're thinking about that right now and working on that right now. Again, some of that is easier. Some of those frameworks are easier than others. Again, I specifically again mentioned capacity costs. Default service procurement, this is a very wonky item. But just know that in Ohio, retail customers, they can shop. They can shop for their power or in the four regulated distribution utility areas, they will just take their power through default service procurements that are conducted by the commission. Many, not really saying this about the PUCOs and the four utilities who are currently working with PUCO to procure default service, but in many of our states, what we are finding in many of our states what we are finding is those default service procurement frameworks are really exposed to a one-year capacity market hike. And so this idea of hedging those procurements longer term is something that we are actively working on with states, and we have actually drafted a white paper associated with this and are shopping that as well to say, like, hey, you can design these differently so that the exposure to the capacity market price, when it filters all the way down to your bill, is greater, is better hedged than the procurements currently are. That is generally it. I'm open for questions. Again, I want to applaud the state of Ohio. I want to applaud all of you for holding these hearings, for listening to your constituents. This is like I continue to reference. This is unlike anything that we've seen. This is not your everyday affair in the electricity business. And so, again, I want to thank all of you for the opportunity to be here and applaud you for holding these hearings. Thank you, sir.

Thank you. Appreciate that. Again, broad view. I want to ask kind of a short-term question and perhaps a little bit longer term. I want to understand the transmission capacity within Ohio. And so by way of analogy, if you take this room here as an illustration of the state, you have maybe 12-gauge wire running around here. The total capacity of perhaps this wired room is maybe 100 amp, but we're only maybe using 20 or 30 amp to light the room and so forth. In the same way, what is the used capacity of the main transmission lines throughout Ohio as far as their average capacity usage? So if you have a 760 KV line cluster throughout the day, what is that being used at capacity-wise?

Well, please. Thank you, chairman. Well, okay, I was hoping for a little better than that because, you know, I was wondering, are we running at 50% of capacity?

We already had a little bit of that last year, and we could have that again. So depending on the capacity of our main transmission and then the offloading of certain high-end users, so for instance, some of the digital Bitcoin and those type of things have these agreements to come offline. And so you mentioned that. I'm trying to figure out in the short run here, what is our risk assessment for going brown or out in the short run based upon these conditions that we have before us?

Can I follow up on that? Please. So asked a little bit differently, what's your day-to-day concern about our grid reliability here in Ohio, separate from coincident issues where it's super cold day and we have a super big storm coming through, what's your day-to-day concern about our...

So, again, getting back to the pecking order in regard to alleviating certain high-load capacities if we have a problem. I'm just trying to get a base understanding here from your perspective of what it's going to look like if we have to ask certain operators to drop off the grid. Yes. So what is that pecking order as PGM understands it in regard to even industries or specific companies that helps us maintain a level of power to you and on that Our co to the representative our grid operators they have a we just characterize it

Thank you, Mr. Chair. So towards the end of your comments, you actually hit on one of the things I wanted to ask about, which is we have a pretty well-developed bill in the House from Representative Kloppenstein for doing residential demand response. In your comments, you used a different term, but it sounded essentially to me that you wanted data centers to also do peak demand response. Is that a correct characterization of what you're – Am I using the terminology wrong?

Follow-up? So is the bottom line that –

Okay. Well, thank you. So I wanted to come back to the transmission questions that were being asked. So when we passed House Bill 15, we created standards for how much these data centers need to generate their own power or get purchase agreements. But most of these appear to be also still not being completely off-grid. They're part of that. They may be responsible for 85%, but they're still part of the grid. How is this working out where you talked about them going back on their batteries, but when a data center is signed up, how are we ensuring that we're not having to do 100% build out of the grid for them that the rest of the ratepayers are paying? You know, if they're most of the time running on their own power, is everyone else still paying for the transmission to exist for them to receive 100% from the grid?

Thank you, Mr. Chair. One other topic I wanted to hit on is on your 10th slide, you have a map of Ohio and talking about agreements and studies. And for project agreements, more than 90% of them are solar storage, wind, or hybrid. I want to make sure I understand what this is telling me. Is this slide telling me that we have 10 gigawatts of, or 8 gigawatts, I guess, of solar and wind that are ready to come online in Ohio, but they're just not there yet? Or are these projects that are being blocked in some way? I mean, we're reading a lot of headlines about renewables being blocked by local restrictions. What is this slide telling me?

Thank you, Mr. Chair.

Thank you, Chairman. Thanks for being here. Good to see you. Question. The new generation that we talk about that we need, I only know of one group that claims we don't need it, but have we talked any more? I know one time there was a discussion of doing separate queues, a renewable queue and a baseload generation queue, and I know at one point, I'm probably getting this wrong, but we had the quick 50 projects that got bumped up.

Thank you.

Thanks for those answers. I been trying to keep track of what I want to ask here When you talk about connect and manage when it pertains to data centers and I want to use the example they had. There's one. It was the buy, bring, or build your generation. Yes. So if a data center has said, well, we're just going to buy, we're going to pay our share, and we're going to buy off of the grid, and they're doing 31,000 credit card transactions a second, which quick math says that's over $111 million per hour. I mean, for those of us that are small business owners, I mean, when your credit card swipe doesn't work, that's a problem. Yes. So how do you address that? Is that – there would almost, I would think, have to be a ranking order within the data center to say if it's a service for an everyday business, how do we curtail their electric usage if they say – wait. I mean if I'm them and I'm trying to play devil's advocate here. If I'm them, I'm saying wait a minute. I played by the rules. I'm buying. So how am I getting curtailed and I don't have any backup?

Sorry, I'm just curious as to how, if I was buying, how do I get penalized? Thank you.

Representative Workman. Thank you, Mr. Chairman. Thank you for being here. Much appreciated. So am I correct when I hear you say that it is our responsibility and authority to maintain or regulate how the allocations are happening for retail costs? Is that correct?

Follow-up? Follow-up. So if Ohio's goal is ensuring existing rate payers do not subsidize the new demand, what policies do you recommend that we prioritize to ensure that doesn't happen?

Senator Reineke.

Co-chairs to the senator, without sort of being overly simplistic about it, I think trying to figure out all the costs that are sort of caused by data center growth and then figuring out how to, from a regulatory perspective or statutory perspective, asking the regulator to do the work. I don't want to get in front of Chair French. Chair French, you tell the senator exactly how you want to get this done. But I think that is going to be very key here, is determining the costs that have been caused and then allocating those costs to ultimately a data center customer class. I think that being able to explain that to a constituent, say, Mr. and Mrs., we understand the concern. This data center demand just came on really truly in 2024. It's a sort of generational challenge, but we have created a committee. We are analyzing this issue, and we are going to take action to try and allocate costs specifically to the cost causers.

Yes.

Senator Wilkin.

Chair to the Senator, I do not handy, but I will get that over to you. I do think that there There is definitely a difference in completion rate and success, and I do think that I'll give you a general answer, which is that your standard fare, for instance, natural gas units that find their way into the queue, have a higher completion, have a higher success rate than renewable projects do.

Chairman Holmes.

Chair Holmes, short answer is yes. Short answer is yes. It would be created to try and funnel the combination of, again, if you go and you look at my slide that lists out the wholesale costs and the retail costs, whatever costs that can be identified that are caused by the data centers to specifically those data centers. It's not that easy. Okay, it'll be – what I mean by that is that I think I don't want to, in your minds, oversimplify the framework. We're going to have to do a lot of work with PUCO in order to get this right. But the short answer is yes. That's exactly sort of the – if the ratepayer protection pledge in tandem with the statement of principles regarding PJM, all of which culminate in this idea that the data centers are going to pay for the electricity and other costs, but the electricity costs that they are causing. We then have to figure out the right statutory regulatory framework in order to make that happen. That is sort of in its infancy across our jurisdictions. But like I said, no air gap between us and the PUC-O. They are aware of it. They are working on it. And that is exactly what that is meant to do, is to allocate the cost caused by the data centers to the data centers themselves.

I apologize for that already.

All right. Good afternoon. Joint Chairs Holmes and Chavez, Ranking Members Glassburn and Blackshear, and members of the Joint Data Center Committee, thank you for the opportunity to speak about data center development as it relates to the Public Utilities Commission of Ohio, as well as the Ohio Power Siting Board. As everyone in this room knows, the reason we're here today is in Ohio and around the country, we have seen significant data center development. So while these facilities are becoming more and more important in our daily lives, they also have, of course, driven electricity demand at a pace that we have never seen before in PJM. In addition to the new generation that's already been discussed, these projects often need additional transmission infrastructure to serve them as well. So what do we do to make sure that these important but energy-intensive businesses are paying their own way so that all of our other rate payers are protected? Well, Ohio is taking a really comprehensive approach to this to make sure that that's exactly what happens. First of all, the data center tariff. I think it's been touched on already today, but I'm going to go into a little bit more detail. So in 2024, AP Ohio reported that it had received requests to connect more than 30,000 megawatts of data center load. This would lead, of course, to significant infrastructure investment and significant associated costs for that infrastructure within this region. It was really difficult for AP Ohio to determine how much of that 30,000 megawatts was real and how much was speculative. And they certainly did not want to build out a bunch of infrastructure for these loads if they weren't real. So in May of 2024, AEP Ohio proposed a tariff specific to data centers that requested long-term commitments that would protect Ohio's other rate payers from the cost of the infrastructure needed to serve them And in July 2025 the PUC approved the new tariff To our knowledge it was the first data center tariff in the country There are numerous other states, as you heard Austin say, now exploring this concept and implementing similar plans to protect their other citizens from increased costs as well. At a high level, what the tariff does is it requires the new data centers to provide collateral, pay a minimum demand charge, sign long-term contracts, generally between 8 and 12 years, and pay exit fees if they leave their electric service early. This is all in an effort to avoid the stranded costs. So in February, AAP Ohio publicly announced that it had signed about 5,600 megawatts of data center load under the new tariff. So 5,600 is a lot. It's a lot of load, but it's not 30,000. So this tariff unquestionably helped AAP Ohio to refine the amount of load in its data center queue, thereby saving other rate payers millions of dollars in unnecessary infrastructure as they will only build out what is necessary from the agreement side under their new tariff, and the majority of those costs will be borne by the data centers under this tariff. So in addition to the tariff saving rate payers by allowing AAP Ohio to refine the load in its queue, there are provisions that ensure that exit fees and collateral collected for load that does not materialize flows back to benefit the other ratepayers and not the utility. We are also applying this approach to other utilities in Ohio. So we have directed First Energy to propose a data center tariff with the PUCO, and the PUCO staff has recommended that the creation of a data center tariff in AES Ohio's recently filed rate case. So again, all of these changes are in an effort to ensure that we do not unduly burden data center development, but also protect other ratepayers from the costs associated with their development and operation. With electric security plans sunsetting because of House Bill 15, we are also re-examining the rate designs for transmission costs. The way cost allocation works, and I think Awesome touched on this, is that the transmission owners build the transmission that PJM and the transmission owners say is needed, and then PJM builds the utilities for their share of the costs. The utilities then come to us at the PUCO for cost recovery. We have a number of different customer classes, and we allocate those costs across the different classes. data centers are currently in the industrial or commercial customer class they are not currently in their own class so we have communicated to the ohio utilities and stakeholders that we will be evaluating how costs are allocated across the different customer classes with the main goal again of ensuring that existing and future data centers are being properly allocated their share of FERC-approved transmission costs. So House Bill 15, it is setting the pathway to new generation in Ohio without question. So as we have seen increased energy demand, we know that Ohio needs to build more generation. It's not just Ohio. The entire country needs to build more generation, but we certainly need it here too. So one of the most important changes that the legislature made was to update timeframes for the power siting board to site generation projects. So the bill also created new types of cases, such as projects in priority investment areas and sites controlled by the developer, which requires expedited decision-making by the Power Siting Board. So historically Ohio has had a policy in place where behind the meter electric generation was required to be located on the same site as the end user And the generation that qualified was historically green energy which then included wind and solar and it could be sited by the power siding board within 90 days as long as there were existing structures on the location. But House Bill 15 expanded the use of behind the meter generation by expanding the types of generating sources that can be used to power businesses while also allowing additional flexibility when the generating source does not share a physical location or is not directly adjacent to the end user. So why is this important to why we're here today? Well, it's important because since the enactment of House Bill 15, Ohio has seen a significant number of data centers choose to provide their own behind-the-meter generation. So in 2025 alone, the Power Siting Board authorized the construction of or had applications submitted for over 2,000 megawatts of behind-the-meter dispatchable generation. And in addition to the behind-the-meter applications, we have been provided pre-application notice that another 2,755 megawatts of traditional gas-generating facilities are also looking to expand their existing facilities or cite new generation in the state. In addition, the Trump administration recently announced a $33 billion, 10-gigawatt natural gas power plant project in Pikedon, Ohio, in partnership with Japan that will power a large data center campus. Ohio is also poised to become a leader in nuclear technologies, with exciting announcements from Vistra and Oklo. In the last two months, the Power Siting Board received separate notices for another three proposed behind-the-meter natural gas-generating facilities, an 800-megawatt facility in Asheville, a 1,300-megawatt facility with approximately 1,000 megawatts of battery storage in Millersport, and a 430 megawatt facility in New Albany. These announcements are not just important because of the energy need for grid reliability. They are investments in Ohio's people and our economy. So with the legislature's leadership in passing House Bill 15 and with the governor's support, Ohio has positioned itself as a leader for these energy-intensive users to have the opportunity to begin operations quickly by providing their own firm supply of electricity, which, of course, helps to maintain a reliable grid for the rest of our ratepayers in the region. So the issues that impact the cost to deliver electricity flow well beyond the borders of Ohio. As such, Governor DeWine joined 12 governors in the PJM region and the White House's National Energy Dominance Council in signing a statement of principles urging PJM to take actions that will protect non-datasetter customers from increased capacity prices caused by data center demand. The principals outlined several provisions, including a one-time auction, which you heard Awesome talk about. This would offer long-term contracts to new generation to serve data centers. Data centers will be responsible for all the costs associated with the new generation. PJM recently announced that it is proposing to move forward with this auction this September. As PJM's auction proposal progresses through PJM and then to FERC, we will also, at the PUCO, advocate for measures that will ensure that data center customers bear all capacity costs for new generation procured through this auction. Additionally, Commissioner Dieters, who's here today, through his leadership in the organization of OPSI, which is the organization of PJM States, and our federal energy advocate, Sarah Perot, who's also here today, continue to engage in federal and regional efforts to make sure that data center customers are allocated the costs to build, operate, and maintain new energy infrastructure necessary to the state. serve them. Through collaborative efforts of the legislature, the governor, the PUCO, and the Office of the Federal Energy Advocate, Ohio is a leader in energy policy in general, but also specifically as it relates to data centers. We have positioned ourselves as an attractive place for businesses to locate, especially given the ability to bring your own generation so that operations can move quickly. Thank you to the members and leadership of the Ohio General Assembly for giving us the statutory direction and guidance that has driven so many of these important energy-related achievements for Ohio. We look forward to continuing this effort together. Thank you, and I'm happy to answer any questions.

payers don't pay for that in Ohio, is that correct? Co-chair, that is correct. Okay, and it's not fair to have all the data centers pay for that either, is it? I mean, what I'm trying to say is we're not going to try to lump all the costs of new generation onto the data centers. We're going to try to allocate it proportionally.

Correct.

Yeah, Mr. Co-Chair, thank you for the question. That's our job. And so what we're doing is, and again, I'm referring to Austin's testimony. He's correct. This is not an easy, necessarily easy task, but what we are doing, and it's one step at a time. So, you know, AAP came in with the tariff request. We approved that. And now we're saying, okay, everybody else needs to do that too, right? So, you know, let's move forward with that. Now we have this backstop auction coming. Okay, how are we going to make sure that that generation, that that capacity that is procured in that backstop auction is allocated to the data centers? We need to make sure we do that. And then we need to make sure the transmission costs, like I talked about earlier, that those are also allocated proportionately and justifiably to the users, the cost causers, as they are. And so if the data centers need this much transmission, they need to pay for it, and they're willing to pay for it. We've heard that. And so, yes, that's our job, is to work with PJM, to work with the utilities, to work within ourselves and make sure that happens.

Thank you, Mr. Chairman. Thank you so much for being here. So I do have a question about the customer classes. It sounds like this is already something that is well underway. and in development. The question I have is, at what point does the PUCO determine a customer type has become sufficiently unique that a separate class structure should be considered?

Senator Blackshear.

Co-chair, Senator, thank you for the question. I think there were a couple of reasons for it. First, to make sure that there weren't stranded costs. So a data center comes in and says, we need this much infrastructure built out to serve us, and then either they don't come here, They go somewhere else or they're not here for as long as they said they were going to be. And then what happens with all that cost for all that infrastructure? It would get assigned to the other rate payers, and that wouldn't be fair. And so we want to make sure that we don't have those stranded costs. And also, it was becoming very difficult to forecast, right? So, again, AEP had 30,000 megawatts that they thought were looking to sign up. And so to avoid that and to be able to kind of narrow that down to what was real, that helped. This tariff really obviously helped because it was 5,600 back in February is what they announced. So I'm sure that's grown since then, but that's what we knew then.

Co-chairs, Senator, thank you for the question. I think it's been very helpful. I mean, we've had, I just want to try to add it up in my head, and I'll probably get it wrong, but I want to say, I don't know, probably seven to ten projects that have asked to be cited or have been cited, most of which are to serve data centers. So what that does is, of course, it takes that demand from the grid because they're providing their own, and it's behind the meter. So it's been very helpful.

Thank you, Mr. Chair, and thank you for coming today. Let me start with a basic of our residential customers today being held harmless from the rapid expansion from data centers.

Maybe. But what are we doing?

Follow up. It sounded like earlier that the data center industry as a whole is willing to make that answer unequivocally, yes, we will hold you harmless. what do we need to do to get to that state?

Maybe asked a different way, what help do you need from the legislature to make sure that you're fully equipped to do that?

I'll follow up. Yes, Mr. Chair. So let's say for sake of argument, we get to that point, we have the framework, whether it's new classes or tariffs or whatnot, you have the authority, you've got all the tools you need. Is there ever going to be a looking back, a reconciling of how much customers have paid that maybe was caused by data centers or other large users in this marketplace? Is that going to be reconciled at any point that we would be able to credibly tell residential and smaller consumers, you may have paid for a moment there, but it's getting evened out?

Representative Workman. Thank you, Chair. Thank you again for all of your clarification here. I want to speak a little bit about the data center tariffs. This sounds like it's a first of its kind in the country, so fantastic work on that. I'm wondering specifically about the provision that they pay a minimum demand charge.

Sure. Co-chairs, Representative, thank you for the question. The contracts are 8 to 12 years, but they can be renewed. And also that time frame is put in there because of the costs that are attributable to them and how long it will take to recoup that Okay and then the minimum demand charge I just going to try in a real simplistic way So if they say they need 100 megawatts and they only use 60, they're still going to pay for 85. That's the minimum demand charge.

We want to make sure that at a minimum, this is what they're paying. to make sure that those costs don't get stranded. And I think maybe that goes back to Representative Glassburn's question, is reconciling. Well, with this, I think it's starting to come out. It hopefully will come out in the wash as they're paying forward with taker pays or tariffs that we're actually clawing back some of that over time. That's what I'm trying to agree.

Thank you, co-chair. Based upon AEP's results, I think it's absolutely working the way it was intended to work, which is why at this point we're saying, okay, it's time to look at the other utilities as well.

Thank you.

Thank you. Good afternoon, Chair Chavez, Chair Holmes, members of the Select Committee on Data Centers. Thank you for the opportunity to testify today. My name is Maureen Willis and I serve as the Director of the Ohio Consumers Council, the the state agency charged by the General Assembly with representing the interests of residential utility consumers. At the outset, I respectfully ask that my full written testimony be entered into the record, but today I will shorten my remarks. Today I will focus on my remarks on what OCC believes is a central question before this committee. That is, how can Ohio support data center growth while protecting Ohio households from unfair costs and reliability risks associated with the extraordinary new electric demand? Ohio can and should do both. We can welcome investment. We can support innovation. We can compete for economic growth. but we must also protect Ohio families that are already struggling with rising utility bills. And that balance matters. Because the issue before this committee is not whether Ohio should welcome data centers. The question is, how do we ensure growth occurs in a way that remains fair to Ohio consumers and sustainable for Ohio communities? Ohio does not face a choice between economic development and consumer protection. Ohio can thread the needle. Let me begin with something important. OCC supports economic development. Ohio should compete for investment associated with cloud computing, artificial intelligence, and digital infrastructure. Data centers represent a real opportunity for Ohio. They bring capital investment, construction jobs, economic activity, and the possibility of strengthening Ohio's position in a rapidly evolving digital economy. These opportunities are real. But because of the scale of these facilities, Ohio must approach these carefully and thoughtfully. Thoughtfully means recognizing both opportunity and responsibility. Data centers are not like traditional industrial customers. Historically, Ohio's industrial growth often meant large employers with substantial permanent workforces and relatively predictable energy demand. But modern data centers are different. But that doesn't make them bad. It just means that they simply raise different issues and different questions. These facilities can involve extraordinary electricity demand, sometimes comparable to the energy needs of entire communities. Some proposals involve hundreds or even thousands of megawatts of projected load. They may require new substations, transmission upgrades, expanded distribution systems, reliability planning, and in some cases, substantial water resources. At the same time, once construction is complete, permanent employment may be comparatively limited relative to the scale of the infrastructure required. Again, construction jobs matter, capital investment matters, economic activity matters, but policymakers should also ask what remains after construction is complete? What long-term public benefit exists? What infrastructure burdens remain? And what happens if projections change? And importantly, who bears that risk? These are prudent governance questions, and because of the unprecedented scale of the projected data center demand, Ohio has every reason to proceed thoughtfully before costs become embedded into residential utility bills. This conversation is happening at a time when Ohio families are facing real affordability concerns. Consumers are seeing increasing utility costs. Transmission costs have risen. Distribution costs have risen, and we see wholesale market pressures continuing. For many Ohioans, particularly seniors, families with children, and households living paycheck to paycheck, utility affordability is already a challenge. Against that backdrop, policymakers must proceed carefully before introducing additional risks into residential bills. And that is why one principle should remain. Central to our discussion, consumer protection matters. Ohio households, as we've heard today, should not bear the costs or risks associated with the extraordinary private demand growth from data centers. That principle is reasonable, and it's simple. Cost causation must remain the guiding principle. Ohio utility regulation has long recognized this as a basic principle. Cost should follow cost causation. Consumers should pay for the cost that they impose on the system. And that principle becomes especially important when discussing the demand from data centers. Data centers may require a lot. New substations, expanded transmission, distribution upgrades, reliability backstop investments, and potentially significant new generation sources. So to the extent that investments are built primarily to serve data centers the data centers should pay that cost Residential customers should not become involuntary financiers of speculative infrastructure tied to private investment That is sound rate-making, and frankly, it provides greater certainty for everyone involved. Importantly, Ohio is not beginning this conversation from scratch. The Public Utilities Commission, the legislatures, utilities, consumer advocates, and stakeholders have already begun addressing many of these issues, and OCC appreciates those efforts. The PUCO has increasingly recognized that extraordinary electricity demand may require different tools and different approaches. We've already seen meaningful steps from the PUCO. As Chair French testified, the commission approved a large load structure tariff involving AEP Ohio, recognizing that extraordinary demand may warrant service arrangements different from traditional customer classes. That was the large load tariff arrangement that OCC supported, along with others. The commission has also directed, as Chairman French noted, that the first energy Ohio utilities should pursue separate tariff approaches in response to extraordinary new load. And as she mentioned in the AES Ohio proceedings, the PUCO staff recognized the need for dedicated approaches that are responsive to these emerging data center growth and associated planning concerns. The POCO has also been grappling with whether data center load should require separate procurement approaches or different reliability treatments so that existing residential consumers are not unfairly exposed to new risks. Again, OCC appreciates that work. It reflects a growing understanding that traditional rate making may need to be adjusted so that customers whose electric demand may materially alter statewide load forecasts are protected. And thoughtful adaptation is appropriate. The General Assembly has already begun important work. For example, Senator Mark Romanchuk's effort related to utility forecasting reflect an important principle. That is, infrastructure should be built based on demonstrated need and not speculation. As utilities cite projected data center growth to justify infrastructure investment, forecasting matters. Announcements are not the same thing as operational facilities. Forecasting should be evidence-based and milestone-driven. Likewise, we have seen bipartisan efforts by Representatives Dave Thomas and Tristan Rader that reflect thoughtful work regarding utility service for data centers. Their efforts in establishing a statewide data center tariff appropriately raise questions involving customer commitments, financial accountability, cost responsibility, and importantly, stranded cost protections. OCC appreciates these efforts because they reflect something really important, that Ohio is already moving thoughtfully in the right direction. I also want to briefly mention the role of regional markets. As Awesome Hawk testified this morning PJM plays an important role Regional coordination matters Reliable wholesale markets matter But affordability is ultimately experienced locally It experienced at Ohio kitchen tables Questions like, who pays? Who bears the risk? How should extraordinary load be treated? What protections exist for consumers? These are fundamentally state questions. PGM cannot answer retail affordability questions for Ohio families. States still have an important job to do. Across the country, policymakers and governors increasingly recognize that affordability and resource adequacy cannot simply be left to the market alone. States, like Ohio, must continue doing the hard work of protecting consumers while supporting growth. Ohio has an opportunity to lead. OCC also recommends consideration of what many refer to, it's been spoken of this morning, as bring your own new generation, or BYONG. At its core, the concept is simple. If a customer requires extraordinary electricity demand, there should be a credible plan brought forward by that customer for how its load will be served. That plan could include long-term supply agreements, dedicated generation, storage, firm procurement arrangements, or meaningful demand flexibility. The objective is straightforward. Extraordinary new demand should not undermine affordability or reliability for existing consumers. consumers should not face unexpected costs because planning assumptions were overly optimistic. I would also briefly mention one additional opportunity that deserves attention, advanced transmission technologies. As Ohio considers how best to respond to increasing data center load, policymakers should give thoughtful evaluation of technologies that can help maximize the use of the grid infrastructure that we already have. These technologies have the potential to improve system efficiency, reduce congestions, and increase transfer capability without immediately requiring more expensive infrastructure expansion. Put simply, Ohio should optimize before it over builds. Before residential consumers are asked to finance significant new transmission or infrastructure expansion, utilities and planners should evaluate whether cost-effective technologies can make better use of the system already in place. OCC appreciates that the General Assembly has already recognized the importance of this issue through HB 15, which included a study requirement advancing transmissive technologies and opportunities to improve grid efficiencies. And so what should Ohio do? OCC respectfully recommends several guiding principles moving forward. First, protect residential customers from cost shifting associated with data centers. Second, require meaningful customer commitments and financial accountability for data center growth, and that is primarily through statewide tariffs similar to the AEP tariff Third consider credible supply planning or bring-your-own-new-generation approaches for data centers. Fourth, explore separate procurement or reliability structures for data center demand. Fifth, and finally, strengthen our forecasting standards so that infrastructure reflects demonstrated need and not speculation. Chair Chavez and Holmes, members of the committee, Ohio has an opportunity here. We can welcome innovation, we can support economic growth, and we can compete for investment in the rapidly changing economy. But growth must occur responsibly. and that means keeping one principle at the center of this conversation. Ohio households should not subsidize speculative infrastructure or bear affordability and reliability risks associated with the extraordinary private load growth of demand centers. OCC or Ohio does not face a choice between economic development and consumer protection. We can do both. We can welcome growth while protecting affordability. We can support innovation while preserving reliability. And we can ensure that Ohio remains competitive without asking families already struggling with utility costs to unfairly shoulder the risks that they did not create. That is how Ohio threads the needle. Thank you for the opportunity to testify. I'd be happy to answer any questions the committee may have.

Representative Claggett. Thank you. Appreciate your help today. It was three years ago in this building we were having significant debates in regard to what happens if the electric car industry explodes and have a need. I would like to refer to data center loads as point loads and that is a potential car issue as you know a evenly distributed load. However it's not even. It's individual houses making their choices about that vehicle. My question is, does your council or did your council three years ago have a position on how in the world we were going to pay for that significant load that was being contemplated at that time? And how does that impact how we're thinking about these point loads for today? Chairs, to the representative, thank you for

So on that point, where do you draw the line on creating customer classes in order to follow the cost causation?

Thank you for that. I get that. I'm just wondering where we draw the line, right? So data centers are one thing. EV cars may be another thing. How far down the line do you take all these customer classifications? I don't want to be discriminatory in the state. I want to be fair.

To your knowledge, is there a definition of what a customer class is or how it's defined, or is that something we need to look to you to help us with?

Speaking of tariffs, what are your thoughts on the AEP tariff?

Representative Glassburn. Thank you, Mr. Chair. So I'm going to ask you a couple questions that are, I think, going to superficially sound contradictory but I think it what to some degree to what the chairman was trying to wrestle with which is so with the PCO chair I asked about how much how are we certain that residential customers aren subsidizing data centers right now? How far do we look back? How do we, do we make any kind of restitution?

Sure. Okay. I want to be clear-eyed in this committee. I want to make sure that we are looking at data centers for what they are and not trying to lump all the blame on them. I'm not being pro. I want to make sure that we're being pragmatic. They may be 50% of the auction price now, but I will submit to you that our prices are high right now, not because of data centers wholly, but because we've taken baseload generation offline. We've taken 14 coal-fired power plants offline in Ohio over the last decade. We have artificially cut supply. Now we're scrambling to get that back. So I just want to make sure I'm not saying that you're wrong. I just want to make sure as a committee we're looking at the whole, and this isn't an inquisition on data centers.

And so that leads me back to where I was going to ask kind of the second half that seemingly go in two different directions, one being how do we keep this neutral for residents and make things whole from the past. But I have seen, at least in the generation and data centers that I have visited or learned about, that there's wide variance in terms of how much they are impacting the grid, whether they're completely off the grid, whether their existence actually might end up being beneficial to the grid and lower cost. I'll take the example of the site where many of us went to, Vistra reinvigorating the perinuclear plant, where they're putting in new turbine blades and that that is a facility that otherwise would have come offline, and it's in the long run possibly going to even lower cost for people in that service market. So let's say we did have some kind of way to make sure residential customers were able to be made whole and not impacted. How do we account for this wide variance in these, even within the subsection of data centers, as to what their structure is, what they're doing, how they're impacting the marketplace?

Senator Reineke Thank you Chairman Thank you Director Willis for being here I appreciate the positivity of your response to the committee And I guess my question goes back to you represent the Consumer Council. You've been pretty direct on saying there's so many benefits to Ohio and to the industry, knowing that the issue that we have with our constituents is trying to neutralize the negativity. How do we do you or how should we approach this with our constituents for them to understand what you're saying on behalf of the consumers in Ohio?

Follow-up. Thank you. Do you see that there is a need for a direct program or a direct, not program, but direct communication that needs to go out from the Consumers Council to address this? Because to me it's a very, very serious issue for lots of reasons.

And certainly, you know, on the local end, there are questions that communities have. But we certainly view our role as important and would love to be able to work more and get that message out about data centers. Again, I think we do, right now we do that. That's, you know, when you go to our webpage, that's the greatest hits is from people looking at our data center information. So, again, that education is important, and OCC would be happy to be part of that education effort. Thank you. Senator Blackshear. Thank you, Chair, and thank you so much for being here with us. Just a couple of questions. how are transmission costs that are necessary for data center and interconnection being funded? Chair, to the Senator, right now there's the transmission costs. that are approved on the federal level. And those, they just, they're approved by FERC after the utilities present forecasts to PJM and PJM approves projects. So, and that's where, again, the importance of forecasting and importance of getting those forecasts right and not overbuilding, because that then becomes the approval process. And those transmission costs, once set by the federal government, the state has no choice. The only choice the state has at the local level, the PUCO, is how are we going to allocate it? What customer gets this piece of the pie? What slice? How big is that slice? So those transmission costs come down. I also will note that there's, in addition to those regional transmission costs, there's supplemental transmission costs that are more local. Those are ones that, again, the costs get passed through very quickly. Those are the ones that there's very little analysis of. In fact, OCC filed a federal complaint with FERC saying we need someone to look at those local costs because right now the Ohio Power Siting Board is not looking at them, the PUCO is not looking at them, and PJM's standard was do no harm, has been not protective of consumers. So we have a federal complaint filed on that. And I think, too, there are some legislative talks underway about getting those local transmission projects under the jurisdiction of the Ohio Power Siting Board, and that would also be helpful. Follow-up? Thank you so much, Chair. Could you also explain the cost shift that residential rate payers are seeing on account of data centers? To the Chair, to Senator, what we're talking about, certainly we talked about the capacity costs, the capacity costs being high. They are passed down to consumers. On the transmission and distribution side, we haven't seen so much, but it's coming. The projects are starting to be built. There's a recognition that in 2030, 2031, 2032, that's when we're going to have an imbalance between supply and demand. And so those projects are starting to be approved at the federal level now. So I think we're just starting to see those. And that's why it's important to get ahead of the ball and to have some preventative measures in place to keep those costs from being shifted to residential customers. One last follow-up. Thank you so much, Chair. I see on slide, page 8, that you have some legislative recommendations and then it goes down to page nine, would you consider those consumer safeguards that should be implemented in a time of increased energy costs? To the chairs, Senator, yes, those are our recommendations. Again, they are principles, and certainly they are principles that are being worked on at the legislature currently. Again, I mentioned Senator Romanchuk's forecasting legislation, which looks at and puts some teeth into the policy and principles. And certainly the legislation being worked on with Thomas and Rader addresses the points that we make about having a statewide tariff and requiring skin in the game So I think those policies and recommendations are being implemented but more can be done So yes those would be our recommendations for this committee Representative Workman. I'll make it very quick. Thank you, Chair. Thank you for your testimony today. Very quickly, one of your pillars of success for the implementation of these recommendations is to strengthen forecasting oversight. You mentioned specifically milestone-based forecasting. What exactly do those milestones look like? Well, to the chair, to the representative, the milestone forecasting means that it's similar to, I think it's similar to the data center tariff approach that was discussed, where we don't take speculation, where we require commitment. We require signed agreements, where we require contracts so that it's, the forecasts are not based on speculation, they're based on clear advancement and clear plans to go forward. And again, I think from, we talked about the data center tariff, how there were, it started out, AEP started out with 30,000 megawatts. They went through the data center tariff, the requirements for load studies, the requirements for signed contracts, for agreements, And at that point, if we think about it as a pyramid, down here's the 30,000. Now they've whittled it down to 5,600. So, again, those are milestones that have to be reached in order to submit that forecast. And, again, we're talking about the forecast that's submitted to PJM, and then PJM approves the project. So really important to get those forecasts right and to be transparent about those forecasts. I think that that hasn't been talked about yet this morning. I was happy to be able to talk about that. I know you've heard a lot, but that is one thing, a really important thing. And again, Senator Romanchuk's bill is really driving toward that and is a very good step to getting those milestones and those protections in place. I see no further questions. I just have a comment and then a question. On your reconductoring for the optimization, I'm with you. That sounds like an easy thing, but in our research and study, that becomes a very difficult cost-benefit analysis as you go through, and that was my comment. The question is related to Chair Holmes asked earlier about, you made a comment about speculative infrastructure and how that could be applied. problem. If we attribute cost to a data center and they build two miles worth of distribution line to their center, is there a way to quantify the benefits to communities that are able to tap onto that line, ancillary benefits that we're gaining from that? Chair, great question. I'm sure there is a way. As I stand here today, I don't know what that way would be, but I would think that our planners, our utility planners could certainly figure that out because it is, you know, if investment is going in and is primarily to benefit the data centers, but there are benefits to other customers, certainly that could be taken into consideration when you're talking about cost causation. And it's not just electric lines. It will be natural gas lines. it be water infrastructure I think we can get some benefit out of this and certainly we can get some benefit out of cost sharing that you know you have fixed costs and if there more customers on the system to share fixed costs then that is also a benefit as well Seeing no further questions, thank you for your testimony. Thank you. Next, we have testimony from Director John Logue. I knew I was going to. Okay. John Logue of the Ohio Environmental Protection Agency. Welcome to committee. You may begin ready. Good afternoon, Chair Chavez, Chair Holmes, and members of the committee. I'm John Logue, the director of the Ohio Environmental Protection Agency. Thank you for the opportunity to testify today. You have our written testimony with you. A quick edit to that. I believe at the beginning it says good morning, so we'll change that to good afternoon. I want to start with a simple but an important point. Ohio EPA does not have a special permitting program for data centers. We evaluate them the same way we evaluate any large industrial or commercial project based on the specific environmental activities and potential impacts involved. Additionally, Ohio EPA permitting is largely shaped by federal law, primarily the Clean Air Act and the Clean Water Act, which set minimum national standards that states must meet. Our work with data centers touches three main areas, water quality, air emissions, and water capacity. To begin with water quality, most data centers in Ohio send their wastewater to municipal treatment systems and do not need a direct discharge permit from Ohio EPA. But for those who may discharge to a stream or a river, we issue a National Pollutant Discharge Elimination System, or NPDES permit that establishes pollutant limits and monitoring requirements. Ohio EPA has only issued one such NPDES permit for a data center at this time. Ohio EPA also requires permits for infrastructure that handles cooling water before it is discharged and stormwater permits for construction sites disturbing more than one acre. If a project affects wetlands or streams, additional federal and state approvals may be required, such as Section 404 permits from the U.S. Army Corps of Engineers and Section 401 water quality certifications from Ohio EPA. Finally, I should mention that earlier this year, Ohio EPA had proposed a draft general NPDES permit to streamline discharge permitting for data centers statewide. Ohio EPA uses general permits to streamline the permitting process with uniform requirements and standards and currently offers them for specific stormwater, wastewater, and air pollution control permitting. General permits do not lower standards as established in federal or Ohio law. Ohio EPA received thousands of public comments and is currently reviewing those. There is no final decision timeline has been set or made at this time. On the air side, Ohio EPA issues permits for emissions sources at data centers, most notably backup diesel generators, as well as for independent or behind-the-meter power stations being built to serve data centers directly. Facilities are classified as major or minor sources based on their emissions levels. major sources are large-scale facilities with potential to emit 100 tons per year of regulated pollutants and are subject to more stringent requirements set under the federal Clean Air Act minor sources have much lower emissions and are subject to fewer federal regulations but are still required to meet all state air pollution control requirements Regarding water capacity data centers often connect to municipal water systems rather than developing their own supply When they do, local utilities must work with us to confirm that they have the sufficient capacity to handle the added industrial demand. If new or expanded treatment infrastructure is needed, those plans must be reviewed and approved by Ohio EPA. Finally, I want to highlight that public participation is built into our process. For major permits, the public receives notice, has an opportunity to comment, and can request a formal hearing. Ohio EPA responds to those comments before issuing any final permit decisions. Thank you for your time today. I'm happy to answer any questions you may have. Thank you for your testimony. Any questions from the committee? Representative Workman. Thank you, Mr. Chairman. Thank you, Director, for your testimony today. I have to admit that I am somewhat shocked because the publicity or the public narrative that's out there right now is that the Ohio EPA was considering allowing the output of chemicals and toxic solutions from these data centers into our public water supply. I guess my question is, how would you respond to that? How do we correct that narrative? And what is your recommendation for trying to kind of clear those waters? Thank you, through the chairs. Thank you, Representative Workman, for that question. The NPDES permitting, we have over 1,500 of those that have been issued in Ohio today for all kinds of businesses and industries. So some of the narrative, certainly what we saw in some of the comments, was that this was some radical departure from permitting that happens in Ohio and across the country all the time. The proposal is not. So a data center potentially could discharge to a stream, a river, a lake today, as can other businesses. So not all of Ohio and not every piece of property has access to a public water system or to a municipal sewer system. So the NPDES permit is structured exactly to allow that. So what gets missed is it's a regulated discharge. It is not a free-for-all. And any discharge may not reduce aquatic habitat, aquatic life, recreation opportunities, etc. So it really gets into potentially the body of water that may be discharged to for any of those permits. What would be discharged to that water? And can that particular body of water handle that? So it is a very much case by case based on what's happening. but it was not, in and of itself, not anything unique from what we do today. Follow-up? Thank you, Chair. Is there a risk of forever chemicals being infused into our public water supply, streams, lakes, anything of the sort? Through the Chair's representative workmen, data centers have not been identified as a source of the PFOS, the forever chemicals, either in Ohio or by the United States Environmental Protection Agency. Federal requirements don't yet exist for discharge of PFAS, but those are coming online under Administrator Zeldin's leadership, so we're actively monitoring that. Senator Wilkin. Thank you, Chair. So just to follow up, to kind of clarify, currently are you aware, you said there's only one permit. Are you aware of either there have been violations or is there anything ongoing or have they been halted from discharging? And do you foresee, with a lot of the changes, do you foresee additional discharge permits being approved? Through the chairs. Thank you, Senator Wilkin, for that question. On the overall, I guess what we're seeing with data centers, I asked our team to take a look. Have we had any events where we've had to deploy our emergency response team that replied to events across the state of Ohio every day and all kinds of spills, discharges, unauthorized activity? We've had a grand total of two that involved spills of diesel fuel during construction. For context, I will probably have five emails in my inbox when I return to the office about diesel spills somewhere in the state of Ohio today. So all things considered, it's pretty manageable. That's the extent that we've had on an emergency response. From an overall compliance standpoint, we have only issued two notice of violations to any data centers in Ohio. Both of those were for late paperwork submissions to us. So from a compliance standpoint, complying with the terms of their permits, we are not seeing any issues writ large. Regarding the question about do we anticipate potentially additional coming online that may be seeking an NPDES permit, hard to say. I think our intent when we went down the process of considering this draft permit was in anticipation that there may be. and are we ready to respond to that if that becomes an eventuality in Ohio? To this point, we're not seeing it. You've heard from some of the other folks testifying today about the increased use of water reuse and closed loop, which we're very supportive of. We think that's a good idea wherever it can be placed in. So right now we just have the one. Hard to say. And oftentimes we're not sure that there's a project happening until we receive a permit application. That's often the first real knowledge we have that there's a project underway. Follow-up? Thank you. So strictly as it pertains to the discharge, has there been any issues with that at the one you have right now? Through the chairs, no, Senator, there has not been. Thank you. Senator Blackshear. Thank you, Chair, and thank you so much for being here. we talked a lot about water. I want to talk about air. Data centers need power 24-7, so I'm curious on what impacts to air quality do their backup generators pose. Through the chairs, Senator Blackshear, to date we have permitted just over 4,900 backup generators that are affiliated with various data centers across the state. Those are backup emergency generators, so they typically are restricted for how often they can be used. Ideally, they're never needed to be used in the event of a power disruption. They will typically operate them once a month to turn them on to make sure they're working and to maintain them. Overall the emissions are pretty low But again within the constraints that we under and Ohio law of the Clean Air Act all of those emissions are regulated and fully comply with all federal and state standards under the Clean Air Act. Chair Holmes. Yes, thank you. This is illuminating. This is great. A lot of constituents have talked about, so you talked about discharge. Their concerns came from data centers accessing aquifers and drilling their own wells for water. I'm assuming that if they're discharging into the sanitary system, they're also drawing their water from that. But is that correct, or should we be thinking about aquifers? Chair Holmes, that's correct in most cases. So I believe Director Mertz is testifying after me. She and the Department of Natural Resources are more handled the kind of water quantity and a withdrawal of water. When there is anyone seeking to withdraw water from any body of water from or potentially digging a new well for an industrial use, we and DNR work very closely on that. Again, most are discharging to municipal systems. Most are getting their water from public systems. Part of what we do, though, is to make sure that that system has the capacity. Do they have enough capacity in their source water, where they're pulling their water from, to supply to their customers? Do they have enough for that potential increased demand? And then if the data center is discharging to that public water system, to that municipal wastewater treatment plant, do they have the capacity to handle that increased load? So sometimes it does require infrastructure improvements. if they want to be able to provide that to a data center, and we would work with them on the engineering plans, the permitting for that increased capacity. Follow-up? Yes, sir. Representative Claggett. Thank you, Chair. In regard to municipal discharge from industrial water users, is it fair to say that data centers or any other product like that is it is a closed loop or some such system that they are discharging some of their water waste to the municipality, that they have to undergo the same standards that any other industrial user has, such that if the discharge water is not of a certain type that the processing plant can handle and filter out any impurities, that they have to site process that. Is that fair? Through the chairs. Yes, Representative, that's correct. Right. And so a follow-up would be, is there any data center discharges that are having to be pre-filtered or changed in some way before the municipality can accept it? Through the chairs, Representative, I am not 100% sure on that, but we can find that information and provide it back to the committee chairs. Please. Representative Glassberg. I'm sorry, Mr. Chair. Thank you. So, okay, so 97, 98% of this is going through municipal sources. Do they have the obligation to serve no matter what? Do they have the flexibility to manage the quality of the water coming back? Can you comment on that? Through the Chair's representative, I am not 100% sure on that obligation to serve question that you asked. so I can find that out, and we will provide that back to the chair. From the discharges they're receiving at a public water system or wastewater treatment plant yes they have obviously some say in what they receiving that coming through them because it will have to be treated at the source or it will have to be treated at the plant. So if the plant doesn't have the ability to treat it, there's going to have to be an arrangement made or circumstances made that's got to be treated somewhere along that line on the discharge. But we can provide more detailed information on that through the committee chairs. Thank you. And maybe it's a question better suited for the data centers themselves, but we're talking about discharge in the abstract, and it sounds scary. I mean, any idea what the discharge is? It is typically what we would expect it to see. Wouldn't necessarily be anything unique or different than we might see in a lot of industrial users. It will be dependent on where they're getting their water from. So the chemistry of the source water will certainly have some impact on what is eventually discharged, so where they're pulling their water from, the concentration that's run, and the chemical treatment that may be in use. So it will vary, but typically it's dissolved solids and some salts is what we would expect to see. Thank you. Seeing no further questions, we thank you for your testimony. Thank you. Last, save the best for last. We have testimony from Director Mary Mertz from the Ohio Department of Natural Resources. Welcome to committee. Thank you for hanging with us all day. You may begin when ready. Good morning, Co-Chair Chavez and Holmes, members of the committee. Afternoon. It was going to be morning. So I'm your last person, so I'm standing between you and escape from here. So I will try to summarize as we go, but I appreciate the opportunity. You know, Ohio's fortunate to have plentiful natural resources, put the state in a position to attract a wide array of prospective businesses. We're considered a water-abundant state, right? We're framed by Lake Erie and the Ohio River, seemingly endless streams and surface water resources, strong groundwater supply. We are starting in a great position. And so we are engaged in steps to protect that water supply for the future, for all the development we see for generations to come. And we have scientists engaged in monitoring and helping us understand Ohio's waters. And today I'm going to focus on just three things. How we, our registration system for water use, our permitting system for water use, and the resources we provide to communities and industry to understand Ohio's resources. So I'll start with water withdrawal registration. So we maintain a registry pursuant to statute, and the owner of a facility that withdraws water at a quantity greater than 100,000 gallons per day is required to register and report their annual water use with our Division of Water. No fee. It's not a decision. It's not a permit. It is just a registration. And these annual reports provide information about the facility and that water withdrawal volumes to the public, including other state agencies, local governments, consultants, anyone involved in the planning and review of proposed developments. We make it available on our website so anyone can access it. However currently we have nearly 4 3 registered facilities 777 of those are public water supply public water systems 465 for agriculture and livestock use, 288 for industry, and 42 for power generation. So power generation and public water systems withdraw the most water of all those sectors. But if you tie into a public water system, you do not register separately. And all of Ohio's data centers currently tie into a public water system. So there are no separate registrations for data centers. So that is just something we do not have visibility into their actual water use. Now what we can see on the public water systems, and I'm happy to provide some follow-up information, but we can see in certain metropolitan areas where you see data centers located that their water use is going up. But we don't know how much of that is due to the data center or something else. They don't need to break it down for us. So while we do have a water use registry with nearly 4,000 registered facilities, there are no data systems because they are included in the public water supply registration. So that's the first way we are involved in protecting Ohio's water resources. The second way is the high-capacity withdrawal and consumptive use permit. So this is an actual permit. It's provided for under two different sections of the code, and the requirements for who needs a permit is a little bit different in the Ohio River Basin and in the Lake Geary Water Basin. And that's really due to, in the early 2000s, the Great Lakes Compact at that time, Great Lakes and St. Lawrence Waterways, Water Resources Compact, had some higher standards for the use and withdrawal of water in the Great Lakes Basin. So under this process, like I said, slightly different requirements for Great Lakes and Ohio River Basin and slightly different thresholds for when you need a permit. And what is required in the permit is outlined in statute pretty specifically. But I can tell you that in Ohio, only six high-capacity permits have been issued since the late 80s, and none of those are in use. So it's fairly rare that somebody would need one, and the ones that have been issued are not currently in use. Either that development went away or things didn't progress as they had anticipated. Today, there are no data centers that are permitted by ODNR under these provisions and statute because of exemptions that are in the statute. So again, they're tied into public water systems, and those are exempt from this permitting requirement, as well as the power generation facilities are also exempt from the permit. So I guess on the registration and permitting front, what I'll tell you is today we have no data centers either registered or permitted. But what we do have is a strong structure that the General Assembly set up years ago. It is workable. We have both registrants and permittees, and we are prepared to use that should a data center ever go to a rural area where they can't count on municipal water supplies. So we have the system in place. It's ready to execute when it's needed. It just has not been used at this time for data centers. and the final thing I want In addition to our permitting authority, we're involved in providing information and data to support industry and communities to understand their water supplies and help them make decisions about what they want to allow with their water supplies. So we conduct groundwater research. We supply publicly accessible statewide groundwater data that can be used to guide industrial and business siting, including data centers. We assess an aquifer's ability for supporting water withdrawals. We evaluate the long-term effects of pumping water above ground. Our website has online mapping tools showing fundamental facts about our aquifers. We show water yield, rate of recharge, how deep the accessible groundwater is, how easily that groundwater can move through an aquifer. So we have a lot of information that can help all entities, including data centers, find the right place to locate because all water resources are very site-specific. We maintain the state of Ohio's ever-growing official database of water well logs. Over a million records currently of private, public, and exploratory wells, all of this available on our website. We operate an observation well network that includes over 150 active monitoring wells. They've been collecting continuous groundwater-level data for decades. We also do additional special research projects. the General Assembly directed us to do a special groundwater monitoring project up in northwest Ohio called the Machindo Aquifer, Michigan, Indiana, Ohio, an aquifer that serves all three. They provided funding. We installed 12 new long-term groundwater observation wells. We did extensive aquifer testing, and we have a really refined understanding of what is going on in that area. And we can do that anywhere the General Assembly would direct us to. So that allows us to assess the impact

Thank you for your testimony. Any questions from the committee? Uh-oh. Okay. I thought you wore us out, but you didn't. Representative Claggett.

That's a big question. I would say as a general matter, we are always concerned about water quantity. So when I see articles about California or Arizona where development has occurred and they don have the water to support it I show that to our staff and I say it is our job if we think this is going to happen to raise the red flags to you so you can make the decisions that you feel like you need to make But we are a very water state I have confidence we can do lots of development We just need to be thoughtful about where it is and we don't inadvertently put it in the wrong place and get impacts we don't expect.

Well, so when we get that data, we provide it to others. So really, I think the question is, could you all use that data, or could the communities, or could the other industries in the area so they have a better understanding of why water is being withdrawn from that area so they can make their decisions. So, I mean, it would certainly provide more transparency to have that registration requirement.

Thank you, Chair. I'll be very quick. Thank you, Director. So, obviously, it sounds as though we agree that the groundwater mapping and aqua-fora research is important to planning and preparing for these developments. Are there geographic regions where additional research should occur proactively before we allow additional data center development?

Any further questions? Seeing none, thank you for your testimony. And thank you for hanging in there with us all day. The Department of Development was not able to send anyone here, but we have written testimony from Director Mihalik. Yes.

Thank you for that. Yes, I know they had personal conflicts. They weren't able to attend today, and you're absolutely right. It is a very important topic. The topic of tax abatement is on the forefront of everyone's mind, so we will re-invite them to come at a later date. and in the meantime, we have written only testimony on your iPads for review and it will also be on the website. So with no further business before the committee, we are hereby adjourned.