All right, I think we're ready. Great. I'm Will Dotson from the Investment Banking Division at Morgan Stanley, and I'm delighted to introduce a first-time participant in the conference, EMCOR, the management team is here. We've got Tony Guzzi, who's Chairman, President, and CEO of the company, Jason Nalbandian, who's the Chief Financial Officer, and Andy Backman, Vice President of Investor Relations. Welcome to Laguna. We are thrilled to have you guys here.
Thank you.
Since it is your first time, Tony, maybe for you, why don't you start out and just give us a little perspective on the EMCOR story, talk a little bit about your businesses, acquaint the group with what's going on at EMCOR.
Yeah. So if you just take it at the top level, EMCOR is a company where we actually do the work. So we're a company of plumbers, pipe fitters, electricians, sprinkler fitters, HVAC technicians, and operating engineers. So said simply, we're a skilled trades company. Highly trained people, sort of if you look at trades, we're up here at the top of the food chain. So we're looking to either build something, fix something, retrofit it, and then do ongoing maintenance. If you look at the company today, we have a mechanical and electrical segment. Give or take, it makes up 70%-75% of the company.
We have a building services segment that the mechanical services business in that is about 70%-75% of the revenues of that business, and the rest of the other stuff is site-based where we have people actually running buildings or doing raw technician work. We have an industrial business, which is a little bit of a misnomer. It's really our oil and gas business, mostly focused downstream doing mechanical work and some electrical, with some midstream and upstream, and then we have a business in the UK that's really a facility services business. The common theme across all of it is, go back to my opening comment, the people that work in each one of those segments are by and large doing the same kinds of things, just whether they're doing it on a large construction project, a retrofit job, or an ongoing maintenance contract.
Now, if you think about overall, at our heart, we're contractors. So that by definition, we're flexible, we adapt, and we don't really create demand. A little bit in the aftermarket, we create demand. We react to demand. And we have to have our company positioned to be able to execute for our customers. In a lot of ways, we do control the mix of what we work on by how we're going to apply our resources. The other part about EMCOR that's core to who we are is this is a values-based company. We very much believe in values of mission-first people always. We believe on the mission-first side that it's things like integrity, discipline, transparency. Think about what we do. We wake up today, Jason and I and our teammates have people working on at least 12,000 customer sites, doing fairly dangerous stuff in some cases.
These are guys that are at the tip of the, as you said, screwdriver. They have welding torches in their hand. They're working in dangerous facilities, lots of heavy equipment around them. The other part of our thing is people always. We do operate in an environment of mutual respect and trust. Secondarily, safety is a core value. We are Six Sigma in safety compared to the rest of the industry. We operate at a TRIR of less than one. And then finally, we very much believe in teamwork. If you think about how we run the company, we have those segments. But how we actually run the company is a combination of how you would run three types of companies. At the top of EMCOR, our top 150-200 people, give or take, feels very much like a professional services firm.
That might seem odd to you, but they are our partners. We operate through 100 subsidiary CEOs. We want them to feel empowered to make decisions at the local level and project level. Then we knit it all together at the segment and corporate level. They are my partners. They're Jason's partners. They're Andy's partners. These are very sophisticated business people. About 60% of them grew up in the trades, and they're running businesses up to a billion dollars right now. Great story, right? Great American story. The next part is we very much, and I think this is what distinguishes us from other people in our space, is we run this like an industrial company. That's my background. 20 years ago, we run this like an industrial company. This is a very metrics-driven company.
And you see that through our cash conversions, up cycles, down cycles, no growth cycles. We typically generate cash at least equal to net income or a little bit above. And again, that's not common with all contractors or people in the services industry. And then finally, in some ways, we look like a platform private equity investor because we have four or five places we relentlessly invest to grow our footprint geography-wise, capability-wise, to be able to do what we need to do for a customer in both an end market and a geographic market. Now, you think about what's driving our business today. We didn't wake up one day and say, "Oh, we're going to be great data center builders." We had built the skills to do that over a long period of time, including we built some of the first data centers in the early 2000s.
There's a little bit of resurgence around 2012 or 2013. But what we learned is, and we knew this, is good industrial electricians and pipe fitters and HVAC people are flexible once they get the rhythm down on that job. And that's what's allowed us to go from serving in 2019, three data center markets electrically and through their search for power to be able to expand, to give or take 15 today. And that's why mechanically we went from servicing one or two to four or five today. A little different. And then fire life safety, we can serve every data center market in the country. And so we also are pretty good at capital allocation. If you look at us over a very long period of time, about this much has to go in, maybe 10% goes into growing the business because we're not a capital-intensive business.
40% or 50%, we return to the shareholder, mostly through stock buyback, a little bit through dividend. And the other 40% or 50% has been through acquisition. So that's a little overview of who we are. We run lean. We have a corporate office that maybe has 100 people in it. And we've been able to grow the company to where this year we think we'll do somewhere between $16.3-16.4 billion and $16.8-16.9 billion in revenue.
So that's a super overview, and I definitely want to come back to the capital allocation piece. But why don't we start macro? One of the very interesting things about the story is the growth drivers that you sit in the middle of right now. Clearly, at the conference, we've talked a lot about data centers. We've talked a lot about reshoring. We've talked a lot about electrification. So I think those are probably. I'll let you explain the nature of those drivers and then give your perspective on where they're going.
Jason and I are going to sort of do this together. I'll talk here, and he'll tell you what that's meant in our business as a follow-up, okay, as far as growth. So we always thought that if you bought and expanded your company and did the right training and you built the best supervision and the best training programs with the union and outside of the union for your trade workers, that you would win. That's been sort of one of the theses of the company all the time. We're long-term investors in leadership development. But if you take those trends and you pull them apart, maybe just because of my bent on the world, I've sort of never been a big China bull. I sort of guess that there would be reshoring happening at some time for a lot of geopolitical reasons.
I'm sort of the guy that will have been in those meetings in another setting. People would say, "Oh, this is all going to happen." I said, "Yeah, but there's one part that just doesn't make sense. Would you have invested in the Soviet Union in 1975?" Well, these are discontinuous political systems that will have a hell of a time working together. So our answer to that over a long period of time on the reshoring front was to invest like crazy and build a footprint where we didn't have one in the Southeast, Mid-South, and Southwest. If you had been in EMCOR in 2007, we almost had no presence. We were California, the Northeast, a little bit the Midwest, but we didn't have any presence in the Southern United States. So our guess was that's where it was going to happen when it happened.
We started seeing it happen sort of 2015, 2016, 2017. You started to see it. And then, of course, COVID acted as an accelerant because I guess we figured single source supply that used to be chapter and verse in 2000 to 2008. We sort of figured single source supply out of one plant with a complicated supply chain really wasn't a good thing. And so it's going to come back. It is coming back with a lot more automation, and it is happening exactly where we thought it would happen, with the addition really of Ohio, Indiana, Iowa, and some midwestern states. So that's the reshoring trend. And Jason, we see that in our backlog in our revenue growth, right?
Yeah. If you look at our remaining performance obligations, which for us is really just a measure of backlog, if you look at our traditional manufacturing, you can see we're up almost 30%. You're a little over 30% year- over- year. And that's where we're seeing some of that manufacturing and reshoring.
If you took it over a five or six-year window, it's probably a high single-digit CAGR.
For sure. Right? I think the thing about us, right, is data centers are definitely a big contributor to our growth, particularly if you just look at this year through the first half of 2025, where we've more than doubled our data center revenues. But our markets are more diverse than just data centers. I think if you look through the broader companies, our healthcare revenues, for example, are up 40% year to date. We touched on manufacturing and industrial work. Our water and wastewater RPOs, as well, are up 20%. So fairly diverse. And I think the thing that we often look at is if you boil down the rest of our business and you take out the high-growth market sectors and you strip away data centers and you strip away high-tech manufacturing, we're still growing mid-single digits. Let's call it 4-4.5% if you look over 2024 or 2025.
And when you strip that out of non-rez, you'll see that we're growing above non-rez about twice, a couple hundred basis points. Then you go to data centers. It's interesting because there's been this disconnection of thought, and I'm going to tie it to what I think is I would call it electrical expansion, not electrical. I've never been sort of the electrification guy. I've been more the, "We need more power," guy. And maybe our team here had a unique seat at that because we knew what people were thinking about as far as data centers and semiconductor plants and manufacturing plants. And then you sort of put that against what people were talking about energy supply, especially from 2021 to 2023, you're like, "This doesn't work. You're going to need more baseload power. You're going to need more fossil power.
Nuclear is at least 10 years away," so let's just focus on data centers. We're pretty good at it, and we're only good at it because of all those things we talked about before: the training, the development, the execution in the field to our subsidiary companies, the way those 10 or 12 EMCOR companies that really service the data center market, how they work together, how they figure out how to take a customer from Virginia to southern Washington, and how they work together to make sure we get to the best answer for our customers, and I don't think a lot of companies can do that because it's our culture. You go back to that values culture. That drives us to work together. We trust each other.
Contracting in the end, or a people business like ours, is a lot about trust, right? So our teams have to trust each other that they can execute for the customer. Then you start thinking about the tools, whether it be for reshoring, well, whether it be for data centers, whether it be for healthcare. The essential tools are essentially the same. To do these complex jobs, you have to be really good at virtual design and construction. You have to be able to integrate yourself back earlier into the process. I wouldn't say we're doing design, but we're finishing the design for constructability and taking the lessons learned and helping our end customer, the owner, for the next build be able to do value engineering. So we do a lot more design assist than we ever have.
And you start thinking about the skills you need to do those big projects, those complicated projects are really the same. It's just what market you're going to apply them to. And so we went from having, give or take, a couple hundred VDC people, virtual design and construction people that actually are working in rooms like this, spread out, designing things, and then in trailers. But we have about 1,500 today. If you look at our headcount growth, right, in our construction businesses, which is 70% of the company, give or take, our headcount's growing. Our revenues are growing mid-teens to 20%. Our headcount's growing a third of that.
So you take that virtual design and construction, and then you put it into fabrication, and that allows you to take hours out of the field. It allows speed to market to happen. You get less errors. You put a smattering on top of that of better contractual terms. Today, we're buying less equipment, mainly because of COVID and supply chains. The owners are buying more to themselves, and you see what the effect can be on our margins.
Talk to me about the regulatory environment is clearly shifting and has been moving over the last, not just the last nine months, but certainly over the last several years. Lots of different legislative initiatives. Talk about what you're seeing today, how that affects execution today, and where you think demand grows as a result of some of these things.
Again, I think by nature, because of who we are, we just call a different play depending on the regulatory environment. We act very quickly. The actual regulatory environment, you can argue they were pushing and pulling at each other over the last couple of years, right?
People ask me, "When do you do better?" I don't know. You do better with Democrats or Republicans, and I shrug my shoulders and go, "I have to do well with both." And our construction workforce is mainly union. Service workforce is probably 50/50. And construction unions, other than most of you are probably from New York or California or Chicago, that's a different world than most of the construction unions around the country. The way construction really works is we work with partners with them. We build the bench. We train people. We recruit people with them.
We try to get to the right mix of workers on the job so we can be competitive. So you'd say under a Democratic administration, because there's more Davis-Bacon work, when they wrote the CHIPS Act, they made it more favorable to union contractors, all those things. Okay, that's great. Those jobs are going to be done that way anyway. 70% of the labor on those kinds of jobs are probably going to be union, mainly because of the upskilling, right? A lot of skill. And we can put together a workforce because the union guys from Michigan can come down to Texas to work and check into the book there. And I don't want to get into a whole foray on that right now. So you say, "Okay, that's better." And some of the laws were more favorable.
But then you get into, "Okay, what's favorable now?" Well, permitting is a little easier, probably. A lot of people want to build gas pipelines right now. So we're going to do some compressor stations. The power sector is more thoughtful, I would say, right now. The engineers are back in charge of that, the people that actually understand the energy industry and what it takes to baseload a grid, what latent power really means in a grid to keep it running, what it takes to power data. So think about a data center. Think about regulation. Data centers went from being 10, 20 megawatts when we were building them for people like Morgan Stanley or JPMorgan, 2005, 2006. Then we started building these cloud storage data centers, which are still growing high single digits. And they were 50, 40, 70 megawatts.
Folks, that's a lot of power. That's one-fifth of a turbine output of a GM, a GE LM8000, right? And now cloud storage is sort of getting up to 100 megawatts. AI is about 200-300 megawatts. So let's put that in perspective for you. 200 megawatts, how many houses? Households. It's about, let's say, four people in a household. 200 megawatts probably powers somewhere the power needed for 15-20 thousand people. So a mid-sized town, small town. Let's talk about a data center cluster like they're building down in Georgia. It'll be a mix of AI and cloud storage. I don't understand how all that works. We just build them and service them. But I know the one that's probably AI is 200 megawatt with liquid cooling. And I know the one that's doing the cloud storage is about 50 or 100 megawatt.
That campus, that campus, will be somewhere between 2,200 and 3,000 megawatts. Or if you're a power guy, they say 2-3 gigs. Think about that. The nuclear plant they put in Georgia, I think that's somewhere around 3,200-3,500 megawatts. So that one data center site, that one data center site is going to take the output of that entire nuclear plant that was built and took 10 years to build. So why are we servicing 14 or 15 data center markets now? Because they're looking for stranded power. That's why we're in South Bend, Indiana. That's why we're in Fort Wayne, Indiana. That's why we're on the Indiana side of Lake Michigan where the steel plants used to be. That's why Columbus, Ohio can boom. Why? Because they got connectivity, probably the best connectivity outside of DC, coupled with Ohio River Valley coal and gas.
To get that same 200 megawatts solar, which is intermittent power, to get that same 200 megawatts with a solar field, if you've ever been to an industrial solar field, I'd charge you to go. Anybody have a guess how many acres it takes? It's not a guess. 1,500. To get the power to power that one data center site, that one data center site in Georgia, and to do it with solar, by the way, you have to build gas backup. But to do it with solar, just do the math. 15-25 thousand acres. So that's why you can't book a gas turbine right now through 2031.
I do think, though, when you go to electrification, it's interesting when you talk to utility executives, which we do, when you talk to people. I do think it's back to a density of power argument, which sort of favors, I think it's going to be gas and then nuclear some. And that's the electrification that's going to take place. And if you think about a data center, it's a great microcosm of how you think about battery power and everything else. A data center dedicates about 15%-20% of its square footage to the battery room or the UPS. How long does that UPS typically run that data center until the diesel generators kick off? Anywhere from 7-15 minutes. That's about where battery technology really is as far as density. So a lot could have gone on. The good news is we're contractors.
We're going to react to whatever demand's out there. We were pretty good solar field developers. I mean, builders, not developers, and I think it is going to be all of the above, but you're going to have to gravitate more towards baseload power or else we can't power these AI data centers, and then if you start thinking about places like New York, they're going to have to figure out what they're going to do for power because you're going to need response AI data centers just outside the city because with what you do, latency, that split second is going to mean something just like it does in high-speed trading.
And when we talk to our customers, and you think about demand, whether it be a chip manufacturer, whether it be a reshoring person, and especially the data center people, what they have planned for the next three to five years to just keep up with demand, both cloud and AI, is stunning as far as the power draw that's going to have to happen. So long-winded answer, but.
No, super helpful. I think one thing I wanted to shift and talk about was workforce strategy. There are two questions in here. One is availability to support all the growth. And then second, related to the growth, as you scale, how do you keep standards high? Obviously, Huck talked about safety and the culture. How does that work as you're getting so much bigger?
Yeah. So the only way we can keep the culture is through our supervision. And so we spend a lot of time training supervision. So I sort of start with a macro view, and then Jason will back me up with some numbers on how our workforce has grown versus revenue and all that on concrete terms. Someone asked me a really good question right after first quarter earnings. We were at an investor conference like this. And I'd never thought of this question this way. And they said, "Yeah, Tony, Jason, Andy, how do you ensure that you have the A team on every job?" And I think reflexively, I wanted to answer, "Oh yeah, here's how we get the A team on every job." But I took a step back and I said, "We can't do that.
Otherwise, we can't build capacity, right?" And our customers want us to build capacity, especially at the supervision level. And we typically try to average to a B to a B plus on a job. And maybe our B or B plus looks like it's better than a lot of because of the way we're growing, it's probably better than a lot of other people's As. Sure, we have A guys on every job that matters and gals. And then we have to expand our workforce. We do a lot of recruiting. The diversity of our workforce really expands every year because we find the people. And if we can get a good mix on that job, we can create another team, right?
So our foreman might be a C plus on that job or a couple of them, but they're working with a superintendent across multiple jobs and a project executive that are A players. Likewise, on other jobs, we may have a higher mix of foremen that are A players. And the whole magic with what we do from workforce development is to be able to build more foremen and more superintendents and project managers. And you think about why we have been successful recruiting really skilled tradespeople or having people come from other careers into the trades. I think it comes down to four things. And someone just initiated a coverage on me. I think they copied what I've been saying for like 10 years, right? They must have listened to the calls because I say it on every call. Why does somebody come to work for EMCOR?
Let's start at the trades level versus another company. I think the first reason they come is they're going to get paid every week. Sounds simple, right, and these are in no particular order. That sounds simple, right? I'm going to get paid. That's not true with all contractors, and if you're a union contractor, they're actually going to pay into my benefits fund too. You're going to keep current. They're going to get paid with EMCOR companies, and we're going to pay the benefits too. The second thing is they do dangerous things. Go on any one of our job sites. You see things moving around all over the place planning. These guys are expert planners. The foremen are planning that work with our superintendents, with our project managers to keep people safe. That's the first condition.
So I want to know that you're investing in the best safety equipment. Something as simple as buying the right hard hat as technology develops there. Are you going to give me the right safety gloves? I mean, everything. Fall protection, and they always have that with us. I've never told people that, "Let's go do a competitive bid on safety equipment." Whatever our guys think are right for that local conditions, they buy. The third thing is, if I want a career, not everybody wants a career, right, but if I want to advance, am I in a company that offers me the chance to advance, and am I working for people that actually know what they're doing at the field level? Go back to my point that, especially in the electrical business, a big 80% of our people that run our electrical company started in the trade.
Mechanical is about 50%. If they weren't that, they were engineers or project managers. Our folks in the field know the work. And then finally, if I do a good job for you, am I going to be part of your ongoing crew? And are you going to have work going forward? And then maybe some hard numbers, Jason.
Yes, so Tony touched a little bit on the investments we made in our fabrication capabilities, our capacity, the construction tools we're using, VDC and BIM, and so over the last several years, we've really invested in those technologies to make us more productive and more efficient and allowing us to do more with less labor, and so if you look over, let's say, a five-year period, when our revenue is growing at a CAGR of 9.5%-10%, our headcount growth is only 3%-3.5%, so revenue is outpacing headcount growth almost 3 to 1, and I think that's telling to some of the efficiencies we've been able to gain on our projects.
And then, if you think about retention, because we can't retain all the tradespeople because work goes like this in a local market, but we're trying to retain a core. So what do we do? We train, right? In any given year, we're bringing somewhere between 125 and 150 EMCOR leaders out of the field at multiple levels in the organization. We're training at the frontline level in a centralized way, anywhere from 20-40 foremen a year, 20-40 project managers, where they actually learn how to lead, right? They know the means and methods. We can train that at the local level. We do that through peer groups. But they learn how to actually build a workforce, retain people, create the environment that people want to work in.
Then we have a second course where we bring about 40 people out of the field a year: CFOs, project managers, a couple of superintendents, some people we think may be CEOs someday, or senior executives in our subsidiaries. There we teach them about how you actually run this business economically. And Jason teaches part of that himself. I do. And we do it through a case study. And we do that down at Wilmer Wehrbach. Now we're going to do it at Georgia Tech. And that's on, okay, now you need to understand the language of our business. I think about contracts, all that. Then we have our capstone, which is up at West Point at Thayer Leadership. We've been doing it about 12 years. We've trained about 550 people. And that is, okay, now you're the senior leader. What does that really mean?
And if you look at retention through people that have been through our courses, and the top two we've had for 10 plus years, our retention of our key leadership, other than retirement, is north of 85%. So people aren't stealing people from us. They're trying, but our people want to stay. And I think they see the workforce development. They see the opportunity. They see that we have an operating one. We also have an EMCOR operating model that basically says, if you don't have a better answer, you decentralize, and you centralize where it makes sense and where you can get scale. And that creates this push and pull between entrepreneurship and discipline and process control that I think we've done extraordinarily well over a long period of time.
Great. I think I wanted to wrap up on the M&A piece of capital allocation. So did a big deal earlier in the year. Yep. Why don't you give an update on the Miller Electric acquisition and then talk about M&A roadmap going forward, at least as you see it?
So Jason, maybe hit some of the top numbers, what our allocations look like over the last.
Yeah, so for us, and Tony touched on this a little bit, right? Capital allocation for us is very balanced. We have a fairly disciplined approach. We're not prioritizing high leverage. We're seeking to maintain liquidity. And over a long period of time, if you look, let's say, 10 years, eight years, we really are fairly balanced. We're almost 50/50 between business reinvestment, so the M&A and CapEx and shareholder return.
Miller, Henry Brown, and his team, I mean, they're known as one of the best electrical contractors in the country. We didn't have a big Southeast presence electrically. It's a third-generation family company that has grown really well. So when we think of acquisition, so big picture, we're going to invest with what we know how to do. Our preference would be electrical mechanical construction and mechanical service or building control. That's where we're going to invest, right? There'll be some stuff on the fringes like tuck-ins, but that's where we invest. But the Miller is pretty much the perfect deal for us. On a company like that, we're not looking to make a bargain, but we're not also looking to be in a heated battle with private equity that has no idea what they're buying. This is a complicated buy in our business.
So I started talking in 2019 to Henry about working together. And we hit the right moment when it made sense for both of us. He wanted to keep growing. It was getting a little bigger than what the family was maybe comfortable with. And off we go. There was no other person on that deal. He was either selling to us or keep growing his company. Batchelor and Kimball, very similar. We bought that in 2019. Great mechanical contractor in Atlanta in parts of the Southeast. Brian Batchelor was running the company and his father. Same kind of situation. And obviously, it's went well because now Brian's the CEO of our mechanical segment. And so when we look at a company of size like that, any company we look at, the first thing we look at is can they execute in the field?
If they have a bad reputation in the field, we're not going to try to fix that. We would rather just move on. Because if they don't know basic means and methods, that gets really complicated. On the smaller ones, if they aren't disciplined in the back office and all that, I'm talking like $30 million purchase price and below, we can usually work with them to fix that. Or we just make them a branch of one of our bigger companies. We can figure that out. On the bigger ones, we actually go to a whole different level. If you and I don't share the same values and you're going to run a company for us, which they are going to run it for us, we're not buying you. I don't care how much money you make. Because it's not worth the reputational risk to our company.
So we start in the field, and we go to the leadership, and then we get a match, and we buy a company. Very rarely are we in an auction. Almost never. They may have an intermediary to help them do the deal because that's the first one they've ever done. But very rarely are we in an auction. And quite frankly, we're pretty good buyers, and I'm a pretty tough grader. I'd say we're a good B plus student in M&A.
All right. Well, we're at time. We're delighted you guys joined us. So thank you for being here for the first time. And congratulations on the S&P 500.
Yeah. Thanks.