Dycom Industries, Inc. (DY) Earnings Call Transcript & Summary

We're very pleased to have Steve Nielsen, CEO of Dycom, with us today. Steve, thank you for joining.

Good to see you.

And maybe -- all right. I'm Alex Waters. Yes. I work at Bank of America with Dave Barden on the communications infrastructure team. So we're very glad to have Steve here. Steve, maybe just to start could -- obviously, you're retiring in November here. You've had a long career at Dycom. Maybe could you just give us the 10,000-foot view of how you've seen the industry change, where we are in kind of a cycle versus previous cycles?

Sure, happy to do it, but I finally learned how to do the forward-looking disclaimer disclosure. And that was the -- that's the highlight of every presentation. So just to remind everybody, in the course of this conversation, we may make forward-looking statements that may entail risks and uncertainties that we've more fully outlined in our SEC filings, which we do recommend everybody review. So with that, -- so if you think about the industry from a 10,000-foot view, this is over at least the last 20 years, maybe longer. Clearly, the country or the culture or the economy however you want to think about it, whenever we think we provided enough bandwidth through the telecommunications network, the country has always demanded more. And so supply begets demand, which creates more supply, which creates more demand. And so if you think about today's announcement with Verizon and Frontier, I think about that just as another in a long series of strategic transactions that have facilitated the country's thirst for more bandwidth over an extended period of time. We took a view in the fall of 2016 that we thought that ultimately, fiber infrastructure would pass 75% or 80% of America without the need for subsidized capital from the government. I think we're kind of on that way to $110 million or thereabouts. When we made that statement in 2016, I think there were something like 26 million homes passed. We're approaching 70 million. So I think we've got trend behind us. And it's always encouraging as it was with today's news where deep pools of capital and strong balance sheets intersect with big trends that create significant opportunity for us.

That's great to hear. And I think just maybe off the Verizon Frontier news this morning. I mean any change that you're expecting from them being kind of under the same umbrella right now or in the next 18 months or so in terms of deployments or impacts that could have happened for Dycom?

Yes, I was unable to listen to the presentation this morning, but I understand that Verizon reiterated the Frontier plans through the closing date. I think that's typical in these types of transactions. And it's interesting. We made a decision a long time ago to stay focused in telecommunications when some of our peers, some of whom are larger, decided to take a more diversified approach to the services that they offered. And one of the benefits of a concentrated focused approach to an industry is we were around when Frontier bought assets from Verizon in 2010, created some opportunity then. We were there in 2016 when they closed a CTF transaction of Frontier to Verizon. And we are a large supplier to both of them right now. And so again, having that historical perspective and the ability to grow with customers as they get larger, we think, has been an important part to the growth story at the company.

When you think about that as well, I mean, is that kind of the same characteristics you're seeing with the Uniti and the Windstream merger again, too, of...

Yes. That's a little bit different in that the companies were together and then they were separated and now they're back together. I think what's encouraging about that transaction as well as a number of others that have been announced this calendar year, is that the ostensible purpose behind the recombination in part is to facilitate another 1 million homes of fiber that were they to remain separate, they would not have decided to do. And so again, there's just been a series of strategic transactions this calendar year. And when you go through all the details, which we're happy to go through, the net result, the headline that comes out of them is moving from the potential fiber passings to the announced fiber passings in each one of those transactions. So if you look at the T-Mobile -- Lumos, T-Mobile, Metronet JV announcements, the Windstream recombination and then the right speed refinancing. The net of those, if you add them all up, is about 9.7 million incremental homes of fiber which is not quite as large as when AT&T made their commitment as a result the FCC approval of DIRECTV, but pretty close. And that was big news. And now it's just a series of announcements.

And then maybe one question that we get a lot is, I mean, with T-Mobile kind of jumping in here with Metronet and Lumos is, do you talk to them? Are they a customer right now or...

We certainly work with both Metronet and Lumos and are encouraged by incremental capital coming into the space. And so I think that's a good thing. We do some work for T-Mobile, but not historically has been that big a relationship.

And then maybe just talking about -- I mean, you've reported earnings late August. I think one of the biggest questions as well that we've gotten is 3Q guide maybe came in a little softer than the Street was expecting. Could you maybe just talk about some of the puts and takes that are associated with the 3Q guide, the year-over-year lapping of some onetimers as well, too.

Yes. So that -- let's start with the last one. So a year ago, we had a closeout in a change order that resulted in about $26 million of onetime revenue. We called it out a year ago. Be nice if we could have one of those every third quarter, but it wasn't going to happen. And so we -- on our adjusted numbers, we backed that out. So that's number 1. Number 2, we talked a little bit we're giving guidance in August, and there was a hurricane that kind of meandered across Florida and up the East Coast, that certainly created a little bit of weather pressure. We also talked a little bit about, and I think this is a good thing, but there has been a little bit of effect on some of the smaller customers that as the cadence of BEAD approvals has picked up at the state level that if your state all of a sudden, you wondered if they were ever going to get approved for BEAD and now they are approved, that you're starting to factor that into how you think about your capital plan over the next 12 to 24 months. And that doesn't mean they stopped anything, that just meant that at least they had an alternative source of funding for what was next and so a little bit of slowness there. And then in general, there's always going to be oscillations from customer to customer. Somebody had a stronger first half this year as opposed to a stronger second half last year. And so you get those kind of year-on-year comps that could influence the number. I think what we always try to remind people when we're working through those kind of shorter-term trends is don't lose sight of the big trend. And so we've talked. If you look at the 10-year period ending January of '24, and we're January year-end. So that's why do it that way. And you say at the beginning of that period -- for the year ended January of '14, company had a little less than $1.8 billion in revenue. We finished last year at $4.175 billion. We had under -- a little under $200 million of EBITDA, grew it to over $500 million in EBITDA. Organic growth of 7.5% CAGR and 9.2% on total. And here's what's really important about the equity. We ended up that 10-year period with about $715 million of net debt. But in that whole 10-year period, we repurchased $650 million of stock. So when you can have that kind of operational performance and then basically generate enough cash flow to either liquidate your debt or liquidate some portion of the equity claims on future growth. I think that's a pretty strong business. No matter if it was a little rainy in August or some client was a little slower in the back half than the front half, right? Don't loose the thread on the big picture.

And then maybe just thinking about the third quarter, we had AT&T this morning, Pascal was on stage with Dave. And they called out that there are some workers that are on strike right now for fiber installations. Will that have any read-through to the third quarter for you guys of -- until there's people in the field.

They self-perform the installation work. So no direct impact. We've been through these things before. They tend to get worked out and whatever didn't get done during the period of time that they were out has to get done pretty quickly. And so we typically see work pick up a little bit when they get it settled, I'm sure everybody is working hard to make that happen.

Okay. That's good to hear. And then maybe just when you're thinking about kind of the -- I know you -- we talked about kind of the 10-year period here. But when you're thinking about the second half of the year and the ramp into calendar '25, fiscal '26. I mean one of your competitors has come out and said that wireline for them, which is a little smaller or much smaller than you guys have, but...

And defined differently.

Yes. It's very robust right now. That demand environment is very robust heading into the next year. I mean, any comments that you guys have on that just as we get into '26 and what that backlog could start to look like.

Yes. I know David is good at this, but we generally don't get people to -- we don't like people talking about us, and we don't talk about them. So good try. I would say there's plenty of demand out there. We have a broader footprint than anybody else. And I can assure you, we have pockets of very substantial businesses when you were comparing them to the market as a whole, who are doing better than average. And we've got other parts of the business that are a little behind the average. And we just don't tend to think about it that way.

Okay. I gave it a shot.

You have an able teacher from what I saw this morning.

I know, yes. The Barden school of questioning. When -- I mean you brought this up before him, but just kind of that 140 million homes.

Yes. And it grows up every year. There's new household formation on top, sure.

So we're around 70 right now, let's just say, 60 in the next 10 years or so. I mean what is the -- once we hit that 140 million, 150 million, what's the opportunity set there? What is that -- is it mainly maintenance work? Is it -- I mean -- can you just walk us through kind of your thoughts on once we hit it?

So a couple of things. So one, and obviously, we've been answering this question for about 25 years. I mean, in the '90s, the question was when first-gen wireless gets fully deployed like nobody is going to need a phone line, right? So we had to deal with that question for probably 6 or 7 years. I think one of the parts that we -- and the way we think about the business strategically is it's all a function of traffic growth. And I saw Verizon was at Oppenheimer here about a month ago and they talked far more eloquently than I can of all the opportunities that they see in the current business for traffic growth to continue, whether that be AI or other applications that a year ago or 2 years ago, nobody even thought about. And the way we thought about the business is as long as traffic is growing at 25% or 30% across a generation. If that continues, there's always going to be upgrade cycles. So whenever think you're done. There's always something else to go to. Tell me what 6G wireless is going require for wireline infrastructure, I have no idea. So one, we've always been surprised. I mean, just one after a number of the long-distance companies fell apart in the 2001, 2002 time frame, I would meet with investors, and they would say, nice story. We used to own your stock, but there's never going to be any growth in traffic again. And even if there was, there's all the dark fiber out there. And here we are at a conference where we're talking installing a remarkable amount of incremental bandwidth to the long-distance network.

That's my next couple of questions, so...

Be prepared to be surprised. The other thing that we've seen and no guarantees here is that as the network has been increasingly fibered, copper is less a portion of the network spend that customers have gone to simplify their supply chains and do vendor consolidation. And so that we have seen over the last 20 years, opportunities to grow market share, where I'm not sure exactly what the CAGR is in the end market demand, but we've certainly seen more demand for services at scale. And so 20 years ago, a customer would have 10 vendors in a state, now we're doing all the work as a single vendor. And I think that all comes about because as they simplify their businesses, they want to take cost out of their management structure that larger vendors that are able to step up to the management role in the process can save them some money. So I think that will be an opportunity, too.

Okay. And maybe just on the opportunity set, I mean, we had Lumen here an hour or so ago. I mean, maybe before I get there, the data center opportunity for Dycom, I think your words were a couple of quarters, it's been modest of kind of those, what, 50, 100-mile connections into some of the data centers. I mean, can you just talk about how that has evolved over the last couple of years or so? And what you're expecting with this kind of new wave or -- I mean, of data center builds, but really just kind of AI that's coming in.

Yes. So -- and certainly, Lumen has spoken to this with a whole lot more wisdom than we have. I guess what I would say is as new data centers have been sighted in areas where there's better power availability. That has created the need to connect the Internet backbone to where those data centers are. And so historically, that Northern Virginia, and there were some areas where there were very dense clusters of data centers as that becomes more distributed, where power is less of a constraint, there are opportunities to connect those data centers with what I'll call fiber laterals. And that's an okay business, but I think what we've said is that's modest, 30 miles here, 100 miles someplace else. When -- what was very interesting about the Microsoft Lumen Corning announcement was that it appears that there are real aspirations to create much broader incremental additions to the capacity of the Internet backbone. And from what we understand. And again, we're not the experts. In part, that's because as AI moves from training to inference, you've got to reduce latency for those applications to become really effective and that's going to be both in the intercity network but also in the metro rings. And I think there have been a number of our customers, again, who spoke at conferences where -- who have large metro ring footprints, where I think that's an opportunity, and that's something that we do every day as part of our master service agreements for those metropolitan areas. And I think that's also -- and as that develops, I think that's a good opportunity.

Maybe just a little bit more of -- I mean, Chris was on the stage saying it's going to be a 2- to 4- or 3- to 4-year build for Lumen, the opportunity is kind of...

I heard the question was 3 to 4 and he said 2 to 3.

2 to 3 was it? Yes. Okay.

We are still paying attention.

When you look at kind of the $5 billion that Dave discussed saying that is on existing conduit and then the $7 billion opportunity, which a good portion of that is new routes. I mean, can you just talk about the differences in that sort of work that Dycom could potentially do for them? I know Lumen is your second largest customer so.

Yes. Look, they have an amazing asset in these incremental conduits that they placed 25 or 30 years ago. and then the demand was not initially there, right? So it's a great asset for them. We participated in the long haul or long distance network built out in the '90s, was a minor part of the business, bigger for some others, but we have experience there. And everybody would always rather install an existing conduit than to go out there and have to build new routes. Just from a permitting, one of the issues around long-haul routes is when you're going between city to city, some of the environmental conditions you're dealing with are very different in that part of the network compared to on paved roadways, right? So if you think about getting a fiber across Louisiana, there's going to be a little part of it [indiscernible]. I mean, that's just kind of the way it goes. And so I think anybody who has the asset that they do is in a great position to be very effective for customers as an option. It's good work. It's something that we talked about on the call that, that type of work, we're in the process of doing a couple of thousand miles over the last 2 or 3 years. So we understand the business. And it's certainly much more cost effective than building new routes.

So maybe just switching topics and thinking about -- I mean you brought a BEAD beforehand, but maybe before we jump into BEAD. Can you just -- I mean, can you talk about how much ARPA dollars are still being deployed, how much RDOF dollars that you're seeing from customers being deployed in some of these rural areas?

Yes. I don't have the exact numbers. I think there's still about $1 billion of TAM that ARPA has yet to disburse. But it's interesting about 3 months ago in Pennsylvania, there was an announcement that both Verizon and Comcast had secured about -- enough with their own capital to deploy something like $250 million or $280 million of rural broadband using ARPA dollars. And what's interesting about that to tie it into BEAD is that the BEAD allocation for Pennsylvania is a little bit under $1.2 billion. So here you are before you even get to BEAD, you're still working through ARPA, which was passed in April of '21. And you've got $0.25 billion plus that's going to be deployed even in anticipation of the BEAD process. And so I think there are analogous situations all over the country where our view of kind of BEAD is that it's a continuation of an increased level of government subsidy to rural broadband that clearly became much more important as a result of COVID than it was before. It was always on everybody's radar. There was $6.5 billion of stimulus money in 2009 that was dedicated to rural broadband, but I think there's much more of a political and cultural commitment that just says we're not going to have rural America at a significant bandwidth disadvantage compared to the rest of the country. And so it's interesting. The original RDOF FCC funding was $21 billion. They've issued about $6 billion. And who knows where that $15 billion will be needed subsequent to BEAD. Maybe not, but certainly always possible.

And then just with BEAD, I think this is another question we get is, I know it's a bipartisan bill, but I mean, do you see any risk if there is a change in the White House in kind of November here of some portions of the program changing?

I mean, the consultants that we use in D.C. tell us that there's broad bipartisan support for it, that if you're a red state governor and you're running for reelection in '26, you'd like to have some accomplishments that you can talk about in rural America or rural portions of your state. And so at best, there may be some changes if there was a change in administration around how the program is managed on an ongoing basis. But we don't expect any significant changes based on what we know at this point.

Okay. So I guess where we are today, 39, I think, states that last time we checked have received Phase 2 approval from the NTA that's around 60% of the BEAD money. I mean you noted on the conference call in August that 3Q of next year start to see some sort of impact. So I guess, what have your initial conversations been like with customers in terms of their deployments. What are you kind of expecting on, I guess, cost for some of these rural builds as well, too?

Yes. I think we've had preliminary discussions with a number of customers. But until the volume 2s were approved and you go to the subgrantee process, that's where it becomes real and so there are Montana, Louisiana and West Virginia are now out with their subgrantee, the windows are open. And I think we look forward to seeing what kind of a market develops for who's interested in what states and at what terms. And I don't think it's going to take all 50 states to go through that process before we know. I think probably after 10 or 15 states, it will be pretty clear where the market is and where programs are successful in that subgrantee process and where some may need a little bit of rethinking. But remember, there are statutory and regulatory required calendars that once you get approval and you go through the subgrantee process, you've got to start seeing activity within that next 12 months.

And I guess just -- I mean you noted beforehand of $15 billion or $6 billion of the RDOF money has been spent. I mean...

No, allocated. They've all been spent.

Yes. I mean -- when you think about the runway for BEAD though, I mean, $42 billion, that's a -- and how...

And it required 25% match.

Yes. So how -- I mean, how long does this last? I mean where do we...

So one of the interesting thing that a number of the customers have talked about is because there are all these interrelated funding sources that even in areas where -- that are not deemed by the -- by somebody as underserved or unserved, you may enable upgrades along the way to get there of plant that's not even covered. And so that's not only is there all the work that has to be done to accomplish the BEAD spending. But there's also other areas that may not be covered by the BEAD footprint that will now be economic to upgrade, right? And I'm thinking of 1 operator, in particular, that looked at 1 metro area and looked at RDOF, looked at state funds and looked at ARPA funds, and there was really a mosaic and when you -- or a puzzle, where you put all the pieces together, all of a sudden there were more economically viable passings even in areas that weren't covered by a government program. So it's a lot of work. And these programs are always eagerly anticipated. One may say over anticipated, but I think generally underappreciated in terms of their long-term duration and impact on the industry and hopefully on us.

And maybe before we switch topics here, just in terms of rural builds, I mean, is there a general rule of thumb that you kind of point to for cost per passing and...

It's very interesting. So one of the things that we're looking to see as the subgrantee process begins in earnest across the country is to the extent the process is attractive to the incumbent operator. They certainly have an advantage in terms of getting started because they have existing aerial facilities that they can attach to. They have pole attachment agreements, they've permitting relationships with the county DOTs, usually in rural America, the county DOTs is a very important relationship as well as they typically have central offices and transport fiber. So if -- to the extent that they're actively involved, I think you get started quicker than the alternative but it will also be a little less costly because they have all that existing infrastructure that they can build on. To the extent that the incumbents have other priorities and somebody who's not an incumbent steps in, it will take longer to get started because you will have to negotiate pole attachment agreements and get together with the permitting agencies and you'll have to get middle mile support for the builds. So it will take longer, but it will obviously be more costly because they won't be able to piggyback on existing infrastructure. So it's going to be interesting to see what the blend is. That's -- I really do think there will be a market that develops state by state that will be an intensive kind of geospatial modeling process to see where it works for the incumbents and where it may work on a more attractive basis for somebody else.

Okay. And then when we're thinking about kind of the third quarter here, you guys made a big splash with the Black & Veatch acquisition, wireless construction operator out West. I mean could you maybe just talk about the acquisition...

Well, also here in New York too, New York and New Jersey.

Yes. And so can you just talk a little bit about that acquisition and kind of what you're expecting from this business now being under the Dycom umbrella?

Yes. As we talked about on the call, we've had a good wireless business for now over a dozen years. And we had invested substantially in systems and capabilities, and we were looking for an opportunity to leverage those capabilities across a broader footprint. This opportunity developed. It's -- nothing is ever easy, but at a high level, because we had -- share a common customer, we already have the systems in place to do the integration. And so that's really where the opportunity for us is to be able to grow the top line and then leverage the capabilities that we've developed individually across a broader footprint to create good returns. The other thing we talked about on the call is that at a relatively middle of the organization level. Oftentimes, we're working for the same person for one of the customers on the wireless side as we are on the wireline side. And so it's really just an ancillary service to sell into the same day-to-day customer. There are also some things that I think we can bring given our wireline capabilities. We've always done lots of small cell work, part of O-RAN initiative is you've got to change out small cells, too, like we bring capabilities to that footprint that may create more opportunity than the assets that we acquired. So a little bit of -- hopefully some revenue synergies by coming together also.

And -- I try to think the best way to ask this of. In terms of...

I think you just spit it out around here. That's what I've been observing while I've been around.

In terms of the -- you noted the wireless piece of the pie is a smaller portion for you. And we had all the tower companies here yesterday and today, and it sounds like there is starting to be those initial conversations of more deployments [indiscernible].

Well, certainly, we're in an aggressive site acquisition phase for O-RAN deployments. I mean when you're doing an equipment change out, you've got to go through and you've got to get site-acq on thousands of towers. So certainly, that's the case. I think because of the commitment and the predictability of the O-RAN equipment change out, I think densification of that comes during or at the end of that's just upside to the model that we put together.

Okay. So that -- I guess, that densification conversation, those new deployments that's, I guess, just ongoing and...

Yes. It's always part of that industry doing sector splits and creating more capacity and density that's just what you do in that business. And so that's always there. This is a significant equipment change out. It's never been done before at this kind of scale.

And if we -- with the last 10 minutes here, so maybe switching to margins. you've worked through the last couple of years of improving margins. I think you've noted on the call a couple of quarters ago that you worked through a program that was a drag on margins. I mean, any other programs that you're seeing a drag on margins right now and maybe where could we see the margin profile go for Dycom?

Yes. So we've always talked about kind of mid-teens EBITDA as kind of aspirational. It's usually a function of good operating leverage across to the majority of the customer set. So there have been periods of time where we had decent organic growth, but that meant the top 2 customers doubled and everybody else declined 10%. That does not give you good operating leverage across your cost structure, right? And so good broadly distributed growth is, for the most part of a prerequisite that those kind of margins. And when you look at the strategic transactions that we've been talking about, if you look at BEAD, you look at our wireless -- increased wireless exposure. We're certainly trying to set ourselves up for a broadly distributed growth without a doubt. I think the other thing is, and it sounds trite, but I think it's true is that outside of an occasional program -- and I mean occasional, we really had not had to call a program out in a long time, prior to the last one. There are always things we're working on to improve the business, right? Things that we can do with our scale that it would be difficult for others to replicate that don't have our scale. I mean I always give the speech to the folks that says, "I want to get to the point where our strategies are so scale enabled that if everybody knew exactly what they were, they still could replicate them." And we're not there yet, but that's where we'd like to get. We can always become more efficient in the back office. I mean there are advantages from scale there. We have common systems, but we can always have better common systems. One of the advantages that we've had in our integrations is being able to take single integrated systems and deploy that to replace disparate systems that have developed in businesses over time. And again, that's because we're investing at a level in the business that, again, is hard to replicate without our scale.

And on the costs, I mean, are you leveraging kind of any sort of AI with people with -- people out in the field and -- can you maybe talk about some of those [indiscernible].

So I mean it was -- we were trying to improve our use of technology to control field operations long before anybody talked about AI. But specifically, with respect to AI, we have a business where we receive something like 180,000 to 200,000 locate tickets a week, and they're all textual instructions to the technician. We acquired a natural language model to read those tickets and risk assess them by the machine about 5 years ago. We have another prosaic application where we did machine learning that can look at images of excavations and determine the depth of the cable to make sure that we're delivering what we've committed to in the specification with the customer. We have a large technician business in California. And not surprisingly, traffic in L.A. is a huge management consideration for timeliness and efficiency and we put an AI fleet routing system in and took about 8% out of travel time without changing anybody's behavior just using real-time inputs for where traffic was congested and routing around it. So -- and that's just scratching the surface. There's a whole lot more that we are working on and can deploy in the business.

And you brought up the locating business. I think maybe one of the underappreciated aspects of the Dycom business is the locating on top of kind of these maintenance programs that you provide to your customers. I mean could you just talk about the recurring nature of these a little bit more and provide...

Sure. Yes, I think It's always great to talk about the big fiber programs, but this is just get another development. We have had businesses that have provided master services work to customers in geographic areas for 70, 80 years on a continuous basis. And so the core of the business is that ongoing relationship programs come and go, but that maintenance is always going to be there. With -- there's always road moves. I mean even on projects that we did 3 and 4 years ago across major metropolitan areas, we're getting calls now that says we've got 30 projects to do road relocations or new bridges or for some reason, the path of the fiber has to be altered to comply with some change in environment. So we always have that. On the locating business, I mean it's an interesting business, a little bit like wireless and that we sell that -- about 60% or 70% of that business is sold to phone and cable companies. So again, it's an incidental or ancillary service to what we do. And the shelf life of those tickets that we respond to is anywhere from 2 weeks to 30 days. So I guess it is the definition of recurring payments because as long as we get a new call at the same location if it was 2 weeks ago, we got to go out and do the work again. And it's something that, again, we invested in lots of technology to improve that business some of what we're applying now in the construction side of the business are lessons that we learned the hard way in the locating business about how to control the workforce and minimize travel time. So I think there are lots of opportunities to deploy more technology in the business.

And then when we talk to the data center operators, there are a long lead times right now for the supply chain. I mean, are you seeing any impacts to your business on the supply chain or even from your customers getting equipment as well for some other fiber builds?

In our business, the biggest effect was really on the automotive side. I mean there just was a while where it was very difficult to get equipment. People forget there was this concept of the golden part. We'd order a $300,000 machine and it would be ready to ship except for $127 part or whatever it was, right? And so that seems to have -- that constraint is relaxed. So we're in good shape in terms of what we need to buy for capital equipment. We're going to spend well north of $200 million this year. So we feel good about that. And there was clearly a cable and associated equipment and devices issues in '21 and '22. I think for the most part, that's been worked through by the equipment suppliers and the cable suppliers. So today, we're not -- there's always a shortage some place, usually because of some logistical snafu. But in terms of overall, the market is in pretty good shape.

And with that $200 million, I mean, is that ratable. Should we think about that kind of the next couple of years of being in that neighborhood or these programs that could be -- it could increase.

Yes. So typically, we own our fleet, we don't finance it and we sell it on a pretty regular basis. So if you see the other income line that relates to the gain on sale of those assets. But typically, we're going to spend as we're growing about depreciation. During the pandemic, we couldn't because we just couldn't get it even earlier this year, there were some things that we couldn't get. They're coming in now. And we feel good about that as a run rate. One of the advantages that approach made evident in the pandemic was we have always aggressively replaced equipment a little bit early because we could get good resale value from the marketplace and when we couldn't get equipment in the pandemic, we were able to extend some useful lives without any significant increase in maintenance costs. Now we're trying to get back to the more traditional approach. And so we're spending a little bit of money there.

Okay. Perfect. I think that's a great place to leave it. Steve, thank you so much. Really appreciate it.

Good to see you all. Thank you.