Equinix, Inc. (EQIX) Earnings Call Transcript & Summary

[Audio Gap] Disclosures are available at the registration desk. For those of you joining us via the webcast, I'm Mike Rollins, and I cover the communications, services and infrastructure categories here at Citi Research. And I'd like to welcome back Charles Meyers, President and CEO of Equinix. Thank you for joining us today.

Always good to be with you, Mike.

Well, as customary, as it's become for the conference, let's start if you could talk about your strategic and your operating priorities for the coming year.

Sure. Before I start, I'll do my duty and read our disclosure, which is some of what I'll talk about today contains forward-looking statements, so please read our riveting SEC filings for more information about factors that could affect these statements. So I think the priorities for 2020, in many respects, don't look dramatically different than what we've articulated as we moved into 2019. I think it starts with people, as it always does. And so I think continuing to invest in making the Equinix culture really a competitive advantage and stimulating our ability to attract and retain and inspire exceptional talent tops the list. And then from there, I think investing in our platform and continue to adding services that make us relevant to helping customers on their digital transformation journey. So I think -- we put Sara into the Chief Product Officer role and are going to continue to invest behind that team. And then the go-to-market side, I think expanding our go-to-market engine, I really think that we are not at all demand limited. I think that we are -- I think there's tons of opportunity out there, and we're seeing great success in terms of customers resonating with our value prop. So I think we're going to scale our go-to-market engine, and that includes a focus on partners. We're driving significant book -- portions of our bookings through channel now. And so I think that's going to be an area of focus for us. And then also driving operating leverage in the business because we've got to fuel those investments somehow, and I think we would like to not do that by compromising on our margins. And so I think we've got to continue to drive operating leverage. And then the last area I mentioned, which also is a carryover from last year, is scaling or advancing the xScale. We really brought the first one out of the ground in terms of the JV in 2019, and I think you're going to see a couple more of those coming up in 2020.

Well, great. There's a lot to unpack there. I think maybe starting at a high level, one of the questions is when you look at the business, what are your leading and your lagging indicators that you look at to measure the durability of your retail colocation model? And whether you're looking at it as the portfolio as a whole, you're looking at the market level, you're looking at a facility level, but maybe give us just a sense of how you're evaluating the strength and resiliency of the business model.

Yes, I mean, I think it's easier to talk about it at a macro level. Obviously, there are sort of market-level dynamics that we look at as well. But if I look at it and just tell us -- sort of helicopter all the way out to the market level, as has always been the history of Equinix, really the drivers and I think the leading indicators for the health of the business and the demand opportunity that we think is going to be in front of us really revolves around data and the creation of data, the movement of data, the manipulation of data through applications and the storage of data. And if you look at it, all of those things literally across the board, I think, are seeing significant tailwinds behind them. You look at what IoT is doing in terms of just the strict amount of data that's being created. And then once that data is created, it's the need to be moved. The provision -- bandwidth has always been something we've looked at, right, whether that be Internet bandwidth or private bandwidth. Our global interconnection index looks at more interconnection and private-related bandwidth and connectivity, all of which is continuing to set new records and grow at an accelerating rate. And then the applications and the ability to act on that data and how that's intersecting with clouds and how people are storing and using data, all of that continues to be, I think, the fundamental drivers for our business and represents, I think, an incredibly healthy demand environment for the business.

And how would you frame the long-term organic growth potential of this business model?

Well, I mean, it's actually probably hard to frame it with precision other than to say I think there is plenty of opportunity for us to continue to grow this business even though it's now quite large, at a very healthy rate. And I look at it and say nothing in my mind that would preclude us from doubling in some reasonable time frame and doing so with very healthy economics, strong returns on capital, healthy margins, et cetera. And so I think the organic -- and I do think, by the way, that is -- any consolidation activity is over and above that. I think we can achieve that organically. And so I think the demand profile, the relevance that we have in digital transformation on the enterprise side and I think the role that we play in service providers seeking to serve those enterprises all combines to be a very healthy backdrop for the business.

When you say doubling, does that include hyperscale? Or what's...

No. No. I think that's separate. Again, that's not even going to be -- it's not consolidated for us. And so really, the -- we'll see some AFFO juice from that. But I think independent of that, we're going to -- we're in a position -- and again, we'll give you guys more precision and -- when we talk -- when we come together for Analyst Day. But I would say in the context of this question, which is what's the long-term opportunity and backdrop for the growth profile of the business, I think if you look at it and say, "Hey, can we just -- can we be -- we're currently a $5-plus billion business. Can we be $10 billion?" Sure. I think there's plenty of opportunity out there to be that.

And get there through development and utilization ex acquisitions, ex hyperscale?

Right.

How much of that depends on value-added services versus the traditional colo and interconnection?

Well, what I would say is that I think it depends on our ability to evolve the way people consume our value proposition. And so I think it's going to be a blend of things that look like traditional colocation and interconnection and things that deliver our same underlying value proposition, which is global reach, access to scaled ecosystems, a comprehensive portfolio of interconnection offerings and a track record of service excellence. Those are what people want from us. Traditionally, we have put them through the wicket of buying colocation to access those things. Network Edge is the first time we've delivered a service that says, well, you can get some of those things without needing to even be a colo customer. And that's a big change for us, right? And so I think that we'll -- colo is here for a long time. So I'm not sort of signaling the depth of that by any means. But I do think that we also need to be more responsive to more adaptive, flexible, agile ways for people to consume our underlying value proposition. And if you look at our edge services aspirations, it really reflects that.

So I think we have our first live survey question, if we could pull that up, please. As we've done in the past, we ask about the organic constant currency revenue growth for the coming years. So this is 2020. And you're welcome to vote. We'll ask, first, 8% or less? 8% -- over 8% to 9%? Over 9% to 10%? Or over 10%? And we'll go to our audience and see how they're thinking about this. [Voting]

This is always helpful for us, Mike. And I'm not going to vote.

You're not going to vote. You're not going to weigh in on this one.

I'm not -- no, I don't trust that you haven't rigged my unit here, so...

We really haven't. So choice 1, 8% or less, 17%; over 8% to 9% is 63%; over 9% to 10% is 6%; and over 10% is 14%.

Interesting distribution.

Yes. So as you think -- now the long-term growth rate guidance, if I recall correctly, is 8 to 10 annual?

Correct.

8 to 10?

Yes.

And so I mean, there are headwinds in 2019 that held back growth. You talked about Verizon, cross-connect migration...

[ 10 to 100 ] migration, right.

And some self-induced optimization. How do those play into now 2020? Are those now in the rearview mirror? How does -- and does that help the aggregate growth for 2020?

I think all of those things will dissipate at some level or sort of begin to taper during the course of 2020. And so I think there -- but I do think that we have a continued -- I think there is a -- there has to be a recognition that if you look at the industry as a whole, inclusive of our role in it, we are still very much under -- in a period of transformation. If you look at 8 years ago, there was this -- the general rally and cry was, hey, 90% of all enterprise workloads are in enterprise data centers. They're all going to go to colocation. It's a beautiful thing. Nobody wants to build data centers. Well, then the cloud intervened. I think that people then said, "Well, wait a second. Is colo relevant?" And I think that -- so we had to bring people around to, "Well, at least some of it is." And I think the notion that -- of hybrid and multicloud as the architecture of choice has really taken hold now. And I believe that we're in a very strong position in terms of that being the preferred architecture. But that does mean that people who have workloads that are currently living either in enterprise data centers or in a colo, many of which might migrate to the cloud over some time frame. And so that creates some underlying level of churn. Now we tend to be a little less exposed to that than most because more commoditized workloads that might be well suited to being operated in the cloud probably weren't landing with us anyway just because of our general market position and price points, et cetera. So I think we're less exposed, but there is some of that underlying bubbling of the movement from IT architecture, 1.0 to kind of where we are today. And so I think there's a set of puts and takes, but I think the macro demand profile for the business continues to be very good. And if we can get new services with reasonable attach rates, I think that's upside opportunity for us, but I think that's a multiyear phenomenon.

And when you think about the growth of the core business, what kind of impact is enterprise now having? And how important is enterprise to the success of performance in the coming year?

It's very important because I think that we're -- inherently, we talk a lot about ecosystems as part of the -- and the enterprises are, in many respects, the buy side of that ecosystem, right? But of course, they fuel the people who want to come to sell to them. And so the service provider side of our business is equally dependent on the enterprise, right? And so I think it's very important. And if you look at how our growth has sort of moved over the last several years, enterprise and cloud, in particular, cloud and what we call cloud and IT, have over-indexed relative to the broader business repeatedly for probably now, I would guess, 12 quarters in a row. I haven't checked that stat, but it's -- that's -- and I think that's going to continue to be the case.

Pull up the next survey, if we could, please. So another question that I would like to ask is what impact will the cloud have on Equinix's business model over the next 3 years: net positive, neutral or net negative? And we'll go to our polls. [Voting]

I'll vote on this one, Mike.

We might have to go back now and get your other vote on the... So 75% net positive, 18% neutral, 7% net negative. I mean have you seen a change in the visibility of what multicloud and hybrid cloud is going to look like in your data centers and how they interoperate in the ecosystem?

Yes, for sure. I think that if you look at how we're prosecuting the enterprise market, it's a bit more from a top of the pyramid down. It's not sort of some of these folks who enter at sort of the small to medium enterprise and then sort of build something that they go after the larger enterprises. We really have more of an approach that is talking to large, sophisticated, global multinationals who have -- and there, as we've been working with them and we deploy our solution architects to talk to them about what is their long-term IT architecture look like, how are they implementing and thinking about hybrid and multicloud. And so we see it in terms of how they're deploying. We've seen tremendous success with ECX Fabric. The port count on ECX Fabric continues to grow at a very, very healthy clip. The virtual circuit count, as you saw in our Q3 results, we generated 1,600 virtual cross-connects in a single quarter. Again, that's way more than probably any of our providers provided in terms of their entire cross-connect count probably for the, in some cases, for the year. And so the level of interconnection and what we're seeing in terms of people utilizing multiclouds but then also placing private infrastructure immediately proximate to those clouds is definitely showing up. And they're, by the way, they're doing it globally. I mean you hear us talk a lot about the number. I think it's that 64% of our revenue now is from customers who operate with us in all 3 regions. And if you look at our top 20 customers, they average something like 30 or 40 locations. And so they're very, very highly distributed, very hybrid, very multicloud.

And if you think about the impact that the hyperscale initiative is having and the clouds are having on that hyperscale or HIT initiative, what have you seen in terms of the follow-through? Are you seeing retail follow-through, which I think was one of the ambitions of the strategy?

Sure. I think that, one, I think it's very important to reiterate that point, right, which is our hyperscale strategy, and we're kind of slowly working the HIT terminology out of the lexicon, but our hyperscale strategy and the xScale offer, which is part of that hyperscale strategy, was really a means to an end. It wasn't a, "Oh, that's an incredibly attractive business. Let's allocate our capital towards that." In fact, as you know, the whole motivation for getting into that was to say we have to relieve the pressure on our balance sheet. We think that's an interesting business with a very deep demand pool, and we think there is capital that we'll want to follow it and participate in it. And that's why we decided to go with a partner. And I think that -- but the objective was to support and extend and augment our position in the overall cloud ecosystem. And so I would say it is paying dividends in that the people that are shaping the cloud ecosystem are excited that we're in the hyperscale space. They are actively sharing their demand profile with us in terms of where they're expanding, when they're expanding, how much capacity they're going to want. We built a very large funnel of opportunities. We're buying land to respond to those opportunities off our own balance sheet even before we get the JVs established. And so we're seeing good momentum. And I think it just is adding to the credibility that we have with enterprise customers of saying, if you want the flexibility to have your architecture -- your IT architecture be future-proof and have flexibility to use the range of digital partners that you want, cloud providers, network providers, SaaS providers, et cetera, Equinix is the place to be. And I think it just adds to our story. And so I think we're at least seeing the results in terms of defending our position as the leader in the cloud ecosystem.

And you mentioned that you're thinking of expanding in 2020. You mentioned there could be a couple of opportunities for more JVs. What regions or countries might that occur with...

Yes, we haven't given any precision on that other than to say we're actively pursuing the JV in Japan. I mean we talked about that on the earnings call -- the last couple of earnings calls. That's progressing well. That will include assets in both Tokyo and Osaka. And so I think -- I'm not putting any time frames on that because we learned through the one last year that the complexity of these things and the local complexities that come with them are often can be different than we anticipated. So -- but I think during the course of the year, we'll see that one come out. And then I think there are other opportunities in both Asia, potentially in Latin and South America as well and potentially others that we're evaluating as well. So I won't give any more precision as to number and timing, but I do think there's plenty of opportunity for us there.

What are you seeing on the pricing front for your business as well as the adjacent, whether it's the wholesale side or the xScale side of the equation?

Yes. What I would say -- generally, I would say we're finding the pricing environment to be pretty stable and favorable, particularly in the retail business because I think that if we're doing a good job of targeting the customers, the workloads and the use cases where Equinix has a clearly differentiated value proposition, then we're able to sustain favorable pricing. We've said repeatedly on our calls that generally, quarter-after-quarter, we have positive pricing actions, which means that the contractual price escalators that we have in our contracts are outweighing any re-rates that we have in the business. And so that puts us in a very, very good position is, I think, a reflection of the strength of our value proposition. And I -- but I also -- I'm seeing, I think, a little bit more stability in -- even in hyperscale and wholesale. I think there's pressure there only because it's a less differentiated offer. I think wholesale and hyperscale are both things that people with the right amount of capital and a little bit of elbow grease can go get done, as evidenced by the fact that in hyperscale, a significant amount of the capacity sold over the last couple of years was sold by companies that didn't exist 5 years ago or 4 years ago. And so that, I think, is a reflection of lower barriers to entry, right? And so -- but having said all of that, I think that we're seeing a little bit of a cooling in terms of the mania of private money flowing in. I think people are waiting to see kind of -- what kind of liquidity those kinds of assets are going to have and at what price points. And I think that's all good news. I think as a market leader in this space, I think we would love to just see a rational environment. But I do think, overall, I would say that even in the environments where -- that are more heavily commoditized, I think we're seeing a stable pricing environment.

What's the risk of wholesale and hyperscale pricing just slides further downward?

Yes, you've heard me maybe use the term price pollution before, right, which is -- and I think it is a reality to some degree. But what it -- the way it really impacts us, I think, Mike, is more in extended sales cycle. Because what we have to do is say -- because somebody says, "Well, wait a second. I can get that for a lot less over here." And we -- and you have to say, "Well, it's not the same, and here's what you're going to -- here is where you're going to fall down if you decide you're going to go do that." Sometimes it actually requires them to go and fall down, and we have to say, "Hey, we're not going to compete for the business at that price point." And they say, "Great, we're going to do it somewhere else." And then it doesn't perform the way they want or whatever, or they -- or again, they tier their architecture and they say, "This stuff works fine here, but this stuff doesn't." And we say, "Well, we could have anticipated that." And then we repatriate some of those back into Equinix. So I think it's less about us because as long as we don't capitulate, at the end of the day, we price our product, right? And so -- and I think we've had pretty good discipline over the years that says we're not going to chase growth for growth's sake. We're going to price, and we think there's a big enough addressable market, but it is missionary selling still.

Let's bring up the next survey question, if we could, please. So what is the biggest threat to the financial results of retail-centric data centers over the next 3 years? So we have a few different choices here: competition, changes in cloud architecture, tech impacts to interconnection, greater power density, higher SG&A or no immediate threat. [Voting]

And the results are 28%, competition; 21%, changes in cloud architecture; 21%, tech impacts to interconnection; 10%, greater power density; 14%, higher SG&A; 7%, no immediate threat.

Interesting.

Did I miss anything, by the way?

No. I mean the way -- I answered number two, by the way, but -- because I think -- and especially when I take this, the way the question is presented, which is more of on a macro sector, not just an Equinix perspective, because I think that there are retail -- more commoditized retail players who, I think, face much greater exposure to workloads that currently live inside of their facilities simply moving to the cloud. And so I think that, that movement -- but that's what creates the competitive intensity because if that's happening, then they're going to try to refill that, and their biggest lever they have at their disposal is price. And I think that leads to potentially a negative spiral. But I think that's more in the case of people if you're -- if you don't have a compelling differentiation for why somebody would keep their workload there versus simply put that into the public cloud, which I think is where the super majority of workloads will end up, I think that's where there's going to be pressure, and we'll see how quickly it manifests.

In the competitive landscape, are you seeing any changes with consolidation? There's the proposed acquisition of Interxion by Digital. There's been other consolidation over time. You've been...

A player there.

Yes, driving some of that. So what do you see in terms of the competitive landscape, whether you look at it globally or you look at it by region or market?

Yes, I mean, you always have to be careful of this question, I think, because it almost feels ridiculous for people if you say, "Well, we don't have competition," which isn't true, and I wouldn't answer it that way. But what I would say is I think there are not a lot of people who can compete effectively in terms of the sweet spot value proposition that we provide to customers. Some of it can do -- some people can do it on a more regionally specific or locally specific basis. If you look at players -- and actually, Interxion is a pretty good example. We have great respect for Interxion as a competitor. I think they have a business that, in many respects, mirrors ours from a value proposition standpoint. They may have been a little more active in terms of large footprint over the last couple of years, but the core of their business still remains a more interconnection-centric, ecosystem-centric business. But of course, that is limited to the European theater. And so I think the competitive landscape for us is a few -- you have CoreSite in L.A. They have a great franchise in L.A., right? I mean if we'd just be pragmatic about it and say because of their ownership of One -- or their position at One Wilshire, they don't actually own it, but that gives them that sort of Equinix-like value proposition in L.A. So we have to use other levers to compete in L.A., right? And so I see it as more pockets of competition that can provide some of that relevant. But I just don't see a global scale and scope player that has the kind of reach and the kind of capability that we do to do that on a comprehensive basis.

So if we go back to some of the priorities that you laid out, you talked to the go-to-market engine.

Yes.

And so incrementally, what might be different in 2020 in terms of leveraging partners? You mentioned that's a focus for you in terms of the types of sales you might be going after differently this year than last year.

Yes, I would say that for the most part, I don't think it's radically different. I think we might put a little more juice behind it in terms of our level of investment because I think we're seeing such confidence in terms of the bookings trajectory, the really strong mix of business. The response of the customer is such that I think we're -- we feel good about investing behind that. And that's going to mean more feet on the street, more quota-bearing heads. It's going to be more support infrastructure around them, including global solution architects and the like. It's going to include investing in our channel capabilities and particularly behind the set of high-profile partners that are really being productive for us. And those are companies like AT&T and Verizon and Orange and as well as our relationship with the hyperscalers themselves who actually are an outstanding sort of channel for us in that they're not really a traditional channel, but what happens is that they and their efforts to sell their own cloud services are every day running into customers who say, "Well, I'm going hybrid. Here's what my needs look like." And in order to not slow down their sales cycle, they need to be responsive to that. And that's why you've seen companies like -- that's why you've seen AWS Outposts. It's why you're seeing -- it's why Azure Stack exists and all these kind of things. And so I think that we're going to continue to invest on the channel side of things. And then the one thing that I do think is going to be a little different is I think we're -- we are seeing that as we bring new services to market, Network Edge being an example and I think there'll be more behind that, our need to probably consider more aggressive overlay-type selling, which is a pretty traditional approach for a lot of companies as they bring new products to market, they don't yet have traction and so they can't get the attention of the sort of main sales team. And in order to sort of get the ball rolling, you kind of need to look at overlay selling, which adds a little bit of cost for sure. But I think it's -- in the early stages of a product, I think that's something we're going to need to be looking at.

And you mentioned investing a little bit more. You also talked about operating leverage and for some time, the company has been balancing growth and trying to work towards your margin goal of 50%. So where does -- where do margins kind of fit in this equation over the next 12 months?

Sure. Same dynamic is very much in play this year as well, which is that is the constant balance that we're looking at, which is we do believe there's a trajectory and ability for us to get to that goal. We haven't time stamped it in any way quite purposely. We are demonstrating operating leverage in the business. If you look at our SG&A line as a percentage of revenue, it has sort of been comfortably coming down or regularly coming down over the last several years, and I think we're going to continue to drive that operating leverage. And we're going to continue to invest in the business in areas that provide operating leverage going forward. I've talked a little bit about things like our billing system, which I think is -- has more complexity and manual intervention in it than it should have today. And those are areas that we're going to continue to augment and invest behind to allow longer-term margin expansion, but they are going to take some investment in the business today. I would tell you, on balance, if I have to -- if I were faced with a Sophie's Choice of do I want to drop -- do I want to give margin expansion or do I want to invest in the business, on balance, I'll invest in the business if I'm confident in the long-term value creation opportunity. But I believe we can deliver both, and that's going to be our aspiration. And our objective is to deliver slow and steady margin expansion but invest in the business. But if I had to make a choice, I'd continue to invest in the long-term growth of the business because I think the market opportunity we have in front of us is substantial.

And when you have these discussions about finding this balance, I'm curious, like when you look at that 8 to 10 long-term annual range you've established, is this a conversation where you're having internally, like, if you give me 10, maybe I'm not as focused on the margin expansion. But if you give me 8, you need to...

Oh, for sure.

So there is a relationship between where you end up in that range and what margin expansion you want to deliver?

For sure. Yes, I mean, we absolutely look at that and say, okay, how are we going to -- we know growth is important. We know growth is something people are looking for. We know that growth clearly factors into people, how they value the business. I do think that primarily, what people want to say is growth on an AFFO per share basis, right? And so that's our kind of laser-focus in terms of a guiding light metric. But revenue growth is obviously a fundamental driver for that as well. And so yes, there's always trade-offs. But I will say that we have really avoided being overly focused on revenue growth because we think it creates potentially unhealthy dynamics in the business in terms of losing your strategy focus and losing your discipline.

As you've had a chance to go through the annual process that the company typically goes through, did you identify any new opportunities or new challenges for the business?

Yes, I mean, I think that we're continuing to -- I think there are plenty of new opportunities. If you look at how people are thinking about digital transformation, how they want to consume services, I think this notion that we have around the edge services portfolio and being able to animate our value proposition through means above and beyond just traditional colo, I think, are things that we think are new opportunities, which is exactly why we're investing in sort of the platform and product side of the business. And so I think there's those opportunities. I think there are opportunities to undertake our own digital transformation to both improve our customer experience as well as drive operating leverage. So those are things that we're going to invest behind. And so I think there's a variety of opportunities that we sort of came upon as we continued our planning together as a group.

And when you look at the capital that you're putting into the business, is the cost to deploy new capacity changing at all?

I wouldn't say very much. I think this notion of is sort of cost per kilowatt generally deflationary or -- I think there's a set of puts and takes, which is, one, continuing optimization enhancement of the design, improve modularity, those kind of things, which help you drive it down. But I also think underlying unit cost in some places are going up in some areas of -- in some areas in terms of the underlying components that go in, but more importantly, labor. Labor is one that is really tough to fully predict and is generally inflationary, and it's a meaningful element of our cost. And in some areas, it becomes quite acute. We're seeing that in Paris today. And any time the Olympics come, all of a sudden, rates go through the roof and it's hard to get anybody to actually come and do work. And so we are seeing that. So labor is the one factor that -- and that is true also of operating the facilities. So labor is generally inflationary, which is why we continue to sort of resist people who are saying, "Hey, we've had great luck in terms of sticking to price escalators because I think there is some inflationary element in the business."

Are you seeing any change in the economic climate based on the decisions your customers are making?

Any change in the economic climate based on customer decisions?

Yes, so based on your customer conversations, your decisions, you have the benefit of being global. Are you seeing any signs of economic weakness or macroeconomic change based on your interactions with customers?

Well, I mean, I would say yes. I think that -- I think, generally, all the things that we know about and read about are on the minds of our customers as well, whether that be U.S.-China relations, whether that be the eventual outcome of Hong Kong, whether that be sort of the macroeconomic climate and negative interest rates and the impact of those in Europe and a variety of things that we all know about. What I would say, though, is even though those are on the minds of our customers and they are a bit of a topic of discussion, what we have not seen is that those leading to them either slowing down or changing their decision relative to their commitment to digital transformation. We've seen that be a strong and steady beacon that says, "We are going to do it. It's a strategic priority for us. We're going to do it." And I don't know if it's come hell or high water, but it certainly is a -- this is something that we're committed to and we're going to figure out how to get it done.

So the towers have talked about the opportunity to try to get edge, compute or storage underneath their sites. Verizon recently announced part of their, what they call their MEC, which basically is putting cloud resources in some of their facilities that are closer to the edge of the customer.

Yes, if you like, AWS Wavelength and what they announced there, similar, yes.

And so with their mobile edge computing efforts. So do you look at this as a complement, an opportunity, a competitive risk? How would you frame this possibility of edge getting further out to the "edge" for Equinix?

Yes, I think that it is -- I think it's something we needed to be very aware of and cognizant of and figure out how it impacts how we execute our business. It's balance of threat, opportunity, et cetera. It's a little hard for me, I think it's far enough off and too sort of murky and opaque to really be fully clear to me because on the one hand, I think, for example, it's -- in fact, if you go back, we had -- in the role before I took the CEO job, I was in this what we call SSI. We created this team called Evolving Edge, which was looking at exactly this. And I think that the reality is -- I believe at some point in the future we are going to want to extend the reach of Platform Equinix to a broader geographic scope for the reasons -- some of the reasons you described. Whether we do that in Equinix-owned facilities or do that in partners or do that in both, I think, are questions that we'll have to answer over time. I would tell you this, though, that the vast majority of use cases we're seeing, the current aggregated edge, which we think Equinix is the best manifestation of, is sort of -- is meeting the vast majority of use case requirements today. In fact, it is a very rare exception that we see people can't get the performance they need by placing infrastructure in 1 of our 54 metros around the world. Now in a fully densified 5G world, will a set of use cases begin to emerge that are going to demand compute and storage and other resources at a further geographically distributed point? I think the answer to that is yes. And exactly how we play, it's not totally clear to me. But I think it is quite possible that, that would be in partnership with companies who with whom I think we have a certain level of shared and aligned interest and whether those be tower companies or carriers. And we have great relationships with carriers, for example. And so -- and I think that a neutral party playing in that space like, well, who we are, is a really likely outcome. So it's a space that we're watching very carefully.

Charles, thanks for joining us today.

Thanks, Mike. Appreciate it. Thank you.