American Tower Corporation (AMT) Q2 2013 Earnings Report, Transcript and Summary

Greetings, and welcome to the CoreSite Realty Corporation Second Quarter 2013 Earnings Conference Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Derek McCandless, General Counsel. Thank you. Mr. McCandless, you may begin.

Thank you. Hello, everyone, and welcome to our second quarter 2013 conference call. I am joined here today by Tom Ray, our President and CEO; Jarrett Appleby, our Chief Operating Officer; and Jeff Finnin, our Chief Financial Officer. As we begin our call, I would like to remind everyone that our remarks on today's call include forward-looking statements within the meaning of applicable securities laws, including statements regarding projections, plans or future expectations. These forward-looking statements reflect current views and expectations, which are based on currently available information and management's judgment. We assume no obligation to update these forward-looking statements, and we can give no assurance that the expectations will be attained. Actual results may differ materially from those described in the forward-looking statements and may be affected by a variety of risks and uncertainties, including those set forth in our SEC filings. Also, on this conference call, we refer to certain non-GAAP financial measures, such as funds from operations. Reconciliations of these non-GAAP financial measures are available in the supplemental information that is part of the full earnings release, which can be accessed on the Investor Relations pages of our website at coresite.com. And now I will turn the call over to Tom.

Thanks, Derek. Hello, and welcome to our second quarter call. Today, I'll discuss key takeaways from our performance in the quarter, provide an overview of our markets and discuss our growth initiatives. Jarrett will then provide a deeper dive regarding our sales results and put our Q2 activities into perspective among our geographies and targeted customer communities. Jeff will then review our financial results, balance sheet position and provide an update on 2013 guidance before we open the call to questions. Our Q2 results reflect continued execution of our strategy to support performance-sensitive applications. In the quarter, we increased earnings meaningfully, recording a 9.9% sequential growth in adjusted EBITDA over Q1 2013 and 25.4% growth over Q2 '12. Similarly, we increased FFO 9.6% sequentially over Q1 '13 and 22.1% over Q2 '12. Growth across our product lines, most importantly in product lines key to our strategy, drove these results. Specifically, sequential growth in Q2 revenue over Q1 was comprised of 7.3% growth in interconnection revenue, 7.1% in power revenue and 3.3% in rental revenue. We attribute the growth in interconnection and power revenue relative to rental revenue to a favorable sales mix and deal structure. Notably, revenue from power in our breakered-amp product line increased 25% in Q2 '13 over Q2 '12, reflecting the benefits of our strategy and go-to-market model. Beyond our Q2 financial results, we recorded accelerating sales this quarter. In Q2, we sold $5.8 million in annualized GAAP rent from new and expansion leases of turnkey data center capacity, a 14% increase over Q1 and the trailing 4-quarter average preceding Q2. The average annualized GAAP rental rate of executions in the quarter was $188 per net rentable square foot. This higher rate relative to the last quarter was driven in large measure by increased power density and…

Thanks, Tom, and hello, everyone. I'd like to start by discussing our sales activity during the quarter. In the second quarter, we executed new and expansion turnkey data center leases representing $5.8 million of annualized GAAP rent with an average rent of $180 per net rentable square foot. The $5.8 million in leasing volume is 14% above our trailing 4-quarter average. The annualized GAAP rent of $188 per square foot was in line with our trailing 4-quarter average when adjusted to account for power density and geographic mix. Regarding renewals, we renewed 44,702 Net Rentable Square Feet at a weighted average GAAP rental rate of $166 per NRSF, reflecting a 5.4% increase in rent on a cash basis and an 11.7% increase on a GAAP basis. Rental churn was 2% for the quarter and 3.2% on a year-to-date basis, in line with our guidance. Lastly, we commenced 42,672 Net Rentable Square Feet of new and expansion leases at an annualized GAAP rent of $147 per square foot, which represents $6.3 million of annualized GAAP rent. We believe our Q2 2013 sales demonstrate the value inherent in our strategy to focus on building customer communities in network-dense, cloud-enabled data centers supporting performance-sensitive applications. We continue to expand the CoreSite Mesh, which is our community of customers and partners that do business in our facilities. During the second quarter, we executed 115 deals across our portfolio, 33 of which were with new customers. Each lease, expansion, customer logo and partner enhances the ecosystem's value and makes the community even more attractive to potential customers. The 115 new and expansion leases executed in Q2 represents a 20% increase over the trailing 12-month average. Further, the 40 multi-site agreements Tom referenced represents a 35% increase in multi-site executions over the last 4-quarter average. Over…

Thanks, Jarrett, and hello, everyone. Thank you for joining our second quarter call. I'll begin my remarks today, first, by reviewing our Q2 financial results. Second, I will provide an update regarding our capital investments and our balance sheet and liquidity capacity. And finally, I will provide commentary on our guidance for the year. In the second quarter of 2013, we recorded revenue of $57.7 million, a 4.7% increase on a sequential quarter basis and a 13.9% increase over the second quarter of 2012. The quarter-over-quarter increase was primarily driven by increases in our interconnection and power revenue, as both Tom and Jarrett discussed earlier. The growth rate year-over-year was diluted by 1.4% due to our customer at SV3 decreasing their metered power draw as they transitioned certain applications out of our facility. This facility is currently fully leased to this customer, with 50% expiring in 2016 and 50% in 2017. Our Q2 operating revenue consisted of $48.7 million in rental revenue from space and power, an increase of 4.4% as compared to Q1; $7.1 million from interconnection revenue, an increase of 7.3% over Q1; and $1.9 million from tenant reimbursement and other sources. Our backlog of projected annualized GAAP rent from signed but not yet commenced leases is $8.5 million. We expect approximately 85% of the backlog to commence in the second half of 2013, with equal amounts in both Q3 and Q4 and the balance to commence in 2014. Our second quarter FFO was $0.45 per diluted share in units, an increase of 9.8% on a sequential quarter basis and a 21.6% increase over the same quarter of the prior year. Our FFO growth in the quarter demonstrates our ability to realize efficiencies of scale by leveraging our current resources to minimize expense growth. On the expense side,…

[Operator Instructions] Our first question comes from the line of Jonathan Atkin with RBC.

A couple of questions. First, on expenses. You talked about G&A trending up. How do we think about sales and marketing and operating maintenance relative to the 7% that we saw in the first half for sales and marketing and the 26% for operating maintenance for the second half? And secondly, I was interested in the new logos that you brought on. And are these customers new to colo, or are you seeing benefit from churn from other colo providers, or was it a matter of those customers wanting more vendor diversity, and you're winning the business in kind of that fashion?

Jonathan, this is Jeff. Let me touch on the first one, and then we'll have Jarrett and Tom address the second. But as relates to our general -- I guess, our G&A cost, as I mentioned in my prepared remarks, we expect that amount to trend back up to a roughly the 12% of revenues, as we've guided to in previous quarters. That lower amount that was reported this quarter was assisted a little bit by about $400,000 credit that we have going through resulting from us adjusting our reserves for bad debt. On the sales and marketing, as I mentioned, we're at about 6.8% of revenues for this quarter. We expect that to be still in that range of 6% to 6.5% per quarter as we move forward and as we continue to look at some additional investments in sales and marketing expenses, where we might want to enhance and/or make further investments. But I think that range still gives you some idea where it expects to be. On the operating expenses, the only thing I would say there is that in the third quarter of our fiscal year, it is traditionally a little bit higher due to the seasonality of some of our power expenses that do -- that we will get hit with in the third quarter. Other than that, that should be relatively consistent beyond Q2 as we head through the second half of the year.

Jonathan, it's Jarrett. And building on the new logos with the 33, we're just seeing progress. We're seeing more at bats [ph] with the sales and marketing coverage and investments that we see there. We're primarily seeing more, I think, new deployments, new application-type deployments, but we also see the diversity of choice, particularly in the network and cloud segment. So we see continued momentum. And again, with the investment and the tenure of the sales force, we see more at bats [ph] and more new logos.

And then on the sales force, can you remind us where things stand on sales headcounts and maybe comment on the sales retention, sales recruitment, any major changes or challenges that you're seeing on that front?

Yes, we're about 90% there on the sales side. You'll see continued ebb and flow, but we're materially there. We've been patient, hiring the right talent, and it's really paid off in the network and cloud vertical segments, where we're seeing that progress and that focus. So we'll continue to be patient and move forward as we go in and close out the rest of this year. The big thing is giving them time and getting them productive and getting them more experienced and trained. That's taking a couple of quarters.

And Jon, it's Tom. I would just add that we felt like we saw pretty good returns on the expansions of the quota-bearing headcount in Q2 as people are coming off of ramp. We believe that what we've been working to do is working. And given that, we're going to evaluate probably adding an additional smaller round of headcounts. But if you think this 30 to 35 quota-bearing heads right now, if we look at another 5 to 10 on top of the historical plan, it's not going to move the expense structure in a huge amount, but that's a -- it could be a meaningful add to the quota-bearing headcount, and we're certainly looking at that.

Great. And then finally, I'm just wondering about connecting your sites together. And to what extent are you seeing carriers deploying or customers looking to purchase or maybe deploying their own dark fibers as opposed to lit services between your sites? And you mentioned the example of NY1 and NY2, where you're doing a tethering. Is that dark as well as lit or just lit capacity at this point?

Jon, it's a combination of dark and lit. In major markets, we want multi-count, high-count fiber between our sites. But think of it as it's really our campus strategy. It's really bringing this large cloud enablement, where we can do the larger deals, and then tethering and connecting back through high-count fiber back to our point of interest, in the case of NY1. So we're getting some momentum with that. And in other markets, with did share with you, where there's key points of interest that are moving, that are running out of capacity, it's a benefit for us to make it easy for them to leverage our campus in NY2 or other markets.

Jon, when you're thinking about intermetro connectivity, we have a link along the coast of California linking Northern and Southern; we have a link over in the Eastern region. Those are lit. Those are lit leases with a little bit shorter-term capacity. Within a metro, intrametro, most of that is dark. And that's been something that's been meaningful for us in L.A. As we mentioned on the call, it's accelerated very well in Los Angeles with our strength at LA2. And so we're replicating that in the New York. Intrametro, primarily dark. That said, Jon, I mean, I guess the bulk of the capital and the bulk of the strands, intrametro, is dark. We do provide lit services intrametro, it's just -- you can put gear on either end of one strand and break it into waves and move a fair amount of smaller retail traffic, but that's one strand. So as a service offering, we are also lit intrametro. But as a capital and if you count fiber, you see the big guys wanted to light their own glass, and they are looking for dark.

Our next question comes from the line of Emmanuel Korchman with Citi.

Just -- Jeff, if we can dig in to guidance for a second. You're now guiding to the lower end of your revenue guidance and of your cash rent growth, but you've raised FFO guidance, and it doesn't look like anything else on Page 21 has changed. So maybe, can you walk us through what's changed between last quarter and this quarter, except for actuals coming in, obviously? And also, what might get you to the high end of guidance if you're thinking lower end on those 2 big items?

Yes. As you pointed out, and as I did say on the call, we've guided towards the lower end of our revenue range. And that's largely due to our customer at SV3 having lower metered power draw during 2013. How -- and really what it comes down to is really product mix. On the positive side, we have had higher revenue from our higher-margin products. And that this a good flow-through to the bottom line. And as a result, we expect to have adjusted EBITDA, reported adjusted EBITDA, at the higher end of that range. And ultimately, that's what's increasing -- or resulting in the increase in the guidance.

You see a deceleration in metered power, which is a 0-margin product, and you saw outperformance on higher-margin products, cross-connect and breakered-amp power. And that's really the reason for the dissonance between the movement in top line revenue and the movement in profit.

Maybe I'm missing something. How do they go from an EBITDA range that doesn't change, though, to an FFO range that increases?

Well, I think Jeff said we expect our adjusted EBITDA to come in at the higher end of the range as well. We're guiding up on adjusted EBITDA. We're guiding up on FFO. We guided down a little bit on revenue, and it's all because of product mix.

I mean, your $1.05 to $1.10 EBITDA range didn't change, though, quarter-over-quarter?

Correct. Just take the midpoint, Manny, of that range, of $1.075. We expect to be in the higher end of that range.

Our next question comes from the line of David Toti with Cantor Fitzgerald. David Toti - Cantor Fitzgerald & Co., Research Division: Jeff, the first question I have is relative to your floating-rate debt level there. I know you've spoken about this in the past, but maybe we could talk about the -- specifically, the line balance. Clearly, funding development through that facility. What are the current long-term financing plans as some of these developments stabilize and get delivered? And is there some impact, given credit market turbulence, in your mind, to some of the spreads that you established going into some of those developments relative to the long-term financing costs?

David, let me give you some idea kind of how we're watching and monitoring things and give you some idea of what we're thinking. But when you look at -- we obviously -- we watch the mark-to-markets very closely. But when we look where we are halfway through the year and what we anticipate spending in terms of development dollars for the rest of this year, we've guided to a range of $200 million to $225 million of CapEx for the year. If you just pick a point somewhere in the mid-range, call it $215 million because it will make the math easier, year-to-date, we've spent about $96 million. So that leaves us about $120 million of additional CapEx that we'll spend in the second half of the year. We've got about $73 million outstanding on the line today. So that's going to put us up somewhere between $190 million and $200 million outstanding on the credit facility by the end of the year, assuming we don't do anything from this point between now and then. That still gives us well above $200 million in total liquidity. However, we're going to continue to watch the markets and ultimately assess when it makes the most sense for us to go to the market, and it's really going to depend on the size of the proceeds that we want to deploy, what pricing is looking like in the marketplace and, ultimately, the term we're looking for when we term out some portion of that short-term debt. I think, as it relates to -- I guess, your second part of the question was really around levels and how is that impacting ultimate returns. We've traditionally said that our returns, which are on an adjusted EBITDA basis, are north of 12%. So that's not going to change that based on what we end up doing from a financing perspective, but that gives you some color in terms of how we're looking at and monitoring the markets. I think late 2013, early 2014, we will continue to look at where an opportune time is to do something from a financing perspective.

David, I'd offer 2 comments in addition. One is, although our debt, if you exclude the preferred, our debt is 100% floating right now, it's also currently 1.2 turns. And we think about when to term out more in terms of the floating as a percent of turns of EBITDA rather than the ratio of floating to fixed. If we were an 8x EBITDA borrower, that ratio might have a lot more meaning, but we really just think more in terms of the risk of interest rate rises relative to our income trajectory. We think through the rest of this year, that's a very manageable risk. But when we do get a handful of turns out or 3 turns out, we look hard at terming out, as you saw us in the past. The second thing I would offer is, as the spreads in the market have widened, certainly, over the last 2 years, the company-specific spread at CoreSite has come in, I think, reasonably nicely. So you have a company dynamic that continues to strengthen relative to a market dynamic that has widened a little bit. I think the strengthening of the company dynamic tempers the risk in the market dynamic. I'm not saying that it overcomes it, but it tempers it. And so we think about that when we think about the risk of widening rates versus near-term dilution to our shareholders. So we think in terms of absolute floating-rate debt relative to EBITDA, and we believe our company-specific spread continues to tighten. David Toti - Cantor Fitzgerald & Co., Research Division: Okay. That's all very helpful. And then I just -- I had one other detail question. Maybe I missed it. Did you talk about the occupancy decline at MI1? Did I miss that?

We didn't. The short strokes there are we had one customer move out. There might have been more. We had one material customer move out. And the decline in occupancy there is really due to the law of small numbers. It's a 20,000-foot building. We had a 5,000-foot lease move out. I think if you look at the rate on what moved out, you can -- I mean, I'd look at the variance between the property table q-on-q. We moved out 5,000 feet, an average of $45 a foot a year, and the in-place base on the rest of Miami is $121 a foot a year. So we had a small guy move out at a very low rate, and we're -- we didn't manage that churn to have a different strategy with the building. It was a not-that-profitable customer, and we feel good about backfilling that over the course of the year, and we think the in-place rents reflect that opportunity.

Our next question comes from the line of Jonathan Schildkraut with Evercore.

I'd like to maybe ask some questions about demand drivers. Tom, going back to our conversation we had at NAREIT, where we talked a little bit about Facebook and their need to kind of reach the mobile customer and that highlighting the value of your dense network facilities. Since that time, I think that digital media and dynamic ad insertion for mobile devices has really escalated in terms of news flow. And then obviously, we had a very good print from Facebook on their mobile ads yesterday. So I was wondering what else you're seeing in that digital media, ad insertion marketplace and how that's driving demand in your facilities. And then sort of on the other side, there's been some news flow about remote peering. And I was wondering if that was at all a headwind or an opportunity for you?

I'm going to turn the mobility question over to our savant here.

Jonathan, it's Jarrett. Yes, in terms of mobile, we are seeing ad-insertion platforms prevalent. There are some assets that were getting some tower deployments as well from the network guys, as well as, clearly, some value in financial services in our assets like our D.C. campus, Chicago, New York. I think we saw mobile deployments from the network provider, from tower infrastructure, from that product line and from the digital content players, I think, in 5 markets this quarter. So it is early days, but it does leverage that community. It does leverage the digital content. You mentioned one player in that space, but our digital content is a very strong base for us. They are pushing that to the edge. And again, from the market perspective, ad insertion is latency-sensitive, and we are focused on those folks and new logos in that space who are coming on-board.

Great. And remote peering?

Oh, in terms of remote peering, I think there's an opportunity. One of the things that we called out was the 40 multi-site deployments that we had this quarter. We have multiple networks in every one of our markets this quarter, and that's pushing out for the IP side and peering community, as well as the private networking deployments that they're putting out there. And so I do think there's opportunity for us there that folks are looking for other places for IP and peering and picking new choices outside of downtown New York. I think that's particularly an opportunity for us in our NY2 facility. We aren't seeing a lot of pull-through that we can see right now from remote peering. There's still major hubs, typically 2 and 3 -- 2 or 3 in every one of the markets. Folks are diversifying in Northern Virginia, New York, Chicago, Santa Clara and L.A. They're looking for multi-site kind of deployments there.

Great. And in the past, you guys have given us some indication of exposure to performance-sensitive applications by discussing the number of -- or percentage, rather, of customers that had multi-site deployments across the base as opposed to just the new contracts. And I was wondering if you might update us on those numbers.

I don't -- we don't have the updated stats at the moment. I would be comfortable that it's moved north, and we'll come back on the next Q with the detail. I don't think it's-- these are fairly large battleships to move, right? So I don't think it's changed in a huge amount. But I think the trend is favorable, and we'll come back on next Q with the clinical data again. Sorry we didn't bring it this time.

Our next question comes from the line of Jordan Sadler with KeyBanc Capital Markets.

I wanted to come back to the question regarding tethering, which you are doing in L.A. and now in New York. I'm curious, one, about sort of the plans to do it in other markets. And then, it's unclear to me, are you connected to third-party facilities in L.A. as well? I mean, it sounds like New York, you're tapping into 111 8th and 60 Hudson and maybe some other. So are you doing the same in L.A.? And maybe just talk about that strategy a little bit more.

We're -- we don't in L.A. I mean, our only tether in L.A. is a point-to-point from LA2 to LA1. I think broadly, the strategy, Jordan, where you're entering a market without a point of network density, then you help seed growth by tethering to other points. When you have established your own point of network density, then you don't really tether to other people. And as you build more capacity elsewhere in the market, you tether back to your point of network density. It's the age-old hub-and-spoke that we've been doing for years now. And the tethering to other folks I view as kind of the first 3 to 5 years, as things get going. And then, as we execute our strategy to bring more networks into our facilities, then the value of that tether decreases.

What's the cost in New York, for instance, of sort of the -- of tethering to those 2 other facilities?

I think the capital on that, Jordan, was on the order of $1 million, $1.5 million, plus or minus.

Per facility?

Yes. And there's a reasonably manageable monthly maintenance cost with the network.

And just on that, if I could build, one of the benefits of that, those 2 campuses happen to be publicly announced where you can directly connect AWS. And again, bringing the network and the cloud community and making that a campus is a great -- not just for AWS, but other cloud providers, giving them that choice of multiple cloud providers and hybrid environments for cloud closer together.

Okay. A separate question is regarding the PLR having no impact. What's the -- I mean, I see that there obviously was some income tax in the quarter. What are the mechanics surrounding, I guess, avoiding the income tax going forward?

Yes, Jordan. Most of the tax expense that you see in the quarter is really resulting from state income taxes. That PLR generally won't have a whole lot of impact on state -- the way states look at REITs. So you're always going to have some level of -- a small amount of state tax expense. In terms of logistics for the PLR, we have been working on transitioning our business out of the TRS and into the REIT, and it just takes a little bit of time. But as I said on the call, near term, we don't think it's going to have a big impact. However, longer term, as that side of the business grows, that's where we will see the benefits on a long-term basis.

Okay. And then lastly is just on SV2. I didn't catch exactly what was going on with that tenant. It sounded like they're moving some gear out, even though they've got later-year expirations. Can you maybe just give us the nature of the type of the tenant or the nature of the move-out?

Jordan, it's SV3. So that's the large single-tenant deal we've had there for some time.

Oh, it is SV3. Okay.

Yes.

And that's the deal you did shortly before the IPO?

That's right. The single-tenant, full building.

Okay. You just said they're scaling up and pulling some of the gear out at this point, right?

Yes. I think that's a customer who is -- no surprise in the marketplace, a number of larger customers are self-sourcing data centers in other locations. And I think this customer has consolidated half of the gear out of that building. They're, I think, looking to sublease the space. We've talked about this in the past. We're also in an open dialogue with them. And our income is secure for the next 4 and 5 years. And there you go. I think, at the end of the day, the timing at which we take that space back to the market, if it's before the lease expiration, will just be driven by when we need inventory in the location. As we've said in the past, wholesale rates have declined meaningfully from when we signed that deal. Separately, though, we believe that our ability to -- that we have an ability to increase EBITDA in that facility off of that inventory with a retail program. So we don't have an intent to take that back to the wholesale market at whatever time that capacity comes back to us. And Jordan, I want to be -- also, a little more color around the cost of the networking, if you will. The fiber is one cost. The gear on either end of the fiber is another cost. And then the capability to manage the network is a component. I think what we shared in the past, if you look at any tier, you look at our packet exchange, you look at our Ethernet exchange, you look at our dark fiber purchases and you look at our gear inside our buildings, it's less than a couple of points of our total capital base. It's -- I'm guessing it's 2 points plus or minus. So it's -- there's a lot more to it than just buying $1 million or $2 million worth of fiber. But the total cost is still a reasonably small amount of our capital.

Was that part of the underwriting of NY2?

It was. We went a little bit longer on it than we had underwritten. So you saw a little negative variance on cost there because we went a little bit bigger on the package, but yes.

Our next question comes from the line of Matthew Spencer with Robert W. Baird. David B. Rodgers - Robert W. Baird & Co. Incorporated, Research Division: It's Dave Rogers here. I wanted to follow up on the developments. You've got 6 projects or so that are complete, yet to stabilize. Another 4 under construction. Maybe taken separately, Tom, can you talk about any activity in the market that either makes you incrementally more positive or negative on the returns coming off of those and remind us, maybe as separate groups, what the overall development returns do you expect on those assets are?

Sure, Dave. I think that -- we think of the development in 2 different groups. There's one class of development that's really just add-ons into existing facilities. And those are incremental builds to meet demand as we start getting short on inventory. Those tend to be the higher-margin, higher-returns -- higher-return developments for us because by definition, they're built inside a facility that already has customers and traction and some interconnection base. Let's leave that category there. Circling back to the other category, you have new large ground-up or full-building rehab deals. And we have 4 of those -- or 3 of those right now. We have the pre-leased deal in the Bay Area, SV5. And I think we've given all the guidance we can give around returns on that. It's an attractive deal for us, and we're glad to have it. And then we have 2 large new TKD data centers going up, one in Virginia, one in New Jersey. We've consistently said we think we can meet or beat '12 [ph] on a stabilized cash yield basis. We continue to believe that. I think as our -- what we do expect, and historically we have seen, higher returns on the incremental adds in existing data centers, and those returns can vary fairly widely. And our objective is to continue to drive a higher component of breakered-amp sales and cross-connect sales as part of the product mix and, hopefully, have some upside across the portfolio. But that's how we view yields. Really no big change other than we think the yields continue to move in the right direction for us across our -- each development, and we're excited about the future. David B. Rodgers - Robert W. Baird & Co. Incorporated, Research Division: Great. And then maybe on the strategic deal that you did in the first quarter, and I didn't hear if you commented on this during your prepared remarks. But can you give more color on kind of when you expect that rollout to be fully deployed and then kind of what the time frame is you're thinking about for the cross-connect? I know we've talked about kind of the timing of cross-connects kind of following up to those deals, but when will they be fully deployed?

Yes, this is Jarrett, Dave. The first phase, over the holiday season. And I think we mentioned this as a cloud gaming platform. And into early next year, you can expect the first phase of that deployment to go out, again, in Q4. But it will phase in and ramp up in the first half of 2014. The first deployments will include power. And again, it's an interconnection -- we anticipate it to be an interconnection-rich application to the networks in those sites. So first phase for the holidays, and then ramping into 2014.

And I think on average, our stats suggest that 80% of total cross-connect growth from all new deployments in any Q burn in over an 18-month period after the deployment. There are -- some are faster, some have more growth for a longer period of time, but on average, you get about 80% of the cross-connects on a new deployment over the 18 months after it deploys. David B. Rodgers - Robert W. Baird & Co. Incorporated, Research Division: And then last question, maybe on the acquisition environment. Any appetite for additional acquisition today? Anything that you're looking that -- coming up?

Well, we wouldn't comment on anything specific, Dave. But I would just say that -- and the message we tried to send last quarter is, look, we just try to be smart and pay attention to the market and pay attention to opportunities to increase shareholder value. At the same time, our internal growth opportunity is quite significant, we believe. And as such, our hurdle for buying something is relative to our hurdle for sticking to our knitting right now. And really, none of that has changed since we've been a public company. I think there's been a little oscillation in the perception of what we're trying to say, but our orientation, I feel, has been pretty consistent.

. Our next question comes from the line of Jamie Feldman with Bank of America Merrill Lynch.

I was hoping you could talk a little bit more about your expirations. I know you touched on some this year, but even into next year, where you've got 22.8% of rent expiring, are there any known move-outs? Or how should we think about that NOI stream?

Yes, Jamie. This is Jeff. We haven't given any specifics as it relates to 2014 roll. We'll plan on doing that probably later in the year or as we give 2014 guidance. But I can tell you that there aren't any significant known move-outs in that fiscal year that we're aware of at least today. So nothing significant that we're aware of that will move the needle there.

Okay. And then, I know you spoke about mix and -- product mix and tenant mix kind of driving better rents. But where would you say tenants are just today in terms of thinking about the economy and whether it's getting better and how much better, and how much of that is driving your outlook -- or I guess, how much of that drove operations in the second quarter, and then your outlook going forward?

Well, I think the North American climate is obviously a little bit brighter. But I think we all read the same papers and see the same economist reports. There's not blistering growth here. But certainly, the outlook here is a little bit brighter. I think more than anything, you just saw -- I would circle back to what we've said pretty consistently. I think overall demand in the industry I expect to grow at similar rates to where it has before year-on-year. And there are some quarterly oscillations. We saw that in Q3 last year. You saw the counter-oscillation in Q4 of last year. We look at Q1 and Q2 results as fairly normalized. So I don't think that there was an exuberant outpouring of executions based on a view of better health in North America. Things are, I think, on the margin, better in North America. I think, more than anything, we're seeing stronger growth in capturing the smaller performance-sensitive deals that we think have a product mix that drives higher margin because we've invested into a platform that captures more of that activity. So that's what we attribute Q2 to. And I think things are better in the States, but I don't see IP spending or the trends behind IP outsourcing, we don't see Gartner's forecast shipments of router and interconnection gear. Those trends still seem relatively consistent, a little bit brighter in the States, a little bit duller in China and Asia. But it's a fairly consistent pattern from what we've seen over the last couple of years.

Okay. And then finally, one of your competitors last night discussed the lengthening the book-to-bill cycles, specifically in the enterprise segment. Is that something you guys have seen?

Yes. This is Jarrett, Jamie. Not really. We have seen consistent, again, the smaller deals and our focus on that are moving -- are consistent. We're seeing those consistently move through to close, and roughly the same time.

And I would say, also, you've got to think about the geography. It -- I think there are other companies in the sector that have more business abroad right now. And maybe some of the lengthening of cycles is due to macro factors elsewhere. My guess is -- my sense would be that the companies we compete against in North America are probably seeing pretty similar things that we are, and they're very able competitors with great business plans. And some -- other times when you have exposure to a market that's a little bit slower, and then things pick up. So that's our sense. I think in North America, we've seen things stay pretty steady.

Our next question comes from the line of Tayo Okusanya with Jefferies & Company.

Two quick questions. First of all, the -- could you talk a little bit about the dollar value of rents that have been signed but haven't yet commenced?

Yes. You're talking about backlog, Tayo? It was -- in the prepared remarks, we has talked about it. At that end of the quarter, it's $8.5 million. And what we gave some color on was that -- of that $8.5 million, we expect about 85% of it to commence in the second half of this year, equally split between Q3 and Q4, and the rest of it in 2014.

So that's helpful. And then the second thing. I mean, there's been some recent chatter around Facebook and then subleasing a lot of their space in Santa Clara. Could you kind of talk about how that may or may not have impacted you?

Yes, I don't think it has impacted us. And we really have worked to message this to the Street for the last 2 years and strongly over the last year. We do expect them to move out of the SV3 building upon lease expiration. You have half of that in 4 years and half of that in 5 years -- or I think it's 3 and 4. And we expect to time our inventory deliveries in Santa Clara such that we would convert that space to our retail colo program. And between now and then, we're collecting rent, and we have a good relationship with the customer, and on we go.

So Facebook is still in your space right now, or have they actually subleased it already?

No. They've been in the market to sublease it. We've been in discussions with them. I mean, they have -- I think they've moved out a portion of the building, and that's that. It is a fairly normal landlord-tenant situation with a tenant who doesn't need some of their space and is probably going to move out the rest of it over whatever time period make sense for their business, and we look at the intelligent way, as the landlord, to either take that space back early and market it early or wait until the lease ends, and that's really driven by our inventory availability in the market. That's what drives our decision-making and our thought of timing and whether we participate in taking it back early or not. And until we need capacity in the market, we're happy to stay where we are with that lease.

. Our next question comes from the line of John Stewart with Green Street Advisors.

This is Eric Frankel here with John. I just have a question regarding LA2. Obviously, you made a lot of great progress in leasing there this quarter. Just curious, were there any customers that migrated from LA1 over to LA2?

Yes. We've had some migrations. We've had a handful, and it's an activity that we encourage.

Great. And just a question on the G&A. I think, Jeff, you might have quoted on G&A, you expect to run at roughly 12% of revenue over the next year. Is that scalable at all in the future, or is that probably a the way to just think about G&A over the next couple of years?

No. I think in the near term, call it 12 to 18 months, I think that's a reasonable percentage. I think once you get beyond that, we do think it's scalable. And ultimately, we'll know better as we get closer and closer to those time frames. But at some point, definitely, it is scalable.

And I would offer that not scaling it, I think, is a good sign. Not scaling G&A means that we believe that further investments create accretive returns. If we wanted to just stay steady state as a business in our current activities, then we could scale G&A down as a percentage of revenues. As we -- if we believe that enhancing our capabilities produces attractive returns, we'll keep G&A as a percent of revenue where it is or evaluate the optimal position for the shareholders.

Great. And just a final question, just on interconnection revenue. Can you just clarify how much of that comes from your more network-dense properties and how much comes from some of the newer enterprise locations?

It's -- we haven't given that specific breakout by property, Eric. I don't even have it at my fingertip. But -- yes, at this point, we haven't disclosed those separately.

Well, I think generally, the longer a property has been around and the more customers it has in it, the more cross-connects you'll have embedded in that building. And we just work hard to systematically add to that dynamic in new buildings and new developments. It takes time, but the trend has been favorable, and we're going to keep doing what we're doing.

Mr. Ray, there are no further questions at this time. I'd like to turn the floor back over to you for closing comments.

Thank you. Well, we're pleased with Q2. We're very pleased with the seasoning of the sales team and with the returns we're getting on what we're trying to accomplish. I think, more than anything, we're very pleased with the sales mix we see evolving in the organization. And we're going to stick to our knitting and keep going. Thank you, everybody, for the support from the analyst group, from the investors. We're going to keep working hard to create good returns, and we'll talk to you guys soon.

Thank you. This concludes today's teleconference. You may disconnect your lines at this time. Thank you for your participation.