Greetings and welcome to the CoreSite Realty's Fourth Quarter 2019 Earnings Call. At this time all participants are in a listen-only mode. A question-and-answer session will follow the formal presentation. [Operator Instructions] I would now like to turn the conference over to your host Carole Jorgensen, Vice President of Investor Relations and Corporate Communications. Please go ahead.

Thank you. Good morning and welcome to CoreSite's fourth quarter 2019 earnings conference Call. I'm joined today by Paul Szurek, President and CEO; Jeff Finnin, Chief Financial Officer; and Steve Smith, Chief Revenue Officer. Before we begin, I'd like to remind everyone that our remarks on today's call may include forward-looking statements as defined by federal security laws including statements addressing projections plans and future expectations. These statements are subject to a number of risks and uncertainties that could cause actual results or facts to differ materially from such statements for a variety of reasons. We assume no obligation to update these forward-looking statements and can give no assurance that the expectations will be obtained. Detailed information about these risks is included in our filings with the SEC. 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 simple information that is part of our full earnings release which can be found on the Investor Relations page of our website at coresite.com. With that I'll turn the call over to Paul.

Good morning and thank you for joining us. Today I'm going to cover our 2019 financial highlights and recap our 2019 priorities and key accomplishments. Jeff and Steve will follow with their respective discussions of financial and sales matters. Our 2019 financial results included new and expansion sales of $55 million, a record which nearly doubled the $27.7 million of annualized GAAP rent signed in 2018, operating revenue of $572.7 million, which grew 5.2% over 2018 and FFO per share of $5.10, an increase of $0.04 year-over-year. A year ago, I shared four priorities for 2019 translating new construction into more abundant sales, acquiring additional new logos, bringing new connectivity products online to increase sales and delivering a great customer experience and ongoing operational efficiencies. I'll review each of these relative to our 2019 accomplishments. We executed well in our first priority of translating new construction into higher sales. In 2019, we placed 224,000 Square feet of data centre capacity into service, including a 108,000 square feet for the first two phases of FDA, our ground up development in Santa Clara and 116,000 square feet of campus expansions in Reston, Los Angeles and Boston. As a result, we restored our available and developable capacity to 25% in our top five markets at the end of 2019, compared to 16% at the end of 2018. And we used the new capacity to achieve a record leasing year, including leasing 100% of the first two phases of FDA and 74% of LA3 phase one a year in advance of its expected completion in late Q3 of 2020. Our 2020 development pipeline continues to be strong as we expect to deliver at least 196,000 square feet of new projects, including two grand up developments CH2 in Chicago and LA3 in Los Angeles, the…

Thanks, Paul. Today I will review our fourth quarter and full year financial performance, discuss our balance sheet, including our liquidity and leverage expectations and review our financial outlook and guidance for 2020. Looking at our financial results, for the full year, operating revenues grew 5.2% year-over-year, reflecting increases in new and expansion lease commencements of 46.8%, growth in interconnection revenue in line with our expectations, offset by elevated churn we experienced in 2019. General and administrative costs were $43.8 million reflecting 7.6% of revenue in 2019 compared to 7.4% in 2018. Net income was $2.05 per diluted share, a decrease of 7.7%. FFO per share was $5.10, an increase of $0.04 over 2018 and adjusted EBITDA margin was 53.8%, a decrease of 60 basis points year-over-year. For the quarter, operating revenues grew 5% year-over-year and approximately 1% sequentially. We commenced new and expansion leases of 86,000 square feet during the quarter, reflecting $16.6 million of annualized GAAP rent. Our sales backlog as of December 31, included $15.6 million of annualized GAAP rent for signed, but not yet commenced to leases or $19.8 million on a cash basis. We expect about 40% of the GAAP backlog to commence in the first half of 2020 with the remaining 60% in the second half of the year, weighted to the fourth quarter with completion of LA3 phase one. Adjusted EBITDA was $79 million for the quarter and increased 6% year-over-year and 1.4% sequentially. Moving to our balance sheet, in November, we amended our credit agreement, extending our near term maturities. We also extended the maturity date of our revolving credit facility to November 2023, with a one year extension option, and we added an additional $100 million of liquidity. We ended the year with $386 million of total liquidity, providing plenty of…

Thanks, Jeff and hello, everyone. I'll start off reviewing our quarterly sales results, and then talk further about our sales successes for the year and key drivers. We had a solid quarter of new and expansion sales, we delivered $6.6 million of annualized GAAP rent, primarily reflecting core enterprise leasing, including 31,000 square feet with an average annualized GAAP rate of $216 per square foot. Included in those results was a sizable multi market hybrid lease, leveraging the unique capabilities of our campus platform to support their dense architecture and complex interconnection requirements. As a second example of how business continues to evolve and leverage low latency, high performance technology, we believe this and others, like this customer, will provide additional interconnection opportunities, as well as stickiness to our platform. Winning new logos has been a key driver for us throughout 2019 and more an important contribution to these quarterly results. In the fourth quarter, we won 36 new logos. Two thirds of these new logos were enterprise customers, which included some notable strategic accounts. Turning to pricing, overall pricing in our markets is fairly stable, except for Northern Virginia where pricing is softer, especially on larger deals, as we've been saying for the last three to four quarters. We were pleased with our fourth quarter sales in Northern Virginia, which outpaced each of the prior three quarters on a volume basis. Renewals are another key aspect of releasing focus. During the fourth quarter, our customer renewals included, annualized GAAP rent of $21.9 million. Our renewals represented rent that decreased about 1% on a cash basis. Similar to last quarter, we had a few customer renewals negatively impact our cash flow growth for the quarter. This includes five customers that went excluded our cash mark-to-market was a positive 3.4%. Churn…

Thank you. We will now be conducting a question-and-answer session. [Operator Instructions] Our first question comes from the line of Colby Synesael with Cowen & Company. Please proceed with your question.

Great, thank you. Power as a percentage of rental revenue came down in the second half of 2019. And your revenue guidance for 2020, I think was below Street expectations. I'm curious if any of this has to do with the customer who's moving out of SV7, I think they may have actually already moved out and that space might be vacant. And that's what I'm thinking is the explanation, but I'm hoping could fill that in and give us some color on what the expectation is for power in 2020, that might be I'm thinking the delta between what the Street was expecting and what you guided to. And then secondly, as it relates to CapEx, I'm just curious what you've assumed for additional land purchases? Thank you.

Good morning, Colby. It's Jeff. Let me see if I can take your first question. You are correct, that some of that decrease in power revenue is directly attributable to the customer at SV7. And so as you think about 2020, I'd probably put it in simplistic terms from the standpoint given the guidance we've given you can see we've guided to an increase of revenue of about $32 million at the midpoint. We've also guided to interconnection revenue growth of about $8 million at the midpoint. The remaining $24 million, I would allocate that pro rata to what our full year 2019 is, the relative portions between rent to power, which is about two third rent, one third power, as you think about modeling and those components for 2020. In terms of CapEx, we have – obviously, the projects that are under development at this point in time. And well obviously, as we work through 2020 will have the need, hopefully in the opportunity to add additional development projects that will disclose at that point in time. At this point in time, we don't have any incremental land purchases in our guidance, not to say we're not looking, not to say we won't execute it on it, but at this point, none of that is embedded into the guidance for CapEx for 2020.

Okay, thank you.

You bet.

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

Hi, so I just wanted to come back to the churn. It seemed like you had the guidance for the full year laid out last year, and we ended up at 11.1 just above the high end of the range. I'm kind of curious – I feel like you ended up a little bit higher for some reason, was there something that was missing? Or something that happened during the quarter that you didn't anticipate 90 days ago, that caused sort of an increase in that churn that you could sort of speak to? And I guess sort of similar question for 2020. I feel like you discussed last quarter on the call the normalized level of 7.5% to 8% plus a little extra from this customer move out. But it even seems like this customer move out exposure in the fourth quarter of '20, it feels like the full exposure of the customer as opposed to half of the exposure that customer. I can clarify that if that doesn't make sense.

Yeah, Jordan, let me let me see if I can address that and, and see if we can clarify some things. But let me just start with the 2020 turn and then I'll come back and talk about the fourth quarter. As it relates to 2020 churn, what we tried to communicate last quarter is that we expect our churn in 2020 to recede back to its normal levels of call it 7.5% to 8% with the exception of the fact that we have to add another, as we just said, on the call 250 basis points related to that specific customer. So all in, you're going to have churn somewhere around 10%, which is what our guidance is. We tried to communicate that last quarter, and I think it got lost a little bit in the translation. And so I just want to make sure we're clear on that. So we expect midpoint at 10%, inclusive of 250 basis points from that one customer.

Is it 250 in 4Q of '20 and 250 4Q of '21, Jeff?

It's 250 in 4Q of 2020 and it'll be less in the fourth quarter of 2021. Just to give you an idea, it's five megawatts this year, four megawatts next year, just to give you an idea how that deployment will, will mature.

Okay.

And then as it relates to the fourth quarter, Jordan, obviously, the full year churn came in slightly ahead of and higher than what our guidance was. And I would point to two things. As it relates to our churn in 2019 and in 2020, we had some of that churn that we expected to ultimately take place in the fourth quarter of this year, so a little bit earlier than we anticipated. And then as I said in the prepared remarks, we expect to also have our churn in the first quarter of 2020 to be elevated. The full year, we still expect it to be in line with what we anticipated. But the timing associated with it got moved into the first quarter where some customers are moving out sooner than what we expected. As a result of that we expect first quarter churn to be somewhere around 325 to 375 basis points. And those two elements of some of that moving sooner than we anticipated are obviously weighing in on some of the guidance we've given for 2020.

Okay, and then I have a bigger picture question for you, Paul. Is a combination I guess of the elevated churn lower releasing spreads, lower CapEx, particularly relative to the size of the company today and higher leverage they all seem to point to slower growth profile for your domestic data centre portfolio. So even though demand does seem like it's going to persist longer term for data centre space, is it fair to say that we've passed the point of maximum valuation and/or maximum growth for data centers in the US?

No. Jordan, it's a good question. It may seem odd, even contradictory that we have, for example record sales and record churn in the same year. But it makes sense if you view them as two sides of the same technology coin. Our record sales reflect new data business models and use cases that seem to have significant tailwinds behind them, as reflected in our growing new logo sales and to edge in major metro cloud sales and continuing growth from customers acquired in recent years. The churn primarily reflects business models that are waning and as Jeff mentioned, have become a significantly smaller part of our portfolio, which means that the amount of that churn should wane, especially to the extended has been accelerated into 2019 and 2020. And the lower CapEx just relates back to what I said in my prepared remarks that we are shifting from the need to build out entirely new buildings, which for over the last year plus this year into a phase where we're building out new computer rooms in existing buildings which we can do more quickly and have higher returns. So I mean, look, we're like every company that has grown, you get bigger, your denominator gets bigger and the industry has matured some in terms of new capital coming into the industry. But we still feel really good about our business model and our markets. We actually have more markets that we can grow in that we practically had or had as a practical matter a few years ago because we used to be highly dependent on three markets. We didn't have scale in Chicago and New Jersey was more urban for a few years. But now we're building scale in Chicago and New Jersey has really picked up as a nice enterprise market. So in terms of being able to deliver, I think, above standard growth to compared to the rest of the reed industry, and serving an industry with significant tailwinds. I still feel very optimistic. But we are impacted last year and this coming year by this accelerated churn from these older business models.

And I guess the only other thing I would add there, Jordan is just to kind of point you back to some of the prepared remarks there as well. And just the overall demand characteristics that we've seen over the especially the last year and new customers coming to the platform we had more customers that are contributing more revenue, significantly more revenue and the average size of those customers is also significantly up. So I think as you look at the overall demand currents of the business and those customers adopting CoreSite it's actually stronger than ever.

Okay, I'll jump back in the queue. Thanks, guys.

Our next question comes from the line of Robert Gutman with Guggenheim Securities. Please proceed with your question.

Yeah. Thanks for taking the questions. So it looks like at the 30,000 square feet leased McWhorter about half that was mostly the Virginia, Northern Virginia, DC market and some Denver, where was – it's hard to discern where the other half of that was markets wise. And second question on renewal pricing, the week renewal pricing or the recurring sort of low renewal pricing in 2020 is it – can you characterize it a little more? Is it broad number of customers, concentrated customer and early, middle, end of the year?

Yeah, just to give you some perspective on the strongest markets, which kind of implies where some of that larger leasing came from were LA, Northern Virginia and New York, so hopefully that gives you a little bit more color as to where those leases came from. And the second part, your question there?

Just some color on the on the sort of muted renewal pricing this year, is it a few customers or is it sort of widespread, is it can be an even pace through the years, is it concentrated with one customer in the second half? I mean, just trying to figure out why, again, the renewal pricing is kind of weak.

Yeah and it's pretty indicative. As I pointed out in the prepared remarks around Q4 leasing where we had a few customers that drug down our mark-to-market to a negative 1% and when you exclude those five customers, we actually had a positive 3.4% cash basis. So as we look at some of those ageing customers, those specific customers side with us back typically in the 2015 or prior era, and as they've been – had annual escalations in their rent and some of the pricing in those markets has been stable or in some cases decreased. Now, we've obviously had to make some adjustments that have brought down the overall expectation there. So there's a little bit of that sprinkled throughout the entire guidance, but that's also baked into the overall assumptions in guidance for 2020.

Okay, thank you.

Our next question comes from the line of Jonathan Atkins with RBC Capital Markets. Please proceed with your question.

Thanks. I got a couple. I was interested on the demand side any markets you would call out has seen elevated levels of current demand compared to last year? And then wanted to get kind of an update on the SV7 backfill situation, if you could provide us a little bit more color than what you've provided in the script if that's possible. And then finally, the customer relocations in 1Q and 4Q, which markets is it occurring in and are these customers that are moving to the cloud or just downsizing the right G going through consolidation? Or what are the factors that are leading to this occurrence? Thank you.

Okay. I mean, the markets where we've seen pickup in demand as we mentioned, the comment I made earlier, New York has really improved. Santa Clara continues to be strong. We obviously had strong demand in LA last year and frankly as Steve mentioned in his sales comments Virginia picked up in the fourth quarter. So those are those are the primary markets that are plus on the demand side. In terms of –

Sorry, I asked about the current demand, so it sounds like – so what you called out was kind of backward looking at that and then continuing prospectively into this year [indiscernible] market in terms of elevators at the end.

Yeah, based on our sales funnel, they still seem to be to be good markets. The relocations, primarily, that's the SV7 lease that you talked about. And as we mentioned in the past, that's just a customer that realized that their applications were not performance sensitive, didn't need to be in Santa Clara moved it to a lower cost market. And to a certain extent, the other relocation is the same kind of thing and it's out of Milpitas and they are making the same kind of cost rationalization decision. No, I mean all I can really say about the SV7 situation and Jon you know as well as anybody how the scale and hyper scale leasing in Santa Clara tends to come in lumps and weighs although they tend to come quarter-by-quarter with very significant amplitude between quarters as opposed to year-by-year. But our sales team is out there, working with our customers and working the market for backfills for that space. The customer itself as you know has hired a brokerage firm which worked for the customer and we don't have any control over what they do. Time will tell if their efforts are a distraction or if they provide value to the customer and indirectly to us. But so there's a good coverage out in the market of all the opportunities that are potentially out there to backfill that space before the end of the year.

Thank you.

Our next question comes from the line of Nick Del Deo with MoffettNathanson. Please proceed with your question.

Hi, thanks for taking my question. First, when you look across your entire book of business, not just lease that we’re doing in 2020. How do you feel about mark-to-market risk and particularly in Northern Virginia?

So you could see how we feel about it that's baked into the guidance Jeff provided. I think it's harder to look beyond that because it really depends upon how supply and demand evolves in each of our markets. But I do believe it's a little bit like the churn thing that some of that will go away as we go past certain vintages of past leasing.

Alright, got it and then Steve, you noted pretty significant growth in deals that came through the channel [technical difficulty]. Do you expect additional growth in channel deals in 2020 and beyond? And how do you think about the cost of acquiring revenue via the channel versus in house and minimizing potential channel conflict?

I'll start with the last part of that question and get into the first part. It definitely is a balance right. So I do see value in extending the channel and we have some efforts internally to continue that growth and actually look at some other channels to broaden our reach and deepen our reach into key customers. As we look at the market, I think there's a great opportunity out there for the mid to large enterprise that is really wrestling with how they evolve their IT challenges and much of that involves partners to help them with that evolution. So working with those partners to make sure that we're part of that conversation is definitely part of the strategy. But it is balanced because in some cases that does come with cost associated with it. I would say that in some cases as well, it actually brings more value and stronger pricing, so it just kind of depends on the channel and how we manage that mix.

Got it, it seems like a number your peers are leaning harder into the channel as well. Do you think that's just kind of a natural function of new enterprises being potential customers going forward or is there a risk that people become too heavily reliant on the channel?

I do think it is balanced, right. And it is important to manage that risk or that mix rather, but I do think that it is such a complex environment out there for enterprises that they need help in managing that evolution of their IT strategy. And so as they look to partners, collocation is part of that and their cloud on ramps and how they bring that into a seamless architecture for their customers and their employees. So they need help doing that. And partners of various flavors are a key part of that, whether it's system integrators, resellers and otherwise. So I think that's a good opportunity to help our enterprises come to our platform, as well as a good opportunity for us to leverage kind of off payroll resources to help us find demand.

Okay, got it. Thanks Steve.

Yeah.

Our next question comes from the line of Michael Rollins with Citigroup. Please proceed with your question.

Hi thanks and good afternoon. I guess two things if I could. First, can you help frame if you take a look at the signed leases in 2019 and the $55 million of annualized rent, can you frame for us what the total opportunity was that you were pursuing in dollars in 2019? It gives us a relative sense. And are you finding that you're – that there's just maybe less available in the market for certain areas that you may want to fill? Or are you purposely moving away from those because of price over churns or some other factors? And then the second thing is just – if you take a step back, are there things in the portfolio that CoreSite would benefit from overtime larger sales force, different systems, certain geographies, are there certain things that you're finding could be incremental to pursue the growth and the wallet share opportunities that you're looking to achieve? Thanks.

So Michael, typically we don't give out specific sales forecasts. And so we probably shouldn't retroactively give out comparisons to sales forecast. But obviously, we're very happy with having a record sales year last year and what it tells us about the market. In terms of the latter part of your question, it's a really good question and kind of balances between strategy and execution, what you're doing, what we're doing, strategy is obviously really important. And we and our board take it seriously and we evaluate it regularly. But there's a – if you'll pardon my use of the expression, there's no Polish proverb to the effect that execution eats strategies lunch. And we have consistent as you pointed out we have put more focus on execution in the last few years. And I think you see some of that showing up in our results already. Well, we're building faster, better and more cost effectively. And capacity is crucial for sales. We're selling more effectively, including the difficult transition to selling hybrid cloud data centre solutions to enterprises who are relatively new to collocation, which shows up in our rising new logo sales. We've made agile changes in our product offerings that we've talked about that are specifically targeted to facilitating the customer journey to the hybrid cloud while easing the transition to a collocation model. Every one of those products has a purpose, making it easier to switch from one cloud to another, making it easier to provision redundancy for both resiliency and cloud SLAs, making it easier for companies to flatten their wand networks and be able to save significantly on their overall network costs as they move to a hybrid cloud model, and even giving them the same visibility into their collocation environment in terms of…

Thank you.

Our next question comes from the line of Sami Badri with Credit Suisse. Please proceed with your question.

Hi, thank you. Looking at your annualized rent mix by customer type and surprisingly cloud revenues keep growing as a percentage of mix now at 33% of annualized revenues as of 4Q 2019. But if I look at the total enterprise rent or the other enterprise mix that moved to 44.6% on an annualized basis, it was actually down slightly 4Q '19 and it was down a little bit more on a sequential basis in 3Q '19. So could you just give us more color on why this is happening and should the enterprise footprint in your facility on an annualized revenue basis continue to decline as a percentage of total mix or should it be stabilizing around this 33% of annualized rent level?

Well, I think what you're saying Sami is just another data manifestation of the churn versus the sales. I believe and Jeff will correct me if I'm wrong, that the vast majority of the churn that we have been experiencing this year and last year that relates to these older business models has been categorized in the enterprise bucket.

That's correct, yeah.

So as that bucket – those enterprises have shrunk as a percentage of our portfolio and eventually that churn wanes because we just don't have that much more left there. Theoretically, that should see – that should drive up the percentage of our share of enterprise with the caveat that as we saw last year, we are seeing more edge cloud use cases and cloud demand for availability zones in these high performance markets. So that may counterbalance the growth in our enterprises.

And Sami, just to give you a little bit more color around the new logos that are buying from us, about 78% of them are enterprise versus 20% on cloud and a few new networks as well. So this gives you a bit of color as to where the new logos are coming from. And I do think you'll see some big lumps in cloud as that's how they buy.

Got it and then just as a follow up, as you look at your interconnection revenue that keeps outpacing rent growth on a year-on-year basis. What is – would you say some of these new data manifestations or what is driving some of the acceleration that interconnection growth? And if you were to pinpoint on customer types, cloud, network and the other enterprise category, do you think that this is – which customer type is driving this and do you think that this could accelerate given the change in the data manifestation business types.

Yeah, Sami let me give you some information related to that and see if Steve has anything else to follow more broadly. But in terms of that interconnection revenue growth, historically, as we've said, you get about two thirds of that revenue growth that come from pure increase in just the volume of the cross connect products. And then you get your other one third of that growth – really comes from those customers moving from a lower price product into a higher price product, as well as customers renew you get some price increases from those conversations, as well as on occasion, we look at what our list prices are for our various products. So that's what captures that remaining one third. As you think about going forward and where we're seeing some of that growth. It's important to look at ultimately, who our customers are ordering connections to and where we're seeing outpaced growth as compared to overall, those customers connected to our cloud customers. And that has been consistent for the last couple years. It's probably two to three times higher from a percentage basis then the overall growth and that just leads to a lot of the ecosystem that Steve and Paul have commented on, but anything else to add there Steve?

I think you spelled it out well there, Jeff. When you think about enterprises connecting their hybrid architecture, as we've mentioned before, it really does need to be well connected in low latency fashion and with some of the upgrades that we've made in our OCX platform, as well as adding new on ramps to our campuses, that's helped facilitate all of that as well as some of the newer products that we provided as far as inter site capabilities and so forth. So we've seen some good uptick in that product as well.

Got it, thank you.

Our next question comes from the line of Frank Louthan with Raymond James. Please proceed with your question.

Great, thank you. I want to go back to the churn and just get a little more detail. You did mention that you have customers impacted going to public cloud, you think this is going to kind of slow down. Give us some more color on what – so what is that based on? Is that customer feedback or is there a particular vertical that you have the exposure to and why you're confident that after these next couple of quarters, were you calling out the elevated churn that this situation you found yourself in is going to correct itself?

Yeah, Frank, this is Jeff. I'll just provide a little bit of additional color around and hope it I hope it helps. We've obviously spent a lot of time, energy and effort in 2019 just continuing to peel back the onion and better understand some of the contributors to the churn and specifically related to some of those business models. And some of what we found, just to give you some color around it are – as an example, some of the resellers in the marketplace are an example of some of those business models that are not as strong today as they were historically and you saw some of that impacting our 2019 churn, some of its impact in 2020. And then there's other business models that rely heavily on the need and desire for burst capacity various times during the year. Some of those customers rely on the cloud; other customers rely on us in terms of adding capacity. And when you look at that those are the areas that have some exposure to migrating some or part of those to the cloud or to just that may not be as strong and may migrate out of out of our data centre entirely. When you look at and quantify what remaining exposure we have, as I commented on its much lower today than it has been historically, worried about 4% to 6% of our annualized rent that's embedded in our base today. That's why we get comfort that as we look beyond 2020, we believe those churn levels will recede back to the historical levels of 7.5% to 8%. And that's what we're modeling and why we continue to be optimistic in – and in relationship to some of the comments we've provided today. So hopefully that helps.

Okay, great. Thank you.

Our next question comes from the line of Mike Funk with Bank of America. Please proceed with your question.

Yeah, thank you for taking the questions. A couple if I could. Yeah, going back to your comments on SV7, maybe just quickly, can you let us know what the expiring rent is there relative to market rates? Then I've got a follow up question after that one.

We try to not talk about customer specifics. We have mentioned in the past that this lease was signed when market conditions were fairly tight in Santa Clara, but they seem to be fairly tight in Santa Clara today as well. So it's really hard to predict.

Great and then second one, you mentioned the – part of the move driver is a performance centric application maybe not being as performance centric as the customers thought. If you look at some of your higher profile locations, how good of visibility do you have into the use cases of your customers in those facilities, and maybe risk of further churn as customers look at the need to be paying peak rents versus being say hundred miles away.

As we've said in the past, there's always a dynamic related to this, because some customers just don't have that type of visibility, especially when they start – when they're in the startup phase and are scaling up. Having said that, we don't have – we've got this one big chunk, I don't believe that we have anything comparably sized that would be subject to this dynamic elsewhere in the portfolio.

And maybe just kind of comment on some other comments in the quarter. So I guess Cummings on their call mentioned they project 5% to 10% decline in power generation because they're seeing some slippage of construction from '20 into '21. I think Google said they expect to have more of their spending budget allocated towards servers over data centers. So I mean, certainly some mixed commentary out there just about the kind of the strength of the data centre market and demand in general. I'd love to get your thoughts or commentary on that.

So again, we have a hard time seeing what's going on outside our markets and our customer activity, but from what we see demand still seems to be strong. Having said that, if you just look broadly from what we can all see there are markets that have low barriers to entry, saw significant private capital going to development platforms over the last two years and probably got over built. To some extent we've mentioned that in Northern Virginia and the spec construction that took place in 2018 and 2019, which appears to have abated quite a bit. So that may be part of it. And I think you've got similar dynamics going on in markets like Phoenix and Dallas and maybe other markets that are less visible to all of us, but that may be a part of it of what you're seeing there Michael. I don't know that we see any slowdown. I mean, you look at the overall growth trends of the major CSPs and they still seem to be in absolute terms growing the same volume year-over-year.

And your comments imply that maybe some of the speculative capital is pulling back out of those markets that are coming in the last two years. I mean, in the markets where you do compete, are you seeing any kind of specific changes in competitor behavior, maybe less aggressive or more aggressive, any kind of change in behavior?

I mean, the only one where we've really seen a significant amount of that has been that we're in has been Northern Virginia, and I believe we have absolutely seen a pullback in privately funded speculative development. And I don't know, it's three or four of those development platforms have explicitly announced that they're not going to go forward with the spec development they're trying to mutate into a build to suit model.

And then one final one, I appreciate of the time here. Just to clarify, so what is your exposure if any to some of the second tier data centre providers, I guess even guys like Flexential, Cyxtera, Internap, do you have any exposure there as far as leasing space?

I think that's covered in Jeff's comment about how our exposure to that kind of sector in general has declined. That has been in some respects, some of our churn – at least with respect to one of those companies. But again, we don't have much of that left.

Right, that's in the – that's in the low single digits that you called out, the 46% of the business, right?

Exactly.

That's great. Thanks for clarification.

Our next question comes from the line of Eric Rasmussen with Stifel. Please proceed you’re your question.

Yeah, thank you. So in terms of your 2020 guidance, how should we think about the range of your revenue guidance 600 to 610? What can drive that to the lower end of that range? And then what are your thoughts about the company's ability to return to low double digit growth in possibly 2021.

Hey, Eric, it's Jeff. Let me just give you some feedback on your first question related to revenue. And I think, similar to what Paul alluded to in his prepared remarks, we've got the capacity today as compared to the last couple of years, where I think it gives us the agility to compete more effectively, where we can see some of that demand. As some of the demand continues, I just think it gives us the capabilities to compete more effectively especially for some of those scale deals that might in the marketplace, so I think we're optimistic about that. But we got to see ultimately how the market evolves and where those opportunities arise. And as we've said, historically, those tend to be fairly lumpy. In terms of your question longer term the ability to return to double digit growth. I would say that as I look near term and call it the next two to three years, I think we're optimistic to get back to those high single digit growth rates, assuming we continue to execute on some of the priorities Paul laid out and assuming that the churn received back to those levels we've seen historically, and I think we're optimistic about getting back to those high single digit growth levels in the near term.

And I'd just add – remind you the comments I made earlier that we're bigger cut made, the denominator is bigger, the industry is more mature and yet relative to other reach sectors I think our prospects for recurring annual growth over the intermediate term are better than probably all the other ones I can think of or at least most of them I can think of. And I think that's a good business model.

Great, thanks and then VA3 seem to make a nice progress despite no scale deals in the quarter. Are you starting to see any movement where leasing dynamics are improving where you would potentially participate in this market when it comes to larger deals?

Yeah, I would just say you're exactly right. I mean, our fourth quarter was actually the best quarter we've had in two years. And if you look at the makeup of that as you look at the overall numbers, it's all retail type of business, good enterprise business that we've been selling there. So we've been able to really sell effectively into our first phase there and now into second phase. And we have seen good pipeline and good results there. So we're optimistic. And as the opportunities present themselves both in retail and scale there is scale opportunities out there that value that type of platform that we both now have the capacity for, as well as the interconnection for those low latency applications that exist out there. We just – as we've said, in the past, we are trying to be diligent and pragmatic about those opportunities and ensure that they add value to the platform and deliver the returns that we expect.

Thanks and maybe just last, just following on that theme. Obviously, last year was a great year for your scale colo business. I mean, I think it was up 4X of what it was a prior year. Is that at where we are today – is this sort of an achievable target for the company this year or it's still too early to tell when we're waiting on sort of some of these key markets to come back?

Well, as Jeff mentioned, we're optimistic about the growth opportunities that we see ahead of us. So it was a record year for us. And as you can back into the math of our current guidance, we have not expected that same type of result in our current sales mix for 2020. But we're optimistic about –that opportunity for us, but those deals and that leaking is very lumpy as we've depicted in the past and so trying to forecast that type of growth is challenging.

Thank you very much.

Our next question comes from the line of Nate Crossett with Berenberg. Please proceed with your question.

Thanks, guys. A lot’s been answered already. But what are you guys expecting for interconnected pricing going forward? So it looks like MMR per Cabinet, the growth rate has trended down I believe. What's kind of a stabilized long-term growth rate that we should be assuming for that?

Hey, Nate as you look at that, MRR per Cabi and you look at the results for the fourth quarter, really what contributed to the – I think it's 4.1% overall growth rate year-over-year is the largest contributor to that was the interconnection revenue growth. And historically we've been in that mid-single digits for that MRR per Cabi growth, I would expect that as we think about 2020 to be probably low single to mid-single – low to mid-single digits for 2020 and largely just because of the guidance we've given around our mark-to-market on renewals being a little bit lower for this year. So that's how I would think about it as you head into 2020 given where we are today.

Okay, and then you just go on that cash renewals, what was the 2020 guidance and then if you stripped out some of the larger roll downs? Or is it just across the board roll downs everywhere? Just trying to get a sense of normalized kind of pricing?

Yeah, I think it's – I would look at it from the perspective of what we experienced the last two quarters, which is both in the third quarter and in the fourth quarter, we had a handful of customers, where we had some roll downs and Steve alluded to it in his prepared remarks and that's dependent upon their size, is dependent upon the term of their leases and more specific into the market, which those renewals are. Once you strip those out, the pricing in both third and fourth quarter was positive. I think it was 3.4% last year – I'm sorry, in fourth quarter and I think it was 2.8% as my recollection the third quarter. So that's the dynamic you're seeing. And I think that dynamic will continue as we go through 2020.

Jeff, would it be helpful for him to understand for every hundred basis points of churn or combination of churn in mark-to-market what that means in terms of dollars for us? And that falls to the bottom line?

Yeah and I think – Nate, specifically, I think what Paul is alluding to is if you just think about our 2020 guidance and forecast, when you look at our churn from 2019, into the year 11.1%, if you get that down to a more normal range where we have been historically, so just strip, four percentage points off that to make the math easy. That type of elevated turn in '19 and that four percentage points adds another $12 million or roughly $0.25 per share to our bottom line in 2020. And I think that gives you an indication of some of those headwinds we're face headed in 2020. And it makes it real when you start to quantify it from that perspective. Is that what you're looking for?

Yeah, I mean, that's helpful. Just completely different topic now, what do you guys doing on the ESG front in terms of commitment to clean energy? What are the targets you guys have put out? And then maybe can you speak to what the financial impact would be or could be by going, for example, 100% green?

So, first of all, I just refer you to the ESG report that we put out every year. Our primary focus quite honestly has been on energy efficiency dollar for dollar, that's the largest impact on our and our customers' greenness. And frankly, we're already providing a lot of that greenness to customers. We know one customer for example that is able to qualify their new compute environment as one of the 10 greenest in the world just by moving into our more power efficient data centers from their on-premises deployment. So that's where our most of our focus has been, as we said in the past, we will purchase renewable energy, where it is allowed by the local utility and where the cost is basically competitive i.e., will not adversely impact our customers. And that tends to be the focus of our customers and that is what we have been – that's what we've accomplished. We also quite honestly encourage our utilities to generate and allocate other low carbon or non-carbon sources of generation. But because we don't have control over the power mix in our more regulated markets, we don't have any specific targets, but we do advocate for non-carbon and renewable generation sources and we do as often as they allow us and we can do so competitively purchase and wheel in those sources of energy.

Okay, that's helpful. Thanks, guys.

Our next question comes from the line of Richard Choe with JP Morgan. Please proceed with your question.

Hi, I just wanted to clarify on the annualized cash rent growth, the five customers and you mentioned the 2015 vintage, is this something that's going to last just through 2020? Or is it going to last for a few years since it just started in the second half of last year, just want to get a sense of how long you might be facing some of these roll downs going forward.

Yeah. So Richard, as I said, I think in response to an earlier question, it's hard to predict beyond 2020 because you don't know what supply and demand will be in each market and what that will mean for market pricing in each market. But we do believe it's like a lot of – like even the churn that we're experiencing is that you do go through these vintages and then eventually the effects of a certain vintage are two go away. And so we don't really have visibility beyond 2020. But if history is any guide, we should see some improvement in these areas going forward past 2020.

Hey, Richard the only thing I'd add is on page 15 of the supplemental, just take a look also at our lease expiration table, it'll give you an idea what those explorations are on a per square foot and compare that to our current pricing. And then just monitor that as we move forward. I think it gives you a better idea maybe where that's going to be longer term.

Got it, thank you.

You bet.

Our next question is a follow up question from the line of Jordan Sadler with KeyBanc Capital Markets. Please proceed with your question.

Thanks. I was – just want to dial in a second on California for a second. Could you give what you think market pricing is for scale in Santa Clara these days?

Don't know if you want to – I don't know if there's – it's difficult to disclose what market pricing is. There are so many factors that go into a price for a given customer, depending upon their density, the size of the deployment, as well as their interconnection requirements. So that's why the moment of silence there because it's really difficult to kind of quantify a given market price given all those different factors.

Is it feasible for a scale requirement to be better than $250 per K Debt [ph]?

All in is that you're talking about?

Yeah. Yeah.

I guess it's –again, it just depends on the deployment. And there's a lot of different ways to configure a pricing parameter around a different given customer based off of their power draw, their density levels, all those different factors. So it's just challenging for me to even give you a straight answer.

And honestly, we encounter different dynamics than other companies do because our ecosystem has a different value than campuses that don't have the same density and diversity of customers who are using all the various services and want to do it with the lowest latency or zero latency in essence highest bandwidth, highest security for exchanging data. And so there's – it's like saying in other forms real estate what's the market price for space in particular mixed use development. A lot of that depends on what the mixed use development is compared to what other alternatives are.

The only other thing I would just kind of – my comments with this, as you build a contractual agreement with a customer, there's a lot of factors that go into that part of it as size and density and so forth, but also expected power draw from that customer and how you manage that in relation to the rest of the draw of that computer room and the building. So it really is very challenging, really impossible to give you just a market price without knowing the exact parameters of the customer.

Okay, and then this is for – maybe for Jeff. Jeff, have you thought about the potential exposure to Prop 13 for CoreSite?

Yeah Jordan, we have looked at it over the last couple years. And basically when you look at that portion of our property tax expense that's isolated to specifically California and then more specifically, to those leases, where we do not have the ability to pass through the cost increases. It's limited to about $3.5 million that sits inside our portfolio that's equal to about 15% of our total property tax expense. And so what's important also about that –

Your direct exposure to California property taxes is 3.5 million.

For that portion in which we cannot pass cost increases to customers.

Right, right.

Yeah, that's correct, so 3.5 million and just to give you some other data around that I think helps as we look at this, the leases associated with that $3.5 million, they're on average ever remaining lease term about two and a half years. So obviously, we'll have plenty of opportunities near term to have – continue to have conversations about maintaining them and pricing it accordingly. But we're watching that closely and continue to monitor the overall effects to our business.

Jordon I'd only add that these are estimates based on what we believe the impact on property tax valuations would be, obviously, if this thing passes and it gets implemented, latter part of 2021 into 2022, some of those things will be determined more specifically.

Right, right. Okay. Thank you.

Our next question comes from the line of Michael Bilerman with Citigroup. Please proceed with your question.

Hey, thanks for sticking on. So I just want to go back on the churn just in terms of – as you discussed last year it caught you by surprise and you have spent a lot of time digging into it. Jeff you peeling back the onion to really understand things. What have you done from an organizational perspective in terms of changes, in terms of reporting lines? Maybe people who've left the organization, the people who've joined the organization, I guess other than just sucking it up and having a lot of churn last year and this year, what are you doing differently to get the Street comfortable that this won't be a recurring problem?

Well, obviously Jeff and his team, as you've mentioned have done a lot more digging and analytics around the customer base. But we have also made it a high priority with all of our general managers in each market, working collaboratively with the sales team to have specific responsibilities and reporting around churn, risk, expectations, what's being done with respect to each customer. We've laid out a new rubric for determining how to address customers, when they come in at renewal and want a pricing change or something like that. When do we give that when do we not? The main focus being at least not be surprised by it, but having said that, I think even had those efforts been in place a year to 18 months ago or those new processes in place a year to 18 months ago, I don't think it would have changed our experience that much because quite honestly, a significant part of the – probably almost all of the unexpected churn is churn that really wasn't to be expected based on lease terms. Customers turned out before their lease term was up or even customers that at one stage in the discussions were planning to renew and then somewhere late in the game changed their mind. So we should be able to have – and I think the biggest comfort, as you said for the Street is just the fact that a lot of these business models have significantly declined, as Jeff described as a percentage of our portfolio.

So I guess there's an element of the leasing decisions initially, right? So in the banking world, we're all familiar with claw backs when something you do turns out to have not been the right thing. And so I wonder just on your initial leasing is there changes there that need to be made in terms of the types of customers you're bringing into the portfolio that may have a higher risk of churn in the future, right. So are you making better leasing decisions today? And also, it sounds like you're taking it a lot seriously, trying to manage the outgoing customers from churn.

So I think we're making good leasing decisions today. I can't really say that they're better because when you look at some of these customers that came in and some of them have been in place for seven, eight, nine years, had vibrant business models at the time. And I think neither they nor most people really expected the significant disruption from cloud that started occurring four, five years ago. So that's always a part of this business. It's why we stress the importance of constantly acquiring new logos. That's why we've tried to build great relationships for edge cloud use cases with the major CSPs who are good to work with and are obviously good credits. So yeah, I think that as we work through these churn episodes we come out with a stronger portfolio. But I think the company has always been pretty careful about its leasing decisions and that shows up in our very low bad debt expenses.

Right and just specifically on the 10% for next year, so it sounds like 370 in the first quarter, 250 for a specific tenant in the fourth quarter, which leaves about 380 spread over the second, third and fourth, just from what I guess I would call normal, so 120 basis points a quarter. Is that – just to make sure I got the math right, is that right?

Yeah, I think that's close Michael. I would say 350 was our midpoint for the first quarter, so you were just a little bit high there. But that might leave you about 400 basis points for the rest of the last three quarters.

And is that – I guess, just knowing that number, that enough, right. I mean, I guess I'd be conservative enough on the level of an anticipated churn that normally would occur outside of what you know today, right. And it's going to be really high in the first quarter. What's to say that it won't stay that an elevated level through the second, third and fourth quarter, right?

Right, yeah. No, it's obviously something we look at as we're going through all of our renewals over the next call it 24 months and it's our expectations that it would be lower. Typically on a typical quarter for us it's usually 1% to 2% is what we've seen historically. So those numbers don't sound out of line with what the remaining portion is.

And Michael, just as Jeff mentioned earlier, the higher churn in the first quarter is basically one customer and the higher churn in the fourth quarter is another customer. And absent that, we don't see any other customers of significant magnitude – of similar magnitude that are in that category.

Obviously, closer to those quarters we get better visibility right.

Okay, alright, thank you.

There are no further questions in the queue. I'd like to hand the call back over to Paul Szurek for closing remarks.

Thank you. Thank you all for your time and your interest today. Those were a lot of good questions. I just like to wrap up by thanking my colleagues throughout the company for all the progress they've made and proactively and efficiently building new capacity, expanding our sales to new enterprise customers, and supporting customers with new connectivity products and exceptional operations. I'm very fortunate to work with this group of people. And just to recap, we believe the ongoing benefits of these improvements along with the secular tailwinds for our campuses should lead to a resurgence of growth if we outpaced churn for our expectations. We look forward to working to achieve that future and we hope all of you have a great day. Thanks very much.

Ladies and gentlemen, this does conclude today's teleconference. Thank you for your participation. You may disconnect your lines at this time and have a wonderful day.