Equinix Inc (EQIX) 2003 Q3 法說會逐字稿

Hello, and welcome to the Q3 earnings release conference call. Following today's presentation, there will be a question-and-answer session. At that time, instructions will be given should anyone wish to ask a question. Until then, all lines will be in a listen-only fashion. At the request of Equinix, this conference call is being recorded.

I would now like to introduce today's host, Director of Investor Relations, Mr. Jason Starr. Sir, you may begin.

Good afternoon, and welcome to Equinix's quarter ended September 30, 2003 earnings call. Before we get started, I would like to remind everyone that some of the statements that we will making today are forward-looking in nature, and involve risks and uncertainties. Actual results may vary significantly from those statements, and may be affected by the risks we identified in today's press release, and those identified in our filings with the SEC, included in our report of Form 10-KA filed on April 25th, 2003, and our most recent report on Form 10-Q, filed on August 4th, 2003.

With us today are Peter Van Camp, Equinix's Chief Executive Officer; Renee Lanam, Equinix's Chief Financial Officer; and Keith Taylor, Equinix's Vice President of Financing and Principle Accounting Officer.

At this time, I'll turn the call over to Peter.

Thank you, Jason. Let me begin today by reporting that we achieved our objective of operating cash flow positive for the month of September. And I am pleased to report that this was the case for the entire third quarter, as Equinix generated $1.9 million in operating cash flow. This clearly supports the message from our last call of the Company hitting an important inflection point in our development.

Now, let's look at some of the other highlights. Total revenues for the Company was $30.9 million. This represents a 53 percent increase over the same quarter last year, and a 9 percent increase over Q2. Equinix closed 60 new customers in the quarter, with 45 of these coming from the U.S. Key wins include Amazon, Apple, Comcast IP Services, Deutsche Boerse AG, Macromedia and Standard Bank Asia Ltd. We saw 49 percent of our U.S. customers order additional services in the quarter, including EDS, IBM and Microsoft. We ended the quarter with over 650 customers. U.S. interconnection services revenue increased by 17 percent over second quarter, growing to 22 percent of U.S. recurring revenues and 19 percent of total recurring revenues. For investors tracking EBITDA performance, the Company was a positive 1.4 million for the quarter. Now, this is calculated by taking our loss from operations, at 14.3 million, and adding back non-cash charges for depreciation, amortization, accretion and stock-based compensation. This result includes a $1 million property tax charge taken in the quarter, which Renee will expand on later.

Now, let me provide some additional color on customer wins for the quarter. This was another great quarter in bookings, which included some sizable commitments, including Amazon, Macromedia and the U.S. government. We do expect to see the U.S. government be an important contributor to growth through 2004, as Equinix was just recently awarded its GSA approval. In Asia, I am also pleased to announce that we have expanded our IBM relationship, with a large commitment in Singapore. We're thrilled to be able to work with IBM in Asia, and we feel it's a strong statement of our value, our service quality and the trust that they have placed in our relationship.

As we mentioned in our last call, we continue to see the network-neutral model win for us. This was instrumental in a number of key deals in the quarter, yet there's another aspect of Equinix that our investors aren't necessarily close to, but our customers certainly are, and that's our service quality. Even in a crisis such as the East Coast blackout, or Hurricane Isabel, there was no impact to our consistently high service levels. The East Coast power outage looked no different to IBM or any other customer in our Newark or Secaucus IBXs than the standard weekly routine of bringing up our backup systems to assure their reliability. And service quality is something more than just an infrastructure issue; it's about process discipline, proactive customer communications and the responsiveness of our team. As I'm out talking to customers, it's very gratifying to hear them rate us highly in each of these aspects. The direct customer feedback around the East Coast power outage, especially from those customers who have come to us from other providers, was a clear example of this. Beyond neutrality, our service quality has become a real differentiation point in a consolidating industry, where our competitors' focus is uncertain. So much of this business is about trust.

Touching on industry consolidation for a moment, I'd like to take the opportunity on this call to talk about the strategic and tactical rationale behind our announcement today regarding the takeover of Sprint's Silicon Valley center. First, from a tactical perspective, Silicon Valley is without question one of our most successful markets, and our pipelines in this market remains strong. With our success in San Jose, we recently began looking for alternatives that would expand our growth opportunity in the Silicon Valley. In a consolidating market, we were able to look at a number of different, fully operational centers. Following our review and significant diligence on a subset of these facilities, we found the Sprint center to be the best answer, consistent with our standard and consistent with three criteria for an IBX. That's proximity to customers and fiber routes of network providers, security and the power, design and reliability to support our customers' increasing power requirements. I believe both Equinix and Sprint are thrilled with this transaction. Right up front, as we began discussions, Sprint placed their customers' well-being as a first priority. As outlined in their letter to customers announcing this transaction, they found Equinix possessed, and I quote, "Number one, proven capabilities to provide collocation services, facility management and high-quality customer services; two, commitment to insuring a seamless transition with undiminished services to the existing customer base; and, 3, financial stability."

Strategically, this acquisition makes a lot of sense. It demonstrates that we have opportunities to expand our growth footprint without significant capital spend in datacenter builds. One by one, other datacenter operators are discovering that success in our business was not merely a matter of building it and they will come. Equinix, unlike other providers in this space, is seeing growth and success. We have continued to grow quarter on quarter, as a result of our network neutrality and superior customer service, and we expect other opportunities of this nature to exist. However, we will be selective and committed to our objectives in increasing cash flows and profitability in any opportunities that we consider.

Now, to outline more of the specifics, I'd like to turn it over to Renee. Then I will come back to cover some other recent developments and outlook for fourth quarter and fiscal year 2004.

Thanks, Peter. Well, without a doubt, this was another strong quarter for Equinix. Before diving into the numbers, let me touch on two key events -- first, the Company's achievement of positive cash flow from operations for the entire quarter and, second, the acquisition of the Sprint center.

First, on the cash flow announcement, let me put some perspective on this accomplishment. On December 31st of last year, we closed the acquisition of Pihana and i-STT. At that time, we said our goal was to be cash flow positive from operations as an entire company by the end of 2003. At that time, this seemed like a formidable goal, given the integration challenges of two newly-acquired companies and the work required to gain the synergies out of this combination. Of course, at that time, Equinix's operations, on a stand-alone basis, were still not yet cash flow positive. Last quarter, we announced that we were ahead of schedule on our consolidation efforts, and we accelerated our commitment to reach cash flow positive from operations by an entire quarter, to the end of Q3. Today, we're very pleased to be able to announce that we exceeded that goal by hitting cash flow positive from operations not only by the end of Q3, but for the entire third quarter. We credit our success to the ability of this Company to rally behind aggressive goals. This is true from those responsible for our topline growth, the sales teams, who delivered another great quarter of new bookings, it, to those responsible for our largest expense line, the operations team, as well as those in our Asia/Pacific headquarters, who came in under budget and who continued to look for ways to trim discretionary spending.

This achievement for the quarter is even more significant, given two other things that happened in the quarter that had a negative impact on our operating cash flow. First, as I mentioned on the call last quarter, we decided to close down our Thailand operations. While we took the charge it Q1 of this year, which is when we made the decision on the shutdown, we did not finalize the shutdown until this quarter, with the payment of approximately $500,000 towards the close of these operations, as the cash outflow in the quarter that we had to absorb on the operations cash flow line. Second, in the quarter, we made a cash payment of just over $900,000 in connection with local property taxes. On a side note, we also increased our tax accrual for our IBX properties by a further $1 million. This decision was based on the advice and analysis of our local tax advisers, and represents what we believe is a prudent approach to property tax estimation. The lack of visibility in this area is due largely to the high level of discretion local authorities have in ascribing property values, exacerbated by the current fiscal squeeze that many municipalities currently face. Given these factors, we're confident that our current reserves are adequate to cover tax exposures on our IBXs. The fact that the Company absorbed both the Thailand shutdown costs and the Chicago tax payment in the quarter, and still hit its cash flow target is what is so impressive about this accomplishment.

Now, let me provide some additional color on the planned takeover of Sprint's Silicon Valley center announced today. The agreement is structured such that we will sublease the entire center, and we'll take ownership of the property, equipment and improvements in the center. The building has 160,000 square feet, slightly larger than our current San Jose IBX. While it has capacity for approximately 3,000 cabinets, it is currently built out to support just over 1,000 cabinets. One of the attractions of the center is that we expect to get the center profitable without incurring CapEx to build out the additional cabinets. We estimate that for us to have built out the center to its current 1,000-cabinet capacity, we would have spent more than $50 million. By not having to incur these costs, and because we expect to have customers in the building from day one, we anticipate that the center will reach cash flow breakeven by the end of Q1, 2004. To acquire the center, we will not incur any debt or use any of our cash initially, as we simply assume the sublease at current fair market rates. We are in negotiations with the landlord to enter into a direct lease at the expiration of the sublease. As always, we intend to put in place multiple options to extend the lease. We expect to close this transaction on December 1st. Going forward, we plan on putting 3 to 5 million of capital investment into the building in the first year, to Equinize (ph) the building and to increase the power capacity for the existing 1,000 cabinets. As we fill up the existing cabinets, we will evaluate whether it makes sense to incur additional CapEx to complete the build-out of the center. However, as I mentioned above, even if the center is projected to be profitable with the current 1,000 cabinets, this is a decision that does not need to be made today.

Now, let me move into the numbers for the quarter. Consistent with past calls, I will review the numbers on both a cash basis and a non-cash basis. In presenting our results on a cash basis, the non-cash expenses that I will separate from our results are depreciation, amortization, accretion, stock-based comp and non-cash interest expenses. This will give you a better sense of our true cash spending levels. The financial statements attached to today's press release contain a similar level of detail.

So, let's start with revenues. As Peter mentioned, Q3 revenue was 30.9 million. This is a 53 percent increase against the same quarter last year, and a sequential 9 percent increase over Q2. Total recurring revenues for the quarter were 28.4 million, or 92 percent of the total, up from 91 percent last quarter, and were comprised of cabinets, power and interconnection services. This recurring component is a 78 percent increase over the same quarter last year, and a 9 percent increase over Q2. Nonrecurring revenues for the quarter were 2.5 million, or 8 percent of total revenues. Of our nonrecurrent revenues, 2 million was from professional services and installation fees, and the remainder was related to a contract settlement with one customer in Asia/Pacific. Excluding any additional large customer settlements, we expect 95 percent of our revenue base to come from recurring revenue sources. Breaking out our revenues by region, 83 percent of our revenues came from our U.S. centers, while the remaining 17 percent were derived from our Asia/Pacific centers. One customer, IBM, represented 15 percent of our Q3 revenues. No other customer accounted for more than 5 percent of our revenues.

Moving on to churn, to remind you, we calculate churn based upon the number of cabinets returned to us by our customers in the quarter, as a percentage of the total cabinets booked at the beginning of the quarter. On our last earnings call, we mentioned that we were in the process of finalizing an agreement with one of our customers in the Asia/Pacific region to reduce the size of their commitment in Singapore, in exchange for a settlement fee and an agreement from them to extend their commitment to Equinix and its other Asia/Pacific IBXs. While we are pleased to have finalized this settlement, the net reduction in cabinet commitment increased our churn for the quarter to 6.8 percent. Excluding this specific event, churn for the quarter was 1.6 percent, well below our expected average of 3 percent per quarter. Going forward, we expect to see churn remain in the range of 3 percent. Continue to expect virtually all of our churn to be a direct result of customers with financial issues, or who have been acquired. We just don't see churn from customers leaving our centers to go to competitors.

Moving to our DSOs, company-wide DSOs were at our best reported level of 27 days, from 33 days in Q2. We expect to see this normalize in the 35 to 40 day range going forward.

Cost of revenues for the quarter, excluding non-cash expenses, were 19.7 million. Cash cost of revenues consists primarily of salaries and benefits for IBX personnel, power charges, rent and security expense. Total cost of revenues for the quarter, including non-cash costs, were 33.3 million. Comparing these results to previous periods, our cash cost of revenues increased by 13 percent quarter over quarter. The majority of this increase was due to the increase in property taxes I discussed earlier.

For the quarter, our cash gross profit margins were 36 percent, as compared to 31 percent for the same quarter last year and 39 percent in Q2. This quarter-over-quarter decrease is primarily a reflection of the additional tax accrual we took in the quarter. As for this accrual, our cash gross margins for the quarter would have been right at 40 percent. Moving to Q4, we believe our cash gross profit margins to normalize and to be in the 40 percent range. Looking ahead to 2004, we expect to see this range continue to increase, as we are now through the inflection point on operating profitability.

Of our approximate 21,000 cabinets, 33 percent were billing at the end of Q3. This was flat, as compared to last quarter, as a result of a large customer in Asia who, as expected, reduced its commitment. This did not have an impact on our revenues, however, as we have been treating this customer on a cash basis since the beginning of the year. In fact, we were still able to increase our recurring revenues in the quarter by 9 percent without using up any more of our cabinets. This means that we continue to see increased revenue per cabinet, which gives us that much more headroom for growth, both in terms of revenues and margins.

Moving back to the expense line, SG&A expenses for the quarter, excluding non-cash items, were 9.8 million, an 8 percent decrease from the prior quarter. We expect our SG&A number to remain fairly constant over the next quarter while revenues continue to climb. Our net loss for the quarter, excluding non-cash expenses and non-cash interest, was 1.1 million, a 53 percent improvement over Q2. Including all non-cash charges, our net loss for the quarter was 19.7 million, or $2.12 per share, a 13 percent improvement over Q2.

Turning to our balance sheet, our cash, cash equivalents and short-term investments, or our unrestricted cash, on September 30th was 25.2 million. This is up by $1 million from our cash balance at the end of last quarter. Capital expenditures for the quarter were $1.4 million.

Next, moving on to our statement of cash flows, our net cash generated from operating activities was 1.9 million. In our press release today, we provided guidance on, among other things, operating cash flow for Q4. Peter will provide more detail on this guidance in a moment. One thing I want to touch on, however, is you will note our operating cash flow for Q4 is expected to be approximately $1 million, which is down from Q3. The reason for this is we will be making a $2 million bond interest payment in Q4. We pay interest on our bonds semi-annually, so no payment was required in Q3. In addition, we currently expect to liquidate approximately 1 million of property tax accruals in Q4. This means we will see an additional 3 million in operating cash outflow that we didn't see in Q3. Our net cash used in investing activities was 1.4 million, primarily related to the funding of capital expenditures. Our net cash generated from financing activities was 432,000. This was primarily a result of proceeds from employee stock plans, offset by monthly principal payments on our debt facilities and capital lease obligations.

Finally, with respect to our equity capital structure, as of September 30th, we had 9.4 million shares of common stock outstanding. Assuming conversion of our preferred stock, as well as the outstanding convertible notes and the warrants associated with those notes, we would have had 18.1 million shares outstanding. In addition, as of September 30th, we had 3.8 million options and other warrants outstanding.

This concludes my prepared remarks. Let me now turn the call back over to Peter.

As everyone on this call has hopefully gotten the message, we have had another solid quarter in Q3. 2003 is really shaping up to be a great year for Equinix, one that is frankly pretty satisfying for the team, as we have all worked so hard, in the face of an extremely difficult and volatile environment, to build a lasting enterprise. With our long-term success in mind, and as I noted earlier, we may encounter new opportunities, perhaps similar to Sprint, or other possibilities where an acquisition is attractive in this consolidating market or, of course, the opportunity to further delever the Company is something we're always evaluating. As I imagine most of you saw, the Company filed a $150 million shelf registration two weeks ago to put the flexibility in place to take advantage of opportunities such as these as they arise.

Now, let me offer our view on the outlook for the Company for the next quarter, and then a high-level view of 2004. This guidance assumes that we close the Sprint center. For the fourth quarter, the Company expects revenue to be in the range of $32 to $33 million. Cash generated from operations will be approximately $1 million. As Renee noted earlier, Q4 includes our semi-annual bond interest payment of $2 million. Cash used in investing activities, consisting primarily of capital expenditures, will be approximately $4 million. This includes the use of success-based capital attributable to IBM's deployment in Singapore. Cash flow used in financing activities will be approximately $1 million, representing payments on capital leases and other debt obligations.

To tie up the full year, revenues are projected to be in the range of 116.5 million and 117.5 million. This is a 52 percent increase year over year. Cash used in operating activities will be less than $20 million. Total cash, as of December 31st, 2003, is expected to be greater than $20 million.

Moving into 2004, one of the strengths of Equinix's model is the recurring nature of our revenues and the relatively fixed nature of our costs. Due to this visibility, I'd like to provide some high-level guidance on next year. Total revenues for 2004 are expected to be in the range of $150 to $162 million. Our cash gross margins are expected to be in the range of 45 to 50 percent. While expenses for SG&A, excluding non-cash depreciation, amortization and stock-based compensation, are expected to increase a nominal 3 to 5 percent over 2003 levels. As we head into the close of a great year in '03, the fundamentals of the Company are as strong as I've ever seen them. Operationally, financially and strategically, this has provided us with the ability to gain key new customers and marketshare, while adding a new footprint in an important Silicon Valley market. 2003 has already been a pivotal year for Equinix, setting the stage for an exciting 2004.

Jennifer, we can take questions now.

(OPERATOR INSTRUCTIONS). Tom Watts (ph).

Congratulations on the great quarter. Just a couple of questions regarding the Sprint acquisition. Were there any customers in place in the facility? Had they all been moved out, or had Sprint lost them all before the sale?

No; we expect customers to come with the facility.

And what impact to revenue run rate would that have? Is this a lot of customers, or how many customers?

Tom, I think the number of customers we're still trying to tie down. I think the guidance we gave today, though, does take into account the customers we're expecting to come over.

So you are in the selling process of that right now?

Correct.

In terms of the actual closing of the transaction, when did that occur -- or will it occur?

December 1st.

And how does this facility relate to your other facility in Silicon Valley? Is it differentiated in some way? Or is it just such a good deal that you couldn't pass it up?

Well, it was an attractive deal. But as I mentioned, we did have a chance to look at a number of facilities as we came to this one. And there were some clear differences in it. One was its power design and ability to support what was an ongoing trend in power demand from customers, is blade servers, and just more high-density implementations require greater power. This facility was well designed to accommodate that. Beyond that, it's in a very fiber-rich area, by its location. So although, out of the gate, obviously, Sprint is the network in the facility, we see an opportunity to just maintain our network-neutral model very well here. And then, we also have a plan in place to connect the center with our San Jose facility, that will give us the full diversity of our networks there, so certainly, it helps our interconnection business, and the ability for those customers to take advantage of anything we've done there. So it's really a good fit for us.

As you suggested by your comments, there are also a number of other distressed facilities out there, that are on the market. One, is there a window of opportunity for you to capture customers from some of those? And second, what do you think the impacts will be on the market once either other Sprint facilities or excess facilities our others come back (indiscernible)?

Well, it's interesting. Just the glut of properties we were able to look at in the Silicon Valley alone tells me that that event has already happened, or will happen shortly. I think we've already experienced the impact of the glut of space, because we just looked at a number of centers that are just lying dormant right now. And obviously, because of our growth from share shift, we have seen the benefit of a lot of customers who have just already made the decision, and we've earned those customers. So I just think the quarters ahead still hold a great opportunity for us to gain share shift while the market just rationalizes itself.

And just one final question. You mentioned your GSA approval. Can you talk a little bit about your U.S. government marketing efforts, what types of opportunities you see there, and how soon we should see contributions from that?

Yes. I think we are largely getting rolling there. We do actually have a few government wins under our belt, but obviously, pursuing the exercise of GSA approval is a statement that we believe this is an opportunity for us. We have a sales office in Reston, Virginia that's turning focus toward government now on a lot of their activities. So, we aren't quoting, by the vertical, any specific percentage of revenue numbers at this time, but certainly, a greater emphasis by the Company to go after it.

Well, thanks very much, and congratulations on getting to your cash flow goals.

Vik Grover, Needham & Company.

Hey, guys, good quarter. Just a couple of follow-ups on the Sprint transaction. Does it have other fiber backbones built to the facility, or will you need to get other backbones to crack that facility? Or will you possibly just back all the traffic to the neighboring IBX in the region to get it carrier-neutral? And was there any impact on the quarter from the blackout, in terms of higher line costs or higher operating costs? I imagine running those facilities night and day on your own power must have cost something. So any kind of feedback there? And then I might have a follow-up.

First off, on the Sprint facility, in fact, there are other fiber providers than just Sprint already in the building. But its location also lends itself very well -- just literally passing outside the door are a number of fiber providers, which will be a simple economic justification for those to build in. Clearly, we see an attractive opportunity to gain the benefit of our interconnection taking place in San Jose, so we look to provide connectivity through that means, as well. But we feel really pretty good about fiber and bandwidth into that building.

Then, specifically on the service outage, although certainly some more fuel than typical was used, it's not a material impact on our cost line, and supporting what happened. Both the IBX and Secaucus, as well as Newark, experienced a disruption in their normal power supply, not in our service quality, obviously, but we did have to shift to backup power in both of those facilities for an extended period of time. And so we did use fuel in that effort, but they transitioned very successfully, no experience to customers. And as I said before, not any different than what we do every week to just test our backup generators. So as a result, you'll hardly notice this on any cost line; this was just the weekly event.

On the STT relationship -- obviously, there's a lot going on at Singapore besides just IBM. Have you had any preliminary discussions with them with regard to how they might approach working with Equinix subsequent to them closing the Global Crossing transaction? And as a housekeeping, I just wanted to clarify -- on the EBITDA line, if we add back the accruals for taxes, are you guys suggesting that your EBITDA would have been about 2.6 million for the quarter?

Okay. First, on Singapore Technologies and Global Crossing, we have talked to them a lot about it, as you might guess. The reality of the business model and what they invested in is certainly our neutral position. It is something they respect a great deal. So it's provided, certainly, an entree for a lot of conversations with Global Crossing, from a customer relationship standpoint. And we've actually seen them come into some more centers with us as all this conversation began, but the reality is that the importance of neutrality in our model is such that we do see them as an important customer. We see them offering bandwidth and value to our other customers in the facility, but you won't see us extend beyond that position. And then, lastly, on EBITDA, the number you had offered was 2.8 million, I think.

The actual EBITDA in the quarter was 1.4 million positive. The tax accrual was about a million. But for that, it would have been 2.4 million positive.

But that also includes the 0.5 million from shutting down Thailand, as well, right?

Correct. But we had accrued for the Thailand in Q1, so I (multiple speakers).

Okay, that was already accrued. Thanks a lot, guys. Great quarter.

(OPERATOR INSTRUCTIONS). Andy Schroepfer, Tier 1 Research.

Good quarter, everybody, excellent work, and congratulate Phil for me. To start off, your recurring revenue number was just under my aggressive number for the quarter, but your inflation (ph) revenue was stronger. So I'm modeling directionally correct, that would mean you had a lot of installations that didn't fully get into third quarter on the recurring side. Is that a fair way to look at the quarter?

I think it's an interesting way to look at the quarter, Andy. We haven't actually analyzed it that way. I don't think that that's the impact we'll have. I think we hit the high end of the range on the recurring revenue and the total revenue for the quarter.

It was definitely strong. On the fourth-quarter guidance, you said Sprint is in there. But I assume they are only in there for one month of the quarter. Is that correct?

Yes.

Out of the 2004 outlook, you had to break out the incremental revenue from '03. How much is coming from existing customers versus new customers? Is there any rough guidance on what you think is happening there?

No. I guess that's one we can look at for you. We'll have to go back and look at the buildup to give you, maybe, something specific on a future call. But I expect the trends that we've experienced will be similar.

I think it was the fourth quarter of last year, you gave a comment that I think it was 75 percent of new customers came from customers that were with other providers. Any kind of stat you can give for third-quarter wins? How many came from customers that were with another hosting provider over to you guys, versus new customers that were in-house?

Actually -- and this includes, I believe, the bookings within the quarter from current customers -- 37 percent of bookings came from competitor wins this past quarter.

And that's counting it on a cabinet basis, right?

Yes.

So it might be higher on a customer count?

Yes.

On the CapEx side for fourth quarter, if took out the additional CapEx for IBM and Singapore, what would a normalized CapEx number be? I'm just trying to compare it to what I was modeling before.

CapEx for the year landed at $7 million, which was largely our annual guidance. So some of it was bleed-through from, I think, being less than guidance in Q3 plus this IBM deployment.

It would have been about 1.5, absent the Singapore buildout, in Q4, if that helps you.

Yes, absolutely. Then, just generically speaking, your model is obviously working from being carrier-neutral, and there's a number of players like Sprint, for example, since they were the main provider in that facility, I would be so bold as to say one of the reasons they needed to exit the business is that they didn't win customers because they were the only network in there. Are you finding that your competition is getting wind of this as an obvious thing they need to do, is provide more carriers in their facilities? Is that changing the competitive dynamics of how you are winning deals, and having to be a little more focused on how you deliver that message?

Well, I think we for some time have seen more traditional providers like carriers who have been willing to accept the second network or maybe even, in some cases, a third network for reliability, or just because the customer had demanded that they had access to others. But what happens there is that second or third carrier are providing local loops and adding costs just to get in that facility to serve them. And the power of having everybody already in the building over a cheap crossconnect, with the flexibility to even change providers and certainly go to a broader list of suppliers, is really what our model is about. So I think we've seen for some time some of the carriers have offered this ability for a second provider, but just replicating our situation at a cost base and flexibility standpoint, is where they just can't match us.

Absolutely. And then, I obviously am a big fan of the Sprint transaction. In terms of others out there, I'd definitely be a fan of seeing you do more. But are you active in those discussions? You don't have to name locations, but you hinted at it -- a little more than a hint. Any more comments of specific markets or any statements of how active that might be?

No more specific statements, Andy.

Excellent. Keep up the great work, guys.

Richard Fetyko, Kaufmann Brothers.

Congrats on the quarter, guys. Just a couple of questions on the Asian business that you have. Could you, first of all, break out what the -- I suppose the EBITDA drag was from the Asian operations on overall EBITDA, and also just give us a little higher overview of what you're seeing out there, in terms of pricing pressures and bringing that portion of your business to EBITDA-neutral position?

Richard, this is Keith Taylor. So on the EBITDA side, if you break it down between the two regions, if you will, roughly 1.8 million of the EBITDA -- excuse me, Asia represented 1.8 million of the EBITDA loss, and therefore U.S. was roughly 3.2 million of EBITDA-positive gain in the quarter.

And then I think, Richard, you had a second question about pricing pressure. Was that specific to Asia, or the business in general?

Mostly in Asia, unless you're seeing something else in the U.S., as well. I'm just wondering how long you estimate it will take you to bring Asia to EBITDA-neutral position -- how soon you see that inflection point?

Specifically in Asia, there is certainly and amount of price pressure that exists over there. Frankly, it's a market-by-market thing. Tokyo is a different place, and Hong Kong is a different place from Singapore, and certainly Sidney. In each market, there are different price points where traditional colo services apply. I would say, though, that in many of our deals, we are still able to gain a premium there, much as we gain a premium here over other service providers. The reasons are a little different in Asia, though, and this gets back to our theme for our customers of just trying to make Asia easy for those multinationals who are going over there. I just came back from a trip a few weeks ago, and what was exciting to me was dinner and customer meetings each night of the first week there, with different U.S. customers that are potential opportunities for us in Asia, and the level of trust in our service levels here, and how we are managing things over there, versus walking into a given market with the incumbent telco, who owns a large percentage of the share and has some pricing power in their market ownership position, is a real differentiator for this multinational opportunity. And so they understand that we need to make money and grow and allow us, I believe, a premium over what are some other opportunities they'll find in a given market that might be real estate-focused and very much about price pressure. So certainly, rack returns there are not the same as the are in the United States, but we have this great advantage of a deep U.S. customer base who recognizes the value of our relationship over there.

Brian Garnick (ph), Corsair Capital.

Hi. Good afternoon. It was a great quarter. Just a couple quick questions on Sprint. Can you tell us how much revenue Sprint will contribute in the guidance for next year, without --?

Not disclosing that at this time, Brian.

Can you tell us, of the thousand racks that are built out, how many are occupied?

I don't want to start doing it for Sprint. We don't offer utilization on a center-by-center basis. It will roll up into the full footprint, obviously, as it all comes into place.

I don't know if I have to wait for the 10-Q for the next one, but I think you've been reporting the acquisition rather than (ph) Asia, in the Q. Is that right? The numbers related to the acquired Asian and partial U.S. businesses?

What we report in the Q now is actually the U.S. operations separated from the Asia/Pacific operations, yes.

So even in the last quarter?

Yes.

Okay. Well, I think you've pretty much answered my questions. Thank you.

(OPERATOR INSTRUCTIONS). Morgan Frank (ph), Manchester.

A couple of quick clarifications, if I could. The 3 to 5 percent operating expense increase you are guiding for, for '04 -- is that from '03 as a whole for the year, or from sort of current levels?

That would be current levels.

Okay. That's what I was hoping you were going to say. And then I'm sort of running through your income statement for next quarter, and trying to figure out where OpEx is going to go in. It looks like maybe you're going to try and pull another $600,000 or so out of your cash operating expenses, to make the sort of EBIT levels you're discussing? Is that about right?

I'm not sure I followed the question.

If you're going to do 33 million in revenue, and then have a 40 percent cash gross, but then take out 3 million in tax accruals and a cash bond payment and still have a million left in EBITDA, that would seem like your cash OpEx would need to drop from the levels it's at this quarter.

It might drop just a little bit, in part because one of the things we saw in Q3 was such a low. We just had incredible cash collections. And as I said, we expect the DSOs to be spread out a little bit. So you might see that coming through the line, as well.

And then, I guess, the last question is regarding Sprint, again. If you guys are at 33 percent capacity utilization overall, what's the big attraction to buying capacity or to increasing capacity? Are you running out within the specific Silicon Valley geography, or what was the -- it seems like capacity is the one thing you've got tons of.

Well, yes. I mean, on a global footprint, clearly. But we've really get two centers now. I'll give you a little bit of data here, in that we have two centers that are over 50 percent utilized now. As we said for some time, Newark is virtually full. Obviously, that justified the buildout in Secaucus. But San Jose would be that second center, and this just continues to be a great market for us. And we want to ensure that we've got plenty of headroom. And when you can take on a center in the form we had, it also gives us time to build against that. But it was clearly addressing a specific market, and not putting any downstream limits on its ability to grow for us.

Great. Good quarter, guys.

Chris Justin (ph), Morgan Stanley.

Hey guys. Great quarter, as has been said. Listen; I think a lot of what I was going to ask has been answered. But do you mind going through -- in the past, you've kind of given an overview of the competitive landscape and some of the sales process. Do you see now that the critical mass, some of the customers bringing in their vendors, their suppliers, things like that -- is that reducing your costs? And if so, kind of the pricing footprint that's out there? Are you seeing less and less? Is it still contracting? That kind of thing.

Well, a couple of things in there. Just generally, in the competitive landscape, the consolidation we saw this year, the announcements by C&W and Sprint, have clearly had an impact on us. And we gained a lot of share shift value out of those announcements, and see ourselves continuing to do that. Specifically, in terms of who, I'd say, we go up against most in an opportunity is AT&T. And certainly, the prevalence of their brand and their sales team is strong, and we will see them in specific opportunities. Of course, we are quickly selling to the neutrality message in our offerings, and also, what's very interesting is that we're even seeing RFPs today that are requesting multiple networks in the RFPs. So I think neutrality as a sector in itself is starting to emerge. And this is holding up very well from a competitive standpoint.

I think your other question was about pricing, and as a result of that neutrality, we are also still experiencing a premium over the other providers that are out there. And the premium is -- I'd just characterize it as a return per rack or a return for our stated capacity, we feel, is greater. 15 to 20 percent may be a fair premium that comes from the access to these networks, versus just a raw solo play with one network applying to it. So I think the value of the model itself, again, is emerging. It's the right one, that customers are trusting. And as a result, just as we win these customers, we are able to justify it more.

Well, I appreciate it. It's obvious in the numbers. And I guess the other thing is -- and leave it to a non-analyst to ask the vague, big-picture questions. But it's not really discussed -- you guys haven't mentioned in a while. Is there any quantifiable number that you could put on this total market that you go after, given you could provide the facilities and space?

Well, all I can do is point to the -- because the Company will have one view, and I -- I think it largely will support this number, but Andy Schroepfer's, research, who's on this call, Tier 1, seems to be the most prolific in doing the research and diligence to come up with industry numbers. And his number was that this market is a $1.5 billion market this year. And so that's out there as a number. Certainly, we're growing in share within that market, and feel good about it as a size of target. It could vary, however, because we have a number of different players in the market that might put some new accounting in that number, and perhaps a conversation with Andy at some time would help clear some of this up for you. But IBM being in our center, are their numbers in this market or not? Certainly, our pipeline reflects that we've got a lot of growth ahead of us. And so I feel good about that number at this time.

Good recommendation. Andy generally takes my call, as you do, so I appreciate it, guys. Good quarter.

This concludes our conference call today. Again, thank you for joining.