Equinix Inc (EQIX) 2005 Q4 法說會逐字稿

Hello, and welcome to the Q4 2005 Conference Call. This call is being recorded for replay purposes. [OPERATOR INSTRUCTIONS.]

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

Good afternoon, and welcome to our Q4 and year ended 2005 results conference call. Before we get started, I would like to remind everyone that some of the statements that we’ll be 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, including our form 10-K filed on March 10th, 2005, and form 10-Q filed on October 26, 2005.

In addition, we’ll provide non-GAAP measures on today’s conference call. We provide a reconciliation of those measures to the most directly comparable GAAP measures, and a list of the reasons why the company uses these measures, in today’s press release and on the Equinix Investor Relations page at www.equinix.com.

With us today are Peter Van Camp, Equinix’s Chief Executive Office, Keith Taylor, Equinix’s Chief Financial Officer, and Margie Backaus, Equinix’s Chief Business Officer.

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

Thanks, Jason.

Welcome, everybody, as we put the final remarks on 2005 and talk about another exciting year of expected growth for ’06. the final quarter of ’05 produced 61.8 million in revenue, which rounds out to a year of 35% growth at 221.1 million. EBITDA in the fourth quarter of 21.8 million contributed strongly to a final result of 70.1 million, which roughly doubles our 2004 number.

Our cash gross margins in the quarter were a strong 59%. To be fair, some seasonality in our costs contributed to this result, such as utility spend and even reduced FICA costs, but I still think it’s an excellent statement about the operating leverage of the model. Our adjusted free cash flow for the quarter, which excludes the purchase and sale of real estate, was 6.4 million, bringing the total year to 35.2 million in ahead of plan. We closed 68 new customers in the quarter, bringing our total customers at the end of ’05 to 1,138.

Also in the quarter, as expected, we completed the transactions outlined for the purchase and financing of our DC area data center campus, the sale and leaseback of our L.A. center, and the planned exit of our San Jose grounds lease. With these transaction completed, ’05 was clearly a big year for us in terms of our expansion efforts to support our future growth. Of course, these included the acquisition of 3 new centers in the L.A., Chicago and Silicon Valley markets, all of which are expected to open in the first half of 2006, and will increase our cabinet capacity by 19%.

We also saw great success in the 2 new properties that opened at the end of ’04 and in March of ’05. Both of these saw solid bookings and installations, which now have been both generating positive cash flow. This all happened in less than half the time we had anticipated.

In other ’05 developments important to our future growth, we hired and integrated approximately 70 people into the company as we grew the team to support our long-term goal of 500 million in revenue. Also in support of this, we undertook significant initiatives in scaling our system infrastructure and business processes. We completed an extensive measurement of our customer satisfaction around reliability, service offerings and future growth plans, which resulted in a 90% endorsement score of our current customers who would recommend Equinix to their peers.

New services released included the upgrade of our exchange platform to support 10-gig capacity for our larger peering customers. Demand for this has been strong and, with 2 additional rollouts in the first quarter, we expect this service to continue to fuel exchange traffic growth in 2006. We also announced a partnership with NeuStar to jointly develop and to bring to market a VoIP peering service through our exchange platform, while in Asia we’ve launched a new mobile roving interconnection hub in our Singapore IBX. This actually complements our overall progress in Asia, where in ’05 the business hit an inflection point, seen by the steady sequential quarterly increase in its contribution to our EBITDA performance.

I’m also pleased to announce that we’ve appointed Samuel Lee, formerly our country manager in Hong Kong, to his new role of vice president and head of our Asia-Pacific region. Samuel has worked in the region for over 10 years and with Equinix since the end of 2002 when we entered Asia. He has shown himself to be a strong and trusted leader by our Asia-Pacific team.

Finally, in the quarter we squeezed in a pretty significant equity offering for our large shareholder, Singapore Technologies Telemedia, where we placed a total of 10.2 million shares through a common and mandatory exchangeable offering. Of course, this has brought a new level of liquidity to the stock, as well as removing any questions or overhang regarding STT’s position in the company. It’s also created an attractive entry point for a number of new institutional shareholders to establish a position in Equinix. It’s good to have all the new shareholders on the call today.

So, let me stop there and turn it over to Keith, then I’ll come back to talk about more specific plans for 2006 and our ’06 guidance.

Thanks, Peter. Good afternoon. After another strong quarter of financial performance, I’m pleased to provide you with our fourth quarter and fiscal year 2005 financial results.

So, let me first start with the revenues. Our Q4 revenues came in at 61.8 million, a 6% increase over the previous quarter and a 37% increase over the same quarter last year. This result was ahead of our expectations, and helped us finish the year at 221.1 million in revenues, a 35% increase over 2004. Our US revenues represented 87% of the total revenues.

The core of our business, our recurring revenues, were 58.4 million, a sequential 8% increase over the previous quarter, and a 37 percent increase over the same quarter last year. For the year, our recurring revenues were 208 million, a 35% increase. Non-recurring revenues for the quarter were 3.4 million derived primarily from installation fees and professional services. For the year, non-recurring revenues were 13.1 million. With regard to churn, in Q4 and in turn for the year, both our cabinet and MRR churn continued to remain below our targeted level of 3% per quarter.

For Q4, MRR churn was 2.7% for the quarter, while our cabinet churn came in at 1.8% for the quarter. For the year, our cabinet and MRR churn was approximately 5.5 and 8.2% respectively, less than the churn experienced in 2004. As we’ve outlined in the past, churn typically arises from 2 sources, customers with financial difficulty or business difficulties, and proactive churn. Proactive churn typically occurs when we raise customer prices to align to current market rates, or to reflect additional resources the customer consumes, such as power.

Moving into 2006, we’re continuing our efforts to optimize returns, given the current market prices and the demand placed on the IBX’s power and cooling design. We anticipate this will result in higher churn levels than experienced in 2005, yet still within our stated guidance of 3% churn per quarter. Recurring monthly revenues associated with these churn cabinets is expected to be replaced with higher priced MRR cabinets, consistent with a scarcity of supply in a high-quality co-location market.

Moving on to gross profit and cash gross margins, the company recognized a gross profit of 20.1 million for the quarter, or 32%, compared to 17.1 million the previous quarter, and 10.3 million for the same quarter last year. Our cash gross margins increased to 59%, up from 57% last quarter, and 53% same quarter last year. This result included a $1 million benefit realized from the San Jose ground lease restructuring charge, but offset by approximately 1.5 million in partial quarter costs related to our expansion efforts in L.A., Chicago and the Silicon Valley IBX expansions.

In addition, we experienced other seasonal benefits, including the anticipated quarter-over-quarter reduction in utility rates over the higher rated Q3 quarter, plus the expected Q4 reduction in our FICA cost. For the year, our cash gross profit was 126 million, or 57%. Looking forward into 2006, we do expect to incur expansion costs in the cost of revenue line of roughly 5.7 million.

I’d like to now detail the 3 real estate transactions that we completed in Q4 in order to help you better understand how these events affect the financial results going forward. First, we purchased our D.C. area campus for 53.8 million, including the purchase cost of approximately $800,000. We financed the transaction with a 20-year, $60 million mortgage at 8%. This mortgage, unlike an operating lease, will have the effect of increasing our interest expense by 4.8 million in 2006, yet will benefit EBITDA by approximately $3 million. Our 2006 cash flow will decrease by approximately $3 million as a result of this transaction.

Second, we completed the lease financing of our Los Angeles expansion center, resulting in a cash gain of 3.4 million, which is being deferred for book purposes. As part of the transaction, we entered into a 20-year lease with total payments over the term of the lease approximating $77 million. Under the terms of this lease arrangement, we recorded a lease financing obligation of 38.1 million and will amortize this obligation over the next 20 years, and will benefit our EBITDA by approximately $1 million in 2006.

We also finalized the accounting treatment of our recently acquired Silicon Valley expansion center, which will be fully accounted for as an operating lease. As a result, our lease cost will increase our cost of revenues in 2006 by approximately 2.5 million over what we originally anticipated. Given this change in the final accounting around our L.A. expansion IBX, our cost of revenues will increase by approximately 1.6 million over our original guidance.

Lastly, we finalized the agreement to amend our Silicon Valley ground lease whereby we recorded a restructuring charge of $33.8 million in Q4. From a P&L standpoint during the full-year term of this agreement, our EBITDA will improve by approximately 6 million a year, yet our cash flow will decrease by $4 million a year.

On to utilization rate. At the end of Q4, our operational cabinet capacity was approximately 26,200, excluding about 4,900 cabinets we expect to add in 2006. At the end of Q4, approximately 14,100 cabinets, or about 54% of our total cabinets, were billing, an increase from the 13,700 cabinets billing over the previous quarter. Breaking out utilization on a weighted average basis, meaning calculating the utilization rate, taking into account when in the quarter a cabinet started billing, our utilization rate was just under 54%, or about 14,000 cabinets billing on average in the quarter. While we consider the cabinets from the expansion IBXs, our utilization rate decreases to 45%.

Looking at revenue per cabinet on a weighted average basis, our average monthly recurring revenue per cabinet increased to $1,390, up 2.1% over Q3 result of $1,361. This is the result of our optimization strategy, customers ordering additional services and, of course, price increases for both our new and existing customers.

Looking into Q1, we expect our MRR per cabinet to remain flat to slightly up. Our SG&A expenses for the quarter, including stock-based compensation expense of 2 million, were 17.3 million. SG&A expenses for the year, including 8.3 million of stock-based compensation, was 65.7 million. In Q4, our cash SG&A expenses was 14.8 million, a slight decline over the previous quarter and a 26% increase over the same quarter last year. This included the addition of approximately 40 SG&A employees during the year in support of our announced $500 million revenue goal. Our total cash SG&A, excluding stock-based compensation expense, was 55.4 million, or 25% of our revenues, down from approximately 28% in 2004.

Moving on to EBITDA. Our EBITDA for the quarter was 21.8 million, slightly higher than our expectation, and a 22% increase sequentially. This represented a flow-through rate on incremental revenues of 106%. This result is a great example of how our quarterly flow-through rates can vary due to seasonal variances in our costs, such as utilities, and the FICA, as Peter mentioned. As we saw in 2005, these costs recur in Q1, which will have the effect of lowering our incremental EBITDA margins from the higher Q4 result, as you will see in our guidance. This variability is a key reason why we track our internal performance and projections on an annual basis. For the year, our EBITDA was 70.1 million, which represented a flow-through rate of 60% on incremental revenues, even when including the significant investments in our capacity expansion and team.

Before moving on, let me just touch briefly on income taxes. Similar to 2005, we do not anticipate paying any significant income taxes for 2006, and these taxes will primarily relate to alternative minimum tax.

Our net loss for the quarter on a total company basis was 32.6 million, or $1.25 per share. Excluding stock-based compensation expense and the restructuring charge, we would have had net income of 3.2 million. Our net loss for the year was 42.6 million, or $1.78 per share. Again, excluding stock-based compensation expense and the restructuring charge, our net loss for the year would have been $528,000.

On a cash basis, our cash net income for the quarter improved to 20.5 million, a 19% increase over the prior quarter. For the year, this result was $66 million, a 116% increase over the previous year. Going forward, cash net income is a metric that we intend to eliminate as EBITDA appropriately reflects the operating performance of the company. Now, as many of you know, beginning in the first quarter of 2006 we will commence expensing stock equity awards provided to our employees in addition to the performance-based restricted stock issued to the executive team. Although our expenses highly depend on stock price on the date of grant, we estimate our 2006 stock-based compensation expense will approximate $35 million. Apart from the stock price, there are 2 other key assumptions that affect this expense that you should be aware of, (1) the expected term of the equity award, and then (2) the expected volatility rate of our stock price. As we finalize the accounting for this expense, we will update you on the next earnings call with more specifics.

Also, one comment on our equity issuance philosophy. Clearly, it’s our objective to hire and retain key employees at competitive compensation levels, including equity compensation programs. However, our compensation plans have been shifting over time to be more cash-based, as it’s our intention to reduce the level of dilution attributed to our equity compensation plans. The net dilution rate has been steadily declining. 2005 was approximately 6.5%. 2006 is budgeted to be just under 5% and, looking forward into 2007, we expect the rate to be approximately 3.5%.

Turning to our balance sheet, at the end of the year our cash, cash equivalents and investments balance was $188.9 million. This includes a $30 million draw-down under the Silicon Valley line of credit, which has now been repaid. CapEx for the quarter was 22.9 million. Our expansion CapEx for our new centers was 14.9 million, while our ongoing CapEx was 8 million. For the year, our total CapEx was 45.4 million, and was comprised of 25.7 million of expansion CapEx and 19.7 million of ongoing CapEx. Included in our ongoing CapEx investments for 2005 was roughly 2.8 million for the initial development of our back office systems, 2.8 million to deliver additional power equipment and supporting improvements in one of our existing IBXs, and 2.2 million related to the upgrade of our exchange platform.

Next, moving to our statement of cash flows. Our net cash generated from operating activities was 18.6 million. This represents a quarter-over-quarter increase of 3.1 million, primarily attributed to an improvement in our operating results and a reduction in our working capital. For the year, cash provided by operating activities was $67.6 million. Net cash used in our investing activities for the quarter was 65.9 million, primarily related to the 53.8 million purchase of our DC campus and the 22.9 million invested in capital expenditures.

These outflows were offset in part by greater than anticipated increase in our accrued property, plant and equipment balance at 10.6 million. For the year, cash flow used in investing activities was $120.9 million. As a result, our free cash flow was negative 47.4 million in the quarter. Excluding the impact of the real estate transactions, our adjusted free cash flow was 6.4 million in the quarter. **For the year, our adjusted free cash flow was 35.2 million, which was higher than planned primarily due to the slower than anticipated pay-down of our capital expenditure accrued balance. This benefit may affect our 2006 free cash flow by up to $5 million.

Our net cash generated from financing activities in the quarter was 127.9 million, primarily related to the $60 million mortgage for the DC campus, the 38.1 million Los Angeles lease financing, and the $30 million draw-down to the SBB credit line. For the year, cash generated from financing activities was 134.6 million.

Finally, with respect to our equity capital structure, with the completion of the STT share offering in November, we had approximately 27.4 million shares of common stock outstanding. This number excludes 2.2 million shares related to our convertible subordinated debentures and about 4.5 million shares related to employee stock plans and other warrants, the majority of which vest over the next 2 to 4 years.

Let me turn the call back to Peter.

Thanks, Keith.

Now, let's shift to 2006. As you saw in a separate press release to today's earnings, we are announcing the build-out of a new IBX within our DC area campus. As we've noted in the financing for the acquisition of this campus, we had an obligation to invest 40 million by the end of 2007. Based on our success in our most recent expansion there and the continued strong demand in that market, we are ready to proceed with this build-out. This is the first Greenfield build we will have done in some time and, to be clear, this is a build-out of what's currently an empty shell versus a retrofit, or the equinization of an existing data center. As many of you know, this expansion space was a very attractive aspect of the purchase of the DC area campus. Its proximity to one of the richest IP network interconnection points on the Internet today, as well as the opportunity to leverage the team in place in the ongoing management of this center, created some clear synergies in our decision to do this build.

As we've noted on prior calls, we've been taking a hard look at the returns of a Greenfield build. Based on continued demand, pricing strength and the value of bringing the right power and cooling design to meet customers' future requirements, Greenfields can present exciting financial returns. Of course, our ability to tether these new centers to our original IBXs is an important differentiation in any center we'll bring online.

So, just to review the specifics of what we see in this center, let me take you through the highlights. As we've now completed the initial design work and have awarded the bid for the center’s construction, we see an ability to deliver a cabinet level of 1,500 to 1,700 sellable cabinets at a cost of 50 to 55 million in CapEx. As the center reaches capacity, we model on a timeline of approximately 3 years for this to happen. We expect that it will generate a range of 35 to 38 million in revenues and deliver cash gross margins greater than 65%. This would produce a 10-year IRR of 40%.

What's exciting about a new build such as this is the ability to put an increased level of power and cooling on a per-cabinet basis in the design. In fact, our plan is to more than double the capacity of our original IBX builds. With this, there are some important points you should consider as you think about the revenue and returns in a build like this. To be clear, the above-mentioned range of cabinets is a sellable cabinet number. Of course, historically, we've guided everyone to utilization levels of 75 to 80% of our available cabinets. The sellable cabinet number reflects the ability to deliver power and cooling across all cabinets in this center. In turn, the revenue, or MRR, per cabinet in this IBX will be higher as we deliver a higher power and cooling design to each utilized cabinet.

I should also note, with an anticipated build schedule of approximately 12 months, which has not yet begun, we don't expect to incur the full expansion CapEx associated with this build in '06. For modeling purposes, we would expect approximately 40 million to be funded this year. We also expect to fund this build from our balance sheet.

Also, as you may have seen in the same release, we are making plans for a future Greenfield build in Chicago. As we've noted before, we view Chicago as a strategically important market and, based on our demand study, we recognize the much larger opportunity than our current expansion would support. As an update, you may have seen we announced significant growth in FX in our financial vertical. This will represent a great deal of our plan growth in our soon to open expansion center.

As we've also announced, we've identified a potential site for further expansion in Chicago. There’s significant work to be done in the diligence for this property before a final decision to expand for this is made but, pending this, the building would cost 9.75 million, representing approximately 228,000 square feet, which presents some interesting options for its development. Clearly, with a building of this size, we will take a phased approach in a traditional IBX build. Yet, other options could include partnering with a real estate concern or even a customer to support both the deployment of an IBX and a large enterprise center to share in the capital build and enhance our returns from this property. Again, we're early in our planning around this expansion, so we'll update you as things firm in this site.

Of course, the build in northern Virginia, or what we potentially might do in Chicago, present 12 to 18 month lead times in delivery of this capacity. So, we are excited about the future returns from the centers, but these are really about our '07 growth.

So now, I'd like to take our focus back on '06. We continued to see a very strong pipeline across our business. This includes Asia, as they've come off their strongest bookings quarter ever. Also, a number of our opportunities are slated for our new expansion IBXs. Again, these come online in the first half of the year. I should note, this will have some impact in how quickly we realize revenues from this pipeline, as a number of key opportunities are timed for the new center openings. As we've talked about in the past, we typically see book-to-bill timelines running 45 to 60 days. In the case of new centers, obviously book-to-bill timelines will be specifically tied to the opening of these new IBXs. Our team is executing well and on plan in the equinization of these 3 centers, while it’s great to see the customer demand that’s lining up.

So now, let’s look at our guidance for 2006 in the first quarter. As we outlined, we expect another year of solid growth as we look at the key trends in our market of expanding broadband access, increasing digital media-rich applications, and the enterprise, leveraging the reach of IT and the Internet to deliver important information that drives their businesses. We believe these trends provide great support for our expected growth. Of course, all of these trends continue to drive additional bookings from our installed customer base, which has always been an added point of visibility in our forward view of the business.

So, in guidance, as we’ve outlined, we anticipate revenue for 2006 to be between 275 and $285 million. We expect cash gross margins to be 57% to 59%. I should note, as Keith mentioned, final accounting treatment of the leases for 2 of our expansions has created approximately 1.6 million in additional net cash costs in the gross margin line. However, our current guidance absorbs this. Cash SG&A will be in the range of 60 to 64 million, or 22% of revenue at the mid-point, down from 25% in 2005. This includes continued investment in the team and the supporting engine for our future growth opportunities.

EBITDA will remain in the range of 95 to 105 million for the year. Ongoing CapEx for the year remains at 20 million. Expansion CapEx for our 3 new IBXs will be 35 to 40 million, while we’re now adding an additional 40 million in anticipated 2006 expenditures for the build-out in our DC area campus. This brings total CapEx for 2006 to be in the range of 95 to $100 million. This does not include any additional expansions that may occur.

Our adjusted free cash flow will be break-even to a negative 5 million for 2006. This new guidance includes 2 elements, the 40 million in additional CapEx for our DC area build, as well as allowing some flexibility for timing considerations related to working capital.

Moving on to Q1, revenues are expected to be in the range of 63.5 and 64.5 million. In the build-up to our biannual guidance, we’ve accounted for the fact that we’re bringing 3 new centers online within the next quarter, which has delayed some revenue in our pipeline. Additionally, because of the high utilization rates now in L.A. and Chicago, we are carefully managing our remaining capacity in favor of our installed base of customers there while we’re also focusing on optimizing returns as below-market contracts come due. These efforts are really what we’re talking about when we discuss proactive churn. Some of this was in the Q4 churn number, and we’ve recognized the full quarter’s impact of this with our guidance.

In cash gross margins for the quarter, these are expected to range between 57 and 59%. Cash SG&A is expected to be approximately $15 million. EBITDA is expected to be in the range of 21.5 and 22.5 million, a similar result to Q4, but again reflecting investments in our cost base and seasonal fluctuations that we’ve noticed. Total CapEx is expected to be between 30 and 35 million, which includes 24 to 27 million of expansion CapEx in support of these new centers.

So, in completing guidance, like to thank everyone for joining the call. As I think you can all see, fourth quarter capped off a great 2005. And now, as we launch into ’06, clearly we’re planning another big year of expansion as we pursue a significant opportunity ahead and solidify a market-leading position. As always, with the opportunity out there, our success will be all about execution. Of course, this team has proven its ability to step up to a challenge, and we’re certainly excited about this one.

Operator, let’s open it up for questions.

Thank you, sir. [OPERATOR INSTRUCTIONS.]

Tom Watts with Cowan.

Hello, Tom.

Hi, Peter. Congratulations on another great quarter. So, for the Washington, DC expansion, we’re looking at about $30,000 per cabinet. Is that fair?

Yes, I think the math cancels out that way.

And for the expansion in Chicago, is there going to be similar prices, and do you have a sense of what the cabinet capacity might be there?

Well, as I said, the size of that building does suggest a phase approach if we were to move forward with that building. But, it’s early to give you some specific CapEx guidance around it. Again, we’re looking at some potential interesting opportunities in the way we would develop and bring that to market, and I think the opportunity for a shared build or support from a real estate concern could present a difference in the way we would deliver those cabinets and the returns available from them.

Okay. And then [you have] other markets besides DC and Chicago. Do you expect-- are you looking at a further expansion in any of those during 2006? Does current utilization and growth justify-- justified in other areas?

Well, those are the key expansions for now. We’re not looking at anything specific that would be announced or starting to develop in ’06.

Okay. And then finally, in the past, you’ve said that you would consider additional markets, such as Atlanta, if you had advanced customer commitments on those. Are you continuing to have discussions in that area, and are you seeing any further developments that could suggest ...

You know, I’ve talked in the past about some pretty significant demand analysis we did over the course of last year, and it’s still leading us to believe strongly in the core markets we’re in, as well as obviously the value of the network interconnection there and the ability to link centers to that is an important part of the model.

Okay. Then, just a final question. The 2.1% increase in MRR per cabinet quarter to quarter, how much of that is price increase, and are you seeing those price increases across the market? Are your competitors enjoying those, as well? And is a lot of this growth industry-wide, or how much did you think is specific to you?

Well, I certainly think the demand is industry-wide, and I-- you know, I think there is a scarcity that exists out there. There are still some regional players. When we go in and look at optimization in a center, now that we’ve got some highly utilized centers and we’re looking to put them to a market price point, a given contract that may be coming up, if in fact we’ve lost one of those customers because they’ve moved on, typically been a regional guy that supports that. but, I think the reality is that it’s the current capacity out there just continues to fill up, and so that will be demand for our competitors, as well as ourselves, as this continues. So, I don’t have specific view on competitors’ pricing other than what they’ve been offering on their earnings calls and what we’ve seen in the market from time to time. But, I think pricing just continues to rise.

[Inaudible] now being able to get a 40% return on capital for Greenfield builds, are you aware of other people in the industry starting Greenfield builds, as well, and do you think those sort of returns are available for them?

No, I’m not aware of specific ones. I think we did hear of expansion coming from-- well, one I heard about, AT&T, don’t know whether that was just a prior data center that they would reconfigure for market or how that would go about. I haven’t heard any new news on this. It’s been a while since I’ve seen that. But, on the other hand, I’d also say that we see a great opportunity for our model in this market and, clearly, our growth rate compared to what the market growth rate has been has been very strong, and thus supporting our ability to fill these centers up and drive the returns I’m talking about. So, I don’t know if we’re unique in being able to produce this kind of return or not but, clearly, I’m feeling good about it just based on the confidence in our own demand and the performance of our business.

Jonathan Schildkraut with Jeffries.

Hi, Jonathan.

Hey, another great quarter. I have a handful of questions. I want to start with cabinet adds. Only 400 cabinets added this quarter. I’m wondering if that is a reflection of the very strong 1,300 in Q3, or if it’s a reflection of kind of that slowdown and pairing of customers that you were discussing as you manage the remaining capacity in some of your higher demand markets?

Yes. The net cabinet adds were approximately 400 for the quarter, and so I think it does reflect some of the churn that we highlighted from fourth quarter, but some of it is just install timing, in the way pipeline just blends in and gets in place. So, I don’t know. I’m looking over to Keith for any more color that might be meaningful there.

Hey, Jonathan. From our perspective we do look at the net, so the net cabinet movement period over period or quarter over quarter, and that, as Peter said, it really does reflect some churn in the quarter that we’ve talked about on the scripts, but also sort of where we are in some of our capacities or utilization capacity levels and utilization levels in some of the IBXs. But, this is something that we were sort of anticipating.

So, do you think that next quarter could look very similar to this quarter in terms of new cabinet adds?

I think realistically, I think it’s going to trend between what we have here and sort of more of the traditional quarter, which is the sort of 700. It will be in that range somewhere. But again, Q3 was somewhat of anomaly because we had some very large installations take place and, as you saw, sort of a bit of pent-up demand. That’s not something that we expect to recur on a regular basis. Obviously, we’re looking for some larger customers, but it’s atypical of a quarter to see that amount of growth.

Okay. A couple questions on the expansion. In terms of the DC facility, you know, you talk about 1,500 to 1,700 cabinets added to capacity. Let’s use 1,600 as our average. In the past when you’ve talked about-- and I’m just trying to drive to an apples-to-apples comparison here-- in the past when you’ve talked about capacity, you’ve talked about it and said, oh, you know, at an 80% utilization level. So, if we were to say the capacity here was 1,600 and you’re saying that you could use all of it, is it then on an apples-to-apples basis somewhere like the equivalent of what you previously would have said was 2,000 cabinets?

Well, yes. I think if you’re looking at a sheer cabinet number, the answer is absolutely yes, Jonathan. The only thing to be cautious about, though, is the price point for these cabinets will be different, too. And so, obviously with bringing more power resource to it, we’ll have a higher MRR per cabinet. So, this capacity won’t be coming online until next year but, clearly, we’re going to need some thought in how-- need to put some thought into how we guide you towards MRR per cabinet and how you look at a cabinet of utilization, because these cabinets are, frankly, worth more than what has historically been an IBX cabinet.

Yes, the concept of a cabinet is rapidly changing.

Yes, so-- well, actually as we work on it, you and obviously the other analysts on the call may have some good insight to how best you want to be able to track cabinet utilization going forward. But, as we bring on these higher valued cabinets, we’ll need to create a-- just a good metric or model for you to follow.

In terms of the Chicago site, I mean it’s an enormous site. I think it might be amongst the biggest that you’re talking about, or relative to your current IBXs. If you were to facilitize that site, and including the DC site, my estimates have you somewhere at around a revenue capacity of somewhere around 460, $465 million. That would imply only 1 to 2 additional IBX expansions. [Inaudible], what are you thinking about?

Well, first of all, the demand that we see in Chicago is great, and it certainly warrants going beyond the expansion that’s coming online here in the first half. But, on the other hand, it doesn’t at this point and time suggest that we’re going to every-- or that we go to Chicago and build this center out entirely as an IBX. And so, it’s an interesting building, and certainly available and on the market, and we took the opportunity to secure it, pending our diligence, and even pending our planning around it in terms of how we’re going to develop and bring it to market. So, that’s in place, and certainly that’s an announcable event for us, so we’ve provided that to you. But, look at this as a phased build is the only thing we’re really speaking to at this point in time in Chicago. And so, that would then give us another IBX in the more traditional size. And then-- so, obviously there is room for other builds as we look towards a 500 million in the way we would shape that long-term revenue goal.

Okay, great. Just 2 more questions, sorry. You talked about the launch of the 10-gigabit Ethernet exchange, launching into a couple of new markets. Is that-- does one connection on the 10-gigabit Ethernet exchange, theoretically can that replace multiple ports on the traditional gigabit Ethernet exchange? And as a result, could that be a net negative impact on a revenue per customer basis?

Yes. I mean, you could start doing math that way, Jonathan, but that’s not the impact we’re seeing. Just an interesting stat since we launched 10-gig, our utilization or capacity flowing across our exchange platform now is up 50%. So, it’s just the classic Moore’s Law of just more demand and more capacity being out there now hitting the switch because we’ve enabled 10-gig. And, of course, all the applications we’re all hearing about of music, video and so forth being introduced to the Internet or driving the utilization and really required us to have this 10-gig support. So, I see it as just continuing to support the health of our exchange business and it continuing to grow alongside the rest of our business.

All right, great. Last question is on Equinix Direct. There’s been a lot of news over bandwidth prioritization and peering arrangements and the possibility of paid peering being a legitimate product sometime in 2006. Equinix Direct has been around for a while. Wondering if there has been any increase [inaudible] in the product, customer demand, etc.

Well, there’s certainly been a lot of increased conversations about the Equinix Direct product, all driven by what we’ve talked about here. I would say the politics of peering, to use that term, have been going on for quite some time, but the reality is is the bandwidth pricing continued to erode. I know a lot of the carriers and access networks are clearly looking at this and saying, “How can I get more revenue for this physical plant that I have out there,” and so that’s certainly stimulated a lot more conversation about billing and settlement actually taking place for [IT] flowing from that, or for [IT] networks.

Now, the reality is that this will take some time to solve. Clearly, if a network was to shut off access to Google, Yahoo or anyone out there, that would impact the value of the access they’re selling to their customers. So, not an easy answer here, and I think even regulatory involvement may be needed to managing billing and settlement. All that said, we’ve already had some paid peering. Actually, AOL offers it today for access to their [eyeballs], and we see other of our customers taking a serious look at how that might work in their business. So, I see it as a potential opportunity with-- for it but, as you talk about it and even as I get excited about the thought of this happening, I do want to just couch it within the potential regulatory and just-- again, call it politics of really where peering may go in billing and settlement. So, just to manage expectations around that.

Michael Rollins with Citigroup.

Hi, how are you? Just a couple questions. I guess the first question I had is I was going through some of your guidance items around free cash flow, and I was just thinking conceptually the mid-point of your EBITDA guidance is around 100, the mid-point of your CapEx value is around 100. I guess you’re going to have some cash interest expense. [Could you] describe that, and also the cost of the discontinued leases? So, when you take all these things in your account, and maybe there’s other items that you may add onto the list that would help or hurt cash flow for ’06, can you get us to the number that’s sort of this working capital and other adjustment? And is that something that simply reverses out in ’07, or does it perpetuate for a little bit more time?

The second question that I had, this is a quick one, is just a breakdown of cabinets added, international versus domestic.

Okay. So, Michael, this is Keith here. So, let me take the first question. Clearly, free cash flow, the way that we have always managed our announcement [inaudible] on free cash flow is really our GAAP cash flow. It takes into consideration all that you’ve said, so we look at our operating cash flow less our investing cash flow absent a real estate transaction, so when we purchase a building. And so, the things that we do look at, of course, is our EBITDA. But, EBITDA is a pretty darn good surrogate today for operating cash flow. So, the one thing that we think catches people a little bit, and maybe this is right to your point, is that there is the assumption that, when we talk about CapEx, that CapEx all gets paid in the year. So, a typical person might say, “Your EBITDA is 100 million, your CapEx is 100 million, therefore you have zero free cash flow.” That’s certainly one way to look at it.

But, the way that we also look at it is how much are we hanging up on the balance sheet at any-- at a given point or at a year-end. So, looking forward into 2006, even though we anticipate spending-- or recording $100 million of CapEx on our balance sheet, we think there’s going to be a portion, roughly $15 million, that will get hung up on the balance sheet and paid in 2007, and we’ll guide to that. But, that is no different than what we experienced in ’05 coming into ’06. There’s roughly $12 million of ’05 CapEx that is being paid in ’06. And so, that is something that we’ll continue to guide you to, but bottom line is we do [think] free cash flow under the scenario of zero to negative 5 for the year.

Just to follow up on that real quick, it’s been the incremental change in the working capital benefit. Would [they] only imply that 3 million, if the 12 million benefit into this year goes to 15 million at the end of this year?

Yes.

And then, you’ve got the discontinued [leads], which is-- I think you guided previously at 10 million?

Yes.

And then, cash interest expense. What are the items that would offset that then to get to the GAAP free cash flow guidance?

Well, again, there’s a lot of other things that go on when we talk about EBITDA. There’s a lot of other non-cash activity going through there. A good example is deferred rent. You’re recognizing more rent expense in your P&L than you would-- than you’re paying on a cash basis. So, there’s a number of things that are taking-- also taking into account. And so, when you net all that out, it basically is giving you that neutral cash flow position at the end of the year.

Let me give you one other example, deferred installation. When we charge our customers, again, for every dollar of MRR we book, a good range is somewhere between $1.00 and $1.50 of MRR revenue. MRR contract will go along with that MRR revenue. So, what happens is we install the customer, as an example, and we bill them on their first invoice. We get all that cash up front, but yet you’ve got a deferred installation revenue that’s getting recognized over the relationship of that customer contract or that customer life. And so, there’s a couple of other nuances that are taking place, and again, we’re really not trying to be [inaudible] here. It’s just there’s a number of ins and outs, but you’re sort of dead on. We look at working capital activity. We also consider interest expense, and we also consider what’s coming out of the restructuring line. That is being offset by things like deferred rent. So, deferred rent, deferred installation, and that’s where we are. So, I don’t know if that-- is that clear enough?

That helps. Thank you.

Okay.

And then international versus domestic cabinets?

Yes. So, basically, the number of cabinets billing on an Asia basis is roughly 2,800, and in the US it’s roughly 11,200 cabinets billing at the end of the quarter.

And then if I can just follow up one more question, if you didn’t just take conceptually that the idea of price increases, on a net basis for all the different moving pieces, what would you expect your average rate of inflation or price increase be for 2006 versus 2005?

Well, I think we’ve been just generally guiding to a 3 to 5% increase in our cabinets. And so, certainly inflation is well within that number. But, that really is the kind of ongoing expectation that we’re still trying to steer people towards, Michael.

Erik Zamkoff with Morgan Joseph.

Good afternoon. Congrats on a nice quarter.

Thanks, Erik.

And I’ll only ask a 2-part question rather than a 6 or 7-part question. Very quickly, can you give us your view on-- and I know you briefly hit on this in your comments-- on power and cooling issues, what’s going to be different in the new centers, what are you-- and sort of how this impacts the way you think about your business and your customers think about their networks. And then, a quick update on the marketplace in Europe and what your thoughts are regarding that area of the world.

Okay. Well, first on the power and cooling, what’s been great as we’ve sat down and done a design for [Ashburn] is just the ability to have taken all the knowledge we’ve learned over the past couple years as we’ve watched the power and cooling issue evolve. And so, with that, as well as the demand analysis that we did, we also use that interaction with our customers to really set, from their view, what they believe their power and cooling requirements would be.

So, with that out there, as I said, on a per-cabinet level in the design of the building, we’re really talking about a twofold increase in the power and cooling capability deployed in a center. And so, this is something that all our engineers are obviously excited about, to bring a product of this nature to market here. But, we just feel it’s the appropriate product for where the world sits with Blade Servers and just [dense] chipsets on networking gear. So, this is going to be a very attractive place for customers to come to for sure, and I think that we will be quite different than what the market currently holds.

And what does that mean for-- I apologize for the follow-up. Does that require going back to a degree and retrofitting some of your older centers?

Well, I mean, that’s why we’ve guided historically to obviously the 75 to 80% of utilization of a center, and it doesn’t mean that we’ll do that, but we have done things such as pricing for cabinet equivalents and really managing the current environment such that our customers can accomplish what they need to, but we’re bringing more in power and cooling than the actual physical cabinets deployed in the traditional centers in a number of just power-dense installations.

And then, well, Europe?

Yes, sir.

Yes. Europe, there’s not a real change in our view there. We still see a number of players who are doing a good job at a more or less break-even level of business, but haven’t seen anybody break out from the pack as being a strong player there. I think Europe obviously will see similar trends to the United States and see similar opportunity. But, right now, we see our best opportunity right here in the US, and are continuing to pursue that.

Andy Schroepfer with Tier One Research.

Hey, everybody. Congratulations.

Hey, Andy.

I looked-- I’ve got surprisingly some more questions on the pricing topic.

Okay.

If you do the math on the [inaudible] and the T-- IBX that you’re going to build with superpower capability, use the $35 million revenue number, use the 1500 rack number, and you get a 1954 per month per cabinet. Since that’s meaningfully above, I think most of us on this call can understand the dynamics of how a dense part configuration could justify that. But, maybe talk about what you see as maybe any differences in the average customer size, the level of value-added service purchasing, whether it’s 10 gig ports or what kind of good bookings have you had, verbal commitments from clients for that kind of new price point and facility.

Yes. Well, the reality of it is, Andy, we’ve already got customers paying the type of numbers we’re talking about as you play with the ranges of what that revenue per cabinet might look like in the new center. And again, they’re paying for it because they’re consuming that level of power and cooling from the design of our current IBXs.

So, although they physically might not have the same level of space as the resource we’re charging them for, at the end of the day, returns for what we see as the price points needed for this new center are already happening. So, we feel very good about the ability to get that pricing.

Now, when you think about just the total revenue per customer in the return-- or in what we see going forward, I don’t see it changing a great deal. Because we’re already experiencing it in our centers. It’s just physically what’s going on is different in the way we support customers.

And so, I know as I said earlier, that does mean we’re going to need to put some good thought into how we track our utilization and just what metrics to follow-- I mean to follow should be. For now, I would break it out separately. I would just say there are soluble cabinets there. They’re going to be at a different price point. And as this center fills up, you’re going to see those cabinets returning a certain revenue per cabinet. And then the traditional business, which-- returning that. But, I agree. We need to somehow get this in a homogenized basis to make sense to everybody.

Okay. The one extra part of that that I just wanted to dive into, the customer that you’re expecting to go into that, is that going to be the same type of customer or is it going to be a smaller number of customers taking a larger than average size footprint? Or any differences you’re expecting with that?

No, I think it’s going to be the customers that we already support today. I think the real bridge that needs to be better defined here is what their footprint and consumption of a current IBX looks like compared to the new IBX.

Got you. I’ve got a few others related to that that I can take offline, but some other ones. As you’re talking to some of these big customers with some big needs for expansion, how much headroom do you feel you have from current pricing to what would be otherwise current build-out prices for the customer to consider building out a 50,000 or 100,000 square foot data center?

So, basically, you’re talking about a customer’s own build model going forward?

Right. What kind of price points are the large enterprises or the large dot-coms thinking of when they do the cost benefit analysis of building themselves versus taking a big footprint with you, or experiencing with you? Is there a big gap still where you still have a lot of room to continue to raise prices? That’s the heart of my question.

Well, yes. I do believe there’s room, particularly with our installed base as we bring more relationships to market to have continued price increases. I also will say-- and I’ve said this in conferences, maybe on prior calls as well, that even from a field sales standpoint for new business, we’ve raised the bar in terms of what they can discount to and where the price points are just as we go to market. So, both in a new booking standpoint, as well as contracts rolling over, just to reflect the current market environment and value resource delivered here, that pricing-- those price increases have taken-- are starting to take effect and we will be doing more of that as we grow.

All that said, as you think about what the choices are for a large user, this comes down to a basic outsourcing question, all right? I mean, if you have scale, expertise and capital available to build a data center, and you’re going to build out 6-700 cabinets requiring 2 to 3 megawatts of power, you could potentially do an ROI that says, yes, I would go build. You certainly have to have a lot of faith in your expertise to management it and all those ongoing things. But, at the end of the day, it’s just a classic outsourcing question.

Now, in our business, we already have a number of guys who are managing in their own data center. AOL, for instance, is a large customer of ours, yet most of their server infrastructure is in their own data center. Yet, they’re in many of our IBXs that sit alongside the networks that are there for their peering and content delivery.

So, we’ll always have a mix of what our just large users of co-location, some of which will sit in our centers, but many of which will be in their own builds when they in-source those.

So, I know that’s high level color, but I think that’s the basic make/buy that any customer could do.

As long as you still feel confident there’s headroom, that’s a great answer. So, let’s see, the next one. I know you guys try to keep your customer contracts to a year or less so that you could continue to up-tick them with the power prices and other increases. To what extent does this next build-out phase change that or potentially view a point in the future where you might want to start taking some deals that are longer than a year?

Well, you know, ultimately we still are landing in a lot of 2-year deals, Andy. As much as we certainly go to market and we start at a year, negotiation, because the customer spends obviously a lot of capital and energy to get into a data center, we’re landing in 2-year terms in a lot of the deals. So, pricing continues to show an opportunity for us which will we’ll pursue. But, no planned change in our current business. Always if we put an anchor in 1 of our centers, then that can be a longer term for sure, a big deployment.

And 3 other quicker ones. How many of your IBXs are essentially fully sold right now?

Don’t have an answer for you on that but, clearly in markets such as LA and Chicago as we talked about, we’re really in a pretty full situation where we’re really just favoring the current installed base and not bringing new customers into those centers, just to manage that growth for those customers.

Let’s see, on your concentration, you didn’t disclose whether or not IBM was 10%, so I’m assuming they’re not. But any update you can give, any kind of stat, top 10 represent X or the largest is X, or ...

No, I’m sorry. IBM I think is 10%, right, Keith?

Yes. They dropped from 11 to 10%, Andy.

But beyond that, we have nothing above 5.

Okay. And IBM, 10% for the quarter and year?

10% for the quarter and-- I’m just checking for you. And for the year it’s about 11%.

Okay, great. And then the last one for now, since you’re opening some other IBXs that you rightly disclosed are going to have some installations, any dependence on a certain key date to open in first quarter for the first quarter guidance?

No. I mean, obviously we’ve-- well, none are opening in first quarter, just to be clear, all right? But centers will begin opening in second quarter. We’re certainly targeting by the end of the first half to have our centers online. So, we’re feeling good about the timeline for those and executing well against them. But, these are construction projects so we always put that caveat out there. But, all this has been contemplated in our guidance. We feel good about the guidance that’s out there.

Great. Well, best of luck to Samuel. He’s got big shoes to fill from Phil.

That’s right. Well, I’ll pass that along to him.

Our last question comes from Greg Mesniaeff with Needham & Co.

Two quick questions and I’ll keep it brief. Could you give us a little bit of color on your VoIP peering service that you’re launching and the partnership with NeuStar?

Sure, yes. We’ve actually now brought the trial up. So, we’re installing our first customers in our Ashburn, Virginia-- this is just 1 site, but the DC area data center campus has a very strong exchange platform there with many networks in it. And so, we’ve already identified the initial customers for it. They’re coming in in a beta and we’ll begin to see that product working for us here very shortly.

With all that said, I would say it’s going to be an evolution over the course of the year. And we’re not giving guidance for significant growth in this area at all. And so, as VoIP peering matures and we just have more success in beginning to market to these customer and so forth, we’ll update you.

I’m assuming there’s not a whole lot of visibility on this-- on the business right now as far as customers.

Well, certainly NeuStar and [AquaNet] share a common set of customers that are targets for this. So, there’s an identified thing that we’ll begin talking to immediately as the trial proves itself. Again, the initial beta are begin installed now. So, we’ll have a good target set to pursue; but again, not giving any hard information about uptake and so forth until we pass the first stage here.

Got you. And then also, another question is, as you continue to focus more on Greenfield opportunities rather than older centers, what should we be looking for in terms of modeling G&A? I mean, are there variables we should be particularly paying attention to, like maybe different property tax assessments or something of that nature?

When you think about the Greenfield, most of that property tax and the like is included in what we call our cost of revenue line. And again, the metric that we try and guide the Street to is that, on incremental revenue, 65% to 70% of the revenue should drop to the cash growth profit line. And that would include things like rent, employees, property taxes, security staff and the like.

So, based on your guidance, I don’t really see any major differences here as far as methodology or...

No, not really. I mean, in the end, as Peter alluded to, we made a large investment in 2005 and we’re continuing to make that investment in 2006. And that’s in the SG&A line and that’s to drive toward that $500 million business plan. So, what you see is basically what we’re anticipating. It’ll be a roughly 6-- I’m sorry, pardon me-- a roughly $15 million [cash] SG&A for Q1 and 60 to 64 for the year.

This concludes our conference call today. Thank you for joining us.