Cogent Communications Holdings Inc (CCOI) 2009 Q4 法說會逐字稿

Good day and welcome to Cogent Communications Group Inc.

fourth quarter and full year 2009 earnings conference call and webcast.

Today's conference is being recorded and will be available for replay on the company's website at www.cogentco.com.

At this time I would like to turn the conference over to Dave Schaeffer, Chairman and Chief Executive Officer.

Please go ahead, sir.

Thank you very much.

And thank you all very much for joining today and good morning.

Welcome to our fourth quarter 2009 earnings conference call.

I'm Dave Schaeffer, Cogent's Chief Executive Officer.

With me on this morning's call is Tad Weed, our Chief Financial Officer.

We are pleased with our results for the quarter and for full year 2009 in an improving but still challenging general economic environment.

During the quarter, traffic grew on our network by 23% from the third quarter of 2009, and by 86% from fourth quarter 2008.

For full year 2009, traffic grew by 75% over full year 2008.

We continue to be encouraged by increases in traffic on our network and growth in our revenue.

We are optimistic about our outlook for 2010.

We are encouraged by the level of order backlogs that we have, orders signed but not yet installed on our network.

Since the end of the third quarter we significantly expanded our footprint by adding another 30 buildings to our network and adding over 5,000 route miles of intercity fiber to our network.

We also acquired, in the quarter, two vacant data centers.

Throughout this discussion, we will highlight several operational statistics we believe demonstrates our increasing market share, expanding scale and size of our network, and the leverage of our business model.

We have assembled a unique set of assets and infrastructure and operations that efficiently generate cash while selling a product whose demand is rapidly growing.

We believe there is very significant barriers to entries to replicating what we have assembled here at Cogent and we are, in fact, the lowest cost operator and the most efficient operator in our sector.

I will review in greater detail certain operational highlights from our operations and continued expansion plans.

Tad will provide some additional details on our financial performance.

And following our prepared remarks, we will open the floor for question and answers.

Now I would like Tad to please read our Safe Harbor language.

Thank you, Dave and good morning to everyone.

This fourth quarter 2009 and full year 2009 earnings report in this earnings conference call discuss Cogent's business outlook and contain forward-looking statements within the meaning of Section 27A and 21E of the Securities Act.

Forward-looking statements are based on our current intent, belief, and expectations.

These forward-looking statements and all other statements that may be made on this call that are not historical facts are subject to a number of risks and uncertainties, and actual results may defer materially.

Please refer to our SEC filings for more information on the factors that could cause actual results to differ.

You should also be aware that Cogent's expectations do not reflect the potential impact of mergers, acquisitions, other business combinations or financing transactions that may be completed after today.

Cogent undertakes no obligation to release publicly any revisions to any forward-looking statements made today or otherwise update or supplement statements made on this call.

Also during this call, if we use any non-GAAP financial measures, you will find these reconciled to the GAAP measurement in our earnings release and on our website at www.cogentco.com.

Now I'd like to turn the call back over to Dave.

Thanks, Tad.

Now for some highlights of our fourth quarter results.

Hopefully, you've had a chance to review our earnings press release.

As with in previous quarters, our press release includes a number of historical metrics.

These metrics will be added to our website.

Hopefully you will find the consistent presentation of these metrics informative and helpful in understanding our financial results and the trends from our operations.

Our fourth quarter 2009 revenue was $62.5 million, an increase of 3.8% from our third quarter 2009 revenues.

For full year 2009, revenues were $235.8 million, an increase of 9.4% from our full year revenues of 2008.

On a constant currency basis, our 2009 revenues increased by 11.2% from our 2008 revenues.

We evaluate our revenues based on product class, on-net, off-net and non-core, which Tad will cover.

We also evaluate our revenues by customer type.

Our customer types include two major groups -- NetCentric customers and corporate customers.

NetCentric customers buy large amounts of bandwidth from us in carrier-neutral data centers.

Corporate customers buy bandwidth from us in large multi-tenant office buildings.

Our corporate customers represent 58.3% of our total customer connections at the end of the quarter.

Our corporate customers represent 46.5% of our revenues for Q4, 2009 and revenues from our corporate customers grew by 2.4% from the third quarter.

Our NetCentric customers represent 41.7% of our total customer connections at the end of the quarter.

Revenue from our NetCentric customers represents 53.5% of our revenues for Q4 2009, an increase of 5% from Q3 of 2009.

Fluctuations in foreign currencies will materially impact our revenues, particularly our NetCentric customers.

Approximately 30% of our revenues are derived from international operations.

From Q3 2009 to Q4 2009, fluctuations in foreign exchange positively impacted our revenues by $600,000.

However, for full year 2008 to full year 2009, fluctuations in foreign exchange negatively impacted our revenues by $3.9 million.

Now for some overall trends and highlights.

Pricing and provisioning.

Pricing for our most widely sold product which is our corporate 100 megabit connection remains at $900 for that 100 Meg connection or $9 per megabit.

We continue to offer discounts related to contract terms for all of our corporate and NetCentric customers.

We offer volume commitment discounts for our NetCentric customers.

During the quarter, 790 NetCentric customers took advantage of these volume and term discounts and entered into longer term contracts with us.

These discounted NetCentric customers helped contribute to a 5.1% reduction in our average price per megabit of our installed base.

The price per megabit in our installed base declined from $6.33 in Q3 2009 to $6.01 in Q4 2009.

The price per megabit for new contracts signed in the quarter was $4.51.

Now to address ARPU.

Our on-net ARPUs declined, however, the rate of decline in those ARPUs decelerated substantially.

On-net ARPUs declined by only 0.3 of 1% from third quarter, as we generally sold larger customer connections in the fourth quarter.

Our on-net ARPU was $982 in the third quarter of 2009 and $979 for the fourth quarter.

Virtually unchanged.

Our off-net ARPUs continued to increase, and increased by 6.9% from the third quarter as the average size of these off-net connections continues to increase, particularly as ethernet services become more ubiquitously available.

Off-net ARPU increased from $1127 in the third quarter to $1205 in the fourth quarter.

Now to touch on churn.

Our on-net and off-net churn numbers both improved in the quarter.

Our on-net churn rate improved to 2.7% in the fourth quarter as compared to the 2.9% in the third quarter.

As a reminder, we consistently report churn numbers on a gross basis, which is inclusive of moves, adds and changes.

As a result, a customer who remains with us but enters into an amended contract is counted in the churn numbers.

Our off-net churn rate also improved to 2.9% in the fourth quarter of 2009 as compared to 3% in the third quarter of 2009.

Our churn rates can be impacted by our continued stringent monitoring of the credit standards of our customers and the collection status from those customers.

Now to turn to traffic.

Traffic continued to increase on our network, and in fact increased by 23% sequentially quarter over quarter, by 86% when fourth quarter 2009 is compared to fourth quarter 2008, and by 75% on a full year 2009 over full year 2008 basis.

In these cost-conscious times, we believe Cogent's low price and value proposition is a distinct competitive advantage.

We anticipate continued growth in our network traffic, as more and more customers take advantage of our value proposition.

Before Tad provides some additional details on the quarter, I would like to address our results against our longer term revenue targets and expectations.

As a reminder, we have previously announced expected top line revenue growth of between 10% and 20%.

For the fourth quarter, our revenues increased sequentially over the third quarter by 3.8%.

If you annualized that rate of growth, it results in approximately a 16.1% annual rate of growth.

For full year 2008 to full year 2009, our revenue growth increased by slightly below our target at 9.4%.

However, adjusting for fluctuations in currency and a very challenging macroeconomic environment, our annual revenue growth was in fact 11.2%.

Tad would now like to cover some additional details concerning our results.

Thank you, Dave.

Again, good morning to everyone.

I would also like to thank and congratulate our entire team on their hard work and efforts not only for the quarter but also for the year of 2009.

I'll begin with providing some additional details on our revenues by product class.

First, regarding on-net revenue and on-net customer connections.

On-net revenue was $49.7 million for the quarter and increased 3.4% from the third quarter.

For the year, our on-net revenues grew by 7.1%.

About 85% of our new sales for the fourth quarter were for on-net services.

On-net customer connections increased by 555 customer connections, or 3.3% for the quarter.

And we ended the quarter with about 17,200 on-net customer connections on our network, and our on-net customer connections grew by 21.5% for the year.

Revenues from our off-net business were $11.8 million for the quarter, and that increased 6% from the third quarter.

Off-net revenues grew by 25.3% for the year.

Off-net customer connections were flat at about 3200 customer connections for the quarter, but grew 6.4% for the year.

Lastly non-core revenues are about $1.1 million for the quarter.

That's less than 2% of our revenues, and represent about 900 customer connections.

On EBITDA margin and gross margin, the operating leverage of our business continued to result in healthy growth in EBITDA margins.

Our gross profit margin declined by 30 basis points from the quarter from 56.2% to 55.9%, primarily due to a continued increase in investment in expansion-related costs.

Our gross profit margin was 57% for 2008 and 56.5% for 2009.

EBITDA as adjusted was $17.4 million for the quarter, which was an increase of 2.2%.

EBITDA as adjusted was $64.9 million for the year, and that increased 8.2% from 2008.

EBITDA margin declined by 40 basis points from the third quarter, and the margin was 27.8% for the fourth quarter compared to 28.2% for the third quarter.

EBITDA margin for '09 was 27.5% compared to 27.9% for 2008.

Quarterly lumpiness in our gross margin and EBITDA margins can and does occur.

If you examined our 19 quarters' quarterly metrics since second quarter of 2005, when we had our public offering, you'll notice occasional lumpiness in our quarterly gross margin and EBITDA margin expansion.

And this lumpiness can occur due to seasonal factors, such as the timing of the performance of certain professional services, including audit and tax services, and the timing of our expansion activities.

In particular, we experienced material increases in audit fees and an increase in payroll taxes from the resetting of payroll taxes in the first quarter.

We expect that to occur again in the first quarter of this year.

However, our long-term trend has demonstrated increasing gross margin and EBITDA margins.

We expect that long-term trend to continue.

Regarding earnings per share, our basic and diluted loss per share for the fourth quarter was $0.03 and our basic and diluted loss per share for the year was $0.39.

For the year, that was compared to a loss per share of 2008 of $0.34.

However, there were non-recurring items that impacted the comparability of loss per share for each of these periods.

Reducing the loss per share for 2008 was a gain of $23.1 million related to purchases of our convertible notes in 2008.

The impact of that on the loss per share was income of $0.52 per share for 2008.

If you adjust for that gain, the loss per share of '08 would have been $0.86.

Reducing the loss per share for the fourth quarter and full year 2009 was an income tax benefit of $1.5 million.

That income tax benefit reduced the loss per share for the quarter and for the year by $0.03.

Without that tax benefit, the loss per share for the full year would have been $0.42.

If you look at the EPS excluding items for 2008 would have been a loss of $0.86 and 2009, that improved to a loss of $0.42.

For the third quarter, the loss would have been $0.07 with no items, and for the fourth quarter adjusting for the tax benefit, would have been $0.06.

Dave mentioned foreign currency, and I'll provide some more details.

About 30% of our business is located outside of the US.

For the quarter, about 24% of our revenues were based in Europe and 77% of our revenues were related to our Canadian operations.

We have just begun to recognize revenue related to our operations in Mexico.

Continued volatility in the euro and Canadian dollar to US dollar conversion rates materially impact our comparable quarterly revenues and annual revenues in our financial results.

The impact on our revenues Dave mentioned was about 600,000 on a sequential quarter basis.

When you compare fourth quarter of '09 to last year's fourth quarter of '08, the impact was an increase of $2.2 million.

However, the euro was weaker for the full year of 2009 versus the full year of 2008.

So on an annual basis, the impact on our revenues was a decrease of $3.9 million and that lowered our annual revenue growth rate from 11.2% to 9.4%.

The euro to US dollar average for the third quarter was $1.43, and that increased to $1.48 in the fourth quarter.

The average was $1.32 for the fourth quarter 2008.

For year of 2008, the average was $1.46 and that declined to $1.39 for 2009.

Regarding the Canadian dollar, the relationships were similar to the differences we saw in the variability in the euro to the US dollar.

The Canadian to US dollar was $0.91 for the third quarter, and that increased to $0.95 for the fourth quarter.

For the fourth quarter of '08, it was $0.83.

The Canadian to US dollar average was $0.94 for the full year of '08, but that decreased to $0.88 for the full year of '09.

As we are here today, the euro to US dollar rate is about $1.35.

If you used that rate for the remainder of the first quarter, including what the rate has been for January and February through today, that would result in an average rate of about $1.38 of the first quarter of 2010.

That is a material reduction from the actual average rate of $1.48 for the fourth quarter of 2009.

As a result, our first quarter 2010 revenues will be facing significant foreign currency headwind related to the impact of the euro.

And at this time we estimate the negative foreign currency conversion impact of sequential quarterly revenues is about $1 million.

Regarding capital expenditures, our capital expenditures totaled $7.7 million for the fourth quarter, down from $16.7 million for Q3.

For the year, capital expenditures were $49.5 million, compared to $33.5 million for 2008.

The annual increase is related to the significant increase in our network expansion activities.

We added over 11,700 miles of fiber to the network in 2009, including material geographic expansion into new locations.

On a quarterly basis, we can and have historically experienced seasonal variations in capital expenditures and construction activities.

Quarterly capital expenditures are, in part, dependant on the number of buildings we connect to the network each quarter and the timing and scope of network expansion activities.

Typically we experience our lowest level of capital expenditures in our fourth quarter.

We added 30 buildings to the network in the quarter.

That was similar to the 32 buildings we added in the third quarter.

For the year, we added 125 buildings to the network.

That was 15% more than the 109 buildings we added in 2008.

We expect to continue our network expansion in 2010 and to add approximately another 120 buildings to the network in 2010.

We also expect to further increase our fiber miles in 2010, but at a lesser rate of expansion than we experienced in 2009.

As a result, we expect our 2010 CapEx to be approximately 20% less than our 2009 capital expenditures.

Some balance sheet items.

We increased our cash by $4.8 million from the third quarter to the end of the year.

This is a true and pure cash flow increase, more cash in the bank at the end of the quarter than at the beginning, and included no financing activities.

As of year end, our cash and cash equivalents totaled $55.9 million.

For the year, our cash declined by $15.4 million as we invested operating cash flow in our network and in our geographic expansion.

We have about $92 million of our original $200 million face value of our convertible notes remaining.

Those notes mature in June of 2027.

They are reported on our balance sheet at $66.3 million, which is net of their unamortized discount.

Capital lease IRU obligations were $109.7 million at year end, and about 5.6 million of this amount is a current liability.

As a reminder, these obligations are for long-term dark fiber leases and are being paid over a remaining weighted average period of more than 13 years.

In October 2009, we entered into a $20 million resolving credit facility that may be used for general corporate purposes, acquisitions and share or note purchases.

We have not borrowed under this facility.

Our days sales outstanding for worldwide accounts receivable was 31 days at December 31, 2009, and that metric is again significantly better than our targeted rate of DSO of 40 days.

This was also an improvement from the 33 days outstanding at the end of the third quarter.

Again, I want to personally thank and recognize our worldwide billing and collections team for continuing to do an unbelievable job on customer collections and credit monitoring.

In this economic environment, we continue to closely monitor our credit standards and our accounts receivable.

Further demonstrating the performance on customer collections, bad debt expense was only 2.1% of revenues for the quarter, and also 2.1% of our revenues for 2009.

Our average contract term continued to increase, and increased by another 2.7% for the quarter.

Increases in average contract term and credit monitoring activities can and have impacted our revenue performance under US GAAP.

For example, increases in our average contract term and estimated customer life results in a longer period over which to recognize installation revenues that we bill to our customers.

An increase in the amortization period reduces our revenues under US GAAP.

Additionally, closer credit monitoring activities can result in the rejection of certain orders or the termination of an account.

Operating cash flow.

Cash flow from operations was $16.3 million for the fourth quarter, an increase from $14.8 million for the third quarter of 10.8%.

For the year, operating cash flow was $56.9 million, an increase of 4.8% from the $54.3 million for 2008.

Lastly, we continued our progress toward achieving net income and positive earnings per share by increasing our operating income from $0.5 million for the third quarter to $1.4 million in our fourth quarter.

I will now turn the call back over to Dave.

Thanks, Tad.

I would like to take a moment and talk about our sales force and the activities and productivity of that team.

We began the fourth quarter of 2009 with 250 sales representatives and ended the quarter with 267 sales reps, quota bearing reps, selling our services.

We began 2009 with 244 quota bearing reps.

We hired 76 reps in the quarter, and 59 sales reps left the company during the quarter.

Our monthly turnover of reps improved slightly to 7.6% monthly from the 7.8% in the third quarter.

Our monthly rep turnover was 7.5% full year of 2008 and 8.1% for full year 2009.

We began the fourth quarter with 215 full-time equivalents, these are fully ramped reps, and ended the quarter with 235 full-time equivalents.

We began 2009 with 219 FTEs.

Productivity on a full time equivalent basis was 3.6 per rep in the quarter.

This is a monthly number.

This performance was relatively in line with our historical organic productivity rates.

As a reminder, our sales rep productivity is not based on contract signing but rather based on the successful installation of our customer orders.

Our productivity per full-time equivalent in 2009 for the full year improved to 4 units per rep from the same metric in 2008, which was 3.6 orders installed per rep per month.

Now to address the scale of our network.

We added 30 buildings to our network in 2009 and we ended the year with 1451 buildings on our network.

That represents approximately 600 million square feet of multi-tenant office space, as well as a number of data centers which we are connected to.

For 2009, we added 125 buildings to our network, 5 more than our previous expectations that were announced of 120 buildings, and 16 more than we actually added in 2008.

Now we've added to our network footprint as well.

We now have over 13,300 metro fiber miles and over 46,000 route miles of intercity fiber, an increase of over 6,000 miles from the beginning of the quarter and over 11,700 miles since the beginning of the year 2009.

We expanded our network into additional markets.

In Europe, we expanded into two new countries, in Poland we have expanded the network to serve Warsaw and Krakow.

We have expanded the network to serve Luxembourg.

In existing countries in Europe, we added Cambridge in the United Kingdom, Antwerp in Belgium, and Rotterdam in the Netherlands.

Here in North America we've also expanded the network, adding two new markets, Boise, Idaho and Memphis, Tennessee.

We continue to evaluate additional fiber routes both in Europe and North America.

The Cogent network is one of the most interconnected networks in the world.

Today we are directly connected to over 2970 unique networks.

We believe that our network has substantial capacity to accommodate additional growth.

We are currently only utilizing about 19% of the lit capacity in our network.

Traffic continues to grow on our network, and our revenues are growing much faster than revenue rates of growth among our competitors, demonstrating our gain in market share.

We believe that Cogent's position as a low cost provider and value proposition we offer our customers makes us unmatched within our industry.

Our pricing strategy has attracted many new customers, increased our average contract term, and significantly increased the volume and revenue commitments from existing customers.

Our business remains completely focused on the fastest growing segment of the telecommunications market -- that is, the internet.

And because of that focus, we are providing to our customers a utility which is necessary for their businesses.

We have a very strong balance sheet especially when compared to others in our industry.

During the quarter, we increased our cash balance with no financing activities by $4.8 million.

We are putting some of that cash flow to work by expanding our international footprint and expanding the number of markets we serve.

We continue to be extremely encouraged by our sales initiatives and the size of our order backlog.

We recognize, however, that the recent strengthening of the US dollar will negatively impact our growth in the very short-term, as Tad mentioned.

We are committed, however, to providing top line revenue growth of between 10% and 20% annually and continuing to expand both our gross margin and EBITDA margins on a long-term basis and most importantly continuing to focus on increasing the amount of free cash flow that we generate for the benefit of our equity holders.

With that, I'd like to now maybe open the floor up for questions.

(Operator Instructions) We will take our first question from Jonathan Schildkraut with Jefferies.

Great, thanks.

Two questions, really.

First, Dave, if you could talk a little bit about the competitive environment, if there have been any changes over the last 90 days?

And then the second question has to do with video growth.

Obviously, we are starting to see more and more news flow around video over the internet, and over the top video could be a very important element of the business model for 2010.

So, I would certainly like to hear a little more on that.

Thanks.

Hey, thanks for your questions, Jonathan.

First of all with regard to the competitive environment, Cogent has two very distinct addressable markets.

And in each of those markets, we generally have different competitors.

For our NetCentric business, we are connected to about 425 carrier-neutral data centers.

In those centers, we compete with about a dozen other global tier 1 backbones, two or three of which actively compete with us for high volume, low price type of bandwidth.

For those customers, we continue to provide the best value and take market share.

You see that in our traffic growth numbers growing 23% and the sequential revenue growth in our NetCentric business.

So if you look at the calculus there, traffic grew 23%, average price per megabit did decline 5.1%, but the result of those two inputs resulted in a 5% growth in our sequential revenue.

A couple of our competitors have spoken about the pressures of high speed IP transit pricing and have pointed to that as reasons for declines in either all or part of their revenue base.

I think many of our competitors believe that the pricing levels today are unsustainable and the rate of decline is also unsustainable.

We are demonstrating that not only can we make money at the current prices, but we can make money in an environment where for the NetCentric base, prices are declining.

I think the competitive landscape has actually gotten much more benign, as many of our competitors continue to focus on trying to upsell customers to other products and did not want to go in and directly compete with Cogent for the products that we sell.

I think on the NetCentric side, we are seeing an improving competitive dynamic.

On the corporate side of our business, we are generally competing with large incumbent providers who sell a complete vertically integrated bundle of services.

Those companies have been reluctant to accelerate the cannibalization of their voice traffic, which is still generating a majority of their revenues.

So in those market spaces, we have seen very stable pricing with Cogent's core product, a 100 megabit per second internet connection for an average of $900 a month being offered at the same price today as 10 years ago, we see stable pricing and a very, kind of, limited set of competitors in that market.

Sometimes we see competitive carriers reselling but they generally cannot compete with us on our facilities based network.

In our off-net corporate foot print, we are encouraged by the fact that ethernet services have become more ubiquitously available, and we have negotiated very favorable terms with our two major ILEC partners to provide those tails.

We then add the IP layer to those services.

In total, the competitive environment, we believe is improving.

Now, with regard to your second question about video.

Today about 70% of the traffic on our network is video.

Virtually all of the bit volume growth is coming from video and we believe, particularly as the FCC continues to support the idea of net neutrality, over the top video will become more and more of a mainstream phenomena, and with that we see years of potential traffic growth rates similar to the growth rates that we've demonstrated this year.

And with our pricing model, that will result in revenue growth in our NetCentric business.

Great.

Thanks, Dave.

Thank you.

And we'll take our next question from Mike McCormack with JPMorgan.

Good morning, Dave.

Scott Goldman here, on for Mike.

A couple of questions.

I wonder if you can talk a little bit about incremental margins?

They seem to have been declining, I think, obviously due to heavy expansion of your route miles and intercity miles this year.

I was wondering if there is anything else we should be thinking about beside that expansion that may be impacting that margin?

And then, if you can set the tone for expectations on expansion in 2010.

Secondly, I just wondered if you can touch base on what percentage of your NetCentric customers are currently on long-term contracts?

Sure.

Let me start with margins.

We sit down and weigh the uses of cash flow that the company generates.

Over time, we have bought back stock, we have bought back our debt.

And in the most recent downturn, we decided in conjunction with the board, that acceleration of our network expansion to ultimately reach our targeted addressable market made the most sense.

As Tad mentioned, we added about 11,700 miles in 2009.

We expect that number to be significantly less in 2010, but still relative to the installed base, it will be a significant number.

The result is we will see a lowering of our capital expenditures, and we should begin to see some expansion both in gross margin and EBITDA margin.

The pressure on our margins comes really from two things.

One, the activation of new portions of the network with very few customers in those locations, very little revenue until they mature to a level similar to other parts of the network.

So if you look at our net building foot prudent, for example, we have 8.5 customers per corporate building and 17.3 customers per data center on average in the network across the 1451 buildings at the end of the quarter.

If we add new buildings, on day one they have no customers, so it takes a while for them to equalize to the rate on the network and we did suffer some degradation in margin as a result of that.

But it was calculated ultimately to help us reach the optimal addressable market.

But we should see probably towards the end of 2010 and 2011 resumption of our continued expansion in margin.

Tad may want to chime in a little more on that and I'll let him talk about the long-term contracts as well.

The other thing that could impact margins is the percentage of our business that is coming from off-net, and as I mentioned, we have entered into agreements with major incumbent providers to sell us tail circuits at very competitive prices that we've had great deal of success in the market with.

That takes our ARPU up for those off-net services, but those services come with a true cost of goods sold, an incremental cost of goods of about 50%.

So therefore that will also degrade our margins from the effective margin rate of our on-net business.

We do expect over the long-term to continue to see our on-net business grow at a faster pace than the off-net.

In the short-term, we've had some acceleration in that off-net as a result of these circuits being available.

And I'd like Tad to touch on the contract question.

A few numbers to give some background.

If you look at the quarterly changes in the overall average contract term.

And this is on an order basis, so if you did it on a revenue basis, the increase would actually be greater.

From the first quarter through the fourth quarter, each quarter sequentially it was about 3%, with a high of 3.8% from Q1 to Q2 then and a low of 2.6% from Q2 to Q3.

We continue to sign longer term contracts on an average basis.

If you look at where we stand at the end of the year, approximately two-thirds of the contracts are one year or longer, and the mix of contracts that are two or three years has been increased, increasing.

Again I don't have those figures here, but if we did it on a revenue basis, it would even be more of an increase because most of the contract extensions are from the NetCentric contracts.

That's very helpful, guys.

Just going back on the margin side, any FX benefit that you guys can quantify as well.

You know, in the past, some of our European operations were of smaller scale and there may have been some distortion, but today at the end of fourth quarter, margins in our European and North American and Canadian businesses all are relatively similar.

And for that reason, FX will probably equally impact top line and bottom line performance.

Great.

Very helpful, guys.

Thanks.

Our next question comes from Michael Rollins with Citi.

Hi, good morning.

Hi, Mike.

I wanted to follow up on some of the comments earlier in the call.

You said you were very encouraged by the bookings but the billing had not hit yet.

I was wondering how we should be thinking about that in terms of sales productivity and in terms of the on-net off-net mix of that incremental bookings you were discussing.

Sure.

Two parts to the question.

First of all, remember our sales force productivity is measured only on install, not sold business.

Today, we actually have the largest backlog, meaning orders sold that have not yet been installed, in the company's history.

We had a very successful set of promotions in December, which is normally a fairly slow selling month, to result in one of our best months in the company's history and probably the best December we've ever had.

That resulted in a significant number of orders in the backlog to be installed.

The mix of those orders is actually distorted by off-net because the provisioning times of off-net are much longer than that of on-net.

So if you looked by unit number of order, probably 55% to 60% of the orders are in the on-net bucket.

Those provision relatively quickly, and about 45% of the orders are off-net orders, which have a much longer provisioning window.

Because almost all of the off-net sales that we are doing at this point are ethernet-based requiring special assembly by the ILEC, the average provisioning time of those orders is about 120 days.

So we should continue to see some significant increases in installed orders as orders in those funnels, both on-net and off-net, continue to come out, and the result will be hopefully in this quarter an uptick in rep productivity just based on what's in the funnel versus what gets installed.

Great.

Thanks very much.

And we'll next hear from Colby Synesael from Kaufman Brothers.

Great.

Thanks for taking my questions.

You talked a lot about obviously how you are expanding the network, and I think some of those purchases are dark fiber as well, some the fiber capacity solutions.

I was wondering if you could talk about the opportunity to reduce the cost of goods by taking some of your lit service capacity today and moving that on to some of the dark fiber you may be purchasing, and maybe some timing in terms of when that could actually start to occur.

And also just a question similar to the same line of thinking, as it relates to CapEx, can you give us some linearity in terms of what CapEx is going to look like in 2010.

Thanks.

Hey, thanks, Colby.

Two different questions, first of all with regard to the fiber purchases, our network is comprised, our backbone network, both in the metro and long haul, is comprised exclusively of dark fiber, no lit services with the exception of transatlantic capacity.

We do buy lit services up between North America and Europe and we have chosen to do so because the prices of those lit services is so competitive that it is more cost effective than owning IRUs on those cables.

But terrestrially, the 46,000-miles of intercity fiber and the 13,300 miles of metro fiber are all dark fibers.

So there is really no opportunity to reduce our backbone costs because we are buying no lit services today.

Our lit services are primarily the tail circuits for our off-net customers.

As I mentioned, virtually all of our off-net customers are corporate.

Our average on-net building today has about 8.5 customers out of 51 opportunities in the building.

Average building of about 580,000 square feet.

If you look at our off-net customer base, it's very different.

As Tad mentioned, we have got approximately 32,000 off-net customers, and those customers earn about 3,030 buildings.

Our customer density in those buildings is only about 1.05 customers per building.

It does not make sense for us to buy fiber to serve those customers because the addressable market in those buildings is just too small to justify our capital outlay.

So we do not see any significant ability to expand margin through dark fiber IRU purchases, but we do see the opportunity to continue to grow our addressable market and improve our connectivity.

That is what is driving our IRU purchases.

Now let me let Tad maybe touch on the CapEx flow.

Sure.

CapEx for 2009 was $49.5 million, and that included material expansion in miles and also in networks and overall footprint.

As we looked at our plan for 2010 and where we believe we're going to be expanding to, we think that level will be approximately 20% less than it was in 2009.

Whether it's 25% or 15% will be in part just dependent upon the opportunities presented to us.

And frankly, the timing of some of these installations are somewhat out of our control because we are dependent upon another provider, but we think at this point about a 20% reduction is what we are planning for.

But there's no sense in terms of this being back end or front end loaded?

Whether or not you are planning to make a big push in adding fiber in buildings in the first half of the year versus the second?

I would say, no.

Although typically the seasonality we see is less in the fourth quarter.

That may recur this year.

It's happened over the past couple years.

And then, just moving back to the operating margins component of the business, I think you mentioned in the answer to one of your questions that you don't expect margins to improve until the end or late 2010.

Is that largely because the fiber purchases we've seen in the third quarter, third and fourth quarter of 2009 are going to be similar in terms of what you'll be doing one, two and three quarters in 2010?

It's actually based, Colby, more on the fact we purchased the fiber and picked up the operating cost, O&M, the operations and maintenance costs when we activate that fiber.

As I mentioned, with an expansion in our footprint of almost 20% in full calendar year 2009, we end up diluting the customer density along those routes, and it takes some time for those customers, those new markets and new routes to have the same kind of performance characteristics as more mature markets.

While there will be continued improvement in the operating efficiency and cash flow production and metrics in terms of margin on the existing footprint, it will be diluted by these new markets.

And we do expect some lumpiness.

As Tad mentioned in Q1, we have a couple of things we know are coming, which are increased audit fees and some of our sales meeting costs.

So we do expect to see some margin expansion throughout the year.

If you look at our long-term historical expansion over the 19 quarters as a public company, it's now roughly about three-quarters of 1%, both gross and EBITDA margins sequentially.

We believe in the long-term we can continue to do that.

But we think because of these expansions in 2009 and 2010, we're going to see that rate of expansion muted in the short-term.

Aggregate cash flow will go up, revenues will grow, and margins will expand, but they'll expand at a muted rate due to the expansions.

Okay.

So you do think that margins, whether we're talking about gross or EBITDA, they go up just at a very slow pace as opposed to what you expect long-term?

I think that's right.

Okay.

Thank you.

We'll take our next questions from Frank Louthan with Raymond James.

Great.

Can you give us an idea of the CapEx that you expect to build out the two data centers to make those operational, and I assume that's included in your CapEx currently.

Thanks.

Sure, Frank.

We increased our data center footprint from 37 to 39 data centers.

These two centers are actually completely built out centers that are vacant, so we did not just take over just shell space and have to build it out.

These centers were operational data centers in which the two operators failed or exited the centers.

So the capital required to reactivate these centers is minimal.

We will go into each of the two centers, recertify the fire suppression systems, recertify the power backup.

Generally the only consumable that needs to be replaced is generally batteries, which are fairly trivial in cost.

So you are correct that capital is included, but as a percentage of the roughly $40 million in CapEx we are expecting in 2010, these two data centers are trivial.

Okay.

Great.

Thanks.

And two other follow-ups.

What is the focus this year going to be from the sales force, is it going to be more from the NetCentric side or more on the commercial customers?

Where are they going to be focused?

And then, can you remind us, you usually have seasonal impact in the first quarter, some non-recurring items that tend to negatively impact EBITDA sort of off trend.

If you can remind us what those are and what you expect that pressure to be in the first quarter of this year?

Thanks.

I'll take the sales question, Tad will take the cost impact question.

In terms of sales focus, roughly two-thirds of our sales force focuses on corporate customers by head count, about one-third of our sales force focuses on NetCentric customers.

However, the quota levels for the NetCentric reps are much higher.

If you look at the quota objectives of the two teams, it's approximately 55% NetCentric and 45% corporate.

In fact, today the customer mix is 53.5%, 46.5%.

We think throughout 2010, the growth rates of the two respective businesses will be relatively similar, and therefore, result in that same kind of 50% to 55%, 45% to 50% split.

If you go back and look at our mix split, end of 2008 versus end of 2009, it's virtually identical.

In any one quarter there can be a little bit of lumpiness, we could find a big corporate customer with a lot of locations.

Actually two quarters ago we had a large retailer.

At the end of last quarter we signed a very large multinational legal and legal services firm with literally hundreds of connections both on-net and off-net.

We can have some big corporate customers that distort things or you can end up with a couple of large video customers such as the types that Jonathan talked about that while taking very few connections will generate tremendous revenue and bandwidth.

But I think the mix that you see at the end of year 2009 is probably pretty similar to what you should you should see in terms of productivity and mix for end of year 2010.

Good morning.

On kind of the, recurring, some of them recurring but at different rates.

The amounts we incur at the first quarter at different rates would be the finishing of the corporate audit which a great deal of that work takes place in the first quarter.

The resetting of the payroll taxes and we have our annual pay meeting -- in the aggregate, that's about $1 million from the fourth quarter to the first quarter you can see.

You can see this kind of trend.

If you look at our press release, the sequential changes in the EBITDA from the first quarter to the second quarter, both we've got the last eight quarters in the current press release, anyway, are probably the most significant increases.

Because you get from Q1 to Q2, we did not experience those seasonal increases.

That's kind of the level of seasonality we see with respect to those costs.

Great.

Thank you.

And our next question comes from Tim Horan from Oppenheimer & Company.

Thanks, good morning guys.

Dave, I'm sorry, can you revisit what your average selling prices are and how that's trending and what you expect here.

I know you gave the numbers.

I wasn't able to write them down.

And what you're seeing from the pricing front from competitors.

Thanks.

Yes.

Sure, Tim.

First of all on the corporate side, the most common product on-net is a 100 meg connection on a one year contract for $900 per month for those customers.

For the NetCentric customers, price per megabit is probably the right metric to look at.

As we mentioned, price per megabit in the fourth quarter of the installed base was at $6.01, down from $6.33 in Q3 of 2009.

So a reduction of 5.1%.

The average new NetCentric sale in the fourth quarter was $4.51 per megabit.

Great.

The new sales seemed to accelerate a little bit versus the last two quarters, but the embedded pricing can be a bit more stable.

Any thinking how that will impact next year's trends?

Would there be much difference from this year?

You know, we continue to offer discounts for term and for volume as Tad mentioned.

We also respond to the competitive landscape.

We have seen, as I mentioned in the answer to Jonathan's question, a very, I would say, benign competitive landscape compared to what we've seen in the past.

I do expect the price per megabit on the NetCentric side to fall perpetually, so some of that fall will be in response to market pressure, some will be to stimulate new business models.

I think the rate of price decline is probably going to be fairly consistent.

I think the important takeaway here is the ability to grow traffic in a declining price environment has resulted in now every quarter for the past year sequential revenue growth in our on-net NetCentric business.

The metrics for last quarter were very similar to the quarter before, which is traffic volume up 23%, price down 5.1%.

The product of those two resulted in growth of revenue from NetCentric customers for Cogent of 5% sequentially quarter over quarter.

I do believe that many of our competitors continue to lose market share, witness lower growth rates and their inability to get to their pricing levels.

On that point, it seems like your competitors have reported weak quarters and seems their trends are unsustainable from at least the public equity that they have.

Any thoughts maybe on the industry structure a couple years from now consolidation in that regards potentially your exit strategy, if any thoughts on that?

Thanks.

Sure.

Two points.

First of all if we look at the industry, I think the biggest structural change which our industry is going through is the migration of applications and traffic away from specific purpose networks, voice and cable TV broadcast, on to IP internet backbones which is extremely deflationary.

For the local access providers, they have a defensible barrier to entry and a natural somewhat monopolistic or duopolistic position.

So I think on the access side, people can be relatively assured that they're going to have a stable revenue stream while they may have to increase their capital cost to deliver those services.

And many of those companies also have a wireless component that they bundle with their access product.

Now it's voice, soon to be purely internet.

But I think the access networks are stable.

Facility-spaced access competition is difficult.

We have chosen to cream-skim the market and focus on a very specific demographic.

And I think we have a good business.

Any carrier who says we are going to ubiquitously go out and compete with the ILEC and build facility is probably going to fail because of the capital intensity of those builds and the relatively low penetration.

On the access side, I think you've got a fairly stable environment.

On the back bone side, I think many of the companies in that space have balance sheet challenges and operational challenges.

And those challenges are accentuated because the majority of their revenue is not internet.

So what they're doing is they are experiencing two forms of pricing pressure.

One they are reprising their own base as they move toward more IP and less legacy services.

And two, they need to become more competitive on their IP pricing.

So I think some of those companies will disappear and go away through either combination and/or liquidation.

So I think this is an industry that has gone through one round of consolidation back at the beginning of the decade, kind of 2001 through 2004.

We actively participated as an acquirer.

I think there will be another round of consolidation that will occur or is already beginning to occur.

In terms of our position -- we are in a great position, we are producing cash, we are growing our business.

It is true our revenue scale maybe smaller than others, but our revenue is growing and theirs is shrinking.

Those two will converge at some point in time.

I don't think we have any external pressures to necessarily do something inorganic, whether to be acquired or actively acquire other assets.

But we always remain focused on our shareholders and creating maximum value.

So other than those broad kind of parameters, I can't give you any more specifics at this point.

Always helpful, Dave.

Thanks a lot.

Thanks, Tim.

And we'll take our final question with Michael Funk with Bank of America Merrill Lynch.

Great, thanks for taking the question.

Two quick ones.

You mentioned you expanded to new markets in the quarter in Europe and also in the US.

I was hoping to maybe get some color around the pricing in those markets relative to your own legacy pricing.

Some of your competitors have talked about greater pricing pressure in Europe, for example, and if it's just a matter of more competition coming in like yourself.

In the second on the backlog, any comment you can give on your hit rate, or the change in your hit rate for converting backlog into revenue would be helpful.

Okay.

Let me take the new markets.

In North America, the markets we add include data centers and corporate buildings.

The new markets of Boise and Memphis here in North America are small compared to, obviously, places like LA and New York, but similar to places like Indianapolis or Louisville.

I think we'll see a competitive landscape very similar in those markets, and there is not a great deal of variability across North America, US and Canada in our corporate competitive landscape and pricing.

In terms of Europe, all of our buildings in Europe and virtually all of our customers are NetCentric.

Our buildings are data centers and our customers are NetCentric.

I think we have been very key in driving down prices in a number of new markets.

If you look at the some of the third party studies such as telegeography, it's pretty interesting to see price per megabit in all the markets.

You can go to our website and see which markets to see which markets we're in.

They're generally are 30% to 40% higher than Cogent's prices on average.

Cogent obviously being the low and they generally give the low and the high.

But we tend to reset the bar in those markets.

If you go to markets we are not in, in other parts of the world, pricing in many cases can be an order of magnitude more per megabit, a factor of 10 more, than what we sell at.

These two markets, several markets in Europe whether it be Luxembourg or the two cities in Poland or places like Rotterdam and Antwerp or Cambridge, we think pricing will be very similar to what it is in markets we are already in, in many of those companies.

For example, in the Netherlands, we are already in Amsterdam.

We don't anticipate any real difference in Rotterdam versus Amsterdam, Cambridge versus London.

Poland is a new country for us, and we'll offer our standard pricing.

We have been extremely aggressive in entering new markets in eastern Europe and driving down prices to stimulate demand.

And again, based on our cost model, we can do that with the margins we've been able to demonstrate.

In terms of the hit rate in backlog, let me separate that in into two parts -- for the backlog I described earlier, we actually have 100% hit rate, these orders are signed and are validated or waiting install.

And the breakage rate there is virtually nonexistent maybe 0.1 of 1% or something.

And so orders that have been validated, signed, customer paperwork, credit check has been done and we are just now provisioning them.

Virtually all of those orders install.

The second kind of backlog are the orders that are in our sales funnel, i.e.

they are being worked by the sales team but not yet closed.

That funnel also is robust, and our close rates have generally been about 50%.

That's been our historic average and we have not seen any material improvement or degradation in that number over the past several years.

And one quick follow-up, then.

Based on your comments from the call and you answered my question, sounds like the NetCentric business, and you have made a few comments about competitors that are stepping away, not wanting to pressure pricing further.

It sounds like a significant revenue growth opportunity there.

Or maybe an acceleration into the beginning of this year in those new markets.

Is that correct?

I think because we're in about 145 markets now, a half dozen new markets isn't going to materially move the needle.

While it would be great for Cogent and our competitors that the total dollar value of that NetCentric wholesale IP transit market grows, it has not over the past 10 years.

And our growth has come almost exclusively by taking share.

I think if you look at our growth rate, it's very consistent, it's very steady, and while I'd love to hope for an acceleration, I think slow and steady wins the race.

And I think you're going to see similar type growth out of Cogent over the course of 2010 and beyond as we've demonstrated in the past.

I don't think there's any one breakout event.

Great.

Thank you for the follow-up.

Tad reminded me here, FX obviously distorts things.

So just be very conscious.

We're not trying to hide behind that or take credit for it but we want people to understand in the non-US markets, while we sell at a constant price, the reported dollars can fluctuate materially.

.

And that concludes our question and answer question today.

I'd like turn the call back over to Mr.

Schaeffer for additional or closing remarks.

I'd like to thank everyone.

I know it's a very busy earnings day today and thank you for taking the time to hear about Cogent, and obviously any follow-up questions feel free to contact the team and we'll try to fill in any blanks.

Thanks a lot, everyone.

Bye-bye.

That does conclude today's call.

Thank you for your participation.