Cogent Communications Holdings Inc (CCOI) 2008 Q3 法說會逐字稿

Good morning, and welcome to the Cogent Communications Group Third Quarter 2008 Earnings Conference Call. As a reminder, this conference is being recorded and it will be available for replay at www.cogentco.com. That's www.cogentco.com.

I would now like to turn the call over to Mr. Dave Schaeffer, Chair and Chief Executive Officer of Cogent Communications. Please go ahead, sir.

Okay, thanks, Lakeita. Thank you, and good morning. Welcome to our third quarter 2008 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. In the third quarter we experienced significant foreign exchange headwinds and certain non-recurring events both positive and negative. Overall, we are relatively pleased with our results for the quarter and very encouraged by the growth in traffic on our network.

Our revenue and EBITDA did not meet our third quarter guidance. However, adjusting for the impact of substantial changes in foreign exchange rates we exceeded our Q3 2008 revenue guidance. Over 30% of our business is outside of the United States. During the quarter traffic grew on our network by 5%. We significantly expanded our footprint, as well as our sales force.

We took advantage of opportunistic market conditions by purchasing over $106 million of our original $200 million convertible notes for about $0.45 on a dollar. These note purchases reduced our true debt by 53% resulting in a gain in the quarter of $9.7 million and an additional gain of $46.8 million in October. The total gains is $56.6 million.

As Tad will outline these transactions resulted in net income for the quarter and for the fiscal year and we were able to affect these transactions in a tax free manner except for AMT. As we outlined in our last call in June 2008, we expanded a pilot program and offered term discounts to all new Cogent [NetCentric] customers who purchased larger volumes and to our existing NetCentric customers who increased their internal contract value with us.

During the quarter, this program continued to achieve great success and our quarterly rep productivity continued to increase on a monthly basis from 3.4 units per rep per month in Q2 to 3.8 units per rep per month in the third quarter. Additionally under the program, NetCentric customers with existing contracts representing approximately $26 million in remaining contract value chose to increase their contracts by an additional $12 million in contract value or nearly 50%. This increased the total remaining contract value from these customers to over $38 million.

Finally, our average contract length for all customers continue to expand and grew by 4% in quarter over quarter as more customers continue to extend their contract and show greater confidence in Cogent.

We ended the quarter with over 220 sales representatives. We continue to invest and improve our sales organization. As of today, we have 235 quota bearing reps selling our services. During the quarter, in addition to the note purchases, we also continued our stock buyback program.

In the third quarter we returned a total of $12 million to our shareholders through stock purchases of approximately 1.2 million shares. As of today, we have approximately $32 million remaining under our authorized programs for stock buyback. Since June of 2007 under our total program, which equaled $150 million in value, we have purchased a total of 6.3 million shares of our common stock for approximately $118 million. These purchases represent over 12.5% of the outstanding common shares since these programs began.

Throughout our discussions as in the past, we will continue to focus our results and impacts on On-net business. Our On-net business represents approximately 81% of our total revenues. We will continue to highlight several operational statistics which we believe demonstrate our increasing market share, expanding scale and operating leverage of our On-net business.

I will view in greater detail certain operational highlights and our continued expansion plans. Tad will provide some additional details on our financial performance. Tad will also walk through some of the foreign exchange assumptions in our guidance for the fourth quarter and update our guidance for full year 2008 and provide our initial guidance for full year 2009. Following our remarks we will open the floor to questions and answers.

Now, I'd like to ask Tad to read the Safe Harbor language.

Thank you, Dave, and good morning, everyone. This third quarter of 2008 earnings report and this earnings conference call are to discuss Cogent's business outlook and contain forward-looking statements within the meaning of Section 27A and Section 21E of the Securities Act. The forward-looking statements are based upon 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 differ 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 revision to any forward looking statement made today or otherwise update or supplement statements made on this call. Also during the 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 Cogentco.com.

I'll turn the call back over to Dave.

Okay, thanks, Tad. Hopefully, you've all had a chance to review our earnings press release. As within previous quarters our press release includes a number of historical metrics. These metrics will be added to our website. Hopefully, you'll find these metrics informative and helpful in understanding the financial results and trends from our operations. As always, if you have additional suggestions or metrics you would like us to add or refine, please let us know.

Our third quarter of 2008 revenue of $54.6 million was $400,000 below our guidance of $55 million. The increase in revenue represented a 1.4% increase over second quarter 2008 and a 16.2% increase over Q3 2007. The impact of foreign exchange materially impacted our revenues for the third quarter causing a revenue reduction quarter-over-quarter of approximately $600,000.

As Tad will detail, a substantial revision in foreign exchange rates materially impacted our previously expected results for the fourth quarter and beyond. EBITDA as adjusted, of $14.2 million was $2.3 million below our guidance for the quarter of $16.5 million. Four factors materially contributed to this EBITDA shortfall.

During the quarter, we incurred substantial write-offs associated with the loss of one large customer. We were actually successful however in picking up additional revenue as many of the customers of that customer chose to purchase from Cogent.

We also incurred a substantial increase in professional fees associated with a legal dispute that has received some press attention rightly. We have accelerated the expansion into certain markets taking advantage of some beneficial pricings. We've also incurred some increased seasonal power costs. And finally, as we experience with our international revenues the FX we noticed a reduction in our EBITDA from the translation of these foreign exchange rates. Many of our costs reside here in the United States and our impact on EBITDA is greater than that of our impact on revenue.

Our overall business grew by 0.001% second quarter of 2008 to second quarter of 2000 -- or third quarter of 2008 and approximately 17.5% from third quarter 2007. The impact of foreign exchange rates reduced the growth in this revenue part of our business by approximately $500,000. Substantially all of our European and Canadian revenues are On-net revenues so changes in foreign exchange rates impact our On-net revenues proportionately compared to other portions of our revenue. Approximately 89% of our new sales in Q3 were for our On-net services.

Our On-net customer connections increased by approximately 6.4% quarter-over-quarter. Revenues from our Off-net businesses increased by 6.3% Q2 2008 to Q3 2008 and our non-core revenues actually increased by 14.4%. Off-net quarter-to-quarter revenue growth was consistent with our guidance for Off-net revenue of approximately 5% and was the highest Off-Net revenue growth rate we have experienced since Q1 of 2005.

Now I'd like to take a moment and talk about pricing and provisioning. Pricing for our most widely sold product remains $1,000 per month for a month-to-month 100 megabit connection or $10 per megabit. We do offer discounts related to contract term, as well as for larger volumes. As a reminder, our NetCentric pricing program does not impact our corporate customer base. Our corporate customers represent approximately 60% of our total customers and represent in Q3 46% of our revenue. This is up from 44% of revenues in Q2 2008.

Our provisioning cycle guarantee remains unchanged and we guarantee our customers who purchase On-net services from us a 17-day provisioning cycle. In the third quarter our provisioning team again consistently outperformed our guarantee averaging ten business days against our 17-day guarantee. Traffic increased on our network by 5% from Q2 reversing the decline we had seen from Q1 2008 to Q2 2008. Traffic increased from the comparable quarter in 2007 by approximately 24%.

Now, I'd like to talk on our ARPU or Average Revenue Per Customer. Our On-net ARPU was approximately $1,210 in the second quarter declining to $1,143 in third quarter, a decline of approximately 5.6%. The impact of foreign exchange as well as discounts associated with contract volume and contract length contributed to this decline. Additionally, we did see an increase in our corporate customer base relative to our NetCentric base and corporate customers, ARPUs tend to be lower.

Our Off-net ARPUs continued to increase as our average Off-net connection size continued to increase. Off-net ARPUs increased from $940 in Q2 of 2008 to $1,001 in Q2 2008, an increase of 6.5%. Substantially all of our European and Canadian revenues are On-net, so fluctuations in foreign exchange have virtually no impact on our Off-net ARPUs.

Our total churn remains consistent at about 2% of revenues. On-net churn is in line with the rates that we've experienced previously for the past two years and our Off-net churn also remains consistent at about 2.5%. Tad will now cover some additional details concerning our third quarter 2008 results. Tad will also provide updated guidance for fourth quarter of 2008, as well as update our guidance for full year and also provide our initial guidance for full year 2009.

Thank you, Dave and again, good morning to everyone and I'd also like to thank and congratulate our team on very hard work and their efforts this quarter. Going back to EBITDA and gross margin. As Dave mentioned, EBITDA as adjusted was $14.2 million for the quarter which was a decrease of 14.6% from the $16.6 million for the second quarter. That resulted in an EBITDA margin contraction of 480 basis points. We were 30.8% for the second quarter and just about 26% for the third quarter.

As Dave mentioned during the quarter we incurred a bad debt write-off of $600,000 from an unpaid accounts receivable balance associated with the loss of one of our large customers. That customer has been replaced by the customers of that customer, which we'll -- I'm sure cover in the question and answer section.

We incurred an increase in professional fees as well associated with a publicly announced legal dispute. We also took advantage of some beneficial pricing and have accelerated our expansion into certain markets. And as we experience with our international revenues, we also experienced an EBITDA reduction from the translation impact of foreign exchange on our international EBITDAs.

Gross margin decreased by 150 basis points for the quarter. Third quarter was 55.9%, 57.4% was the second quarter. The decline is in part due to foreign exchange. The additional costs associated with our accelerated network expansion and seasonal increases in power and utilities costs. Additionally, because most of our fixed network costs are located in the US, unlike the foreign exchange reduction in our comparable revenues there's less of a corresponding translation decrease in costs of goods sold in Europe.

Our direct incremental On-net gross margins continued to be almost 100% and direct incremental On-net EBITDA margins continued to be approximately 95%. Earnings per share. Earnings per basic and diluted common share was $0.05 positive for the quarter. The $9.7 million gain for the quarter on the buyback of convertible notes increased EPS by about $0.22 per share. If you exclude this gain, our loss per share would have been about $0.17 for the quarter.

Weighted average shares decreased by approximately 1.8 million as a result of the weighted average impact of shares purchased under our stock buyback program. And as Dave mentioned, we purchased an additional 1.2 million shares in the third quarter and our plan for the remainder of '08 and '09 anticipates and includes the estimated impact of additional share repurchases.

Non-cash equity based comp expense for the quarter was $4 million or $0.09 a share and that was slightly less than our guidance of about $4.5 million for the quarter. Depreciation and amortization expense was $15.5 million. Again, slightly less than the guidance we had which was $16 million for the quarter.

We've covered this -- a little more detail on foreign exchange. About 30% of our business is outside of the US, 22% being in Europe and 8% related to Canada. And as Dave mentioned, these unprecedented rapid changes in the euro, Canadian dollar, the US dollar conversion rates negatively impacted our comparable revenues by about $600,000.

The euro to USD average was $1.56 in the second quarter and it dropped to $1.50 in the third quarter. As many of you know, the current euro to US dollar rate is about $1.30 and it's fluctuating wildly. Variations in foreign exchange materially impact our US GAAP revenues. For the 22% of our business in Europe, for each penny decline in the euro to US dollar we have a negative impact of our euro based revenues of about 100,000. For the 8% of our business in Canada each penny decline in the Canadian to US dollar rate negatively impacts our Canadian based quarterly revenues by about 40,000.

Capital expenditures. Capital expenditures totaled $9.5 million for the quarter, up from $9 million for the second quarter. And as a reminder, on a quarterly basis we can and have historically experienced seasonal variation in our capital expenditures and construction activities. Typically, we experience our lowest level of CapEx in our fourth quarter.

We expect to add slightly more than the 100 buildings originally targeted for 2008, and we'll have added more routes than originally planned as well. As a result, we expect our 2008 capital expenditures to be slightly over the $30 million estimate.

Regarding cash, debt and our balance sheet. As of September 30, our cash and cash equivalents and short-terms investments were $109.2 million. In September, we purchased the $20 million face value of our notes for $9.9 million in cash. After the quarter end in October, we purchased an additional $86 million in the face value of our notes or $37.8 million in cash. So in the aggregate, we purchased 106 million of our original 200 million face value of notes for $47.7 million in cash. That result in a gain of $56.6 million in net income for the quarter and for the year 2008.

So after these transactions we retired 53% of our true debt, which is the convertible notes, and we have about 94 million of the original 200 million remaining. As a reminder, the remaining notes are not due until June 2027. Our capital lease IRU obligations totaled $103.4 million at the quarter end and about $6 million of this is a current liability, so due within a year. These lease obligations are for dark fiber and being paid over a remaining weighted average life of more than ten years.

Days sales outstanding for worldwide accounts receivable continued to improve and was 31 days at September 30 better than our target of 40 days, and an improvement from the 32 days outstanding at the end of the second quarter. And despite the one isolated incident we are not experiencing a deterioration in accounts receivable collections.

And, again I need to thank our worldwide billing and collections team for continuing to do a fantastic job on customer collections. Our service as a necessary utility for our customers and our customers pay their Internet access and voices, and our team does a great job in following up on collections.

Operating cash flow. Cash flow from operations improved significantly for the quarter by 25% and was almost $18 million, compared to $14 million for the second quarter. Regarding guidance for the fourth quarter, updating the year and for 2009. The guidance is based upon our current and expected run rates of our business, includes the estimated impact of our share repurchase program, planned increases in our sales and marketing programs and the estimated impact of our current and anticipated network expansion projects and current traffic trends.

Our estimates are based upon current foreign exchange rates. Any material changes in these rates will materially impact our forecasts as it has materially impact our results. Any share and note purchases will also impact our forecasts. Share and note purchases primarily impact our estimates with interest income, interest expense and any gains on future note repurchases. Also, the share repurchases impact our weighted average shares outstanding for the period and in the aggregate these transactions obviously impact our EPS estimate.

So for the fourth quarter of 2008 we expect our fourth quarter 2008 revenues to be over $55 million. We expect EBITDA as adjusted for the fourth quarter to be over $14.5 million. Our gains on the fourth quarter note purchases are expected to be about $47 million. These results combined are expected to result in the basic and fully diluted EPS per share of over $0.90. Our loss per share assumes approximately $42 million weighted average shares outstanding.

Updating the guidance for fiscal 2008. Total revenue for 2008 is expected to now be over $215 million. Our previous guidance was $218 million. The significant change in the FX is caused over a $3.0 million reduction in our previous forecast. We expect EBITDA as adjusted to now be over $60 million. Our previous guidance was over $65 million.

Gains on our notes will be approximately $56.6 million. As a result we expect net income for fiscal 2008 and expect our 2008 earnings per basic and diluted common share to be between $0.55 and $0.60. The previous estimate was a loss of between $0.50 and $0.60 per share. The gain on the note purchases represent about $1.25 per share for fiscal 2008. Our EPS estimate assumes approximately 44 million weighted average shares outstanding for the year.

Initial guidance for 2009. Total revenue is expected to be over $250 million. EBITDA as adjusted is expected to grow to between $75 million and $80 million. As was mentioned in our press release, we have a required accounting change that's going to result in about $4 million of additional non-cash interest expense in 2009 and this is related to what is now our 94 million of convertible notes.

Cogent and companies with convertible notes similar to ours are required to adopt this accounting change on the first of the year 2009. The pronouncement in this call the APB 14-1, Accounting for Convertible Debt Instruments That May Be Settled In Cash Upon Conversion. The pronouncement essentially requires that the convertible feature of our notes be bifurcated or separated from our notes and then separately valued. The accounting results in an increase in the debt discount and a corresponding increase in paid-in capital. That increase in the debt discount will result in additional non-cash interest from amortizing that increased debt discount.

The accounting rules further require that the discount be amortized to the earliest put date of our notes which is June 2014, instead of the actual maturity date, which is June 2027. There's no change or impact at all for cash flow. The pronouncement also requires, which is unusual retroactive treatment, so Cogent and all other companies impacted by this change will need to restate their 2007 and 2008 operating results for the period that the notes are outstanding.

We expect our 2009 loss per share to be between $0.15 and $0.20 per share and the impact of this accounting change APB 14-1 will contribute about $0.10 to that loss per share for 2009. Any additional note purchases will impact that estimate. Our 2009 EPS estimate assumes approximately 40 million weighted average shares outstanding. And then lastly, we expect our 2009 CapEx to be similar to the 2008 rate.

And with that, I will turn the call back over to Dave.

Thanks, Tad. I'd like to take a moment and discuss sales force productivity. As many of you know we have implemented a number of measures to improve our sales organization and we are seeing the benefits of those improvements. We began the third quarter of 2008 with 208 sales reps, and ended the quarter with 223 quota bearing sales reps.

We hired 62 reps in the quarter, the most reps we have ever hired in a single quarter in the Company's history. 47 reps left the Company during the quarter slightly up from the 41 reps that left the Company in Q2. Our monthly sales force turnover remains approximately the same at about 7%.

We began the third quarter of 2008 with 177 full-time equivalent sales reps -- those are reps that are fully ramped, and ended the quarter with 194 full-time equivalent sales reps. Today, we have 235 quota bearing selling our services.

Productivity on a full-time equivalent rep basis for the third quarter of 2008 was 3.8 unit sales per rep per month. This rate was substantially higher than the 3.4 units per rep per month that we experienced in the second quarter. As a reminder, these rates are derived based upon customer installations, not upon contract signing so all installations are then counted.

We continue to have a disciplined approach to managing our sales force and monitoring their productivity. We continue to add sales reps in 2008, and we expect to have approximately 250 quota bearing reps by the end of the year. We also continue to enhance our sales training and sales management infrastructure.

Now with regard to our network. We added another 27 buildings to the network in the third quarter of 2008 and ended the quarter with 1,301 buildings directly attached to our network with fiber into the buildings [to the floor] to each customer. For 2008 we expect now to exceed our goal which was initially 100 buildings and for the first nine months of 2008 we have added 84 buildings to the network.

For 2009 we expect to continue this pace and add approximately 100 buildings to our network. We have secured additional fiber and construction is underway in extending our network into a number of new markets.

In Europe, we're extending into Tallin in Estonia, the first time we have entered the Baltics. We are extending our network in Romania to Timisoara, Lucerne in Switzerland. In North America a number of extensions as well adding cities such as Worcester, Massachusetts, Louisville, Kentucky, Nashville, Tennessee, Colorado Springs, Colorado and Albuquerque, New Mexico. We continue to evaluate additional routes both in Europe and North America.

Now I'd like to take a moment and comment on the size and scale of our network. We have over 12,000 miles of metro fiber. We have over 33,000 route miles of inter-city fiber. We are directly interconnected today to over 2,400 networks. Approximately 320 of these networks are settlement free peering partners of Cogent. The remaining almost 2,100 networks are today customers of Cogent. We believe that our network has substantial capacity available for it to continue to grow and accommodate our growth plans into the foreseeable future.

Today, we are utilizing approximately 20% of the [link] capacity in our network. This is actually slightly down from the 21% last quarter due to expansions in our network, even though we have experienced substantial traffic increases.

On-net revenues, our growth business continues to grow and traffic once again is growing on our network. We believe that Cogent is the low cost and low priced provider of Internet services. Our recent pricing changes have reemphasized our position and attracted new customers with longer term contracts in increased volume commitments from new and existing customers.

We have revised our 2008 plan to represent the impact in recent currency changes that impacted our international operations. Our business remains completely focused on Internet based services. We do not sell other forms of telecommunication services. We believe that these services remain isolating from the current economic turmoil that many other businesses are facing.

Our business provides a necessary utility to our customers, and the strength in our receivables and reduction in our DSOs demonstrate that our customers pay for the services that we deliver. We have strengthened our already strong balance sheet, especially when compared to others in our industry. We have eliminated over $106 million of face debt of our original $200 million face value of convertible notes for $0.45 on a dollar, generating over $56 million in gain and reducing our true debt by 53%.

And again, we are extremely encouraged by the sales force initiatives that we implemented earlier in the year. The 13% improvement in sales force productivity growing from 3.4 units per rep per month to 3.8 units per rep per month. We also saw a continued increase in traffic on our network. For the full quarter our traffic grew at 5%, but that somewhat understates the acceleration in traffic growth that we are now seeing.

On a monthly basis we grew 1% -- July over June, again, 1% in August -- August over July, and over 9% in September over August resulting in the full quarter rate of 5%. We continue to see this growth rate accelerate and, in fact, from September to October we have seen traffic in our network grow on a monthly basis by another 10%, where today we are carrying approximately 14 petabytes a day of information on our network.

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

Thank you, ladies and gentlemen. Our question and answer session will be conducted electronically.

(Operator Instructions)

We will now take our first question from Frank Louthan with Raymond James.

Great. Thank you very much. Can you give us an idea of the flow through on the FX impact down to the EBITDA line? And then going forward it's the -- how much do you think of the traffic growth the nice increase you've seen in the last couple of months how much of that has been due to Olympics and election coverage, and do you think that is going to subside at all, or is that traffic related to something else? And if you can give us an update on the announcement from -- with regard to Sprint last week that would be helpful. Thank you.

Sure, Frank. Thanks. It's Dave. I'll take those in reverse order. We have had for some time a settlement free agreement with Sprint. Sprint elected to try to charge Cogent for our interconnection which was not called for in our agreement. That matter has been in litigation in Northern Virginia, and continues to be in front of the courts where we are vigorously protesting that litigation.

Last Thursday, Sprint, at 4 o'clock elected unilaterally to sever the connections between the two networks. Three days later, on Sunday at 4 o'clock Sprint unilaterally elected to reestablish those connections. I think that reestablishment was primarily as a result of the impact that the disconnection from Cogent caused to their customers.

We received literally hundreds of calls from Sprint wireless customers who could not reach content or customers on the Cogent network. While we remain uncertain of Sprint's intentions, we hoped that they'd wait for the final ruling from the court and Cogent remains absolutely committed to abiding by whatever ruling the courts have.

Now let me touch on traffic. Part of traffic growth is associated with seasonality. Not so much specific events such as the Olympics or the election, but rather just the return of people from vacations in the summer. And because Cogent has a large concentration in the educational sector, where we serve over 600 universities and a number of K through 12 systems, we tend to see a pick up in September. So, I think part of our September pick up was as a result of seasonality. I think also part of it was a result of the volume based pricing that we put in place earlier in the summer.

In 2007, we actually saw a negative traffic growth in August. This year, we saw positive growth so you know I think that shows that the seasonality is only partially a result of why we grew traffic. And I think the growth that we continue to see in October is as a result of the value that we deliver to our customers.

The Olympics, the election or any specific application will drive short-term traffic growth, but I think the long-term trend is that more and more content and more and more information that people continue to go after they will go after on the Internet, as opposed to other distribution media. So I think we continue to see long-term traffic growth, and hopefully the continued movement of video to the Internet which will allow us to kind of resume the long-term historical trend rate.

Now let me touch on the FX impact on revenue and EBITDA. The FX impact in our full year guidance was about $3 million. That's on total revenue. In terms of EBITDA flow through there is some disproportionality in that many of our costs are based here in the United States and are therefore not given the beneficial advantage of a lower FX rate whereas revenues are disproportionately impacted.

We also have going on here with our margins all of these other extraordinary events such as the increased litigation costs for Sprint. The disconnection and bad debt associated with our one large customer. The utility cost increases in costs of goods sold and finally we've actually seen some increased taxes from some local jurisdictions.

These are not income taxes or even property taxes, but many of these jurisdictions had previously waived taxes on Internet service providers and now in an environment where they're seeing a tighter budgetary environment they are implementing or rescinding kind of the tax (inaudible) they had in place. While these are not material impacts, all of these things have resulted in the change in EBITDA. And it's very difficult, however, because we were report things on a worldwide basis and many of our costs are spread throughout the world to totally isolate the FX impact on a line by line impact.

Okay, great. And just one follow-up. Give us your thoughts on uses of cash now going forward -- buying debt at $0.45 on the dollar is a pretty good deal. Is that going to be or continue to be your focus, or be more aggressive on the share buyback [with] the stock rate here is? Can you give us an idea of what your thoughts are going to be for that?

Sure. With regard to the debt purchases we have not issued a tender so these have all been private transactions. We have absolutely tried to capitalize on the extreme dislocation in the credit markets and the distress that some of the bondholders were facing. We only have now 94 million of face left.

We will continue opportunistically to talk to those bond holders. We are not initiating a tender. We do have some authorization from the board on continued debt buyback and believe we could go back to the board for even greater authorization but it has to be on a opportunistic basis.

With regard to the share buyback we do have $32 million left in the buyback program and we fully intend to deplete that program over time. Clearly with the stock at this levels we absolutely want to be opportunistic and aggressive. Tad?

Just add a little bit. At the end of the quarter we had $109 million in cash. As we mentioned in October we spent $38 million on the subsequent note repurchase so -- call it 70 million after that point. And we are generating our own cash, so we've also included assumptions on the remaining share purchase program within the plan for the remainder of 2008 and also into 2009.

Great. Thank you.

And we'll now go to James Breen with Thomas Weisel Partners.

Thank you very much. Just a few questions. One, it looks as if sales force churn came up a little bit in the quarter. Can you talk about what's happening there? And with the ARPUs, you touched on it a little bit -- it looks like ARPUs are down in the 70s percent range sequentially in your On-net business.

Can you talk about it sort of a currency environment where if the currency stayed constant quarter to quarter what would that number more look like? And then lastly with respect to guidance for '09, what kind of assumptions are you making with currency either improving it or deteriorating further for next year to get to where your current numbers are? Thanks.

Hey great, Jim. I'll take the first two and then I'll let Tad touch on the FX assumptions going forward. First of all, with sales force churn it actually was -- actually down slightly on a percentage basis quarter-over-quarter. While the absolute number of sales people that we turned over was slightly higher at 47 as opposed to 41, it was because it was off of a larger base.

We continue to see sales force turnover being flat for three of the four quarters or for the two middle quarters. Tends to be very low in the fourth quarter as we generally try not to turnover people in the fourth quarter of the year and then it picks up a little bit in the first quarter.

I do think that some of the training and management initiatives we've put in place have improved things. We've actually seen the percentage of the sales force achieving quota continuing to rise. It went up from kind of the mid-50s to today being 64% of the sales force is at quota so that all, I think, bodes well and there has been no real increase in sales force turnover.

With regard to ARPU, you've really got three things going on simultaneously. First and absolutely the most significant is FX. For our On-net revenues, virtually all the revenues in Europe are On-net because ARPU is just simply the number of customer connections divided into total revenue because that revenue goes down because of FX ARPUs go down.

Secondly, the percentage of customers that are corporate actually increased from 44% to 46% of revenues. For our corporate On-net customers, they tend to take smaller connections than our NetCentric customers so as we see an increase in the percentage of our On-net customer base that is corporate that will pull down ARPU that has not changed the effective price per megabit sold. That number remains about constant at a little over $9. I think our average price is at $9.30 right now.

And then finally, you've got an increased discount associated with contract lengthening. As we mentioned earlier, contracts have lengthened on average by about 4% in the quarter now getting up to just under 12 months. So that has, I think, has positive impact in locking in more revenue, but it does slightly degrade ARPU. But on an ongoing forward basis, if there were no FX changes we would expect ARPU to be effectively constant I think at this point. Those other two impacts are fairly minimal.

Now, let Tad just kind of touch on what he assumed in guidance for '09.

Sure. As we have done in the past since we're taking kind of the current rates in using our forecast and we'll use the same amounts in the forecast for the fourth quarter and obviously the year of '08 and then in '09. So I've not assumed further deterioration or increase depending upon which side of the pond you're on.

Specifically -- and we've used the rates as they are today, so as a reminder Europe is about 22% of our business and it's got $1.30. So that's what we have in our assumption with no change in that rate going forward. To the extent it changes it's going to impact those estimates. Canada is about $0.90 on the dollar and that's about 8% of our business. So we're using the current rates with no forecasted change up or down in the currency rate for the remainder of this year and next year.

Great. Thank you very much.

And we'll now go to Michael Roland with City Investment Research.

Hi. Good morning. Just a couple of quick questions. First of all, as you look at what's going on in the competitive environment? Where do you see in terms of pricing maybe on a 12 or 24-month view of how tough it could get, or do you think we're one year of bottom on what pricing is? And then the second question I had is just if you could just -- I think you touched on this briefly earlier. Would you talk a little bit more about the assumptions underlying On-net and Off-net in the '09 guidance? That would be fantastic. Thanks.

Sure, Mike. I'll take the first one and let Tad take the assumptions in the guidance question. With regard to pricing we have two very different customer bases and addressable markets and we are seeing different dynamics in each of those two markets.

For these 46% of our revenues that are derived from corporate customers we have seen a stable pricing environment and I think that environment based on everything we can see will remain stable for the next 12 to 24 months. Remember, our corporate customers buy not on a per megabit basis, but rather on a per connection basis. What Cogent is offering is a completely alternative connection at a price that is equivalent to what will be provided either by the ILEC or a competitive carrier using ILEC facilities that is much less capacity and generally less robust.

Because of some of the regulatory changes and the lack of competitive overbuilds on true facility based networks, we've seen vast placing environments be stable. And I would expect with the increased costs of capital you will continue to see few overbuilds to compete on a facilities base basis and that environment to be stable.

I think the second part of our business which is a larger portion of our business, it is 54% of our revenue today which is our NetCentric or service provider business. That is where we sell in a carrier neutral data center. That business has experienced substantial price reductions on average over the years, and it's generally where investors focus when they talk about price per megabit.

What we are selling is a fungible commodity at the same location as typically a handful of other competitors. In that environment, we have seen the competitive landscape change quite a bit. We've seen a lot of the legacy carriers elect not to compete for new business and just totally withdraw from the market.

We do see typically a couple of active competitors in that environment who have lowered their prices to get closer to Cogent, but we remain absolutely committed to being the price leader. In that market, we offer both volume discounts and term discounts that have taken our $10 price down to as low as $4 per megabit. In that environment we're probably today selling at an average of about $6.5 a megabit based on contract value and term.

We will continue to be the price leader. I think we have not seen are competitors able or willing to get down to our price points. We remain committed to beating their price points. And I think based on the reported financial performance of a couple of companies we actively compete with, I think their ability to compete is greatly restricted and we think we're going to see probably less competition going forward as people focus on other parts of their business that are much higher margin.

Let me turn it over to Tad now, and let him touch on these assumptions -- questions.

Sure. With respect to how we build our revenue forecast we start with a trailing six-month average and then often we include since we're now -- October is behind us, we have that data. We incorporate that data into that so often it's trailing six-months including October in most cases.

We take our historical mix. We take our historical pricing. We take our historical churn rates and then we take our assumptions on rep churn and hires. For example, our disclosed increases to get to 250 reps by the end of '08. Using those six-month averages and not blindly using them we look for any outliers, because you will get certain events that create an outlier in any one month and we'll discard that, and that's what we use to build our plan going forward.

So it's based upon the past. We do incorporate percentage changes. So if we've -- rep productivity has been trending down we would use that and incorporate that in the forecast going forward. If ARPU has been trending down that will be incorporated in the forecast going forward. If mix has slightly changed in terms of new sales that will be incorporated into going forward, and the churn rate.

So, if you will, it's kind of built as a revenue go forward with your starting run rate, your forecasted ads by product site and your forecasted cancellations by product type, getting you to the end revenue for the following month and that's where we start from. And we have, as we mentioned, used in the past and have used in our current plan the current foreign exchange rates. So, frankly, that has had the most material impact on what we forecasted for this quarter versus what we experienced.

Thanks.

And we'll now go to Mike McCormick with JPMorgan.

Hey, thanks, guys. Maybe just quickly on the economy. Anything on the corporate customer side as far as either pulling back on scale or scope or canceling products all together? And then secondly, looking into the fourth quarter guidance it sounds like you're saying pricing is pretty stable at least on the corporate side. Traffic seems to be improving pretty nicely here, but we're looking for margin compression versus prior guidance. I'm assuming that's FX, but if you can just clarify that would be great.

Yes. Hey, sure, Mike. It's Dave. First of all, it is absolutely clear that there is a lot of economic turmoil and the country is in a serious recession if not something approaching a depression. While there may have been some ambiguity around that a quarter ago, I think the events of the past two months have absolutely convinced everyone that we're headed into a dramatically slowing environment.

We, however, have been somewhat immune from that. The service we sell is a utility meaning I think virtually everyone who buys our product needs it to run their business whether they be a corporate customer or a NetCentric provider. We are the lowest cost provider and I think customers are more value conscious, so if they are looking for Internet connectivity they will generally buy the best value and therefore switch to Cogent.

And I think we're also seeing an acceleration in the long-term trend of shifting traffic away from other higher costs types of networks to the Internet where you're getting product displacements so things are moving away from TDM frame relay, ATM and private LAN networks all to the Internet. All of these things are contributing to a healthy traffic growth and a healthy demand environment.

If you look at our corporate footprint -- you can go to our website and see the buildings we're in. The vacancy rates have not really ticked up in those buildings. Maybe rents are not rising as quickly, but because we focus on class A skyscrapers we tend to have a little bit more isolation, and we haven't seen a huge uptick in empty floor space in those buildings which would make our selling more difficult to the corporate customers. In fact, our corporate productivity is as good as ever.

On the NetCentric side, if unemployment goes up people will have more time on their hands probably spend more time on the Internet using Internet more to find a job or download entertainment, looking for lower cost ways to get content and information. All of these things kind of bode well for traffic.

Anecdotally, in meeting with our sales organization -- we had a meeting about four weeks ago where we brought all of our directors together from around the world. And as we brought those people together there were directors who were meeting their goals, some exceeding it -- some quite honestly weren't meeting their goals.

So I started quizzing people starting with the people that were doing the worst saying, have you see the lightening of sales cycle, or is the reason why you're not hitting your targets the economy? And not a single one of the directors said that they had seen the economy as a problem. There may be other issues, but the economy is not a problem.

Let me switch now to the FX and our guidance. Clearly, the top line reduction is entirely attributable to FX. For the EBITDA and margin compression, FX is part of the story, but we also have seen this increased litigation costs which we have to be conscious of and budget for. We have some seasonal audit type things that are factored into the budget, but overall, you should continue to see going into 2009 a resumption of the margin expansion that we've demonstrated.

If you took the past year for example, we've guided to about 100 basis points quarter-over-quarter both gross margin and EBITDA margin expansion. That's in fact exactly what we did, yet it's lumpy. Some quarters it will go up 200% and another quarter it may go down a little bit, but if you look over a long-term you are able to demonstrate that type of operating leverage in business and we expect that to continue going forward.

Great. Thanks, Dave.

Thanks, Mike.

And we'll now go to Jonathan Atkin with RBC Capital.

Yes, good morning. I wondered if you could clarify the situation about the large customer that you mentioned that left. And then the accelerated entry into new markets and the associated OpEx burn --I assume that also is one factor behind the compressed EBITDA margins. Is that basically long haul route construction or is that accelerated pace of On-net buildings?

And then my final question relates to the traffic growth trends that you referenced, and just kind of taking a step back how [usage] sensitive are your revenues compared to prior quarters? It strikes me that in the wake of the weak pricing particularly the volume based discounts that customers are taking advantage of maybe your business is less usage sensitive than it use to be?

Okay. Good, Jon. Those are all very good questions. Let me start with what has been out in the public domain. The customer that we lost was a company called Alpha Red. They were a large hosting operator based in Texas. They ran into some financial difficulty. We had kept that customer on a fairly short leash, although it can never be short enough in the sense that we did end up taking a charge for the bad debt that we had to write-off from that customer and that did impact our margin in third quarter.

We had a little bit of a ramp up as we then basically focused the sales force on going to their customer base. Now we were actually very fortunate, and we were able to sign up most of their very large customers and get actually more of a percentage of their business than we previously had. So in terms of the net recurring revenue that we booked out of the customer base that was abrogated through that company actually went up so I think that's good.

Our customer concentration actually continues to decline. If you look at our top customers, our top five customers today are about 8.5% of revenues. If you look at our to 25 customers they're about 15% of revenues. We have a great deal of customer diversity and particularly it's a percentage of our business is more corporate we see an even greater amount of diversity.

Now with regard to new markets, the costs that we bear in those markets -- the building costs are pretty much equivalent everywhere. We have added a few more markets than we originally planned at the beginning of the year. That's part of the reason why we're going to add more buildings than we initially planned and a few more routes. That will all affect capital that those are really operating costs.

What we really pick up in operating costs are two things. One, the field tax and field support people necessary to support those remote markets whereas if we add additional buildings in existing markets we don't need to add more field people because from within driving distance, but as we extend into places like Kiev or Bucharest it just becomes too logistically challenging for a guy to support those say out of Germany or Austria so you just need to have more field people. And, we now service customers in a total of 23 countries.

Each of these countries has their own kind of legal requirements in terms of entities and minimum presences. There's just some kind of base case number that we have to spend in each country and we've probably added more countries than we originally planned. I think that's had a little bit of an impact. I would expect to see those startup costs going down next year. That's part of the reason why we'll see some improvement in margin.

And then finally, let me touch on your usage sensitivity question. Historically, most of our customers buy fixed commits. In fact, all of our corporate customers buy fixed commits. Our NetCentric customers however typically will have a fixed commit and then will use more than that fix commit paying a slight penalty. Historically, about 2% of our revenues came from those burst fees.

We saw that drop last quarter to 1.5% and it's dropped even further. Part of the volume based pricing initiative has actually incented people to take greater commitments. So while it's good for aggregate revenue the amount of bursts or overage revenue has substantially decreased where today it's probably less than 1% of revenues and we expect that trend to continue. By incenting people to take these fixed commitments we've, I think, isolated ourselves from short-term traffic fluctuations.

Thank you very much.

Hey, thanks, Jon.

And we'll now go to Nick Netchvolodoff with Barclays Capital.

Hi, Dave. How are you? I just wanted to follow-up or get some more detail on 2009 expansion plans kind of where you're thinking of going, US versus abroad? Some detail there.

Sure, Nick. As we look at our footprints, Cogent is becoming very ubiquitous both in Europe and in North America. I would say most of the cities that we are adding are secondary cities. Some of those are fill-ins along existing routes. Some are on new routes. I think there are some additional routes in North America where we own fiber that we have not led that we will be adding into some smaller markets.

We also will be adding in Europe primarily not so much smaller markets in existing footprint, but probably some new countries there's still a few markets that we don't have probably the ubiquity that we'd like and there are one or two markets where we don't have the redundancy that we would like to add primarily in Eastern Europe so we will probably close some reins.

Our European expansion will be almost entirely NetCentrics of connecting to data centers. We have today I think a grand total of two corporate buildings in Europe. We are considering adding some additional corporate buildings in London in the Canary Wharf area for -- to try to test that market, but that's a limited market for us. Whereas in North America, we are pretty ubiquitously penetrated in the corporate buildings. We have added some smaller markets like Des Moines, Minneapolis, Omaha and we are continuing to add corporate buildings in those markets.

So for example in Minneapolis, I think today we have a half a dozen buildings On-net and there's probably about 20 targeted buildings. So we will substantially increase the building footprint in a place like Minneapolis.

Right.

And then in North America finally, we are reactive to the growth plans of the data center operators. A number of data center operators Equinex, Switch and Data, Digital Realty Trust, DuPont Fabros have all announced a number of expansion plans and we typically want to serve 100% of their footprint, so if they expand we'll expand.

Okay, great. Thanks.

Thanks, Nick.

And we'll now go to Ken Lee with Jefferies.

Hey, guys. Most of my questions have been asked and answered, but I just wanted to get your thoughts on debt repurchase versus share repurchase. It seems that your -- I mean your convert is 1% covert. It strikes at 49. It seems like it's almost free money at this point. Why the repurchase now versus buying back more shares?

Sure, Ken. It's Dave. You are correct. It's close to free money, and we're very fortunate to be able to time the markets and acquire that capital at a very low cost. That instrument however is puttable in about five in a half years, so we do think about that. We think about kind of the yield to worse just on the debt portion. While the convert is very far out of the money we hoped that within that five and a half year period that's not always the case although it's not really the equity component of the convert that driving our decision. It's really the ability to buy at a very good yield to worse which is approximately 20% at the prices that we've been buying.

The convert is opportunistic. We are generally tendering for it. We are being contacted by convert holders who have their own liquidity situations and we base those purchase decisions on a case-by-case basis opportunistically create value and strengthen the balance sheet. Basically for each dollar we spend we create about $1.20 of enterprise of value in purchasing the converts so that is very accretive, and as Tad pointed out we are generating free cash in addition to having excess liquidity on our balance sheet.

While it's free money, the opportunity costs of that capital is high if we can put it to work. As we think about the share repurchase program, we view that as a measured program that is designed to return value to shareholders. We've been consistent. We've bought back over 12.5% of the Company in the past 18 months. We absolutely plan to be consistent and have a measure pace to continue to do buybacks.

Clearly, we'd like to be buying at today's price as opposed to where we bought in the past, but the purpose of the buyback program is to return value to shareholders in an orderly way. And, I can't profess to be smart enough to time the market. So we will consistently buy back stock. We will opportunistically buy back the debt. And quite frankly there's more of an opportunity right now in the stocks and the debts simply because there's not that much debt left out there.

Great. Thank you.

Hey, thanks, Ken.

And we'll take our next question from Robert Dezego with SunTrust Robinson.

Just had two quick questions. One, you spoke of the increase in the contract value of the customers that were taking advantage of the price point. Could you talk about the change you're seeing in the monthly spend of these customers versus the actual contract value?

Sure, Rob. It's Dave. In the vast majority of cases we've seen the monthly spend go up. There have been some instances where the customer has elected to increase contract value solely by extending churn. However, there is no commission paid in that situation so the way in which our commission structure is based is a rep who would be processing that is only paid on the increase in monthly recurring revenue from an existing customer and obviously the entire revenue from a brand new customer.

For our sales force on average 50% of their compensation is fixed or their based. 50% is variable so they have a strong incentive to get a greater amount of recurring revenue on a monthly basis from the customer. So the majority -- the vast majority has resulted in both a lengthening of the contract -- it's gone up on average about 4% as I'd said earlier, as well as an increase in monthly spending. But there have been a limited number of cases where the spend has stayed constant or even in few cases gone down that has been impacted -- was impact on our ARPU and is reflected in number.

But we also see that the majority of existing customers taking advantage of this program has subsided. And most of the people taking advantage of the term and volume discounts now are brand new customers, not existing customers as was the case when we initially rolled the program out in June.

Okay, great. So your existing customers are not calling in and looking for this new plan, your're saying it's mostly for new customers?

That's correct.

Okay. And my second question is you'd previously indicated that you expected corporate revenue to continue to decline as a percentage of revenue longer term, but the last few quarters we've seen obviously a change in that as it's risen. Could you maybe just let us know what your expectations are as far as revenue growth or percentage of revenue for your '08 and '09 guidance for corporate revenue?

Sure. Long-term we believe that our two addressable markets will determine the split in revenue. Our NetCentric addressable market is larger than our corporate market simply because of the discipline we apply to the different types of buildings that we will go after. What we have seen however is a slowdown in the rate of traffic growth on the Internet.

To counter that and to accelerate our gain in market share, which I think you've seen in the numbers, we've demonstrated particularly in the September and October monthly sequential traffic growths we're actually countering those trends. But in an environment where the Internet is not growing as rapidly we would see corporate pick up as a percentage. Long-term, I think NetCentric will still increase as a percentage from the 54% that it's at today to probably close to 65% of our revenues. Whether that occurs in '09 or not is unclear.

In the short-term it's basically a result of existing customers not growing as fast, putting the price volume breaks in place and gaining share. And if we look at penetration on the corporate side we have today about 7.5 customers up from about 7.2 per building so it continues to increase, but that's out of an addressable market of 51 customers per building. So we have a long way to go on the corporate side.

On the NetCentric side, we're at about 15 customers per data center out of an addressable market of about 200. So again, I think long-term we have a greater addressable market that's unserved on the NetCentric side of the business. Short-term, because of a slower growth environment we may see corporate pick up for the next quarter or two, but I think long-term NetCentric will be dominate. And our guidance -- I'll let Tad touch on that, but I don't think there's a whole lot of variance baked into our guidance.

We actually don't build it by corporate versus NetCentric. It's kind of happens by a function as what's happened in the past and kind of baked into the mix that we focus on as On-net and Off-net, as opposed to corporate person NetCentric.

And then the impact of pricing differences between them would be dealt with by including the historical trend on ARPU changes, as I said, over the trailing last six months. So that's kind of how we look at it even though we do evaluate results based upon mix of corporate and NetCentric. In terms of forecasting future revenue we do it by product class on that (inaudible).

Okay. And now just two quick housekeeping follow-ups. Did you guys -- I think talked up the number of buildings you're going to do for '08. Could you talk about the number of buildings you're expecting to add in 2009?

Sure. At least 100. 100 will be a -- it may be slightly better than that.

Okay. And did you guys give a total traffic carried on your network for the quarter?

We did. We said as of today we're carrying about 14 petabytes a day, and traffic for the quarter increased from the previous quarter by 5%.

5%. Okay. Thank you very much, guys. Best of luck.

Thanks, Rob.

(Operator Instructions)

We'll now go to David Dixon with FBR Capital Markets.

Thanks, gentlemen. Just a question if I could -- just looking at the customer turning there in the quarter I wanted just to make sure I was interpreting the recurring revenue lists correctly there. Dave, I think you'd mentioned previously that you had to kept this company on a tight leash reflecting some credit concerns so [admirable] there.

I think you mentioned it was on a week to week basis. So should I be interpreting that 600,000 impact as a weekly revenue commit, and therefore calculating that as around 2.5 million per quarter? That was question number one. And then just confirming that you've achieved a rate recurring revenue lift off that base, and whether that occurred in the same facility -- is it certainly a good result because that business as I understand it was split evenly between yourselves and Level 3.

So that being the case, just wondered if you could tell us a little bit about how you achieve a greater mix of that customer business from obviously a very -- within a very challenging transition there. Thanks very much.

Sure. Good to hear from you, Dave. First of all, with regard to the customer that particular customer has been a customer or had been a customer of Cogent for about four years and had grown rapidly with us. They had experienced consistent payment issues and -- initially were like all other customers on a monthly payment plan. And as they fell behind we realized that to keep them on a short leash we wanted to put them on a weekly payment plan.

That did not mean, however, that their balance was zero. They were carrying a balance forward. So the 600,000 that we lost represented several weeks or almost a month of their revenue that was not their weekly run rate. I wish we were that good in collecting and people were basically current by the week. So while they had a weekly payment plan to Cogent that revenue was multiple weeks. Probably close to a month's worth of recurring revenue even though they had to make weekly payments to make it.

Now with regard to picking up business; the business that we picked up was both in the two facilities in Texas that that operator controlled, as well as a third facility in Texas where they were subleasing space. So the traffic that we ended up winning from their customers was in all three of those locations, as well as those customers chose to diversify adding additional capacity with Cogent both in Europe, on the East Coast and the West Coast.

So the traffic wins that we got from those customers bases are now spread over a half a dozen locations, as opposed to being concentrated in three. And as a percentage of the dollar spend by those customers, because of the volume based pricing structure that we have in place they were highly incented to give us a greater percentage of their business. So while I won't say that we have 100% in every one of those customers cases, I think we have a much greater percentage than we had previously with the one customer.

I'll just add one thing that -- just the impact on the fourth quarter. So while we did lose this customer and we had to write-off their outstanding balance, and we were able to replace the revenue actually with greater revenue in the long-term and improve concentration replacing one customer on a payment plan with a series of customers who actually paid in advance. There's a ramp up for the month of October in replacing that revenue as those connections were installed. So as you're here today the run rate is actually higher, but for the month of October accumulative revenue because of the install period is actually less.

And that was baked into our guidance that that ramp for those series of customers.

Right. Okay, that's helpful. And then just -- on the 600,000 being a monthly revenue amount at $4 a meg, I calculate that as being around 50 gigs. Is that equivalent to the amount of traffic that we would have had previously as a bytes reference point?

Well, a couple of things, Dave. First of all, it was slightly less than a month, so it's slightly more than that in terms of traffic. And they had multiple ports, so they were not necessarily getting the $4 price point. In other words, they spread their traffic over a larger number of ports. So to get that price point you have to have a full 10-gig port and concentrate your traffic on those ports.

Right. So -- and then, despite the reduction there we've actually seen -- I think you mentioned a 10% increase in October which would have incorporated the effect of the turndown of that customer.

Not only the turndown, but also of counts -- the ramp that Tad talked about. So if, for example -- if that particular customer that was distressed went away and all of their customers instantaneously turned back up at the exact same moment the rate of traffic growth in October would have actually been greater than the 10% that we reported.

Right. Okay, that's great. And then just one quick follow-up, Dave. Any sense of a continued cost estimate to Sprint, especially with this dispute that's underway? How we should be thinking about that going forward, whether you're accounting for any contingency there? Or, [if] could you give us some assistance with what we would see potentially if you do see an adverse outcome there from the courts?

Sure. So we are not accruing -- although we have listed in our disputes the Sprint litigation the actual principal amount. We have no contract with them that calls for us to pay money, so we find it very difficult that they could present us a bill since there's no contract for services with the dollar amounts associated with it. So I don't know under what theory they are litigating. That is separate and somewhat distinct from the interconnection litigation, which is basically a series of requirements that we met that they are alleging we did not meet.

We had reserved money for and have budgeted monies for the legal expense associated with that litigation, and that's on the order of several hundred thousand dollars a quarter. Obviously, with any kind of litigation we always hope it settles prior to being fully adjudicated, but the only thing that we think we're going to be spending, unfortunately, is money with lawyers. And, it is unlikely that we will be able to recover those legal expenses.

Right. Okay. Does seem a little odd. I know you probably can't comment on too much, but it does seem a little odd that related to -- obviously what had been occurring over a year ago and for the claims to be made now just seems -- a significant amount of time has elapsed from what was the trial period that they refer to.

Right. We fully paid for the trial period. There's no dispute about the payments that were made during the trial. The dispute is for the periods after, but you are correct that it is a bit unusual for someone to wait that long before raising the issue, which I think also focuses on the weakness of their case.

All right. Well, thanks very much, gentlemen.

Hey, thanks a lot, Dave.

And this concludes our question and answer session. We have no other questions at this time. I would like to turn the conference back over to Mr. Dave Schaeffer for any additional or closing remarks.

Well, thank you very much. Thank you, everyone, for listening. I know these calls sometimes go a long time, but we do try to be complete in answering questions. I'd just like to take this opportunity to thank the entire Cogent team. I mean, this is a difficult economic environment. I think the team has performed exemplatory.

We're doing I think a very good job. Our business continues to grow. We are focused on the right markets, the right customers, and most importantly the right products at the right price point. And I think you'll see continued strong performance from the entire Cogent teams. So, I'd like to thank the roughly 500 members of that team that are responsible for us being able to deliver these results to you. Thanks a lot, everyone.

This concludes today's conference. We thank you for your participation. Have a nice day.