Description

Thesis
XO Communications (XOCM) owns a network which is a low cost supplier in a commodity business: providing communications services to businesses. XOCM may be purchased for $700 million, less than 20% of the replacement cost of its state-of-the-art network. In February, the company announced that it would acquire most of Allegiance Telecom’s (ALGXQ) assets for $600 million. The combined companies have a current enterprise value of $1.3 billion. On a pro forma basis, the two companies would have had about $100 million of EBITDA in 2003, arguably the worst year in the telecom industry history and a likely tough in the industry’s cycle. Carl Icahn currently holds 60% of its shares and 95% of its debt.


Background
XO Communications (XOCM) is a facilities based competitive local exchange carrier (CLEC) with a new network which is 90% depreciated, a new CEO and a pristine balance sheet. XOCM enjoys a cost advantage verses its peers from lower depreciation costs as well as new, better equipment. Unlike long-haul communications companies e.g., Global Double Crossing and Level 3, XOCM owns networks which connect directly to thousands of customers in every major metropolitan region in the US – the company owns the last mile. In addition, for customers who are not close enough to connect fiber directly to XOCM’s network, the company owns licenses to wireless spectrum which may be used to connect to their customers. The property, plant and equipment owned by XOCM cost more than $4 billion dollars to construct; much of the cost was financed through debt. XOCM filed Chapter 11 on 7/22/02 when the debt burden became too large and the capital markets were closed.

The company emerged from Chapter 11 in January 2003 and recently completed a two stage rights offering in January 2004. As a result, XOCM’s debt is now extinguished along with the pesky interest expense. As of this writing XOCM has cash net of debt of approximately $181 million.

Chapter 11 has also relieved the company of most of its ongoing depreciation expense: the network which cost over $4 billion is on the company’s balance sheet for $473 million. As a result, XOCM’s D+A expense comes to about 2.6% of the original cost of building its network. For comparison, D+A costs for other CLECS runs 10% - 15% of the cost of their networks; D+A costs for RBOCs runs 7% - 8% of the original cost of their networks. The difference is even more stark when one considers that the oldest portions XOCM’s network are 6 years old, while the RBOCs have large amounts of decades-old equipment.



ALGXQ acquisition
On February 13, XOCM won a bidding process to purchase substantially all of the assets of ALGXQ out of bankruptcy for about $600 million. XOCM’s winning bid consisted of $311 million cash plus 45.38 million shares currently worth about $292 million. The company used equity to avoid levering the balance sheet. In light of what we have learned about highly leveraged telecom companies over the last three years, this seems prudent. After the acquisition closes, XOCM will have debt in excess of cash of approximately $155 million

ALGXQ has a collection of network assets which cover substantially the same footprint as XOCM. As a result, XOCM is removing a competitor as well as reducing costs by eliminating duplicate functions. XOCM paid a bit less than book value for the ALGXQ assets; in this case, book value represented about half the gross cost of assembling the assets. The ALGXQ assets acquired produced about $521 million of revenue and probably broke even on an EBITDA basis during 12 months ending 12/31/03.


XOCM is breaking-even on an EBITDA basis, but will soon make money
Like many other telecom companies, XOCM appears to be burning cash. Upon closer inspection, note that XOCM actually turned EBITDA positive during the fourth quarter of 2002 and continued to be EBITDA positive for the first half of 2003. After digging a bit more, two points become clear:

1. XOCM is currently breaking–even on an EBITDA basis.
2. On a proforman basis, XOCM + Allegiance would have generated about $100 million of EBITDA during 2003.
3. XOCM’s controlling share holder, Carl Icahn, stood to benefit if the first stage of the two stage rights offering was not fully subscribed i.e., poor 2003 results were helpful to him.

Like many companies which emerge from Chapter 11, XOCM issued rights to former shareholders so that they could purchase shares in the recapitalized company. In this instance, however, the rights offering was conducted in two stages. During the first stage, former shareholders were issued one right to purchase a new share of XOCM at $5 for every 116 shares of old XOCM they held. Up to 40 million shares of new XOCM could be purchased during stage 1. Rights holders were permitted to bid for more shares than they were entitled to, so that if the stage 1 rights offering were undersubscribed, they would be allocated the additional shares. On the other hand, their rights were not transferable. Shares which were not bid for during stage 1 became available for a second stage of bidding to be done by senior credit holders, which in this case, was Carl Icahn. As a result, lower bidding during stage 1 would result in more shares becoming available for Mr. Icahn during stage 2. This had a number of positive aspects for Mr. Icahn, including:

1.Icahn’s equity position stood to be diluted an additional 7.6% if stage 1 were oversubscribed.
2.Having shares available for stage 2 permitted Icahn to swap his XOCM debt for equity at $5 per share.
3.If Mr. Icahn were able to continue to own more than 80% of XOCM’s shares, he could continue to use XOCM’s net operating losses to shelter income generated by other businesses he controls.

The point is, until the rights offering was completed and the new shares distributed (which happened on 1/23/04), Icahn’s interests were not aligned with those of other shareholders. This situation has been remedied and we expect to begin hearing good news from XOCM. For instance, on February 4, XOCM management announced that they were applying to be listed on the NASDAQ.

Economics
XOCM’s earnings potential is directly linked to management’s ability to either increase revenue without increasing SG&A or cut SG&A without losing revenue. The Allegiance acquisition facilitates cost reductions and adds $525 million of revenue. XOCM’s SG&A is running 55% - 60% of revenue; the CLEC industry average SG&A is 30% of revenue. Our base case pro forma income statement showing the combination of XOCM and ALGXQ as if it happened on 12/31/02 has SG&A of 56% of revenue; under this scenario, the company would have generated $100 million of EBITDA (combined enterprise value of $1.3 billion, for an EBITDA multiple of 13).

As one would expect, small improvements in SG&A have a significant improvement in earnings performance. For instance, if management were to reduce SG&A to 49% of revenue, XOCM would currently be trading at 5.8x EBITDA; if management were able to reduce XOCM’s SG&A to 30% of revenue, XOCM would currently be trading at 2.4x EBITDA. Of course, this cuts both ways, i.e., small increases in SG&A have will have a profound negative impact on financial performance. Also, it is one thing to model cost cutting in a spreadsheet, and quite another to successfully implement a cost cutting initiative. Below we review SG&A costs at comparable companies, outline the components of XOCM’s SG&A costs and discuss some of the ways in which XOCM’s management is likely to achieve cost savings.

Cost Cutting: SG&A
A key assumption of our analysis is that XOCM’s management i.e., Carl Icahn, will bring XOCM’s SG&A costs, currently about 60% of revenue, inline with its competitors. Comparable CLECs average SG&A costs of 31% of revenue; RBOCs average SG&A costs of 27% of revenue. While the company has not disclosed its specific plans for cutting SG&A, two potential ways that SG&A costs are likely to go down:

1) During 2003, XOCM had a great deal of customer churn. One insider we spoke with indicated that the company lost $250 million in revenue through customer attrition (on a little over $1.25 billion of revenue). It took their sales force of 1,100 people working full time to replace these customers. Apparently, a large percent of this attrition (maybe half) was due to non-facilties based CLEC customers (resellers) and dot bombs going out of business. It is likely that the rate of customers going out of business will decrease going forward, if for no other reason, there are not many non-facilities based CLEC customers and dot bombs left.


2) XOCM currently has 1,100 sales people; Allegiance currently has 700 sales people. The two companies service the same geographic areas. XOCM is likely to keep the 800 highest producing sales people from the pool of 1800 sales people between XOCM and ALGLX. On paper at least, XOCM will simultaneously upgrade their sales force and cut $25 million from SG&A.

We presume that management is reviewing the rest of the SG&A categories for ways of reducing costs.


Cost Cutting: Cost of Services
XOCM Chief Executive Grivner indicated that the ALGXQ acquisition will help XOCM cut network costs i.e., cost of sales, by $60 million. A large chunk of this reduction is likely to come from reduced co-location costs. These are costs that both XOCM and ALGXQ pay to rent space at the central offices of ILECs. Since there is a very high overlap between the networks of the two companies, in many cases, the equipment can be consolidated into one space, resulting in lower costs. (Please see exhibit in the appendix showing market overlap). Additional costs savings may be achieved by cutting duplicate service personnel and the associated overhead.

Valuation today: $5.21
At $5.21, XOCM has a net enterprise value of $704 million (170 million shares fully diluted minus cash of $520 million plus debt of $340 million. Assuming the Allegiance acquisition were completed as of this writing, the combined companies would have a net enterprise value of $1.3 billion (215.3 million shares minus cash of $184 million plus debt of $340 million).

Valuation potential: $12 - $20
So what is XOCM worth? Since the company is not currently making money, earnings multiples don’t make much sense. On an EBITDA basis, the company, post ALGXQ acquisition, is probably trading about something like 13x ’03 EBITDA, not especially interesting from a value perspective. On the other hand, as discussed above, if management succeeds in bringing SG&A costs in line with industry averages e.g., SG&A 30% of revenue implies that the company is trading at 2.4x EBITDA, obviously, a very interesting situation.

Other methods of estimating XOCM’s valuation rely on asset metrics. For instance, consider the price one would pay for XOCM’s assets if they were owned by one of its competitors i.e. enterprise value / gross property plant and equipment. This method results in implied share values for XOCM of $23 to $26 when compared to other CLECS and $$23 when compared to the RBOCs. On a value to revenue basis, the current price one pays for other CLECs would imply that XOCM shares would be worth $12 to $13, the current price one pays for RBOCs would imply that XOCM shares would be worth $19 - $20. There are three key assumptions underlying these methods:

1.Mr, Icahn can figure out how to make as good a use of these assets as his competitors do of their assets.

2.Mr. Icahn, as the control shareholder, will treat minority shareholders fairly.

3.The market is fairly valuing the competitors assets e.g., the competitors will not fall to XOCM’s valuation.


GLBC settlement
An investment in XOCM is de facto an investment in Carl Icahn. Mr. Icahn effectively bought XOCM out of bankruptcy and, in the process, is pursing other telecom asset plays. His handy work has already paid a dividend: XOCM made a $158.5 million investment to acquire 24% of the pre-petition senior debt of Global Double Crossing (GLBC); while Mr. Icahn failed to buy GBLC (a good thing) the investment appreciated to $192 million, which was paid to XOCM on 12/15/03 in the form of $164.8 million cash plus 824,000 shares of GBLC.

Market Dynamics
During the first stage of the rights offering, 32.5 million shares of XOCM were sold for $5 per share and delivered to the purchasers on or about January 16, 2004. A look at the trading volumes of XOCM shares prior, during and after the rights offering indicates that there is likely a number of people who exercised their rights to buy shares at $5 and began selling their shares at a profit i.e., when XOCM was trading north of $6. Note that daily trading volume has increased by 3.5x since the closing of the rights offering (all shares from both rights offerings were distributed between 1/7/04 and 1/23/04). While part of this is likely due to a slight increase in attention to XOCM, a portion of this volume is likely to be the result of participants in the first stage rights offering flipping their shares for a profit. After all, these are people who were previous shareholders and creditors to XOCM and were burned, so it is understandable why they may be motivated to take profits and exit their positions. To the extent this happens when the shares trade for more than $6, it will put downward pressure on the share price. I t will take about 77 trading days (assuming all excess shares being traded are rights recipients selling their shares) to work through the 32.5 million shares distributed to stage 1 rights holders. This means that the share price could very well remain under pressure though the middle of May: ample time to build a position.

Catalyst

1.Issuance of updated financial statements showing $181 million of cash, net of debt.
2.Greater transparency now that the rights issue is complete; all shareholders’ interests are now aligned.
3.EBITDA positive going forward.
4.Listing on the NASDAQ (application process announced 2/4/04).
5.Succesful completion of the Allegiance acquisition, Q3 2004.
6.Cost cutting which will result in positive GAAP earnings within 18 months.