XFERA:WHAT’S A FAIR PRICE?. THE CASE OF SPAIN'S 4TH UMTS LICENSE
By: Ignacio de la Torre
Francisco J. López Lubián
Finance Department
Instituto de Empresa. Madrid
Address: C/ Castellón de la Plana, 8
28006 Madrid
Spain
e-mail:
web site:
XFERA:WHAT’S A FAIR PRICE?. THE CASE OF SPAIN'S 4TH UMTS LICENSE
It is often the case that the common value is not universally known, but instead each participant has an internal estimate of the item’s value. The most likely scenario is that there is a range of such estimates. Barring a systematic bias in the estimates, the average of the estimates should be closer to the true common value of the item than any of the individuals’ estimates. If each participant bid is close to their estimate, the winner is likely to have over-bid and suffer what is known as the winners’ curse. In other words, the winner is the bidder who has made the largest possible error in his estimate of the common value[1].
What went wrong?
On December, 2002, John Silverman, VS' telecom equity research analyst was studying the whole process of bidding and allocation of European UMTS licenses, which had taken place during the year 2000. Silverman’s boss, James Mackenzie, had asked John to write a research report on valuation pitfalls that occurred in said process, focusing on a specific example. John centered his attention on the case of Spain, where the bidding process used the beauty contest format rather than that of auction[2]. A beauty contest allowed for a much lower license fee[3], resulting in a higher valuation upside for a UMTS license shareholder. Xfera obtained one of these licenses in Spain. The company’s rich shareholder mix (including private equity firms, savings banks, telecom operators, and companies pursuing a diversification strategy), and the fact that all these institutions had clearly got their numbers wrong when they had valued this project aroused Silverman’s interest. If the cheapest UMTS license in Europe was a clear valuation anomaly, what had happened in Europe to produce a transference of value[4] from the private sector to public hands of close to €170 billion., one of the largest transferences of wealth in history?
Digging deeper into the question, Silverman found that this error of calculation was incurred not only by the referred shareholders but also by the commercial bankers which had lent the money to finance the ventures, by the strategic consultants which had advised them on the bidding processes, by the Governments (almost producing a number of bankruptcies in the telecoms industry), and by the investment banks, which had mistakenly valued the businesses from the perspective of both the equity and debt investors.
Silverman thought again about the report he was about to write, wondering what might have caused the huge differentials between expectations, realities and the assumptions that never materialized. Were those assumptions realistic at any point in time? Was risk well assessed? Were valuation mistakes the result of psychological factors? Was the credit crunch that impacted the economy during 2001-2002 the driving force behind the differences in valuation? Again, John challenged himself to ask the key question: What went wrong?
Industry background
Current cellular phones use either first generation (1G) analogue technology or second-generation (2G) digital technology. The 2G standards vary. For example, Europe uses the TDMA-based Global System for Mobile Communications (GSM) standard and the US uses both the IS 136 TDMA standard and the IS-95 CDMA standard. 3G cellular standards will permit the same cellular phone to be used in any country without modification. 1G and 2G technologies, on the other hand, were developed hastily resulting in a lack of standardization that led to incompatible cellular networks. Significant standardization efforts are underway to avoid the same problem with 3G networks. 3G (also known as UMTS) are a range of technologies which provide high speed wireless access, hence allowing for end user mobility[5]. The main feature of UMTS relies upon the fact that a base station allocates a dedicated communication channel to each cellular device it serves. Before 3G networks were available, a technology known as 2.5 G (or GPRS) was used as a development of the GSM standard. Theoretically, it was able to transmit data at 170 Kpbs (versus 9.6 Kbps of GSM), although most of the systems available in the market work at 40 Kpbs.
UMTS (Universal Mobile Telecommunications System) was a new third generation mobile communications system. UMTS had the support of many major telecommunications operators and manufacturers because it represented a unique opportunity to create a mass market for highly personalized and user-friendly mobile access.
UMTS was to deliver pictures, graphics, video communications and other wide-band information as well as voice and data, direct to people on the move. UMTS was to build on and extend the scope of mobile technologies (such as digital cellular and cordless) by providing increased capacity, data capability and a far greater range of services. The launch of UMTS services as from 2002 was to mark the evolution of a new, "open" communications universe, with players from many sectors (including providers of information and entertainment services) coming together to deliver new communications services, characterized by mobility and advanced capabilities.
Why UMTS?
UMTS was expected to offer high-value broadband information, commerce and entertainment services to mobile users via fixed, wireless and satellite networks. UMTS offered new applications by accelerating convergence of telecommunications, IT, media and content industries to deliver new services and create fresh revenue-generating opportunities. UMTS was to deliver low-cost, high capacity mobile communications, offering data rates up to 2 Mbit/sec with global roaming and other advanced capabilities. Furthermore, UMTS could offer a global network available and accessible from anywhere in the world for users who wanted to carry with them just one item of terminal equipment. UMTS also permitted easy inter-operability when changing from one network to another.
UMTS Licenses in Europe
55 licenses were to be issued to operators and consortia in 13 countries throughout Europe, which implied the entry of 13 to 17 new entrants to the European cellular stage. The total cost to new and incumbent operators for this was estimated to reach around €170 bn. Spain awarded its license by March 2000 by means of a beauty contest. Following UK's pricing through an auction process that spring, Netherlands and Germany used the auction process. Meanwhile, Italy reshaped a beauty contest into a crypto-auction. These auctions took place at a moment when investors felt dissatisfied with TMT, with WAP and its associated services.
Apart from the bidding process, telecom operator managers had a clear motivation to enter multiple markets, given that it was perceived mobile operators who did not possess a UMTS license ran the risk of disappearing in 10 years. Additionally, it was widely understood that entering a market through a bidding consortium was much cheaper than through a greenfield entry. Furthermore, a mobile company without the kind of product range that UMTS technology was going to make available would certainly face a serious competitive disadvantage.
The UK auction started on 3 March, but in April the Nasdaq crisis began and people began to think twice about the price to pay on subsequent licenses. The final price paid in the UK was GBP 22.5 bn, totaling €656 per POP for five licenses. Different spectrum allocations led to different total amounts being paid. License B, the largest of the licenses, was sold for €174 per POP. License A, reserved for new entrants, hence, lower synergies, was sold for €128 per POP, hence, the average price was €131 per POP, and the total price was €37.823 bn. UBS Warburg[6], based on these multiples, valued the German license at €53 bn. (€9-13 bn per license), the Italian at €2.6 bn. per license, the French at €7.7 bn., the Dutch at of €1.6 bn and the Swiss at €800m. Since Sweden and Portugal were to opt for beauty contests (similar to those applied in Spain), the costs for licenses were expected to keep being minimum (see table 1 for the estimated cost of licenses based on the price paid on the UK auction).
Table 1: Estimated cost of licenses based on UK auction[7]
Country / Pricing Relative to UK / Cost of license Eur m / License valuation per pop (EUR)Austria / 90% / 970 / 120
Belgium / 90% / 1,511 / 150
Finland* / nm / - / -
France / 80% / 7,674 / 133
Germany / 100% / 10,813 / 133
Greece / 90% / 1,556 / 150
Italy / 60% / 4,558 / 80
Netherlands / 90% / 1,831 / 120
Portugal / 10% / 0 / 17
Spain*º / nm / 0 / 3
Swedenº / nm / - / -
Switzerland / 70% / 815 / 116
UK / 100% / 7,674 / 131
* Actual. º Beauty Contest
In that same report, UBS maintained that[8]
We would make the point that UMTS for most incumbents is value-enhancing. In return terms, this means that double-digit IRRs should be possible. For Vodafone, we estimate an IRR from UMTS of 19.6% […] The market's focus on returns all of a sudden is symptomatic of a general unwillingness to look outwards (or rightwards) on the P&L. This represents a real change of heart as far as cellular is concerned. Capex has consistently overshot forecasts (with subscriber growth admittedly) over the past five years, yet the stocks have performed well… We wonder how many investors looked at Vodafone's return on capital after the Mannesman transaction […] Buying Vodafone involves buying the goodwill and other group level features unique to the company.
The Spanish Mobile Market
In 2000, the cellular business in Spain was showing spectacular growth. By that time, penetration was 38%, having effectively more than doubled in 1999, and Spain was the fastest growing country in Europe in that regard. It was estimated that penetration in Spain was to keep growing very fast over the next few years. Furthermore, Telefónica Móviles was going to keep a leading position in market share, although Airtel, Amena and the fourth operator were to keep gaining market share. This and more data is shown on table 2.
The successful launching of Amena (Spain's third GSM operator) needs to be taken into account when analyzing the environment in which Xfera was launched. Amena had become EBITDA positive only 23 months after start-up. By Q1 2001, Amena, with a 16% market share in volume (client base) and 13.3% in value, had an ARPU of €26.3, an estimated customer’s acquisition cost (CAC) of €122[9], leaving a “pay-back” period (CAC/ARPU) of 5 months. The company’s strategy consisted of moving from volume growth to value growth, and achieving a 16.5% value (not volume) market share of the Spanish industry by 2005, obtaining positive net profit by 2004 and investing around €7 bn in 2001-2005, 70% of which was to be used for network deployment (mainly UMTS).
1
Table 2:Spanish mobile market. Main market assumptions[10]
Main Figures / 1996 / 1997 / 1998 / 1999 / 2000E / 2001E / 2002E / 2003E / 2004E / 2005EMarket Penetration (%) / 7.5% / 10.9% / 17.8% / 37.8% / 52.8% / 64.8% / 72.8% / 79.8% / 85.8% / 90.3%
Penetration Added / 5.2% / 3.4% / 6.8% / 20.0% / 15.0% / 12.0% / 8.0% / 7.0% / 6.0% / 4.5%
Market Subscribers (000) / 2,995 / 4,338 / 7,051 / 15,005 / 20,956 / 25,717 / 28,891 / 31,669 / 24,049 / 35,834
Market net adds(000) / 2,051 / 1,342 / 2,714 / 7,954 / 5,951 / 4,761 / 3,174 / 2,777 / 2,381 / 1,785
Market Shares:
Telefónica / 78.3% / 73.5% / 69.4% / 60.3% / 56.0% / 53.5% / 51.0% / 49.0% / 46.5% / 44.5%
Airtel / 21.7% / 26.5% / 30.6% / 32.9% / 33.0% / 32.0% / 31.5% / 31.0% / 30.0% / 29.5%
Amena / 0.0% / 0.0% / 0.0% / 6.8% / 11.0% / 13.5% / 14.5% / 15.5% / 16.5% / 17.0%
4th Operator / 0.0% / 0.0% / 0.0% / 0.0% / 0.0% / 1.0% / 3.0% / 4.5% / 7.0% / 9.0%
1
Xfera & Spain’s UMTS License
By 2000, it was estimated that the valuation of an independent new UMTS license was to reach between € 1.3-3.1 bn, this being the net value of estimated investment required (€ 5.5 bn) and the fee required by the Spanish government (€ 129 m)[11]. Three existing mobile operators (Telefónica Móviles, Airtel and Amena) and three new consortia (Movi2, Movilweb 21 and Xfera) were bidding for four new licenses. Xfera’s shareholders are shown in the following table:
Table 3: Xfera's shareholders structure
Vivendi was diversifying from its French GSM business into different European countries. It had a Joint Venture with the Spanish construction and services company FCC, similar to the agreement reached between the finnish operator Sonera and the Spanish construction company ACS. There were two financial partners (Mercapital and CF Alba), a toll-motorway company (Acesa) seeking diversification and some local savings banks owning a 5% stake. Other partners controlled a 7% stake.
Once all the bids were submitted, a "beauty contest" was to take place, the final result of which would to be based on three technical aspects, i) technology of the project, ii) financial and business strategy and iii) contribution to domestic output. On 15 March, 2000 the Spanish Government awarded Xfera the new UMTS mobile license, with the remaining three licenses being awarded to the three existent operators. The legal framework included an annual UMTS license tax, which was imposed later on the year[12]. UMTS operators were to launch operations by August 2001, and the Government did not rule out the granting of two additional GSM licenses by April 2001. In addition, Xfera, as well as the other winning UMTS consortia, were to offer guarantees of €3 bn to the Government to secure the compromised Capex, and each consortia agreed to fund and build its own network, with no sharing of the networks allowed.
Presentation of Xfera to the capital markets
Before addressing the valuation of Xfera at the time the license was granted, it is important to see how Xfera presented itself. On May 2001, Mr. Carrión (one of Xfera's managers) presented the company to the financial markets. By that time, the legal framework of the Spanish mobile industry had changed considerably, with the annual UMTS license tax under revision, the delaying of the launching of UMTS services to June 2002, the granting of permission to Xfera to use other operators’ networks in order to start operating (through the postponed granting of the two new GSM licenses which the Government had been thinking of giving a year earlier)
Mr. Carrión's presentation used graphs which showed how the mobile data penetration was supposed to reach 34% in Europe and 32% in Spain by 2004. In turn, by 2004 mobile devices should have overtaken PCs as a means for Internet access. Additionally, the presentation stated that "telecom services move towards integration of traditionally separated platforms". A new value chain was being created in the sector, which rendered operators as no more the central piece controlling the user ownership, with multiple different new players trying to gain some customer market share (e-tailers, financial services & banks, content providers, etc.). This value chain was supposed to be the result of the convergence of telecom & Internet business value chains, with many new and different players vying for space. Hence it was imperative to go from being a business concentrated in access to one concentrated in content and portal revenues. In order to secure portal value, players would have to compete for customer ownership, which was by no means guaranteed, given that customer options increase and competition intensifies. Hence, competition through partnerships and alliances was sure to become intense.
When comparing fixed Internet with mobile Internet, the first was offering an endless supply of content options, whereas mobile Internet was driven by applications and the ability to access information anytime and anywhere. For this reason, if, under the WAP business model, between 90-95% of the business was to be generated in access with only 0-5% in content and portal, under UMTS, access was to represent 40-60% of the business, content 10-20% and portal 30-40%.
Mobile technologies evolved from 2G, a service concentrated on voice, SMS and slow WAP that would provide give access by Q2-3 2001, to early 2.5 G tech, which offered fast WAP and bluetooth applications; the latter was to give access to mature 2.5 G (also known as GPRS) in the market as from Q4 01 to Q3 02, which would offer color screens, music and optional cameras; then early UMTS cameras, in the market by Q4 02 until Q2 03, would offer streaming images and music and downloads, and, finally, mature UMTS, in the market by 2004, would provide video-clips, videoconferences and interactive multimedia applications.
Regarding challenges, this presentation divided them into four areas,
i)those related to mobile-internet business convergence (which implied new revenue models, partnership structures, outsourcing strategies, new profile headcount to be recruited and a new structure for the organization able to answer to this environment),
ii)the movement from operator-centric strategies towards customer-centric ones (involving changes in the offer design –packaging, pricing, distribution-, as well as a new communications strategy),
iii)a new competitive landscape (a new player had to set a strategy to compete with international operators such as Vodafone, or with integrated telecom companies, such as Telefónica, which could use a bundling strategy; additionally, a well-defined niche strategy was needed, and the question of whether it was better to wait for UMTS technology or launch GPRS services needed to be answered), and
iv)the challenges of new technologies (UMTS tech development calendar was uncertain, GPRS’s list of services was still not fully developed, and IT system integration for the new services was of key importance).
Xfera planned to offer services by 2001 targeting top of the range mobile users, by offering fast WAP. Over 2001-2005, Xfera had announced investments of € 7,813 m. Main financial targets included breakeven at EBITDA level by 2004 and bottom line profits one year later, by then, Xfera expected to generate revenues exceeding € 3,005 bn.
Valuation of UMTS licenses in Spain
Silverman read one UBS Warburg[13] report that assumed two different scenarios. First it took the estimated UBS value of TEM[14], Airtel and Amena to calculate the equivalent values per subscriber in 2005, estimating a market penetration at 90% for 2005, and assuming a market share for the fourth operator at that time of 9%. This methodology relied on an EV/sub by 2005 of € 2,647 for TEM, € 2,420 for Airtel and € 1,493 for Amena. This valuation was undertaken on a homogeneous European basis, with the European valuation model assigning US$ 3,500 (US$ 2,000 for traditional cellular and US$ 1,500 for mobile data) to a mobile subscriber at maturity (90% penetration), and then discounted back to arrive at today's valuation. The standard EBITDA margin was set at 40%, so an operator with a higher margin could claim more value per subscriber. In addition, operators were graded by a time value of money factor (year to maturity) and a long-term market share.