1. Introduction

The UK staged the first of the European 3G auctions and it set the expectations for the auctions that were to follow in the rest of Europe.

As with any auction it is the last entity to leave the auction empty handed that sets the price for those who emerged with a 3G licence. This last entity out was a joint venture between ntl and France Telecom, with the activity based largely in ntl.  ntl itself was created by two entrepreneurs who bought a few disparate UK cable TV franchises with borrowed money and grew at a frenetic pace to take over the entire UK cable TV industry…and burnt through $34 billion dollars in the process. It had to make a Chapter 11 bankruptcy pit stop to survive. But it could have been a lot worse…it might have won a 3G licence!

In 1998 I found myself head of advanced technologies in the UK cable TV operator ntl. One of the jobs I had was to prepare the company for the UK 3G auctions. I led this activity until just prior to the auction itself when a colleague from our corporate HQ I’d been working with took it over.  This account and analysis views the 3G auction from this unique vantage point…the company that set the final price of a 3G licence.


2. Industry Preparations for the 3G auction

A large number of companies were tracking the 3G opportunity since the DTI had issued its consultative note in 1997.  By early 1998 most were stepping up their preparations.  The starting point was to construct a business model to see if the opportunity would be profitable. This required working out:

  • The cost of rolling out a 3G networks and subsidising hand sets
  • Dimensioning the 3G market opportunity
  • Putting a value on the 3G Licence

Cost of rolling out a 3G Network

Network experts created a number of network models to give their companies a feel of the cost of the 3G networks, how much traffic they could carry and what area they might cover. This brought out a number of important features of the 3G technology affecting its economics:

  • The customer speed of “up to” 2 Mb/s being touted around was not all that it seemed. To obtain the 2Mb/s the customer would have to be standing right under the base station mast. One might as well install a public payphone box if 3G required all its customers to crowd under a radio mast to enjoy a 2 Mb/s service. A customer situation 5 miles away from the base station mast might only get 64 kb/s. That was hardly better than the GSM networks could be stretched to achieve.
  • The key to getting high speed was to cut the maximum distance a customer was likely to be when connecting to the network. This in turn required a city or town to be covered with an immense concentration of small cells. These all cost money. A more practical speed expectation might be 384 kb/s ie five times less. Even this would involve massive investments just to cover the main cities, motorways and towns. Forget ever covering rural areas with a density of 3G networks cells to achieve this sort of high speed!
  • The next surprise to emerge was the cost of connecting all the 3G cells back to the public networks. The 3G frequency channels were 25 times wider than the 2G GSM channels. This pushed up the cost of the telecommunications “back-haul” links to connect the base station cells to the cellular radio operator’s core network.

Most of the companies would have arrived at a plan for around 6000 base stations and a price tag of around £1 billion would have been earmarked for the 3G network.

Everyone knew that it was inevitable that they would have to subsidise 3G handsets. This was the way the GSM market worked and all the 3G operators would have to fall into line. Indeed a new entrant 3G operator was likely to have to pay bigger subsidies since it would have to buy its way into a market share. The comfort was that the semi conductor industry would be working over time to bring down the number of chips needed to build a 3G hand set and this would bring the cost down in due time. In this way the £400-500 initial cost of the first 3G handsets would inevitably fall to under £200 in time. Everyone also knew that all 3G handsets would have to have a GSM capability built in from the very outset since the first 3G networks would only have limited coverage of the country.

On this basis each company was likely to have earmarked at least £500 million of hand set subsidies to get its 3G customer base started.

The 3G Market Opportunity

There would have been a lot of head scratching when it came to defining the market opportunity over and above simply grabbing a share of voice telephone calls. Into this vacuum the manufacturers exercised a lot of influence and nobody more so than Nokia.

Nokia’s “3G technology” propaganda was in a class of its own. They had done some quite innovative work in stimulating a number of small Silicon Valley companies to try out ideas for Internet type applications on the mobile phones. They were also one of the champions of getting Internet Web site text onto mobile phones using a protocol they called WAP (Wireless Application Protocol).

They showed interested potential bidders a graph of the steep world-wide growth in the numbers of mobile radio customers. This was a curve soaring upwards. They then showed a graph of the steep world-wide growth in the number of customers connecting to the Internet. It was another curve soaring upwards. The 3G operator hopefuls were being invited to imagine the potential growth that would come from putting the Internet on a mobile phone. A market that would be the imposition of one high growth curve being overlaid on another high growth curve was a very compelling image. The 3G opportunity was an enabling technology to do this.

The brouhaha going on in media on WAP phones in the GSM market lent credibility to the vision.  There was something tangible happening even as Nokia spoke. If the industry was convinced that WAP would work for customers at data rates of 9.6 kb/s and tiny black and white screen – getting much higher data rates coupled with the best colour screens that Japanese industry were capable of producing…. this looked promising.

Nokia did a good selling job, as was their role in the market and made a significant contribution to stimulating interest in the potential of the 3G opportunity. They also probably did much to convince operators of the mistake they might make by staying out of the opportunity.

The other major influence from manufacturers at the time was the prospect of “Japan Ltd” taking the 3G opportunity seriously and putting into the market compelling 3G handsets, leveraging their world lead in small colour display technology and digital camera technology.

The next task for the various 3G auction preparatory teams was to look at what revenues they could expect to drive through a 3G network. The dismal news from all the market survey companies was that the UK cellular radio market was heading for saturation within two years or so. What this meant was that an operator could only gain a customer if one of the other operators lost one.  This was not promising for any new entrant as, by then, all four of the incumbent operators were performing well. The trick any new entrant had to pull off was to hold its churn down to a few percent below the incumbent mobile operators and this would drive up the number of customers it gained. At best it could hope for a 16% market share as its optimistic target with 8% as the pessimistic lower bound. Telephone calls remained the bedrock of the business case.

The really tricky bit was how much revenue a 3G operator could expect to extract from each customer from new data services?    The application that most preparatory teams would have been comfortable with was ones that combined location information in the service being offered. This seemed “a natural” to offer people “on the move”.

Views on the idea of using 3G to broadcast video entertainment or for use as videophones were mixed.   The sceptics would have recalled that the wire line world had made three serious attempts to get videophones off the ground. None of them had succeeded. Video conferencing had crept its way to a niche market and stubbornly refused to break out of its niche status. Pricing a video call at a realistic level, noting that it would consume 8 times the capacity of a normal phone call, seemed an improbably mass application. There was probably nobody, who had run the numbers, left with any delusion of full-length movies being sent to 3G mobile phones.

There was one video mobile phone concept that did appear credible. It was an idea known as “You see what I see”. The mobile phone would have a built in video or still image camera. The user would point at something that they wanted a person they were calling to see. A still picture or a short moving image would be captured, turned into a video file and the file sent over the 3G network to the person at the other end of the call. It was possible to conceive of a variety of circumstances where this might have application in business or meet a social need. And people would willingly pay for the video file to be sent. The 3G technology was offering a network that could send that file much faster.

There remained other revenue opportunities to put into a business case.  Advertising was one. The mobile phone is a natural “reminder” medium. For example an event organiser might want to suggest to customers that “tonight was the night of the big sports match” or whatever.

Cost of 3G licences

The final major input to the business case that had to be assumed was the price the 3G licences would cost. The Government’s advisor, Rothschild, was busy using the media to set industry expectations. The £250m to £500m range appeared the current buzz. For most preparatory teams their business case stood up well with the assumption of £500 million in the business case.  In fact the business case could sensibly absorb around £1 billion. Nobody was thinking of that sort of price in the 1999.

3. The “Mannesmann Catastrophe”

On the 19th October 1999 an event occurred that was to prove cataclysmic for the 3G auction prospects, the mobile radio industry and wider parts of the global telecommunications industry.  It was probably the biggest single contributor to the crazy prices paid at the 3G auction. Mannesmann announced to the market that it was in talks to take-over the UK cellular radio company Orange.  Two days later it announced that it had clinched the deal for £19.8 billion. The offer comprised 0.0965 new Mannesmann shares and £6.40 in cash for each. This made an Orange share worth £16.29.   Mannesmann was also to shoulder around £2 billion pounds of Orange debt.

The effect of this news was electric. In 1996 Orange floated shares on the Stock Exchange at a price of £2.05. The price Mannesmann was offering amounted to nearly an 800% gain in 3 years.  It was jackpot time for the lucky Orange shareholders. One journalist worked out that for the same price Mannesmann could have bought Marks & Spencer, Sainsbury and Next, with Rolls Royce thrown in.

Even more stunning was the price Mannesmann paid for each of Orange’s current customers. At the time of the take-over Orange had 3.5 million customers and a modest 17% share of the UK market. The purchase price of £19.8 billion worked out at over £5,000 for each customer. In that month Orange happened to be selling a pre-paid package that allowed a customer to have a mobile phone for as little as £50 a year. In other words a customer on that tariff would have to live for another 100 years to pay their contribution towards the purchase price of Orange!  Even that calculation assumed that the Orange business was paying its way. It wasn’t. In 1998 Orange had turned in a loss of £105.7m on sales of £357.9m. Orange was still in the middle of its investment cycle for its network.

The effect of the Mannesmann purchase of Orange within the companies preparing for the 3G auction was immediate. Within hours of the announcement, most of the preparatory teams would have worked out:

  • to secure Orange’s 3.5 million customers would require a much bigger 3G network than it had hitherto planned for say double the size ie £2 billion
  • £1.4 billion would allow it to give absolutely free 3.5 million 3G terminals at say £400 each,
  • throw in a huge marketing budget of say £600 million and
  • add the £1 billion for the DTI licence.

For a mere £5 billion a company could create a mobile entity that Mannesmann had just paid £19.8 billion for!

The world had just re-set the value of a 3G licence. The price started with a “billion” on the end of it. The DTI auction was suddenly offering the key to a money printing press. Mannesmann had inadvertently fired the gun that started the stampede for a 3G licence at prices that would have been laughed at only a few weeks earlier. The mechanism was not to distort the 3G business models being prepared but to remove the lid on what the majority of companies were prepared to pay in the heat of the auction.

Why did Mannesmann bid such a huge amount of money for Orange? What did they see in the value of Orange to make it worth such a sum?  The reality was that the short to medium term profitability of the business they were buying was not a factor driving the decision. The Orange purchase was just an enormously expensive tactic to fight off the possibility of an unwelcome hostile take-over bid. Vodafone was the unwelcome stalker.

Mannesmann was rightly the pride of Germany. A leader of a sunset industry had re-invented itself as a successful industry of tomorrow. Mannesmann owed its success to one man more than any other – Klaus Esser. The day the deal with Orange had been announced Mr Esser and his colleagues were celebrating the tenth anniversary of Mannesmann’s entry into telecommunications. Just as Vodafone had been selected by the UK Government to be the new entrant in the UK market, so the German government had chosen Mannesmann as the new entrant for the German market. Vodafone would take on British Telecom in the UK market. Mannesmann would take on Deutsche Telekom in the German market.   Like Vodafone, they had realised that success in cellular radio was to invest heavily. By 1991 they had spent £600m putting in a GSM network across Germany. These efforts were rewarded and the German second cellular operator went from strength to strength.  It was the “Vodafone style success” story of Germany.

The key difference between Vodafone and Mannesmann was that Vodafone had got its chance five years ahead of Mannesmann. That was a result of the Thatcher Government throwing open the UK telecommunications market to competition much earlier than the rest of Europe. It was the decisive factor on who was to come out on top in the battle of wills between the two companies.

Once Vodafone had secured its leadership of the UK cellular radio market it looked overseas for expansion.  Vodafone chased every opportunity vigorously. One of the rules they had worked out early in the game was to pay for acquisitions with their own shares and not to pay cash (or high yield bonds). To make this work required that the Vodafone share price had to continuously rise. They had luck on their side with the exceptionally long bull market that the world economy was enjoying. But they also worked hard for it. Vodafone was well run. They had a management team that worked well together. It was also a remarkably stable management team. This allowed a long-term strategy the time to work. By the late 90’s Vodafone was ready to make the leap from large regional player to global giant. Mannesmann, running 5 years behind, was still in its “regional” growth phase.

In January 1999 Vodafone announced to an enthusiastic market that it had acquired the US cellular radio operator AirTouch for $62 billion in stock and cash and around $4.5 billion in assumed debt. Vodafone shareholders and AirTouch shareholders were each to own approximately 50 percent of the capital of the combined company.

AirTouch itself had been no sloth in the race to gobble up stakes in cellular radio companies around the globe.  The AirTouch acquisition gave Vodafone not only a strong position in the U.S. cellular market but new positions in India, Italy, Japan, and Spain. Germany was the only market where the two company’s operations overlap. Vodafone announced immediately that it planned sell its 17 percent stake in German mobile phone firm E-Plus so it could keep AirTouch’s stake in market leader Mannesmann Mobilfunk.  This set the alarm bells ringing in Mannesmann. The Mannesmann Board had a certain comfort that German take-over laws stacked the odds heavily against a hostile take-over. But perhaps it was time to build up the defensive barricades a bit higher.

One way to beat off any unwelcome bid from Vodafone was to buy-up a UK competitor to Vodafone and enjoy the protective shield of the UK or European regulatory authorities blocking any take over on competition grounds. The only problem was that there was no obvious way to buy any of the UK cellular radio operators. Vodafone was the only wholly publicly owned company. The AirTouch acquisition had put it right out of reach. Cellnet was owned by British Telecom. No chance there! Orange was considered relatively bid proof because it was 44.8 percent-owned by Hong Kong conglomerate Hutchison Whampoa, which had consistently said it wanted to use the British company as its expansion vehicle in Europe. This left One2 One that was jointly owned by Cable & Wireless and the MediaOne from the USA.

In March 1999 developments in the USA freed up the situation. The US Cable TV Group Comcast took over MediaOne and made it clear that it did not regard the mobile business as part of its core business. Cable & Wireless was about to make another of its strategy zig zags. It decided to get out of the UK mobile market altogether.  One2One was put up for sale.

Five European companies expressed interest in buying Mercury One2One including Deutsche Telekom, France Telecom and Mannesmann.  Deutsche Telekom was keen to break into the UK market. It was flush with cash from a secondary share offering.  It was still suffering the humiliation of its botched merger with Telecom Italia. It was time for a bold knock-out blow.

They stunned the market with an £11 billion bid. This valued each One2One customer at £2700. For the mobile radio industry it was a staggering premium to pay. For the financial institutions the calculators came out and mobile radio stocks were re-valued upwards.

France Telecom had put a sensible sum of money on the table and had been left out in the cold as a result. Putting sensible valuations on acquisitions in the telecommunications sector was going out of fashion!

Mannesmann had also let One 2 One slip through their fingers. They had been taken by surprise at the amount of money that Deutsche Telekom had put on the table. Worse, they could now feel Vodafone breathing down their necks. With One 2 One now out of play Mannesmann had to go for Orange. There had to be a price that made it worthwhile for Hong Kong conglomerate Hutchison Whampoa to sell its stake in Orange.  By October Mannesmann had clinched crucial agreement with Hutchison for its 44.8 percent controlling shareholding in Orange. The price was £19.8 billion. Hutchison was given a 10.2 percent stake in the enlarged company.

For the extraordinary price Mannesmann paid for Orange it won three things. First the combined value would make it much more expensive for Vodafone to take it over. Second, its ownership of Orange made things more complicated for Vodafone. The regulatory authorities would not allow the Vodafone, the UK’s market leader, to own one of its UK rivals. Third Mannesmann became the largest of the European regional cellular radio players. It had also won parity with its domestic competitor Deutsche Telekom. Both owned a UK cellular radio company. Absent from this list was any notion of ever making a profit from this staggering “investment”.

Vodafone was stung by the Mannesmann acquisition of Orange. The combined assets also dislodge Vodafone from its top spot in Europe. Mannesmann now had 7m more mobile subscribers in Europe than Vodafone. This had longer-term implications for the competitive position of Vodafone in the European market. Advisers to Vodafone put the word around that the `jury remained out” over whether or not to proceed with an attempted hostile take-over of Mannesmann.

A hostile approach would be fraught with difficulties. A hostile suitor would have access to only limited information on a target company, making it difficult to work out the full consequences of a take-over. The structure of German management boards and the complex cross-shareholdings between banks and industrial groups also posed formidable barriers to hostile take-overs. A shareholder needed more than 75 per cent of voting shares under German law to exercise management control. Without it, it would be unable to remove members of supervisory boards, which are invariably dominated by representatives approved by a company’s workforce. In general the workforce representatives resisted take-overs because they usually led to job cuts. All this was in addition to a curious shareholding structure that stated that no Mannesmann investor could control more than 5% of votes. That rule expired in the summer of 2000, which has led some commentators to believe that Vodafone would bide its time before striking. Klaus Esser was to tell reporters at a press conference “I don’t see any realistic risk of a hostile take-over“.  At least history appeared to be on his side.

At the beginning of November rumours circulated in the media that top executives from Vodafone AirTouch and France Telecom had met within the previous fortnight to discuss a near-£70billion joint break-up bid for Mannesmann and Orange, its new British mobile-phone subsidiary. Mannesmann shares duly surged nearly 30% over two weeks.

On the 14th November 1999 Vodafone’s Chris Gent flew to Dusseldorf and met Klaus Esser. He tabled a “friendly” bid for Mannesmann’s telecommunications group of £64 billion. Klaus Esser,Mannesmann chairman, immediately threw out the offer as “wholly inadequate”. He said later in a public statement that “A deal would solve Vodafone‘s problems in Europe, but what does it offer Mannesmann?”  This was followed up by a letter urging Vodafone to withdraw its proposal.

Six days later there was a parting of the ways between the Anglo-US and German mergers and acquisitions culture. Vodafone tabled a £79 billion hostile offer. A wave of anger swept through German media and political circles. The offer was firmly rejected by Mannesmann. Vodafone’s share price slipped during the day reducing the value of the offer to £75 billion. This gave Mannesmann an immediate line of defence. Esser told the media: “The lack of cash makes such an offer very risky for our shareholders.”  Mannesmann‘s supervisory board supported the Executive Board’s view that the offer was “entirely unattractive”.

Vodafone set out a time-table for its bid. The offer would be made on Thursday 23rd December and it would be closed Feb. 7 2000. Vodafone would meet with its own shareholders on the 24th January 2000 and ask them to endorse the offer. In the meantime it set about wooing Mannesmann shareholders, many of which were institutions outside of Germany. This put centre stage the question of who were the ultimate decision takers of a German based public company – the shareholders or the management/employees.  To emphasise where Vodafone believed the power of decision should rest it offered a sweetener to its deal that valued Mannesmann at £95 billion. A strengthening of the pound had allowed Vodafone some margin to improve its offer. The improvement in the offer had strings. It only “kicked-in” if the Mannesmann Board recommended the deal to its shareholders. Vodafone reasoned that failure of the Mannesmann Board to do this would expose them to their shareholders as standing in the way of them getting a really great deal.

For weeks Esser and the Mannesmann Board tried to persuade investors that the German engineering and telecommunications concern would do better on its own. It also pursued other possible alliances. The German concern had held talks with French concern Vivendi about an Internet alliance in the hopes of staving off Vodafones pursuit. Instead Vivendi formed an alliance with Vodafone.  The damage to the credibility of the Mannesmann strategy far outweighed the potential value of the deal. Klaus Esser and the Mannesmann Board were fast running out of option. They knew that if the Vodafone bid were to fail the short-term prospect for their shares was not just a dip. They would dive vertically downwards.  They could write off any notion of loyalty from the global financial institutions. But they could not depend upon German institutional shareholders to stand firm either. Even they had wider shareholder interests to protect. The raw power of Anglo-US capitalism was starting to bear down on the “consensus” German industrial structure.

As the closing date for the offer loomed in the beginning of February Klaus Esser and the Mannesmann Board threw in the towel. There was no conceivable Mannesmann interest in prolonging the fight. Mannesmann shareholders emerged owning 47.2% of the world’s largest cellular radio operator. If that prospect did not appeal to them they had to option to sell their newly acquired Vodafone shares in an exceptionally bullish stock market for cash.

Amongst those likely to have been taking some of their winnings in cash at the earliest opportunity would be the Hong Kong conglomerate Hutchison Whampoa, who had finished up with a lot of Mannesmann shares from the earlier Orange deal.

Everyone knew that the Vodafone purchase of Mannesmann would never get past the competition authorities unless Vodafone disposed of Orange. Disposing of Orange was therefore part of Vodafone’s plans. In fact Vodafone was depending upon the disposal of Orange to help pay for the cost of taking over Mannesmann. In that respect the Mannesmann ploy of buying Orange had failed as a hostile bid defence.

In the meantime, Vodafone was the temporary owner of Orange.

4. DTI gets its auction back on track

As enthralling as the Vodafone battle for the control of Mannesmann was to the rest of the world, it was a nightmare for the Department of Trade and Industry. They were impatient to get their 3G auction underway.  They knew the market conditions were right for the auction. Like everyone else they did not know how long this would this last. On the other hand pushing ahead with the auction with two of the prospective lead bidders, Vodafone and the owners of Orange, involved in a hostile take-over battle involving vast sums of money would itself be risky.  There was considerable relief in the DTI when the battle was finally over. They quickly pushed ahead with plans for the 3G auction.

The DTI received applications to take part in the auction from all four of the existing UK cellular radio operators: Vodafone, British Telecom (Cellnet), Orange and One2One.

In addition applications to take part in the auction were submitted by:

  • MCI Worldcom
  • SpectrumCo Ltd (a consortium comprising Sonera Corp., Tesco, EMI Group and Nextel Communications and the Virgin Group)
  • Telefonica (the principal Spanish telecommunications operator)
  • 3G (UK) Ltd (owned by Ireland’s eircom)
  • Crescent Wireless (with the same principal shareholders as a new alternative optical fibre telecommunications carrier Global Crossing)
  • Epsilon Tele.Com Plc, (a bidding vehicle set up by the Japanese bank Nomura);
  • One.Tel Global Wireless Ltd (from Australia); and
  • TIW UMTS, (a subsidiary of the Canadian enterprise TIW)
  • ntl in partnership with France Telecom

In early February 2000 the DTI E-Commerce Minister Patricia Hewitt said that she was delighted to announce that the DTI had been able to qualify all the applicants and were on track for the auction to begin in March 2000.  They held their breath as they awaited the reaction of the industry to their decision to allow Orange to enter the auction. British Telecom immediately protested. How could a company appear twice in the auction? They threatened to take the Government to Court. BT contacted a number of the other registered bidders for support. Whether it was over confidence, a reflex action that anything BT demanded from the Government had to somehow be against the interest of the other players or whether most were simply fast asleep but BT got a dusty answer from the other registered bidders. They were on their own. They backed down. The DTI had won their bluff. They were, without doubt, on shaky ground. As it was, all the bidders then went into the auction having accepted the decision.

In cloaks of massive secrecy almost all the bidders come to similar conclusions on the value of a 3G licence. They thought they might have to pay as much as £1 billion. Most were prepared to stretch themselves to as much as £2 billion if they really had to. The DTI’s own economists were still projecting £1 billion per licence as the very best case outcome for the DTI.

5. The Biggest Gambling Game in Town

The mobile manufacturers had done their best to inject some dot.com fizz into the mobile radio auction. The Government’s “Multimedia on the move” picture of increasing synergy between telecommunications and broadcasting had enriched the vision. By the day the auction the media had created the usual aura around 3G of “new services one could not yet even dream of”. Sitting well above the reality (or otherwise) of business cases, an image had been created around 3G of something desirable and indeed essential to be a part of.  The bidders at the DTI 3G auction were without doubt in the high stakes, high risk and high reward business. Put another way they were gamblers. The casino owners were the Government.

The DTI Radiocommunications Agency has always had one of the best UK Government Internet Web sites. They used it to great effect to keep the world informed of what was happening at the auction.  Each round of the auction was reported upon within minutes by means of colourful charts. The UK DTI was running this particular casino in style.  It was by far the biggest game in town.

Round 1 of the auction saw Orange emerge as the only existing operator tabling a lead bid.  A clutch of companies most people had never heard of also emerged as lead bidders. TIW, Crescent and Epsilon were not exactly leading names that would set the mobile radio world alight. More surprising was that one of the licences failed to attract an initial bid. This must have caused a flutter of nerves amongst the casino owners.

Round 2 cheered up the ntl audience. ntl emerged as the lead bidder for Licence A. This was the licence set aside for a new entrant and it had by far the most radio spectrum attached to it.  By Round 5 ntl had been pushed out of Licence A. By Round 6 it had hopped over and re-appeared in Licence C as the lead bidder.

It was Round 7 of the bidding before BT appeared with a lead bid. They were pitching for the B Licence. This had the most spectrum of the licences available to all comers, including the existing UK mobile radio operators.  Surprise! Surprise! Vodafone appeared in Round 8 and knocked BT off their Licence B perch.

In the first 20 rounds it appeared to be anybody’s game.  ntl and a company calling itself Spectrum Co slugged it out for the pole position for the A licence. Licence B, the best licence for the incumbent mobile operators, saw a contest between Vodafone and its “arms length” subsidiary Orange.  By round 22 Orange seemed to get the hint that its new owner did not want it bidding on Licence B and it switched its predominant attention to Licence E.

A few companies made much of their super gamesmanship they played in their auction bid strategies. For all the good it did them. Vodafone’s strategy was announced well before the auction began. They would simply keep their hand in the air for Licence B until the last of the other bidders dropped out.  For the entire 150 rounds of the DTI auction they pitched for Licence B. They were lead bidder for 108 of the 150 rounds for this licence, including the crucial final round.   It evident from the record that Orange pitched into bidding seriously for the lesser Licence D from Round 22. They were lead bidders for Licence D for 90 out of the 150 rounds.

The mood of the most impartial observers of the auction gradually changed from one of excitement to concern.  It took an incredible 77 further rounds before the field started to seriously thin out. The £2Bn barrier had been crashed through well before this. Epsilon’s last appearance was in Round 84, Crescent in Round 88 and One.Tel in Round 97.

The Merchant Bank advisors to the DTI had spent months generating an expectation within the industry that each licence was likely to fetch between £500m to as much as £1Bn. The DTI’s own economists saw £1Bn as the maximum figure they were likely to extract for any of the licences.  Most bidders had decided that, at a stretch, they could bid up to £2Bn.  If a figure of £2Bn was the very top figure that any sense could be made of acquiring a 3G licence, why was anyone still at the auction table in Round 100?

The exit of SpectrumCo in Round 93 left WorldCom, Telefonica, TIW and ntl struggling to seize Licence A. By Round 105 Worldcom had made their last appearance in the lead for Licence A and Telefonica had switched to one of the lesser licences by Round 109.  Winning Licence A became a two-way fight between TIW and ntl.

The mood of the impartial observers had changed yet again. Concern had given way to incredulity as another £1Bn was put on the table. In Round 126 TIW knocked out ntl as the lead bidder on Licence A. Relief swept though everyone at ntl not concerned with the auction. Common sense has arrived at last. It was short lived. The relief gave way to disbelief as ntl swept back into the bidding for Licence A in Round 127 and held sway until Round 131.  In Round 130 ntl had put on the table a mind blowing £4.277 Bn.  In Round 131 TIW had put on the table an even more massive £4.384 Bn.  It clinched Licence A for them.

This was an auction that the ntl auction team seemed unable to leave. It switched its attention to one of the lesser licences. It went for Licence C. In Round 132 it managed to push Telefonica out of the auction. ntl had stretched its expectations as a result of bidding £4.277 Bin for Licence A. Scaling the amount of spectrum from Licence A to Licence C made it seem worth having a go for Licence C. Any hope of a licence was short lived. It had BT to contend with. BT had just been shoved out of the Licence B competition by Vodafone with a grand slam bid of £5.3848 Bn. BT got renewed energy by scaling its licence B losing bid to where Licence C now stood. It certainly was not going to be outbid by a relatively small market entity like ntl, even less was it going to let France Telecom into the party.  By round 148 ntl put in a last gasp bid of £3.9705 Bn. In round 149 BT knocked ntl out of the auction entirely with a bid of £4.031 Bn.

By Round 150 of the auction there were now no challengers left.

The biggest gamblers in town emerged with their licences:

Licence A – TIW at a cost of £4,384,700,000

Licence B – Vodafone at a cost of £5,964,000,000

Licence C – BT at a cost of £4,030,100,000

Licence D – One2One at a cost of £4,003,600,000

Licence E – Orange at a cost of £4,095,000,000

The Casino owners – the UK Treasury added up their “takings” from the auction. A staggering £22,477,400,000! Their expectation was a mere £5,000,000,000.  Even that seemed a large sum of money to pay for a bit of paper that hitherto had been given away almost free.

The moods at the senior management levels in the UK operation of ntl immediately after the 3G auction was one of relief if not joy. There was a perverse feeling that ntl had been lucky to lose at the 3G auction.  Fortune had indeed smiled on them and they all knew it. They had not won a 3G licence.  What a narrow escape! What a slice of luck!

6. And the rest of Europe

Those who thought that the results of the DTI 3G auction in the UK would shock industry into moderating its behaviour at the other 3G auctions being held across Europe were mistaken. The Dutch auction was a bit of a disappointing affair for the Dutch Finance Minister with the number of bidders matching the number of licences. It raised only $2.7 billion for five licences.

The German auction showed that industry had not shifted from its excessive valuations of the 3G licences.  The bidders for the German licences included the existing German mobile phone operators – Deutsche Telekom AG , Mannesmann Mobilfunk (read Vodafone) , E-Plus-Hutchison, Viag Interkom (read British Telecom) and MobilCom (with France Telecom support).

New entrant bidders included Group 3G , a consortium linking Finland’s Sonera and Spain’s Telefonica, and Swisscom-backed debitel . The character of the German auction was different from the British experience and led to the following outcome:

Block          Bidder                            High Bid        Backers

1              Group 3G                           $3.33B           Sonera, Telefonica

2              Group 3G                           $3.33B           Sonera, Telefonica

3              Mobilcom                          $3.26B        France Telecom, Orange

4              Mobilcom                         $3.41B            France Telecom, Orange

5              Mannesmann                   $3.38B           Vodafone AirTouch

6              T-Mobil                               $3.39B           Deutsche Telekom

7              Viag Interkom                   $3.25B           British Telecom, Telenor

8              Mannesmann                    $3.32B           Vodafone AirTouch

9              Mannesmann                    $3.32B          Vodafone AirTouch

10             T-Mobil                             $3.25B           Deutsche Telekom

11             Viag Interkom                 $3.26B           British Telecom, Telenor

12             T-Mobil                              $3.25B           Deutsche Telekom

Total                         $42.75B or £28.5 billion

Perhaps the most interesting outcome was the Italian auction. It was held in October of 2000. The expectation of the Italian finance ministry was high when, like the UK, it put up 5 licences for sale. Unlike the UK auction it only attracted six bidders rather than the 13 that pitched into the UK auction. Perhaps financial fatigue was setting in across the industry.  Having said that, Italy is one of Europe’s major cellular radio markets. Lots of cars, lots of traffic jams and lots of people talking on their mobile phones.  The bidders were all serious players and, for a while, things looked promising. The total bids quickly mounted to $11.6 billion. Then a spectacular public row clattered across the media between British Telecom and its Italian partner. Claim and counter claim of bad faith shot backwards and forwards.  Whatever the truth of the matter the consortium just disintegrated. The Italian Ministry was left with five licences and five bidders. The auction stopped dead in its tracks. The finance ministry was mortified. Politicians struggled hard to find a conspiracy. There was none to find.

The five emerging winners were:

  • Telecom Italia Mobile (TIM) – TIM is Italy’s largest mobile-phone company, 60% controlled Telecom Italia, and which in turn is 55% owned by Olivetti.
  • Omnitel Pronto Italia – Italy’s number two wireless firm, 76% owned by Vodafone, and 23% by Verizon.
  • Wind Telecomunicazion – Wind is majority-owned by Enel, the Italian State electric utility with significant ownership by France Telecom (43%).
  • IPSE 2000 – A consortium headed by Spanish telecom Telefonica (39%);
  • Finland’s Sonera (19%); and a group of Italian industrialists.
  • Andala – A consortium 51% controlled by Hong Kong’s Hutchison Whampoa, along with Tiscali (25.5%) and CIR (15%)

The most interesting thing about the Italian auction is the price that the winners finished up paying. The total of $11.6 billion equates to roughly £1.5 billion per licence. The irony is that the fair price actually resulted from a failure of the auction process.

The Danish Auction left its mobile radio industry with mixed blessings. The bids were sealed bids. Four licences went to the four enterprises that had bid the highest sums of money. Except under the Danish rules they all paid the fourth lowest bid price. That still led to a high cost of a 3G licence in Denmark of around 1 billion Krone.  Much more “industry friendly” was the payment method. The winners paid 25% of the money up front and the remaining money was to be paid at 7.5% a year for 10 years. There had to be bank guarantees to cover off the first three years of payment and then the operators were free to walk away from the rest if their businesses were not working.

7. Why the 3G auction bids went so high

If all the bidders entered the first of Europe’s 3G auction expecting to pay £1 billion and set their upper at £2 billion, it takes some explaining as to why they all emerged paying between £3.97 billion and £5.96 billion.  A good place to start an analysis of why the auction prices went so high is the UK auction itself.

The length of time it took to get to all the bids into the £500m territory showed 13 players all feeling for where the natural ceiling might be influenced by the expectations Rothschild had originally set for the industry. That was Round 44. It was not until Round 65 of the auction before the first bidder broke the £1 billion barrier with their bid.

On 1st April, four weeks after the start of the auction the total of all bids hit £10 billion and not a single one of the original bidders had left the auction. By round 97 and two days later all the bids except one were just over the £2 billion mark. The first sign a faltering occurred. All the bidders could exercise an option of staying out of a particular round of the bidding on up to three occasions. This was the point where bidders had hit the limits that they had set for themselves and wanted time to consult with their financial backers on whether to continue.

By round 100 the field had thinned considerably. All the UK cellular radio operators were still in the fray. Worldcom were still there. This would not have surprise anyone in view of their ability, at the time, for raising huge sums of money from the global financial markets. Telefonica were still in They had a bit of cash in the bank play with. ntl was still at the table with their France Telecom partners providing the financial muscle. TIW was a dark horse and there must have been surprise from the other bidders to still see them at the table. Within the space of a few more rounds the list of bidders had thinned to just six.

This was the point in time when the design of the auction is likely to have had an impact on the prices being bid. The DTI decided that one licence should be set aside for a new entrant and this would have more spectrum than the other four licences open to all. This looked to be an act of kindness to the new entrant. In reality it created 4 incumbent players going for 4 licences and 9 new entrants all bidding for one licence.  The result was to push up the price of the new entrant licence to over £4 billion and effectively re-set the price per MHz price expectation for the other licences.

Having carved out a 5th licence having more spectrum the DTI found the remaining spectrum was not exactly divisible by four. The result was that one of the four remaining licences had more spectrum than the other three. This set in train a competition amongst the lead incumbent operators, Vodafone and BT, as to whom was to emerge with more 3G spectrum than the other.  There was more in this than simply corporate pride.  Both Vodafone and Cellnet (BT) had learnt in the 1980’s that the amount of frequency channels they had very tightly constrained the number of customers they could provide a good quality telephone service to.   In other words the DTI had tilted the competitive playing field in the way they packaged up the spectrum. One operator was about to get a structural advantage in the 3G market for the next 15 years.

This accounts for the £1.9 billion premium Vodafone paid to see BT off and seize this larger spectrum package.

At this point an auction round by round analysis gives some clue (although far from conclusive proof) as to who were the aggressive bidders and who were being reluctantly dragged along. Top of the list was Vodafone. They were the lead bidders for 108 rounds. Orange claims the second position by being the lead bidder for 104 rounds. ntl gets to third position with a lead bid position in 91 rounds reflecting  France Telecom strenuous efforts to get into the UK cellular radio market. TIW came next as lead bidder in 80 rounds. Almost at the bottom of the league we find One2One. They were the lead bidders in only 45 of the rounds. Clearly here was one company being dragged along doing the minimum to just emerge with a 3G licence.  The surprise in the lead bid analysis is the position of British Telecom. They were the real laggards. They were the lead bidders in a mere 41 rounds.   This was the least number of lead bids from all those who eventually won a licence. They were evidently not the leaders of the 3G price stampede.

Finally it is the time to look at the last 6 bidders individually:

In the run-up to the auction most observers expected the four existing UK cellular radio operators to win a licence. It was a rational expectation. Like everyone else the frequency channels translated directly to the number of customers they could support with a new service. In addition winning a licence also kept a new competitor out of the market. That was every bit as important in terms of future revenues. They also had a large customer base of existing GSM technology customers ready to hand to load the new 3G networks with, once they were built.

By far the most powerful reason the existing UK cellular radio operators would emerge from the auction with a licence was the anticipated effect on the companies share price of not getting a licence. The Wall Street Journal captured the mood just before the auction when it quoted a Chase Manhattan analyst as saying   “If you lose a 3G license, you’re out of the game for ten years.”  In other words any senior executive of any of the existing UK operators had to expect that the stock markets would massively hammer their share price if they didn’t get a licence. The same stock market that was later to be aghast at the debt they clocked-up to pay the prices they finished up paying.

The fact was that, for the incumbent UK operators in 2000, leaving the UK 3G auction without a licence was simply not an option. The choice for a chief executive was between instant death or carrying a debt that might prove terminal later…the decision was entirely logical (in the context of the craze world of a bubble). The theory of the Treasury economists that those that won a licence would be those that put the most value on the 3G mobile opportunity proved fatally flawed. Against the backdrop of a likely 10% drop in the share price of a large public corporation that the market was valuing at close to $200 billion, a bid of £5.964 billion was not entirely irrational. As will be shown later, this had precious little to do with the real economic benefit flowing from running a 3G mobile business.

TIW is the next bidder worth examining in more detail.  Amongst the potential new entrants to the UK public cellular radio market probably TIW had the most to say in public before the auction. They ran a specialist private mobile radio network aimed at the “fleets of trucks” market.  The service, known a Dolphin, was struggling. A 3G licence would allow TIW to put new life into its UK venture and even make sense of it. Well that was the public plan.

Unbeknownst to all the other bidders at the auction Hutchison Whampoa had cut a deal with TIW just before the start of the auction that had the effect of bankrolling their 3G bid. To get around the auction rules the deal was that TIW, if it won, would immediately lease its entire network to a joint venture with Hutchison Whampoa in which Hutchison would own 90% of the new network and TIW the other 10%. It was a legitimate thing to do but here was a player coming to the table with a secret backer that had recently had the good fortune of a huge windfall of money. When Klaus Esser and his Mannesmann Board launched their unprecedented large bid for Orange they did this by buying out Hutchison. The result was to unleash a chain of events that was to lead to a new entrant coming to the auction supported by a bankroll north of £9 billion.

Orange is the next bidder to look at in more detail. Here a quite extraordinary situation reveals itself. The result of the DTI decision to allow Orange into the auction was to effectively let a player into the auction that was in fact bidding with somebody else’s money. Even more bizarre, the anti collusion rules meant that Orange could not discuss with Vodafone the maximum price that they would pay for a licence. That would have been collusion between bidders. So in effect the party that would have to make out the cheque was not allowed to have a say in just how high Orange could bid. The inside story is as remarkable as it is mysterious. Prior to the auction Orange set up two teams. One was tasked to work out the Orange strategy if they won a licence. They carefully evaluated the opportunity and saw a business to be made that valued the 3G licence at a maximum price of £1 billion. The second group was charged to work out the Orange strategy if they failed to get a 3G licence. They had a capital sum of £1 billion to play with. This Plan B was to upgrade their GSM network to the faster EDGE standard and saturate towns with public WiFi hot spots. It may not have matched the promise held out for 3G but could be delivered far faster. Speed would be of the essence to pre-empt the arrival of 3G. The top management decision was that Orange would go into the auction and if the bidding exceeded £1 billion they would withdraw from the auction and the second team would immediately be brought back together to implement Plan B. The second team watched the 3G Auction break the £1 billion and waiting for the call to assemble. It never came. One of the Engineers put a call in to a colleague in the first team to see what was going on. He replied that they had no idea. An instruction was coming from on high to stay in the auction and bid whatever it took to get a 3G licence. Nobody in Orange had any idea where this instruction had come from. Where could this instruction have come from? That is one of the great mysteries of the UK 3G Auction.

Again the theory of the Treasury economists that those that won a licence would be those that put the most value on the 3G mobile opportunity proved fatally flawed. Here was an entity in the auction actually bidding with somebody else’s money with instructions from a mysterious source to ignore the business case and pay whatever it took to get a 3G licence. 

Mercury One-2-One had been purchased by Telekom of Germany. This was another incumbent former monopoly operator on a buying spree. Having said this the evidence clearly points to One-2-One being dragged along and paying the minimum just to stay in the game.

This leaves ntl. In round 130 they bid £4,277,700,000 for Licence A. This seems far removed from the £2 billion ceiling figure that they, like everyone else, had entered the auction as the top whack they were prepared to pay.

There were three reasons for this escalation under the heat of the auction:

First, France Telecom knew that there were now only two routes it could take to get into the UK cellular radio market. It could buy Orange off Vodafone. Since Mannesmann had paid £19.8 billion for Orange in the first place it was hardly likely that Vodafone would accept any less from France Telecom.  The second route was to get a 3G licence and work with ntl to create the fifth network. Mannesmann had created a wholly inflated point of comparison for the France Telecom Executives sitting in the heat of the 3G auction wondering whether to leave the table.

Second, mobile radio was the missing piece in ntl’s highly successful bundling strategy. ntl were able to offer their customers a combined telephone, TV and Internet service. A mobile offering in the mix could be exceptionally compelling. Barclay Knapp would hang on in at the auction for as long as France Telecom would stump up the finance. So it really came back to France Telecom’s appetite for continuing the fight.

Third was enthusiasm of the bid team itself. With each round of the auction, as the bids ascended, the ntl inner team tasked their KPMG advisor with re-working the business case. Perhaps ntl could expect a bit more revenue here, perhaps the average revenue per customer could be stretched a bit there, perhaps all direct debit could get the bad debt down to almost zero, maybe the churn could be a bit lower….. By the time their bid hit £4 billion every parameter had been taken to its limits of plausibility. All that remained was the profit that might come by selling the network in 15 years time, the so-called “terminal value”. That is what sat behind the final bid of £4.2 billion, an investment that would never pay for itself over the life of the business plan and only justify itself when it was eventually sold to a wealthy investor in 15 years time. Even then every optimistic assumption had to come good at the same time.

Finally France Telecom decided that buying Orange from Vodafone was the better options. In Round 148 ntl withdrew from the 3G auction.

France Telecom later bought Orange for £31 billion. Vodafone paid about £25 billion for Orange itself, it took on about £2 billion of Orange’s debt and paid the £4.1 billion that Orange had paid for its 3G licence in the UK. France Telecom finished-up paying 35 per cent more for Orange than Mannesmann paid, excluding the UMTS third-generation mobile licence. Put another way it paid £6,700 for each Orange subscriber.  Our Orange light user paying his or her £50 per year would now have to live on another 134 years to do his or her bit for France Telecom. France Telecom found itself in 2002 saddled with a debt of more than £40 billion.

France Telecom contribution to Vodafone’s balance sheet, whilst helpful, did not get Vodafone entirely out of the mire affecting the rest of the telecommunications sector. On May 29th2002Vodafone unveiled the biggest loss in UK Corporate history after it wrote down nearly £20 billion relating to its acquisitions. Its net debt was £12 billion.  However, its balance sheet was in much better shape than a number of other companies in the sector.  It had a turn-over of £22.85 billion and free cash flow of £2.4 billion.

As a tale of contrasting strategies MMO2 (formally Cellnet owned by British Telecom) had a net debt of £617 million in mid 2002. Modest compared to Vodafone’s debt of £12 billion. On the other hand MMO2’s turn-over was £2.8 billion compared with the £22.85 billion for Vodafone.  Contrasting positions to finish in 2002 for two companies given a level playing field by the Department of Trade and Industry in 1984.

When Rothschild’s were pressed by the media part way through the auction to justify the massive bids that were being put on the table, they were to patiently explain to the media that these were the worlds most sophisticated companies.   And how!

The above analysis shows that it was Mannessmann that was largely responsible for triggering the mobile asset inflation that was to feed through into the astronomic bids that were made for the 3G licences. The Government and particularly its advisors probably contributed their share to the media hype that made corporate managers fearful of the financial market reaction to coming out of the 3G auction without a 3G licence. The design of the auction itself made a contribution to excessive prices being paid for particular licences. This was more accidental than deliberate. The administration of the auction itself by the Radiocommunications Agency was highly professional.

8. The Consequential Damages

What seems to have been missed entirely by most observers was that the amounts of money paid for the 3G spectrum was a factor but not the main reason for the severe collateral damage to the entire telecommunications sector.

The real damage was done by the Treasury rules on how the spectrum had to be paid for. As with most public institutions, the rules of engagement tend to get set by solid historic precedent rather than speculative future consequences. In particular, the FCC had an experience a few years earlier where companies overbid and then simply defaulted when it came to paying their dues. Bidders for the DTI auction were therefore given two options. The first was to pay the entire amount up-front immediately. The second was to pay half immediately and the other half could be spread over 10 years. The interest rate on the deferred amount was deliberately set at penal levels (from memory it was 3% over commercial rates). This was set deliberately above the amount that could be obtained from the commercial markets. Even then it had to be supported by a “bank guarantee”. This made it a “no-brainer” to borrow the money on the commercial markets and pay the entire money up-front.

The result of the Treasury effort to collect all the money instantly and reduce risk of default was to hit one single industry sector with a £22 billion levy in a single financial year.  This was in part replicated across Europe. It is not possible to take out from a single industry sector in one financial year such a vast financial spike of cost and for there to be no collateral damage. As we will see below it drove the telecommunications sector into a recession.

In fairness to the Treasury they were not to know that the auction was to spin out of control. But their fear of a bidder defaulting and forcing a re-run of the auction led them to applying unreasonable terms.

The financial markets (that must also take some blame for creating the conditions for an asset bubble) were not slow to over react in the other way. The markets applied enormous pressures on the mobile phone operators to reduce the debt on their balance sheets. The easy way for them to do this was to cut back on their capital expenditure programmes. They did this savagely.

The large system vendors went from full order books to nearly empty order books in the space of 12 months and no company, however well run, could adjust to this catastrophic change in market conditions. They then all found themselves in deep financial trouble and the next move was to make redundant thousands of their staff. The sector went into recession. The long term consequence was for the industry to lose two of the finest telecommunications suppliers in its history…Lucent (who were taken over by Alcatel) and Nortel.

The Chairman of US Fed Alan Greenspan was quick to loosen monitory policy to help the stock market recover and shield the wider economy from the fall-out of the dot.com and telecommunications bubbles bursting. The apparent success of this created a complacency amongst financial regulators that asset bubbles did not matter. The warning bell was not heard. The world was to pay a very high price indeed for this when the next asset bubble, this time in the property market, came along. So loose was the monitory policy and out of control leveraging that “the monster” of the unregulated financial markets had grown even bigger…and this time nearly brought the entire global economy to the point of melt-down.