Trading on an illusion at the start, but shifting to success

Trading on an illusion at the start, but shifting to success

24 June 2020 | Alan Burkitt-Gray


Born as the dot-com storm raged, Capacity is still serving the industry after 20 years, though it changed direction in its first year. Alan Burkitt-Gray talks to some of the people who were there at the time.

Capacity was launched in 2000, 20 years ago this year, partly as the result of a delusion. But the founders soon realised their mistake and repositioned the business so successfully that it has been able to serve the wholesale telecoms industry for two decades.

The delusion was the myth that minutes of voice traffic were like oil, gas and electricity, so they could be bought and sold by eager young traders on exchanges. Energy trading company Enron went into the voice-minutes market on the same belief.

Capacity started as a bandwidth trading publication and shifted to wholesale telecom when it became apparent that bandwidth trading wasn’t a runner,” says Eira Hayward, the first editor.

Richard Elliott, who then ran a bandwidth trading company called Band-X, confirms: “Yes indeed, the name Capacity referred initially to the trading of capacity, which was the founders’ original focus. They cleverly adjusted the focus post-2001.”

Elliott recalls an encounter with Enron at the time. “They were great commercial traders in oil, and did due diligence on [buying] Band-X and then started up in competition [with us].”

It all changed in 2001. The dot-com crash brought so many telecoms and tech companies down after a period of corporate madness. Before then there had been a surge of proposed mergers: Deutsche Telekom failed to buy Telecom Italia; BT failed to merge with MCI to create a company called Concert; Deutsche Telekom tried to merge with Telefónica.

BT’s MCI bid failed because Worldcom, run by Bernie Ebbers, snatched it from under its nose; Sprint then tried to merge with MCI Worldcom. Meanwhile BT went on to create a joint venture, also called Concert, with AT&T.

“I was at BT,” says Phil Mottram, later of Zayo and now with HPE. He’d been recruited to manage its relationship with internet service providers (ISPs) such as AOL and Uunet. “It was via a dial-up connection and the ISP would get a split of revenue. Then they went free, and it was a split of the interconnection revenue. Growth was tremendous as people started to use the internet more.”

AT&T and BT and the others “thought we would need a global network” for voice, he says. “We did need a global network, but it was the internet.”

Carl Roberts, who was then with GlobalOne, a collaboration between Deutsche Telecom, France Telecom and Sprint, recalls: “The M&A guys were completely out of control.” They measured their success in terms of “how many acquisitions do you make each week?” he says.

“This all led to a glut of assets. Companies were left with a lot of assets sitting there soaking up cost. People built out huge amounts of capacity and there is still lots of dark fibre across Europe not doing anything.”

It wasn’t just in fibre. Roberts moved from GlobalOne to Worldcom to manage international data centres. “So many people had built data centres that the prices went from $3,500-$4,000 per rack per month to $280-$300.”

Murder in Russia

Chris Harper had also worked in GlobalOne but left to help Sprint build its own international network with “my black book of people”. (Earlier he’d worked for NEC, selling radio paging in Russia, where the local mafia moved into the market and murdered people they didn’t get on with.)

He joined a company called Iaxis, which was competing with Colt, Hermes, Level 3 and others to build out across Europe as fast as they could. Iaxis was a re-creation of Telemonde, run by Kevin Maxwell, son of the great publishing fraudster Robert Maxwell, who’d drowned mysteriously in 1991.

Iaxis received takeover offers, including “one for $800 million from Bell Atlantic” – later a constituent of Verizon – but it turned them down. “This was weeks after Carrier1 had an initial public offer (IPO) of its shares.” The Wall Street Journal archive reports: “The IPO of 9.375 million shares was priced at €87 a share ($87.27) raising $818.2 million and giving the company a stock-market value of roughly $3.5 billion.”

Back to Iaxis: “We didn’t have customers,” recalls Harper. “We had what Enron called swaps.” In a swap in that fevered era of the telecoms industry, company A sold $10 million worth of capacity to company B, which sold something similar to company C for the same sort of amount, which in turn sold something back to company A. That made $30 million worth of turnover on the books, though nothing actually happened.

Iaxis crashed, owing money to Ciena and Nortel, its equipment suppliers – and providers of vendor finance – and to the German network operator GasLine.

Hedonistic times

There’s a novelisation of the Iaxis story, Harper mentions at the end of our call. Seven Floors High is by Steve Goddard – clearly a pen-name for someone who worked in the company. According to the blurb on the back cover, it tells the story of “a hedonistic company called Iaxis where everyone is set to become a millionaire”. I read the almost 300 pages on a Sunday afternoon: it’s a better account of those mad times than you’ll find in any book by an MBA. Spoiler: they didn’t become millionaires.

And then in 2001 the tragedy of 9/11 happened. Some of the people murdered at the World Trade Center were from a London conference company, Risk Waters. Fortunately for two former executives, Rachel Jones and Mark Kemp, they had left the year before to nurse their idea of starting a publication about capacity trading.

Hunter Newby, who was CEO of Telx at the time, remembers the time when Enron and others were “very heavily influenced by the energy traders”. Texaco had built over many years a feature called Henry Hub along the shore of the Gulf of Mexico, for natural gas to be collected for distribution.

The hub “existed a long time before gas was traded”, says Newby. But “gas traders with no understanding of how long the infrastructure took to build thought the telecoms industry could do it in months. They were going to fix our market. They thought we were all idiots.”

Incidentally, I’d like to thank Hunter Newby for not only keeping many early copies of Capacity in his loft in New York, but for couriering them over to me in London after our call. After 20 years and at least four office moves, early copies of Capacity are hard to find. All the scans on these pages are of his copies. There might be some in a corner of our Bouverie Street office in central London, but that’s locked down for now.

Camille Mendler, now an analyst at Omdia, was also involved in the early days. She recalls meeting Elliott and the founders “at a café in Kingston upon Thames”, the south-west London suburb where Capacity was based for more than a decade. “They approached me to be editor,” she says. “But I didn’t want to be an editor, so I recommended Eira Hayward.”

The first conference

Capacity’s allegiance to bandwidth trading survived at least until the April 2001 issue, when Jones, as editorial director, reported our first big conference with the words: “Never before have so many carriers talked openly in the same place about bandwidth trading as at Capacity 2001 in Atlanta.”

Until then the incumbents had ruled the telecoms world, by supplying half-circuits on routes such as London to New York. “The idea was a neutral exchange with a fixed margin,” says Elliott, “with bids and offers, such as for 2Mbps from Singapore to Hong Kong.” There had been “a global wave of deregulation. If trading was going to be possible, that was the right time to do it.”

It didn’t stay like that for long. By July/August 2002 Jones was writing about “the stark contrast between the bankruptcy regimes in Europe and the US”. She compared the way European carriers such as Storm and Carrier1 had vanished into receivership with the way US-based Global Crossing, Viatel and McLeod were “allowed months to restructure and reorganise their debt-ridden businesses”.

And the prices had fallen. One company’s infrastructure was sold for five cents on the dollar – just 5% of what it had cost to build – recalls one person in the thick of it at the time.

High margins

But the wholesale telecoms business “had the best year ever” in 2001, says Roberts. “Most of the wholesale business was largely about voice, and the margins were in the mid teens to lower 20s per cent – it was good business.”

Although another person in the industry told me over a lunch several years ago – much wine was consumed – of the night Worldcom nearly lost most of its European connectivity, because it hadn’t paid its bill to Deutsche Telekom. Only a last-minute call to Ebbers, as CEO of Worldcom, saved the company’s European customers from being cut off overnight. (Ebbers later went to jail. He was released in December 2019 after serving half his 25-year sentence, and died in February.)

Why didn’t trading work? Of course, it was different from trading in oil and gas, which you can store in tanks until they’re needed. Yesterday’s voice minutes, if not used, are of even less value than yesterday’s newspapers: they have disappeared into the ether.

Go to jail

Enron closed its broadband arm in mid-2001 and then in December 2001 the whole firm went into bankruptcy because of massive frauds in its energy business. The CEO, Kenneth Lay, was convicted but died of a heart attack before he was sentenced. The COO, Jeffrey Skilling, came out of jail in 2019 after serving 12 years.

The playwright Lucy Prebble wrote a brilliant play about Enron in 2010. It was premiered in London, bombed in New York, but has been revived at least once in the UK; if you ever get a chance, go to see it.

“Band-X survived the crash very well,” recalls Elliott. “Its primary customers were carriers’ carriers.” But then a host of companies such as KPNQwest – a European joint venture between KPN of the Netherlands and Qwest of the US – collapsed. Joe Nacchio, former CEO of Qwest, also went to jail. That was for insider trading, and he served six years.

Meanwhile, the reason for trading in long-distance voice minutes disappeared. The cost of long-haul shrank so much that there was no longer much sense in comparing one supplier with another. People realised that the main cost of a call is the local-access element, fixed or wireless.

Voice is dead, isn’t it?

Christian Michaud, who is now with Tata Communications, had been at Teleglobe in Montréal in the 1990s, but moved to a Canadian IP start-up called Cescom. “That was the first time I heard that voice was dead. It was a booming time,” he says wistfully. “The challenge was to make the technology stable. I was there until the bubble exploded in 2001.” He moved back to Teleglobe after 9/11. “It was restructured in 2002, and five years later was acquired by Tata.”

Roberts adds: “Here we are 20 years later and voice is still going, but IP transit cost is now less than a thousandth of what it was in 2000.”

But, thankfully, Capacity is still here. Newby says: “It’s a testament to Capacity that we’re on this call right now.” In boxes in his attic “there are many other magazines that don’t exist anymore and conferences that don’t exist. There are so many companies that don’t exist.” But he had copies of Capacity, and we’re still here, and he had early Capacity conference programmes.

Capacity has remained throughout as a resource for the industry,” says Newby. “Great articles and great panels at industry events.”