CHANGE
The 3G auctions were the last party of an old regime.
Of all the sins and excesses of the tech bubble, perhaps the most damaging took place not in Silicon Valley, New York, or Houston, but in a government office in central London. There, on April 27, 2000, the UK auctioned off five licenses for 3G wireless spectrum, raising a whopping $35 billion. Bureaucrats fell over themselves with joy, economists toasted the success of auction theory, and British taxpayers celebrated the historic transfer of wealth to the public purse. Over the next year, a half-dozen other European countries held their own auctions, raising a combined $100 billion in a frenzy of overbidding.
Where did all that money come from? Loans, which have become crushing, industry-killing debt. Now, in what appears to be the ruins of the 3G dream, European wireless business is suffering. Some carriers may go bankrupt, potentially causing Enron-like shock waves. Even if 3G works - so far it hasn't - auction winners can't afford to buy the equipment necessary to roll it out.
The marketplace didn't ask for 3G. European regulators created it to extend the success of GSM to high-speed data. But cheaper routes to always-on dataphones, like Japan's i-mode, have emerged, and as a result 3G may become one of history's greatest failures of industrial policy. Technocrats have blown it before, but the difference between mistake and disaster this time is the crippling, out-of-control auctioning of spectrum.
Economists love auctions. They see them as a market-based (read: fair) alternative to "beauty contests," in which regulators hand out licenses to the most attractive candidates. Auctions are thought to reveal a spectrum's true value since bidders set the price, paid in the form of a charge against future users (who ultimately will bear the cost in higher rates). Because they pay up front, the winners have an incentive to bring services to market quickly as a way to recoup costs. Or so the theory goes.
In the past, auctions were justified by the premise that airwaves can be so easily mucked up by interference that they must be fenced off into regulated monopolies. Such is the case with the analog world. Dumb radio signals need protection because they can't protect themselves; they overlap and are drowned out, like voices in a crowd. But digital signals are defined by the information they contain, not frequency or amplitude. Many transmissions can share the same space, just as unrelated packets share the Internet's fiber arteries. Smart digital radios are adaptive, constantly negotiating between transmitter and receiver to find pockets of clearer air, extending their range.
Think of the difference between an old analog cordless phone and today's digital versions. To keep the old phone from interfering with a neighbor's, you'd select a free channel. Modern phones use digital spread spectrum technology, which allows many calls to share the same frequency. Technology obviates the need for policy.
Most digital cell phones, including those with the proposed 3G standards, work the same way. So why are they being regulated? The usual answer: Even in the digital world, there are limits. Coding may allow signals to share the airwaves, but the bits continue to ride on radio waves. Spectrum remains vulnerable to overexploitation; it can't be shared by everyone all the time. Lacking an easy way to control dynamic sharing, regulators take the road they know best: granting monopolies.
But two technology trends may finally topple this makeshift structure and signal the end of the auction. The first is 802.11b (aka Wi-Fi), which uses spread spectrum in a special unregulated band to send broadband data over a short range. Over the past few years, Moore's law has driven down the cost of Wi-Fi network equipment, allowing the technology to expand from offices and homes into public spaces. Each added node raises the possibility of an ad hoc microcellular network that can achieve much of what 3G has promised - but below the regulatory radar.
The other blow to scarcity-driven policy is ultrawideband, which extends the spread spectrum concept over a wider range of frequencies, enabling more devices to share the airwaves. In February, the FCC approved the first ultrawideband devices. Only very low-power gadgets with short ranges are being allowed on these frequencies initially, to avoid spectrum swarming. But if all goes as expected, the same technology could be given free range.
Both technologies take lessons from Internet economics - favoring pragmatism over paternalism - and neither needs the protection of billion-dollar, government-imposed monopolies. Imagine tiers of service with users paying to avoid congestion rather than being sheltered from it by the FCC.
Technology can even have the effect of expanding available spectrum. As the cost of transmitters drops, networks can be subdivided into smaller and smaller cells. Smart antennas will allocate signals directionally on the fly.
When the telecom world recovers from its hangover, the mania of 3G auctions will be seen as the last party of an old regime. Where airwaves still need protecting (voice, for now), beauty contests may return, rehabilitated, based on successful models in Sweden and Japan. Where they do not, technology will help markets self-regulate. And politicians will just have to go back to the dirty work of balancing budgets and justifying taxes, shedding the occasional tear for the golden calf that got away.
