Google's Wireless Auction Play
Think of Google as Garry Kasparov facing Bobby Fischer. Exposing its queen, the Web search giant bids $4.7 billion in a government auction of new wireless spectrum. Google's opponent, most likely Verizon or AT&T, pounces on the queen, raising the bid to corner Google's king.
Checkmate? Actually, this chess game ends in a draw. The high bidders, as yet undisclosed, are required to build an open wireless network allowing all sorts of new devices and software to run over it. And so in losing, Google (GOOG) still achieves its primary goal, revving up competition in the wireless market, without spending a dime.
No D Block Auction?
And yet that doesn't mean the victors in the Federal Communications Commission (FCC) auction will be afflicted with a severe case of buyer's remorse. If AT&T (T) and Verizon (VZ) emerge as the top spenders in the auction, which has already raised a record $19.4 billion even though an entire block of licenses may not sell, the two biggest U.S. mobile carriers will have scored in important ways.
For starters, they will have secured another big chunk of an especially scarce resource while impeding the expansion ambitions of smaller rivals such as Alltel. MetroPCS (PCS), and Leap (LEAP). And despite the record auction proceeds, a quirk in the auction rules may hand AT&T and Verizon a huge bargain on the "C" block licenses, which carry the open-access requirements Google had championed.
The bidding, which began Jan. 24 and is drawing to a close, has been anonymous. But the FCC has disclosed the latest offers several times a day, prompting a parlor game among industry observers to discern the names from the bidding patterns. The current betting among these experts is that this auction is unlikely to produce a new national wireless provider. "My guess is [the auction] will not transform the industry in a fundamental way," says Peter Cramton, an economist at the University of Maryland who was instrumental in designing the first FCC auctions back in the 1990s.
Blame Google's apparent withdrawal and the auction's convoluted rules for that. While FCC Chairman Kevin Martin recently lauded the auction's high gross, Stanford University economist Greg Rosston says "this auction is a total disaster from an auction efficiency standpoint." Rosston had been an adviser to Frontline Wireless, a prospective new carrier that applied to participate in the auction, but apparently failed to raise enough backing and shut down before the bidding began. Frontline had been eyeing the "D" block of spectrum, which the FCC was selling with a burdensome stipulation to provide services for public-safety agencies. That condition appears to have scared away all bidders, with the sole bid coming in below the FCC-set minimum. So barring an eleventh-hour bid, the D block will not be sold in this auction.
Obstacles for Smaller Companies
Critics are pointing to the FCC's unusual slicing of the spectrum, a new approach that had been pushed by Google. In this auction, one license covers the tiny Caribbean island Culebra, seven miles by five miles in size, while another covers the entire Great Lakes region. There's even a nationwide package that was seen as possibly leading to the creation of a new national provider. The FCC has never offered such a hodge podge in a single auction before.
Despite the criticisms, the FCC chairman tells BusinessWeek.com that he's been pleased with the auction's set-up and outcome. "It's just as possible that we were able to increase the overall (bids)…by bringing in some of the new bidders by the way we structured some of our rules," Martin says.
"If you compare the (average bid) for each of the different blocks to what we've raised in past auctions, each block, including the C block, raised more."
Still, there's little doubt that the diverse array of licenses posed problems for smaller companies, as bidders were still subject to an old auction eligibility rule stipulating that they cannot bid for larger licenses as the auction progresses. So, for example, if you placed a bid for Culebra in round one, you couldn't then bid for Great Lakes in round two.
When that rule was implemented years ago, it was intended to help the FCC wrap up auctions featuring more uniform blocks in a timely manner. But applied to odd collection of spectrum licenses offered in this auction, the rule might have hampered buyers like Alltel and Leap, says Rosston.
Rosston speculates that some would-be national providers may have started out bidding for large licenses in the C block, the open-access swath purportedly coveted by Google. But as the auction progressed, these bidders may have switched their focus to smaller licenses in other blocks that, in the early rounds, seemed to be going cheaper. Most bidders expected Google or one of the big phone companies to ultimately snap up the nationwide package of 12 C licenses anyway, leaving bidders for individual C licenses empty handed.
But then Google apparently withdrew and the bidding for C licenses stalled at what looked a tasty bargain. Yet any bidders that might have switched their sights to smaller licenses would have lost their eligibility to reenter the auction for the C block's larger chunks of spectrum. The big phone companies, meanwhile, are believed to have decided it would be cheaper to pursue the dozen C licenses individually rather than all together, supplementing their network capacity in regions rather than nationally.
The eligibility rule "can prevent arbitrage from taking place," says Stanford University economist Paul Milgrom. "It's as if the auctions [of block C and other blocks] were run in sequence."
Were different rules applied, "[this auction] would have likely raised more revenue," says Cramton, who also advised Frontline. Wall Street analyst Thomas Watts of Cowen & Co. figures that Block C alone should have fetched, at going market rates, between $7 billion and $8 billion—far more than the current top bid of $4.74 billion.
The end result: One or several winners will get block C licenses at bargain-basement prices, while several would-be rivals are still duking it out for smaller licenses—many of them now grossly overpriced—in blocks A and B. For example, while the top bid for a B block license covering Seattle and its suburbs had already reached $219 million, a C block covering all of the West Coast and offering nearly twice as much broadcast capacity was selling for $320 million. "You look at this, and you say, 'Something is wrong,'" says Stanford University economist Andrzej Skrzypacz.
So even if a player like Alltel manages to substantially increase its coverage with B licenses, "it's definitely going to cost more money" than it might have, says Rebecca Arbogast, principal at Stifel, Nicolaus & Co. "This will drive up prices for consumers or reduce profits."
Yet some bidders may have simply avoided the C block because of the FCC's complicated open-network requirement. "Any red tape or requirement potentially has a cost," says William Samuelson, a games theorist at Boston University. One uncertainty is whether the FCC will regulate the network-access fees that carriers will charge outside providers of devices and services, and it's those fees "that will determine if open access has any teeth," says Cramton.