The major US telcos were both breaking ranks in different ways last week, Verizon by signing an inflated royalties deal with Broadcom, AT&T by unexpectedly endorsing FCC chair Kevin Martin’s proposals for open access in the 700MHz band ‘ ideas roundly condemned by the other cellcos. But it was Google, once again, that really upped the ante in the increasingly acrimonious debate over the rules for the upcoming auction of this beachfront mobile spectrum, pledging to spend at least $4.6bn on frequencies if the FCC adopts its proposed rules.
From Google’s own point of view, the acquisition of spectrum is a questionable and risky strategy; but in the developing battle between the traditional cellular business model and the internet community’s push to apply its own norms to mobile networks, the search engine giant is a hugely effective champion. The 700MHz auction will be one of the first major tests of how far the developed mobile economies are prepared to accept the internet models that appear to be essential for future growth, but which will disrupt all the existing assumptions about how to make money from wireless. Even the public safety aspect of the issue has become tied up with this more fundamental stand-off, because proposals such as Frontline’s recommend a wholesale, multiple access approach that is supported by the internet players ‘ and as in other trends such as metrozones, public safety requirements can often open a back door to new commercial models (see Wireless Watch June 4 2007).
The basic demands of the wireless radicals are that the rules for the 700MHz auction and subsequent spectrum ownership should favor flexible ownership, wholesaling and open access across at least one-third of the band, in order to encourage more operators with a wide variety of models and applications, and to boost uptake of ‘real’ mobile internet, with unfettered access for consumers and either free or low flat rates, balanced by new commercial mechanisms such as advertising revenue.
The carriers have been predictably hostile to such ideas, especially in the wake of recent comments by Martin that supported a flat rate, open IP architecture with services delivered over the web and devices needing only a cursory adherence to the air interface (see Wireless Watch July 16 2007). They argue that rules that favor entrants do not reflect the investment that the cellcos have put, and will continue to put, into wireless infrastructure, and that they risk allocating valuable spectrum to companies with no prospect of making a commercial success of the networks.
Now AT&T has not only denied speculation that it would lead litigation against the FCC should it adopt Martin’s proposed rules, but has endorsed the Martin plan. “We think Chairman Martin’s plan is a creative compromise that balances the interests of companies and consumers,” Jim Cicconi, AT&T’s senior executive vice president of public policy, told USA Today. The super-carrier is probably reassured that Martin did not go further ‘ his proposals, while liberal by FCC standards, were criticized by Google and others for not pushing open access requirements sufficiently aggressively. For instance, Martin’s ideas fall short of the fully wholesale model advocated by Frontline Wireless, Google and others ‘ this would split spectrum ownership entirely from service delivery and could lead to future innovations such as short lease or ‘on-demand’ access to bandwidth for more flexible operator models than are possible with the current system. Instead, he recommended open access on two 11MHz blocks in the 700MHz band, but these would not have to be wholesaled. His definition of openness is largely centered on the ability to access networks with any device rather than having these dictated by the operator, while the internet lobby also want full transparent roaming across all 700MHz networks, and the end to any walled gardens.
The compromise may have convinced AT&T to support the plan, and implicitly to accept the inevitability of mobile internet evolution and to try to take a proactive role in that. But Verizon Wireless remains intransigent, reiterating its position in a statement that said: “We remain guided in our principled belief that open and fair auctions serve everybody best: consumers, innovators, US taxpayers. To rig the 700MHz auction in any way – to limit its value and potential to deliver exciting new products and services – would be a huge disservice to the nation. Corporate welfare for Google is bad public policy.”
Faced with such opposition from one of the most effective lobbying companies in the US, Google stepped up its own war of words, with the pledge to spend at least $4.6bn on spectrum should the auction rules reflect its own proposals ‘ and so go further than Martin’s current position. Google is not only putting its money where its mouth is, but effectively allaying government fears that an open access auction would deter deep pocketed bidders and therefore raise less money for the treasury. The sum offered exactly reflects the $15bn sum the government expects to raise from the total auction, (since it refers to about one-third of the spectrum – 22MHz out of 60MHz). In a letter to Martin, Google CEO Eric Schmidt said the company is making the financial commitment to ‘to address unsupported claims about any impact from adopting open platforms conditions’.
Google’s conditions were listed in its FCC filing of July 9, the most important being that, on at least 22MHz of the total 60MHz to be auctioned at the turn of the year in the 700MHz band, there should be mandating of:
Open applications, with consumers able to download and use any software applications, content, or services they desire unrestricted by the operator
Open devices – consumers should be able to use whichever communications device and wireless network they prefer
Open services, allowing third parties to acquire wireless services from a 700MHz wholesaler on ‘reasonably non-discriminatory commercial terms’
Open networks, enabling all service providers to be able to interconnect at any technically feasible point in a 700MHz licensee’s network.
Schmidt said Martin’s proposal to attach limited open access to 22MHz in the upper 700MHz band is a step in the right direction, but ‘falls short of including the four tailored and enforceable conditions, with meaningful implementation deadlines, that consumer groups, other companies and Google have sought. In short, when Americans can use the software and handsets of their choice, over open and competitive networks, they win.’
‘We hereby inform you that, should the commission expressly adopt the four license conditions requested in our July 9 letter’with specific, enforceable and enduring rules’Google intends to commit a minimum of $4.6bn to bidding in the upcoming auction,’ stated Schmidt in his letter to Martin.
The CTIA, the association representing the cellular operators, responded immediately, claiming Google was scheming to have the auction ‘rigged’ in its favor and was now offering financial incentives for this. ‘The veil has been lifted,’ said president Steve Largent. “If Google is willing to commit almost $5bn for spectrum that it wants encumbered with various requirements, then let it win that spectrum in a competitive auction and choose that business model. Google and its allies, with their collective market capitalization approaching half a trillion dollars, don’t need a government hand-out at taxpayers’ expense. The competitive wireless industry welcomes all new entrants, but no company should be able to buy a custom-fit government regulation that suits their particular business plan. Consumers should decide if they’re right, not the federal government.”
The battle reprises many of the arguments that were fought between cellcos and internet players when free access, municipally funded metrozones came on the scene a couple of years ago. Operators pleaded unfair competition from publicly funded new operators with open access models; metro network backers argued that without such innovations, the traditional providers would never provide broadband services that were affordable by most of the population and supported new norms of consumer behavior. Both had some points to make, as they do in the 700MHz debate, which is far more serious for the telcos because it concerns licensed spectrum in which very attractive services could potentially be offered ‘ rather than metrozones, which come with all the downsides of license-exempt spectrum, and have largely failed to deliver commercial success.
Open mobile internet leaders such as Intel, Microsoft and Google have long argued that the US will fall behind the world in this important trend if it is left to the telcos to implement next generation networks and services ‘ an argument that has been supported by the slowness of the carriers to bring even fixed broadband to non-premium user bases, and by the increasing shortage of competition to AT&T and Verizon. However, while Intel in particular has considered acquiring spectrum in order to support carriers that support its favored business models – a time-honored tactic used by companies like Qualcomm – Google has certainly gone a step further by demanding that the basic auction rules are written to its liking, rather than using its considerable funds and market weight to push its approach in the open market. The major carriers are facing unprecedented commercial pressure to adapt to internet models, from activities such as those of Sprint and Clearwire in WiMAX, and from the new consumer expectations of open, fast access, largely set by Wi-Fi. So it can realistically be argued, as perhaps it could not five years ago, that the market really can decide this time, provided the auction rules are fair to everyone. The FCC undoubtedly needs to be even tougher than it was in last year’s AWS auction in preventing the telcos twisting the rules to their advantage, but robbing one set of bidders of unfair advantages just to give them to another seems out of step with modern thinking on auctions. And Google, unlike some of the disruptive players of the past, certainly has enough financial clout to compete on a level playing field with the telcos in any bidding war for spectrum.
And if Google does acquire its 22MHz, by fair means or foul, will this be a good thing for US mobile internet? The company has undoubtedly done the open access cause many favors to date, by bringing the issues to the table and exposing many of the weaknesses in the telcos’ traditional positions. Martin may, beneath his rhetoric, remain largely on the telcos’ side ‘ as the AT&T climbdown suggests ‘ but he can no longer ignore the demands of the internet and neutrality brigades. But the fact that Google has belatedly learned to lobby and play poker like a telco does not mean it has also learned to operate in the telco market effectively. If it were to acquire the spectrum, it then has to put into action its mechanism for wholesaling it to myriad small operators, possibly on an on-demand basis (see Wireless Watch May 28 2007) ‘ something which will be technically and commercially complex, and which Google has shown no evidence of developing beyond the theoretical stage. The greatest risk of any strategy that relies on small operators is that, while they may avoid the upfront burden of spectrum acquisition, they still have to create networks and services, and may lack the funds and/or market experience to do this effectively. Google may well take on the network building itself ‘ hence its intense recent interest in wireless network technologies ‘ and so create an internet version of the approach that will dominate all 4G wireless sectors, with network ownership and management separated from service provision (as in most public utilities in developed economies). But not only does the company lack any experience of the business of running a utility ‘ and the experiences of electricity, transport and even fixed broadband operators in many countries highlight the pitfalls it could face ‘ but the desire to do so points to a company that is losing sight of its core business.
Obviously, Google stands to benefit from any major expansion of open internet access, since this will drive greater usage of its services and will almost certainly involve an element of advertising business model, something the search giant is in pole position to dominate. But is controlling access networks really the way to make itself a universal feature of the mobile internet experience? It has been doing a fairly good job of getting its advertising engines and applications on to the wireless networks via partnerships with conventional operators, but those carriers may be more tempted to support challengers to Google should they perceive the company as a competitor. Its fear is that, in the hands of the telcos, new spectrum will not be used to broaden internet access as rapidly as it would like, but the evidence for this is dwindling as the cellcos are forced by the Wi-Fi explosion, by consumer demand and by the decline in their own core service revenues to accelerate their provision of mobile broadband facilities. Google may be making the same mistake that Microsoft made at the turn of the decade, going after total control in order to create new channels for its core technologies, rather than infiltrating them through partnerships and through appealing to consumers. Microsoft invested billions in broadband wireless operators such as NextLink, which proved to have inadequate technology and commercial experience to make a success in their spectrum, and subsequently went bankrupt. Microsoft has proved far more successful in pushing its Windows and MSN platforms and PC-centric vision via support for technologies such as Wi-Fi, and for partners such as Vodafone, than via control of the actual networks.
Google will need powerful friends if it is really to make its wholesale model work through owning the spectrum and network itself. One potential ally it is clearly targeting is Nokia. Chris Sacca, head of special iniatives at the search giant, mentioned Nokia by name as a possible customer for Google’s prospective network. Nokia has struggled to make its presence felt among US cellcos and has pledged improvement in its US position as a key objective for 2007-8. The Finnish company is also moving more rapidly than its handset rivals to embrace the open model, releasing non-cellular devices such as the Internet Tablet, supporting Linux and pronouncing itself a ‘mobile internet’ rather than a handset player. While it clearly sees this as a way to reduce its dependence on cellcos, broaden its revenue base and increase its US market share, it would still be a major step for Nokia to throw in its lot with Google and a host of small operators, despite the added bonus of wrongfooting their mutual enemy Microsoft. Nokia would have to make a more definitive statement than it has to date that it was breaking with the US cellcos and pursuing a new model ‘ possibly a risk worth taking in the US, but one that would have unwelcome impact on its better established relationships with operators in other parts of the world.
Picking on Nokia also seems to contradict one of Google’s basic tenets, open access from any device (which implies that the users choose and therefore fund the CPE, as in Wi-Fi). Nokia’s main interest in a deal with Google would presumably center on the search company guaranteeing it some level of handset sales with its client operators, which could point to Google acting as a central procurement agency for handsets. Such an idea might be more practical in reality than the fully open system Google advocates, at least in the short term as the industry (and the consumer base) makes the painful adjustment to a whole new model of telecoms. It is also more likely, as Sacca admitted, that early customers for the Google spectrum would be existing cellcos such as MetroPCS and Leap, who want a lower cost way to expand their territories, rather than the entirely new breed of players the giant envisages in its more idealistic moments. These flat rate carriers, while undoubtedly disruptive to the major cellcos, are hardly poster children for advanced mobile internet services ‘ but could Google actually be too choosy about the models pursued by its customers? In other words, there will be many compromises along the way, and Google will find itself having to play by many of the rules adopted by the more conventional spectrum holders. Its acquisition of spectrum will put several cats among pigeons, in the US market at least, and may hasten the move towards the fully open world it advocates, but this will be a long and painful process, and in the meantime, there is a very real risk that, in its desire to control that transition, Google will take a commercial hiding.
Google earnings disappoint:
Google’s second quarter profit fell short of expectations, reflecting its massive spending levels, necessary to support its high growth rates. Now that revenue growth is slowing, though, investors are getting uncharacteristically nervous about the much-hyped stock. Google posted a 58% increase in revenue to $3.87bn, driven mainly by its core search and online advertising activities, but rapid staffing increases inflated costs and CEO Eric Schmidt admitted ‘we overspent against our own plan in the area of headcount’. Galloping expenses are often the major bugbear of internet majors as they mature and are forced to expand their business models, but Schmidt said no cost cutting programs were planned. Analysts were disappointed that revenue growth slowed to 58% year-on-year, compared to 63% in the first quarter and 73% for full year 2006. This was mainly down to a slowing of revenue growth from partner sites that carry Google advertising. Google will need to get the shareholders and markets fully back onside if they are to tolerate the massive spending and risk that a major move into wireless operations would entail.