India’s 3G spectrum auction is, by most counts, a meticulously designed exercise that seeks to serve many competing goals across the equity/efficiency divide.
For instance, almost in a textbook manner, it seeks to discourage collision among players, deter predatory pricing, and ensure efficient price discovery – through simultaneous auction of the country’s 22 operating areas, or circles, over multiple rounds of bidding – so that the public treasury is not shortchanged.
Last but not the least, it seeks to select winners in a transparent manner. This last goal is particularly laudable, given recent allegations of ministerial misconduct in allocation of 2G licenses.
But that said, it is difficult to shake the nagging suspicion that the auction may have been designed more to help the government maximize its revenues than to help the industry. This is because some design elements of the auction seem to carry the potential to push the bidding process into the realm of the irrational.
The government and its auctioneer should actively seek to steer the process away from such an outcome since resulting pathologies could hurt the pace and/or extent of 3G network rollout and, consequently, hamper the rapid and extensive rollout and diffusion of higher-end mobile and broadband services in the country. In short, they could undermine stated policy goals.
The government would do well to trust the competitive dynamics that obtains in the industry, and demonstrate the confidence that this dynamic would be sufficient – as I, indeed, believe it would be – to serve the public exchequer’s interests. It can do so by swiftly tightening the bidding activity requirement, as discussed below.
But before we discuss the design of the auction and its potential consequences, a quick update on the Indian spectrum auction might be in order.
Auction update
By the end of the 8th day and at the end of 52 rounds on Monday, the price for a single pan-India spectrum license stood at INR 6,067 crores (or $1.36 billion) – about 73 percent more than the government’s reserved, or base, price of INR 3500 crores (or $780 million). However, not all of the country’s 22 operating circles being equal, bids have varied markedly across circles.
For the uninitiated: For purposes of telecom regulation, India is divided into operating areas – 3 metros (Delhi, Mumbai, Kolkata), five A circles, eight B circles and six C circles. Each circle is roughly coterminous with a state boundary, with the exception of the North East region, which has several small states clubbed together into a ‘C’ circle. Chennai, once recognized as a separate metro, has been subsumed into Tamil Nadu ‘A’ circle.
Not unexpectedly, Delhi and Mumbai have garnered the highest bids till date – owing, in part, to the 2G spectrum crunch that operators face there and, in part, to the revenue potential of 3G services in these metros. At INR 691.20 crores and INR 635.38 crores, Delhi and Mumbai have garnered bids that are, respectively, 115% and 98% more than the base price of INR 320 crores.
Some circles yet to see bids
At the same time, some circles are yet to see bids while some others have only received lukewarm bids. At the time of writing this column, while a few B circles had begun to see bids, none of the six C circles had received even a single bid – probably because the bidding activity requirement may still have been rather lax.
The bidding activity requirement, which is scheduled to tighten in stages , from 80% to 90% to 100%, is the percentage of eligibility points that a bidder must spend in a given round to qualify for bidding in a subsequent round. The auction concludes when bids have peaked at 100% activity.
By most indications, the bidding activity requirement for the past week has been set at 80%. I believe the bidding requirement will have to be tightened before B and C circles see serious bids. However, as these requirements are tightened, bids for Delhi and Mumbai could easily rise to anywhere between 2.5 to 3 times their base price.
Design and consequence
Before we discuss the design of the auction and its potential consequences (potentially unintended, but not necessarily unanticipated), it might bear reiterating that the current 3G spectrum auction is one of two spectrum auctions slated in India. The second auction, for BWA (broadband wireless access) spectrum, is scheduled to start two days after the completion of the 3G auction currently under way.
Assuming the government is serious about its stated goal of early deployment of 3G and mobile broadband services – and despite often publicly aired inter-ministerial differences over the direction of the industry, there is little reason to believe it is not – we shall term any imposition of excessive economic burden on industry players as an unintended consequence.
The first unintended consequence is that two auction design parameters , their individual merits notwithstanding, when conjoined with strategic intent of players and the transparency of the process, perversely create the opportunity for gamesmanship – in fact, almost invite it.
The second unintended consequence is that the sequential auctioning of 3G and BWA spectrum, given the potential fungibility of 3G and BWA spectrum over time, is likely to result in intense bidding for BWA spectrum – once again leading to expensive outcomes that might undermine the government’s stated policy goal of early broadband deployment in the country. Financially strapped players rarely opt for speedy network deployments.
Potential for gamesmanship
The design parameters that present cause for concern are 1) the bid escalation process and 2) the bidding activity requirement.
Under the auction rules, only the auctioneer can raise bids across rounds, and that, too, in a phased, pre-determined manner, using a publicly stated formula –10 percent when the variance between number of bidders and number of slots in a given circle is three or more, by 5 percent when such variance is two or more, and by 1 percent, when the variance is less than two. The design element is intended to contain irrational exuberance on the part of bidders and, at the same time, limit the opportunity for predatory bids.
The structure of the bidding activity requirement – to be set rather arbitrarily by the government — is intended to promote price discovery while preventing a bidder from waiting till late in the auction to bid on his/her target area/s – in short, it is intended to prevent cherry-picking by players.
Now, consider the following: The auctioneer raises the price for a circle in a given round by say 10 percent. Whatever else that does, it signals to players that there are at least four players bidding on that circle. Now, if the bidding requirement remains lax, a bidder can choose to be a spoiler, and indulge in gamesmanship, raising the spectrum cost for that circle for the eventual winner.
The scope for such gamesmanship and mischief remains particularly alive the longer the government keeps the auction activity requirement at a relatively-lax 80%, since the bidders are not likely to be stretched in their utilization of eligibility points, and can risk acting as spoilers.
True, if there are ‘N’ slots, the gamesmanship of the ‘Nth plus 1’ player is likely to drive the final price in any auction. However, the point being made here is that the government, to avoid sins of omission, should actively refrain from aiding and abetting any opportunity for such gamesmanship. It can do so by tightening the activity requirement sooner than later.
What about the legitimate goal of price discovery, one might ask? Well, we believe that strategic and competitive interests of participants would be sufficient to ensure that. There are, after all, nine players bidding for three nationwide spectrum slots, and at least six of these well-capitalized players have signaled their interest in acquiring a pan-India spectrum slot. The genuinely competitive dynamics, by and of itself, should serve the public treasury well.
The BWA auction
A similar argument should hold for the BWA spectrum action that is to follow. The fact that 11 players are competing for 2 nationwide spectrum slots should be reason enough to assume that the auction will likely invite fairly aggressive bidding. More so, given the fact that many of the operators participating in the 3G spectrum auction are also slated to participate in the BWA spectrum auction.
Given the potential fungibility of spectrum across 3G and BWA – India is auctioning BWA spectrum slots in the 2.3GHz band, instead of the 2.5GHz band – some of the players who may have failed to secure 3G spectrum slots are likely to view BWA spectrum auction as a second lease on life.
These players, driven as much by concerns for their own future as by the desire to keep the BWA spectrum out of the hands of competitors, are likely to make the BWA auction an intense affair. The competition between WIMAX and TD-LTE backers is likely to add its own intensity to the BWA spectrum auction process. I will discuss the WiMAX/TD-LTE issue in India at greater length in the next column.
Competitive dynamics & price of a 3G slot
Six of the nine participants in the 3G spectrum auction – Aircel, Bharti-Airtel, Idea Cellular, Reliance Telecom, Tata Teleservices, and Vodafone-Essar – have signaled their intent to bid for all 22 circles in the country.
While it remains to be seen how far each player might go in pursuing its signaled intent, media reports suggests that two bidders will end up with nationwide 3G spectrum slots – Bharti-Airtel and Vodafone or Bharti-Airtel and Reliance, depending on whom you believe – with the third slot likely shared among the remaining players. That seems unlikely, given the number of players.
A more likely scenario, I believe, would have at most one player, probably Bharti-Airtel, end up with a pan-India 3G spectrum. The remaining two spectrum slots – three in a few circles – would likely be fragmented across the rest of the players, with each players seeking to secure 3G spectrum in circle or circles in which it sees revenue potential and has 2G incumbency.
I believe there is likely to be intense bidding for the top 12-13 circles – the three metros, the five A circles, and the top four or five of the eight B circles – for the revenue opportunity they represent. And, as the auction activity requirement is tightened – from 80% to 90% and then 100% – each player will likely compete more fiercely for 3G spectrum in the circle or circles it deems strategically important.
As players strive to secure circles of interest the bidding war is likely to intensify, raising the price for all. The eventual cost of a pan-India 3G spectrum slot, by my calculations, is likely to be close to $2 billion.
Thoughts on ‘lesser’ circles
While services revenue potential is going to drive 3G spectrum bids for the top 12 or 13 circles, bids for ‘lesser’ circles – the ones with scant revenue opportunity – would likely be driven by industry players’ desire to qualify for monies from the USO (Universal Service Obligation) Fund – currently about $4 billion, by some conservative estimates.
It would be a mistake to assume that the bidding for the lesser circles might be any less intense than that for other circles. The case of the state-floated tender for rural telephony that had to be discarded a couple of years ago might be instructive in this regard.
The government, seeking to stimulate rural telephony, had decided to float a public tender and invite bids. The endeavor was going to be subsidized by USO fund, and the party seeking the least amount from the USO fund was going to be awarded the tender. The ensuing competition was so fierce that bids ran into the negative – that is, operators indicated a willingness to pay the government for the privilege – if, of course, they could lay claim to the USO fund.
As the example indicates, the competitive dynamics in India can be fierce. The government should demonstrate its appreciation of this dynamics and tighten the bidding activity requirement sooner rather than later.
Dr. Shiv Bakhshi is founder and principal analyst of Mobile Perspectives, a research and analysis firm serving the mobile industry. He can be reached at shiv@mobileperspectives.com
Analyst Angle: India's 3G spectrum auctions: design, intent and consequences: Gov't should trust competitive dynamics, tighten bid activity rules
ABOUT AUTHOR
