Evaluating spectrum auction design

In the run-up to the next-generation mobile spectrum auction due on April 7, BR Research would analyse a number of auction modalities, related telecom phenomena, and market propositions. However, at this point, uncertainty prevails over market response, and there are questions more than the answers.
Lets start with the auction design. The auction design directly impacts market response, revenue mobilisation and future competitive landscape.
The Information Memorandum (IM), released by the PTA on February 25, shows that a total of 57.38MHz spectrum is on sale: 30MHz for 3G services in 2,100MHz band; 20MHz for 4G services in 1,800MHz band; and 7.38MHz (defunct-Instaphone) in 850MHz band. Only the winner of a 3G license can bid for a 4G license. Only a new entrant can bid for the Instaphone-legacy spectrum.
Its a two-stage auction. Essentially a silent bidding phase, stage 1 would see qualified bidders submit sealed bid offers at base price commensurate with their 3G spectrum demand (minimum 10MHz). If interested, they can also bid for a 4G license. If the bidders spectrum demand is lower than the spectrum on offer, provisional winners will be announced on the base price. Some spectrum will remain unsold.
If the bidders demand exceeds the spectrum on sale, the auction will transit to stage 2 for a computerised, multiple-round, ascending auction. There, simultaneous incremental bids would take the price higher over the base price, with the magnitude depending on bidders appetite, until there are no more bids.
This auction design makes one curious about the likely final price. Theoretically, the government is guaranteed to raise $1.6 billion if the entire spectrum gets sold on base price. Theoretically, a maximum of six licenses (three 3G licenses, two 4G licenses and the Insta-license) are on sale for five existing operators and any number of potential new entrants.
But the final outcome will depend heavily on three factors: level of new entrant participation, existing operators bidding for 3G and additional preference for 4G. If the three factors are positive, the $1.6 billion figure will surely be breached. Expect a big pay day!
However, in a scenario where no new entrants participate and existing operators only bid for 3G, the auction design suggests a gloomy picture. The auction would likely end at stage 1, and the government would raise a maximum of $885 million, selling 3G licenses at a base price of $295 million per 10MHz.
If recent telecom developments are any guide, the actual outlay may fall somewhere in the middle. So far, there is no news, not even a rumour about a potential new entrant--international or local--showing noticeable interest in this auction. But one cannot count out new entrants right now.
Its a mixed picture at home. Warids financial troubles and uncertain future will most likely keep it on the sidelines. So, expect four, not five, existing operators to participate in the auction. But the remaining existing operators are not letting on much right now, holding cards close to their chest. There is naturally more operator interest in 3G over 4G at this stage.
In the past, Telenor has given the impression that it will participate in 3G auction if the price is right. Mobilink may not hold the same view, but don forget that its holding group VimpelComs 43 percent shares are owned by Telenors holding group (source: Reuters, July 2013).
Another problem is that Etisalat, which also controls Ufone, has its own long-running dispute with the government on PTCLs $800 million privatisation settlement and related property transfer. That issue can potentially flare up close to the auction. China Mobile (Zongs sponsor) may, however, aggressively participate, taking this opportunity to better differentiate itself in the market.
April 7 is still a month away. Telecom authorities know too well that the only way to make the auction competitive and successful (in terms of both money and market) is through ensuring high participation from new, international operators. They have less than a month to do that. Good luck!