A fundamental flaw in the Indian 3G spectrum auction designAugust 4, 2010
The Indian telecoms market is divided into 22 telecoms circles and with a population of over a billion each circle has an average population of 52 million which means that the average circle looks like a country somewhere between Italy and South Korea. Indeed many of the larger circles have populations similar to Germany and some are even twice as large. The sheer scale of the Indian mobile market is staggering.
It is not surprising therefore that the auction designers assumed that operators such as Reliance, Bharti and Vodafone would treat the circles as prizes worth winning in isolation and not as part of a pan-Indian spectrum strategy. To put it another way they assumed that the value to an operator of Circle A plus Circle B would be no greater than the sum of the parts.
Where no synergies are assumed to exist, auction designers often opt for a simultaneous multi-round auction or SMRA design where bidders place bids on individual circles in isolation. This design works well provided that there are no synergies or complementarities between circles. However, as the recent comment from Bharti below reveals some operators aspired to a pan-Indian footprint as they clearly anticipated synergies from combining circles.
“The auction format and severe spectrum shortage along with ensuing policy uncertainty drove the prices beyond reasonable levels. As a result, we could not achieve our objective of a pan-India 3G footprint in this round.” Bharti
Those synergies may arise from purchasing scale economies as well as marketing and branding benefits but the important thing from an auction design perspective is that the value of Circle A plus Circle B is greater than the sum of the parts.
When synergies exist a SMRA auction gives rise to what game theorist’s call aggregation risk. Aggregation risk is the risk that arises when you place bids on a combination of individual circles which collectively justify paying a higher price due to synergies than simply the sum of all the prices you would have been prepared to pay for circles based on their stand alone valuations. The risk is that you are then outbid on one of the circles in your desired combination which destroys the value of your synergies but you cannot step back from your other winning bids on the remaining circles. A situation can arise where the bidder ends up having to pay more than the value that he can generate from his sub optimal combination of circles. The situation can also arise where the bidder has to continue to bid on key circles such as Delhi and Mumbai in order to protect the value that is embedded in his bids on other circles – this may well have been the case in the Indian 3G auction.
Had the auction designers taken greater account of the value of a pan-Indian footprint they may well have selected an alternative auction structure such as some form of combinatorial auction. In a combinatorial auction bidders bid for packages of circles and they are guaranteed to receive the whole package of circles or nothing at all. Had a combinatorial approach been adopted Bharti would not have faced aggregation risk and it would have had the opportunity to express with greater confidence the value it placed on a pan-India footprint. A combinatorial approach may not have resulted in the frustrations that they are currently expressing and could have resulted in a more economically optimal allocation of spectrum.