5G auctions must realise technology, market realities

Different generations of mobile technologies have remarkably changed the way we live and communicate.

Second generation (2G) mobile networks, prevalent prior to the millennium, democratised voice telephony; 3G, during the decade of the new millennium, made it possible for users to access the Internet, albeit at rather slow speeds; and 4G made India the global leader in data usage during this decade. Each of these technologies are macro cellular technologies, providing capacity or coverage depending on the frequency bands at which they operate.

However, the future 5G technology needs deployment of small cell access networks using high frequencies to complement the traditional macro cellular deployments. It also needs a robust and high capacity backhaul to complement the high capacity of access networks. The roadmap for public policy must factor in lessons from the past and the current reality.

Policy priorities: The ‘National Broadband Mission’ was launched in 2019 to realise the lofty objective of ‘broadband for all’ enunciated within the National Digital Communications Policy, 2018. Mostly, it would be based on wireless networks, and would therefore need radio spectrum, a natural resource, to be allocated through auctions following the Supreme Court judgment in 2012.

Past performance: The simultaneous multiple round ascending auction, as used in India, has proven to be a robust mechanism for price discovery via a non-discriminatory and transparent process. However, unfortunately, the government sees spectrum sale as a cash cow to bridge the account deficit. Debuting in 2010, spectrum auctions did result in windfalls for the government, half a century after Nobel Laureate Ronald Coase recommended it to the US Federal Communications Commission (FCC) in 1959. Unsurprisingly, telcos resorted to irrational exuberance, vying for the limited slivers of spectrum on offer in each round.

Ironically, most of the winning bids turned into a ‘winner’s curse’, as telcos found it difficult to invest in expansion and upgradation of networks after having emptied their pockets in spectrum auctions. Decreasing ARPUs, continuous pressure to maintain multiple generations of technologies and the prevalent hyper-competition only exacerbated their woes.

During the latest round of spectrum auctions held in 2016, 60 per cent of the 2,300 MHz remained unsold and the 700 MHz band completely was unsold.

Willingness to pay for 5G spectrum: High prices for licensed spectrum arise due to exclusivity of the spectrum rights and the critical nature of the communications involved. Further, the demand for spectrum is derived from the competitive dynamics and level of profitability in the downstream market for services.

It is in this context that the huge possibility of simultaneous use of licensed and unlicensed spectrum changes the nature of competition in that market. A lot of data traffic is already being offloaded to the WiFi networks in homes, offices and increasingly in public places. A much greater use of unlicensed bands is in the offing. Just like 5G, the new WiFi standard for unlicensed use, 802.ax (also called WiFi 6), promises greater speeds and low latency, but requires delicensing of additional bands. In addition, spectrum in the V-Band (57-71 GHz) and E-Band ( GHz) could also be delicensed as already recommended by TRAI. Recently, the FCC decided to allocate 1.2 GHz of spectrum in the 6 GHz band for unlicensed usage.

Imagine two go-carting event organisers who can choose to send their carts on either a personally owned track or on a jointly owned track. In equilibrium, the congestion on each of individual tracks and the jointly owned track must be the same, otherwise customers will switch tracks, disrupting the status quo. However, given that each owner would try to push as many of their customers onto the jointly owned track, it is very unlikely that an equilibrium will be found. Instead, we will see unstable patterns of behaviour that will hurt profitability.

There is certainly increased demand for telecom services due to intense data usage, especially on account of the Covid pandemic. However, the ability of users to bear the proportionate increase in overall charges is in doubt, given the state of the economy.

5G is much more than a communication network: A 5G world will be characterised by M2M, IoT, digitisation of supply chains,value chains, and life in general. Telecom will be part of a greater value network comprising every aspect of the economy. Hence, a more long-term view of the returns to the nation and the society is needed, rather than a narrow focus on the proceeds from spectrum auctions.

Future readiness: To make 5G and other technologies viable for operators and beneficial for the users, we need to accelerate the path to a critical mass of use cases in order to trigger virtuous cycles of adoption. This requires coordination and subsidisation, including subsidisation of spectrum to avoid suboptimal growth in the early stages. For example, an analysis of recent spectrum auctions for 700 MHz in countries such as Denmark, Norway, Switzerland, and Uruguay indicates that the average winning bid price is about $125 million per MHz, whereas the reserve price fixed by TRAI is about $500 million!

Why can’t the auction be conducted without any reserve price, instead of fixing it at very high levels based on prior winning bids? This will, in turn, induce truthful bidding instead of a recurring winner’s curse.

Prasad is Professor at MDI. Maheshwari is Public Policy Consultant and Senior Visiting Fellow at ICRIER, Sridhar is Professor at IIIT-Bangalore. Views are personal

COMMENT NOW