Analysing the 5G spectrum auctions

The success of telecom auctions has been judged only through the lens of spectrum sold, but it has neglected the opportunity cost of the unsold spectrum due to high reserve prices

The ongoing 5G spectrum auctions have quite a stamp of history on them. Those who keep a tab on the sector will know how spectrum was initially bundled with licences and then separated because of the unworthy exploitation of this regime. Subsequently, the bitter political economy associated with administrative spectrum assignment, including the half-baked and bungled first come, first serve method, led the Supreme Court to pronounce auctions as the only transparent assignment tool for the government. The auction is, therefore, here to stay, but its value in the current environment is another story for another day.

That every auction is different is a cliché. Success depends on several factors, including the demand and supply of spectrum, the number of operators, and economic conditions. All these dynamic factors make auctions dissimilar. But is the 5G auction significantly different from the others? Indeed, it is. (Shutterstock)

Have these telecom auctions served us well? Not always. Yes, they reduced corruption in the assignment, but as a revenue-generating mechanism for the State, the outcomes have been mixed. Revenue generation has been an avowed goal for the government, especially because of an unbalanced fiscal situation and the inability to plug the direct tax gap. The success or lack thereof of telecom auctions has been judged only through the lens of spectrum sold, but it has neglected the opportunity cost of the unsold spectrum due to high reserve prices.

The auction regime worked well when demand exceeded supply, either genuinely or artificially. It is revealing that in only one of the seven auctions that have been held, 100% of the spectrum was sold. That was in 2010, when the first spectrum auction was held in India, coming close on the heels of the infamous 2G scam. In two others, 2014 and 2015, significant amounts (around 75% or more) were sold, largely because 2014 and 2015, marked the end of the 20 year-licences for operators and, therefore, business continuity required operators to bid.

That every auction is different is a cliché. Success depends on several factors, including the demand and supply of spectrum, the number of operators, and economic conditions. All these dynamic factors make auctions dissimilar. But is the 5G auction significantly different from the others? Indeed, it is.

This is the first time the 5G spectrum has been put on the block. The spectrum is qualitatively different because it will bring the Internet of things (IoT) to life enabling machines to talk to each other, thereby ushering in a new era of breakthrough innovations across different industry verticals including health, manufacturing, transportation, and the list goes on. The other difference is that non-licensed enterprises have been allowed to bid for 5G to provide private network solutions for specialised areas such as airports, ports, logistics, power, manufacturing, hospitals, and outsourcing units — services that telcos believe are their preserve. There was passionate discourse in the run-up to the auctions to prevent this, but, in the end, it turned out to be a storm in a teacup. Only Adani Data Networks put up its hand with a paltry ₹100 crore as earnest money deposit and has bid merely ₹700 crore, mainly for Gujarat. So, why did it enter the fray? More on this later.

And finally, supply seems far greater than demand in this auction (at the prevailing reserve price). A total of 72 gigahertz (GHz) of spectrum worth ₹4.3 lakh crore in 22 telecom circles at the reserve price is up for sale to four bidders: Jio, Airtel, and Vi (Vodafone and Idea) as licensed telcos, and Adani as the enterprise.

Although it was expected that the auction would be over in two days, it has gone on for four, and counting. But for all practical purposes, it is over, save the ongoing battle between the three telcos in the densely populated UP (E) circle. And while more bids may come, the revenue boost will be limited.

So how should we read this auction?

For a start, almost all spectrum has been sold at its reserve price, indicating how reserve prices have effectively become ‘sale prices’ in a supposed auction, an eventuality that was predicted owing to the steep reserve prices to begin with. However, there seems to be an almost dogmatic inflexibility in the reserve price-setting. The history of unsold spectrum suggests that the reserve price-setting regime is subservient to potential imputations by the Central Bureau of Investigation, Central Vigilance Commission, and Comptroller and Auditor General of India. In this auction, the government stands to realise around Rs. 1.5 lakh crore as opposed to ₹4.3 lakh crore (i.e. about 35%). While the figure is higher than what it had budgeted in its non-tax revenue from spectrum this year (Rs. 75,000 crores) , it begs the question of whether a more rational approach to reserve price setting might have netted more revenue especially given the current ongoing economic scenario.

What can we say about the revealed preference of bidders? Firstly, since all spectrum sold at reserve price, there has been no strategic bidding to raise costs for rivals. Secondly, acquired spectrum for 5G services — in the mid (3.3-3.67GHz), higher (26GHz), and the lower band (700MHz) is reflective of operator strategy. Jio bought the most expensive block of 10 MHz in the 700MHz band in 22 circles (the frequency went unsold in the previous two auctions), in addition to spectrum in the mid-to-high bands. Airtel and Vi have bought 5G spectrum in the mid-and-high bands – Airtel to perhaps compete aggressively with Jio, and Vi to stay in the race for 5G, ostensibly to minimize future losses rather than maximise gains. Except for the purchase of 10 MHz in the 700 MHZ band, none of these has been a surprise, including the sale of spectrum in the low bands for 2/3/4 G services.

The biggest revelation, however, has been the nature of Adani’s bidding. It is clear that the intent is to enter and prepare for a longer and more substantial innings in the Indian telecom market. Not as a fresh licensee, but to either purchase the 33% stake the government has in Vi, and, then finally, take over a willing Vi. Alternatively, the group could save the public sector combine of BSNL/MTNL, similar to what Tata is doing with Air India. Either way, some more churn is afoot in India’s telecom sector.

Rajat Kathuria is dean, School of Humanities and Social Sciences, Shiv Nadar University, and senior visiting professor, Indian Council for Research on International Economic Relations (ICRIER) The views expressed are personal