Why Spectrum Bidding Should Be A Mix Of Base Price Plus Revenue Share

The Telecom Commission’s decision to cut the spectrum usage charge (SUC) from five percent to three percent for future auctions is a step in the right direction. The SUC is levied on the adjusted gross revenues of telcos.

With the next round of auctions barely a month away, this relief makes sense, but it does not go far enough. If spectrum is going to be sold to the highest bidder and at very high prices, and if every operator has to re-bid for spectrum every time his licence runs out in any circle (once in 20 years), it hardly makes sense to charge a hefty SUC on available revenues on top of that. There is a case for reducing SUC further to 1 percent or something nominal.

In the next round of auctions, the Telecom Regulatory Authority of India (Trai) has proposed a huge offering of spectrum, including the hitherto unauctioned 700 Mhz spectrum with a hefty base price of Rs 11,485 crore for one Mhz of all-India spectrum. The 700 Mhz band is very useful for mobile broadband and 4G services, which is the future growth trajectory for the mobile internet.

Apart from the 700 Mhz band, also on offer will be large chunks of 800 Mhz, 1,800 Mhz, 2,100 Mhz, 2,300 Mhz and 2,500 Mhz spectrum (plus a bit of 900 Mhz) at prices that could theoretically raise over Rs 5 lakh crore for the exchequer if all of it gets sold, and assuming government sticks to the Trai-recommended base prices.

Given the revenue expectations, one wonders where the annual SUC earning of Rs 7,000 crore really fits into the overall picture.

If all spectrum will have to be sold by auction, where the winner’s curse is a possibility, there is another option before the government: to lower the fixed reserve price to sensible levels, and allow bidding on the basis of share of revenues offered. If this is done, it would mean that small operators will pay high spectrum prices only when their revenues grow, and not before that. Government will get more when the business actually grows, and it may even get better prices for spectrum.

The Delhi and Mumbai airport privatisations under UPA were done not of the basis of cash bids, but share of future revenues. The advantage of bidding based on future revenues, subject to a base price, is that it will not make bids immediately unviable even if the bids are high.

Whichever way the government decides, one thing is clear: trying to push up spectrum charges sky-high could wound the sector before it has reached its true potential. And don’t forget, it is banks who have to finance these bids, and banks are in no condition to take on much higher risks in telecom when they are staggering under bad loans.

Keeping spectrum charges rational is not a favour to telcos; it is a favour to the government itself, as it will spread revenues over years and help banks owned by it from getting into further trouble. Telecom revenues should be considered annuity incomes, and not upfront bonanza.

As we have noted before, the golden goose should be expected to lay eggs regularly; trying to extract all the eggs in one shot is not a great idea.