Spectrum usage charge has a huge bearing on the operator. It is a perpetual outflow of cash to the government — dampening telecom revenues by approximately 3.5 to 4.5%. This note is an attempt to explain how acquiring a large quantum of millimeter wave spectrum in the 26/28 GHz band can help contain (reduce) this outflow. Also, here we try to analyze the possibility of the government tweaking the rules to prevent this revenue loss. And is this their intent? If yes, what methods it can adopt if any? And will this be against the spirit of the telecom reforms?
Background
On 15th Sept 2021, the cabinet, through a press note, announced a set of reforms for the struggling telecom sector of India. In this press note, the reforms were bucketed under two heads — a) Structural Reforms; b) Procedural Reforms. Point number 7 of the structural reform is reproduced here for easy reference.
“No Spectrum Usage Charge (SUC) for spectrum acquired in future spectrum auctions.”
Later on 8th Oct 2021, DOT came out with an office memorandum to elaborate on this further, which is reproduced below.
“For spectrum acquired in future auctions, no Spectrum Usage Charges (SUC) will be levied. The condition of a minimum 3% weighted average SUC rate and SUC floor amount will also be removed. Guidelines will be issued by DOT to operationalize the decision.”
Note the line highlighted in bold. This is important, else (without this condition/caveat) the cabinet press note on reduction of SUC for future spectrum has no meaning. You will appreciate this as you read on. But since DOT is yet to issue the guidelines, the question that comes naturally to the mind is — Can DOT tinker with the existing rules such that the gains to the operators (as visible by perusing the existing rules) are minimized? And if yes how? To answer these questions we need to first understand the fundamentals.
SUC Concept
Every spectrum band auctioned in the past was encumbered with a levy. This levy was set as a percentage of the operator’s TOTAL revenue (% of AGR), which the operators needed to pay quarterly to the government. Note — This levy is in addition to the auction price. Since every block of spectrum acquired at a different point in time has a different SUC rate, and that too, anchored with different conditions, it has made the SUC formula quite complex and difficult for a layman to understand. I had made an attempt in the past to simply this— “Telecom Reforms — No SUC for future spectrum, but will it reduce outflows?”. Now the real challenge is how to use these diverse rates of SUC, for different blocks of spectrum (acquired at different points in time) to arrive at an integrated value? But even before that, why there is a need for an integrated value? The reason — Since it is technically impossible to slice and map the revenue of the operators to the spectrum blocks acquired at different points of time, therefore, the SUC by definition is a % of the Total Revenue. Hence, these individual SUC rates can’t be operationalized without integration, or in absence of a clearly defined formula.
Current Formula
Fortunately, the DOT has devised a robust weighted average formula to drive the integration of various SUC rates. And the philosophy of which has remained consistent since the beginning. In order words, whatever tinkering that DOT has done on the SUC rates for various spectrum bands acquired at different points of time, it has never tinkered with the principle logic used in the formula. The logic used in the formula is pretty simple. Let's say an operator has different blocks of spectrum (a, b, c, d) acquired at different points of time at different SUC rates (w, x, y, z).
The integrated SUC rate = (a * w + b * x + c * y + d * z)/(a+b+c+d)
Only Caveat — Spectrum in TDD bands need to be halved before embedding into the above formula. Why? When you have 10 MHz of FDD spectrum, you actually have 2x10, and when you have 10 MHz of TDD spectrum, you just have 10 MHz. Hence, before plugging the TDD spectrum value into the above formula we have to divide it by 2.
Proposed Change
The strength of the SUC formula is its simplicity of logic, which is simple to understand and operationalize using just two attributes — a) Dividing the TDD spectrum by 2 as explained above; b) Floor of 3% irrespective of what value emanates out of the formula. In other words, if the formula throws an integrated SUC rate of 2.5%, even then the applicable SUC rate for the operator will get capped at 3%. This is the existing rule. Note that the DOT’s office memorandum dated 8th Oct 2021 has done away with the 3% floor. But Why? The reason — if the floor rate of 3% continues then it will make the cabinet’s decision of charging 0% SUC for future spectrum acquired through auction totally null and void. Let's understand.
Benefit of Reform
We all know in order for the operators to roll out 5G services they will need large blocks of spectrum in various bands (compared to the legacy technologies) to drive high data speeds. Such large spectrum blocks are available only in 3.5 GHz (370 MHz) and millimeter waves (26/28 GHz — 4.25 GHz). Now, assume an operator buys in total 500 MHz of 5G spectrum (100 in 3.5 GHz & 400 in 26 GHz). As per the cabinet press note, these blocks will be attached with 0% SUC. Also, assume that the operator has an existing 50 MHz of legacy spectrum at an integrated SUC rate of 3.5%.
SUC % rate (after 5G auction) = (50 x 3.5% + 250 x 0%) / (50+250) = 0.58%
Now if the government continued with the existing 3% floor, then the operator’s benefit from the telecom reforms would be next to nothing — just just 0.5%, compared to 2.92% without the floor. This is why the floor has been removed now.
Possibility of a Tweak?
In govt anything is possible, and this can’t be totally ruled out. But given the past, the history of the SUC formula, and the decision of the government to do away with the 3% floor, the possibility of a tweak looks extremely difficult if not impossible. The only way to tweak this formula (without breaking the logic) is to attach different weights to different kinds of spectrum bands. Note in the past, the DOT chose not to apply any weights to distinguish one band from the other. For example, the 800 MHz band is better than the 2.5 GHz band in “coverage”, but poor in “capacity”. But the blocks of spectrum in both the bands are treated the same for the purpose of calculation in the SUC formula. So what makes us think that DOT will now decide to treat the spectrum of different bands differently? Even if it wants to attach weights, it will totally arbitrary, as there is no effective scientific basis for implementing such a scheme. And if the intent to prevent revenue loss was paramount, then why did the DOT announce to do away with the SUC floor?
Conclusion
Based on the above logic, I think that as the operators acquire more spectrum in 5G bands, especially in the millimeter-wave spectrum the SUC rates will fall drastically. The reason why the millimeter band is important as it has 11.5 times the quantum compared to the 3.5 GHz band and is priced optimally by TRAI. The 0% SUC rate for the future spectrum is an extremely important reform that the government has decided to implement, and looks like they have thought this through and have understood the implication before making this announcement.
(Views expressed are of my own and do not reflect that of my employer)
PS: Find the list of other relevant articles in the embedded link.
