The Changing Face of Indian Telecom

The Indian Telecom Industry has come a long way. It has transformed the way people communicate. However, the sector is facing contradictions. On the one hand, competition is making services affordable, but on the other hand, the exponential cost of bottleneck resources (i.e spectrum and license fees) is forcing the sector to consolidate - making affordability unsustainable. The technology, however, is now more empowering than ever before. It is capable of driving new services (VoLTE, IOT, M2M), which were virtually incomprehensible just a few years back. Given this situation (things that are in store in near future, i.e with 5G knocking our doors) how should we prepare ourselves to fully capitalize on this opportunity? The current note attempts to investigate this question. In the process, it also tries to visualize the regulatory changes needed to leverage these advanced technologies for the benefit of the consumers in 2018 and in future years.

Affordability

The services can become affordable only if the input costs for producing these services are optimal.

Why is input cost very high?

Outflows on account of buying spectrum and paying license fees are a large component of the overall input cost. The following figure provides a conservative estimate of the outgoing cash flow in past and in the future years.

Please note that in future years the estimated average cash outflow on account of these components alone will be approximately 27% of the total industry revenues (see red line in the graph above). Note that this is a conservative estimate, as no new spectrum bands (700, 3400 etc) have been considered while making these estimates. Also, the price of 900 MHz (which is 1.8 times that of 800 MHz) has been aligned to that of the 800 MHz for driving realistic estimates.

How to optimise spectrum cost?

The input cost around spectrum can be optimized by radically changing the model for calculating reserve prices. Unfortunately, the current model has contributed to raising the reserve prices exponentially in the past auctions. Why? As it (the current model) links reserve prices for future auctions to the previous auction prices, thereby preventing the opportunity of curating distortion in pricing triggered by irrational bidding by the operators. Please read my note titled - What is wrong with spectrum prices?

The proposed model for calculating reserve prices is as under.

The advantages of the proposed model are that it will make reserve prices rationale, i.e more for circles with higher revenue and for spectrum in lower bands (with better propagation characteristics). It will also motivate the bidders to bid responsibly, as future reserve prices will get automatically corrected (reducing pricing volatility), thereby penalizing irresponsible conduct during the auctions by rewarding those acting responsibly, who can buy the same spectrum at lower prices in future auctions. However, the biggest advantage of this model its restores the confidence of the government officials in taking firm decisions. Currently, most officials are afraid of tinkering with spectrum prices due to the fear of being accused of corruption/scam. The simplicity of the proposed formula and the absence of discretion in deciding reserve prices will help dilute that fear.

Quality

It is also imperative that quality of services offered on the ground be significantly increased. But how do we improve quality? We can do so by enhancing the capacity of access and backhaul networks used to carry data traffic.

How do we enhance the capacity of access networks?

This we can do by improving the quality of existing spectrum bands, and by releasing spectrum in new bands (both licensed and unlicensed). The quality of existing bands can be improved through the harmonization process.

In 2016, harmonization of 1800 MHz band enabled all the players to offer 4G services and also enriched the govt by more than Rs 20,000 Cr. Similarly, harmonization of 900 MHz will increase the quantum of spectrum for auction by 12.5 times than what was available in the 2016 auctions at the same time enrich the government by Rs 27,000 Cr (@ 800 MHz prices). A detailed plan to harmonize 900 MHz is available in the embedded link.

New spectrum bands of 700 MHz and 3400 MHz will enable 2x35 MHz and 300 MHz respectively on a Pan-India basis. This will greatly enhance the capacities of access networks for the purpose of deploying 4G/5G technologies.

Releasing unlicensed spectrum in 60 GHz can significantly enhance the capacity of WiFi networks. The current WiFi networks use only 20/40 MHz of spectrum at a time. With 60 GHz spectrum, this quantum can be increased by 50 to 100 times to around 1 GHz. Please note that 60 GHz has total 9000 MHz of spectrum (approximately), but auctioning this band will be of little value (less than half a billion dollar) due to the poor quality of this band due to oxygen molecule interference. Please refer to my earlier note titled - Valuation of 60 GHz spectrum (V Band).

How do we enhance the capacity of backhaul networks?

We can do so by increasing the density of optical fiber networks in the country and by releasing spectrum in E band (70 GHz). OFC will help backhaul traffic emanating out of access networks which are increasingly becoming thicker due to heavy usage of data. The problem will exponentially increase with the deployment of 5G, as the current non-OFC based backhaul networks will be totally incapable of carrying the high volume of 5G data originating out of these access networks. Hence, the government needs to facilitate an optimal cost of right of way for cable laying - a must for deploying OFC networks.

Flexibility

The flexibility needs to be on two accounts. Firstly, it must enhance the service provider's ability to enter into business arrangements with other licensees (VNO, M2M etc). Secondly, they must be fully empowered to offer all kinds of innovative services - a must when 5G gets rolled out.

What is the major barrier towards fostering business partnership?

The key barrier is the increased operational cost triggered by double taxation both on payment of license fees and spectrum usage charges. As on date, the license fees (including spectrum usage charges) is paid at a certain % of the total revenues (approximately 11.6%). Now assume that a VNO collects Rs 100/- from the customer on behalf of the operator. He pays Rs 8/- (@ 8%) as license fees. He passes Rs 90/- to the operator after deducting his contribution towards enabling that service. The operator holding the spectrum now has to pay Rs 10.5 (@ 11.6 % of Rs 90 = 8% LF + 3.6% SUC). Hence, the net outflow to the government is now Rs 18.5 against Rs 11.6 had the setoff being allowed. This is 60% more from the case when the license fee already paid by the VNO partner was allowed to be set off. Please refer to my earlier note titled - How Effective is the VNO Guidelines?

How to prevent double taxation on license fees?

Double taxation can be prevented by creating a model for payment of license fees which will automatically allow the operator to set off license fees already paid by its partner. Before the advent of GST, calculating operator revenue and the applicable value of setoff was a challenging exercise. That is why the consultant hired by DoT in 2003 (for implementing revenue share) advised that no setoff should be allowed. But with the rollout of GST, the situation has changed, i.e revenues and setoff are easy to calculate and should be fully leveraged for the purpose of determining license fees. Doing so will become extremely crucial once the 5G gets rolled out, as most 5G services will be offered by VNOs riding on top of the service provider's networks. Please refer to my earlier note titled - License Fees @ % of Operator Revenues Should Go, But Why?

How to increase the operator's flexibility to offer new services?

This can be done by redefining the rules governing "net neutrality". The principle of net neutrality prevents an operator from tying up with another operator/content provider for providing differentiated services - the packets of which needs to be prioritized over the operator's network in order to make these services meaningful. In this regard, TRAI has recently recommended that the "specialized services" be kept out of the purview of "net neutrality". This is a welcome development. Also, the Feb 8th, 2016 order of TRAI on differential services needs to be amended to include the specialized services. Please refer to my earlier note titled - Analysing TRAI's Recommendation on NN.

Manufacturing

Manufacturing of electronic products in India is extremely crucial, as this will enable India to move up the value chain, conserve precious foreign exchange and create high-quality jobs (R&D related). But the current intensity of manufacturing is very low - resulting in weak investments, a high outflow of foreign exchange and low-quality jobs.

How to increase the intensity of manufacturing in India?

The current policy of manufacturing in India is anchored around PMA (Preferential Market Access) and creating a duty barrier on a selected set of components used for manufacturing handsets. Both approaches have its limitations. Why? Firstly, the PMA is not applicable in the manufacturing of mobile phones (very little government procurement), the turnover of which is significantly high compared to that of network equipment (2.5 times). Secondly, keeping the duties high for a long time is unsustainable, as most components used for manufacturing of mobile phones are covered by ITA (bound rates), thereby likely to get challenged soon. Hence, the best way is to offer a financial incentive in the form of GST refund, as it (GST) is the best measure of value added by the manufacturer. Hence, more the value one adds more will be the financial incentives - resulting in increased manufacturing in the country. This will be easily auditable, as there will be no ambiguity in quantifying the gains realized due to this incentive - preventing the government officials from unnecessary scrutiny. Please refer to my earlier note in this regard titled - Incentivising Manufacturing of Mobile Phones in India.

Conclusion

The advent of new technologies have already changed the face of Indian telecom, and now is the time to overhaul the regulations to support the deployment of these technologies. The old regulations have outlived their utility and now is the time to change them to enable the consumers to fully realize the benefits enabled by these new technologies. It seems that the government is seriously working in this direction. The new NTP (National Telecom Policy), likely to be unveiled in early 2018, can be used as an opportunity for triggering these changes. Once that is done, India is poised for a new era of telecom revolution - driving economic growth and prosperity.

(Views expressed are of my own and do not reflect that of my employer)