Four percent and an urban surcharge
Published 11 October 2026
TRAI released its recommendations on May 9 after receiving thirty sets of comments, twelve sets of counter-comments and holding an open-house discussion. For non-geostationary fixed satellite services, it recommended a spectrum charge of four percent of adjusted gross revenue. This was higher than the one-percent-or-less approach Starlink had supported. It was a recommendation to the government, not a notification adopting the charge. TRAI's reasoning included keeping the revenue-share rate aligned with geostationary fixed satellite services using similar bands, encouraging adoption in underserved places and avoiding arbitrage between the two. The recommendation therefore rejected the idea that administrative assignment needed to be priced solely at the cost of administering access. trai.gov.in, Press Release on Satcom_09.05.2025 trai.gov.in, TRAI recommendations, May 2025 trai.gov.in, SpaceX and Starlink counter-comments, November 2024
The regulator nevertheless saw non-geostationary systems as having advantages over geostationary providers, including lower latency, wider coverage and potential economies of scale. It judged that those advantages could give the spectrum greater economic value for the newer operators. TRAI proposed an additional per-subscriber charge as a way to recognize that value and protect a fair competitive position for existing systems. Its discussion also considered the possibility of promotional prices during market entry reducing the revenue on which a percentage charge would be calculated. That was a forecast in the regulator's economic reasoning, not a finding that Starlink had engaged in unlawful pricing in India. The extra charge was proposed before a domestic consumer market for the service had developed. trai.gov.in, TRAI recommendations, May 2025
The proposed addition was five hundred rupees per subscriber per year, restricted to urban areas. Rural and remote subscribers would be exempt. TRAI connected the exemption to the objective of extending reliable access to underserved places. It feared that greater purchasing power and demand in urban areas could draw providers away from that objective. The proposed charge was therefore both a contribution from a commercially stronger segment and an incentive to expand beyond it. The five hundred rupees was not an announced Starlink tariff paid by a household to the company. It was a charge the regulator recommended levying on the service provider for its urban subscriber base. Whether and how a provider would reflect its costs in retail pricing was another question. trai.gov.in, TRAI recommendations, May 2025
The recommendations included a minimum annual charge of three thousand five hundred rupees per megahertz. For non-geostationary fixed satellite service, the comparison was between that minimum and four percent of adjusted gross revenue plus the urban subscriber charge. The greater amount would apply. The minimum mattered because a revenue-share system could otherwise produce a small payment when revenues were low. It also made spectrum bandwidth part of the annual obligation. This was not a per-megahertz auction bid or a one-time purchase price. It was part of a proposed recurring charging structure, set alongside the percentage of revenue and the urban subscriber calculation. Describing the proposal as simply four percent would omit those other elements. trai.gov.in, TRAI recommendations, May 2025
TRAI also specified payment terms. Revenue-based charges would be paid quarterly in advance, within fifteen days of the quarter's beginning. Minimum charges would be paid at assignment and at the start of each year, with adjustments confined to that year's minimum. The urban subscriber component would be one hundred and twenty-five rupees multiplied by the number of urban subscribers at the end of the previous quarter. Those terms translated the annual proposal into a schedule an operator could apply. trai.gov.in, TRAI recommendations, May 2025
User equipment created another affordability problem. TRAI's international comparison put hardware costs for a major non-geostationary operator between two hundred and four hundred and fifty US dollars across the countries examined. The regulator did not present those figures as Indian retail prices. It used them to identify a potential barrier for rural and remote users with limited purchasing power. A network's coverage could become available across a region while the terminal needed to access it remained beyond a household's means. The regulator therefore proposed that the government consider a subsidy for targeted users in unserved or underserved places. The policy question had moved beyond making a signal available to making the equipment accessible to those the signal was intended to reach. trai.gov.in, TRAI recommendations, May 2025
TRAI suggested possible routes through direct benefit transfers or the Digital Bharat Nidhi fund and recommended reviewing any subsidy as technology and market conditions changed. These were options for government consideration. The recommendation did not create an entitlement, assign a subsidy amount or state that every rural customer would receive assistance. Its target was narrower: selected users in places where service was absent or insufficient. That focus distinguished a public-access measure from a general reduction in an operator's business costs. Subsidizing equipment could help a specific customer cross the entry barrier, while lower spectrum charges might affect the economics of the provider as a whole. The two measures addressed different points in the path to adoption. trai.gov.in, TRAI recommendations, May 2025
On duration, the regulator recommended assignments for up to five years, with a possible extension of up to two more after considering market conditions. That was shorter than the long terms many operators had requested. TRAI linked the shorter period to a developing market whose demand and competitive effects would become clearer through operation. It also considered the duration of space authorizations and the useful lives of low Earth orbit satellites. The proposed review window preserved room to change terms as the market matured. Long-term investment certainty was not the only objective. The regulator wanted to observe a new service before committing the charging and assignment framework for a much longer period. trai.gov.in, TRAI recommendations, May 2025 trai.gov.in, SpaceX and Starlink counter-comments, November 2024
Sharing was to come with obligations. TRAI recommended that operators coordinate in good faith in shared higher-frequency bands and that the government retain the power to prescribe a binding framework if needed. It proposed examining power-flux limits and, as a last resort, splitting spectrum when non-geostationary fixed satellite systems failed to coordinate. This addressed the concern that calling a band shared would leave coexistence entirely to private negotiations. The recommendation recognized that those negotiations could fail. It gave a role to technical rules and government intervention without making an auction the default answer for the shared bands. The proposed framework combined permission to coexist with conditions intended to make coexistence workable. trai.gov.in, TRAI recommendations, May 2025
TRAI's examination of foreign rules showed that sharing did not have one universal fallback. Its account of the American framework described a default split when operators failed to coordinate and interference exceeded a six-percent increase in system noise temperature. The affected networks would divide the relevant spectrum while the threshold was exceeded, then could resume use of the full assigned bands once it was no longer exceeded. Selection order depended on the first operating satellite's launch date. The British regulator had taken another view. In the account quoted by TRAI, Ofcom rejected spectrum splitting because it could be inefficient, affect different systems unequally and reduce service quality or competition. The comparison exposed a policy choice inside a shared regime: forcing a division might resolve an immediate conflict, but the division could also reduce what the networks supplied to customers. TRAI was examining that choice, not announcing that the American threshold was automatically India's rule. trai.gov.in, TRAI recommendations, May 2025
The technical discussion explained why coordination depended on geometry as well as the name of a band. TRAI described angular separation as allowing different geostationary systems to use the same frequencies. Non-geostationary constellations instead required coordination techniques as spacecraft moved. During an in-line event, a satellite could lie along the same sight line between another constellation's satellite and an earth station, creating a need for additional mitigation. The discussion also described the international framework's protection of geostationary networks from unacceptable interference by non-geostationary systems, with other coordination provisions applying in bands outside that rule. These were constraints on how systems could coexist, not evidence that all sharing would fail. They made clear why a licence or a price could not answer every practical question about access. Operators needed procedures for particular encounters between links, including encounters that arose only as spacecraft moved into a different alignment. trai.gov.in, TRAI recommendations, May 2025
Gateways received their own recommendations. Operators should coordinate with one another, and DoT should study whether minimum coordination distances between gateways were needed. Sites using certain bands also used by terrestrial mobile systems would require careful location choices, including remote or uninhabited areas where coexistence was more feasible. TRAI proposed that permitted gateways be installed and commissioned within twelve months to reduce the risk of scarce sites being held without use. These conditions concerned the ground network's geography. They showed why satellite access was not a decision made only at the level of a frequency table. The position and operation of equipment on Indian soil could affect interference, coexistence and the ability of other providers to build. trai.gov.in, TRAI recommendations, May 2025
For bands already identified for terrestrial mobile use, the regulator proposed more than a general instruction to place gateways carefully. It named 42.5 to 43.5 gigahertz and suggested case-by-case permission at remote or uninhabited locations where mobile use was less likely to emerge. DoT would prescribe an exclusion-zone requirement for coexistence. TRAI also proposed a portal containing the coordinates of mobile base stations in the relevant ranges. A proposed earth station's geofencing coordinates could then be checked through an automated process to produce a feasibility result. This was an administrative design for making the conflict visible before installation. The recommendation combined a geographical limit, an engineering assessment and a database that could support repeatable decisions about where a gateway was possible. trai.gov.in, TRAI recommendations, May 2025
TRAI addressed an operator's exit as well as its entry. It recommended allowing surrender of spectrum before the assignment period ended, with at least thirty days' notice to the government and regulator. If surrender would disrupt or close service, the notice to authorities and subscribers would be sixty days. A reasonable processing fee could cover administration. This proposal made continuity part of the spectrum framework. The right to stop using a frequency was not treated solely as a transaction between the assignee and government. Customers could depend on the link, so a change in the operator's rights needed notice at the point of use. That concern was different from the initial decision on how the rights were priced. trai.gov.in, TRAI recommendations, May 2025
The regulator also recommended a defined assignment timetable no longer than thirty days after an application, provided the network had received in-principle clearance. If there was an objection, it should be communicated within that window. The purpose was procedural clarity rather than automatic approval. The proposal gave an operator a point at which it should hear either an assignment or what needed attention. TRAI further asked the government to explore assigning spectrum in blocks rather than carrier by carrier. Both suggestions addressed the practical work of administering radio rights once the broad policy choice was made. A framework could support administrative assignment yet still need predictable processes for applications and changes in network operation. trai.gov.in, TRAI recommendations, May 2025
TRAI's minimum-charge rationale was more specific than a general wish to collect revenue. It wanted operators to obtain only the bandwidth they needed, reducing unnecessary coordination burdens and encouraging rollout after assignment. At the same time, it warned that high minimum payments could deter small providers and services used for defence, emergencies, public utilities and remote access. The amount proposed was five percent of an existing seventy-thousand-rupee annual royalty per megahertz, giving the three-thousand-five-hundred-rupee floor. The five percent in that calculation was not five percent of adjusted gross revenue. It was a fraction used to derive the minimum from an earlier per-bandwidth royalty. The two percentages measured different bases and served different roles in the proposed charge. trai.gov.in, TRAI recommendations, May 2025
The recommendations distinguished the frequencies used by fixed services from those used by mobile satellite services. For non-geostationary fixed links carrying data and internet, TRAI proposed considering Ku, Ka and Q/V bands for both user and feeder links, subject to the national frequency plan. For mobile satellite service it listed L and S bands for user links, and C, Ku, Ka and Q/V for feeder links. A household broadband terminal and a mobile satellite device could therefore sit within different radio arrangements even when both carried data. The validity provisions also had two clocks. An individual assignment could last up to five years with a possible two-year extension, while the recommended policy regime's terms were to remain valid for five years from notification, also potentially extended. Later revisions after that period were to apply to existing entities as well. The proposal did not offer a permanent exemption to the first operators entering the market. trai.gov.in, TRAI recommendations, May 2025