Satellite internet’s regulatory battle

For most of Pakistan’s history, space and connectivity were kind of separate. Satellite was largely a business-to-business affair, with local players running VSAT [Very Small Aperture Terminal] links, while the Space and Upper Atmosphere Research Commission (Suparco) handled orbital assets under a security chain and the Pakistan Telecommunication Authority (PTA) policed the terrestrial market. The two worlds intersected only at the margins: no consumer satellite market and no foreign constellation selling internet directly to Pakistani households.

The commercialisation of low Earth orbit (LEO) broadband has upended the old order. When Starlink first engaged with PTA in December 2021, there was no framework to govern it since PTA’s categories were mostly terrestrial and Suparco’s mandate was research. To address this, the National Space Policy was approved in December 2023, followed by the Pakistan Space Activities Regulatory Board (PSARB) rules two months later. The PTA’s licensing framework, finalised by April 2026, then made satellite broadband a licence category for the first time.

The new licence allows broadband, backhaul, bandwidth provision, and corporate intranet, while explicitly excluding direct-to-device, mobile satellite service, Earth stations in motion, and broadcasting. That means, at least for now, don’t expect connectivity on a Karachi-Islamabad flight or your phone catching a signal where there’s no tower. To be fair, most markets treat these as separate licence categories rather than lumping them into one permit.

To offer services, every LEO operator obviously has to be locally incorporated and licensed. More critically, they are required to build a gateway earth station on Pakistani soil within 18 months, route all domestic traffic through it, and keep user data in-country.

The final piece remaining is the detailed regulatory framework

By design, a LEO network with inter-satellite laser links has no natural point to tap: traffic can hop from one satellite to another and come down anywhere. Forcing it through a gateway inside Pakistan is what enables interception and content-blocking. This brings us to another critical requirement: the licence requires an interception capability live before launch. In plain English, the operator must build a system that lets law enforcement agencies monitor a specific user’s traffic on request.

Big brother-y as it may be, it’s not exactly novel. Every mobile operator and internet service provider (ISP) in the country is already bound by the same rules; the difference is that terrestrial traffic is already domestic and tappable, whereas LEOs need the gateways and data localisation to enable it in the first place. For context, most countries — be it India, Bangladesh or the US — have identical requirements for satellite operators.

As for the commercial terms, there is a one-time licence fee of $500,000 and about 2.5 per cent of annual gross revenue in recurring charges, which includes spectrum charges and the Universal Service Fund contributions. On top of that, there’s a separate 6pc levy into a Strategic Plans Division-controlled research fund for space development, translating into an aggregate take of roughly 8.5pc, compared with about 4pc in India and 3-5.5pc in Bangladesh.

Moreover, the Fixed Satellite Services explicitly carves out Azad Jammu & Kashmir and Gilgit-Baltistan from the licensed area altogether. Ironically, these happen to be the regions where the case for satellite is quite strong, characterised by weak infrastructure, sparse towers, remote valleys and border settlements where fibre deployment will be tough.

Security is the obvious objection, as these are sensitive areas bordering a hostile neighbour and thus can be tricky places to allow a foreign constellation an open channel. However, we are far from the only country to deal with such a situation. India kept its border areas inside the framework and roped in additional controls instead — monitoring zones within 50km of the border, geo-fenced and location-locked terminals, and the power to suspend service during hostilities.

Currently, the process also looks quite complicated and time-consuming. After incorporation with the Securities and Exchange Commission of Pakistan, the operator must be cleared by the PSARB to utilise Pakistani space and subsequently obtain the relevant operational licence from the PTA, while the Frequency Allocation Board of Pakistan assigns spectrum.

Even more critical is the question of who will administer the rules. Consider the PSARB board, where five out of eight seats are reserved for the security and space establishments — including three Suparco officials and a co-opted Inter-Services Intelligence member — alongside two civilian ministry reps. Meanwhile, private-sector, academic or telecom-industry representation is completely missing.

This potentially creates a major conflict of interest, as Suparco, which runs PSARB’s secretariat, is also the operator of PAKSAT and holds the first right of refusal, ie government bodies must use PAKSAT before seeking connectivity from anyone else. In other words, a competitor has been entrusted as the regulator and is currently in charge of authorising the rules of the game.

Contrast this with India, where licensing was vested in a newly created body, the Indian National Space Promotion and Authorisation Centre, separate from the Indian Space Research Organisation, which gets one seat. An industry figure chairs it and answers up a civilian chain to the prime minister’s office. Bangladesh took an even simpler approach, granting powers to the existing telecom regulator instead of forming a dedicated space authority, and sped up Starlink’s launch in seven months.

Despite this, global operators have shown keen interest in participating. At least four are actively seeking a licence — Starlink, Amazon’s Kuiper and China’s Qianfan among them. Preliminary conversations with them point to a broad willingness to play by the rules even where they chafe: demands such as local gateways, data storage, and lawful interception have generally been accepted in markets like Nigeria and Bangladesh.

The commercial rollout requires the final piece of the jigsaw, ie the detailed regulatory framework by the PSARB. For this, it has engaged the services of the London-headquartered Access Partnership, which submitted its recommendations a while back. But months have passed while the document now bites the dust on the desk of the powers that be. In the meantime, the rest of us are probably busy on our 20 Mbps internet connection.

Abdul Aziz Malik is an ex-analyst at Data Darbar
and Mutaher Khan is co-founder of Data Darbar

Published in Dawn, The Business and Finance Weekly, September 7th, 2026