On July 6, 2026, Bangladesh's Posts and Telecommunications Division approved a request from the Bangladesh Telecommunication Regulatory Commission (BTRC) letting Starlink export unfiltered international bandwidth from Bangladesh into neighboring markets — India's northeastern states, Bhutan, and Nepal — via International Private Leased Circuit (IPLC) links drawn from BSCCL's submarine cable landing stations at Cox's Bazar and Kuakata. It is the first time a regulator has authorized a satellite ISP to re-export terrestrial submarine-cable bandwidth across a land border this way, turning Bangladesh's fibre-optic geography — two of South Asia's few deep-sea cable landing points — into a wholesale on-ramp for a company still negotiating its way into several of the very markets it will now reach from the outside.
The mechanics matter. Starlink is not building new capacity; it is drawing wholesale bandwidth from BSCCL, the state-owned cable operator, under a three-year IPLC supply agreement, with Summit Communications and Fiber@Home as backup suppliers if BSCCL cannot meet demand. BTRC Chairman Md Emdad ul Bari was explicit about the boundary: "The unfiltered bandwidth will not be supplied within the territory of Bangladesh. It will go to other countries." Domestic Bangladeshi retail service — which currently draws roughly 80 Gbps from two licensed gateway operators — is walled off by design; BTRC's conditions require Starlink to submit network diagrams, run real-time traffic-monitoring tools, and give Bangladesh's National Telecommunication Monitoring Centre lawful-interception access and a compliance API before a single export packet moves.
Why Dhaka Said Yes
The case for the deal is straightforward economics. Bangladesh has spare submarine-cable capacity and little market to sell it into beyond its own borders; converting that surplus into a satellite-relayed export product lets state-owned BSCCL earn foreign currency without laying a single new fibre strand. It also fits a pattern BTRC had already signaled: the regulator finalized its NGSO satellite-operator licensing guidelines in March 2025, well before Starlink applied for export rights — meaning this was the next stage of a framework the commission had already built, not an improvised carve-out. For landlocked Nepal and Bhutan, and for India's northeastern states long constrained by distance from India's own cable landing points, a Bangladesh-routed satellite backhaul is a genuine capacity shortcut, bypassing thousands of kilometres of terrestrial haul to reach the coast.
The Steelman: Borders Still Matter for Bytes
The strongest objection isn't paranoia about Musk's politics; it's a coherent regulatory principle. Every market Starlink is about to reach from Bangladesh still requires its own license to sell retail internet service, and for good reason: licensing is how a state fixes accountability for lawful-intercept obligations, content compliance, spectrum coordination, and consumer protection to an entity actually before its courts. Nepal's Telecommunications Authority ordered an immediate halt to unauthorized Starlink use in Himalayan expedition camps as recently as May 2025, citing Section 21 of its 1997 Telecommunications Act, and it still requires any foreign ISP to cede 20% ownership to a local partner before selling service commercially. India, for its part, has already put Starlink through a formal GMPCS licensing process — the Department of Telecommunications lists Starlink Satellite Communication Private Limited alongside Jio and OneWeb as one of just three authorized satellite broadband operators, a process that came with a ₹1 crore entry fee, security conditions, and years of review. A bandwidth pipe reaching a dish in Assam or Kathmandu via a Bangladeshi IPLC, rather than through that domestic licensing track, is — on paper — reaching a market the receiving government hasn't yet formally opened. Regulators who worry that satellite backhaul can outrun the licensing regimes built for terrestrial telecom aren't wrong that the plumbing has changed faster than the rulebook.
Why the Approval Is Still the Right Call
BTRC's conditions actually answer that objection rather than ignore it. The bandwidth is contractually restricted to customers outside Bangladesh's borders, meaning the receiving jurisdictions retain full authority over who is allowed to buy and resell it inside their own territory — Bangladesh is exporting wholesale capacity, not retail access. Nothing in the approval lets Starlink sell a subscription in Guwahati or Kathmandu without that country's own license; it only changes where the underlying bits enter the sky. That's a distinction India's and Nepal's regulators are already equipped to police at their end, the same way they'd police any wholesale capacity purchase from a foreign carrier. Treating a wholesale export deal as equivalent to unlicensed retail entry conflates two different regulatory questions, and risks blocking exactly the kind of infrastructure-sharing that makes remote, high-cost-to-serve geographies — the Seven Sister states, the Himalayan interior — cheaper to connect. If the goal is affordable connectivity in terrain terrestrial fibre struggles to reach, an approval that keeps retail licensing intact while unlocking wholesale capacity is the proportionate outcome, not a loophole.
What to Watch
The real test isn't Dhaka's approval; it's whether India, Nepal, and Bhutan enforce their own retail-licensing lines once cheaper, unfiltered bandwidth is one satellite hop away. Nepal's NTA has already shown it will act against unlicensed Starlink use on the ground. Whether it — and India's DoT — extend that enforcement posture to bandwidth arriving via Bangladesh, rather than treating BTRC's approval as a fait accompli, will determine whether this becomes a template for regional satellite backhaul or a persistent friction point between regulators reading the same deal very differently.