Starlink won the war for public attention in Ukraine. Japan's three major carriers — KDDI, NTT Docomo, and SoftBank — have all signed partnership agreements. Governments worldwide are treating it as critical infrastructure. But investors should ask a cold-eyed question: can SpaceX actually recover the cost of building and running this network?

The Replacement Treadmill

Starlink currently operates over 6,000 satellites in low-Earth orbit at approximately 550km altitude. The problem is physics. At that altitude, atmospheric drag is significant enough to limit satellite lifespan to roughly 5–7 years. To maintain a constellation of 6,000 satellites, SpaceX needs to replace approximately 1,000 units per year — indefinitely.

A Falcon 9 rocket carries between 20 and 60 Starlink satellites per launch depending on the satellite variant, with a listed price of approximately $67 million per flight. Replacing 1,000 satellites annually requires somewhere between 17 and 50 launches — a range wide enough to matter, and at the lower end alone, roughly $1 billion in launch costs before accounting for satellite manufacturing, ground station operations, and personnel. Total annual operating costs are estimated in the range of several billion dollars.

SpaceX is a private company and does not publish financial statements. Reports of profitability vary widely and cannot be independently verified. For institutional investors evaluating Japanese telecom stocks with Starlink exposure, this opacity is a material risk factor.

The Ghost of Iridium

The satellite communications industry has been here before. In the 1990s, Iridium promised "global connectivity anywhere on Earth" through a constellation of 66 satellites. Commercial service launched in 1998. The company filed for bankruptcy in 1999.

The causes of Iridium's failure are instructive. Handsets were heavy and expensive. Call rates were prohibitive. And the fatal miscalculation: cellular networks expanded so rapidly during the construction period that by the time Iridium launched, the market it had been built for had largely ceased to exist in developed markets.

How does Starlink differ? Three meaningful ways: dramatically lower pricing, a purpose-built terminal that — in its current Direct to Cell variant — can work with existing smartphones, and a clearer addressable market in rural, maritime, and disaster-affected areas where cellular simply does not reach. These are real improvements.

But the fundamental question — can massive fixed infrastructure costs be recovered by sustainable recurring revenue — remains unanswered. Iridium failed on the same question with different surface features.

The Musk Risk Premium

Starlink carries a risk that Iridium never faced: founder concentration risk with geopolitical dimensions.

Musk's reported decision to restrict Starlink access near Crimea demonstrated that a single individual's judgment can determine battlefield communications in an active war. His subsequent role in the U.S. government through DOGE has made the political entanglement more complex, not less. Countries that have built critical infrastructure dependency on Starlink face a scenario their risk frameworks were not designed for: a private operator whose decisions are shaped by the domestic politics of a foreign country.

The European Union's IRIS² constellation program and Japan's defense ministry discussions about "dependency on specific private operators" both reflect the same anxiety.

The Revenue Model: B2C, B2B, B2G

Starlink's revenue comes from three sources: consumer (B2C), enterprise (B2B), and government and defense (B2G).

B2G is currently the most stable. U.S. government subsidies for Ukraine, military contracts, and municipal disaster-preparedness agreements generate revenue that is less price-sensitive and more durable. B2C faces price competition from established carriers in urban markets where cellular works fine. B2B — maritime, aviation, enterprise remote sites — is the most commercially interesting growth vector.

The problem with B2G dependence is political volatility. Contract terms can change with administrations — a risk that has become more concrete since Musk's entry into U.S. government through DOGE, which has simultaneously deepened and complicated the Starlink-Pentagon relationship. Commercial sustainability built primarily on government revenue introduces a structural fragility of a different kind than Iridium faced, but fragility nonetheless.

The Bottom Line for Japanese Investors

The partnerships that KDDI and others have signed are real, and the demand they serve is real. But the underlying infrastructure provider is a black box — financially opaque, politically entangled, and operating on a cost structure that may or may not be sustainable.

Investing in Japanese telecoms on the basis of Starlink upside means implicitly trusting that SpaceX's economics work, that Musk's political positioning remains favorable, and that the constellation can be maintained indefinitely. None of these can currently be verified.

The dream is compelling. The business case deserves more scrutiny than it is currently receiving.

Next: #03 — The Spectrum Problem — and the Case for Laser

Series: - #01 — How the War in Ukraine Changed Everything - #03 — The Spectrum Problem — and the Case for Laser - #04 — Who Wins the Space Race?

Disclaimer | This article is for informational purposes only and does not constitute investment advice. URL: analysis/2026/04/satellite-dream-02-starlink-roi-en/Save_As: analysis/2026/04/satellite-dream-02-starlink-roi-en/index.html