PARIS — The growth of satellite telecommunications over the past 25 years is the story of government research and development, and technology-demonstration missions, aided by the occasional military system, gradually giving rise to a commercial space business.

There are those who say even in 2014 that there is no such thing as a commercial space market.

Their argument: Satellite builders still depend on government business, mainly but not only military, to help with nonrecurring engineering costs. Launch service providers cannot survive without a reliable government market willing to pay premium launch prices. Satellite fleet operators — the people with the 80 percent earnings before interest, taxes, depreciation and amortization (EBITDA) margins — depend on government-subsidized launchers to grow and maintain their fleets and to keep their already high capital spending from exploding.

All that is true, but a 25-year perspective provides a clear measure of how far the commercial satcom sector has come, most of it on the back of the astonishing growth in demand for television. The television demand in the United States has been enough to sustain two commercial satellite-television providers, DirecTV and Dish Network/EchoStar, which are both expanding into Latin America.

If satellite communications isn’t commercial at this point, it’s a pretty good imitation.

It’s true that launch service provider Sea Launch’s financial troubles were due at least in part to the fact that it operated in international waters with no government customers, only commercial. The launch business remains tricky and those navigating without lots of government support are hard-pressed to make a go of it. How close was SpaceX of Hawthorne, California, to the edge, with founder Elon Musk’s funding about used up, when NASA arrived with a contract?

But satellite builder Space Systems/Loral of Palo Alto, California, after going through Chapter 11 bankruptcy, has emerged as a major commercial-market player despite a relative lack of government business compared to its competitors.

Higher-efficiency transmissions and digital compression notwithstanding, satellite fleets continue to grow when measured in megahertz delivered to customers. The near doubling of the life of the average satellite in the past 25 years and the increased per-satellite capacity have not prevented fleet sizes from swelling.

The two biggest commercial fleet operators, SES and Intelsat, both of Luxembourg, each manage a 50-satellite fleet. Eutelsat of Paris and Telesat of Canada combined operate 50 more. But this 150-satellite quartet’s market share in terms of available bandwidth is now slipping as emerging nations launch their own satellites and established regional operators expand.

Point of reference: One of the first commercial/civil telecommunications satellites scheduled for launch in 2015 will be for Turkmenistan, whose total population is about 5.2 million — the same as Berlin’s.

From today’s vantage point, the time when Intelsat, Eutelsat and Inmarsat were intergovernmental organizations, the latter in mobile satellite services, seems quaint.

In the United States, pioneer Rene Anselmo used much of his fortune to launch PanAmSat-1 — uninsured; there was no money for that — on an untested Ariane 4 rocket in 1988 and sent a warning to Intelsat that the old days were over.

It was the start of a great run for PanAmSat, and for Europe’s Arianespace, which transformed a government research effort into a launcher that became dependent on commercial launches for 90 percent of its business.

In Europe, it was the small SES of Luxembourg that used a loophole in European telecommunications regulations to outmaneuver government-run systems in France and Germany.

If it’s commercial, there must be the occasional failure — and not only creative destruction. Outside of Inmarsat of London, which was privatized along with Intelsat and Eutelsat, the mobile satellite services sector for a time resembled a multi-car accident.

The Iridium and Globalstar constellations emerged from Chapter 11 under new owners willing to bet they could avoid the missteps of the first-generation owners, diversify the customer base and launch second-generation constellations. U.S. projects using S- and L-band spectrum went bust and now survive mainly as free-spectrum tickets for their new owners.

But this is not the 1990s mobile satellite communications market. VSATs, or very small aperture terminals, installed on ships now compete with Inmarsat for market share as commercial maritime fleets demand broadband and commercial airlines do as well.

Peter B. de Selding has been writing for SpaceNews from Paris since 1989.

Peter B. de Selding was the Paris bureau chief for SpaceNews.