New Subsea Cables RFS 2025: Unitirreno

Most subsea cables connect countries, but increasingly we are seeing cables that serve only a single country for a variety of reasons.Either the country is not contiguous like Indonesia or the Philippines or the country is exceptionally large with isolated densely populated cities like Australia. Or it is sparsely settled like Alaska's coast which has no significant land infrastructure like roads or gas pipes to serve as telecom rights of way. Unitirreno belongs to the former category. This SDM (spatial division multiplexing) 24 fibre pair cable will link Italy's principal territories, the Boot, Sicily, and Sardinia. The design throughput per fibre pair is 20 terabits or 480 terabits per second for the cable. Unitirreno, if built, will be a very high capacity system. Nearly half a petabit. 

Here is the company's key sales pitch and commercial justifications:

1. Unitirreno cuts the latency in half between Sicily and Genoa and provides a completely diverse path to the terrestrial fibre connecting Sicily and Rome to Northern Italy. 

2. To quote the website, "The strategically planned Unitirreno submarine cable network will ... deliver strategic landing points for intercontinental cables from Asia, the Middle East and Africa while providing a new route to major digital infrastructures across Europe."

Undoubtedly, this design will provide unique physical diversity to the North-South terrestrial routes connecting the country. Moreover, the latency savings will be attractive to Internet backbones and ISPs in general as well as financial market low latency traders. What is not convincing is that this network would be a 'digital bridge' for intercontinental cables. No major intercontinental cables terminate in Sicily or Sardinia. So what is there to connect? 😀 No major continental cable connecting Southeast Asia, India or the Middle East to Europe today terminates its cable in Sicily, Sardinia or Rome and buys the last leg to Europe from a third party.  Major international cables are always built end-to-end because it is cheaper and provides end-to-end control and transparency. The key word is this discussion is 'seamless' and that requires one operator, not two. For example, the main work horses of Asian-Europe traffic are AAE1, SWM5, EIG, and soon SWM6. All of them have a Marseille POP. While it is possible that Unitirreno could market fibre pairs as a protect path for big capacity cables, it is not clear to me there is enough demand to sustain the business. 

There is certainly demand for alternatives to Marseille and Italy is the most logical choice. But doing-it-yourself is likely to be smarter than outsourcing to Unitirreno. If this envisioned cable is non-repeatered, its operating expenses might be low enough to make it viable. But it requires an extremely lean and efficient wholesale operation together with creative deal making. It is worth noting that the company's press releases are a bit vague. There has been apparently progress in securing landing rights with Sparkle for Genoa, the marine survey is complete, and manufacturing is underway. The last item suggests the project has legs and will come to fruition. However, it is not clear if there is sufficient demand for the cable's routes and if its management is sufficiently good to make the project work. Successful telecom wholesale operations require alacrity, pragmatism, and a no-nonsense attitude. It is not compatible with a rules-based corporate culture, constant meetings, and thinking in the box. The bios of the management team do not convince me that Unitirreno has the 'right stuff'. There is no evidence of wholesale experience. Furthermore, an infrastructure fund is backing Unitirreno and these funds do not have a good track record in picking management. They often end up picking senior managers from big carriers with no track record of success in an entrepreneurial enterprise. The sort of entrepreneurial spirit that made Hibernia Atlantic an amazing success (bought for $25 million in bankruptcy with zero customers and sold for over $600 million) is difficult to find.