I guess we can agree that the comparison between Sat internet and physical links depends a lot on the physical situation in the target region, and the regulatory frame work.
And please keep in mind that while you are right that there is a risk investing into physical infrastructure also applies to Starlink. It's worth remembering here that all Sat Internet companies prior to Starlink had failed and needed to be rescued with tax payer money.
I don't have exact numbers, and it's a bit muddy due to state subsidiaries, but in Germany the average cost to connect a subscriber in a medium density town with fiber, with given that nothing was prepared and you have to open the street etc appears to be in region of €/$ 2,000 or so.
I don't know if that is done in the US, but also in Europe we now do "trenching". It has some downsides and pitfalls, but this reduces the upfront infrastructure cost for fiber massively.
> while you are right that there is a risk investing into physical infrastructure also applies to Starlink
Absolutely. It's why I think assuming the WACC of a highly-leveraged telecom (around 10%) is appropriate.
> this reduces the upfront infrastructure cost for fiber massively
Fibre makes sense where there is density. It's higher capacity and cheaper. That doesn't mean it makes sense everywhere. And a lot of that everywhere will pay a lot of money for connectivity.
The global telecom market generates trillions of dollars of annual revenue [1]. There is a lot of fruit for the picking.
And please keep in mind that while you are right that there is a risk investing into physical infrastructure also applies to Starlink. It's worth remembering here that all Sat Internet companies prior to Starlink had failed and needed to be rescued with tax payer money.
I don't have exact numbers, and it's a bit muddy due to state subsidiaries, but in Germany the average cost to connect a subscriber in a medium density town with fiber, with given that nothing was prepared and you have to open the street etc appears to be in region of €/$ 2,000 or so.
I don't know if that is done in the US, but also in Europe we now do "trenching". It has some downsides and pitfalls, but this reduces the upfront infrastructure cost for fiber massively.