New Zealanders will soon have new, affordable satellite broadband services to choose from.
At the time of writing, there’s no clear threat to urban broadband service providers. It could be another story outside the cities and towns.
The further you are from established networks, the better satellite looks.
Satellite is not new
Satellite broadband has been available here for years. In the past it has been a last resort for people who can’t get a decent connection any other way.
Established services are expensive and need specialist equipment. They’re not easy to use. Even lining up a dish can be troublesome.
When you get a connection, it is slow by today’s broadband standards and has terrible latency.
Earlier satellites use something called a geosynchronous equatorial orbit or GEO. Geosynchronous means they stay in the same position above the earth’s surface.
As the other part of the name suggests, they orbit close to the equator.
For New Zealand users this means your dish need to be able to see the skies to the north. A GEO isn’t much help if you are south of high hills or buildings.
GEO satellites have to orbit at a great height. Which means they are a long way from a dish. Their signals may travel at the speed of light, but long distances mean they have high latency.
In comparison, GEOs have a huge footprint.
Dig the new breed
The new wave satellites are LEOs. They fly in a low earth orbit. This reduces latency.
LEOs offer much more bandwidth than GEOs. The broadband experience is closer to, say, that offered by fixed wireless.
A satellite company can cover a lot of territory with one or two GEO satellites. LEOs have a smaller footprint.
To make LEOs useful, the companies running them need to have large numbers. You need enough so that customers are always covered as they fly past.
At least three companies plan LEO networks that can deliver broadband in New Zealand.
One, Starlink, has published a New Zealand price list and is taking orders. The others include Oneweb, which aims to start services later this year.
Amazon has Project Kuiper which is well behind its rivals, but brings the clout of a tech giant. And don’t underestimate the connection with Amazon Web Services.
Both the European Union and the State Grid Corporation of China plan LEO networks. These may not offer services here.
It starts with Starlink
The most advanced is Starlink. It has 1300 satellites at the time of writing. The company has permission to increase that to 4400. That is around 50 percent more than the total number of satellites currently in our skies.
Starlink plans to charge NZ$160 a month for an uncapped broadband plan. Customers have to buy the company’s earthbound hardware to use the network. That’s an upfront cost of more than NZ$900.
According to Starlink, customers can expect speeds of between 50Mbps and 150Mbps. Let’s take it as read that most users will be closer to the bottom of that range.
There will be times when there is no satellite coverage. Which ruins activities like streaming TV, especially live sports.
Given that uncapped fibre max plans cost around $100 a month, Starlink looks expensive. The proposed charge is high compared with fixed wireless broadband.
If you are a rural user beyond the reach of the mainstream broadband networks, NZ$160 is a bargain. The hardware cost is nothing compared to having fibre connected to a remote property.
While Starlink does not pose an immediate threat to mainstream ISPs, it will be a headache for the innovative Wisps who string local networks up and down remote valleys in rural areas.
Customers in these hard to connect rural areas will be the first targets for LEOs. One way or another, they will upset the Wisps’ business model.
Small local service providers have a get out. They may be able to set up as resellers of satellite services. They have limited, but concrete, options to add value. Margins will be low, but that’s often the way in the telecommunications sector.
Fixing the fixed wireless lottery
Another group of potential early satellite customers are those people on RBI fixed wireless broadband towers who don’t have good coverage.
Rural fixed wireless broadband is a lottery. If you’re on a non-congested tower and you can see the antennae from your property, then the chances are you get decent speeds.
If you are further away, or your tower is congested, fixed wireless speeds can be lousy. For many of these people, in particular, those who need broadband for work, that NZ$160 a month 50mbps uncapped satellite plan looks like a bargain.
It will take a time for those RBI customers to wake up to the charms of satellite broadband, but once word gets out, you can expect an exodus from poor quality fixed wireless connections.
Spark and Vodafone will be the most affected. Losing large numbers of fixed wireless customers can change the economics of operating rural towers.
It also changes the value of the spectrum used to serve these customers. They may be able to put those frequencies to better use elsewhere. It’s not inconceivable that one of the big telcos partners with a satellite company to retain customers if their mobile phone business is valuable.
While the costs and performance stay at NZ$160 a month for 50–150Mbps, satellite won’t make huge inroads in places where there is fibre. Nor is it enticing compared with the better fixed wireless connections.
Yet nothing is fixed. Starlink has said it plans to double speeds over time and to drop rates. And who knows what could happen if there are three competing LEO networks?
A 300Mbps connection for NZ$160 a month starts to look good against the better performing RBI wireless service.
If the price drops to, say, NZ$100 a month, then the satellite companies could eat into Spark, Vodafone and 2degrees fixed wireless profits.
We know the carriers have room to lower their costs. Look at the annual results for the carriers and you’ll see fixed wireless has reasonable margins.
It’s possible New Zealand carriers could find themselves in a price war with global scale satellite ISPs.
That could, over time, have a knock-on effect that reaches into the fibre broadband sector. Paying $100 a month for an uncapped fibre max connection looks good value against today’s fixed wireless plans. This might not always be the case.
A lot has to happen before satellite broadband has any local impact, let alone affect prices for terrestrial services.
The most likely outcome is that satellite will alter the economics of rural broadband and not change much in our towns. Yet, it will add a further layer of competitive pressure.