[Starlink] SpaceX ordered to explain pricing strategy

[Starlink] SpaceX ordered to explain pricing strategy

[Daniel AJ Sokolov]

Hello,

the Canadian regulatory authority CRTC has ordered SpaceX to reveal how 
its Starlink prices "may change within the next two years".

However, SpaceX will likely file this under seal, meaning it will not 
become public information.

Technically, the order only refers to prices charged in the Far North of 
Canada (The Yukon, the Northwest Territories, Nunavut, Northern 
British-Columbia and one community in Alberta). But as long as 
Starlink's prices are global, this geographical restriction in the order 
is meaningless.

The order is part of CRTC proceeding 8646-N1-202108175, and SpaceX' 
answer is due today, April 8.
Docket at 
https://services.crtc.gc.ca/pub/instances-proceedings/Default-Defaut.aspx?S=C&PA=T&PT=PT1&PST=A
(incomplete due to various 404 errors)

The order to SpaceX has came about after I filed a procedural request in 
this proceeding.

Here is the background:

In most of Canada's Far North, a company named Northwestel has a 
monopoly on landline internet. Also, Northwestel owns the backbone to 
large parts of the area. Northwestel is actually Bell Canada, but with 
much higher rates. It's only been a bit over year that Northerners can 
even buy unlimited internet access, and it is not cheap. (Northwestel 
also has a resale agreement with OneWeb.)

In some areas, one small competitor is trying to hold on: SSi Micro.

They and a few others would want to buy wholesale data transfer from 
Northwestel at regulated prices, so they can mount some competition.

Because Northwestel has a monopoly, they are not allowed to sell 
internet access below cost, and they have to obtain permission from the 
CRTC to change rates. Rates must be "just and reasonable" under the law, 
for whatever that means. The CRTC proceedings to permit rate changes are 
unreasonably slow - a real problem for Northwestel.

However, Northwestel would also love to sell below cost, so they can 
extinguish the little competition they have, and make sure no new 
investor even thinks about entering the market. Northwestel runs a very 
profitable cable TV operation, and they charge business users more than 
double the residential rate for internet access - so they have plenty of 
revenue to cross-subsidize internet, if they would be allowed to do so.

In January, Northwestel applied to the CRTC for permission to change 
this regime. Explicitly, Northwestel wants to be allowed to sell 
residential internet access below cost (cross subsidized from cable TV), 
and to reduce rates or increase data allowances or increase bandwidth at 
any time without another CRTC proceeding.

This, Northwestel argues, is necessary, otherwise Starlink will eat 
Northwestel's lunch. Because Starlink is awesome and cheaper.

Such permission, of course, would be great for consumers in the shortrun 
and awful in the long run. Because it would kill competition.

Most participants in the consultation to Northwestel's application fail 
to understand that. They are jubilant for potentially lower internet rates.

In my filing in February, I asked the CRTC to deny Northwestel's 
application. It is bad policy in the long run.

Also, Northwestel has many options to fight against the (perceived) 
competitive threat from Starlink. Currently, the cheapest unlimited use 
access is a 100 MBit/s down and 12.5 MBit/s up line. They offer plans 
with less bandwidth, but all of those have a usage cap. And overages are 
crazy expensive. It's a topsy-turvy world, where the rich users with fat 
pipes, who can put huge stress on the network, get a free-for-all, 
whereas less affluent users with thin pipes get charged extra per GByte.

In addition, I argued that Starlink does not have the capacity to be a 
real competitor to Northwestel's fat pipes - unless one takes the 
Premium version. Now Starlink Premium is geared at businesses and govs, 
for which Northwestel does NOT ask for permission to lower rates.

Plus other arguments. If you are so inclined, you can find my submission 
in the aforementioned docket under "Interventions".

There, I also pointed out that Starlink's current price point is 
unsustainable, and that they will have to raise prices.

Low and behold, while everyone was waiting for the CRTC's decision on 
Northwestel's application, Starlink increased prices.

So I filed a procedural request to obtain permission to add that 
information to the docket (after the official deadline to add 
Interventions do the docket).

The CRTC has granted my request, added Starlink's price increase to the 
docket, and has ordered SpaceX to explain their pricing plans for the 
next two years by today. Other parties will have until April 18 to 
comment on SpaceX' submission - which may be difficult, because I expect 
all interesting bits to be filed under seal.

Cheers
Daniel

Re: [Starlink] SpaceX ordered to explain pricing strategy

[David Lang]

Why are you so sure that Starlink's current prices are unsustainable?

That's an assertion that requires prove, not just assumed.

David Lang

On Fri, 8 Apr 2022, Daniel AJ Sokolov wrote:

> Date: Fri, 8 Apr 2022 01:59:21 -0700
> From: Daniel AJ Sokolov <daniel@falco.ca>
> To: starlink@lists.bufferbloat.net
> Subject: [Starlink] SpaceX ordered to explain pricing strategy
> 
> Hello,
>
> the Canadian regulatory authority CRTC has ordered SpaceX to reveal how 
> its Starlink prices "may change within the next two years".
>
> However, SpaceX will likely file this under seal, meaning it will not 
> become public information.
>
> Technically, the order only refers to prices charged in the Far North of 
> Canada (The Yukon, the Northwest Territories, Nunavut, Northern 
> British-Columbia and one community in Alberta). But as long as 
> Starlink's prices are global, this geographical restriction in the order 
> is meaningless.
>
> The order is part of CRTC proceeding 8646-N1-202108175, and SpaceX' 
> answer is due today, April 8.
> Docket at 
> https://services.crtc.gc.ca/pub/instances-proceedings/Default-Defaut.aspx?S=C&PA=T&PT=PT1&PST=A
> (incomplete due to various 404 errors)
>
> The order to SpaceX has came about after I filed a procedural request in 
> this proceeding.
>
> Here is the background:
>
> In most of Canada's Far North, a company named Northwestel has a 
> monopoly on landline internet. Also, Northwestel owns the backbone to 
> large parts of the area. Northwestel is actually Bell Canada, but with 
> much higher rates. It's only been a bit over year that Northerners can 
> even buy unlimited internet access, and it is not cheap. (Northwestel 
> also has a resale agreement with OneWeb.)
>
> In some areas, one small competitor is trying to hold on: SSi Micro.
>
> They and a few others would want to buy wholesale data transfer from 
> Northwestel at regulated prices, so they can mount some competition.
>
> Because Northwestel has a monopoly, they are not allowed to sell 
> internet access below cost, and they have to obtain permission from the 
> CRTC to change rates. Rates must be "just and reasonable" under the law, 
> for whatever that means. The CRTC proceedings to permit rate changes are 
> unreasonably slow - a real problem for Northwestel.
>
> However, Northwestel would also love to sell below cost, so they can 
> extinguish the little competition they have, and make sure no new 
> investor even thinks about entering the market. Northwestel runs a very 
> profitable cable TV operation, and they charge business users more than 
> double the residential rate for internet access - so they have plenty of 
> revenue to cross-subsidize internet, if they would be allowed to do so.
>
> In January, Northwestel applied to the CRTC for permission to change 
> this regime. Explicitly, Northwestel wants to be allowed to sell 
> residential internet access below cost (cross subsidized from cable TV), 
> and to reduce rates or increase data allowances or increase bandwidth at 
> any time without another CRTC proceeding.
>
> This, Northwestel argues, is necessary, otherwise Starlink will eat 
> Northwestel's lunch. Because Starlink is awesome and cheaper.
>
> Such permission, of course, would be great for consumers in the shortrun 
> and awful in the long run. Because it would kill competition.
>
> Most participants in the consultation to Northwestel's application fail 
> to understand that. They are jubilant for potentially lower internet rates.
>
> In my filing in February, I asked the CRTC to deny Northwestel's 
> application. It is bad policy in the long run.
>
> Also, Northwestel has many options to fight against the (perceived) 
> competitive threat from Starlink. Currently, the cheapest unlimited use 
> access is a 100 MBit/s down and 12.5 MBit/s up line. They offer plans 
> with less bandwidth, but all of those have a usage cap. And overages are 
> crazy expensive. It's a topsy-turvy world, where the rich users with fat 
> pipes, who can put huge stress on the network, get a free-for-all, 
> whereas less affluent users with thin pipes get charged extra per GByte.
>
> In addition, I argued that Starlink does not have the capacity to be a 
> real competitor to Northwestel's fat pipes - unless one takes the 
> Premium version. Now Starlink Premium is geared at businesses and govs, 
> for which Northwestel does NOT ask for permission to lower rates.
>
> Plus other arguments. If you are so inclined, you can find my submission 
> in the aforementioned docket under "Interventions".
>
> There, I also pointed out that Starlink's current price point is 
> unsustainable, and that they will have to raise prices.
>
> Low and behold, while everyone was waiting for the CRTC's decision on 
> Northwestel's application, Starlink increased prices.
>
> So I filed a procedural request to obtain permission to add that 
> information to the docket (after the official deadline to add 
> Interventions do the docket).
>
> The CRTC has granted my request, added Starlink's price increase to the 
> docket, and has ordered SpaceX to explain their pricing plans for the 
> next two years by today. Other parties will have until April 18 to 
> comment on SpaceX' submission - which may be difficult, because I expect 
> all interesting bits to be filed under seal.
>
> Cheers
> Daniel
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Daniel AJ Sokolov]

To be clear, I am focusing on the consumer product, not the Premium product, which is not available yet.

Mr. Musk himself said last year that he will have to invest an additional USD 20-30 billion to make Starlink survive. After all, he has to build ground stations and launch some 23,000 satellites by the end of the decade to meet his FCC license obligations (not counting any satellites lost to unplanned events, such as a solar flare, warfare, or rocket loss).

If you can finance at 5% (which is optimistic), debt of 10 billion costs you 500 million a year. 

At the same time, the number of clients paying USD99 is limited. The system has limited bandwidth. As the IEEE paper shows, users can expect 25 Mbit/s IF there are no more than 0.1 clients per square kilometre, and IF only 5% of these clients actually use the bandwidth - given a complement of 5040 operational satellites. Currently, 1421 satellites are operational (according to starlink.sx).

The global landmass is about 134 million square kilometres, including all uninhabited areas except Antarctica. The number of humans who can afford USD 100 a month is limited (and don't forget the initial investment, the cost of Dishy's significant power draw, and taxes). Those who are willing to pay all that AND be happy with, say, 30 Mbit/s, is even smaller. And the additional bandwidth per satellite added diminishes as the network grows. You can't double the bandwidth by doubling the satellites, because the available spectrum and the spectral efficiency are given.

At the moment, Starlink revenue is at a runrate of about 25 million dollars a month. A terribly negative cashflow. Yes, there is additional demand, but, like everyone else, Starlink suffers from chip shortages, so they can't make as many terminals as they would like to. And some of the demand they can't fulfill without massively oversubscribing. If there are, say, 10,000 New York City residents on the waiting list, Starlink can't serve them.

In the given setup, churn will be high. When a client moves, there is no guarantee they can keep their Starlink account. When a taller building goes up next door, the connection my be interrupted for good, etc.

Having clients scattered over dozens of countries comes with massive overhead in the legal department. Starlink is just being kicked out of France because their radio license is invalid. I believe they can fight their way back in, but still. This costs time and money. Multiply this with dozens of countries with different legal regulatory regimes, consumer protection laws, tax and filtering/surveillance requirements, etc. 

At 99 dollars is not enough - which is why Starlink had to raise the price. And unless they have tremenduous success with Premium subscriptions, or larger business accounts, they will have to raise prices again.

The good news for Starlink is: They have a captive audience, who has significant sunk cost and often little alternative. So Starlink will be able to raise prices.

BR
Daniel AJ

 

On April 8, 2022 8:15:44 p.m. GMT+02:00, David Lang <david@lang.hm> wrote:
>Why are you so sure that Starlink's current prices are unsustainable?
>
>That's an assertion that requires prove, not just assumed.
>
>David Lang
>
>On Fri, 8 Apr 2022, Daniel AJ Sokolov wrote:
>
>> Date: Fri, 8 Apr 2022 01:59:21 -0700
>> From: Daniel AJ Sokolov <daniel@falco.ca>
>> To: starlink@lists.bufferbloat.net
>> Subject: [Starlink] SpaceX ordered to explain pricing strategy
>> 
>> Hello,
>>
>> the Canadian regulatory authority CRTC has ordered SpaceX to reveal how 
>> its Starlink prices "may change within the next two years".
>>
>> However, SpaceX will likely file this under seal, meaning it will not 
>> become public information.
>>
>> Technically, the order only refers to prices charged in the Far North of 
>> Canada (The Yukon, the Northwest Territories, Nunavut, Northern 
>> British-Columbia and one community in Alberta). But as long as 
>> Starlink's prices are global, this geographical restriction in the order 
>> is meaningless.
>>
>> The order is part of CRTC proceeding 8646-N1-202108175, and SpaceX' 
>> answer is due today, April 8.
>> Docket at 
>> https://services.crtc.gc.ca/pub/instances-proceedings/Default-Defaut.aspx?S=C&PA=T&PT=PT1&PST=A
>> (incomplete due to various 404 errors)
>>
>> The order to SpaceX has came about after I filed a procedural request in 
>> this proceeding.
>>
>> Here is the background:
>>
>> In most of Canada's Far North, a company named Northwestel has a 
>> monopoly on landline internet. Also, Northwestel owns the backbone to 
>> large parts of the area. Northwestel is actually Bell Canada, but with 
>> much higher rates. It's only been a bit over year that Northerners can 
>> even buy unlimited internet access, and it is not cheap. (Northwestel 
>> also has a resale agreement with OneWeb.)
>>
>> In some areas, one small competitor is trying to hold on: SSi Micro.
>>
>> They and a few others would want to buy wholesale data transfer from 
>> Northwestel at regulated prices, so they can mount some competition.
>>
>> Because Northwestel has a monopoly, they are not allowed to sell 
>> internet access below cost, and they have to obtain permission from the 
>> CRTC to change rates. Rates must be "just and reasonable" under the law, 
>> for whatever that means. The CRTC proceedings to permit rate changes are 
>> unreasonably slow - a real problem for Northwestel.
>>
>> However, Northwestel would also love to sell below cost, so they can 
>> extinguish the little competition they have, and make sure no new 
>> investor even thinks about entering the market. Northwestel runs a very 
>> profitable cable TV operation, and they charge business users more than 
>> double the residential rate for internet access - so they have plenty of 
>> revenue to cross-subsidize internet, if they would be allowed to do so.
>>
>> In January, Northwestel applied to the CRTC for permission to change 
>> this regime. Explicitly, Northwestel wants to be allowed to sell 
>> residential internet access below cost (cross subsidized from cable TV), 
>> and to reduce rates or increase data allowances or increase bandwidth at 
>> any time without another CRTC proceeding.
>>
>> This, Northwestel argues, is necessary, otherwise Starlink will eat 
>> Northwestel's lunch. Because Starlink is awesome and cheaper.
>>
>> Such permission, of course, would be great for consumers in the shortrun 
>> and awful in the long run. Because it would kill competition.
>>
>> Most participants in the consultation to Northwestel's application fail 
>> to understand that. They are jubilant for potentially lower internet rates.
>>
>> In my filing in February, I asked the CRTC to deny Northwestel's 
>> application. It is bad policy in the long run.
>>
>> Also, Northwestel has many options to fight against the (perceived) 
>> competitive threat from Starlink. Currently, the cheapest unlimited use 
>> access is a 100 MBit/s down and 12.5 MBit/s up line. They offer plans 
>> with less bandwidth, but all of those have a usage cap. And overages are 
>> crazy expensive. It's a topsy-turvy world, where the rich users with fat 
>> pipes, who can put huge stress on the network, get a free-for-all, 
>> whereas less affluent users with thin pipes get charged extra per GByte.
>>
>> In addition, I argued that Starlink does not have the capacity to be a 
>> real competitor to Northwestel's fat pipes - unless one takes the 
>> Premium version. Now Starlink Premium is geared at businesses and govs, 
>> for which Northwestel does NOT ask for permission to lower rates.
>>
>> Plus other arguments. If you are so inclined, you can find my submission 
>> in the aforementioned docket under "Interventions".
>>
>> There, I also pointed out that Starlink's current price point is 
>> unsustainable, and that they will have to raise prices.
>>
>> Low and behold, while everyone was waiting for the CRTC's decision on 
>> Northwestel's application, Starlink increased prices.
>>
>> So I filed a procedural request to obtain permission to add that 
>> information to the docket (after the official deadline to add 
>> Interventions do the docket).
>>
>> The CRTC has granted my request, added Starlink's price increase to the 
>> docket, and has ordered SpaceX to explain their pricing plans for the 
>> next two years by today. Other parties will have until April 18 to 
>> comment on SpaceX' submission - which may be difficult, because I expect 
>> all interesting bits to be filed under seal.
>>
>> Cheers
>> Daniel
>> _______________________________________________
>> Starlink mailing list
>> Starlink@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/starlink
>>
>

Re: [Starlink] SpaceX ordered to explain pricing strategy

[David Lang]

outside major cities, starlink costs and bandwidth are attractive compared to 
existing options (and frequently cheaper, even including the initial cost over a 
year or so)

your claim that launching additional satellites will not increase bandwith is 
directly coutnered by Starlink's desire to launch additional satellites.

As a private company, we don't know their finances, but we've seen enough to 
know their profit margin on launches is quite high, so it's not a given that 
they are borrowing billions of dollars.

I don't see premium chaning things much, it's 5x the price, but also 5x the 
bandwidth, just in one easy-to-use dish vs configuring load balancing across 5 
dishes. So it looks like a wash to me.

0.1 clients by sq km seems like an incredibly low density. I haven't seen that 
paper, so I can't argue with it's assumptions directly.

I think the number of people who can afford the $100 (or if not individuals, 
then communities sharing a dish) is much higher than you are estimateing. As you 
say there are not a lot of other options.

I agree that the current level of service/pricing is touch-and-go for starlink, 
but with cheaper launches (Starship) and more satellites (and the satellite 
lasers to reduce the need to have a ground station near you)

I've seen people talking about revenue on the order of $30B/year as a 
possibility, and while I think they are probably optomistic, $10B/year on $30B 
in additional investment is a fairly short payback period.

David Lang


On Fri, 8 Apr 2022, Daniel AJ Sokolov wrote:

> Date: Fri, 08 Apr 2022 23:04:04 +0200
> From: Daniel AJ Sokolov <daniel@falco.ca>
> To: David Lang <david@lang.hm>
> Cc: starlink@lists.bufferbloat.net
> Subject: Re: [Starlink] SpaceX ordered to explain pricing strategy
> 
> To be clear, I am focusing on the consumer product, not the Premium product, which is not available yet.
>
> Mr. Musk himself said last year that he will have to invest an additional USD 20-30 billion to make Starlink survive. After all, he has to build ground stations and launch some 23,000 satellites by the end of the decade to meet his FCC license obligations (not counting any satellites lost to unplanned events, such as a solar flare, warfare, or rocket loss).
>
> If you can finance at 5% (which is optimistic), debt of 10 billion costs you 500 million a year.
>
> At the same time, the number of clients paying USD99 is limited. The system has limited bandwidth. As the IEEE paper shows, users can expect 25 Mbit/s IF there are no more than 0.1 clients per square kilometre, and IF only 5% of these clients actually use the bandwidth - given a complement of 5040 operational satellites. Currently, 1421 satellites are operational (according to starlink.sx).
>
> The global landmass is about 134 million square kilometres, including all uninhabited areas except Antarctica. The number of humans who can afford USD 100 a month is limited (and don't forget the initial investment, the cost of Dishy's significant power draw, and taxes). Those who are willing to pay all that AND be happy with, say, 30 Mbit/s, is even smaller. And the additional bandwidth per satellite added diminishes as the network grows. You can't double the bandwidth by doubling the satellites, because the available spectrum and the spectral efficiency are given.
>
> At the moment, Starlink revenue is at a runrate of about 25 million dollars a month. A terribly negative cashflow. Yes, there is additional demand, but, like everyone else, Starlink suffers from chip shortages, so they can't make as many terminals as they would like to. And some of the demand they can't fulfill without massively oversubscribing. If there are, say, 10,000 New York City residents on the waiting list, Starlink can't serve them.
>
> In the given setup, churn will be high. When a client moves, there is no guarantee they can keep their Starlink account. When a taller building goes up next door, the connection my be interrupted for good, etc.
>
> Having clients scattered over dozens of countries comes with massive overhead in the legal department. Starlink is just being kicked out of France because their radio license is invalid. I believe they can fight their way back in, but still. This costs time and money. Multiply this with dozens of countries with different legal regulatory regimes, consumer protection laws, tax and filtering/surveillance requirements, etc.
>
> At 99 dollars is not enough - which is why Starlink had to raise the price. And unless they have tremenduous success with Premium subscriptions, or larger business accounts, they will have to raise prices again.
>
> The good news for Starlink is: They have a captive audience, who has significant sunk cost and often little alternative. So Starlink will be able to raise prices.
>
> BR
> Daniel AJ
>
>
>
> On April 8, 2022 8:15:44 p.m. GMT+02:00, David Lang <david@lang.hm> wrote:
>> Why are you so sure that Starlink's current prices are unsustainable?
>>
>> That's an assertion that requires prove, not just assumed.
>>
>> David Lang
>>
>> On Fri, 8 Apr 2022, Daniel AJ Sokolov wrote:
>>
>>> Date: Fri, 8 Apr 2022 01:59:21 -0700
>>> From: Daniel AJ Sokolov <daniel@falco.ca>
>>> To: starlink@lists.bufferbloat.net
>>> Subject: [Starlink] SpaceX ordered to explain pricing strategy
>>>
>>> Hello,
>>>
>>> the Canadian regulatory authority CRTC has ordered SpaceX to reveal how
>>> its Starlink prices "may change within the next two years".
>>>
>>> However, SpaceX will likely file this under seal, meaning it will not
>>> become public information.
>>>
>>> Technically, the order only refers to prices charged in the Far North of
>>> Canada (The Yukon, the Northwest Territories, Nunavut, Northern
>>> British-Columbia and one community in Alberta). But as long as
>>> Starlink's prices are global, this geographical restriction in the order
>>> is meaningless.
>>>
>>> The order is part of CRTC proceeding 8646-N1-202108175, and SpaceX'
>>> answer is due today, April 8.
>>> Docket at
>>> https://services.crtc.gc.ca/pub/instances-proceedings/Default-Defaut.aspx?S=C&PA=T&PT=PT1&PST=A
>>> (incomplete due to various 404 errors)
>>>
>>> The order to SpaceX has came about after I filed a procedural request in
>>> this proceeding.
>>>
>>> Here is the background:
>>>
>>> In most of Canada's Far North, a company named Northwestel has a
>>> monopoly on landline internet. Also, Northwestel owns the backbone to
>>> large parts of the area. Northwestel is actually Bell Canada, but with
>>> much higher rates. It's only been a bit over year that Northerners can
>>> even buy unlimited internet access, and it is not cheap. (Northwestel
>>> also has a resale agreement with OneWeb.)
>>>
>>> In some areas, one small competitor is trying to hold on: SSi Micro.
>>>
>>> They and a few others would want to buy wholesale data transfer from
>>> Northwestel at regulated prices, so they can mount some competition.
>>>
>>> Because Northwestel has a monopoly, they are not allowed to sell
>>> internet access below cost, and they have to obtain permission from the
>>> CRTC to change rates. Rates must be "just and reasonable" under the law,
>>> for whatever that means. The CRTC proceedings to permit rate changes are
>>> unreasonably slow - a real problem for Northwestel.
>>>
>>> However, Northwestel would also love to sell below cost, so they can
>>> extinguish the little competition they have, and make sure no new
>>> investor even thinks about entering the market. Northwestel runs a very
>>> profitable cable TV operation, and they charge business users more than
>>> double the residential rate for internet access - so they have plenty of
>>> revenue to cross-subsidize internet, if they would be allowed to do so.
>>>
>>> In January, Northwestel applied to the CRTC for permission to change
>>> this regime. Explicitly, Northwestel wants to be allowed to sell
>>> residential internet access below cost (cross subsidized from cable TV),
>>> and to reduce rates or increase data allowances or increase bandwidth at
>>> any time without another CRTC proceeding.
>>>
>>> This, Northwestel argues, is necessary, otherwise Starlink will eat
>>> Northwestel's lunch. Because Starlink is awesome and cheaper.
>>>
>>> Such permission, of course, would be great for consumers in the shortrun
>>> and awful in the long run. Because it would kill competition.
>>>
>>> Most participants in the consultation to Northwestel's application fail
>>> to understand that. They are jubilant for potentially lower internet rates.
>>>
>>> In my filing in February, I asked the CRTC to deny Northwestel's
>>> application. It is bad policy in the long run.
>>>
>>> Also, Northwestel has many options to fight against the (perceived)
>>> competitive threat from Starlink. Currently, the cheapest unlimited use
>>> access is a 100 MBit/s down and 12.5 MBit/s up line. They offer plans
>>> with less bandwidth, but all of those have a usage cap. And overages are
>>> crazy expensive. It's a topsy-turvy world, where the rich users with fat
>>> pipes, who can put huge stress on the network, get a free-for-all,
>>> whereas less affluent users with thin pipes get charged extra per GByte.
>>>
>>> In addition, I argued that Starlink does not have the capacity to be a
>>> real competitor to Northwestel's fat pipes - unless one takes the
>>> Premium version. Now Starlink Premium is geared at businesses and govs,
>>> for which Northwestel does NOT ask for permission to lower rates.
>>>
>>> Plus other arguments. If you are so inclined, you can find my submission
>>> in the aforementioned docket under "Interventions".
>>>
>>> There, I also pointed out that Starlink's current price point is
>>> unsustainable, and that they will have to raise prices.
>>>
>>> Low and behold, while everyone was waiting for the CRTC's decision on
>>> Northwestel's application, Starlink increased prices.
>>>
>>> So I filed a procedural request to obtain permission to add that
>>> information to the docket (after the official deadline to add
>>> Interventions do the docket).
>>>
>>> The CRTC has granted my request, added Starlink's price increase to the
>>> docket, and has ordered SpaceX to explain their pricing plans for the
>>> next two years by today. Other parties will have until April 18 to
>>> comment on SpaceX' submission - which may be difficult, because I expect
>>> all interesting bits to be filed under seal.
>>>
>>> Cheers
>>> Daniel
>>> _______________________________________________
>>> Starlink mailing list
>>> Starlink@lists.bufferbloat.net
>>> https://lists.bufferbloat.net/listinfo/starlink
>>>
>>
>

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Dave Collier-Brown]

[-- Attachment #1: Type: text/plain, Size: 12548 bytes --]

Starlink's sustainability is an interesting question for us to debate, but normally uninteresting to the
Canadian Radio and Television Commission. The fact that they have raised their prices forces the CRTC to address it, and may give us some hints.

What is more interesting to me is seeing Northwestel on the radar again. Friends (hey, Jim!) have encountered them in the past, as part of the Canadian "duopoly" that keeps internet prices high and software cheap, ancient, and bloated. A little competition over quality could be a very good thing for single- and two-supplier regions of the Canadian North and the US West.

--dave

On 4/8/22 17:45, David Lang wrote:
outside major cities, starlink costs and bandwidth are attractive compared to existing options (and frequently cheaper, even including the initial cost over a year or so)

your claim that launching additional satellites will not increase bandwith is directly coutnered by Starlink's desire to launch additional satellites.

As a private company, we don't know their finances, but we've seen enough to know their profit margin on launches is quite high, so it's not a given that they are borrowing billions of dollars.

I don't see premium chaning things much, it's 5x the price, but also 5x the bandwidth, just in one easy-to-use dish vs configuring load balancing across 5 dishes. So it looks like a wash to me.

0.1 clients by sq km seems like an incredibly low density. I haven't seen that paper, so I can't argue with it's assumptions directly.

I think the number of people who can afford the $100 (or if not individuals, then communities sharing a dish) is much higher than you are estimateing. As you say there are not a lot of other options.

I agree that the current level of service/pricing is touch-and-go for starlink, but with cheaper launches (Starship) and more satellites (and the satellite lasers to reduce the need to have a ground station near you)

I've seen people talking about revenue on the order of $30B/year as a possibility, and while I think they are probably optomistic, $10B/year on $30B in additional investment is a fairly short payback period.

David Lang


On Fri, 8 Apr 2022, Daniel AJ Sokolov wrote:

Date: Fri, 08 Apr 2022 23:04:04 +0200
From: Daniel AJ Sokolov <daniel@falco.ca><mailto:daniel@falco.ca>
To: David Lang <david@lang.hm><mailto:david@lang.hm>
Cc: starlink@lists.bufferbloat.net<mailto:starlink@lists.bufferbloat.net>
Subject: Re: [Starlink] SpaceX ordered to explain pricing strategy

To be clear, I am focusing on the consumer product, not the Premium product, which is not available yet.

Mr. Musk himself said last year that he will have to invest an additional USD 20-30 billion to make Starlink survive. After all, he has to build ground stations and launch some 23,000 satellites by the end of the decade to meet his FCC license obligations (not counting any satellites lost to unplanned events, such as a solar flare, warfare, or rocket loss).

If you can finance at 5% (which is optimistic), debt of 10 billion costs you 500 million a year.

At the same time, the number of clients paying USD99 is limited. The system has limited bandwidth. As the IEEE paper shows, users can expect 25 Mbit/s IF there are no more than 0.1 clients per square kilometre, and IF only 5% of these clients actually use the bandwidth - given a complement of 5040 operational satellites. Currently, 1421 satellites are operational (according to starlink.sx).

The global landmass is about 134 million square kilometres, including all uninhabited areas except Antarctica. The number of humans who can afford USD 100 a month is limited (and don't forget the initial investment, the cost of Dishy's significant power draw, and taxes). Those who are willing to pay all that AND be happy with, say, 30 Mbit/s, is even smaller. And the additional bandwidth per satellite added diminishes as the network grows. You can't double the bandwidth by doubling the satellites, because the available spectrum and the spectral efficiency are given.

At the moment, Starlink revenue is at a runrate of about 25 million dollars a month. A terribly negative cashflow. Yes, there is additional demand, but, like everyone else, Starlink suffers from chip shortages, so they can't make as many terminals as they would like to. And some of the demand they can't fulfill without massively oversubscribing. If there are, say, 10,000 New York City residents on the waiting list, Starlink can't serve them.

In the given setup, churn will be high. When a client moves, there is no guarantee they can keep their Starlink account. When a taller building goes up next door, the connection my be interrupted for good, etc.

Having clients scattered over dozens of countries comes with massive overhead in the legal department. Starlink is just being kicked out of France because their radio license is invalid. I believe they can fight their way back in, but still. This costs time and money. Multiply this with dozens of countries with different legal regulatory regimes, consumer protection laws, tax and filtering/surveillance requirements, etc.

At 99 dollars is not enough - which is why Starlink had to raise the price. And unless they have tremenduous success with Premium subscriptions, or larger business accounts, they will have to raise prices again.

The good news for Starlink is: They have a captive audience, who has significant sunk cost and often little alternative. So Starlink will be able to raise prices.

BR
Daniel AJ



On April 8, 2022 8:15:44 p.m. GMT+02:00, David Lang <david@lang.hm><mailto:david@lang.hm> wrote:
Why are you so sure that Starlink's current prices are unsustainable?

That's an assertion that requires prove, not just assumed.

David Lang

On Fri, 8 Apr 2022, Daniel AJ Sokolov wrote:

Date: Fri, 8 Apr 2022 01:59:21 -0700
From: Daniel AJ Sokolov <daniel@falco.ca><mailto:daniel@falco.ca>
To: starlink@lists.bufferbloat.net<mailto:starlink@lists.bufferbloat.net>
Subject: [Starlink] SpaceX ordered to explain pricing strategy

Hello,

the Canadian regulatory authority CRTC has ordered SpaceX to reveal how
its Starlink prices "may change within the next two years".

However, SpaceX will likely file this under seal, meaning it will not
become public information.

Technically, the order only refers to prices charged in the Far North of
Canada (The Yukon, the Northwest Territories, Nunavut, Northern
British-Columbia and one community in Alberta). But as long as
Starlink's prices are global, this geographical restriction in the order
is meaningless.

The order is part of CRTC proceeding 8646-N1-202108175, and SpaceX'
answer is due today, April 8.
Docket at
https://services.crtc.gc.ca/pub/instances-proceedings/Default-Defaut.aspx?S=C&PA=T&PT=PT1&PST=A
(incomplete due to various 404 errors)

The order to SpaceX has came about after I filed a procedural request in
this proceeding.

Here is the background:

In most of Canada's Far North, a company named Northwestel has a
monopoly on landline internet. Also, Northwestel owns the backbone to
large parts of the area. Northwestel is actually Bell Canada, but with
much higher rates. It's only been a bit over year that Northerners can
even buy unlimited internet access, and it is not cheap. (Northwestel
also has a resale agreement with OneWeb.)

In some areas, one small competitor is trying to hold on: SSi Micro.

They and a few others would want to buy wholesale data transfer from
Northwestel at regulated prices, so they can mount some competition.

Because Northwestel has a monopoly, they are not allowed to sell
internet access below cost, and they have to obtain permission from the
CRTC to change rates. Rates must be "just and reasonable" under the law,
for whatever that means. The CRTC proceedings to permit rate changes are
unreasonably slow - a real problem for Northwestel.

However, Northwestel would also love to sell below cost, so they can
extinguish the little competition they have, and make sure no new
investor even thinks about entering the market. Northwestel runs a very
profitable cable TV operation, and they charge business users more than
double the residential rate for internet access - so they have plenty of
revenue to cross-subsidize internet, if they would be allowed to do so.

In January, Northwestel applied to the CRTC for permission to change
this regime. Explicitly, Northwestel wants to be allowed to sell
residential internet access below cost (cross subsidized from cable TV),
and to reduce rates or increase data allowances or increase bandwidth at
any time without another CRTC proceeding.

This, Northwestel argues, is necessary, otherwise Starlink will eat
Northwestel's lunch. Because Starlink is awesome and cheaper.

Such permission, of course, would be great for consumers in the shortrun
and awful in the long run. Because it would kill competition.

Most participants in the consultation to Northwestel's application fail
to understand that. They are jubilant for potentially lower internet rates.

In my filing in February, I asked the CRTC to deny Northwestel's
application. It is bad policy in the long run.

Also, Northwestel has many options to fight against the (perceived)
competitive threat from Starlink. Currently, the cheapest unlimited use
access is a 100 MBit/s down and 12.5 MBit/s up line. They offer plans
with less bandwidth, but all of those have a usage cap. And overages are
crazy expensive. It's a topsy-turvy world, where the rich users with fat
pipes, who can put huge stress on the network, get a free-for-all,
whereas less affluent users with thin pipes get charged extra per GByte.

In addition, I argued that Starlink does not have the capacity to be a
real competitor to Northwestel's fat pipes - unless one takes the
Premium version. Now Starlink Premium is geared at businesses and govs,
for which Northwestel does NOT ask for permission to lower rates.

Plus other arguments. If you are so inclined, you can find my submission
in the aforementioned docket under "Interventions".

There, I also pointed out that Starlink's current price point is
unsustainable, and that they will have to raise prices.

Low and behold, while everyone was waiting for the CRTC's decision on
Northwestel's application, Starlink increased prices.

So I filed a procedural request to obtain permission to add that
information to the docket (after the official deadline to add
Interventions do the docket).

The CRTC has granted my request, added Starlink's price increase to the
docket, and has ordered SpaceX to explain their pricing plans for the
next two years by today. Other parties will have until April 18 to
comment on SpaceX' submission - which may be difficult, because I expect
all interesting bits to be filed under seal.

Cheers
Daniel
_______________________________________________
Starlink mailing list
Starlink@lists.bufferbloat.net<mailto:Starlink@lists.bufferbloat.net>
https://lists.bufferbloat.net/listinfo/starlink



_______________________________________________
Starlink mailing list
Starlink@lists.bufferbloat.net<mailto:Starlink@lists.bufferbloat.net>
https://lists.bufferbloat.net/listinfo/starlink

--
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
dave.collier-brown@indexexchange.com<mailto:dave.collier-brown@indexexchange.com> |              -- Mark Twain


CONFIDENTIALITY NOTICE AND DISCLAIMER : This telecommunication, including any and all attachments, contains confidential information intended only for the person(s) to whom it is addressed. Any dissemination, distribution, copying or disclosure is strictly prohibited and is not a waiver of confidentiality. If you have received this telecommunication in error, please notify the sender immediately by return electronic mail and delete the message from your inbox and deleted items folders. This telecommunication does not constitute an express or implied agreement to conduct transactions by electronic means, nor does it constitute a contract offer, a contract amendment or an acceptance of a contract offer. Contract terms contained in this telecommunication are subject to legal review and the completion of formal documentation and are not binding until same is confirmed in writing and has been signed by an authorized signatory.

[-- Attachment #2: Type: text/html, Size: 14843 bytes --]

Re: [Starlink] SpaceX ordered to explain pricing strategy

[David Lang]

[-- Attachment #1: Type: text/plain, Size: 2263 bytes --]

Larry dug up this paper, is it the one that is being referred to?

https://urldefense.com/v3/__https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9568932__;!!P7nkOOY!qET50RjNacuVwWboH6o-ccJn3yQcttBzx_KR4d_dd2ANtk8ddNdOkK7xNwJFuMrkde30zl-ZZCr7KQ$


lots of 'interesting' assumptions in this paper (right off the bat, that 
everyone is spending the same amount per satellite, which I think makes the 
non-starlink systems look much better than they are)

I also suspect that they are over-estimating starlink costs in a fairly 
substantial way. They say that falcon 9 launches are going to be cheaper because 
it's so reliable, so insurance rates are going to be lower. I'd assume that 
SpaceX is not paying for insurance on the satellites (I'm sure they have 
insurance against damage on the ground)

but even with all this, they list a NPV cost of $0.6M per starlink satellite 
over a 5 year period supporting 2500 subscribers @0.1/km^2 or $200/subscriber 
for the satellite assets with each subscriber paying $6000 over that timeframe 
(there are ground station costs, etc, but with 5k satellites, the satellite 
costs dominate)

Musk has said that without Starship and the v2 starlink satellites, the finances 
barely work, but Starship will significantly decrease the per-satellite costs, 
and the v2 satellites will increase the bandwith available per sq km, and the 
increase in the number of satellites from ~5k to ~40k will increase the number 
of satellites and therefor the bandwith per sq km again.

it doesn't seem such an open-and-shut case that starlink will have to increase 
prices dreastically to survive. They may, but it's not that obvious that they 
must.

Also, that 25Mb/s bandwidth figure is what happens in the peak hour that 
everyone is using the system. If that does not suffer from bad bloat, that's 
actually a fairly comfortable rate, enough for several people to be streaming HD 
video (although for 4k video it gets tighter, but still works) When my 
cablemodem drops out and I fall back to 8/1 DSL, my zoom calls notice when I 
have other people streaming video (along with email/etc), but I'm still usually 
not the worst on the call. 3x that bandwidth (unbloated) would be quite 
comfortable for several people.

David Lang

[-- Attachment #2: Type: text/plain, Size: 149 bytes --]

_______________________________________________
Starlink mailing list
Starlink@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/starlink

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Jeremy Austin]

[-- Attachment #1: Type: text/plain, Size: 1920 bytes --]

On Mon, Apr 11, 2022 at 5:56 PM David Lang <david@lang.hm> wrote:

> Musk has said that without Starship and the v2 starlink satellites, the
> finances
> barely work, but Starship will significantly decrease the per-satellite
> costs,
> and the v2 satellites will increase the bandwith available per sq km, and
> the
> increase in the number of satellites from ~5k to ~40k will increase the
> number
> of satellites and therefor the bandwith per sq km again.
>

Curiously, Starlink so far is not licensed for frequency reuse, which makes
me quite curious how an 8x increase in satellites will result in a similar
increase of bandwidth per square km.


>
> Also, that 25Mb/s bandwidth figure is what happens in the peak hour that
> everyone is using the system. If that does not suffer from bad bloat,
> that's
> actually a fairly comfortable rate, enough for several people to be
> streaming HD
> video (although for 4k video it gets tighter, but still works) When my
> cablemodem drops out and I fall back to 8/1 DSL, my zoom calls notice when
> I
> have other people streaming video (along with email/etc), but I'm still
> usually
> not the worst on the call. 3x that bandwidth (unbloated) would be quite
> comfortable for several people.
>

Completely agree about 25 Mbps being comfortable when latency is good. We
(at Preseem) have a lot of data on how much bandwidth is used per nominal
speed plan, and while the initial increase is steep (a 10 Mbit user is
often constrained these days), above 25-50 Mbps, the slope is about 2:1 --
that is, a 200 Mbit user causes only about 2x the load of a 50 Mbit user.


-- 
--
Jeremy Austin
Sr. Product Manager
Preseem | Aterlo Networks
preseem.com

Book a Call: https://app.hubspot.com/meetings/jeremy548
Phone: 1-833-733-7336 x718
Email: jeremy@preseem.com

Stay Connected with Newsletters & More:
*https://preseem.com/stay-connected/* <https://preseem.com/stay-connected/>

[-- Attachment #2: Type: text/html, Size: 3430 bytes --]

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Mike Puchol]

[-- Attachment #1: Type: text/plain, Size: 3190 bytes --]

The frequency re-use is a factor of steering angle. Take these two spot beams, and the cells that they cover. The left side is a beam almost at nadir, the right side is close to maximum steering angle. Any hexagonal cell inside the beam’s footprint cannot be covered by another beam (from same or other satellite), as the overlap would violate the restrictions on Starlink’s license:



The hex cells are Uber’s H3, so while close to Starlink’s not the same - for the purpose of illustrating the issue, they are good enough. The spot beam size and shape is 100% accurate, as it is taken from the FCC filings - here is the representation of four different steering angles:



It is obvious that the higher the constellation density, the more tight spot beams you have available, increasing re-use considerably. A bonus advantage is you have way less chances of obstructions if you are using satellites almost directly overhead.

Best,

Mike
On Apr 12, 2022, 22:39 +0200, Jeremy Austin <jeremy@aterlo.com>, wrote:
>
>
> > On Mon, Apr 11, 2022 at 5:56 PM David Lang <david@lang.hm> wrote:
> > > Musk has said that without Starship and the v2 starlink satellites, the finances
> > > barely work, but Starship will significantly decrease the per-satellite costs,
> > > and the v2 satellites will increase the bandwith available per sq km, and the
> > > increase in the number of satellites from ~5k to ~40k will increase the number
> > > of satellites and therefor the bandwith per sq km again.
> >
> > Curiously, Starlink so far is not licensed for frequency reuse, which makes me quite curious how an 8x increase in satellites will result in a similar increase of bandwidth per square km.
> >
> > >
> > > Also, that 25Mb/s bandwidth figure is what happens in the peak hour that
> > > everyone is using the system. If that does not suffer from bad bloat, that's
> > > actually a fairly comfortable rate, enough for several people to be streaming HD
> > > video (although for 4k video it gets tighter, but still works) When my
> > > cablemodem drops out and I fall back to 8/1 DSL, my zoom calls notice when I
> > > have other people streaming video (along with email/etc), but I'm still usually
> > > not the worst on the call. 3x that bandwidth (unbloated) would be quite
> > > comfortable for several people.
>
> Completely agree about 25 Mbps being comfortable when latency is good. We (at Preseem) have a lot of data on how much bandwidth is used per nominal speed plan, and while the initial increase is steep (a 10 Mbit user is often constrained these days), above 25-50 Mbps, the slope is about 2:1 -- that is, a 200 Mbit user causes only about 2x the load of a 50 Mbit user.
>
>
> --
> --
> Jeremy Austin
> Sr. Product Manager
> Preseem | Aterlo Networks
> preseem.com
>
> Book a Call: https://app.hubspot.com/meetings/jeremy548
> Phone: 1-833-733-7336 x718
> Email: jeremy@preseem.com
>
> Stay Connected with Newsletters & More: https://preseem.com/stay-connected/
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink

[-- Attachment #2.1: Type: text/html, Size: 5419 bytes --]

[-- Attachment #2.2: Spot beam comparison.png --]
[-- Type: image/png, Size: 282215 bytes --]

[-- Attachment #2.3: Spot beams.png --]
[-- Type: image/png, Size: 950806 bytes --]

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Jeremy Austin]

[-- Attachment #1: Type: text/plain, Size: 1207 bytes --]

On Tue, Apr 12, 2022 at 1:25 PM Mike Puchol <mike@starlink.sx> wrote:

> The frequency re-use is a factor of steering angle. Take these two spot
> beams, and the cells that they cover. The left side is a beam almost at
> nadir, the right side is close to maximum steering angle. Any hexagonal
> cell inside the beam’s footprint cannot be covered by another beam (from
> same or other satellite), as the overlap would violate the restrictions on
> Starlink’s license:
>

 I don't disagree with the description of this type of spatial reuse.  It
will be interesting to see how nonlinear the capacity growth due to tighter
angles is, however, and whether it's more like a 4x increase or an 8x
increase.

The question of whether they can achieve their goals without additional
reuse (for example, covering a given hex with more than one channel) is
open. Fun times.


-- 
--
Jeremy Austin
Sr. Product Manager
Preseem | Aterlo Networks
preseem.com

Book a Call: https://app.hubspot.com/meetings/jeremy548
Phone: 1-833-733-7336 x718
Email: jeremy@preseem.com

Stay Connected with Newsletters & More:
*https://preseem.com/stay-connected/* <https://preseem.com/stay-connected/>

[-- Attachment #2: Type: text/html, Size: 2579 bytes --]

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Mike Puchol]

[-- Attachment #1: Type: text/plain, Size: 1573 bytes --]

Oh, one thing we know for sure is they are already covering certain cells with multiple channels (different frequencies, as mentioned). We have observed two terminals in the same location having different POP latency patterns, which can only be explained by having two separate paths.

Best,

Mike
On Apr 13, 2022, 00:30 +0200, Jeremy Austin <jeremy@aterlo.com>, wrote:
>
>
> > On Tue, Apr 12, 2022 at 1:25 PM Mike Puchol <mike@starlink.sx> wrote:
> > > The frequency re-use is a factor of steering angle. Take these two spot beams, and the cells that they cover. The left side is a beam almost at nadir, the right side is close to maximum steering angle. Any hexagonal cell inside the beam’s footprint cannot be covered by another beam (from same or other satellite), as the overlap would violate the restrictions on Starlink’s license:
> >
> >  I don't disagree with the description of this type of spatial reuse.  It will be interesting to see how nonlinear the capacity growth due to tighter angles is, however, and whether it's more like a 4x increase or an 8x increase.
> >
> > The question of whether they can achieve their goals without additional reuse (for example, covering a given hex with more than one channel) is open. Fun times.
>
>
> --
> --
> Jeremy Austin
> Sr. Product Manager
> Preseem | Aterlo Networks
> preseem.com
>
> Book a Call: https://app.hubspot.com/meetings/jeremy548
> Phone: 1-833-733-7336 x718
> Email: jeremy@preseem.com
>
> Stay Connected with Newsletters & More: https://preseem.com/stay-connected/

[-- Attachment #2: Type: text/html, Size: 3518 bytes --]

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Daniel AJ Sokolov]

On 2022-04-12 at 13:39, Jeremy Austin wrote:
> Completely agree about 25 Mbps being comfortable when latency is good. We
> (at Preseem) have a lot of data on how much bandwidth is used per nominal
> speed plan, and while the initial increase is steep (a 10 Mbit user is
> often constrained these days), above 25-50 Mbps, the slope is about 2:1 --
> that is, a 200 Mbit user causes only about 2x the load of a 50 Mbit user.

I find that data very interesting.

Is there any publicly available data I could refer to in a regulatory 
filing?

Thank you
Daniel

Re: [Starlink] SpaceX ordered to explain pricing strategy

[Daniel AJ Sokolov]

Erroneously sent to David only. Now to the list;

On 2022-04-08 at 14:45, David Lang wrote:
> 
> your claim that launching additional satellites will not increase 
> bandwith is directly coutnered by Starlink's desire to launch
> additional satellites.

I never claimed such a thing.

Please read again what I wrote: "And the additional bandwidth per 
satellite added diminishes as the network grows. You can't double the 
bandwidth by doubling the satellites, because the available spectrum and 
the spectral efficiency are given."

Of course adding additional satellites adds bandwidth - but the more 
satellites you already have, the less each additional group of 
satellites adds. Diminishing returns for additional satellites are not 
zero return.

There are additional ways to increase bandwidth: Use more spectrum 
and/or increase the spectral efficiency. Nobody knows if or when either 
can happen, and what tech changes that would require on satellites and 
ground equipment, so we can't build projections on that.


> I don't see premium chaning things much, it's 5x the price, but also
> 5x the bandwidth, just in one easy-to-use dish vs configuring load 
> balancing across 5 dishes. So it looks like a wash to me.

It is not 5x the bandwidth. Starlink promises 100 to 200 MBit/s for 
consumers, and 150 to 500 MBit/s for Premium. Taking the midpoints, 
Premium is 2.17x the bandwidth.

In any case, even 5x the bandwidth would not equal 5x the load on the 
network. So Premium access brings in more $ per kbit than consumer accounts.

Premium clients will get 24/7 tech support (which should not stress the 
network much, but will add OPEX).

> 0.1 clients by sq km seems like an incredibly low density.

It is. Note that this number refers to the area covered by a satellite 
(101,000 square km), not to landmass. Depending on how much of those 
101,000 square km are water, you can adjust the number for density over 
landmass.

>I haven't
> seen that paper, so I can't argue with it's assumptions directly.

https://ieeexplore.ieee.org/document/9568932

> I've seen people talking about revenue on the order of $30B/year as a
> possibility, and while I think they are probably optimistic, $10B/year
> on $30B in additional investment is a fairly short payback period.

I think that $10 billion revenue/year is possible - if they get to 8 
million consumer accounts or 1.75 million premium accounts, or a mix 
thereof, without degrading the user experience too much, and before 
running out of money.

With an IPO, SpaceX can maybe raise $20 billion net to achieve low 
capital costs. (That would be one of the top 4 largest IPOs in history, 
and Mr. Musk's shareholdings would be diluted accordingly.)

If SpaceX can achieve a sufficient manufacturing rate of Raptor engines, 
can get to a reliable Starship quickly, and can find a launch facility 
that allows for at least biweekly launches, I think they can make 
Starlink work out.

These are significant vagaries.

However, an ARPU of $99 was unsustainable. I thought so, and Starlink 
thought so. They introduced Premium, and raised the consumer rates. I 
don't think this was the last rate increase.

BR
Daniel AJ