Net Neutrality and Internet Packages

For those that don’t know, I’m a professional web developer for my day job (currently working desktop development, but this is the first time in my 11 yr career that I’ve done that). I also happen to be an experienced cruiser, spending over two months of my life at sea – 4-8 days at a time.

Which actually combine to offer a somewhat unique perspective on the Net “Neutrality” debate.

You see, one oft feared aspect of not having Net “Neutrality” is that you will eventually have Internet packages similar to cable packages now. I actually think that is a good thing, as those sites that use more bandwidth (and thus data as well, which is one of the current measures of bandwidth usage) *should* cost more to consume than those that use less. It also *should* cost more to use even low bandwidth (think: primarily text and even static images) sites so much that they become a drain on bandwidth. It should also cost *less* to use lower bandwidth sites and/ or not use the higher bandwidth sites frequently. And I say this as both a purveyor of a little trafficked, low bandwidth site *and* as a person who has “cut the cord” and now streams digital video – in 4K Ultra HD at that – daily. Particularly in the era of “unlimited data” on nearly any device that connects to the Internet (be it via landline or cell connection), this is the only real way to have price do its only true job – to allocate resources as effectively as possible, using basic economic theories of supply and demand.

Some will argue “but the providers will make a killing”. Maybe, maybe not. But maybe they are *losing* money hand over fist right now and are simply trying to slow the losses or even make a profit?

Now, at 300 words in, let’s get to the point, shall we?

I’ve already seen Internet packages work very well. In the particular scenario I am about to describe, I choose not to partake, but other passengers clearly partake quite a bit. This scenario is Wifi on board Carnival’s cruise ships. Which again, I’ve spent nearly 75 days on across over a dozen cruises.

Carnival only has a set amount of bandwidth possible for their satellite-based Internet, and it has to work for everything from ship navigation and communication to employee personal communication to passenger use. Because of this, they charge passengers to use their internet in a mechanism to allocate a scarce resource as efficiently as possible.

Rather than being the one price does everything for a set speed, as land based Internet providers currently do, Carnival instead uses Internet Packages, which are currently defined thusly on their site:

Social Wi-Fi Plan ($5 USD per day)*
Access the most popular social websites and applications including Facebook, Twitter, Instagram, Pinterest, Linkedin, Facebook Messenger, Whatsapp, SnapChat and the most popular airline sites. Does not include access to most websites and email. Does not support access to video and music streaming (such as Netflix, Hulu, Spotify, Pandora), iMessage or Skype. Cellular-network-dependent Wi-Fi calling and Facetime services are also not supported.

Value Wi-Fi Plan ($12 USD per day)*
Access popular social websites, email and applications including Facebook, Twitter, Instagram, Pinterest, Linkedin, Facebook Messenger, Whatsapp, SnapChat and the most popular airline sites as well as news, entertainment (not including streaming sites), sports, weather, banking and finance. Post pictures of your trip and make your friends jealous. Faster speeds than the Social Plan. Does not support Skype or video and music streaming (such as Netflix, Hulu, Spotify, Pandora). Cellular-network-dependent Wi-Fi calling and Facetime services are also not supported.

Premium Wi-Fi Plan (17.70 USD per day)*
Our Premium Wi-Fi plan provides you access to all sites under the Social and Value Plans at the fastest possible connection on board (3 times faster than the Value Plan). Our Premium Plan also supports Skype calling where coverage allows. Does not support video and music streaming (such as Netflix, Hulu, Spotify, Pandora). Cellular-network-dependent Wi-Fi calling and Facetime services are also not supported.

* Price based on cruise-long plans, are subject to change without prior notice and may vary throughout the fleet.

With this mechanism, Carnival is thus able to allocate its scarce resource as efficiently as possible. Note that Carnival does not allow any streaming – music or video – at all, no matter the purpose. And yes, Carnival makes a lot of money for comparatively little daily upkeep.

I don’t know their internal numbers, but based on personal observation I would assume that somewhere between 50% and 90% of cruisers have some version of the available packages. Based on my knowledge of human behavior and the apparent socioeconomic status of my fellow passengers, for purposes of the math below let’s assume that just 50% of passengers have a WiFi plan, and that among that group, 50% of them have the Social plan, 25% have the Value plan, and the remaining 25% have the Premium plan.

Carnival currently has 25 cruise ships, with two more coming online over the next 18 months. Their highest passenger capacity ship is the Carnival Vista at 3934 passengers (a boat I was just on in November 2017), and their lowest passenger capacity ships are most of the Fantasy Class of ships, specifically the Carnival Fantasy, Carnival Ecstasy, Carnival Sensation, Carnival Fascination, Carnival Imagination, and Carnival Inspiration, each with a passenger capacity of 2056. The average across the current fleet is 2656 per boat, which is the number we will use for our following calculations.

Assuming every boat is fully filled every day – not always the case, but clearly Carnival’s goal – 25 ships multiplied by an average capacity of 2656 per ship gives us a total of 66400 passengers per day across the fleet. Using our assumptions above regarding how many people have the various internet packages, we find that 33200 people have some internet package. Thus, the range of money Carnival could be making *per day* is between $166,000 and $587,640, with a likely number based on our assumptions above of $329,510. Per year, this would come out to a range of $60,590,000 to $214,488,600 and a likely of $120,271,150. Not a bad chunk of change for little overall cost, and very likely a very high ROI.

But how does this play into land-based Internet?

On land, we no longer price by data usage now that even mobile carriers have largely gone to unlimited internet. Instead, we price by speed, and buried in the contract it tells you that at a certain data point, your speed will be lowered. It is possible even land line Internet does this, but if so it is less notorious. Because of this and the wide disparity in how people actually use the Internet (and even how often, even now), our pricing model is actually dramatically inefficient. A person who uses little data and largely avoids personal use of the Internet is still charged the same as a person who streams 4K Ultra HD video to multiple devices while streaming audio to multiple devices while perusing various social media platforms on multiple devices, all at the same time.

We need a better way to capture this disparity and use price to allocate resources more effectively and more efficiently.

And that is where packages similar to Carnival’s currently come in. I don’t know what the accepted prices will eventually come to stabilize at, and I don’t know exactly what the accepted packages will be, but I strongly believe that this system will be the most efficient at resolving this disparity. With this package based approach – not allowed under Net “Neutrality” – people who rarely get online or get online often but to little overall impact would be allowed to pay less, because they consume less resources. Others would still be allowed to consume to their heart’s content, but would be expected to pay for their consumption.

This has historically been the most equitable way to allocate resources, and it is a shame we do not have such a system for Internet consumption. Even worse, that so many oppose even allowing such a system to exist.