• December 7, 2022

AI Shake-Up As Prominent AI Guru Proposes Mind-Bending “Mortal Computers” Which Also Gets AI Ethics And AI Law Dug In

Here’s something that you probably hadn’t been yet mulling over: Mortal computers. But maybe you should be. The heady topic came up at the recent and altogether quite prominent annual conference …

Piiano Releases Secure Database For Enterprises And Developers

Data protection company Piiano has released Piiano Vault, a secure database that’s designed to give enterprises the ability to store safely sensitive personal data in compliance with the EU’s GDPR, California’s …

How Creative Artists Agency (CAA) Is Employing Data, Analytics, And AI To Shape The Culture Of Our Times And Inspire The World

George Clooney. Cate Blanchett. Beyoncé. Lady Gaga. Brad Pitt. Tom Hanks. What do these artists have in common? Each of them, among others, in addition to a roster of professional athletes, …

A series of recent papers sponsored by Big Tech attempts to discredit the effort to find sustainable financial models for broadband. Broadband fair cost recovery is about closing the digital divide through two-sided markets business model innovation. There are a variety of models, whether ad taxes (in which advertisers bear the costs, not end users), traffic assessments, and ideally market-based, good faith negotiation and pay as you go (PAYG) models as used im cloud computing. Essentially fair cost recovery is an evolution of policy to catch up with 40 years of internet development and to allow broadband provision to enjoy two-sided markets, just as Big Tech does today.

The end game of the recent papers is to ensure that Big Tech pays as little as possible for the use of others’ networks, to avoid participation in reducing the cost burden for end users, and to insulate Big tech from financial efforts to close the digital divide, (e.g. universal service obligations, social responsibilities, tax etc.). It is rational and understandable that corporations want to reduce cost and increase profitability. What we should not accept, however, are claims unfounded on fact. Here are Big Tech’s key arguments versus the facts.

Claim: Big Tech already invests in infrastructure, so it should be free of further obligation.

Fact: Broadband providers invest 20-30x of revenue in infrastructure compared to Big Tech.

In a recent report, Analysys Mason writes that it was requested by its Big Tech clients to update their earlier report on infrastructure investment. The report observes that companies like Google and Amazon invest approximately $88 billion annually in hosting (i.e. data centers), transport (i.e. submarine and terrestrial cables), and delivery (i.e. peering and caching). However impressive, this amounts to just about 1 percent of Big Tech’s $6 trillion annual revenue. In comparison, the US broadband providers alone invest some $90 billion in the country annually, as much as 20-30 percent of annual revenue, to say nothing of the rest of the world’s broadband providers.

Moreover, Big Tech invests where there is little to no regulation or where they can set up tax free or with free/low-cost energy (e.g. undersea cables, tax haven islands and so on). They purposely avoid the expensive, regulated places to invest like the last mile. By contrast, broadband providers are required by government to provide coverage on the condition of doing business. Big Tech has no such “coverage” obligations. More largely, Big Tech can confine their investment to their proprietary content and thus enjoy an efficiency advantage, unlike broadband providers which must use their networks for the content for all comers.

Naturally we should welcome all investment in networks, but Big Tech wields their storytelling it such a way to insulate themselves from tax and social obligations. If we accept Big Tech’s argument, then we should remove the infrastructure obligations on other parties.

Claim: Exploding internet data is a revenue opportunity for broadband providers.

Fact: Mobile operators report 20 years of declining average revenue per user (ARPU) in the face of exploding mobile internet data. Other broadband providers report declining profitability.

It is a fact that internet data has exploded. The video data of the internet of 2022 alone is greater than the entire internet of 2018. Analysys Mason, Brian Williamson, and others say that broadband providers can and should make money off this exploding data. I’m sure broadband providers would agree, but the dynamics have not worked in their favor. Just because data is growing doesn’t mean that broadband providers make more money.

The following graph illustrates the trend taking the example of the United Kingdom (UK, pop. 67 million). It shows the explosion of mobile internet traffic in petabytes (y-axis right hand side) and the decline of ARPU (y-axis left side) for the years 2007-2021. In 2007, ARPU was £23; today it is £14, some 40 percent decline. BofA Global Research reports declines in ARPU for other countries over the period; Ericsson notes the mobile internet traffic increase.

Part of the decline in ARPU can be attributed to competition; not only do users have multiple choices for mobile broadband, they can get broadband on multiple networks (fiber, cable, satellite, fixed wireless access etc). Competition is also driven through technology; if one mobile operator invests in the next generation technology (3G to 4G for example), the others will follow. In practice, mobile operators upgrade because, otherwise, their networks will degrade like a carton of milk. More largely, mobile operators also compete with innovations in other technologies like WhatsApp, fiber to the home, DOCSIS 3.0 in cable, and so on.

ARPU decline is also related to physics: Moore’s and Metcalfe’s Laws proscribe efficiency gains—you can process more data for more users with more sophisticated technology, though the unit cost of processing declines. Of course ARPU is not the only metric, but it is indicative of profitability, and indeed, as a sector, broadband telecommunications has flat, if not declining returns. While broadband providers invest in networks and add customers, the overall profitability of any one user has declined. This creates a serious financial challenge about how to recover costs in the future, particularly as policymakers expect continuous upgrades in networks.


In the short run, ARPU decline is great news for consumers. They enjoy more data on better quality networks at a decreasing price. However, declining revenue coupled with an ill-fitting regulatory regime that constrains the ability to recover costs creates an unsustainable future for broadband. If operators don’t earn sufficient revenue, network upgrades go slower, and consumers miss out on next generation innovation. This is observed in many rural areas where operators may have low prices, but build-out is incomplete because of limited customers and the lack of participation by Big Tech. Fortunately a bipartisan group of Senators will investigate the issue in the US in the coming year with openness to a variety of solutions.

While ARPU has fallen for broadband, we see the opposite for Big Tech which enjoys increasing ARPU and other favorable metrics over time.

Claim: Big Tech making financial payments to reduce the cost of broadband for end users is harmful to end users and the internet ecosystem.

Fact: South Korea employs network usage fees for broadband and enjoys increasing network traffic, significant next generation broadband network penetration, and high user adoption.

The markets for peering and interconnection are important topics, but obscure. Reports by WikConsult and ITIF attempt to explain these but hurry to preferred conclusions without using empirical data. They likely fear the mainstreaming of policies from South Korea, widely considered the world’s leading nation for broadband as measured by the International Telecommunication Union’s (ITU) scoring for broadband access, use, and skills. South Korea enjoys the highest rate of 5G adoption of any nation, 47 percent, besting the US and China which boast about one-third of subscribers on 5G.

Importantly, fees from technology companies have not slowed the rate of adoption of fiber to the home (FTTH) subscriptions in Korea, which have increased for the last three years and now stand at 86.6 percent of total broadband connections, the highest in the Organization for Economic Cooperation and Development (OECD). Korea’s National Assembly currently pursues the Network Free Ride Prevention Act to ensure it remains the world’s premier broadband nation, a follow on to its 2020 Network Stabilization Act. These are important, iterative multi-party efforts which attempt to enshrine a market-based framework for parties to negotiate and cover network costs. Importantly Korean policy recognizes that the quality of content delivery is a shared responsibility between content and broadband provider.

In any event, it seems that the authors of the recent papers misunderstand the policies in Korea, for example calling “sending party pays” the dominant regime when it only applies to the same class of broadband providers. For some time, Korean content providers have paid usage fees, in part to keep costs low on end users. This has not harmed the Korean content market which has become a global force in film, music, fashion, and other domains. Indeed Korea now attempts to resolve the asymmetry that only domestic providers contribute while foreign content providers account for greater traffic levels. After more than a year in the preliminary new regime, Google and Netflix enjoy 90 percent operating profit in the country.

The claims that traffic has declined are belied by the data: On mobile networks alone, monthly Korean traffic tops 976,414 terabytes (TB) in 2022, up from 200,000 TB/month in 2015, a near quintupling increase.

There would be no need to intervene if Big Tech companies played fairly. Consider Netflix in South Korea which sued a local broadband provider and declared in court that Netflix had no obligation to pay or negotiate for the cost increases driven by its content. This follows the near overnight, 24-fold increase of Netflix traffic to a small set of users costing the broadband provider some $40 million in equipment to provision and deliver. The court wisely rejected Netflix’s argument and left parties to sort out the costs.

Big Tech’s claim that all data is equal and that its delivery should be free is predicated on a view of the internet when the killer app was email (not video), and when there was no platform oligopoly as we know it today. No one-size fits all solution prevails for fair cost recovery, but it is worthwhile for policymakers to engage with the subject and the data, as they do in South Korea, rather than do what Big Tech wants, which is to shut down the discussion from the start.


Leave a Reply

Your email address will not be published.