Response to Norwegian telecom regulator on zero rating

The Norwegian Communications Authority (NPT) recently published a blog on Net neutrality and charging models in which it suggests that operators should be prohibited from using zero rated subscription models. NPT can be commended for its opennes by posting its blogs in English (making the content more accessible and searchable) and by publishing the direct contact information of its employees. Norway has an impressive record on net neutrality, not only for its leadership in addressing the issue early on, but in a voluntary, co-regulatory approach which has succeeded to deter violations for 5 years and running. The Norwegian approach proves that supporting an Open Internet is already in the business interests of operators, and ex ante regulation is not needed to force operators to do what they want to do anyway.

Creating regulations on net neutrality is a complex and difficult endeavor, not the least of which because there are many definitions of the term. The NPT blog does a good job to address various perspectives. There are some additional points to be considered on zero rating, the practice of exempting some content or applications from charge in a users subscription. However the concept of zero rating, essentially that of sponsorship, has been around for some time. This research note addresses zero rating in various ways: by the media industry, in the developing world, in its relation to video and data caps, and in the evidence for consumer welfare/harm.

Zero-rated models in traditional and new media
One of the key benefits of zero rating is that it allows more actors to participate in the Internet ecosystem. There is no reason why consumers should have to bear all the costs for their internet subscriptions, especially for entertainment. If we look at the history of the media industries, advertisers played a major role in radio, print, and television. Having advertisers sponsor radio programs and place advertisements in print helped to increase the distribution and diversity of content. If we only relied on consumers to pay for the cost of content, there would be much smaller selection of content today.

Consider what net neutrality would mean for newspapers. There would be a rule that would ban advertising outright and require that newspapers only earn revenue from readers. Taking an example with the zero rating rule applied to newspapers, it would observe that it is discriminatory to allow some advertisers to buy full page ads while others purchase small ads, so only one size of ad can be purchased, and the ads are randomly allocated across the paper.

Advertising has certainly helped the growth of Norwegian television, and today in Norway both sides of the market pay to support content. People purchase television packages, but there are also advertisements on television. Until the 1980s, Norwegians had only one TV channel, and it was funded through compulsory media license fees, or taxpayers had to bear all the costs. To be sure, there are benefits to publicly funded television, but allowing different parties to participate has increased the amount and diversity of TV content.

Most internet companies that offer free services employ a form of zero rating by having advertisers fund their activities. For example Google offers zero-rated searches with advertisers subsidizing the cost. Indeed the Google AdWords bid engine works on the concept of quality; those advertisers that create better quality ads (e.g. more relevant to the consumer search query), pay less for each click. This is certainly a non-neutral system, but it drives relevant advertising, and as Google would argue, a better user experience. There is no doubt that usership of the search engine increases because advertisers subsidze the cost, making it free to users.

If dozens of other media can use zero rating to serve consumers, it is illogical that the telecommunciations industry should barred from the same practice. Indeed a consumer advocate could argue that prohibitions against zero rating violate consumer sovereignty.

Zero rating in the developing world
Between one-third to forty pecent of the world’s population is online today. Getting the rest of world online is an important goal, and zero rating is one way to achieve it. For people not online today, cost and lack of relevance are the reasons why. A zero rated program is a way to overcome those barriers. A zero rated mobile plan is essentially a free trial, just as we might request before purchasing a health club membership or a car.

A key application of zero rating is practiced by operators in India, in which grandparents purchase mobile subscriptions with zero-rated versions of WhatsApp to message with their grandchildren abroad. Without the incentive of a cool service offered for free, the grandparents might never try the Internet. To outlaw the service would be depriving people of valuable communication.

Zero rating can be employed for marketing purposes such as leveraging the popularity of Facebook and Twitter, but it is also valuable to promote locally-produced content, government services (e.g. bus schedules, health applications), or Wikipedia. Indeed the Chilean regulator outlawed zero rating, but it has since made an exception for Wikipedia, because of popular protest.

As NPT states, the goal of net neutrality is to preserve the internet as an open platform, but in many instances in developing countries, users may never try the internet at all without a zero rated experience. Thus prohibitions against zero rating may be an elitist European or American notion of deciding how the internet should be for others, rather than letting people decide for themselves.

A full discussion of the topic can be found at the recent Internet Governance Forum. See the video or read the transcript.

Zero rating video and data caps
Another hot issue is zero rated video and data caps. NPT presents the example in which a provider favors its own video service by exempting its service from the data cap while competing services apply. This issue could be handled without net neutrality. Indeed competition law has provisions under tests of margin squeeze, tying, and discrimination. However we can imagine a situation where a non-dominant operator, or even a municipal operator, created a compelling service with local content, for example local news, government, and school activities. The operator wants to zero rate the service to encourage viewing. Competition law would say that exempting the data from the cap is not a problem if the provider does not have a dominant position in either the broadband market or the video market. However net neutrality rules would prohibit this valuable public service if zero rated.

There is a contradiciton in expecting an operator to be competitive but eliminiating the ways in which it can compete. Net neutrality says that operators should only be able to differentiate on broadband speeds, volume and price. This makes for a dreary commoditized market in which the incumbent with the largest network has the advantage. More often than not, the innovative packages in the marketplace come not from the incumbents, but from upstarts. Zero rating is more frequently employed by the #3 and #4 operator as a form of differentation.

Net neutrality advocates focus on hypothetical cases where an operator ”could” favor its own service. They should instead focus on the record. Many operators have attempted to make competiting services to over the top providers, but have failed miserably. Which would you rather use: Facebook or Telenorbook? Last month Verizon ended its proprietary video service RedBox because Netflix, Amazon, YouTube and others are just more compelling. Operators launch new services every day, most of them fail. There’s a natural process in the marketplace where consumers decide. Regulators don’t need to intervene.

As for the assertion that zero rating program will be out of the reach of new entrants, the reality is the opposite. We see a number of small and niche content providers using these models today as the way to compete with the established players or to serve its audience, for example a health care insurance company that wants to encourage its users to watch videos that encourage healthy habits.

In any case, connectivity is a minor cost for startups. Moreover internet transit fees have fallen by 99%, and they continue to fall.

Lack of data on zero rating programs
NPT asserts that net neutrality is needed to prevent discrimination or fragmentation of the Internet, however differentation of packages based on quality and service is not discrimination. Consumers should have diversity of choices, not just packages based on speed, but packages based upon their preferred service and lifestyle. A one size fits all internet package as proposed by net neutrality may serve some users but not others.

Consider the lifestyle package suggested by EPlus, the German operator which has a partnership with WhatsApp to offer zero rated mobile subscriptions. The WhatsApp funcitonality works even when the balance is not paid.

Anecdotally we know that these programs tend to drive adoption. After trying the zero-rated program, users tend to upgrade to full packages. This is a desirable outcome both for net neutrality advocates and operators. Net neutrality advocates want people to pay ”full price” for a ”full internet” experience. Operators also want to ensure that they can migrate their customers on discount programs to full-paying programs.

We also know that users with zero rated products also avail themselves to internet via wifi, internet cafes, and libraries. This counters the assertion that if people use zero rated products they won’t know that there are other offerings online than Facebook.

We can conclude that zero rating delivers some level of consumer welfare, however there are still more questions that deserve investigation. For example, do zero rating programs eliminate the next Facebook? On the other hand, does the opposite happen? An app developer may use a zero rated program to study Facebook and then launch his own version. One of the contradictions of net neutrality is that even with rules in place, the most popular websites still get the lion’s share of the traffic, and increasingly so. Maybe net neutrality is having the opposite of the intended effect.

This is the case in the Netherlands. The Dutch expected their net neutrality rule to herald a flowering of Dutch content and applications. Instead they have experienced the ”Netflix Effect” in which traffic from the American giant has ballooned from zero to 20 percent of all downstream network capacity almost overnight with just a small percentage of subscribers. As Netflix grows to its stated goal of one-third of all households, literally the entire network will be consumed by its video streams.

NPT should do its utmost to be indepedent and to make fact-based regulation. While we have anecdotal stories about zero rating, there are no definitive studies. As such, it is premature to make a regulatory pronouncement on the issue. Until NPT has studied the evidence and quantified consumer harm, it should refrain from making a rule or recommendation.

Rather than make hypothetical statements, conjecture, and prediction, why don’t we study the matter and collect data? Don’t ban what you don’t understand. Strand Consult is already engaged in researching this topic and can assist. To learn more, contact Strand Consult to order our net neutrality report and participate in one of our workshops.