Why ban competitive offers in the online world when they’re allowed offline? Big tech wants plain vanilla broadband pricing because it forecloses platform competition.
In the 1950s, the European Economic Community decided that trade of food, medicines, books, and wheelchairs would not be taxed, as they were “essential” items after World War II. This lack of tax came to be known as zero rating.
That meaning carries through today with the US Department of Veteran’s Affairs (VA) Video Connect program, which emerged during the pandemic so that some 3 million US veterans could meet with caregivers via computer, tablet, or mobile device without data charges. Similar zero rated e-government and public interest programs are considered to help Americans find housing (Department of Housing and Urban Development); workforce training or credentialing via state and federal workforce sites; to apply for benefits and monitor health (Department of Health and Human Services and local health departments); and access early learning and child development services (Department of Education).
However, such beneficial social programs are illegal in the European Union, banned under hard net neutrality rules. Ironically the regulation proffered to protect end users rights, blocked end users from receiving free healthcare and education online during the pandemic. After some time and effort, the UK telecom regulator Ofcom suspended the rules and is now modernizing the policy such that beneficial services are allowed.
It’s estimated that as many as 1 billion people have adopted the internet in some 60 countries for the first time because of zero rated data partnerships between mobile operators and content providers. Generally, these are in-kind programs in which no payment is made between parties. That said, adoption rates have stalled from an infrastructure investment gap estimated at $2 trillion as noted by the International Telecommunications Union-UNESCO Broadband Commission report. The report advocates for predictable, sustainable financial participation from global technology platforms and innovative corporate partnerships to close the digital divide for the world’s remaining unconnected 3 billion people. This is the life’s work of Doreen Bogdan Martin, now the Chair and first female head of the ITU.
A Free Sample Or A Discriminatory Practice?
For people who have never tried the internet, zero rating is like a free sample of a new flavor of ice cream. It reduces the consumers’ cost to try something new. Free samples are needed most by entrant firms and new products, and they help create competition.
Despite the demonstrated benefits of zero rating, a recent ProMarket article by Bruno Renzetti argues for global bans on zero rating, claiming that while the programs appears to be beneficial because they lower consumers’ mobile subscription cost, they would strengthen platform dominance in the long run and hence reduce innovation and consumer welfare. He posits that dominant web platforms would pay mobile operators for exclusive programs and hence this exclusionary conduct would push competitors out of the market or raises barriers to entry of expansion.
Having measured these programs globally for a decade, I disagree. For one, Big Tech lobbies governments all over the world to avoid any financial contribution which could help build broadband infrastructure or improve affordability. Moreover, they argue that like Netflix, they have no obligation to negotiate or pay for the use of other’s networks.
Renzetti’s aversion to zero rating appears to be related to the view that any broadband subscription in which all broadband data is not price equally is “discriminatory” and hence harmful. This is likely a misunderstanding of the word “discriminate”, frequently misunderstood in economic and technical context. The primary definition of “discriminatee” is to recognize a distinction, to differentiate, or to perceive differences, as in babies can discriminate between different facial expressions.
To price discriminate is predicated on this definition. When a vendor can perceive differences between two or more customers, then it may be possible to charge them different prices for the same product (price discrimination) or to customize the product offered to each in a manner that reflects the difference (product differentiation). If the customized products impose different expended or expected costs on the vendor, then charging different prices is not harmul does not constitute price discrimination. In fact, the flat and unlimited plans that users enjoy are forms of price discrimination, just as zero rating is. From an economist’s perspective, these offers are not different.
The economic definition is not the same as the legal meaning of the word, which has to do with prejudice, e.g. “an employment policy that discriminates against women.”
Critics suggest that established providers could use zero rating to make it more expensive for consumers to access nonproprietary content or new content where its providers could not pay to subsidize users’ transport costs. For this argument to hold, one must assume that there is perfect competition in content, that any piece of content can be substituted for any other, that users have perfect information, and there are no transaction costs. In such a market, consumers are indifferent to content, and would only choose one over the other because of price.
While this perfect competition model is interesting for academic purposes, it doesn’t exist in the real world. The better economic model is imperfect competition, which accounts for product variation that assumes that both users and content providers know the relative strength of users’ preferences for the different variations, and they each know where the best matches will be made. As a result, they will match up in a manner that maximizes total welfare.
The debates about free and subsidized content are not new. Licensed television providers lobbed a similar criticism that free or advertising-supported TV would put them out of the business. Rather the opposite has happened. There is a market for both because the two kinds of TV programming are not perfect substitutes; they produce different content. While users avail themselves to both, the advantage of the two models is that advertisers are allowed to participate, and hence more content is created overall.
The same analysis applies to free and subscription newspapers. There would be no room for subscription newspapers if they offered the same information as free newspapers. It is frequently the case that volunteers and local advertisers subsidize local newspapers. Making it free may increase the likelihood that community members will read it in that town, but that does not mean they will give up their subscription to a national newspaper. Moreover, just because the local newspaper is free does not mean the people in the next town will want to read it. The content in that local newspaper has a particular interest to the people in that community.
Internet traffic patterns show that some content is highly valuable, but most content is not. In fact, much of the content on the Internet has no value to most users. This is particularly the case for much of the world’s unconnected people, as they speak a language for which there is no content on the Internet, and often they don’t write, so they have challenges in navigating on the Internet.
However, zero rating bans are also pursued for anti-competitive reasons. The textbook case is in India, where Google-funded advocates succeeded in prohibiting the launch of an ad-free, zero-rated version of Facebook, arguing that India’s poor were better off with no internet than free Facebook. This worked to slow Facebook’s growth in the ad market and to ensure market share for Google. However, all zero rated partnerships with Indian platforms were also banned, so Google remains the dominant player today.
Campaigns against zero rating are also advocated in countries to make flat rate internet subscriptions and high data caps (preferably no data caps) the norm, if not the law. While such offers have appeal, they necessarily force low volume users, whether by choice or budget constraint, to pay more for internet access. Meanwhile, high-volume users, those who want to stream movies or play video games, pay proportionately less for their service. Such bans benefit Big Tech, whose video services comprise most of the world’s internet traffic. Ironically, the argument is made to ban Big Tech from engaging in free data programs primarily to benefit Big Tech.
As a related point, Big Tech fights against any effort to differentiate broadband prices on the underlying data because the advertising on its platform devours as much as 25 percent of the traffic on a mobile subscription. Advertising is supposed to work such that the advertiser would pay the connectivity cost and hence the advertising data is not charged to the user. Don’t be fooled. The “investigations” against data caps are driven exactly for this reason. Big Tech wants its ad data to be treated equally to the data the user actually wants to see.
Bans on zero rating bans would eliminate many benefits that consumers enjoy today. Instead, competition authorities can adjudicate their concerns with a simple five-question test of whether the practice is harmful. See the tool I developed with my colleague Bronwyn Howell.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.