A defense of ad blocking and consumer inattention

Abstract

Ad blockers are a category of computer software program, typically run as web browser extensions, that allow users to selectively eliminate advertisements from the webpages they visit. Many people have alleged that using an ad blocker is morally problematic because it is bad for content providers and consumers, and it is morally akin to theft. We disagree. In this paper, we defend an independent argument for the conclusion that using an ad blocker is morally permissible. In doing so, we respond to the criticisms that ad blocking is bad for content providers and consumers, that it is morally akin to theft, and that it violates a contract between consumers and web publishers.

Introduction

Ad blockers are a category of computer software program, typically run as web browser extensions, that allow users to selectively eliminate advertisements from the webpages they visit. These programs target a variety of ad formats: banner images, pop-ups, sponsored links, and videos, and they are designed to improve web-user experiences by shortening site load-times, eliminating distracting webpage clutter, and curtailing the amount of information collected about the user’s internet activities. Although criticism of consumer ad blocking is nothing new, there was a sudden resurgence of interest in this topic in the second-half of 2015 when Apple released an update for its iOS9 operating system which, for the first time, allowed consumers to install ad blocking software on Safari, Apple’s default internet browser. Within days of this move, a flurry of articles and think-pieces had been published condemning the use of ad blockers.¹

In opposition to these critics, this paper is primarily a defense of the moral permissibility of consumer ad blocking. We begin by considering one prima facie plausible, but ultimately unsatisfactory, argument for the moral permissibility of ad blocking. We then lay out our main argument for the moral permissibility of ad blocking: that utilizing ad blockers is morally permissible because it is morally analogous to other forms of permissible ad avoidance behaviors, such as TV channel surfing and radio dial adjusting during commercial breaks. We contend that these common ad avoidance behaviors are permissible because they neither harm people in unacceptable ways nor otherwise wrong them. Therefore, ad blocking is also morally permissible. In “Objections”, we consider a variety of direct objections to our main argument and argue that none of them undermine our argument that ad blocking is morally permissible; we then consider (and reject) a few additional indirect objections that seek to challenge the ethical assumptions undergirding our argument. We conclude by offering some remarks about the likely future of ad blocking.

Why ad blocking is morally permissible

Although little has been written about the particular question in which we’re interested—viz. under what conditions, if any, is a consumer permitted to block advertisements—there has been some discussion in the last 30 years of the morality of advertising itself.

Several authors have argued that the very practice of advertising is morally objectionable because ads tend to undermine consumer autonomy.² If these authors are correct about advertisement having a morally objectionable effect on consumer autonomy, then we are in a position to formulate a prima facie plausible argument for the permissibility of ad blocking.

Crisp’s (1987) case against persuasive ads is a clear and typical example of this sort of objection. To act autonomously, Crisp claims, one must be acting on the basis of autonomous desires—which are, for Crisp, first-order desires (e.g. I want ice cream) that the agent accepts and that she possesses for reasons which she regards as good or justifiable. For Crisp, the problem with persuasive ads is that they tend to create non-autonomous first-order desires in those who view them. When individuals act on these non-autonomous desires, they are acting non-autonomously. To illustrate, Crisp considers an ad for a male hair product which depicts a man (a presumptive product-user) in a position of social power, surrounded by attractive women. According to Crisp, this ad is implying that using the hair product will give one social status and make one more attractive to others. If the ad viewer forms a desire for the hair product grounded in this specious association then, Crisp worries, his later purchase of this product will be non-autonomous. If so, the ad will have undermined his autonomy in a morally objectionable way.³

We might note that Crisp and Lippke (1989) were addressing these arguments before the advent of online targeted advertising. (“Targeted” ads are ads that have been narrowly tailored to a consumer on the basis of her, e.g., demographics, buying habits, hobbies, interests, and geographical location. The advertiser crafts the ad, using gathered information about the consumer, to maximize the chance that it will impact the consumer’s choices.) One might worry that targeted ads are especially likely to undermine autonomy. The thought might go: the people who understand you well are often best-able to manipulate you into doing what they want. By playing on your desires, insecurities, identity, and biases they can get you to act on the basis of reasons you may not rationally hold or accept. One might be concerned that an advertiser who understands you is, likewise and to that extent, able to push your buttons and coax you into buying her product. If so, targeted ads may raise especially pressing concerns about consumer manipulation. That would strengthen the autonomy-based defense of ad blocking.

If one agrees with Crisp that ads undermine autonomy in a morally objectionable way, then there is an obvious and uncomplicated argument one can make to justify the permissibility of ad blocking. The argument runs: individuals are prima facie permitted to omit or avoid any action which would undermine their own autonomy. Viewing advertisements undermines individual autonomy. So, individuals are prima facie permitted to avoid advertisements. Therefore, individuals are prima facie permitted to avoid advertisements online using ad blockers.

Although we are sympathetic to this line of argument, we regard it as a relatively infirm position on which to rest a defense of ad blocking. First, it’s not clear that advertisements do undermine individual autonomy in a morally objectionable way. Recall that, according to Crisp, an agent acts autonomously only when she acts on the basis of first order desires that she accepts and possesses for reasons which she regards as good or justifiable. But in the example Crisp provides, the man who buys the hair product does have reasons that he regards as good and justifiable—to wit, he believes there’s a tight association between wearing this product and being attractive/having status. Assuming his desires for attractiveness and social status are themselves autonomous, it seems that the man’s autonomy is not undermined in the way Crisp suggests—because the man regards his reasons for wanting the hair product as good and justifiable.

Of course, even if the ad isn’t objectionable for generating non-autonomous desires, we might object that it promotes false or deceptive beliefs. The man has been led to believe that the hair product will improve his attractiveness when, probably, it won’t. On the off-chance it does, it will do so only indirectly.⁴ It’s possible that the man’s mere belief in the product would lead him to act in ways that improve his overall attractiveness. But that’s just a case in which his attractiveness was inside him the whole time, and so he’s still been deceived about the role the hair product is playing.

If the ad is objectionable for promoting false or deceptive beliefs, we’re looking at a different (and far weaker) argument against ads. What this argument would show is that ads which deceive are morally objectionable—and it would show that only if we grant the dubious claim that merely associating one’s product with certain images or concepts counts as a morally objectionable variety of deception. But anyway, the original proposal was that ads like those in the hair-product-case generate first-order desires which are not held for good or justifiable reasons. It is to this claim that we object.

A second problem for using these autonomy-based arguments as a starting point for an argument for the permissibility of ad blocking is that these arguments, if successful, only appear to establish that particular sorts of ads undermine autonomy. The concern about targeted ads, for instance, must be specifically addressed to ads that seek to manipulate you using gathered personal information. That’s not the same as saying we’re permitted to block all online ads. Some targeted ads may be purely informational, may merely inform you about the existence of a brand or product pertinent to people-like-you, and that sort of ad doesn’t seem objectionable. Are we permitted to block all ads just because we’re permitted to block manipulative ads? That depends on at least two things. First, are the opponents of ad blocking right that ad blocking is morally analogous to theft? Second, what percentage of online ads are of the morally objectionable sort? If ad blocking is akin to theft and if many online ads are either purely informational or poorly targeted, we presumably won’t be able to justify committing many acts of theft just to avoid a handful of bad ads. This is another reason why the autonomy-based defense of ad blocking seems infirm.

In view of these shortcomings, we intend to mount a more robust argument for the same conclusion: one that does not require us to take a stance on whether ads do, in fact, undermine autonomy in a morally objectionable way, one that rebuts charges that ad blocking is theft, and one that defends the permissibility of blocking all ads, informational as well as persuasive.

To begin, it will be helpful to situate the conversation about the permissibility of ad blockers within a broader discussion about the moral dimensions of what we’ll term ‘advertising-subsidized content’. By this we mean media content whose creation or dissemination is at least partially funded by external firms, firms who demand, in return, that the subsidized content be conjoined with ads for those firms’ products.

We can usefully model advertising-subsidized content as emerging from a three-way trade among consumers, firms, and content providers. The firms, in this case, are interested in increasing sales via advertising; the consumers want to consume more media content; and media producers would like both to receive subsidies for producing their content and to distribute that content to a wider audience. Ad-subsidized content is a mutually-beneficial arrangement that satisfies these coincident interests: firms agree to subsidize content producers on the condition that they distribute that content with advertisements. Content producers are freed to produce more content, and consumers can consume more content at less cost.⁵ Everyone wins: more content is produced; more content is consumed; more ads are viewed.

Regardless of whether this model faithfully describes the historical circumstances under which advertising-subsidized content emerged, it does seem to capture an important feature of this phenomenon. Namely, that it represents a mutually-beneficial market solution for all three parties. We introduce this model in order to reveal how consumer ad blocking can be fit into a larger pattern of consumer ad-avoidance. When a consumer chooses to avoid the ads that have been conjoined to the advertising-subsidized content that she consumes—as by, say, channel surfing—she apparently undermines and destabilizes this market arrangement.

Our central claim in this paper is that such acts of ad avoidance are morally permissible.

Before we explain our reasons for holding this position, we would like to introduce a few analogy cases (outside of the context of ad blocking) where this judgment seems more apparent. Consider first: Tanya is a resident of a fictional world in the early days of broadcast television. The three-way trade among firms, consumers, and content providers has produced the following arrangement: broadcast TV stations run pure, ad-free content for 18 hours every day, and for the remaining 6 hours—say, 10 a.m. to 4 p.m.—the stations run a solid block of advertising. Most evenings, Tanya watches 3 hours of prime-time television: the evening news, a variety hour, and a couple of her favorite sitcoms. On those days Tanya also watches an hour of advertisements during her lunch break, reasoning that since she watches one-sixth of the station’s daily media content, she should watch one-sixth of the station’s daily advertisement content. After a few years of this, however, Tanya becomes less diligent in her ad-watching. She doesn’t especially like watching ads, and more and more she finds herself doing other things during her lunch break—say, reading the newspaper. Or talking with coworkers. Or enjoying the afternoon weather.

Does Tanya behave immorally on the days when she skips advertising-hour? Surely not.

Now consider Tony, a resident of the actual world. Like Tanya, Tony watches 3 h of prime-time television every night: the news, a variety show, his favorite sitcoms. Like Tanya, Tony was once assiduous about sitting through commercials, reasoning, like Tanya, that this was only fair. Also like Tanya, Tony has recently become a bit lax in his ad-watching. Sometimes he gets up during ad-breaks to go to the bathroom. Other times he goes to the kitchen to make tea, or goes upstairs to tend to his laundry, or simply mutes the TV to have a conversation with his wife. On other occasions, he finds himself channel-surfing through commercial breaks—which is to say, he watches content on other channels instead of watching the ads on the primary channel that is broadcasting his shows.

As before, we can ask: Does Tony behave immorally when he skips ads by channel-surfing or going to the bathroom? And again, the answer seems to be, Surely not!

Other sorts of ad-avoidance behavior elicit similar judgments. It seems morally permissible for Tony to purposely arrive late to the movie theater and miss all the pre-film trailers and cola ads; or for Tony to use TiVo to fast-forward through the ad-portions of pre-recorded shows; or for Tony to read his copy of Time magazine without so much as glancing at the full-page advertisements; or for Tony to strip the advertising sections from his Sunday newspaper. Each of these is a form of ad-avoidance. Each seems to undermine a market arrangement whereby media content is subsidized with ad dollars. And none of these seem morally wrong. Generalizing from these cases, one would conclude that ad avoidance behaviors (as a category of action) are prima facie morally permissible.

This conclusion seems cohere with our judgments about cases like Tanya’s and Tony’s, but we’ve yet to say anything about why ad avoidance might be permissible (and we certainly don’t mean to imply that the mere prevalence of ad avoidance behaviors makes them permissible). We submit that the explanation is this: Ad avoidance is morally permissible because under normal circumstances a person who avoids viewing ads does not wrong anyone (by violating their rights or treating them unfairly) nor harm anyone in an unjustifiable way. In short, there is no party with a legitimate moral claim against the ad-avoiding consumer that would obligate her to view ads. Since ad blocking is a species of ad avoidance, morally analogous to other members of this kind, we conclude that ad blocking is morally permissible.⁶

Formally, our main argument:

• (P1) Ad-avoidance is morally permissible (because it neither wrongs nor harms anyone in an unacceptable way)

• (P2) Ad blocking is a species of, and morally analogous to other acts of, ad avoidance

• (C) Therefore, ad avoidance is morally permissible.⁷

In the next section we consider a variety of objections to our main argument. We argue that none of these objections undermine our main argument that ad blocking is morally permissible.

Objections

In this section we consider three different objections. Each of these seeks to undermine our argument by arguing that there is a morally relevant difference between common acts of ad avoidance and ad blocking—such that acts like Tony’s and Tanya’s are permissible for the reason given, while acts of ad blocking are not. The first objection, which we call ‘the Monetary Argument’, claims that the moral difference between ad blocking and common ad avoidance behaviors is that typical ad avoidance behaviors do not impose serious financial harm on content providers, while ad blocking does inflict serious financial harm on web content providers. The second objection, known as ‘the Contract Argument’, says that ad blockers violate a contractual agreement between themselves and content providers, but other acts of ad avoidance do not violate any contracts. Finally, the Theft Argument claims that ad blocking is unique in that it is a form of theft—taking content without paying—but typical ad avoidance behaviors do not involve taking content without paying.

The Monetary Argument

First, an objector might argue that the relevant difference between ad blocking and common acts of ad avoidance is that the former inflicts serious financial harms on content providers and advertisers while the latter do not. Therefore, even if common acts of ad avoidance are morally permissible, ad blocking is impermissible. Call this ‘the Monetary Argument.’

The Monetary Argument begins by noting that common acts of ad avoidance, such as channel surfing or radio-dial adjusting during commercials breaks, cannot be expected to harm anyone. Advertisers and content providers have no idea whether a particular consumer is channel surfing, and so the channel surfer can reasonably expect that her behavior will not be bad for either party.⁸ For all the innocuousness of channel surfing, ad blocking is often decried as “devastating for publications who derive much or all of their revenue from advertising.”⁹ This claim is empirically well-supported. One recent survey found that 47% of Americans already use ad blocking software on their PC, mobile phone, or tablet, and the estimated financial losses are enormous.¹⁰ One source put Google’s 2014 revenue losses at the hands of ad blockers at $6.6 billion.¹¹ A different estimate put total revenue losses in 2015 (from all web publishers) at $22 billion.¹² Given the magnitude of those figures, there is a serious and well-founded concern that ad blockers will lead to the shuttering of web publications and to the end of the era of free and plentiful web content. And this would be very bad for the people directly involved with the failing publications. According to the Monetary Argument, these bad outcomes are what render ad blocking impermissible.¹³

However, the Monetary Argument fails to show ad blocking is impermissible. This is for two reasons.

First, one can reasonably doubt the predictive assertion that ad blockers will impose terrible costs on content providers and consumers. Certainly, it might turn out that the ad revenue lost to ad blockers leads to the destruction of web publications and the diminution of free and quality web content. But it might just as well turn out that content providers will replace the revenue lost to ad blocking with alternative advertising revenue sources—sources that circumvent or obviate ad blockers. One such alternative already exists in the so-called “native ad.” Native advertisements circumvent ad blockers by mimicking the form and editorial style of a web publisher’s non-ad content. These ads are loaded in the user’s browser in the same manner as the publisher’s non-ad content and are, at first glance, nearly impossible to identify as ads. Ad blockers and consumers alike tend to mistake native ads for normal content, and, thus, these ads tend to attract far more consumer attention than do traditional banner ads. Many content providers see native ads as the future of online advertising.¹⁴

To be clear, we do not mean to endorse native ads as an attractive alternative to traditional ad formats. Our point is rather that the demise of traditional ads at the hands of ad blockers need not also precipitate the demise of web publications. Publishers may survive by turning to alternative ad formats.¹⁵ Native ads are just one alternative among many.

But even granting the Monetary Argument its predictions about the future of web content, there is a second and more decisive reason to reject it. To wit, it is not plausible that the cited harms are sufficient for generating a consumer obligation not to block ads. In general, the fact that a particular act will make a firm worse off does not give the individual consumer a significant moral reason to abstain from acting. To illustrate: suppose you’ve been shopping around for a new stereo. You have decided to either buy from Firm A, from Firm B, or not to buy a stereo at all. If you buy from Firm B, you will make Firm A worse off; if you buy from Firm B, you will make Firm A worse off; if you decide not to buy a stereo at all, you make both firms worse off. Yet these considerations surely do not obligate you to do anything in particular—not even if Firm A really needs your business. In general, the sorts of comparative harms that market participants impose on one another in the course of making decisions about when and with whom to transact do not generate significant moral reasons. And this is why the Monetary Argument fails. Even if consumers who use ad blockers do harm content providers, this doesn’t mean use of ad blockers is impermissible. Thus, if there is an important moral difference between ad blocking and common acts of ad avoidance, it is not that the former harms people in unacceptable ways and the latter does not.

Of course, content providers aren’t the only ones who might be harmed by the death of free content on the web. Consumers stand to lose, too. Ad blocking could make things worse by creating a web environment in which there are more intrusive ads, less content, or less affordable content. Maybe this is why ad blocking is impermissible. But the concerns we raised in response to the original Monetary Argument apply equally well to this slightly-modified version. Most trenchant: the mere fact that an act threatens to make many consumers worse off does not typically give the individual consumer a significant moral reason to abstain from that action. A consumer is morally permitted to, say, buy the last 5 pairs of flip-flops from the store, even though doing so makes things worse for those who come to the store later looking for same. And a consumer is permitted to cancel her subscription to the newspaper, even though doing so will contribute to the shuttering of the publication and subsequent loss to the community.¹⁶

In sum: the Monetary Argument and variants thereof fail because the harms in question are not sufficient for generating moral obligations.

The Contract Argument

One natural response to these comments about the Monetary Argument is to point out that ad blocking is, in at least one moral respect, way different from cancelling a newspaper subscription or making extravagant flip-flop purchases. Consumers who use ad blockers are being unfair. They aren’t being unfair in the sense of generating comparative harms or failing to abide by principles of distributive justice. How are they being unfair? First, one might say that ad block users are being unfair by violating the terms of some explicit contract with content providers. Second, one might say the unfairness at issue is a sort of implicit contract violation or promise-breaking. Third, one might say the unfairness takes the form of “free-riding”, of failing to do one’s share to support a shared resource or good, and in this sense, the ad block user is being unfair to consumers rather than to content providers. For now, we will call this family of objections, ‘the Contract Argument’. We will consider these three variations of the Contract Argument in turn.

To be convincing, the explicit contract version of the argument will need to spell out how individual consumers find themselves contractually obligated to view online ads while a TV viewer, a magazine subscriber, a newspaper reader, a radio listener, and a theater go-er are not. Presumably, the objector would have to make some empirical assertion to the effect that consumers who visit websites enter into morally binding agreements and that consumers of other types of media do not. However, as any experienced web user can attest, the sort of contracts which would generate a promise-like obligation to view (or at least not-block) ads are by-and-large absent on the web. This is not to say that websites never publish user agreements with clauses that ban ad blockers—perhaps on some out-of-the-way Terms and Conditions page. For all we know, tons of websites may do this!¹⁷ We claim only that such clauses are not morally binding. It’s well-known that web users do not comb through the Terms and Conditions pages for each new web publisher they visit. Hence, a web publisher who uses a neglected nook of its website to ban ad blockers does so with the understanding that the ban will go unread. More significantly, the publisher does this while being fully empowered to confront web users in a more obtrusive way (by, say, placing information about the ad blocker ban on its homepage.) Given these facts, it is unreasonable to think that a hidden ban on ad blockers could generate a binding agreement between publisher and consumer. This, at least, is true for websites who hide or omit agreements not to use ad blockers. But we are happy to concede that a consumer who enters an explicit agreement not to use an ad blocker as a condition of accessing a particular website incurs a special obligation to turn off her ad blocker while on that site. As things now stand, though, the majority of web publishers do not ask from web users anything that could fairly be called “an explicit agreement not to block ads.” Thus, we submit that people who use ad blockers are not guilty of wrongfully breaking an explicit contract.

But consider the possibility that consumers who use ad blockers are violating an implicit contract. What might that contract look like? Under what conditions do two parties enter into an implicit contract? Here’s a necessary condition: both parties must be in a position to reasonably infer (a) that they’ve entered into a mutually binding agreement with one another and (b) what the terms of that agreement are.

When Connie Consumer walks into a rug store, there needn’t be (and usually won’t be) any explicit agreement or even exchange of words between her and the rug merchant. Nevertheless, there is an implicit contract formed. Connie can reasonably infer the terms of said contract: she is permitted to browse the rug merchandise and to interact (in limited ways) with the products and employees, but in return, she is expected to pay for products she’d like to take from the store in whatever currency and amount the merchant specifies.

One might argue that web users also meet this necessary condition. After all, web users are arriving online already having had experiences like Connie’s of offline business interactions. These experiences will guide what they infer about their online business interactions, and what we can expect them to reasonably infer about those interactions. When the web user wanders onto a web publication, she can reasonably infer that she’s permitted to browse the content and interact (in limited ways) with the content and publisher, but in return, she is expected to pay for any content she takes in whatever currency and amount the merchant specifies—to wit, the currency of attention and in the amount of all the ads on the page. If the price is too high, the web user (like Connie) is free to exit. But not free to take without payment.

If those are the terms of the implicit contract, then we’re inclined to say that this version of the Contract Argument amounts to a re-description of what we call ‘the Theft Argument’—which we consider in the next section. To see why, consider what we’d say if Connie took a rug from the store without permission. Although we could describe that as an act of contract-breaking or of being-unfair—and those might both be true descriptions!—it seems more apt, more helpful, to describe it as an act of theft. Of course, it is literally an act of theft, but, more importantly, calling it ‘theft’ underscores the morally relevant features of the case and draws attention to morally analogous cases—that is, cases of people taking things to which they’ve no right and without permission. That seems equally true of ad blocking. It seems equally perspicuous—given what’s been said about implicit contracts—to describe ad blocking as theft.

Finally, there is a third way of understanding the charge of unfairness that generates the Contract Argument: that ad block users are “free-riders” who enjoy free web content but do not shoulder any of the costs involved in maintaining this content. They are being unfair in that they are shifting their share of the costs onto others.¹⁸ This sort of free-riding, the argument concludes, is impermissible.

As a first response to this objection, one might point out that this description of ad block users as free-riders seems to apply equally well to other kinds of ad-avoidance. Channel surfers and tardy movie goers can aptly be characterized as free-riders who fail to do their part to support the content distribution system from which they are benefitting. To defend the permissibility of channel-surfing, the objector must provide some reason for thinking the sort of free-riding involved in ad blocking is morally worse than the sort of free-riding involved in channel surfing.

Here’s one suggestion: ad blocking is morally worse because the use of an ad blocker does not require repeated, intentional effort on the part of the consumer. The channel surfer has to effortfully choose to surf each time she wishes to avoid ads; the ad blocker decides to ad block once and then permanently tunes out. We acknowledge that this is a difference between ad blocking and other kinds of ad avoidance, but we deny that it’s a moral difference. It’s not clear why free-riding would be especially unfair, unjust, or morally bad whenever free-riding is easy. (c.f. fn. 7) Consider, for instance, a clever programmer who writes a bit of code for automatically muting her TV during a commercial, or a newspaper subscriber who builds a robot that clips all the ads out of her morning paper. It is implausible that these one-and-done acts of free-riding are appreciably morally worse than more effortful acts of free-riding.

Even if we could identify a morally relevant difference between free-riding ad avoiders offline and free-riding ad avoiders online, there is a second and more decisive reason not to base an argument against ad blocking on the unfairness of free-riding. To wit, the mere fact that an action can aptly be described as “free-riding” does not entail that that action is morally forbidden. Here are two examples: it is well-known that part of the business model for low-price fitness gyms requires that a majority of members rarely use the facilities. This means that the small minority of members who actually do use the gym regularly are free-riding—since the amount they pay for access to an exercise facility is artificially deflated by the glut of paying-customers who stay home.¹⁹ Regular gym users are free riders. Yet they do not seem to be behaving impermissibly. While there’s a sense in which it does seem unfair for a fitness gym to be largely funded by people who pay but never come, it’s implausible to say that gym regulars are being unfair to gym absentees. An absentee gym member is fully empowered to eliminate unfairness, if there is any, by either cancelling their membership or going to the gym. The fact that regular users are free riders doesn’t imply that they’re treating anyone impermissibly or unfairly.

Another example of permissible free-riding can be found in the “freemium” video games which have become increasingly popular in the last decade. There are no upfront costs involved in downloading and playing these games, but they are designed with limitation mechanisms (e.g. fixed numbers of lives, resources, in-game currency) that can only be surmounted by waiting or paying money. The vast majority of players choose to wait rather than pay, but a small minority of users spend real currency to keep playing. An even smaller minority (called ‘whales’, by those in the industry) choose to spend thousands of real dollars to get ahead in these games.²⁰ These whales account for around 2/3 of the revenue from such games and are by-and-large the only reason why these games exist and are free-to-play for most users. The thousands of consumers who enjoy freemium games without paying are free-riding on the backs of these whales. As in the gym case, though, this does not seem to be an impermissible behavior. The existence of free-riding users does not at all hinder a whale from uninstalling the game or reducing their payments. As before, the fact that some users can be aptly described as “free-riding” does not by itself imply that they’re behaving impermissibly. Similar remarks apply to ad block users: the fact that they can be aptly described as “free-riding” does not by itself imply that they’re behaving impermissibly. The existence of free-riding ad block users does not at all hinder other web users from installing ad block software or avoiding web publications.

The Theft Argument

A third objection to our argument goes like this. What’s uniquely wrong with ad blocking isn’t that it’s harmful or that it’s unfair. Rather, it’s that ad blocking is a kind of theft. The ad blocker, unlike the channel-surfer, can fairly be accused of stealing content—or, at least, of doing something morally analogous to stealing content.²¹ Call this ‘the Theft Argument.’

According to this objection, those who consume media content online are engaging in an economic transaction and are therefore subject to the moral norms that govern such transactions. In this case, the transaction is an exchange of media content for consumer attention. Online media content is heavily subsidized by advertisers, and advertisers only pay for ad space on the assumption that their ads will actually be viewed, so web-users who seek out this content thereby enter a transaction where they pay attention-to-ads and receive content. A consumer who employs an ad blocker ceases to transact with the content provider and instead takes the content without payment. And that’s more-or-less what stealing is: taking a good without surrendering the expected payment. Hence, consumers who use ad blockers are stealing. Stealing is prima facie wrong. So ad blocking is prima facie wrong.

The Theft Argument is, at first glance, rather compelling. There does seem to be something to the thought that ad blocking—uniquely among ad avoidance behaviors—violates some norm of transaction, breaking an agreement between consumer and content provider. Nonetheless, we argue that the Theft Argument fails.

According to the Theft Argument, the fact that a consumer can choose to avoid ads does not give her permission to do so. However, we suggest that this claim should be rejected. In particular, we want to say that insofar as the consumer has not entered into any explicit agreements to view ads, the content provider’s failure to signal that ad avoidance is an unacceptable option really does give the consumer permission to avoid ads. One relevant analogy here is with so-called ‘pay-what-you-want’ schemes. In many markets, the sellers have full freedom to choose how and what to charge for their product One option is what we might call the ‘closed sale’ option: a price is set, and the product is given to only (and all) those people who pay that price. Another option is what we might call the ‘open sale’ option: the seller does not restrict access to her product to only those people who have paid a set price; instead, the seller implements a “pay-what-you-want” scheme where consumers who take the good are allowed to choose how much to pay the seller.²² It seems true to say that taking a product without paying is theft and therefore prima facie wrong at least when that product is governed by the closed sale option—i.e. when the seller has set a firm price and thereby signaled her intention to distribute her good to all and only those who pay that price. But what is the moral status of taking a product without paying when that product is governed by the open sale scheme?

It seems to us that it must be permissible to take such a product without paying. In general, if someone has a right to demand that you Ψ, fails to demand that you Ψ, and thereby successfully waives their right to demand Ψ of you, then you are morally permitted to refuse to Ψ.²³ Call this ‘the Option Principle.’ To illustrate, suppose your brother promises to help you move this weekend. You tell him that there’s a chance you won’t need his help, so you’ll call him on Friday if you want him to come. You later learn that your brother work night shifts all weekend and, feeling bad, you decide not to call him on Friday—even though you do need his help. It seems: your brother is permitted to stay home and not help you move. You had a right to demand his help (because he promised!), but you didn’t make that demand. By failing to make that demand, you successfully waived your right to his help. And now your brother is off the hook.

One upshot of this principle is this: one may sometimes be able to infer facts about what one is permitted to do from facts about what one has been given the option to do (or, from facts about what has NOT been demanded of one.) This seems to be what’s going on in the brother-case and in the case of a “pay-what-you-want” scheme. Since the seller presumably does have a right to demand payment from all those who wish to acquire her product and also presumably does have the freedom to waive that right, it seems that the consumer is able to infer from the fact that the payer has adopted an open sale scheme that she—the consumer—is permitted to pay nothing for the good. By giving the consumer that option, by failing to demand a minimum price, the seller waives her right to demand a minimum price. (At least until she changes her mind and institutes a new pay-scheme.)

We submit that the same lesson applies to avoidable ads. When content providers conjoin ads to content in a way that knowingly leaves consumers the option to avoid ads (and when the providers otherwise fail to explicitly demand of consumers that they view the ads), the providers thereby waive their right to demand that consumers direct attention at the relevant ads. In contrast, if a content provider does make their ads unavoidable or does extract from the consumer an explicit agreement to view ads, then it seems right to say that the consumer really is obligated to view ads in exchange for receiving the relevant good.²⁴ We conclude, then, that the Theft Argument fails because it incorrectly dismisses the fact that consumers descriptively can avoid ads as having little bearing on the question of whether they are morally permitted to avoid ads. In this case, the consumer’s ability to block ads (and the absence of a demand to not-block ads) is normatively significant.

There are four apparent objections to this argument. First, one might claim that a publisher who does nothing to preclude consumers from avoiding ads has tried but failed to waive her right to demand of consumers that they view ads. But this is not plausible. First, even if this objection is accepted it does not show that a consumer is obligated to view ads. What it shows is, at most, that a publisher who fails to prevent ad avoidance retains the right to demand that consumers view ads. But retaining a right-to-demand isn’t the same thing as actually having issued a demand. If the publisher, retaining their right, subsequently fails to demand that ad blockers be turned off, then it seems the consumer is permitted to leaver hers on. A more serious problem with this objection is that it is not plausible to think that a publisher could be trying and failing to waive her right. In general, as long as a seller could have chosen to gift her product, the seller should be presumed to be competent to waive her right to payment. And since there is every reason to think that web publishers could give away content for free (however imprudent that might be as a business strategy), there is no reason to think them incompetent to waive their rights to payment, and so, no reason to think they somehow fail to waive their right.

As a second objection, one might claim that even though the seller in these cases has given the consumer the option not to Ψ, this act does not count as an attempt to waive the seller’s right to demand Ψ of the consumer. But this seems wrong. Consider again the case of the pay-what-you-want scheme. The seller who adopts such a scheme has left it open for the consumer to pay nothing for her good, and—more importantly—the seller genuinely could have chosen an alternative payment scheme. She could have, for instance, instituted a “pay-what-you-want (as long as it’s more than $1)” scheme if she wanted to open-up payment options but found the thought of NO payment unacceptable. Applying this to the case of web advertisements: there is nothing preventing a publisher from expressing on her web-page that web users are allowed to access her site only if they watch a particular video or only if they disable their ad blocking software. Again, the publisher’s failure to demand payment from the consumer is normatively significant.

Third, one might object to our argument by calling into question the plausibility of the Option Principle, which we defended above. Suppose you fail to lock your front door one morning and thereby fail to prevent a stranger from walking into your home. You have in some sense given the stranger the option to enter your home (failed to demand that he stay outside), but obviously this omission does not give permission to the stranger to walk into your house. This seems to show that the Option Principle is false. That is, it is apparently not the case that giving someone an option to Ψ thereby gives them permission to Ψ. So, the objection concludes, the fact that web publishers give consumers the option to block ads does not show that web publishers have given consumers permission to block ads.

This objection fails because it overlooks a vital disanalogy between failing to lock one’s front door and failing to preclude ad blocking. According to the Option Principle, an individual is permitted to not-Ψ if the person with a right to demand Ψ has successfully waived that right. But it is not plausible to think that a person who fails to lock her front door is thereby trying to waive her right to demand that others not enter her home. A more illuminating way to put the point: failing to lock-up is not generally regarded as an act of invitation. That is, there is no convention or norm according to which a homeowner who fails to lock her door is thereby inviting a stranger to open that door and walk in. By contrast, acts of price-setting are consistently and conventionally regarded as acts of invitation—in particular, they are invitations to consumers to buy the priced product. If you post a sign asking $100 dollars for your mattress, you are offering and inviting others to exchange that amount of money for your mattress. Similarly, a seller who institutes a pay-what-you-want scheme for her product is inviting interested consumers to transact with her at a price of their choosing. The case of ad blocking is like this. When content providers knowingly allow their content to be viewed without ads, they are in effect inviting consumers to do just that, and consequently, they have waived their right to demand (for the moment, at least) that those consumers view the ads that appear on their websites.

One might object: our reply only shows that if acts of ad-blocker-non-prevention are acts of price-setting, then they are acts of invitation (and so the Option Principle applies.) But, according to the objection, we have not shown that acts of ad-blocker-non-prevention are acts of price-setting, and therefore, our response is implausible.

This objection fails, and it fails precisely because we’re working under the assumption—given to us in the Theft Argument—that there is an expectation that the web user will pay some price for the content they consume. That is, the Theft Argument is premised on the claim that the introduction of web ads is an act of price setting. But since ad blocking is not precluded (in usual cases, on most websites), the price that’s being set is effectively “pay-what-you-want”—i.e. “direct as much, or as little, attention at our ads as you see fit.” Therefore, the Option Principle applies and entails that ad blocking (when neither prevented nor explicitly rejected) is morally permissible.

Finally, one might argue that web publishers do demand that readers view ads in exchange for content. In particular, when web publishers make an effort to organize their web pages in such a way that ads and content are served together, this might count as a way of tacitly demanding that users view those ads in payment for content. But this objection is inadequate. Since web publishers can check whether a particular layout or web design successfully circumvents ad blockers, they can tell in advance whether they’ve successfully tacitly demanded that users view ads. If they find that the new design doesn’t avoid ad blockers, and they choose to leave the design as-is, that seems normatively significant. They could do otherwise! They could redesign ad/content presentation to force consumers to view the two together—and thereby issue a tacit demand. Insofar as they choose not to, we conclude, it is false that web publishers tacitly demand that readers view ads in exchange for content.

One final disanalogy

We close this section by considering one final way in which ad blocking may be morally disanalogous to offline ad avoidance behaviors. One might argue that online ad viewing usually incurs two costs: the normal cost in attention and a cost in some loss-of-control over the web user’s personal information. Web advertisers often record and track a user’s search history, time-on-site, web history, and—perhaps most concerning—physical location via the user’s IP address. No such tracking is possible via newspapers or televisions. Thus, this may be a real moral difference between online and offline ads such that we can’t claim that these behaviors are morally analogous.

However, this difference, if anything, would make the case for the permissibility of ad blocking stronger than the case for the permissibility of offline ad avoidance. How much stronger depends on how morally bad it is for a web user to lose control of their personal information in these ways. Thus, while this objection may succeed at establishing a disanalogy with offline ad avoidance behaviors, that difference arguably supports, rather than undermines, our case that ad blocking is permissible.

Questions about methodology

We have argued that none of the above objections undermine our argument for the moral permissibility of ad blocking. In closing, we would like to consider one final sort of objection, which claims that our argument relies on some dubious or misleading assumption or makes some methodological error. A vigorous defense of our assumptions and methodology is well beyond the scope of this project, but there are four particularly conspicuous concerns worth addressing.

First, one might respond to our argument by granting us our conclusion while simultaneously regarding that conclusion as somehow uninteresting or beside-the-point. The most plausible version of this strategy would involve the claim that moral criticizability is the moral concept of primary interest and importance, and so our efforts to establish that ad blocking is morally permissible are unproductive. We have not established that ad blocking is morally uncriticizable, and that is the far more important and interesting question.²⁵

We should first like to note that it’s not obvious that the ad avoidance behaviors we have considered—e.g. channel surfing, theater tardiness—are themselves morally criticizable. If the objector agrees that these behaviors are morally analogous to ad blocking, then she will need to argue that such behaviors are, in fact, morally criticizable. This, we think, is an implausible implication.

Secondly, we would like to register our strenuous disagreement with the claim that the permissibility question is of little interest or importance—in this context or in others. No doubt, this is not the place to wade into the difficult debate between those who accept and those who reject the usefulness of moral permissibility as a focus of moral inquiry. But insofar as the reader shares our concern with questions of right and wrong, of permissibility and impermissibly, the present objection need not be taken too seriously.

As a related second objection, one might argue that ad blocking is morally criticizable precisely because it reveals an ungrateful character trait. It is ungrateful for the same reason that refusing to tip or not applauding after a symphony is ungrateful. Even if you, strictly speaking, are not obligated to endure these small sacrifices, doing so shows gratitude to parties that have benefitted you, and therefore it shows good character. Our reply to this objection is more-or-less the same as before: the question, Are we morally permitted to use ad blockers? strikes us as, by itself, interesting and non-obvious in answer. We do not claim that the question about ad blocking and an ungrateful character is uninteresting. We merely claim that it is not the question of primary interest.

Here is a third objection. Some philosophers have defended the position that computer ethics is unique—which is to say, that the involvement of computers generates entirely new moral problems and questions that cannot simply be answered by reference to traditional moral theories or categories (Tavani 2002). Hence, one might object to our argument on the grounds that we have falsely assumed that traditional ad-avoidance and online ad-avoidance are morally comparable.

A first response is to point out that ethicists are divided on this issue. Some hold that the introduction of computers requires the introduction of new methods and frameworks; others strenuously disagree (Tavani 2002, p. 37–38). And, second, even if one grants that computer technology can generate sui generis moral problems, it does not follow that every moral problem involving computers is a new moral problem. It is not plausible that, for instance, the ethics of ordering meat from a local butcher depends on whether one places the order in person, over the phone, or online. Hence, pointing out that our target case of ad avoidance (ad blocking) involves computers while our analogy cases (e.g. channel surfing) does not is not enough. The objector must provide some sort of reason for thinking that the involvement of computers is a morally complicating feature in this case.

Here is a final objection. One might claim that the ad blocking issue is really just an instance of a collective action problem. Collective action problems are known to raise unique moral concerns and have special policy implications, and so one might claim that adequately grappling with the ethics of ad blocking means considering it as a collective action problem. We have not done this, so our discussion is inadequate.

We deny that ad blocking constitutes a collective action problem—at least, not in the relevant sense. Collective action problems raise unique moral concerns because they share the following features. (i) There is some bad outcome that an individual would be morally forbidden from bringing about on her own (e.g. massive climate change.) (ii) She has the option to contribute to a collective action which will bring about the bad outcome. (iii) She is justified in believing that her contribution will not have a morally significant effect on the magnitude of that bad outcome. The unique moral problem lies in explaining whether and why the individual has a moral reason to abstain from participating in the collective action when her participation appears not to make a difference.

So understood, the case of ad avoidance is not a collective action problem. Ad blocking does not satisfy the first condition. We do not claim that an individual would be forbidden from blocking ads if doing so would bring about bad consequences (see our discussion of the Monetary Objection in “The Monetary Argument”) Hence, there is no unique moral question about whether an individual is allowed to bring about those consequences as part of a collective.

Conclusion

We have argued that consumers are permitted to use ad blockers because this behavior is morally analogous to and a species of ad avoidance. And ad avoidance is permissible because it neither wrongs nor harms others in unacceptable ways.

One apparent implication of our argument is that firms who do want their ads viewed are free both to create ads which cannot be avoided and to extract agreements from consumers that they’ll view ads as payment for the goods they receive. Indeed, if the history of ad-subsidized media is any indication, we can reasonably expect that firms will increasingly respond to ad blockers by taking these sorts of actions.

History is replete with examples of firms short-circuiting ad avoidance by creating unavoidable ads. There are three general strategies. First, content providers can try to integrate advertising into their content in such a way that the two must be consumed together. This is one of the motivations for product placement. If a movie prominently features characters drinking branded soda, then the viewer cannot help but be exposed to branded-soda advertising. A similar thought justifies practices like placing banner ads in sports arenas and placing ads directly into a person’s social media news feed. Product integration is an even more extreme example of this strategy. This occurs when a product or firm becomes an integral part of the substance or content of a media product (e.g. a sitcom episode about X’s new job at <popular sandwich franchise>.) A second strategy for limiting consumer ad avoidance involves disabling the tools of ad avoidance. Examples include the unskippable pre-feature ads on DVDs (as opposed to VHS ads which one could fast-forward through), the way that Spotify pauses ads if the user attempts to mute or volume-reduce while the ad is playing, and “anti-ad blocker” script which some sites have been experimenting with as a way of disabling ad blockers.^26,27 A third strategy for preventing ad avoidance involves limiting the consumer’s ability to distinguish ads from content. This is the strategy adopted by the developers of the aforementioned “native ads.” This strategy also has some historical precedent in the Special Advertisement Sections that magazines like Time and Newsweek occasionally run and which are designed to look virtually indistinguishable from the magazines’ regular content.

We predict that such methods will become increasingly common on the web. Indeed, we foresee a sort of arms-race developing between web users and advertisers as each party develops ever-more-sophisticated methods for avoiding and delivering ads. Likely, this arms-race will mirror similar escalations seen between consumers and advertisers in other domains (e.g. television) which saw the introduction of many of the formats and strategies just discussed.²⁸

It may well be that this arms-race makes things worse for consumers. And the result could be a web with less quality content and with diminished access to what content there is. This would be regrettable. But to claim, as we have, that ad blocker usage is permissible is not necessarily to claim that it’s ideal or better for everyone than non-use. We claim only that it’s allowed.