On November 10, 2014, the US president called the agency in charge of regulating the media, the Federal Communication Commission (FCC) to adopt clear rules on net neutrality. A few months ago, the FCC had proposed to give Internet Service Providers (ISPs) the freedom to charge for favored access to bandwidth the content providers such as Netflix or YouTube which need a large bandwidth. The risk is that some players would get preferred access (‘fast lanes’) while others would have to deliver their services and opinions on standard ‘slow lanes’. Acting as gatekeepers, ISPs would then be able to pick the Internet winners (in exchange for money) and the losers (such as startups and individuals who cannot afford to pay).

Last week President Barack Obama released a statement urging the FCC to reclassify broadband networks as utility. Traditional utilities have special obligations because they enjoy a monopoly in delivering some goods on a network, such as electricity on the grid or voice on the telecommunications network. In practice this reclassification would lead to more federal regulation and stricter rules on the ISPs.

Net Neutrality is not a new concept. It has been coined by some academics to refer to the principle or ideal that a public network such as the Internet should treat all content, sites, platforms, modes of communication, etc. equally to as to prevent any restriction in the communication process, safe those required by network management (e. g. reduction in spam, etc.). Nowadays the legislators in the U.S. and Europe are struggling to strike a right balance between the need to ensure non-discrimination between the content ISPs deliver and the need to allow them to earn enough so that they can invest in the next-generation infrastructure.

The strong signal sent straight from The White House generated strong reactions not only in the U.S. but also in Europe where the regulatory framework is currently under discussion. In September 2013, the European Commission (EC) has published a proposal for a Regulation revising the telecommunication regulation. The text of the proposal  explicitly allows telecom operators and content providers to agree on dedicated “quality of services” levels to provide so-called “specialised services”. For President Obama, no service should be stuck in a ‘slow lane’ because it does not pay a fee. On the contrary some form of paid prioritization would be allowed under the EU proposal.

In April 2014, just before the end of the legislative term, the European Parliament adopted a draft legislative compromise reflecting more restrictive views on how specialised services with enhanced quality could be offered. The MEPs prohibited any kind of restriction on internet traffic by ISPs. Traffic management measures are allowed in a close list of restricted cases (to implement a court order; to preserve the integrity and security of the network; to prevent and mitigate network congestion) and only if they are transparent, non-discriminatory, proportionate and necessary.

The proposal is currently being discussed by the Member States and the net neutrality seems to be one of the most difficult issues at stake. So far only two countries out of 28 – Slovenia and the Netherlands – introduced a law for safeguarding an open internet.

We shall wait and see how the debate continues and whether the Council will find the way to tackle this legislative challenge. One thing is certain, it is difficult to ignore Obama’s announcement even if the regulatory context is different in Europe.

Anna Trzop and Alain Strowel

For further reading:

• Net Neutrality of presidential importance, The Economist
• Monitoring quality of Internet access services in the context of net neutrality BEREC report
• The European Commission website
• Strowel A., Net neutrality: where do we stand in Europe?

See also the book Alain Strowel edited: Net Neutrality in Europe – La neutralité de l’Internet en Europe, Bruylant, 2013.

In particular, the CJEU was asked to balance several fundamental rights in actions against online intermediaries for copyright infringements committed by third parties. In SABAM v. Scarlet (case C-70/10; see also SABAM v. Netlog, case C-360/10), the CJEU defined the outer limits for “injunctions against intermediaries whose services are used by a third party to infringe a copyright or related right” (Art. 8(3) of the 2001/29 Copyright Directive). On 27 March 2014, the Court of Justice in UPC Telekabel (case C- 314/12) went further in delineating the requirements for copyright injunctions against online access providers and the implementation of filtering measures.

In the last years, the ECtHR ruled three times on the proportionality of some interferences with the freedom of expression online resulting from the protection and enforcement of copyright. One case (Fredrik Neij and Peter Sunde Kolmisoppi v. Sweden, 18 Febr. 2013) involves the operators of the infringing website The Pirate Bay. In Ashby Donald v. France (10 Jan. 2013), the question was whether the sanctions imposed by the French courts on the managers of a website containing photographs of fashion shows were passing the test of proportionality for restricting freedom of expression. In a March 11, 2014 decision (Yaman Akdeniz v. Turkey), the ECtHR considers as inadmissible the request made by the user of musical websites to which access has been blocked.

Here are a list of questions for you to consider at the light of (at least) the two decisions UPC Telekabel and Fredrik Neij and Peter Sunde Kolmisoppi v. Sweden:

• What are the fundamental rights that are most often referred to by the Court in Luxembourg and by the Court in Strasbourg?
• Is Article 10 of the ECHR applicable to the Internet in the same way as it is applicable to other means of communication?
• Does the protection of property as a fundamental right play a role in the balancing made by the CJEU and the ECtHR?
• Is it possible to rely on other fundamental rights to support the view that copyright is a fundamental/human right?
• Do you see a difference in the way the balancing is done by the CJEU and the ECtHR?
• What are the implications of a reasoning based on the balancing of fundamental right?
• As a result of those decisions, does copyright move from the sphere of private law to the realm of public law?

You do not necessarily have to respond to all those questions, you can select a few of them. What is important: your text should be concise, well-structured and to the point. Also it should flow well.

You do not have to be too … balanced in your response!

(You might as well take into account this additional question: Do you think the decision Akdeniz is completely in line with the previous case law of the ECtHR?).

Every academic year at the Louvain School of Management, about 500 master theses are defended. Research topics span all disciplines of management science. In their research, students either address a practical issue raised by a particular organization or aim at contributing to the theoretical or empirical literature in management.

This year, four theses had to do with crowdfunding. We describe here three of them (the fourth, by Camille Storme, details the business processes of MyMicroInvest and is, quite naturally, confidential); we also make the full texts available.

• In “Facteurs de succès et d’échec des campagnes de crowdfunding : Les enseignements de la base de données KissKissBankBank”, Charlotte Lamine replicates the analysis of Mollick (2013) by using recent data from the French crowdfunding platform KissKissBankBank. Charlotte’s analysis mainly confirms Mollick’s results; some extensions are also provided to complement the analysis.
• In “Le crowdfunding: bilan et perspectives”, Michael Riga examines the development perspectives for crowdfunding in Belgium. To this aim, he performs a qualitative analysis of the Belgian crowdfunding scene by interviewing a number of key stakeholders: two executives from two crowdfunding platforms (MyMicroInvest and Look&Fin), three entrepreneurs who raised funds through crowdfunding, one senior official from the FSMA (Financial Services and Markets Authority), and the President of Euronext Brussels. One of the main conclusions drawn from Michael’s analysis is that Belgian crowdfunding platforms currently operate at an insufficient scale. To develop, they should expand outside Belgian but the difficulty there comes from a lack of harmonization of the regulations at the European level.
• Aurelie Bekaert further develop this issue in “Drivers and Barriers for Cross-Border Crowdfunding: Evidence from Belgian Stakeholders”. Her research aims at identifying features that impact the practice of cross-border crowdfunding and at providing evidence from Belgian crowdfunding stakeholders (platforms, projects initiators and funders). The interviewees express their need to expand crowdfunding over Belgian borders: platforms want to expand their activities in order to generate more revenues; projects initiators see international crowdfunding as a vector to reach more people; funders hope to find a wider set of interesting campaigns on the international scene.

Because of their original data sets and their complementary nature, these three theses should be of interest for all researchers in the field of crowdfunding. Furthermore, all three theses start with comprehensive and up-to-date reviews of the literature on crowdfunding.

Economic issues

The picture of the copyright industry that we outline in the third post of this series has many economic consequences. An important one deals with the economics of mediatisation and its impact on creation or diversity. Another with the role of copyright and related internalisation strategies.

Each copyright industry is now confronted with a change of its ecosystem, which affects both mediatisation and distribution. And which in turn, loops back onto production. This process is uneasy to address in a general way because each industry evolves within its own ecosystem. Moreover, each public figure – artist, author, columnist, sportsman, politician, etc. – behaves like an autonomous brand, generating its own topicality, building its own community on social networks.

intellectuallicensing.com

From a microeconomic standpoint, the utility of a copyrighted works can be termed as the sum of the expression and its associated mediatisation. When the mediatisation is subject to network effects, its utility grows with its coverage and with the consumption of the mediatised good. Copyright industries – from the serialized novels published in the newspapers to the Hollywood star-system (Caves, 2000) – have always been confronted with the necessity to generate network effects around brands and branded products. In the analogue world, such network effects were obtained through well-tested strategies combining a release investment calibrated to reach a critical mass and a media reverberation buzzing on the product. And when the buzz was feared to be negative, then the release investment was made to avoid it. (Cf. the case of the Hollywood “ten-ton turkey” release commented by Richard Caves (2000) p.166.)

An open question concerns the achieving of such network effects in digital ecosystems. According to a recent consulting study,

“total global artist income from recorded music in 2013 was $2.8 billion, down from $3.8 billion in 2000 but up slightly on 2012. Meanwhile artists’ share of total income grew from 14% in 2000 to 17% in 2013 (Mulligan, 2014).

But the top 1% artists grasped 77% of this revenue making digital recorded music a “superstar economy”. This suggests that network effects may be more massive but less accessible in digital ecosystems than in analogue ones. Such a trend would favour on the one hand, big incumbent artists whose long lasting presence grants a critical mass (a lower access cost to media), and, on the other, new artists showing extravagance enough to saturate news and social media (Rihanna, Psy, Lady Gaga, Miley Cyrus…). In other words, the competition for network effects in global digital media seems to induce more “winner-takes-all” behaviours than in the analogue walled ecosystems.

The current paradox of digital ecosystems is that works have become accessible through nearly infinite databases but they are less extensively consumed than in the analogue system. Some more empirical evidence is required to validate such hypothesis, but whether true or not it means that the “long tail” theory (Anderson, 2004) is seriously challenged by the economics of mediatisation. In effect, the diversity of creation – when related to effective consumption – relies more on mediatisation economics than on the productivity of search. A key issue both for economists and media is to design mediatisation models that could enhance the notice of niche products.

Mediatisation economics has long been associated with marketing or advertising whose external effects on sales are difficult to quantify. Simply because the utility of the meaningful complement relies on network effects that are as hazardous to predict as the success of creative works. However, more and more consulting firms are now measuring brands digital presence and their apparent valuation by consumers. Such data, if their collection costs are affordable, might be used to model the network effects associated with mediatisation procedures. (According to our experience, the collection of homogeneous data picked up from various digital media sources requires a dedicated task force with a high fixed cost. The specificity of each mediatisation scheme tends to inhibit such an investment.)

Copyright policy implications

Last, but nor least, the growing importance of mediatisation changes the traditional vision of copyright exploitation. While copyright was for long identified as a way to internalise creation efforts, it is also a way to internalise mediatisation strategies. In a media ecosystem, a copyrighted good can be given for free if its consumption increases its media component. This means first, as noticed by various authors, that piracy or counterfeiting can prove socially useful. Not because it provides access to a lower segment of consumers (Watt, 2000). But because it can raise the utility of the good (the media component) and then, generate a higher willingness-to-pay that is collectable where excludability can apply (as already explained by Paul Belleflamme on this blog).

It also means that the previous consent of copyright holders may raise transaction costs preventing them to benefit from mediatisation opportunities. The “notice and take down” procedure applying on digital platforms corresponds to an ex post consent that allows some right holders to better internalise many spontaneous mediatisation efforts. At the cost, of course, of a careful watch of their product mediatisation. The case of the countless parodies posted on YouTube that can be withdrawn on right holders’ request illustrates this trend. Parodies are pieces of the informational complement that add value to the original copyrighted expressions.

The previous consent rule should then apply but with lenience. A too stringent behaviour would prevent right holders from valuable media externalities. Conversely, take down notices corresponding to an effective internalisation decision of the right holder should involve the responsibility of the noticed intermediary. This is the point where copyright enforcement now meets tort law (Lefort, 2013).

More generally, the dissemination of copyrighted works is anything but natural. Following the B&M premise, the utility of copyrighted expressions relies on a media complement produced through publishing and progressive dissemination. Such complement carries network effects and, thanks to its contextual dimension, allows right holders to gain from economies of scale in publishing other works. This point is clearly underestimated by the economic literature. It explains why the Copyright Term Extension Act (CTEA, 1998) has been mocked by so many conventional economists while supported only by a few copyright experts (Liebowitz and Margolis, 2004): the loss of property rights on the Mickey Mouse cartoons would have, no doubt, spoilt the Disney brand and its ability to mediatise new works. Hence, the economics of mediatisation deserve more research and quantitative analysis. It should also promote a better coordination between trademark and copyright law.

Media ecosystems

The term ecosystem is primarily used in biology do designate a community of living organisms (plants, animals and microbes) in conjunction with the non-living components of their environment (things like air, water and mineral soil), interacting as a system (Wikipedia). These biotic and abiotic components are regarded as linked together through nutrient cycles and energy flows. Willis (1997) traces back the history of the concept and its adaptation to the description of natural systems. Ecosystems are defined by the network of interactions among organisms, and among organisms and their environment. They can be of any size but usually encompass specific, limited spaces.

By analogy, a business ecosystem applies to a community of economic agents interacting for the balance of all, in a framework favourable to externalities (Teece, 2007).

A business ecosystem finds its roots in the idea of value networks and can be seen as a group of companies, which simultaneously create value by combining their skills and assets. Business ecosystems create value for a participant only when the participant is not capable to of commercializing a product or service relying on its own competences. (Clarysse, Wright, Bruneel, Mahajan, 2014, p.1164)

The concept of ecosystem differs from the one of industry, market or sector, because it relies on the prevalence of external effects. Quite frequently, the externalities come from private or informal information exchanges – on individuals, organizations or institutions – that decrease transaction costs and favour various forms of business clustering. Such kinds of ecosystems could be called “transactional” since most of the externalities come from the impact on transaction costs of circulating knowledge, informal relations and institutional information.

Similarly, one can consider that a communicating society runs like a biotope in which public messages are received and re-emitted by media. Media interaction participates in society coordination although media are not systematically linked together by formal or contractual relations. As such ecosystem contributes to the production of a “favourable notice”, the participants rely on it to commercialize their products. Quite often, “the skills and assets” that are shared by media deal with trademarks or talent names equivalent to “brands” that provide a meaningful context to goods or expressions (Bomsel, 2013). The mediatisation of people, firms or goods proceeds from ecosystems dedicated to specific products. One can distinguish the ecosystem of books, the one of the news, of scientific papers, of plastic arts, of music, of cinema, of sports, etc. But there is also a media ecosystem for fashion, luxury goods, cosmetics, automobile, food, drugs, etc. While transactional ecosystems decrease transaction costs between the participants, media ecosystems produce the public image – the informational complement – of business or institutional agents and their related products.

Digital technologies are bringing new media in all ecosystems: online newspapers and magazines, blogs, science portals, audio and video platforms, social networks… The echoing of the news, for instance, can follow many new routes and reach consumers faster and wider than could the pre-digital media. Moreover, the connection between advertising or mediatising a product and its actual purchase by a customer can be internalised through online procedures: an html link will transform a viewer into a purchaser, an audio or video platform teases for free until the viewer decides to pay for a service. As a consequence, the borders between mediatisation and sales tend to blur. This trend affects the economic means used to internalise the externalities of the media ecosystem. This is particularly sensitive in all the copyright industries that produce expressions to be distributed in a dematerialised form.

The mediatisation of copyrighted expressions

Copyrighted expressions are informational goods that have to be mediatised. An inspired poem or a scientific theorem posted on an anonymous blog will remain unnoticed. The release of a novel, of a scientific article, of a music tune, of a movie or a TV series, of a sports event, follows a symbolic route constituting a publishing protocol. The public image of a movie disclosed at the Festival de Cannes will differ from the one of a blockbuster presented in a Hollywood showy opening, and even more from a B movie released straight on DVD. The mediatisation of copyrighted expressions adds meaning and value to the works. It is achieved through a media ecosystem providing a direct and indirect echo to the works.

Copyrighted expressions differ from other goods by the fact that, being non-rival, they can be mediatised through free consumption. Thanks to external effects, the supply of free copyrighted works can be financed by ads paid by other firms, by authors (gold open access reviews), or by other contents bundled on a publishing platform. Digitization, which extends the non-rivalness thanks to the dematerialisation of the supports, increases the potential of mediatisation through free consumption. Hence, the coexistence of free and pay content – news, music, video, scientific articles, sports, etc. – on the same digital platforms, tends to narrow the gap between mediatisation and distribution.

In effect, a media ecosystem reflects a state of the technology. Let’s take the case of music. Under the phonographic period, music was taped on records and mediatised through record sleeves, radios, concerts, magazines, video clips and record shops. The radios had to notice the release of the records and to start the buzz. Competition between record producers gave even rise to the payola system – a bribe to disc jockeys in order to get aired – that was depicted by Ronald Coase (1979) as a hidden publicity. Concert tours were keeping the topicality of the artists. Magazines were feeding fans with stories related to music performers. At the end of the chain, record shops were the ultimate media displaying the sleeves and their assortment in racks. In this ecosystem, most of the relations between copyright holders and media were internalised by formal arrangements. The production and sales of records was shaping the ecosystem. Broadcasters were paying license fees while getting paid through publicity. Showcase and tour operators were hiring performers as to sell tickets and drinks. For the most popular artists, record producers invested in video clips and TV ads. Distributors were rewarded on record sales. Only a few players (magazines, talk-shows, variety-shows) benefitted from informal cross-externalities.

With digitization, the phonographic ecosystem gets supplemented both on the media and the distribution side. On the media side, streaming platforms and social networks can create and amplify a buzz that was formerly produced by showcases and radios. The online press will link a music review to a playlist that is just one-click from the reader. TV promotion is re-focussed on ads and talent shows whose viewers argue on social networks. On the distribution side, music downloads and, increasingly, streaming services substitute to CD sales.

A key pattern of this trend is the convergence of mediatisation and distribution. Streaming platforms deliver both mediatisation – through free access service and discovery or thematic apps – and distribution. On such platforms, a subscriber can share his playlists with his social network’s friends. Thus, a listener can become a disc jockey, now called a curator, getting followers for his selection service. On video platforms, amateurs can post video parodies that will increase the popularity of the original. Right holders can either accept or refuse such parodies and even, for a commercial use, license performers their music publishing rights. In other words, with the continuous surge and rollout of new media, the music ecosystem gets denser and denser making the internalisation strategies more open and flexible than in the past.

What happens with music also happens with books, scientific articles, news, sports, cinema, TV series… Each copyright industry – characterised by the production and the publishing of specific expressions – is confronted with the surge of new media that change its former ecosystem. For a while, this change has been perceived as a dematerialisation arousing competition or cannibalisation between distribution channels. This view was suggested by the vertical design of the industry resulting from the copyright transaction chain and the distribution of material supports.

Actually, the change is far more sophisticated. The newness of the digital ecosystems is that mediatisation gets blurred with distribution. In other words, digitization makes copyrighted works distribution a particular case of mediatisation. This already existed when, for instance, radio stations had to pay license fees to mediatise music. But with digital ecosystems, such practises are widespread and continuously expanding. This in turn, makes internalisation – copyright exploitation – an iterative process adapting to new media, new practises, requiring new contracts and new pricing schemes.

The last post of this series examines economic issues and policy implications related to media ecosystems.

Advertising as a complementary good

Since George Stigler’s paper on information costs (1961), the issue of advertising is widely discussed among economists. For many of them, advertising is a means to reduce consumer’s transaction costs and therefore, to increase demand by decreasing the real price of the product. However, as many advertisements do not provide objective information, some economists see them also as a means to change consumer’s preferences about the product. Therefore a distinction is made between informative and persuasive advertisement: informative ads are perceived as socially useful because they cut transaction costs, while persuasive ads generate artificial, if not lousy, needs (Dixit and Norman, 1978). This approach raises two issues: first, it is nearly impossible to separate informative from persuasive advertising as most messages carry the two aspects; second, from a theoretical point of view, if the consumer has free will, his or her preferences cannot be manipulated by third parties (Stigler and Becker, 1977).

This is why Becker and Murphy (B&M, 1993) uphold that advertising should be seen as a complementary good being positively or negatively valued by the consumer. Advertising is then an informational complement, comparable to a quality surplus or deficit, that helps to better discriminate among consumers: consumers who value more advertising will get more of it attached to the product while others will buy the product according to their valuation of its plain utility. Since the meaning associated with information can have a social value, the information complement may carry network effects increasing the utility of the good with the number of consumers. Such analysis of a meaning value brought by advertising to the product is both in line with the free will consumer hypothesis, and with the change in product utility that publicity can bring.

In their approach, B&M define advertising as giving a “favorable notice” to the good advertised so that an increase in ads raises the marginal utility of this good. They underline that such “notice” can either be paid by the advertiser and given away to the consumer, or sold to him or her through various media channels.

For example sports columns in newspapers provide plenty of notice about local professional teams, even though sports sections are not free to readers, and team owners do not pay for the columns.” (p.946).

Such notice generate additional audience to sporting events, whatever their distribution channel.

Marketers commonly distinguish three kinds of channels: paid media that is used for traditional advertising (TV, radio, Press, etc.); owned media, that is the information delivered by the product itself, by an integrated distributor, by an owned website and all the channels related to it; earned media that is the information published by reviewers or consumers, more and more through social networks (Edelman and Salsberg, 2010). In other words, the information complement associated with the product is not only paid advertising it is the accrued meaning resulting from all media channels. This is why we prefer to call it “mediatisation”. Mediatisation economics consists in getting the most valuable complement at the cheapest cost.

thinkofthat.net

A good illustration is given by the historic launch of the Apple Macintosh by Steve Jobs in January 1984 (Isaacson, 2012, Chapter 15). At that time, Jobs had financed a commercial movie shot by Ridley Scott setting up the myth of the liberating machine. The ad was such a storybook that the machine did not even appear in it. Jobs decided to broadcast it during the Super Bowl, reaching at once 96 million Americans. The ad created a massive expectation enabling Jobs to disclose the product in a theatrical keynote echoed as a global event. In other words, while the paid ad was focussed on persuasion, the informative side was covered by the keynote, which got relayed for free all across the world. This “publishing protocol” became a routine that helped Apple to establish its brand as a powerful context adding value to all the new products marketed by the firm. The innovative genius of Steve Jobs did not only focus on products but also on their mediatisation that catalysed their commercial success.

The multiplicity of media channels and the differences in their pricing schemes underline the importance of external effects in providing the media complement. The quest of such effects does not concern only marketers, but all the talents involved in the product, as well as media professionals and consumers. In the Apple example, the media earned by Jobs’ keynotes are external effects of the initial ad investment. The heuristics of marketing aim at generating the most valuable mediatisation as possible. It is frequently related to the maximisation of the externalities associated with the release investment.

Mediatisation results from a publishing protocol

www.dulwichclassics.com

Schematically, two components contribute to the media complement value: one is the myth, the story conveyed by the media; the other is the reach, the social coverage of the message. The myth, which starts with the design or the packaging, the visible appearance of the product, relies on a creative work. It adds a universe that can be more or less valued by the consumer. (For instance, Alfred P. Sloan (1965, pp. 272-273) explains how the curvaceous design of the 1929 Buick has been mocked as « the poor Buick enceinte » leading to a pitiful commercial flop.) The social coverage can either be obtained trough paid, owned or earned media. It may generate network or snob effects according to the positive or negative valuation of the number of consumers. In some cases, the buzz corresponding to earned media can get hi-jacked by angry consumers willing to spoil the image of the product.

However, these components are not independent. For instance, the disclosure of the product relates both to the story and to the coverage: the product’s entry into the public sphere, its arrival to the world takes part in its mythology. The delivery of the Macintosh coming after the Super Bowl announcement and brilliantly staged by its creator, is fully part of Apple’s myth. The disclosure sets up a before and an after which has the dimension of an event. If spectacular, such event can be covered and echoed freely through the news. A resounding release establishes both the newness and the coverage of a product whatever its objective innovation.

In other words, mediatisation proceeds from a “publishing protocol” which provides the product – whether material or not – its meaningful complement (Bomsel & al., 2013). We use this expression by analogy with a correspondence protocol, which connects together two identified agents who agree to start an information exchange. A publishing protocol connects an identified emitter to indistinct receivers. It announces the disclosure of a message and establishes its public status. By doing so, it creates a context adding meaning to the published expression. Hence, such protocol has the economic dimension of a creative work: nobody knows in advance what its utility will be and how much it will add to the product. Although a mediatisation is needed to give the product its public status, it can be perceived either as a good or as a bad.

In this process, the information complement (the myth) is embedded in a set of names: the emitter of the expression who can be a single artist, a sportsman, a politician, or a cast, a team, provides a context which adds sense to the expression. It is the role of the publishing protocol to make the emitter identified by the public and to provide his name – in fact, his brand – with a contextual value. A publishing brand can be seen as an informational asset providing the branded expression with a context valued by the consumer. Such context has cumulative properties: its meaning value changes at every occurrence. A football team, where each player gets mediatised at every game, is a contextual asset portfolio (Bomsel & al., 2013).

Once the utility of the myth is tested, many techniques can be used to maximise its social coverage and the sales revenue. Because a “publishing protocol” deals with disclosure, it is also a piece of news. Hence, it can benefit from the externalities of the news echo. In effect, the news is not restricted by copyright and fuel media competition: a sensational event is instantly relayed on all the news channels including social networks (Facebook, Twitter, Instagram, Weibo, Weixin, QQ, etc.). Moreover, the brand value of popular artists attracts audiences on every platform that takes up the news. The importance of such externalities in producing the media complement deserves a specific focus. It will refer to the concept of ecosystem, which we explore in the next post of this series.

The objective of this series of posts is to show that in the case of copyright, the value added to proprietary expressions is produced through media ecosystems. The prevalence of external effects associated with the media coverage – the mediatisation of the works – reinforces the internalisation function of copyright. But in a way that differs from the traditional vertical chain of contracts. And that makes copyright exploitation a highly empirical activity.

Copyrighted expressions are usually created to address audiences, e.g. receivers that are unknown from the emitter. They are then brought to the attention of the public through media.

Media provide to the expression its public status. Conversely to business talks, or private correspondence, mediatised expressions are messages formally addressed to the public.

Of course, this is not only true for expressions. It applies to any kind of goods addressing indistinct consumers on competitive markets. Consumers are noticed through media of the availability and characteristics of tradable goods and services.

However, expressions are specific because they consist of information, a non-rival commodity subjected to peculiar property rights. As long as copyrighted expressions were published on material supports, the symbolic dimension of the mediatisation was hidden by the economics of the support. The support was bringing rivalness to the expressions, which were then considered as the informational component of a delivery media (Watt, 2000). In a digital environment, the publishing of copyrighted expressions adds information to information.

This series of posts deals with mediatisation that is the operation carried by media so to give material or immaterial goods a public status. Mediatisation indeed covers signalling or advertising, but also all the kinds of information – design, brands, names, reviews, buzz, word-of mouth, social image – attached to a public or tradable item.

Our thesis is that mediatisation is an information complement attached to the mediatised item. It is a complementary good that alters – increases or decreases – the utility of the plain product. The complement is produced through a set of media packed with information externalities. In the analogue world, such ecosystems used to be rather stable. So were the relations between producers, media and distributors, allowing fixed arrangements to internalise the externalities. In the digital environment, media ecosystems get continuously denser which makes internalisation strategies a moving target.

Mediatisation adds meaning to functional goods or services. In the case of copyrighted expressions, this meaning is added to products whose utility is already meaning. Besides, as we will show, the dematerialisation of media tones down the borders between mediatisation and distribution. Hence, mediatisation economics are even more difficult to assess for copyrighted goods than for material ones.

The rest of this series of posts (to be released in the coming weeks) is structured as follows:

• The second post recalls the economic literature presenting advertising as a complementary good instead of a transaction cost. This approach will be extended to unpaid advertising or mediatisation. This post also examines the mediatisation content and its production through what we call “publishing protocols”. Publishing protocols will be used to describe the information externalities that shape the media ecosystems.
• The third post describes the specificities of the mediatisation of copyrighted goods and the changes brought by digitization.
• The fourth post underlines the importance of mediatisation economics and its impact on creation diversity. It also presents our concluding comments, regarding the internalisation function of copyright in media ecosystems.

Source: triplepundit.com

A number of posts on this blog have already dealt with multisided platforms (see here) and with crowdfunding (see here). My aim with this post is to link the two topics and show that crowdfunding platforms are a prime example of multisided platforms.

Crowdfunding platforms (CFPs) facilitate the interaction between entrepreneurs trying to raise funds and consumers/investors willing to participate in the financing of new projects. Why can they be seen as multisided platforms? As Evans (2011) explains it, a business opportunity emerges for a multisided platform when three conditions are met:

1. There are distinct groups of customers.
2. A member of one group benefits from having his demand coordinated with one or more members of another group; in the jargon of the economics literature, it is said that each group exerts indirect, or cross-side, network effects on the other groups.
3. An intermediary can facilitate that coordination more efficiently than bilateral relationships between the members of the groups.

As I now show, CFPs meet these three conditions.

1. Distinct groups of customers

CFPs link at least two distinct groups: entrepreneurs (fundraisers) on one side and contributors (funders) on the other side. Some platforms have brought an additional side on board, namely sophisticated investors (such as venture capitalists, business angels, and institutional investors). For instance, two Belgian platforms have chosen this route: MyMicroInvest allows projects to be funded by the crowd together with a professional venture capitalist; Angel.me has established a partnership with the bank Belfius. The objective is clear: the participation of sophisticated investors is meant to reassure individual contributors; because these investors have much larger capacities and experience to investigate the reliability and success probability of proposed projects, crowdfunders can infer useful information from the choices that these investors make. Note that the opposite may be true as well: sophisticated investors may use the “wisdom of the crowd” as an indicator of the potential success of a new product (something that they may have a hard time to evaluate otherwise).

In the same vein, Indiegogo (a CFP popular with hardware projects) has added some celebrity investors (such as Virgin Group founder Sir Richard Branson, or Pay Pal Founder Max Levchin) to its $40 million round in new funds from investors.

2. Presence of externalities

Each group’s valuation of the platform depends on the participation of the other group(s).

As far as contributors are concerned, there are two reasons for which they are likely to prefer platforms with a larger number of entrepreneurs: such platforms provide them with a wider set of campaigns that they can choose to support and, in the reward-based model, such platforms also increase the probability that contributors will obtain rewards that fit their tastes. Yet, another force may play in the opposite direction: the chances that any given campaign will be successful (i.e., will reach the required threshold) are inversely related to the number of campaigns that the platform hosts; for that reason, contributors may prefer platforms with a smaller number of entrepreneurs. We may conjecture that the former effects outweigh the latter (in particular if contributors can find some way to coordinate on projects that are more likely to be successful). It is thus reasonable to say that entrepreneurs exert positive indirect network effects on contributors.

The same obviously applies in the opposite direction: contributors exert positive indirect network effects on entrepreneurs: entrepreneurs value platforms that are able to attract larger crowds of contributors as they increase their chances to raise the targeted funds; platforms that attract a larger number of contributors are also more interesting because they allow entrepreneurs to showcase their products and to “test the waters” on a larger scale.

It is important to note that agents on both sides also care about what the members of their own group do; that is, within-side external effects are also present. Such effects are likely to be negative among entrepreneurs as they compete for the funds that the crowdfunders are willing to contribute: the more campaigns the platform hosts, the tougher the competition. In contrast, positive within-group effects exist among contributors: the project that a particular contributor has chosen to support is more likely to reach the required threshold the larger is the number of contributors who may choose to back this project too.

All these effects are summarized in the next figure:

Externalities on a crowdfunding platform

3. High transaction costs

As for the third condition, even though entrepreneurs may be able to connect with the crowd by their own means (here is an example), CFPs undoubtedly offer them both higher prospects of success and lower costs. In particular, CFPs are able to mitigate the problems raised by information asymmetries much more efficiently than any individual fundraiser could do on his/her own (on this topic, see Agrawal et al., 2013).

In sum, it seems clear that CFPs belong to the broad class of multisided platforms. If you want to dig deeper, you may want to examine how  CFPs differ from well-known multisided platforms such as dating sites, real-estate platforms, or videogame consoles.

This series of posts is motivated by the pricing model the telecommunications provider AT&T employs for its GigaPower broadband service. Just to recall, AT&T offers two different pricing plans that are based on a consumer’s preference for privacy. For instance, its standard service is priced at either $70 or $99 a month, depending on whether or not the consumer agrees to let AT&T monitor his/her surfing habits (see, for example, here or here). It thus appears that AT&T puts a price on privacy that corresponds to a monthly mark-up of approximately thirty percent.

As I argue in my previous post, the question of how to value private information is however far from obvious. In the following we will have a closer look at some theoretical and empirical approaches that study this topic in a more formal manner.

Theoretical Approaches

Rather than trying to give a precise answer to the question of how to value private information, theoretical approaches focus on the individual cost-benefit trade-offs of privacy related decisions. In this context, one example is the Study on monetising privacy – An economic model for pricing personal information by Jentzsch et al. (2012).

The theoretical part of the study focuses on the competition between two service providers that offer potentially differentiated goods to a group of consumers. Similar to what I explained before, a purchase is here seen as a composite transaction that not only involves the good itself, but also the disclosure of personal information. That is, a consumption bundle is composed of a good and a data requirement. The data requirement is a choice variable of a service provider. Thus, service providers decide about prices and whether to choose a high or a low data requirement. In case a high data requirement is chosen, the service provider receives some constant benefit per unit of the good sold. It follows that consumers select one of the two service providers depending on the price and their privacy concerns. Privacy concerns are exogenously given and for simplicity may either be high or low. A last ingredient of the model are the costs consumers face for disclosing their personal data. Those costs are such that a given consumer faces higher costs for a high data requirement and, further, that for a given data requirement costs are higher for consumers with high privacy concerns.

The study also looks at a two-period variation of the model that introduces the possibility of product personalisation. Consumers, at the end of the first period, may choose to have the good personalised in the second period. For this it is necessary to stay with the service provider that was selected in the first period. Personalisation involves a trade-off. On the one hand, it increases the value of the product to the consumer, on the other hand, it involves a cost in the form of an increased data requirement.

The authors test their model in different experimental settings and based on the results of the theoretical and empirical part derive, among others, the following implications.

• Different data requirements essentially function as differentiation devices that may allow service providers to obtain positive equilibrium profits even for otherwise undifferentiated goods.

• “If there are little to no differences in the prices offered by service providers on homogeneous goods, a competitor who has a reduced data requirement […] can obtain a competitive advantage […]. The reason is that consumers can – by choosing the service provider with a lower data requirement – reduce their costs of disclosure of personal data”.

• “Personal profiles are often the base for personalisation of products or services. If portability of profiles among firms is mandated, consumers will face decreased switching costs and benefit from intensified price competition in the market”.

Empirical Approaches

Empirical studies take the question of how to put a price on privacy more literally. One example is a recent study by the OECD that discusses six potential ways of estimating the monetary value of personal data. Those approaches may be divided into two broad categories: estimates based on market valuation and estimates based on individuals’ valuation. I implicitly discussed the latter in the beginning of my last post (Privacy attitudes, privacy behaviour and contextual effects). In the following I will therefore focus on estimates that are based on market valuation.

The most direct way to obtain an idea about the monetary value of your personal data is to examine the market prices for data. The previously cited OECD study summarises some estimates from different data warehouses in the US. As such different types of data trade at the following prices; personal address at $0.5, date of birth at $2, social security number at $8, driver’s license number at $3, military record at $35.

In an article in the Financial Times, Emily Steele provides some further insight into what what drives the market value of personal data.

Certain milestones in a person’s life prompt major changes in buying patterns, whether that’s becoming a new parent, moving homes, getting engaged, buying a car, or going through a divorce. Marketers are willing to pay more to reach consumers at those major life events. […] The more intimate the information, the more valuable it is. Some of the most personal and secretive troves of data rank as the most expensive.

You want to know the market price of your particular data? Then have a look at this article in the Financial Times that provides a tool that allows you to calculate the market price of your own personal data.

The study by the OECD suggests that, next to market prices, one may equally look at the market capitalisation, revenue or net income of a data record. Also an examination of the costs of a data breach may provide further insight into the monetary value of personal information.

This post only scratched the surface of a highly interesting, but equally complex topic. In your opinion, how should we value private information in the digital economy? Are there other approaches than the ones I mentioned in the post? And what are the advantages or drawbacks of the methods I presented? As a final point, how should we measure consumer surplus when privacy concerns are important? And what do you think is the impact of the recent developments in technologies such as data lockers that aim at giving consumers more control over their private information?

I started my previous post by asking how much you would be willing to pay in order to prevent your Internet provider from tracking and analysing your browsing history. This question was motivated by the pricing model the telecommunications provider AT&T employs for its new GigaPower broadband service. To recall, AT&T offers two different pricing plans based on a consumer’s preference for privacy. For instance, its standard service is priced at either $70 or $99 a month, depending on whether or not a consumer agrees to let AT&T monitor his/her surfing habits (see, for example, here or here). That is, if it were up to AT&T the answer to my question would be a monthly mark-up of approximately 30 percent.

Of course, my question was rather a theoretical one. And indeed, although the question itself sounds simple, the issue of how to value private information is highly complex. That is why it is in my eyes instructive to first introduce you to some basic notions of what is called the Economics of Privacy. The third and final part of this series will then discuss some theoretical and empirical approaches that try to shed light on how to value private information in the digital economy.

Willingness to pay versus willingness to accept

I want to start with a small experiment. In the beginning of this post I asked how much you would be willing to pay in order to prevent your Internet provider from analysing your browsing history. Take a moment to think about this question and write down the monthly amount. Now let us look at the problem from a different angle. How much would your Internet provider need to pay you so that you accept this invasion of your privacy? Again, write down the monthly amount. If you compare your two answers, you might realise that they do not coincide.

This experiment illustrates that it is crucial to distinguish between an individual’s willingness to pay (WTP) and willingness to accept (WTA). Or, to put it differently, how much individuals are willing to pay in order to protect their private data typically does not coincide with the amount of money that makes an equivalent decrease in their privacy acceptable. And indeed, as a recent study by the OECD points out;

Generally, the fraction of consumers who will reject an offer to obtain money in exchange for reduced privacy is larger than the fraction of consumers who will accept an economically equivalent offer to pay money in exchange for protection of privacy (Acquisti et al. (2009)).

The importance of distinguishing between WTP and WTA is further highlighted by the fact that in our daily life privacy decisions are typically present in both forms (Acquisti et al. (2013)). In some situations we have to decide how (or whether) to secure our personal data (WTP), while in others we are asked to reveal private information in order to receive some type of benefit (WTA).

These differences between an individual’s WTP and WTA cannot be explained by standard economic theory. As Kahneman et al. (1990) explain;

the standard assumptions of economic theory imply that when income effects are small, differences between an individual’s maximum willingness to pay (WTP) for a good and minimum compensation demanded for the same entitlement (willingness to accept [WTA]) should be negligible (Willig 1976). […] and there is wide acceptance of the Coase theorem assertion that, subject to income effects, the allocation of resources will be independent of the assignment of property rights when costless trades are possible.

What then could explain the observed WTP-WTA discrepancy? A potential answer to this question is given by Kahneman and Tversky (1979) and Thaler (1980) and comes in the form of the endowment effect. In a nutshell, the endowment effect captures the observation that individuals attach value to objects simply because they own them. Meaning, property rights do matter. The presence of such an endowment effect may in turn be explained by the concept of loss aversion (Kahneman and Tversky (1984)). According to the latter individuals attach more importance to losses than to gains. Acquisti et al. (2013), apply this concept to privacy concerns.

Applied to privacy, this explanation of the WTA/WTP gap would predict that someone who enjoyed a particular level of privacy but was asked to pay to increase it would be deterred from doing so by the prospect of the loss of money, whereas someone who was asked to sacrifice privacy for a gain in money would also be reluctant to make the change, deterred in this case by the loss of privacy.

Composite transactions, salience and privacy preferences

To illustrate how privacy concerns may matter for our daily consumption decisions, let us have a closer look at online purchases. Online purchases may best be described as composite transactions. So as transactions between a consumer and an online service provider that not only concern a particular good or service, but also, as a by-product, information (Jentzsch et al. (2012)). Following Jentzsch et al. (2012), composite transactions (T), such as an online purchase, may consequently be expressed as the sum of the transaction of the good (GT) and the transaction of the information (IT), i.e., as T=GT+IT.

One way to add privacy concerns is to include a parameter x in the latter expression so that T=GT+(1-x)IT. Here, x is a salience parameter (DellaVigna (2009)) that expresses the relevance of the information disclosure in the overall context of the transaction (and by this a consumer’s privacy concerns or privacy awareness). For x=0 a consumer attaches equal importance or attention to the good and the information, for x=1 the information disclosure does not matter. Intuitively, the salience parameter may not only vary across transactions or online service providers, but also across individuals. All in all, this illustrates in a simple way how different individuals may choose different service providers depending on their privacy concerns (we will see a theoretical model that formalises this point later on). Moreover, it accounts for the fact that privacy concerns need not to be stable, but may vary across transactions. This directly brings me to my next point.

Privacy attitudes, privacy behaviour and contextual effects

Judging from the public discussion surrounding topics such as Facebook’s privacy rules or the NSA scandal, we are living in a society that attaches a high importance to digital privacy. This is also supported by various studies (see e.g., Harris Interactive (2001) or Jentzsch et al. (2012)) that report privacy concerns for the majority of the surveyed individuals.

There is, however, no lack in anecdotal evidence that calls this conclusion into doubt. To give an example, more than 70 percent of office workers at London’s Liverpool Street station were willing to exchange their password for as little as a bar of chocolate in a 2004 field experiment by Infosecurity Europe (see here for a summary). Rest assured, this example is not a particular case. Also other studies show that supposedly privacy-concerned individuals are happy to exchange personal information for small rewards (see Acquisti et al. (2013) for an overview). From a WTA perspective individual privacy valuations hence appear to be low. Studies that focus on the WTP draw a similar picture. Here, the bottom line is that although the majority of the surveyed individuals expresses privacy concerns, only a small percentage is willing to invest in the protection of their personal data.

If based on this evidence one were to draw a conclusion regarding consumers’ privacy preferences, one would come to the paradoxical and maybe disillusioning result that consumers place little value on their personal data. Once again Behavioural Economics comes to the rescue and argues that the apparent discrepancy between privacy attitudes and privacy behaviour may be explained by contextual effects. In other words, the value consumers attach to their personal data most likely depends on things such as the type of information or to whom the information is revealed. For instance, according to a recent study by PwC

Consumers are more willing to share broad demographic data and information about their use of media content because they see it as being less personal and more anonymous. However, consumers are less willing to share more sensitive information that might compromise their private interests, such as their web browsing history; information about their personal social lives, such as mobile texting data or call history; or information related to their identity or security, such as social networking passwords, banking or financial information or their social security number.

Interestingly, a set of experiments by Frog (2011) shows that consumers in the United States, India and China put more trust in financial institutions or telecommunications providers than in government agencies or social networking sites.

What about you? What personal data do you value? Do you observe differences in your privacy preferences depending on whether you look at the issue from a WTA or WTP perspective? And finally, do your privacy preferences vary across transactions or with the context? Do you have examples for which type of online transactions you are more privacy-concerned or privacy-aware?