Anthony Luzzatto Gardner1
(1)
London, UK
Anthony Luzzatto Gardner
On February 15, 2015, President Obama gave an interview to the technology news outlet Re/Code in which he made a statement that triggered immediate criticism from the European Union:
…sometimes [European] vendors…who, you know, can’t compete with ours – are essentially trying to set up some roadblocks for our [internet] companies to operate effectively there…oftentimes what is portrayed as high-minded positions on issues sometimes is just designed to carve out some of their commercial interests.
The president was in Silicon Valley to participate in a White House Summit on Cybersecurity and Consumer Protection. His remarks in the interview were probably aimed at US technology firms with which relations had become strained in the wake of the Snowden revelations about US government surveillance. Some of these firms were alarmed that many European governments and consumers no longer trusted US firms to handle their data and were shifting their business to European competitors. While the context of the interview was important, the remarks were unfortunate. Many European concerns about US technology firms were sincerely held, and even legitimate, and not merely motivated by protectionism.
I had heard the same protectionist charge leveled against Europe so many times before. In the late 1980s, many observers warned ominously that the effort to create a Single European market without barriers to the free flow of goods, services, workers, and capital was in fact an effort to create a “Fortress Europe.” According to this narrative, Europe would pursue an industrial policy, including regulations and standards that would discriminate against non-EU firms, as a means of nurturing European business champions.
To the contrary, the single market program dismantled barriers for EU and non-EU firms alike, deepened economic integration, increased European competitiveness, and made the EU into a global force for open, rules-based trade and investment. US firms have been great beneficiaries of this single market because they are unencumbered by historical, linguistic and emotional attachments. They were therefore naturally inclined to look at Europe, unlike some of their European competitors, as one big geographic area without internal borders.
The charges of protectionism and anti-Americanism have always been easy to make. When the European Commission found in 2015 that Ireland had to recover €13 billion (plus interest) from Apple for back taxes that it should have paid over a decade, Apple’s CEO Tim Cook remarked that “It’s total political crap…I think Apple was targeted here…[and] I think that [anti-US sentiment] is one reason why we could have been targeted.”1 Cook’s prior meeting with Margrethe Vestager, European Commissioner for Competition Policy, had gone reportedly gone poorly.
The finding against Apple may explain why President Trump told European Commission President Jean-Claude Juncker at the G-7 summit meeting in Canada in June 2018 that “your tax lady, she really hates the US.” President Trump had evidently forgotten the name of the “lady” or the fact that she was responsible for competition policy. When reporters asked her later about the comment, Vestager responded with her usual poise: “I’ve done my own fact-checking on the first part of the sentence, and I do work with tax and I am a woman, so this is 100 per cent correct.” But the second half of the sentence is “not correct,” she added, saying: “I very much like the US.”2
Trump’s allegation was Fake News. His attack seemed based on the view that no regulatory authority of a foreign jurisdiction should have the temerity to tax, let alone regulate, US companies. The president seemed to think that only the US should investigate alleged tax and antitrust abuses by US tech firms, regardless of where that abuse occurs. This is a misunderstanding of how US and EU antitrust laws work. Both jurisdictions regulate activities such as mergers and alleged antitrust infractions, even by foreign companies, when they have effects on their territories. Given its long history of extraterritorial jurisdiction, the US is perhaps the last country that should complain about foreign countries exerting jurisdiction over US companies.
As a lawyer practicing EU antitrust law in Brussels during 1992–1994 and then subsequently as government official and investor following EU issues, I never detected any evidence that the flag of the companies’ headquarters influences EU competition decisions. Academic studies confirm this, specifically that there is no anti-American bias. In the most exhaustive analysis to date, covering over 5000 merger control decisions over 25 years, the authors could not find any evidence that the European Commission had intervened more frequently, or more extensively, in transactions involving a non-EU or US-based firm’s acquisition of a European target.3
One dramatic recent example of how the flag of companies’ headquarters does not determine the result of EU competition decisions was the decision of the European Commission in 2019 to block the merger between the German and French high-speed train manufacturers Siemens and Alstom. Despite intense lobbying at the highest levels of the German and French governments urging the creation of a European champion to compete with state-backed rivals from China, the European Commission insisted that the merger would harm competition in markets for railway signaling systems and very high-speed trains. If the European Commission is as political as some of its critics suggest, it would have heeded the pressure of the EU’s most powerful members and would have cleared the merger.
Concerns About GAFA
Those looking for evidence of a concerted anti-American campaign against leading US technology firms in the fall of 2014 could point to some concerning developments, however. Government officials and the media were frequently referring to Google, Apple, Facebook, and Amazon in shorthand as “GAFA” in frequent warnings about the threat to Europe from US dominance in the digital economy. The fear only seemed to grow over time: The Economist, for example, carried two striking front covers in late 2017 and early 2018 that proclaimed “Social media’s threat to democracy” (depicting Facebook’s “f” trademark pointed like a pistol) and the threat from the “New Titans” (depicting the GAFA as giant robots eating every asset in their path).
The fear was understandable because Europe was feeling self-conscious about its failure to create many leading technology firms, especially online platforms, and because the wealth and market power of GAFA seemed threatening. Online platforms headquartered in the EU represent only a small fraction of the total global market capitalization of such platforms. Apple’s cash reserves are larger than the GDPs of many European countries, and the market capitalization of GAFA exceeds the GDP of the French economy. Sometimes I wonder what Americans would feel if nearly all large players in the US digital economy were European. Wouldn’t they feel slightly uneasy?
The well-connected German publishing industry appeared to be behind many of the repeated attacks on Google. In April 2014, Mathias Döpfner, head of Europe’s largest newspaper publisher Axel Springer that owns Europe’s best-selling newspaper tabloid Bild and the broadsheet Die Welt, penned an open letter to Google in which he alleged the company’s total dependence on Google: “Google doesn’t need us. But we need Google.” Google was operating a business model that “in less reputable circles would be called a protection racket” because it discriminated against competitors in its search rankings.4 He compared Google to Fáfnir, a character from a Norse myth that inspired Richard Wagner to compose the Ring of the Nibelung. The son of a dwarf king, Fáfnir, became so jealous of the family’s treasure that he killed his father to take sole possession of it. Greed turned Fáfnir into a dragon, who guarded the treasure with fire and poison.
At the end of November 2014, the European Parliament overwhelmingly passed a non-binding motion that called upon the European Commission to unbundle search engines from their other businesses. (This call was repeated in following years). While the motion didn’t mention Google by name, Google was clearly the target. Eric Schmidt, Chairman of Google, called Secretary of Commerce Penny Pritzker to raise his concerns and numerous members of Congress fired off letters in protest.
Although the reaction was overdone because the resolution had no legal effect, I too was rather concerned about the general climate. The European People’s Party issued a remarkable press release stating that “Even Uncle Google has to play fair…Europe must show its teeth against US giant groups…the Internet is not the Wild West.” The press release was objectionable because it fanned the flames of envy about the success of US technology firms and alleged (without any proof) that they were acting in a lawless manner.
It would have been extraordinary, and unacceptable, for the US Congress to issue a resolution instructing the Department of Justice or the Federal Trade Commission to take specific action in an ongoing competition case. Yet the European Parliament had basically done just that by instructing a European Commissioner, who had taken office only a few months previously, what measures to take in one of its investigations into Google’s business practices. In several calls to the European Parliament and in my public statements, I said that in a state of law the accused should be found guilty only after, and not before, a proper trial. The resolution was misguided, moreover, because it would complicate efforts to conclude a transatlantic free trade deal by undermining the support for it from the US technology industry.
When some members of the European Parliament suggested that I was lobbying for Google, I replied that I had recently bought a T-Shirt with the message “I Don’t Need Google: My Wife Knows Everything” emblazoned on the front. None of Google, the other GAFA companies or leading US technology firms ever asked me to lobby on their behalf at the time. Even had I been asked I would not have done so because it would have been counterproductive. Such lobbying would have confirmed suspicions that the US government takes its orders from Silicon Valley.
It would have been inappropriate for me to take sides between the US technology firms on the receiving end of the attacks and the US firms who were frequently bringing complaints against them to the European Commission. Most important, I felt that EU legal procedures deserve respect and should not be politicized. However, I did often publicly defend the principle that all US firms are entitled to an impartial and transparent hearing, based on objective facts, and that the playing field should not be tilted against them in order to give European competitors an advantage.
It is true that GAFA has been under fire in the EU for a long time. Google and Facebook (the “G” and “F”) have alternated as the main object of criticism by regulators, the former for issues relating to its dominance in search, and the latter for issues relating largely to data privacy. But there is no evidence that this scrutiny reflects anti-Americanism.
Facebook has been subject to just as much criticism in the United States as in Europe over the past few years over Russian ads used to influence the 2016 US national elections and over the misuse of Facebook customer data by itself and third parties. In his testimonies before the US Senate and the European Parliament in 2018, CEO Mark Zuckerberg was subject to hostile questioning, especially about the ability of Cambridge Analytica, a data analytics firm, to use personal information harvested from more than 50 million Facebook profiles without permission to target US voters with personal political advertisements based on their psychological profile. In 2019, the Federal Trade Commission launched an antitrust investigation of Facebook (and Amazon).
Google has also had to appear before the US Senate investigation about social media influence in the 2016 national elections. In 2019, the Department of Justice launched an antitrust investigation into Google (and Apple). Attorneys-general from 50 US states and territories have launched an investigation into Google’s dominance in both search and advertising. There are many US firms concerned about Google’s practices and some have brought complaints to the European Commission.
One consortium, including Oracle, Expedia, and Microsoft (before Microsoft subsequently dropped out following its settlement with Google), played an important role in convincing the European Commission to launch investigations into Google’s practices in November 2010. The consortium alleged that Google was manipulating its online search algorithms to display search results from its comparison shopping product (called “Google Shopping”) more prominently than results from its competitors. The Federal Trade Commission had decided unanimously in 2013 to terminate its two-year investigation into similar allegations when Google agreed to make several modifications to its business practices. Despite having decided in 2012 that Google’s actions did not warrant a fine and despite having agreed a settlement with Google in 2014, the European Commission subsequently reversed course, proceeding with its investigation and fining Google €2.42 billion in 2017 in a decision that Google appealed to the European Court of Justice.
Several US firms also helped convince the European Commission to open a second investigation into Google in April 2015, leading to formal legal proceedings one year later. That investigation related to Android, the mobile operating system developed by Google. The Commission alleged that Google was restricting competition by requiring third party manufacturers of mobile devices to pre-install Google’s search app and Chrome browser as a condition for licensing the Play Store (Google’s App Store for Android). The Commission also alleged that Google prevented manufacturers from selling smart mobile devices running on competing operating systems based on Android open source code and that it provided financial incentives to these manufacturers and mobile network operators to exclusively pre-install Google search on mobile devices. It fined Google €4.3 billion in July 2018 in a decision that the company also appealed to the Court of Justice. Rather predictably, President Trump tweeted after the decision that “I told you so! The European Union just slapped a $5 billion fine on one of our great companies, Google. They truly have taken advantage of the US, but not for long!”
Taxation of the Digital Economy
Apple, another member of GAFA, has been in the European Commission’s firing line for its taxation policies. Apple’s highly technical dispute with the EU is worth reviewing in some detail because it raises fundamental issues of joint concern to the EU and the US about how US technology should be taxed. The perception has been growing in Europe and the United States that some of these companies, especially powerful and profitable ones, are able to pay low rates of tax by arbitraging the differences in national tax systems because of their ability to do business without a physical presence.
The European Commission’s interest in Apple was triggered by public hearings held by the US Senate in May 2013 on tax avoidance by major US companies. That hearing was principally focused on the level of US tax paid by US multinationals from their international operations. However, the hearing also disclosed details on the amount of Irish tax paid Apple on its international revenues.
The EU’s investigation concluded in August 2016 that Ireland had granted Apple illegal tax benefits, enabling it to pay substantially less Irish tax than other similar businesses from 1991 until it changed its tax structure in 2015 following a change in Irish law. During that period, Apple had organized much of its sales operations in Europe from Irish subsidiaries. In that way, Apple attributed much of its sales and profits from Europe and Asia to two of its Irish subsidiaries.
According to the Commission, two tax opinions issued by Ireland enabled Apple to attribute nearly all profits from sales in these Irish subsidiaries to “head offices” that existed only “on paper” (because they had no premises, assets, or employees).
Those Irish subsidiaries were incorporated in Ireland but managed and controlled in the United States. Apple was benefiting from a mismatch in tax laws: The United States considered the companies Irish because it determines tax residency on the basis of where companies are incorporated; Ireland considered them American because it determines their residency on the basis of where they are managed and controlled. The result was that much of their profits was not subject to Irish tax on the one hand and was deferred from US tax on the other (until the US ended tax deferral in enacting tax reform in December 2017).
The Commission concluded that Apple should have attributed to Ireland and paid 12.5% Irish corporate tax on all of its European profits derived from its Irish subsidiaries’ sales into Europe and Asia. The difference between the amount of taxes that Apple should have paid and the amount it did pay in Ireland equaled €13 billion over the decade ending in 2014. Ireland therefore had to recover that amount from Apple (plus interest). Both Apple and Ireland appealed to the General Court in Luxembourg. A judgment was expected as this book went to press. An appeal and several more years of litigation appear likely.
The European Commission concluded that this tax arrangement was a hidden subsidy and that, as such, it distorted competition in a manner that contravened the “state aid” provisions (akin to anti-subsidy rules) in its competition laws. The practical effect of the subsidy, some critics claimed, was the payment to Apple of about €220,000 in cash annually for the creation of each of 6000 jobs in Ireland. Seen in that light, it was a questionable deal for Ireland.
Apple argued that the tax was due to the US and was deferred under US tax law. By the end of 2017, Apple had accrued $36 billion of deferred US taxes from its international business which became payable after US tax reform. Apple asserts that the European Commission’s finding violates the basic principle in international tax that profits are taxed where the value is created. “The vast majority of the value in our products is indisputably created in the United States – where we do our design, development, engineering work and much more – so the majority of our taxes are owed to the United States.”5
While the retroactive recovery of unpaid taxes was a long-standing feature of the European Commission’s “state aid” provisions, the gigantic size of the recovery request and several features of the decision made it noteworthy.
One such feature was the tension between the EU’s ability to investigate the use of “sweetheart” tax deals under its competition laws, on the one hand, and EU member states’ exclusive powers over taxation, on the other hand. The “state aid” provisions are a core tool of the European Commission to ensure that companies can compete in the single market on a level playing field. Even though Ireland stood to recover €13 billion (with interest), a huge amount for a country with a GDP of €280 billion, its government appealed because the decision challenged its taxing powers, a core element of sovereignty, and because it could undermine the long-term continued attractiveness of Ireland as a destination for foreign investment. Ireland was not the only EU member state to be concerned: Luxembourg, the Netherlands, and other member states had long offered international investors favorable tax treatment.
I believed then (as I do now) that Apple has a strong case. I contributed to a very detailed critique of the Apple decision that the US Treasury published on August 24, 2016.6 It argued, for example, that the European Commission had departed from prior EU case law and Commission decisions, specifically in its analysis of when a company illegally benefits from “selective” tax advantages not available to other companies. The US government had strong doubts about whether Apple was really getting a unique “sweetheart” deal and whether other companies had been unable to request similar tax treatment. It questioned the European Commission’s conclusion that a tax advantage can be “selective” merely because a multinational (rather than a national “standalone”) company benefits from certain tax arrangements.
The interests of the US government in this case were not principally about Apple’s specific situation. After all, Apple had skilled lawyers to defend it in EU courts. The government’s concerns have been about the impact of the European Commission’s decision on US-EU efforts to drive international consensus on key taxation issues (especially in the digital economy) and on the amount of US tax that it could recover from Apple if the EU’s decision were upheld. The more Apple pays in Irish corporate income taxes, the less it will pay to the US.
Among these key taxation issues was the highly technical area of transfer pricing that lay at the heart of the dispute. Transfer pricing refers to the rules and methods for pricing transactions within and between enterprises under common ownership or control; transfer pricing determinations are usually governed by the “arm’s length” principle that prices must be set as if the parties were independent from one another and reflect “market” conditions. The Treasury alleged that the European Commission was substituting its own “arm’s length” principle derived from EU law for well-established international norms established (under US and EU leadership) by the OECD and the G-20.
The Apple decision appeared to undermine the basic principle that profits should be taxed where economic activities deriving profits are performed and where value is created; in the case of Apple, that was the United States, where its intellectual property is housed and where the essential research and development occurs. According to the Treasury, the European Commission was expanding its role as enforcer of competition and “state aid” law into a new role of a supranational tax authority that can review member state transfer pricing determinations without prior notice about how it would apply its powers.
The Treasury also argued that the European Commission’s application of a novel “state aid” approach in a retroactive manner is inconsistent with the principles of fairness and the legitimate expectations of companies that rely on member state tax rulings. While the European Commission denied that the approach was novel, I thought that there were strong arguments as to why the decision should have had a prospective, rather than a retroactive, effect. With the comfort of Ireland’s tax rulings, Apple had invested a significant amount of capital and had created many jobs in Ireland. Ireland had gotten the “benefit of the bargain” that it struck with Apple; it seemed unfair that Ireland should become a windfall beneficiary of €13 billion (plus interest) in back taxes after it had benefited from Apple’s investment.
Surely, Ireland was well placed as a member of the European Union to understand whether or not its tax rulings were compatible with EU law. It was not persuasive to suggest, as the European Commission did, that Apple’s extremely low effective tax rate should have put it “on notice” that the tax rulings were likely to contravene EU law. As a result of the decision, companies may no longer choose to give any weight to tax rulings granted to them by member states without getting a “good housekeeping seal of approval” from the European Commission.
The Treasury also pointed out that any repayment of back taxes by Apple would be considered as foreign income tax and hence creditable against Apple’s taxes owed in the United States. The Treasury challenged the European Commission’s assertion that Apple’s offshore earnings, accumulating over many years to avoid paying US corporate taxation of 35% on remittance to the United States, would never be taxed in the United States; on the contrary, those earnings were simply deferred and would be repatriated as part of US tax reform. That is precisely what happened in 2018 when Congress passed the Tax Cuts and Jobs Act that imposed a one-off, obligatory 15.5% repatriation tax on foreign earned income.
Apple argued with justification that if the European Commission and EU member states don’t like certain tax regimes, they should change them. And indeed that’s just what Ireland did in 2013 when it modified its tax residency rules and eliminated the infamous “double Irish” tax loophole. Other countries, like Luxembourg and the Netherlands, have also eliminated certain loopholes. And the US followed suit in 2017 when it enacted US corporate tax reform which ended the long-standing US tax policy of deferral of international taxation for US multinationals. Any company has an obligation to observe the law and to pay taxes that are legally owed, but not more. That amounts to appropriate tax planning, rather than to tax evasion. Many companies, including some of leading US online platforms, have been proactively changing their tax structures to comply with new laws and minimize the risk of public scrutiny.
The United States, like the European Union, is increasingly focused on the critical issue of how their tax systems can properly fund social programs, including pensions and protections of labor, health and the environment. This is especially the case in the EU due to its rapidly aging population and the rising burdens on its working-age population to support retirees. Tax systems need to ensure that multinational companies pay their “fair share” of tax and that countries do not engage in destructive tax competition by pursuing “beggar thy neighbor” policies. The challenge is considerable in the case of multinational companies that have significant intangible assets, such as intellectual property, as they have an easier time shifting profits to lower tax jurisdictions.
According to the European Commission, companies with digital business models pay an effective average tax rate of 9.5%, less than half the tax rate of businesses with traditional business models. That figure is strongly disputed, not only by many leading US technology firms but also by the authors of the EU study on which the figure is based. Most seriously, the European Commission averaged the tax rates IP-heavy companies enjoy in a range of countries, without acknowledging that US technology companies actually pay the majority of their income tax in the US, where rates are higher than in the EU.
The United States and the EU have a common challenge in updating current international tax rules that were designed for the “brick and mortar” economy and do not properly take into account business models in which companies can supply digital services in a country without being physically present there. There is a growing belief that too much tax is paid where these services originate and too little where they are consumed and users generate significant value. The European Commission therefore proposed in 2018 a comprehensive long-term solution that would impose tax on businesses where they have a significant digital presence, even if they do not have a physical presence. That presence would be deemed significant if revenues from the supply of digital services, or the number of online users or business contracts for digital services, exceed certain thresholds. This solution is intended to shape consensus at the OECD and then at global level.
Recognizing that this solution will require difficult discussions with global partners, especially the United States, including about the need to change bilateral double-taxation treaties, the European Commission also proposed a short-term solution that would impose a 3% tax on gross annual revenues in the EU on specific digital services. That blunt instrument would likely raise only €5 billion across the EU, while undermining multilateral efforts at tax reform. The proposal has fizzled out because of significant opposition from member states, not only the “usual suspects” of Luxembourg, the Netherlands, and Ireland (who host many digital companies’ holding companies), but also other member states who are concerned about remaining or becoming high-tech hubs. Ursula von der Leyen, the current President of the European Commission, has favored the idea of an EU-wide digital services tax in 2020 if there is no global solution by then.
The risk of uncoordinated action on taxation of digital firms is real: Tired of the likely stalemate at the EU, several member states (including France, Italy, Spain, and the United Kingdom) have moved ahead with plans to impose tax on the revenues of digital companies, and more are likely to follow. That can only fragment the EU even further and lead to frictions with the United States. USTR has opened an investigation of the French proposed tax regime that might could lead the US to impose unilateral tariffs against French imports under the 1974 Trade Act. President Trump threatened to do just that, before calling agreeing a truce with French President Macron until the end of 2020 in order to give time for negotiations toward a global agreement. In the meantime, the French tax will accrue but will not be payable. If a global agreement is not achieved, the tax under the French law would become payable and the EU would retaliate against unilateral tariffs by the US. As a result, transatlantic trade relations would be significantly impaired. A similar UK digital tax is also complicating negotiations toward a US-UK free trade deal. While the US has legitimate concerns about these and other digital tax regimes, it would be better to bring complaints to the WTO than to resort to unilateral trade measures. Technology firms should also ask themselves whether it would not be wise to work with (and improve) the EU proposed digital tax regime rather than to play whack-a-mole with multiple national tax regimes.
The EU’s Digital Single Market
Just as some politicians and business leaders in the United States feared that the scrutiny of GAFA presaged some broader attack on US technology firms, some also feared that the European Digital Single Market (DSM) program would create barriers to US investment in favor of local incumbents. I never believed it would turn out that way, and it hasn’t. The DSM made just as much sense and is likely to be just as positive for US interests, as the 1992 Single Market Program that largely eliminated obstacles to the free flow of goods, services, people, and capital within the EU. One can perhaps quibble with some of the specific proposals, the lack of sufficient prioritization or the focus on more regulation when less regulation or other tools (such as self-regulation) might be appropriate; but the real risk is that the EU does not move fast enough in its implementation.
When launching the program in the fall of 2014, President Juncker announced that “the Internet and digital communications can transform our economies as profoundly as the steam engine did in the 18th century or electricity did in the 19th century.”7 That sounded like hyperbole, but it reflected reality: The growth of digital technologies contributed roughly 30% to economic growth between 2001 and 2011. As the new Commission was beginning its mandate in the fall of 2014, it was self-evident that an insufficiently dynamic digital economy was one key factor why growth, productivity and job creation in Europe had lagged behind the US.
A study by the European Parliament’s Research Service on “non-Europe” in the digital economy (i.e., the absence of a single rule book) showed that the creation of a DSM would increase economic output by between €35 billion and €75 billion euros per year, would raise EU GDP by 0.4% and would create 223,000 jobs by 2020.8 Other studies showed that a deeper EU digital economy, harnessing the full potential of online services and digital infrastructure, could raise EU GDP by more than €500 billion euros. The boost to growth from a DSM would not be evenly distributed, however: The largest benefits would be concentrated in the so-called Digital Nine group of largely Northern European countries.9 The impact of DSM, therefore, made nearly every other growth-boosting initiative (including a US-EU free trade agreement) pale by comparison.
It was very revealing that Germany chose to place its Commissioner, Günther Oettinger, as the head of the digital portfolio at the beginning of the Juncker Commission. Berlin clearly understood that this portfolio, despite its comparative lack of glamor (compared to the globe-trotting role of EU foreign minister that Italy was keen to claim), was the real epicenter of influence over the EU’s economic future, including the competitiveness of the German economy. The importance of the issue extended far beyond the service economy but included the future of value creation in heavy industry, as the German automobile industry well understood. This industry was living in fear that US technology firms would dominate the value-added software of automobiles and relegate German automakers to the role of commodity metal-bashers with low margins. The fact that Tesla’s market capitalization equaled that of Volkswagen and BMW combined in February 2020, despite selling for fewer vehicles, illustrated the concerns.
President Juncker identified the creation of a DSM for 500 million EU citizens—the creation of a single rule book replacing 28 different national regulatory regimes relating to digital services—as a top priority to stimulate cross-border sales, more choice of goods and services at lower prices, investment in new technologies, productivity gains, entrepreneurship and ultimately job creation. The more active regulatory approach in the EU compared to the US on issues relating to the digital economy is largely driven by the need to prevent market fragmentation and boost growth.
A DSM would enable individuals and businesses to access cross-border digital services irrespective of their nationality or place of residence, under conditions of fair competition and with the benefit of a high level of consumer and personal data protection. One of the reasons for Europe’s sluggish economic performance has been the difficulty for consumers to buy and for small business to sell cross-border online; such transactions can be complex and expensive due to differing, and sometimes contradictory, consumer protection, data privacy, and contract laws. On top of these barriers, shopping habits are shaped by linguistic barriers and familiarity with home markets. As a result, only 15% of EU consumers buy online from another EU country, whereas nearly 44% do so domestically.
The DSM program covered a broad range of complex issues from its inception, ranging across many fields, including copyright, e-commerce, competition policy, online platforms, parcel delivery, telecommunications (including broadband services and radio wave spectrum management), data privacy, data flows, and cybersecurity.
One of the core reasons why the Digital Single Market is so important is that it can unleash private capital into the economy, especially in higher-risk and faster-growth technology sectors that have been held back by regulatory barriers and fragmented markets. While many actions lie with the member states, including reform of national tax, bankruptcy, and insolvency laws, the EU is making contributions in many areas, including the creation of a substantial pan-European venture capital fund of funds in 2018, the implementation of cross-border insolvency rules in 2017, and proposals to give small and medium-sized enterprises better access to financing through public markets.
The US government, including the US Mission to the EU, was watching closely as the European Commission rolled out sixteen detailed proposals grouped around three pillars of the Digital Single Market Program. The potential impact on US economic and political interests was (and remains) enormous.
The first pillar aims at promoting better access for consumers and businesses to online goods and services across Europe. One important initiative aimed at ending unjustified “geo-blocking.” That term refers to discriminatory practices used by businesses in one member state to block or limit cross-border access to their online interfaces (such as Web sites and apps), as well as discriminatory practices to apply different conditions of access to goods and services sold by businesses online or offline to customers from different member states. Other initiatives aimed at making cross-border e-commerce easier and reforming European copyright law to reduce the difference in national copyright regimes, allow for wider public access to copyrighted works (including for educational and research purposes) and address the use of copyrighted content (such as news and music) by online firms. The initiatives also proposed the simplification of value-added tax regimes; the improvement in the efficiency and affordability of parcel delivery; a general investigation into the e-commerce sector to identify competition concerns; and a review of measures to boost cross-border access to broadcasters’ services.
The second pillar aims at creating the right conditions for digital networks and innovative services to flourish. One initiative proposed measures to reinforce trust and security in digital services, notably concerning the handling of personal data. Other initiatives included a comprehensive review of online platforms that would cover issues such as the non-transparency of search results and of pricing policies, how such platforms use the information they acquire, the relationships between platforms and suppliers, and the promotion of platforms’ own services to the disadvantage of competitors. The initiatives also proposed a review of rules on audiovisual media services (including a quota for European content in on-demand catalogs) and an overhaul of the EU’s telecoms rules, including the more effective coordination of radio spectrum, the creation of incentives for investment in high-speed broadband and the creation of a more level playing field between “traditional” telecoms firms and “over the top” providers that distribute film, video and messaging content over the Internet.
The third pillar aims at maximizing the growth potential of the European digital economy by investing in information and communications technology (ICT) infrastructure and technologies such as cloud computing and big data, as well as research and innovation to boost industrial competitiveness. Proposals for action included important initiatives to promote the free movement of data within the EU and prevent unjustified data location restrictions for data storage or processing; to adopt common standards in priority areas, such as electronic health records (“e-health”), transport planning, mobile payments and energy efficiency; to study the emerging issues of ownership and access to non-personal data transmitted between businesses and generated by machines; and to ensure that more European citizens have the digital skills to seize the opportunities of the Internet and more governments harness digital tools to improve their public services.
The Digital Single Market is a vision that the US government and the vast majority of US business support and for good reason. Stronger European growth, resulting from a stronger digital economy and greater innovation, means more opportunities for investment and sales into Europe. Moreover, the DSM offers numerous opportunities for US and European businesses to collaborate and innovate. It also offers opportunities for the US government and the European Commission to adopt common, or at least compatible, policies that can set global standards for the digital economy. And if a DSM can contribute to the free flow of ideas, innovation, competitiveness, and lower youth unemployment, it will result in a more stable, peaceful, and democratic Europe.
There were, nonetheless, devils in the details. In my cables to Washington and my speeches to the US business community, I welcomed the DSM but added that we needed to keep a watchful eye on its development. There was the risk that certain powerful lobbies, especially in Germany and France, would seek to turn the DSM into a tool to advance their interests. Even within the European Commission, there appeared to be signs of conflict over the DSM’s objectives.
Andrus Ansip, Vice President of the European Commission and Commissioner in charge of the Digital Single Market, had previously served as Prime Minister of Estonia, one of the most innovative countries in the world in the field of digital services. He was clearly committed to a non-discriminatory and lightly regulated EU digital market; his mantra was that the DSM should aim to create openness and opportunities, not obstacles. He also saw opportunities, not risks, in transatlantic cooperation. In nearly every meeting with visiting business leaders or US government officials, he would tell stories of how he grew up in Estonia listening to Voice of America and believing in the United States as a beacon of liberty.
Commissioner Oettinger, who reported to Ansip in the European Commission’s structure but who carried significant weight because of the country he represented, was similarly supportive in front of an American audience. But when speaking to his home audience, especially the German publishing and automotive industries, his messages were different. He had, after all, been the Minister President of Baden-Württemberg, the headquarters of many iconic German car manufacturers. He would repeatedly stress that US technology firms were too large and powerful. European online businesses are too dependent on a few non-EU players, he warned. “This must not be the case again in the future.” It was necessary, he claimed, “to replace today’s web search engines, operating systems, and social networks with European ones.”10 Playing on European, especially German, emotions around data privacy, he warned that US online platforms like Facebook, Apple, and Google “will go to the member states where data protection is least developed, come along with their electronic vacuum cleaner, take it to California and sell it for money.”11
German Minister of the Economy Sigmar Gabriel routinely described the DSM as a way for the EU to set its own globally accepted digital standards and to achieve market dominance against US “digital imperialists” in order to achieve “digital sovereignty.” This Germanic view was often shared at the highest levels of the French government. Minister of the Economy Arnaud Montebourg warned that Europe risked becoming a “digital colony of the global internet giants.”12 More recently, French President Macron has also argued that France is fighting a battle for digital sovereignty. He has launched a $5.5 billion fund to help start-ups scale in key technologies.
I appreciated that many European (and Americans) have real concerns about the digital economy that are as much cultural as they are economic. Many are not comfortable that critical decisions affecting their lives are being taken by powerful foreign companies without being able to shape them. For example, many Europeans are not happy with a deregulated model that permits abuses from privacy intrusions, decisions based on black-box algorithms, and winner-take-all outcomes.
Nonetheless, as ambassador I also frequently heard the narrative of a zero-sum DSM. According to this narrative, Europe could only lose if US technology firms (especially online platforms) flourished and, conversely, Europe could only “win” by handcuffing their US competitors. A climate of fear and envy was not what Europe needed. It was no good looking into the rear-view mirror and asking: “Why don’t we have a Silicon Valley? Why haven’t we created a Googlesmann, an Apfel or an Amazonne?” Silicon Valley is the result of many unique factors that will be extremely hard to replicate, including an ecosystem of talent, a culture of risk-taking, favorable tax, labor and bankruptcy laws, access to capital and a large, unified internal market.
Efforts to promote “digital sovereignty” through regulation to tilt the playing field against non-European companies, identify national champions through an industrial policy, and restrict data flows by forcing companies to store their data locally or in a European cloud are not what Europe needs to become competitive in a global digital economy. On the contrary, these policies would squander the transformative potential of big data analytics (the examination of large amounts of data to uncover hidden patterns, correlations, and other insights), the Internet of Things (the interconnection via the Internet of computing devices in everyday objects, enabling them to send and receive data) and artificial intelligence.
Europe needs to promote innovation through increased research and development, as well as to invest in ICT and broadband services and connectivity. According to one study, if four of Europe’s six largest economies (France, Germany, Italy, and Spain) could raise their “digital density” (the level at which they consume, process, and share data) to the height of the United Kingdom, they could add roughly €200 billion to their economies. If those six countries could reach the level of “digital density” of the United States, they would increase their annual economic output by €460 billion.13 Moreover, Europe needs to promote entrepreneurship and digital skills; around 40% of people in the EU workforce do not have adequate digital skills; 14% have no digital skills at all.14 And Europe needs to promote more liquid equity and debt capital markets and encourage entrepreneurship.
The problem with the zero-sum narrative of the DSM is that it is based on an overly gloomy assessment of Europe’s achievements. The narrative doesn’t give enough credit to many noteworthy start-up successes in Europe: During the period 2000–2014, Europe produced 40 technology companies worth more than $1 billion (so-called unicorns), only slightly less than the United States.15 Founded in 2006, Dutch payments processing firm Adyen garnered worldwide attention when it launched an €7 billion IPO on the Euronext Amsterdam stock exchange in 2018. Sweden’s Klarna, an online payments solutions provider, is planning an IPO that may be of similar size. There are many other examples, including household names such as Sweden’s Spotify, France’s Deezer, the UK’s King Digital, and Germany’s Zalando. There are flourishing hubs of entrepreneurial start-up talent in places like London, Berlin, Stockholm, and Paris (especially following the election of Emmanuel Macron as President of France).
Moreover, success in the digital economy is not only about creating or maintaining strong positions in the markets for search, e-commerce, and social networking. While these are high-profile, consumer-facing parts of the value chain, industrial, business-to-business, and back-office digital services, as well as the hardware and infrastructure that supports them, are no less important. The reality is that Europe is seizing the huge opportunities that digitalization offers in sectors as varied as financial services, including cryptocurrencies, insurance, health care, manufacturing, design, and engineering. The zero-sum narrative also ignores the fact that many European businesses, including thousands of small and medium-sized businesses, have benefited from participating in the ecosystems created by US digital firms.
I was also concerned that the DSM would give opportunities for narrow interest groups to press for over-regulation that would create, rather than eliminate, obstacles. To avoid that outcome, the interagency group established by the US government to analyze the DSM regularly stressed several high-level messages to the European Commission. Here are a few examples:
When the DSM was announced, it rapidly became evident that the fault lines were mostly dividing US business interests, rather than splitting one side of the Atlantic from the other. US businesses had different views about the need to regulate online (largely US) platforms in order, for example, to ensure the fairness and transparency of search and the collection and usage of data. They differed about whether Internet platforms should compensate publishers for the use of short quotes of copyrighted news content (“snippets”) and share with content creators more revenues derived from user-uploaded audiovisual content. They also differed about proposals to oblige online platforms to assume greater liability for illegal content (such as pirated music videos).
During my mandate in Brussels, online platforms were the regular subject of impassioned speeches and commentary that failed to recognize their many forms, each requiring specific and detailed analysis of benefits and harms. Early definitions were imprecise, applying to a vast array of online actors whose activities were far removed from those of the GAFA firms being targeted in the zero-sum narrative.
As one observer has rightly observed, “it is easy to be uneasy about Internet platforms” because they use vast amounts of data in mysterious ways and make enemies of incumbent firms that lobby for protection.16 Many of the concerns raised about online platforms in the DSM are being debated in the United States, and the giant US platforms have plenty of US enemies that approved of the DSM’s critical look. Although I didn’t want to take sides, I was concerned that the issue of online platforms, above all others, could spin out of control due to its emotional content.
Under pressure from their publishing and telecommunications industries, the French and German governments were lobbying the European Commission to take broad action to curb the power of online platforms. The governments’ key argument was that many large platforms had become critical gateways or “essential facilities”—akin to critical infrastructure like pipelines, rail, and ports—that required new legislation because existing legislation (including competition laws) was not “fit for purpose.” The “essential facilities” doctrine has a long history in EU (and US) antitrust law and typically requires that a monopolist owning a facility essential to competitors must provide (if possible) reasonable use of that facility to such competitors if they are unable to duplicate it. The French and German governments argued that new legislation would be necessary because in the fast growing and constantly evolving technology sector it might be difficult to show either monopoly power or a competitor’s inability to duplicate a platform.
One of the main problems with the effort to “do something” about digital platforms is that they come in so many varieties that a “one size fits all” regulatory approach makes no sense. Some, like Google and Facebook, monetize their free services to users primarily through advertising; some, like Uber, intermediate among various parties and charge a fee when they strike a deal; some, like Amazon and eBay, are open marketplaces that charge a fee when sellers find customers for their products; and some are “true platforms,” such as cloud businesses and app stores.17 Each of these models poses very different issues that require a different and nuanced response.
The European Commission has fortunately declined the invitation to extend the “essential facilities” doctrine to platforms. Rather than wield a sledgehammer, it seems more inclined to use a scalpel based on an in-depth look at the market. While it has remained convinced that existing antitrust law is capable of addressing most abuses in the online world, it has accepted the need for new legislation in certain areas. In 2015, the European Commission launched an investigation into the many forms of online platforms, including online advertising platforms, marketplaces, search engines, social media and creative content outlets, application distribution platforms, communications services, payments systems, and platforms for the collaborative economy. Its conclusions the following year acknowledged some of the important benefits of online platforms, such as revolutionizing access to information, efficiency gains, and increased consumer choice. They also recognized that simplifying, modernizing, and lightening existing legislation, rather than creating new legislation, might be the appropriate policy.
But the conclusions also highlighted concerns that such platforms impose unfair terms and conditions, such as unilaterally modifying the conditions for market access, including access to essential business data; unfairly promoting their own services to the disadvantage of suppliers with whom they compete; restricting suppliers’ ability to offer more attractive conditions through other channels; and failing to provide transparency about tariffs, use of data, and search results. And the conclusions noted that while competition policy contains flexible tools that can be applied to online platforms, there may be instances where suppliers to platforms can be “disproportionately exposed to potentially unfair trading practices” even when the platforms are not dominant. A specific concern related to the dependence of small and medium-sized enterprises on a small number of platforms that have access to “datasets of unprecedented size.”18
The concerns from the investigation led the Commission to propose legislation in 2018 relating to business users of online “intermediation” services (that enable businesses to offer goods or services to consumers). This legislation aims, for example, to increase the fairness and transparency with which goods and services are ranked on search sites. It seeks to clarify the manner in which business users can request access to data held by online service providers and it requires such providers to justify restrictions on the ability of business users to offer the same goods and services to consumers under different conditions through other sales channels.19
The issues of transparency, fairness, discrimination, and data access in the online world are certainly worthy of careful scrutiny and debate. So is the frequently voiced concern about “network effects”—the phenomenon that the quality of goods or services improves as increased numbers of people or participants use them. The downside of such a virtuous circle is that economies of scale may be self-perpetuating and very rapid in the online world, thereby shutting out potential competitors. But such risks should be balanced against gains for consumers and business users.
Benefits from Online Platforms
These gains come in several forms, including improving the use of resources (such as by enabling the temporary use of assets), increasing competition and market efficiency, reducing transaction costs, and reducing asymmetries of information between buyers and sellers. Online platforms can help lower income and less privileged users because lower prices, access to information, more economic opportunities, and lower market barriers benefit them the most.
This is the case, for example, in the sharing economy for rides and accommodation. French President Emmanuel Macron has recognized that Uber has been one of the biggest employers of young (and often marginalized) people from minority backgrounds in the Paris region. The French job market (like many others in Europe) is polarized between insiders with secure long-term contracts and outsiders who live precariously from one short-term contract to the next. Uber identified and exploited a desperate need in Paris for rides, as anyone who has been stranded looking for a taxi in Paris knows all too well. In 2015, there were 18,000 taxis in Paris, fewer than in 1920, and the number did not adjust to meet demand at peak hours. This insufficient supply naturally led to high prices and an uneven level of service. That suited the founder of the private cab company G-7, who benefited from close relations with the pinnacle of French power and for years remained the only dependable, quality taxi service.
When Uber challenged the French labor law system built around contracts of indefinite duration, incumbent taxi drivers erupted in street violence. The French State publicly sided with them, but privately recognized that the sharing economy offered enormous flexible working opportunities and is the labor market’s fastest-growing sector. The dysfunction in the French taxi industry enabled Uber to grow faster in France upon its launch in 2011 than in any other country in continental Europe. Uber has spawned numerous French copycats (including Heetch, Chauffeurprive.com [rebranded Kapten] and Lecab.fr). As of 2015, around 50,000 people in France earn regular revenue from providing rides through online platforms. Many more do so through car-pooling platforms, such as through French “unicorn” BlaBlaCar. Fears that Uber would dominate the taxi industry have proven ill-founded, as competitors (including taxi companies themselves) are providing similar services, such as Lyft, MyTaxi, and Gett.
Travis Kalanick, the founder of Uber, was a genius in re-envisioning an entire industry to the delight of consumers. I have grown accustomed, like many of us, to the convenience of calling a car with a few taps on the screen of my smartphone and of paying for the ride through an automated debit to my bank card, as well as of receiving a summary by e-mail of the cost and itinerary of the journey. In my experience, Uber drivers almost always provide a high level of service, and I have rarely found one who is less than enthusiastic about earning extra cash in a flexible way.
Kalanick once described his opponent as “an asshole named Taxi.” He used similarly colorful language when he slouched in a chair in my office and described European regulators. That is grossly unfair, as there are real public policy issues that require careful assessment (including the significant investment that taxi drivers have made to acquire their licenses and the degree of control that an online riding platform can exert over its drivers without converting them into employees).
Regulatory authorities and courts in the US are struggling with these and other issues, just as they are in Europe. In 2018, the California Supreme Court made it harder for Uber to qualify its workers as contractors rather than as employees. The following year the California state legislature codified and broadened that decision and decided that anyone deemed an employee must be afforded broad protection under labor laws.
An astonishing 12 million French people have Airbnb accounts, about 500,000 of them as hosts; Paris is Airbnb’s leading destination, with about 65,000 listings. Naturally, not everyone is happy, including hotels, citizens concerned about tourist hordes in city centers and city administrations that want to crack down on unreported income. Barcelona, another top destination, was one of the early cities to curb the rapid growth in the supply of unlicensed short stay tourist accommodations. But the service has been a cash machine for hosts and has offered a more personal service increasingly desired by travelers.
Thanks to Google, I was able to research and fact-check my book from a terrace outside of Florence without ever going to a library. I am prepared to let Google collect some data on me as the hidden cost of that invaluable service; in any event, I can activate the “incognito” browsing mode within the Chrome browser if I want to surf the Internet without creating permanent trails of my activity, getting rid of my browsing history, cookies, site data, or other online data when I close the browser. Privacy settings enable me to choose what data gets saved and how long the data will be kept, as well as to delete my data altogether. Many other platforms have also made their privacy settings much easier to use. These positive developments are the results of legitimate concerns and pressure by consumers that they should remain in control of how their data is used and with whom it is shared.
While online platforms are, of course, capable of anti-competitive behavior and bad businesses practices, it is not evident why existing legal instruments available to injured parties and government regulators are insufficient to deal with actual harms. These platforms often face strong competition from incumbents, copycat businesses, and the ability of consumers to spend their money elsewhere. Perhaps the best guarantee against the risks of online platforms is to ensure an open and vibrant economy, one that encourages innovation and investment in ICT (especially broadband) and promotes the rapid scaling up of start-up ventures. There are many examples of online platforms like Alta Vista that used to be a dominant player in online search but disappeared when consumers flocked to more innovative and better alternatives.
One of the aspects often missing from the debate in Europe about the role of (mostly US) online platforms is that many European businesses have benefited from them. The launch of the iTunes App Store in 2008, for example, created an industry from scratch. According to a report prepared in 2014 for the European Commission, EU app developers earned €17.5 billion in 2013, a figure that was forecast to increase to €63 billion by 2018. The report predicted that the EU app developer workforce would grow from 1 million to 2.8 million over the same period.20
Amazon has enabled European small- and medium-sized businesses to sell billions of euros of their goods on the Internet, including cross-border within Europe and even across the Atlantic, for the first time. Amazon already employs around 83,000 people in Europe and is adding thousands more every year; since 2010 it has invested over €55 billion in infrastructure and operations.
According to a study by Deloitte in 2015 on Facebook’s global economic impact, Facebook enabled, directly and indirectly, $51 billion of revenue and 783,000 jobs in the EU.21 Thanks to the Grow with Google program, over three million people across Europe acquired new digital skills between 2015 and 2017; of these, 188,000 found a job or started a business within 14 weeks after their training. Thousands of European micro-entrepreneurs have launched their own YouTube channels and are making six-figure salaries annually. There are many other examples.
Conflict over Copyright
In addition to the issue of online platforms, there were a few other issues in the DSM that divided US businesses. Following the European Commission’s announcement in 2014 that it was planning its first big update of copyright rules since 2001, US content creators (especially in the music industry) joined with their European counterparts in a battle against (mostly US) Internet platforms about the correct balance between the need to share digital content, on the one hand, and the rights of content creators to be compensated fairly and incentivized to do their jobs, on the other.
The Directive on Copyright in the Digital Single Market that was ultimately passed after furious debate in 2019 will enter into force in 2021 when member states implement it into national law. The final text, far more balanced than the prior proposals, includes a provision requiring Internet platforms (especially Google News and Facebook) to obtain a license to use snippets of news articles. Hyperlinks and “very short extracts,” however, are excluded. News publishers have long argued that the news and other journalistic content is one of the main things that people search for online; such content has value and its distribution should be prohibited without a license, just as is the case with music, film, and books.
Google and the publishers claim that the other is getting the better of the bargain: Google states that it doesn’t sell advertising alongside news and that the publishers benefit from the traffic it sends to their sites (8 billion clicks per month to EU news publishers); the publishers claim that most readers of online news only read the snippets with few clicking on the links that take them to the publishers’ Web sites (where the traffic would generate advertising revenue).
There is no doubt that the state of the print industry is precarious everywhere, including in Europe. According to PriceWaterhouseCoopers, print revenues at Europe’s newspapers and magazines declined by €14 billion between 2010 and 2014, while digital revenues rose by €4 billion, leaving a €10 billion hole. A broad coalition of European newspaper and magazine publishers, especially Axel Springer in Germany, enjoyed enormous influence in shaping public opinion. The US print industry has had similar concerns about the use of its content by online platforms.
Critics of the EU’s legislative proposal on copyright argued that the parlous financial state of the print industry was due to factors unrelated to the use of news snippets. The industry’s business model was simply becoming out of date. For years, the industry had made money by being the central one-stop-shop for information about news and much else besides, such as movie listings and classified ads. With the emergence of many other information sources, including proliferating television, cable, online sites, and social media, that business (and associated advertising revenues) had shriveled. For years, the print industry had given away content for free in the hope of being able to monetize it through larger audiences, higher ad revenues and (eventually) paid subscriptions. But that had not worked out because audiences became accustomed to free content.
Critics also argued that the option of publishers to demand a fee could turn out to be counterproductive. They could point to two concrete examples. When the Spanish government had introduced a similar (mandatory) law in 2014, Google had simply shut down Google News in Spain. While readers continued to go directly to the Web sites of large newspaper and magazine publishers, the business of smaller news sites and blogs that rely on traffic from news aggregators and search engines declined. When Germany passed a similar (voluntary) law in 2012, Google responded by carrying news of only those publishers who agreed to be featured on its site for free. Those who refused saw their traffic plummet and eventually backed down.22
The US music industry also joined with their European counterparts in clashing with (mostly US) Internet platforms over provisions of the Copyright Directive concerning the liability of online platforms for pirated audio content uploaded by their users.
The US music industry had become increasingly disillusioned with the “safe harbor” provisions of the 1998 Digital Millennium Copyright Act (DMCA). Those provisions shield online service providers from liability for the activities of their users that infringe copyright as long as those providers are nothing more than neutral hosting providers (in other words purely passive intermediaries). One of the conditions of the safe harbor is that they implement “notice and takedown” procedures to remove content if they have knowledge of copyright infringing material, for example, because of notice from the owner of the content or of the copyright.
The DCMA, the music industry claimed, has proven to be completely ineffective. The “safe harbor” provisions, the music industry believes, are being interpreted too broadly, essentially providing protection to online sites that play an active role in curating or generating advertising revenues that monetize user-uploaded content. The provisions give digital platforms an unfair negotiating advantage in determining royalty rates for the use of copyrighted content. Moreover, according to the industry, the “notice and takedown” procedures don’t work. Of the hundreds of millions of DMCA notices sent to Google (the owner of YouTube), the vast majority relate to songs that have already received a prior notice (before being uploaded again).
The problem of pirated content is significant: The vast majority of music revenues comes from digital distribution (mostly from subscriptions). Online platforms have been major distributors of music and user-uploaded content services such as YouTube are a main source of music consumption. YouTube has argued that it has paid billions in royalties to the music industry (more than $3 billion in 2019 alone) and that the money represents “value added” because YouTube helps the recording industry monetize those consumers (80% of the total) who are not willing to pay for music.
As the music industry had been unable to amend the DMCA despite declining revenues (at the time) due to piracy, it looked to the EU with hope when it announced its intention in 2014 to re-assess similar “safe harbor” rules established under EU law in 2000.23 Many European songwriters, artists, and firms also lobbied heavily to defend their interests. The industry urged the EU to replace the “notice and takedown” procedures with a requirement on online platforms to implement effective “staydown” technologies—typically by means of software that automatically filters uploads—to detect whether content they are hosting contains copyrighted audiovisual material. Whereas YouTube already does this with its Content ID software, some smaller online companies objected that effective software is expensive, that they could not assume the legal risk of being liable for the wrongdoing of their users and that they would therefore have an incentive to block as much as possible.
The music industry believed that tougher requirements on online platforms to remove infringing copyrighted material would also strengthen the hand of the music industry in demanding more compensation for the use of its content, thereby reducing what it perceived to be a mismatch, or “value gap,” between the royalties that online platforms extract from paid-for subscription audio-streaming services (such as Spotify and Deezer) and from online sites like YouTube that generate advertising from content uploaded by their users.24 The final text of the Copyright Directive is a compromise, granting online platforms like YouTube an exemption from copyright liability only if they make best efforts to implement “takedown” and “staydown.”
While there were issues in the DSM on which US business was divided, the DSM also posed a few risks on which US businesses had an almost uniformly negative view. These risks included a requirement on streaming-video providers (like Netflix and Amazon Prime Video) to dedicate at least 30% of their on-demand catalogs to European content. They opposed, but failed to prevent, the insertion of this requirement in the revised Audiovisual Media Services Directive. The US entertainment industry (especially film and TV studios) did succeed, on the other hand, in partnering with like-minded European industries to prevent overbroad restrictions on “geo-blocking” that would interfere with their ability to distribute content on a territorial, rather than pan-European, basis (e.g., through exclusive licenses with the online platforms of Europe’s national broadcasters).
“Geo-blocking” needs to be distinguished from “data portability.” The latter term refers to the ability of consumers to access their online (copyrighted) content when they travel in the EU in the same way they access them at home. For instance, when a French consumer subscribes to CanalPlay films and series from an online service, the user will be able to watch them in France, or when he or she goes on holidays to Croatia or on a business trip to Denmark. The US content industry did not oppose the ability of European consumers to access their content throughout Europe, and the regulation entered into force in 2018.
When the campaign against “geo-blocking” was first announced, it appeared to assume that all forms of “geo-blocking” are unjustified and it targeted access to copyrighted and uncopyrighted products and services. As time went on, finer distinctions were made about what practices should be prohibited and which permitted. Certain forms of geo-blocking were clearly unjustified when applied in the context of uncopyrighted goods and services. It was clearly unacceptable, for example, for a Belgian customer who finds an attractive deal for a refrigerator on a German Web site to be unable to order the product and collect it at the seller’s place of business; or for a Bulgarian consumer who wishes to buy hosting services for her Web site from a Spanish company to be unable to order these services at the same price as a Spanish consumer; or for an Italian family visiting a French theme park to be unable to take advantage of the same family discount on the price of the entry tickets offered on the theme park’s French Web site.
Copyrighted material was very much targeted in early opposition to “geo-blocking.” In his manifesto laying out his priorities upon election as President of the European Commission, Juncker emphasized that consumers should be able to “access services, music, movies and sports events on their electronic devices wherever they are in Europe and regardless of borders.”25 The Juncker Commission considered discriminatory practices interfering with that ability as a significant cause of consumer dissatisfaction and as a contributor to fragmentation of the internal market. Vice President Ansip went even further, declaring that such practices are a violation of the fundamental rights of European citizens to be treated the same wherever they are located.
That view challenged the existing means of distributing digital copyrighted content in Europe. Whereas consumers can purchase physical products (such as CDs and DVDs) throughout Europe, they can’t always buy comparable online content because digital distribution rights have been licensed on a national basis. That means that consumers who want to watch movies or TV shows online are limited to the content that they are permitted to see in their home country. Netflix libraries may differ from one country to another, for example, in cases where a series is already licensed to the online platform of a local broadcaster. Apple has multiple iTunes stores around the EU and charges differing prices depending on the consumer’s residence.
European content producers and broadcasters resisted Ansip’s pressure. An executive for European broadcaster TF1 spoke plainly: “Pan-European licenses are not a viable business model for TV broadcasters since television markets are either national, regional or even local.”26 He went on to note that even Netflix realized this, leading it to invest in local stories with local actors. European film producers strongly backed this line.
In the United States, by contrast, most licenses for digital content cover the entire territory (except, for example, with regard to some live sports broadcasts). However, as US government experts explained to EU officials on several occasions, this has much more to do with culture and language than it does with any legal restriction. Legally speaking, content providers in the United States are free to make decisions about territorial coverage based on commercial demand, free from any government mandate to license nationwide. Indeed, US antitrust authorities point out that territorial limitations on copyright licenses “may serve procompetitive ends by allowing the licensor to exploit its property as efficiently and effectively as possible.”27
Like their European counterparts, US content providers, especially the film and television industry (among the largest US exporters), were also concerned that applying a prohibition on “geo-blocking” to audiovisual content would undermine their flexibility to negotiate distribution deals at a member state, multi-state or pan-European basis—choices shaped by Europe’s remarkable diversity of languages, cultures, consumer preferences, and purchasing power.
For smaller producers the risk was existential. Making content, such as films and TV series, is a risky activity, largely because of the need to commit funds up front (i.e., at the pre-production phase). Reducing the financing risk requires distributors and broadcasters to contribute toward production costs; but they are only willing to do so in return for territorial exclusivity. Smaller players with shallower pockets are especially sensitive to the risk of losing access to that critical financing model.
Mandatory pan-European licensing would interfere with the ability to sell content in different formats (theatrical, DVD, home entertainment, pay-TV) at different times and price points to maximize returns. And it would reduce the incentive to invest in the development of customized content for local markets, including dubbing films into local languages and developing original content for local markets. Some studies (funded by the content providers) have concluded that prohibiting “geo-blocking” of audiovisual content would lead to higher prices and a reduction in the amount, quality, and diversity of content production. Consumers in lower income member states would be hurt because they would pay a (higher) common EU price rather than a lower price set on the basis of national purchasing power.28
US content providers were joined by their European counterparts in making these arguments with the EU institutions. Their successful lobbying campaign removed copyright-protected content (including films, TV shows, music streaming services, e-books, online games, and software) from the Geo-Blocking Regulation that entered into force at the end of 2018. But these exclusions will be reviewed at the end of 2020.
In summary, the DSM has identified key ways for the EU to reduce its digital fragmentation and, as a result, to grow faster. It has been, and deserves to be, supported by US business and the US government. There have been very few instances where US business has had a unified and critical view of its proposals. Now that the DSM is largely on its way by 2020, the question is: What comes next? The answer is a renewed focus on transatlantic digital cooperation.
Transatlantic Digital Cooperation
I recently saw a world map that shows the intensity of Internet connectivity. Most of Europe and much of the United States were colored in bright red to show intense connectivity. Whereas some other parts of the world were also depicted in red, such as East Asia, India, Malaysia, Indonesia, and the East coast of Australia, the rest of the world was depicted in black to show weak or absent connectivity.29 The message from the map was clear: The United States and the European Union are natural partners in addressing the opportunities and challenges of the digital age. The statistics bear out what the map suggested. Over 80% of EU households have broadband connection (versus roughly 70% in the US), and 80% of EU citizens have smartphones connected to the Internet (similar to the US). Over 90% of European businesses are online (similar to the US).30
Transatlantic data flows are the highest in the world. They are 50% higher than the data flows between the US and Asia in absolute terms, and 4 times as large on a per capita basis. The US and the EU are each other’s largest partners in digital trade. The US exports roughly $200 billion in digitally enabled services to Europe and imports about $120 billion of such services from Europe, generating a trade surplus in this area of about $80 billion. And demonstrating the interconnectedness of US and European industries, over half of digitally deliverable services imported by the US from the EU were used in the production of US products for export, as well as vice versa.31
With the growth in mobile computing and the advent of the Internet of Things, big data analytics, and cloud computing, transatlantic data flows are projected to grow substantially over the next decade. But this growth will depend on whether the US and the EU can work together to promote flexible, interoperable standards for these services, promote high-speed Internet access and Internet-friendly rules. It will also depend on whether they can address a number of privacy and security issues to maintain the confidence of ordinary citizens, businesses, researchers, and innovators.32
It is clear that the Internet and the digitalization of everything are revolutionizing the economy and society globally. Cheap new communications technologies feature accurate sensors and enormous computing capabilities that are bringing more and more people and things online. These developments are not only relevant in the sector of high tech; they are fundamentally reshaping traditional industries like health care, finance, transport, agriculture, energy, and even retail. This point was driven home to me on one of my last visits to Seattle, the headquarters of Amazon and Starbucks.
Like many of us, I have long enjoyed the convenience of ordering books and a wide range of other goods and receiving delivery within two days (on the Amazon Prime service) with just a few clicks. While we may be concerned about the future of physical bookshops and the high street, the reality is that online shopping for books and other goods saves us the hassle of driving to a shop, looking for a book (and perhaps not finding it), waiting in line and paying in cash or by card, and then driving back home. Amazon is also investing in bricks-and-mortar bookstores, not just driving them out of business. Amazon’s stores make sense because the company knows more than any competitor about reading habits, even in the specific catchment area of every store, so that it can ensure that there is the right inventory. And, of course, it can also sell its wide range of other products, especially electronics such as the Kindle and Alexa.
I dropped in to experience one of the AmazonGo grocery stores that allow consumers to download an app, wave a barcode at a turnstile to walk in, select grocery items and simply walk out without waiting at a checkout counter. This is thanks to the company’s Just Walk Out technology that automatically detects when products are taken from shelves (and not returned), as well as keeping track of them in a virtual cart. Within seconds of my departure, a useful summary of my purchases appeared on my iPhone. Amazon has managed to completely reimagine the grocery shopping experience thanks to advanced digital technology.
On that trip to Seattle, I also dropped into one of the Starbucks Reserve stores. Seconds after I walked in, a smiling attendant approached me with a small tablet in hand to inquire what I would like to order. I was able to customize my order from numerous options for my cappuccino and pastry. After a few taps of his stylus on the tablet to note my choices (that were electronically communicated to the baristas), the attendant invited me to take a seat in a comfortable leather chair and said that my order would be brought to me. Starbucks has managed to reimagine the experience of ordering a coffee thanks to digital technology.
Consumer experiences like these shape our views about the utility of the digital economy. But it is elsewhere that the really life-changing developments will occur. Many experts believe that the Internet of Things, more commonly known as the “Industrial Internet” in Europe, will soon outstrip the value of the consumer Internet and will reshape the global, including especially the transatlantic, digital economy in profound ways in nearly every area of commercial activity. According to McKinsey, the IoT could add as much as $11 trillion per year to the global economy in nine different areas of application, most significantly in factories and cities.33
Reliance on data generated by sensors has been a long-standing feature in industry. IoT is new, however, because it features the ever-closer marriage of the digital and physical worlds through a combination of network connectivity, super-fast computing, the ubiquitous placement of cheap sensors on objects, and sophisticated data analysis techniques, known as big data, that allow insights to be derived from vast pools of data.
Around the world, there are estimated to be around 50 billion connected devices—including heavy machinery, vehicles, and home appliances—embedded with sensors and software that enable them to capture and communicate data. In the automotive industry alone, it is projected that 250 million (or one in five) cars worldwide will be connected to the Internet by 2020. Only a sliver of the “digital exhaust” generated by connected devices is actually used; of that amount, most is applied for the purposes of detection and control (i.e., maintenance of assets), rather than optimization and prediction of performance, where the real value lies.
IoT is capable of delivering efficiency gains across a wide range of sectors. Although consumer-facing applications such as personal health monitors and home appliances garner significant publicity, business-to-business applications are forecast to generate the bulk of the value. Of the total value that the IoT may generate, McKinsey estimates that about 40% will come from factories and work sites, with transport and urban infrastructure accounting for a further 30%.
The use of IoT is expected to boost productivity, as well as living standards, substantially in the same way as the Industrial Revolution and the Internet Revolution did. Although US labor productivity grew at 2% during the period 1995–2013, roughly double the rate in Europe, both sides of the Atlantic have struggled since then to grow productivity at a faster rate. If IoT could double productivity, then over the next twenty years it could raise average incomes by 25–40% above the current trend. It is hard to overstate what is at stake for both sides of the Atlantic: the competitiveness of our economies; our ability to pay for our high standards of environmental health and consumer protections; our ability to fund pensions for our swelling ranks of retirees; and even our ability to protect our democracies from the attack of populists and demagogues who are increasingly darkening our political landscapes.
Many companies on both sides of the Atlantic, especially in industrial sectors, are racing to seize the opportunities and are fundamentally reshaping themselves in the process. GE, like its European competitor Siemens, has been transforming itself into a “digital industrial” business. That has not only meant recharacterizing the company, but also focusing on data as a source of enormous value generation. As Jeff Immelt, former CEO of GE put it, a modern locomotive is not just a means of freight or passenger transportation; it is a “rolling data center.”34 Thanks to Predix, its software platform for managing and analyzing industrial data, GE is striving to become as much a software and analytics company as an industrial firm.
Europe is well placed to benefit from the Internet of Things. IoT applications may add $2.8 trillion to European GDP by 2030. European industry has significant technological assets to build on, including leadership in industrial robotics and factory automation, embedded digital systems, enterprise and design software, and 3D- and laser-based manufacturing. It has an installed base of over half a million large machines with rotating parts that can be made more efficient.
At the 2016 Hannover Fair, I saw many examples of how US and European businesses are partnering to reap the benefits of IoT and big data. For example, I saw how IBM and John Deere are piloting new technology in Europe that increases efficiency in the production of client-customized tractors and troubleshoots manufacturing problems. IBM has also teamed up with Siemens to develop a technology-solution that allows building owners to gauge a building’s performance and better forecast operational budgets. It also provides better predictive analytics for fault detection and diagnosis so that potential problems can be addressed before they emerge. I saw how Microsoft and Rolls Royce are partnering to improve aircraft efficiency and on-time performance through data analytics. Armed with up-to-date information about how an engine is performing during a flight, mechanics are ready to start on repairs as soon as the plane arrives at the gate.
One of the great opportunities and challenges facing policy-makers on both sides of the Atlantic today is how to promote a transatlantic partnership on IoT. Both the US and the EU are addressing the promise and challenges of self-driving cars, for example. The United States is working to help create consistent state regulations on autonomous vehicles across the United States and speed along testing of such vehicles. Similarly, the EU member states and the European Commission are working on developing common traffic and transport rules and a common digital communication system, so that cars in Europe can “talk” to one another and travel seamlessly across national borders. And both sides of the Atlantic are investing in Smart Cities, where digital technologies provide better public services for citizens, better use of resources, and less impact on the environment. This focus is appropriate in light of the fact that nearly 70% of the global population will live in urban areas by 2050.
A main focus of our transatlantic agenda should be to deepen the transatlantic digital economy by analyzing the prospects and challenges together, as well as implementing complementary solutions. Policy-makers bear significant responsibility to establish a regulatory context that facilitates the Internet of Things, while dealing with key economic and legal issues, including data privacy and data security, intellectual property rights, legal liability for the harmful outcomes of automated systems, the displacement of certain manual work by greater automation, and the risk that social and economic inequalities will exclude certain groups from the benefits of accessing data. We should be cautious about regulating on the basis of speculative concerns, rather than known, demonstrated risks.
The full economic potential of the Internet of Things can only be captured if the US and the EU continue to adopt policies that support expanded high-speed Internet access and the free flow of data. Although the European Commission has earmarked significant capital to research so-called fifth-generation (“5G”) telecommunications technology and has promoted a faster pan-European strategy for its rollout, Europe suffers from fragmented markets, complicating the ability of operators to see a compelling business case for investment, and a lack of an EU strategy toward the usage of spectrum. Instead of there being several large pan-European fixed and mobile operators, multiple (often sub-scale) players compete in each country. EU competition laws have prevented consolidation from happening based on concerns about price increases. In the United States, by contrast, three large players (AT&T, Verizon, and Comcast) have been racing toward nationwide coverage with download speeds of around 400 Megabits per second. China, South Korea, and some other parts of Asia are also moving fast.
The US and the EU should be making cooperation on 5G a priority. Such cooperation would focus on interoperability, namely through common standardisation and spectrum harmonization. Together they can drive further work in international bodies such as the International Telecommunication Union to set a global vision for 5G. The future of telecommunications and computing infrastructure connecting billions of users and trillions of devices will rely on whether they are successful.
The EU has successfully provided regulatory certainty around the free flow of data. The Regulation on the free flow of non-personal data in the European Union, passed in 2018, removes all restrictions imposed by member states’ public authorities on the geographical location for storing or processing non-personal data, unless justified on grounds of public security. The next step should be to ensure the free flow of data between the EU and the US.
Some Digital Challenges We Face
There are many challenges ahead. In addition to the transatlantic divergences about data privacy, described in Chapter 5, three are particularly significant. The first is that the IoT ecosystem will employ hardware and software from many different vendors. The ability of these devices and systems to work together is critical to realize the full value of IoT applications. Without interoperability, at least 40% of the potential benefits of the Internet of Things will not be realized.35
It would be detrimental to tie the IoT ecosystem prematurely to burdensome or conflicting standards coming from different jurisdictions. Collaboration in the development of international standards enables technological innovation by defining and establishing common foundations upon which product differentiation, innovative technology development, and other value-added services can be developed.
This collaboration is possible when standards are developed in organizations that are transparent in their decision-making and open to participation by all interested stakeholders. The development of open and voluntary standards, led by the private sector with active government participation, is more likely than government-led mandatory regulation to address issues with the speed and flexibility required in a fast-changing Internet environment and, therefore, to meet the needs of industry and government. Input from all experts, regardless of where they are based, should be welcomed.
The second major challenge will be for the US and EU to deepen their collaboration on cybersecurity. The Internet highway connects an increasing number of devices is therefore becoming increasingly vulnerable to those seeking to disrupt it, steal data or influence the outcome of elections. As one report points out, “Cyberattacks are becoming more prevalent, more sophisticated and – in an era of networked cars, medical devices, and appliances – increasingly capable of causing significant and potentially physical harm.”36 Moreover, US and EU ICT networks are interconnected and face cyber threats that are global in origin, indifferent to national borders and common to both sides of the Atlantic. The US and EU have increasingly collaborated to establish stronger frameworks to protect critical infrastructure and data-rich assets from, and improve resilience to, cyber-attack and espionage. Both sides of the Atlantic have adopted similar pieces of cyber legislation.37
The third major challenge will be to prevent the misuse of the Internet by international terrorist groups and those who propagate online racist and xenophobic hates speech. Leading Internet companies (such as Facebook, YouTube, Twitter, Microsoft, and Instagram) have agreed on a voluntary basis to work in an EU Internet Forum including governments, EU agencies, and academics to remove online terrorist content and to respect a Code of Conduct on Countering Illegal Hate Speech Online. Although the companies have made significant progress in satisfying the stringent obligations relating to the detection and effective removal of content, the European Commission introduced a Recommendation in March 2018 that included operational measures to remove illegal content more swiftly and effectively.38 It has warned that additional steps, including the proposal of binding legislation, might be required. The US government should look carefully at the lessons from the EU’s approach to determine how to attack the same challenge domestically.
The US and the EU have the skills, the technology, and the industrial base to drive a third innovation revolution, following the industrial and ICT revolutions. Each leads in specific fields: the United States in the Internet of Things and data analytics, the EU in quantum computing and telecommunications.39 The Digital Nine group of digital frontrunners in the EU is particularly well placed to exploit the opportunities of the digital economy. Thanks to the high degree of transatlantic digital integration, the US and the EU have the opportunity to define the architecture and rules of the evolving digital world. They have a shared stake in building a global digital marketplace based on openness, dynamism, and innovation. Ambitious chapters on information and communication technology (ICT) services and e-commerce in a US-EU free trade agreement would be important steps toward this goal. Both sides tabled proposals during the TTIP negotiations, but little progress was made.40
Cooperating with the EU on the global digital agenda is important for the United States for several reasons. The EU’s regulatory activism on digital issues impacts the interests of many US businesses, most obviously in Europe, the largest market for US investments and digital exports. This activism also has an impact abroad, outside of Europe. The recent case of the GDPR is a good example: Well over one hundred countries have adopted data protection laws broadly modeled on the EU approach. And this activism also has an impact in the United States. Some multinational US firms have adopted the GDPR as their data privacy standard, either out of conviction that it has been become the de facto global gold standard or because it is too difficult to maintain multiple data privacy standards. Leading US Internet platforms have also tailored their global terms of service to conform with the voluntary codes of conduct on hate speech that they have signed with the EU.
At times it may appear that there are divisions between the US and the EU on digital policy. These differences are marginal compared to the commonality of views between them, especially compared to the rigid command-and-control approaches to digital policy in China and Russia. Both countries appear determined to assume sweeping powers over the Internet and the collection, storage and use of data. By contrast the US and the EU are innovation-driven economies, with shared interests in spreading the rule of law, robust intellectual property protection, and sensible regulation.
Regulation in the EU of the digital economy, often impacting US firms and core US interests, are neither manifestations of anti-Americanism nor (except arguably in the case of taxation) of profound differences. Both the US and the EU are grappling with difficult issues such as: How should digital companies, among the largest and most profitable companies in the world, be taxed so that they pay their fair share? Do online platforms and data-rich companies pose specific challenges for antitrust policy that has been shaped in a pre-digital age? Should our copyright laws be updated to strike a different balance between the need to promote innovation and the diffusion of knowledge, on the one hand, and the incentivization of artists, writers, and journalists, on the other? What responsibilities do online platforms have to ensure that they do not host illegal material uploaded by their users?
In some cases, the EU has taken a more proactive approach than the US, inspired in part by US investigations (as in the Apple tax case) and by complaints brought by US firms. The EU has taken antitrust action against Google for behavior that the FTC cleared (although at the time of writing the Department of Justice has announced an investigation into Google). The EU has reformed its copyright laws to give publishers and content owners greater leverage against online platforms to require stronger action against pirated content and to pay higher compensation. Whereas several EU member states and the EU itself have passed laws criminalizing hate speech and imposing penalties on online platforms for hosting such content, the United States has not done so because of the guarantee of free speech in its constitution.
The political guidelines for the current European Commission indicate that the EU will continue with its proactive approach. Priorities include an EU-wide digital tax (if there is no global solution by 2020) and a new Digital Services Act that will “upgrade” regulations governing digital platforms, services, and products.41 Margrethe Vestager has been appointed a senior “Executive Vice President” of the European Commission to lead the EU’s efforts in these areas, along with her prior competition portfolio. The European Commission may well continue to set global standards in the digital economy, including standards affecting the responsibility of online platforms to remove illegal content, similar to what has happened for privacy in the case of GDPR. EU courts are promoting the global application of EU law. In one recent judgment, for example, the European Court of Justice ruled that nothing in EU law prevents national courts from requiring online platforms to search and delete duplicate posts of illegal content worldwide.
The Digital Services Act also appears aimed at overhauling the EU’s current rule book—largely contained in the 20-year-old E-commerce directive—governing the liability of online platforms for hate speech, other illegal content (such as child pornography and incitement to violence) and political advertising. A new centralized EU regulator may be created with significant powers over a large swathe of the digital economy, including Internet service providers, search, cloud services, and social media platforms. The Act might contain some far-reaching provisions such as a requirement that platforms subject their algorithms to regulatory scrutiny and comply with mandatory “notice and take down” orders forcing them to remove illegal content. Not only would the Act have a considerable impact on major US companies in Europe, it might also help shape regulatory outcomes in the US and around the world.
US businesses go to Brussels to litigate issues that will have an important impact on their business operations in Europe. They also do so in the hope of establishing EU norms that will apply in the United States and beyond. For example, new EU copyright rules that enhance the bargaining power of news publishers to demand payment for the use of news snippets or of music companies to demand more compensation for content uploaded by users on online platforms might eventually lead Congress to amend the “safe harbor” in the Digital Millennium Copyright Act. If it does so, however, the balance struck among competing interests is likely to be different than the one in the EU’s Copyright Directive.
In the area of taxation, the transatlantic rift threatens to grow wider. Actions by the EU and its member states to impose taxes on digital companies, based on their revenues wherever they have an online presence (even without a physical presence), could lead to similar action by other countries around the world. More than 110 countries have agreed to review the international tax system, parts of which have been rendered obsolete by the digital economy. There is deep disagreement about whether the EU approach is fair. The United States and other countries (including within the EU) have consistently opposed the creation of a different standard for Internet companies in which gross taxes are levied on revenue. The profound disagreement could undermine what has been a coordinated, global effort to crack down on tax avoidance. Worse still, it could help ignite a transatlantic trade war.
A global approach would be preferable, but the reality is that reaching global consensus is very (perhaps too) time consuming. The reality is that Europe will act soon, if not at EU level then at member state level. The question is whether US digital giants would be better off with one consistent set of pan-European rules that they can help craft or with multiple inconsistent national rules.
The EU’s proactive approach toward regulating the digital economy may also be explained by the fact that the EU, especially the European Commission, is widely perceived as having the tools to address public concerns in a swift and decisive manner. The EU institutions are widely criticized as incapable of responding to concerns about safety, migration, economic growth, and so forth. But on some issues of economic regulation, particularly competition law and state aid law, the EU is often perceived as more effective than the member states and, indeed, as the sole vehicle to apply pan-European solutions.
The EU’s proactive approach on the digital economy, moreover, reflects the EU’s essence as a rules-based organization whose function, in large part, is to write and enforce regulations. As the guardian of the EU treaties and like any other bureaucracy, the European Commission has a natural predilection to regulate. EU policy-making does not reflect the widely shared conviction in the United States about the ability of self-regulating markets to solve problems; to the contrary, it reflects the view—strongly held in at least a few key member states—that the market is a jungle that needs to be tamed and that government intervention is often necessary to ensure that narrow private interests do not hijack the broader public good. Despite the rise of populism in Europe, there is still a greater respect there for top-down solutions crafted by bureaucracies, and less trust in market-based solutions, than in the United States.
In order to influence the EU’s regulatory agenda, US firms need to engage with EU institutions early (i.e., at the stage of strategy papers and early proposals) because significant changes in EU regulations become rapidly less likely following internal agreement among the main legislative actors. Where possible, they need to partner with like-minded European allies. I witnessed frequently when I was in my post that strong and early engagement of this kind can help moderate European policy while ensuring greater compatibility across the Atlantic.
There are several critical areas—including artificial intelligence, blockchain, and cryptocurrencies—where regulation is still in its early stages and where the US and EU need to cooperate to ensure consistent regulations. Where regulations are inconsistent, they need to minimize the negative impact on the transatlantic digital market. Because of a change of policy under the Trump administration, for example, there is widespread skepticism that the “net neutrality” principles will be enforced in the US to the same extent as in the EU. According to those principles, Internet service providers should enable access to all content and applications regardless of the source and without favoring or blocking particular products or Web sites. It is important for the US and the EU to remain aligned on both regulation and enforcement to promote an efficient transatlantic digital marketplace for content and services. It is also important for the US and the EU to remain aligned regarding global “internet governance”—specifically the transition from an Internet largely under US control to the creation of a multi-stakeholder system that ensures that decision-making about the Internet does not fall under the control of governments.
One of my distinguished predecessors, Bill Kennard, has proposed twenty concrete steps in five areas to create a “transatlantic digital single market stretching from Silicon Valley to Tallinn.”42 One valuable recommendation is the creation of a US-EU Digital Council, directed by a senior official on the White House staff and a senior member of the European Commission. This council would play the same role in the digital sector as the US-EU Energy Council (described in Chapter 9) plays in the energy sector. It would serve as a platform for high-level discussion of critical regulatory issues and as an early warning system for regulatory divergences. It might prepare joint impact assessments of measures planned by either the US or the EU that have a clear and substantial impact on the transatlantic digital economy. Its mission would be to enhance the coordination of existing transatlantic digital dialogues43 and to break down bureaucratic silos.
It is in the interests of the United States that the DSM succeeds quickly. There are some areas in which the US can impart to Europe its learnings about how to create a dynamic digital economy. These include the development of innovation clusters, improving access to capital, and stimulating entrepreneurship through more flexible tax and labor policies. Even more significant would be an effort by the EU to disseminate the practices of the Digital Nine to the rest of the member states.
If the EU and the US do not work faster to strengthen their digital economies and to align their regulatory frameworks, they will be left behind by their competitors in Asia (especially China). Only by building a partnership, based on the two largest economies of the world and a population of 800 million consumers, can the US and Europe ensure that they retain leadership in this critical area.
Given its own challenges of being left behind in the race with Asia to exploit the opportunities of big data, the Internet of Things and artificial intelligence, the US has a fundamental interest in partnering with the EU because they share largely identical values about society and technology. Moreover, their combined consumer market is a far stronger base from which to compete with China in attracting investment and promoting innovation. The digitization of our economies is no longer a choice; it is an imperative. A deeper transatlantic digital market would not only help ensure that the US and the EU capture the benefits of higher growth rates and standards of living but would also help set the rules for the digital economy globally.
Footnotes
1
Interview with the Irish Independent, September 1, 2016.
2
https://www.politico.eu/article/margrethe-vestager-i-do-work-with-tax-and-i-am-a-woman-donald-trump-google/ and https://www.reuters.com/article/us-usa-trade-juncker/trump-told-me-youre-a-brutal-killer-eus-juncker-says-idUSKBN1JA2K.
3
Anu Bradford, Robert Jackson, Jr., Jonathan Zytnick, “Does the European Union Use Its Antitrust Power for Protectionism?” April 3 2018. https://promarket.org/european-union-use-antitrust-power-protectionism/.
4
Mathias Dopfner, “Why We Fear Google,” Frankfurter Allgemeine Zeitung, April 17, 2014.
5
“The Facts About Apple’s Tax Payments,” Apple Newsroom, November 6, 2017.
6
“The European Commission’s Recent State Aid Investigations of Transfer Pricing Rulings,” US Department of the Treasury White Paper, August 24, 2016. https://www.treasury.gov/resource-center/tax-policy/treaties/Documents/White-Paper-State-Aid.pdf.
7
“Europe’s Next Frontier: Creating Digital Jobs,” http://juncker.epp.eu/my-priorities/digital.
8
“The Cost of Non-Europe in the Single Market: Digital Single Market,” European Parliamentary Research Service, September 2014. http://www.europarl.europa.eu/RegData/etudes/STUD/2014/536356/EPRS_STU(2014)536356_REV1_EN.pdf.
9
The Digital Nine consist of Belgium, Denmark, Estonia, Finland, Ireland, Luxembourg, the Netherlands, Norway, and Sweden. One report in 2016 estimated that their growth rates until 2020 could be boosted by as much as 40%. Boston Consulting Group, “Digitizing Europe: Why Northern European Frontrunners Must Drive Digitization of the EU Economy,” May 2016.
10
“Europe v. Google: Nothing to Stand On,” The Economist, April 18, 2015.
11
Tom Fairless, “Europe’s Digital Czar Slams Google, Facebook,” The Wall Street Journal, February 24, 2015.
12
Juliette Garside, “From Google to Amazon: The EU Goes to War Against Power of US Digital Giants,” The Guardian, July 6, 2014.
13
Paul Hofheinz and Michael Mandel, “Uncovering the Hidden Value of Digital Trade,” The Lisbon Council, Issue 19/2015.
14
https://ec.europa.eu/commission/commissioners/2014-2019/ansip/blog/digital-skills-jobs-and-need-get-more-europeans-online_en.
15
Report by G. P. Bullhound, “European Unicorns: Do They Have Legs?” Independent Technology Research, 2015. http://www.gpbullhound.com/wp-content/uploads/2015/06/GP-Bullhound-Research-Billion-Dollar-Companies-2015.pdf?utm_source=Unicorns%20Report%20PR%2015.06.15&utm_medium=Pure360&utm_campaign=Unicorns%20Report%20PR%2015.06.15%20Pure360.
16
Joseph Kennedy, “Why Internet Platforms Don’t Need Special Regulation,” Internet and Technology Innovation Foundation, October 2015.
17
Cristina Caffarra, “‘Follow the Money’—Mapping Issues with Digital Platforms into Actionable Theories of Harm.” https://ecp.crai.com/wp-content/uploads/2019/09/e-Competitions-Special-Issue-Cristina-Caffarra.pdf.
18
Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, “Online Platforms and the Digital Single Market: Opportunities and Challenges for Europe,” May 25, 2016. https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52016DC0288&from=EN.
19
Proposal for a Regulation of the European Parliament and of the Council on promoting fairness and transparency for business users of online intermediation services of April 26, 2018. https://ec.europa.eu/digital-single-market/en/news/regulation-promoting-fairness-and-transparency-business-users-online-intermediation-services.
20
Mark Mulligan and David Card, “Sizing the EU App Economy,” Gigaom Research, February 2014. https://ec.europa.eu/digital-single-market/en/news/sizing-eu-app-economy.
21
Deloitte, Facebook’s Global Economic Impact, January 2015. https://www2.deloitte.com/content/dam/Deloitte/uk/Documents/technology-media-telecommunications/deloitte-uk-global-economic-impact-of-facebook.pdf.
22
Matthew Karnitschnig and Chris Spillane, “Plan to Make Google Pay for News Hits Rocks,” Politico, January 28, 2018. The impact of the law in Spain is detailed in http://www.aeepp.com/pdf/InformeNera.pdf.
23
Revenues of the recorded music industry in the United States have been recovering since 2015, including (in part) by streaming services. https://www.riaa.com/u-s-sales-database/.
24
According to IFPI, a trade association that represents the recording industry worldwide, the main user uploaded content services have around 900 million music users globally, compared with 68 million users for music subscription services such as Spotify or Deezer (that do not invoke the “safe harbour” provisions). The “value gap” is the difference between the €16 of average yearly revenue per user Spotify (the market leader in Europe in paid-for subscription audio-streaming services) generates for record producers versus the less than €1 of average yearly revenue per user YouTube generates for record producers. Figures as of May 2016.
25
http://juncker.epp.eu/my-priorities.
26
https://www.netopia.eu/cannot-force-creation-european-ghost-market-transmission-rights/.
27
US Department of Justice and the Federal Trade Commission, “Antitrust Guidelines for the Licensing of Intellectual Property,” January 12, 2017. https://www.ftc.gov/system/files/documents/public_statements/1049793/ip_guidelines_2017.pdf.
28
“The Impact of Cross-Border Access to Audiovisual Content on EU Consumer,” Oxero and O&O, May, 2016. https://www.oxera.com/getmedia/5c575114-e2de-4387-a2de-1ca64d793b19/Cross-border-report-(final).pdf.aspx.
29
http://www.businessinsider.com/this-world-map-shows-every-device-connected-to-the-internet-2014-9?IR=T.
30
Statistics date from 2016. “Transatlantic Digital Economy and Data Protection: State of Play and Future Implications for the EU’s External Policies,” European Parliament Committee on Foreign Affairs (AFET), July 1, 2016. http://www.europarl.europa.eu/RegData/etudes/STUD/2016/535006/EXPO_STU%282016%29535006_EN.pdf and http://www.telecompetitor.com/pew-u-s-smartphone-ownership-broadband-penetration-reached-record-levels-in-2016/.
31
Daniel Hamilton and Joe Quinlan, The Transatlantic Economy 2019 (Johns Hopkins University SAIS, 2019), p. ixx. https://transatlanticrelations.org/wp-content/uploads/2019/03/TE2019_FullStudy.pdf.
32
The growth in the global digital economy is also projected to be rapid. It is estimated that by 2021 global internet traffic will be 127 times the volume of 2005. “Cisco VNI Forecast and Methodology 2016–2021,” Cisco, September 15, 2017. The size of the “digital universe” (the data created and copied every year) will grow nearly 20 times during 2015–2025 to 180 zettabytes (or 180 trillion gigabytes). “Data Is Giving Rise to a New Economy,” The Economist, May 6, 2017, citing IDC, a market research firm.
33
“The Internet of Things: Mapping the Value Beyond the Hype,” McKinsey Global Institute, June 2015.
34
Ed Crooks, “General Electric: Post-Industrial Revolution,” Financial Times, January 12, 2016.
35
“The Internet of Things: Mapping the Value Beyond the Hype,” McKinsey Global Institute, June 2015.
36
“Building a Transatlantic Digital Marketplace: Twenty Steps Toward 2020,” Atlantic Council, April 2016.
37
The Network Information Security Directive (the NIS Directive) in the EU, and the National Institute of Standards and Technology Framework for Improving Critical Infrastructure Cybersecurity in the US (the NIST Framework). The US and the EU have been discussing cyber issues since the launch of the US-EU Working Group on Cyber Security and Cybercrime in 2010, formalized in the US-EU Cyber Dialogue launched in 2014.
38
https://ec.europa.eu/digital-single-market/en/news/commission-recommendation-measures-effectively-tackle-illegal-content-online. Under the Code of Conduct, Internet companies were (as of February 2019) removing an average 72% of illegal hate speech notified to them. About 89% of the notifications were being assessed within 24 hours and between 70 and 80% were estimated to be satisfactory removal rates.
39
G20 Innovation Report 2016, Organisation for Economic Co-Operation and Development, November 4, 2016. http://www.oecd.org/innovation/G20-innovation-report-2016.pdf.
40
However, the US and the EU have negotiated Trade Principles for Information and Communication Technology Services that set forth some of the key principles on ICT that could be reflected in a bilateral free trade agreement. http://trade.ec.europa.eu/doclib/docs/2011/april/tradoc_147780.pdf. Rules on e-commerce are also an important feature of the ongoing negotiations toward a Trade in Services Agreement (TISA) at the WTO. Such rules might cover issues such as the conclusion of contracts by electronic means, electronic trust authentication and electronic signatures.
41
Ursula van der Leyen, “A Union That Strives for More: My Agenda for Europe.” https://ec.europa.eu/commission/sites/beta-political/files/political-guidelines-next-commission_en.pdf.
42
Carl Bildt and Bill Kennard, “Building a Transatlantic Digital Marketplace: Twenty Steps Toward 2020,” Atlantic Council, April 2016.
43
The dialogues are the Information Society Dialogue, the Cyber Dialogue and the Transatlantic Economic Council. The Innovation and Investment in the Digital Economy Dialogue, launched in 2016 by Secretary of Commerce Penny Pritzker and European Commissioner Günther Oettinger, had only one meeting.