Can tax avoidance be anticompetitive?
Google has been in the news for more than one reason. For one, its tax planning has attracted the attention of the public media, and prompted European officials to charge Google with tax avoidance or even tax evasion. The record fines imposed on Google pursuant to the EU’s competition law have made at least as many headlines; Google has appealed, but more cases are pending. These seemingly unrelated events under different legal regimes may be connected, however. The fines levied against Google under EU competition law can be seen as making Google pay its ‘fair share’ to compensate for exploiting loopholes in tax law. However, if the objective is to make Google pay its ‘fair share’, why couldn’t this objective be achieved under tax law?
As discussed below in section 2, the foundation for the answer lies in the ‘corporate group’ approach. While the tax laws of European countries do not authorise the imposition of taxes on the profits of entire corporate groups, EU competition law endorses the concept of the ‘economic unit’, which enables the European Commission to fine corporate groups as a whole and target a non-EU parent. This explanation raises the second question discussed in this post: can the Commission’s reasoning to make Google pay under the EU competition laws be understood from a political economy perspective? Currently, the EU does not have power to tax corporations. By imposing fines on corporations under competition law, the Commission is able to extract money from corporations it cannot obtain otherwise. As discussed in section 3, the Commission’s reasoning to impose fines on Google might be a way of making Google pay its ‘fair share’.
- Google’s Tax Planning and Record Fines
Google, among other foreign tech giants such as Apple and Facebook, has been accused of avoiding taxes. With its European headquarters in Ireland, Google benefits from advantageous Irish corporate tax rules. For instance, from its 2015 annual revenue of €22.6 billion Google subtracted €16.9 billion in costs, including royalty fees. These fees, however, are paid to Google subsidiaries and make their way to the corporate tax-free Bermuda islands. From its remaining €341 million taxable profits in 2015, Google Ireland paid €47.8 million in taxes. The alleged scheme is known as the “Double Irish with a Dutch Sandwich”, which is perfectly legal under the domestic tax laws. However, due to pressure from European governments missing out on tax income and to avoid further scrutiny, Google reached settlements with the UK and Italian governments. The French minister of Finance denounced these deals as “more the product of a negotiation than the application of the law”. In contrast, France’s tax administration would “not negotiate the amount of taxes owed” as it would merely “apply the rules.” The French Administrative Court of Paris, however, did not find any infringement of the French tax laws on Google’s part. Instead, it has ruled that Google is in compliance with French tax law and international standards.
Meanwhile, the European regulators did not sit still. Bowing to their pressure, Ireland closed its tax loophole; companies currently using the structure may benefit from it only until 2020. In addition, the Commission submitted a proposal “to ensure that all companies pay fair tax in the EU”. One of these proposed measures is a “digital tax”, which would be imposed by EU Member States. This tax targets revenue from digital activities – activities “which are currently not effectively taxed,” but under the new tax law “would begin to generate immediate revenues for Member States.” In short, Europe felt that Google has not played by the rules, even though its ‘aggressive’ tax planning may be entirely legal. Through these tax avoidance schemes, as European governments see it, Google has failed to pay its fair share of taxes.
Over the last several years, the European Commission has gone after Google under the EU competition laws. It has launched investigations into several Google services, including Google shopping, Android and AdSense. Following these investigations, the Commission has decided to fine Alphabet Inc., Google’s parent based in Silicon Valley. In its latest Decision, the European Commission fined Google €4.34 billion for “illegal practices regarding Android mobile devices to strengthen dominance of Google’s search engine”. One year prior to the record Android fine, the Commission slapped Google with a €2.42 billion fine for “abusing [its] dominance as search engine by giving illegal advantage to [its] own comparison shopping service”, known as ‘Google Shopping’.
There is no one-to-one relation, however, between the taxes of which EU member states were deprived and the fines imposed by the European Commission. Fines imposed by the European Commission on companies are paid into the Community budget. Since these fines help to finance the European Union, they reduce the financial burden for the Member States. Furthermore, with the United Kingdom leaving, the EU has a gap to fill. Google’s deep pockets present an opportunity to raise substantial funds for the EU at a time when there is both a gap in the EU budget and pressure to reduce member state EU contributions.
- The Corporate Group Approach
Article 102 of the Treaty on the Functioning of the European Union (TFEU) prohibits that “undertakings” abuse their dominant position in the market. According to the European Court of Justice, the term “undertaking” refers to an “economic” – rather than a legal – “unit” – this is known as the ‘single economic unit’ doctrine. It does not matter whether this economic unit is managed from inside or outside the EU. An undertaking like Google consists of a large number of legal entities, which are managed from its Silicon Valley headquarters. Thus, if EU-based Google entities commit an act in violation of EU competition law, fines computed on the basis of the turnover of the entire group can be imposed on its parent, Alphabet.
Moreover, the amount of fines under EU competition laws is a function of the undertaking’s revenues. Under the Commission’s 2006 Guidelines on the method of setting fines, the “basic amount” of a fine is the sum of (1) a percentage of up to 30% of the relevant value of sales, depending on the seriousness of the infringement and multiplied by the years of duration, and (2) a percentage of up to 25% of the relevant value of sales, to ensure a deterring effect. Due to aggravating circumstances or an undertaking’s large turnover, the Commission may increase the basic amount. In setting the fine amount, the Commission has much discretion. In the Google Shopping case, the Commission based the value of the relevant sales on the revenue generated by ads and paid results on Google’s comparison shopping service in each of the thirteen national markets, but it did not disclose the corresponding numbers. In addition, to ensure a “sufficiently deterrent effect”, the Commission increased the basic amount of the fine by more than €500 million, based on the ‘particularly large turnover’ of Alphabet Inc. in 2016, approximately €81,5 billion.
Unlike competition laws, tax laws do not provide for a corporate group approach that results in the mandatory taxation of the profits of the entire group; each legal entity is in principle a separate taxpayer. Any mandatory corporate group approach that imposes a tax on the entire group’s profits would also be inconsistent with tax treaties, which carefully allocate the power to tax corporations to avoid double taxation, and would create strong disincentives for successful corporate groups to do business in countries other than those in which they generate substantial profits.
The corporate group approach should be distinguished from the ‘consolidation method’. Under the consolidation method, a parent company combines its own revenues with the revenues of its subsidiaries and submits one tax return. Such consolidation allows a parent to set off the profits of one subsidiary against the losses of another subsidiary with the group. Upon consolidation, the revenues of the subsidiary will not be taxed in the country in which the subsidiary is a resident. In international tax planning, consolidation is an option available to corporate groups; it is not mandatory, and, thus, will be used only if it helps a corporate group to meet its tax objectives.
A form of mandatory cross-border consolidation has been tried, but ran into multiple problems. In the 1970’s, the Californian legislature introduced ‘unitary taxation’ on multinational corporations doing business in California; other states followed California’s lead. Through a method called ‘Worldwide Combined Reporting’ (‘WWCR’), all of the multinational corporation’s domestic and foreign revenues and losses were (1) pooled together, and (2) allocated among the worldwide taxing jurisdictions. Although the US Supreme Court ruled the legislation constitutional, the European Commission claimed it violated tax treaties since California’s ‘arbitrary’ apportionment rules would expose corporations to double taxation. Due to pressure from foreign regulators and businesses, states amended their WWCR laws to only include income and losses from domestic affiliates of the unitary group (i.e. ‘water’s edge reporting’).
In 2011, the Commission proposed a Common Consolidated Corporate Tax Base (‘CCCTB’) within the EU, inspired by the unitary taxation approach. According to the CCCTB, the taxable profit from a company operating in the EU is based on the aggregate of revenue from the different member states and is then allocated between the member states. This system makes ‘profit-shifting’ and tax planning within the EU more difficult. The European Council blocked the proposal, however, mainly due to objections by the UK and the smaller member states. Five years later, the Commission revived the proposal. The new proposal requires the mandatory adoption of the CCCTB by multinational groups with consolidated annual revenues exceeding €750 million (such as Google, Apple, Amazon and Spotify). The proposal is now on the Council’s desk for further examination.
The bottom line is that current taxation regimes have enabled Google to avoid taxation of its substantial EU revenues, As discussed, the root cause of tax law’s inability to collect a fair amount of tax revenues from Google is the way tax laws treat corporate groups. Under competition laws, however, corporate groups are handled differently.
- Competition Law Fines as Compensation for Loss of Tax Revenues
EU competition law is vague, works with abstract concepts, and requires the application of theories, methods, and empirics about competition in the marketplace. There are multiple theories and methods, and there often is much empirical data available – selection thus is unavoidable. Consequently, although the application of competition law is intended as a neutral, objective act, the Commission has much discretion and policy or political choices or biases are unavoidable.
In applying competition law to EU-based corporate groups, the Commission may face political pressure from member state governments. The government of the member state where the corporation is a resident may push for ‘reasonable’ application and mild treatment, while national governments of the victims may argue for ‘strict’ application and substantial fines. Google has not made any significant investments in the EU, and is not a large employer in any member state. As a result, Google cannot count on much support from within the EU. To the contrary, Google is seen as a ‘big, bad bully’ that violates users’ privacy, avoids tax, and crushes competitors. The mindset of decision makers is not immune to such ‘associative priming’, which may introduce bias into its decisions.
These factors are likely to influence the way the Commission exercises its discretion in competition law cases. The Google shopping case is not a clear-cut competition law violation. The Commission’s decision is not supported by unassailable legal reasoning, and reflects choices. Its reasoning can be viewed as working towards the desired outcome: to make Google pay its fair share. Examples of this confirmation bias reasoning can be found in the Commission’s definition of the relevant product market and the strength of the empirical evidence regarding consumer behaviour. In addition, the Commission’s findings of harm and causation raise questions. Below, each of these examples are briefly discussed. They illustrate that competition law is no exact science, and that there is much judgment involved.
A Non-existent ‘Market’
The Commission found that there is a market for ‘general search services’, in which Google is deemed to be dominant. To reach this conclusion, the Commission had to dismiss the argument that Google faces fierce competition from ‘specialised’ or ‘vertical’ search websites such as eBay, Amazon, Yelp, TripAdvisor, Zalando etc. There is much empirical evidence that Google is subject to such competition – for instance, recently, Amazon surpassed Google as the prime destination for shopping queries. On feeble grounds, without proffering empirical evidence, the Commission found that such specialised search services cannot meet the consumers’ need in the general search services market. However, as the US Federal Trade Commission observed, vertical search engines present consumers with an alternative to Google for specific categories of searches. Indeed, specialised search services provide results relevant to users for the queries related to the category in which they specialise. And if a specialised search service generates better search results, consumers will switch to that specialised search service for future queries relating to that category of information. This explains Amazon’s success.
In addition, even if general and specialised search services are different, what really matters is how consumers search and what consumers consider substitutes, which requires empirical evidence. Without much proof, the Commission argues that these services are more like complements than substitutes. To support this argument, it refers to a decision from the German Monopolies Commission which states “[t]here are other types of information (such as products, hotels and restaurants) that can be found through general as well as specialised search engines.” This finding does not support the Commission’s theory, however, because it is a generic, theoretical statement, not disproof of substitution.
The Commission also invokes a study showing that many users click on the first link shown to them. One might think that this corroborates Google’s business model, which is based on providing the most relevant and high quality search results to its users. However, the Commission suggest that users do not click on the first generic search result because it is “objectively what they are looking for”, but because of “a propensity to click on such a result.” Consequently, the positioning of Google’s own comparison shopping service at the top of the page would constitute an artificial competitive advantage. To prove its point, the Commission refers to a study indicating that “the rank of a given link in generic search results on the first general search results page has a major impact on the click rates of that link, irrespective of the relevance of the underlying page.” The reference to this study is unable to support the Commission’s bold conclusion though. In the study the first and second links were switched, and the authors found that the second link then attracted the same number of clicks as the first link. This does not demonstrate, however, that it is only the ranking that determines clicks – it may merely reflect the fact that the first few generic links are equally relevant to the users. Thus, this study does not prove that users brainlessly click on the first link.
In addition, the Commission’s analysis is based on an assumption that many consumers looking to buy a product will complete the purchase after clicking on one of the first few search results Google provides. However, clicking on one of the top results may be only the start of the purchasing process as conducted by web-consumers in the real world. In a first search, consumers may be just trying to get familiar with various offerings without buying anything. They may have no interest in comparison shopping services at this juncture. By implicitly dismissing this possibility, the Commission makes a subjective choice in construing EU competition law.
Competitive Harm and Causation
Contrary to what the Commission suggests, as soon as Google’s general search services fail to meet user expectations, users can and will switch to other search services – such an increase in demand could easily be met by other general and specialized search services. Google’s history shows that it has consistently focused on meeting user expectations. When Google entered the market in the late nineties, AltaVista and Yahoo! were the leading search engines in Europe. Google outcompeted these other services by offering superior quality. Outcompeting competitors is not a violation of competition law.
The Commission, however, argues that Google behaved abusively by positioning and displaying, in its general results pages, its own comparison shopping service “more favourably compared to competing comparison shopping services.” According to the Commission, Google’s conduct may artificially divert traffic from rival comparison shopping services, hindering their ability to compete. Users would not necessarily see ‘the most relevant comparison shopping results’ in response to their queries. No empirical evidence for this finding is provided, however.
Furthermore, the Commission confuses correlation and causation (“post hoc, ergo propter hoc”). A drop in referrals from Google does not mean Google caused competitive harm – alternative factors may come into play. To find causation, the Commission merely refers to data showing a diversion in user traffic following the introduction of Google’s new shopping service. A change in user traffic, of course, can have many causes, including a failure of competitors to make the necessary investments or meet customers’ needs. To complete its causal reasoning, the Commission has to make several assumptions, which are backed up only by weak, suggestive evidence. In particular, the Commission has not been able to reject possible alternative explanations for the demise of Google’s competitors.
- Conclusion: Make Google Pay Its Fair Share
Viewed from the perspective of political economy, the record fines levied against Google can be seen as making Google pay its fair share. European governments believed that Google ‘cheated’ on tax. Through the corporate group approach, competition law was used to remedy the gaps in taxation regimes. EU competition law gives the Commission sufficient discretion to engage in result-oriented reasoning. Associative priming (“Google is a big, bad bully”) and confirmation bias may have helped the Commission to get comfortable with the weakness of its legal arguments and the lack of empirical evidence.
Effectively, the European Commission’s competition law decisions against Google can be viewed as an application of the unspeakable rule that any corporation violates EU competition law if it doesn’t pay its fair share of taxes. If this theory sticks, it raises the question whether the use of competition law as a remedy for weak taxation laws will be condoned. The European Court of Justice’s judgments on the appeals filed by Google will give us an answer.
One thought on “EU Competition Law as a Taxation Regime”
Your article is very interesting and clear.
I am very interested to know if you have an update or a book in english about the subject ( the connection of competition law and tax in european union) .
Leonardo de Andrade Costa