The Limits of Australia’s Digital Platforms Inquiry

By Dirk Auer (International Center for Law and Economics)1

In July of 2019, the Australian Competition and Consumer Commission (“ACCC”) concluded its high-profile 1.5-year investigation into the effect of digital platforms on competition in media and advertising markets.2

The ACCC’s Final Report3 claims that competition is “not working” in the media, communications, advertising and other markets it investigated,4 and that substantial regulatory and legislative changes are necessary to solve — and would solve — the problems caused by ineffective competition.5 These changes include mandated “platform neutrality” obligations,6 tighter merger rules,7 firm- and industry-level codes of conduct over which the ACCC would have a discretionary oversight,8 and the adoption of more stringent privacy rules.9 The ACCC’s proposals would thus tilt the scales of competition and regulatory enforcement in its favor.

However, as discussed below, the premise that markets are failing is not well supported by the ACCC’s report.10 Moreover, even if the ACCC’s premise were accurate, its conclusions miss the bigger picture:: Government intervention is appropriate only if it produces net social benefits. The fact that a market does not satisfy some idealized benchmark (for instance, because firms have some degree of market power) is irrelevant if regulation merely exacerbates the perceived failure or creates new and greater costs.11 Yet the ACCC Final Report almost entirely omits consideration of possible regulatory failure. It is thus of little practical value in evaluating the merits of potential regulatory interventions.

The Overlooked Costs of Enforced Platform “Neutrality”

The ACCC Final Report puts forward many recommendations to address perceived competitive concerns in digital platform industries. These include merger law reforms, and changes to search engine and internet browser defaults.12 In doing so, it disregards traditional tools of competition policy in favor of a much looser approach. Most notably, the ACCC relies upon a piecemeal market definition exercise and departs from established competition law theories of harm. This methodological frailty also ripples through the report’s recommendations. Indeed, the report does not sufficiently defend its conclusions that mandated platform neutrality and the establishment of a digital platforms branch would alleviate the concerns identified by the ACCC.

The report notably claims that Google controls a large share of the “general search advertising” and “vertical search advertising” markets.13 It similarly asserts that Facebook commands a 51 percent share of “display advertising.”14 However, the report fails to put forward a coherent — let alone quantitative — argument that these are indeed relevant markets for competition policy purposes (as opposed to, for example, a market encompassing all online advertising). Readers are left to guess why it is that large players, such as Amazon, as well as the massive range of offline advertising and marketing outlets, are excluded from this analysis. It is true that the report mentions — and then dismisses — the possibility of competition between online and offline advertising.15 But the report’s means of distinguishing between the two markets does not turn on the competition-relevant criteria of demand and supply substitution; rather, it turns on the differences between the specific matching mechanisms employed by each.16 Further gradations (e.g. between online search and display advertising) turn on similarly inapposite characteristics.17

Likewise, the ACCC’s “theory of harm” amounts to little more than an observation that digital platform markets have features that purportedly complicate market entry by rivals. For instance, the ACCC concludes that “consumer inertia” and “default bias” cause users to stick with Google’s search engine and browser, rather than switch to its rival’s products.18

In fact, it is likely that a majority of consumers simply prefer Google’s browser and search offerings. In the desktop environment, Google’s ChromeOS has less than 1 percent market share in Australia, whereas Microsoft’s Windows has a market share of 69 percent and Apple’s iOS a 24 percent market share.19 The default browser on Windows is Edge, while the default browser on iOS is Safari, yet Chrome is chosen by about 70 percent of users in Australia.20 Meanwhile, about 90 percent of desktop searches in Australia are performed using Google.21

Given the demonstrable preference for Google’s browser and search engine, forcing consumers to go through a choice screen may at best be a waste of time, and at worse a paralyzing choice between indiscernible options.22

In spite of having failed to demonstrate actual harm to competition or consumer welfare, the ACCC calls for the imposition of stringent “platform neutrality” obligations. Most notably, it would require Google to provide users of the Android platform with a choice of internet browsers and search engines, rather than make its own services the default option.23 The ACCC would also investigate whether similar obligations should be imposed upon other digital platforms, via a newly established digital platforms branch.24

Unfortunately, the ACCC mostly ignores the potential costs of platform neutrality obligations, as well as the potential rent-seeking opportunities created by a branch dedicated to digital platforms. From an error-cost perspective, it is thus unlikely that the ACCC Final Report achieves the right balance between over and under-enforcement.

Micromanaging the News Industry

Much of the same can be said about the ACCC’s analysis of platforms’ dealings with traditional news outlets. In a nutshell, the ACCC is concerned with the bargaining power of small media firms (or lack thereof) relative to global internet platforms, and the potential effect that these platforms may have on the market for journalism.25 To address these alleged problems, the report recommends that platforms should be made to establish regulator-approved codes of conduct that would bind them in their future dealings with media outlets.26 It also recommends the introduction of fiscal policies that would prop up Australia’s struggling local journalism industry.27

Even if one accepts the ACCC’s assertion that there is a market failure in the provision of certain kinds of content, it is far from certain that its recommendations would produce the desired outcome. Take the ACCC’s rosy view of local journalism. The report hardly considers the possibility that the relative decline of local journalism has not been caused by online platforms (correlation is not causation).28 Nor does it envisage that local newspapers may no longer be the optimal way to stay informed about local events. Finally, it is not clear that shifting additional revenue towards Australia’s struggling local news industry will necessarily lead to the type of content that the ACCC deems essential for Australian consumers.

To make matters worse, the report also fails to grapple with the risk of regulatory failure. Subsidizing local news and favoring public broadcast channels raises the prospect of “white elephant” spending — that is, of public expenditures that exceed the social value they create.29 The ACCC’s fear that local news is being underprovided is not set against a relevant economic benchmark. Instead, the report merely observes that local news coverage has declined compared to its previous highs.30 But this is hardly a convincing argument. All industries go through ebbs and flows. Nor is there any basis for presuming that previously established practices were optimal — especially when they were established under very different technological conditions. Accordingly, if the government acts on the ACCC’s calls to subsidize local news and boost public broadcasting, the resultant expenditures may well harm taxpayers through waste rather than create socially valuable pieces of news.

Finally, the ACCC’s plans create unavoidable incentives for rent-seeking. Rivals will surely attempt to game the approval process for platform codes of conduct. And decisions regarding the allocation of public funding will also be vulnerable to undue influence. Under the ACCC’s proposals, the assessment criteria for grants would be designed by a committee comprising journalists, academics, and former industry participants.31 But allowing industry insiders to play an active role in the allocation of grants is problematic. For instance, it is far from clear that these stakeholders will be willing to embrace the potential disruption that digital platforms may bring to the local news industry.

Privacy Regulation is No Free Lunch

The report also touches upon the topic of online privacy. But while its recommendations are very similar to the privacy regulations recently introduced by the European Union32 and California,33 the ACCC overlooks the tremendous compliance costs created by those regulations. Yet its own recommendations would likely entail proportionally similar costs.

For instance, it has been estimated that American S&P 500 companies and UK FTSE 350 companies alone spent a combined total of $9 billion to comply with the EU’s GDPR in the year running up to its entry into force.34 Likewise, a survey of American companies with over 500 employees found that three out of four planned to allocate over $1 million to GDPR compliance.35

Things are similar for California’s CCPA. One estimate places the upfront costs of compliance with the CCPA (including lost advertising revenue) at $24.5 billion — even as it identifies comparably modest benefits of roughly $6-9 billion.36 Another survey estimated that 70 percent of firms would pay at least $100,000 to comply with the CCPA, while a further 20 percent would spend more than $1 million.37

And compliance costs are not the only problem. One study found that, following the GDPR’s entry into force, weekly venture capital deals in the EU decreased by 17.6 percent, compared to the US, and the amount raised in an average deal decreased by 39.6 percent.38 Furthermore, it has been argued that data regulations may have a significant impact on competition between platforms, arguably limiting the entry of smaller firms, entrenching established incumbents, and disincentivizing increases in product quality on privacy dimensions.39

Conclusion: The Limits of the ACCC’s Digital Platform Inquiry

In his groundbreaking “The Limits of Antitrust,” Frank Easterbrook famously argued that the task of competition enforcers and regulators is not crudely to maximize certain policy preferences.40 Instead, their task is to maintain a delicate balance between the societal harms of anticompetitive conduct on the one hand and, on the other, the administrative and societal costs of unavoidable, if occasional, enforcement errors that invariably follow even modest enforcement activity.

The ACCC Final Report fails to grapple with this complex tradeoff. While it claims to show pervasive market failures throughout the digital economy, it ignores the specter of regulatory failure. This goes against a commendable trend that has seen Australian competition enforcement move towards a more effects-based system.41

The ACCC’s lackadaisical assessment of regulatory costs is all-the-more troubling given that its report focuses on an extremely dynamic industry. What is only a small regulatory cost today could severely hamper competition in the future. Yet, rather than reconsider how ham-fisted regulation might be throttling traditional media firms, or how it may further entrench today’s dominant technology platforms, the ACCC advocates more of the same.

Could excessive media regulation be preventing traditional media from competing against digital platforms (and might the ACCC’s recommendations have a similar effect in the future)? Has this type of regulation really produced net social benefits? And could ill-designed regulation act as a barrier to entry? By almost completely ignoring these questions, the ACCC Final Report all but guaranteed that its conclusions would be heavily skewed in favor of increased government intervention — and potentially ineffective or even harmful intervention, at that.

Click here for a PDF version of the article


1 Dirk Auer is a Senior Fellow at the International Center for Law and Economics (ICLE), and a Guest Lecturer at EDHEC Business School and UCLouvain. ICLE is a nonprofit, nonpartisan research center based in Portland, OR. ICLE promotes the use of law & economics to inform public policy debates. We believe that intellectually rigorous, data-driven analysis will lead to efficient policy solutions that promote consumer welfare and global economic growth. ICLE has received financial support from numerous companies and individuals, including firms with interests both supportive of and in opposition to the ideas expressed in this and other ICLE-supported works. Unless otherwise noted, all ICLE support is in the form of unrestricted, general support. The ideas expressed here are the author’s own and do not necessarily reflect the views of ICLE’s advisors, affiliates, or supporters. Please contact the author with questions or comments at icle@laweconcenter.org.

2 The investigation was initiated on December 4, 2017, by then-Treasurer, the Hon Scott Morrison MP.

3 See Australian Competition and Consumer Commission, Digital Platforms Inquiry Final Report (June 2019) (“ACCC Final Report”), available at https://www.accc.gov.au/focus-areas/inquiries/digital-platformsinquiry/final-report-executive-summary. (“ACCC Final Report” or “the report”).

4 Id. at 2 (“The ACCC considers that the regulatory frameworks governing media, communications and advertising also need to be addressed, as they do not allow competition on the merits.”); id. (“The competition concerns extend beyond specific sets of advertisers.”); id (“The opacity of this ad tech supply chain leads participants to question its efficiency. Where problems do occur, they may be impossible for participants to detect.”).

5 Id. at 3 (“The ACCC is concerned that the existing regulatory frameworks for the collection and use of data have not held up well to the challenges of digitalisation and the practical reality of targeted advertising that rely on the monetisation of consumer data and attention.”); id. (“Policy makers must ask whether the principles that have applied in the past are still fit for purpose and must review legislative tools, principles and oversight to address further technological and consumer-driven developments.”); id. (“The pace of technological change needs to be matched by the pace of policy review. As digital markets and the use of data continue to grow and change, governments need to continue to consider the appropriate level of oversight. The recommendations in this Report allow for this: they both address current problems and allow the Government to identify and address new problems as they arise.”).

6 Id. at 30.

7 Id.

8 Id. at 32 & 34.

9 Id. at 35.

10 See also, INTERNATIONAL CENTER FOR LAW & ECONOMICS: CONCLUDING COMMENTS: THE WEAKNESSES OF INTERVENTIONIST CLAIMS (June 2019), available at https://laweconcenter.org/wpcontent/uploads/2019/07/Concluding-Comments-The-Weaknesses-of-Interventionist-Claims-FTC-Hearings-ICLEComment-11.pdf, p. 5. (“Some participants claimed that network effects create winner-take-all markets for tech platforms. While it may be true that some markets are naturally winner-take-most, from the perspective of consumer welfare, this is not necessarily a negative outcome. In the case of communications networks, it is intuitively obvious why consumers would be better off participating in a few large networks than in many small ones. Furthermore, critics ignore the potential for Schumpeterian innovation, i.e., when innovation occurs at the platform level, and platform owners leapfrog each other, maintaining a temporary monopoly long enough to be compensated for putting startup capital at risk.”). See also, Catherine Tucker, Digital Data, Platforms and the Usual [Antitrust] Suspects: Network Effects, Switching Costs, Essential Facility, 54 REVIEW OF INDUSTRIAL ORGANIZATION 684 (2019). (“I find little evidence that digital data augments market power due to either network effects or switching costs. Instead, digitization may have weakened these two economic forces, because it frees a user from a particular hardware system. Last, it seems unlikely that data can ever meet the criteria for an essential facility, simply because it is often not very valuable and because, since digital data is non-rival, many sources usually exist.”).

11 See, e.g. Harold Demsetz, Information and efficiency: another viewpoint, 12 Journal of Law & Economics 1 (1969). (“This nirvana approach differs considerably from a comparative institution approach in which the relevant choice is between alternative real institutional arrangements. In practice, those who adopt the nirvana viewpoint seek to discover discrepancies between the ideal and the real and if discrepancies are found, they deduce that the real is inefficient. Users of the comparative institution approach attempt to assess which alternative real institutional arrangement seems best able to cope with the economic problem”).

12 See ACCC Final Report, supra note 2, at 30.

13 Id. at 94.

14 Id. at 97.

15 Id. at 92.

16 Id. at 91 (distinguishing between the degree — not even the existence per se — of “targeting” in online versus offline advertising).

17 See, e.g. id. at 92-94.

18 Id. at 110.

19 See Desktop Operating System Market Share Australia Aug 2018 – Aug 2019, statcounter GlobalStats, https://gs.statcounter.com/os-market-share/desktop/australia.

20 See Desktop Browser Market Share Australia Aug 2018 – Aug 2019, statcounter GlobalStats, https://gs.statcounter.com/browser-market-share/desktop/australia/#monthly-201808-201908.

21 See Desktop Search Engine Market Share Australia, Aug 2018 – Aug 2019, statcounter GlobalStats, https://gs.statcounter.com/search-engine-market-share/desktop/australia/#monthly-201808-201908.

22 This is akin to the contested behavioral economics literature on choice overload. See, e.g. Benjamin Scheibehenne, Rainer Greifeneder & Peter M Todd, Can there ever be too many options? A meta-analytic review of choice overload, 37 J. Consumer Research, 409-425 (2010).

23 See ACCC Final Report, supra note 2, at 112. Id. at 110.  Moreover — and contrary to what will likely be the case in Europe — Google would not be able to auction off these “ballot box” slots. See Paul Gennai, “An update on Android for search providers in Europe,” Google Official Blog, Aug. 2, 2019, https://www.blog.google/around-the-globe/google-europe/update-android-search-providers-europe/. Instead, Google would have to include internet browsers and search engines based on their popularity, showing them in random order. See ACCC Final Report, supra note 2, at 112.

24 Id. at 135-136.

25 See id. at 16.

26 Id. at 17.

27 Id. at 19-20.

28 See id. at 5.

29 James A Robinson & Ragnar Torvik, White elephants, 89 Journal of Public Economics, 197 (2005). (The authors define white elephants as “investment projects with negative social surplus”).

30 See ACCC Final Report, supra note 2, at 320.

31 See ACCC Final Report, supra note 2, at 332. (“To ensure independence from the Government, the assessment criteria for these grants should be designed by an independent expert committee made up of journalism industry representatives and other independent experts such as academics and former industry participants. The committee should also make decisions on the allocation of grants”) (emphasis added).

32 See Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), O.J. L. 119/1.

33 The California Consumer Privacy Act of 2018 requirements will take effect from 1 January 2020.

34 See Oliver Smith, The GDPR Racket: Who’s making money from this $ 9 bn business shakedown, Forbes, May 2, 2018, available at https://www.forbes.com/sites/oliversmith/2018/05/02/the-gdpr-racket-whos-making-money-from-this-9bn-business-shakedown/#33232d9834a2.

35 See PWC, GDPR compliance top data protection priority for 92% of US organizations in 2017, according to PWC survey, PWC.com, Jan. 23, 2017, https://www.pwc.com/us/en/press-releases/2017/pwc-gdpr-compliance-press-release.html.

36 See Roslyn Layton, The costs of California’s online privacy rules far exceed the benefits, AEIdeas, March 22, 2019, http://www.aei.org/publication/the-costs-of-californias-online-privacy-rules-far-exceed-the-benefits/.

37 See Ginny Marvin, Businesses prepare to spend heavily on CCPA compliance, MarTechToday, March 19, 2019, https://martechtoday.com/businesses-prepare-to-spend-heavily-on-ccpa-compliance-231938.

38 See Jian Jia, Ginger Zhe Jin & Liad Wagman, The short-run effects of GDPR on technology venture investment, NBER Working Paper Series, 4 (2018).

39 See James Campbell, Avi Goldfarb & Catherine Tucker, Privacy Regulation and Market Structure, 24 Journal of Economics & Management Strategy 47, 68 (2015). See also, Antonio Garcia Martinez, Why California’s Privacy Law Won’t Hurt Facebook or Google, Wired, Aug. 31, 2018, https://www.wired.com/story/why-californias-privacy-law-wont-hurt-facebook-or-google/.

40 See Frank H Easterbrook, Limits of antitrust, 63 Tex. L. Rev., 16 (1984). (“The legal system should be designed to minimize the total costs of (1) anticompetitive practices that escape condemnation; (2) competitive practices that are condemned or deterred; and (3) the system itself.”).

41 See Ian Harper, Peter Anderson, Su Mccluskey and Michael O’Bryan, Competition policy review, Commonwealth of Australia (March 2015), available at http://competitionpolicyreview.gov.au/files/2015/03/Competition-policy-review-report_online.pdf.