Together with leading international competition economists, we not only pointed out Google’s dominant position in the general search market in a policy paper by the Yale Tobin Center Digital Regulation Project. We have also developed several suggestions on how competition in the search market could be increased for the benefit of users and advertisers.
Google’s Vice President for Europe, Philipp Justus, responded in guest comments and – perhaps not surprisingly – argued that the facts are different and that regulation is superfluous. Most people use the Google search engine when they do some research. There are alternatives, especially for special searches, as Justus rightly writes in his guest commentary. But no competition authority would think that Coca-Cola has no market power just because you can alternatively drink tea.
The economically relevant question is which alternatives consumers choose between when they search for a search query, and how many would migrate in the event of a smaller deterioration in quality, i.e. use a search engine other than the default search engine. As various antitrust agencies document, Google’s market share for general searches is usually 90 percent.
Justus argues that this is not due to direct network effects (in which the usefulness of a good increases with the number of users) or the anti-competitive behavior documented by competition authorities and courts and condemned by the EU Commission in the Google Android antitrust proceedings, but solely due to technical progress from Google. In short: the users have alternatives and would only choose Google’s search engine because it delivers the best results.
Top jobs of the day
Find the best jobs now and
be notified by email.
The importance of the default search engine
But if that’s true, the question arises: Why is Google paying Apple significant sums to be installed as the default search engine on Apple mobile devices? According to the estimates described in our policy paper, there are between eight and ten billion US dollars per year in the American market alone. That wouldn’t be necessary if the searchers, as Justus argues, preferred Google and therefore would always switch to Google search on their own initiative.
However, Google’s managers seem to know what has long been known in behavioral science: Most people don’t bother to actively switch from the standard search engine to an alternative. Which is why it is so important to be the default search engine if you really want to reach a large number of users and earn money with them.
This is exactly why it is so difficult for new search engines to capture market share. Entering the market in and of itself is not difficult, as Justus rightly points out, but searchers can only win new search engines if they don’t all get stuck with the standard search engine – which is called “Google Search” on practically all smartphones in the western world: iPhones because of the payment to Apple, on Android smartphones because of the licensing of Android and the provision of important Google apps.
This licensing is de facto linked to the requirement that Google search be installed as standard. It is true that the group waives expressly forbidden contractual clauses following the Google Android process initiated by the EU Commission. However, as described in our Policy Paper (https://tobin.yale.edu/digital-regulation-project), Google has adjusted other behaviors at the same time so that nothing changes in the result.
The market result remains the same
Google is still installed as the standard search engine, and Google’s practices ensure that new providers do not get involved in contract negotiations with smartphone manufacturers. Therefore nothing has changed in the market result. Justus emphasizes that nobody would search Google just because others do the same. In doing so, of course, he ignores the indirect network effects. Google’s high market share in the search market in particular improves Google’s search results because a particularly large amount of data can be analyzed.
This advantage is less important for frequently asked questions, but even more important for rare questions or special questions. Not least because of this, we suggested that Google make its click-and-query data available to other search engines for a reasonable fee. This would give search engines with a previously small market share and a correspondingly small database the chance to score with the quality of their algorithm and the presentation of their search results.
If, as Justus argues, Google is ahead of the competition through constant technical progress and not through these indirect network effects, Google would not suffer any significant disadvantages from this regulation. It is particularly important for Justus to point out that many medium-sized companies in particular like to take the opportunity to expand their customer base through advertising on Google’s search results pages. And they allegedly do this with high profit, as he emphasizes with reference to long-term studies, unfortunately without a specific source.
The figure he cited seems to be based on Google’s Global Impact Report, which in turn builds on a 2009 article by Google’s chief economist Hal Varian. In this, Varian describes a method – based on a number of assumptions – to derive the advertisers’ profit from their bidding behavior. He tried this method on confidential data and found a high profit. How representative the data are, how they were selected – the article says nothing about that, much less is it available to other researchers for review.
Online advertising prices could be lower
What is also left out of consideration when considering: If there were more competition in the market for online advertising, the prices for this advertising would be lower. So it always depends on the comparison with the counterfactual situation. And yes, the medium-sized company could also advertise elsewhere, as Justus emphasizes. But the searchers: inside on Google it would no longer be reached. And that’s where most of them look.
In this respect, it seems misleading to us when Justus warns against tightening competition rules because this would not only affect US tech companies, but also German medium-sized companies. It can be doubted that, with the millions that it spends on lobbying in Brussels, Google is primarily concerned with the wellbeing of medium-sized companies.
According to a study by Lobby Control and Corporate European Observatory, the ten largest online platforms and infrastructure companies have recently spent 32.75 million euros per year on lobbying in Brussels, including 5.75 million euros from Google alone. The market capitalization of just under 1.9 trillion US dollars suggests that at least investors expect high future profits and do not see them jeopardized by strong competition.
The authors: Paul Heidhues is Professor of Behavioral and Competitive Economics at Heinrich Heine University in Düsseldorf.
Monika Schnitzer teaches economics at the Ludwig Maximilians University in Munich and is a member of the Advisory Council for the assessment of macroeconomic development.
More: India’s competition watchdogs accuse Google of abuse of market power.