Does Google abuse its dominant position in the search engine market?

Outline

List of Abbreviations

1 Introduction

2 Relevant Market
2.1 Theory
2.2 Google

3 Dominance
3.1 Theory
3.2 Google

4 Abuse of dominance

5 Conclusion

References

List of Abbreviations

illustration not visible in this excerpt

1 Introduction

“Googeln” is a common verb you can look up in the German encyclopedia Duden now- adays (BI). Google is a very well known player, not only in the market for search en- gines, that made it possible throughout the last decade to increase its revenue to nearly 45 billion $ in 2012 (REV). While 14 years ago there were about 10.000 Google search- es a day, nowadays this number is reached within one-hundredth of a second (STA). All the information gained is extremely valuable for advertisers that use Google AdWords to place consumer matched advertisement. This is also one of the reasons consumers can enjoy products such as internet searches without any charge and therefore might not notice if a player, such as Google, abuses its dominance since it can’t be done through excessive pricing.

In the beginning of 2010 eJustice, Ciao and Foundem, three price comparison websites, claimed that Google has been downgrading their websites in the search results (Brian 2013). All three are so-called vertical search engines, which deal with specific content, such as flights/hotels “[…] rather than dealing with general search requests.” (Van Loon 2012, p. 16). As a result, the EU announced the opening of an antitrust investigation against the search engine pioneer in November 2010. Here the EU will examine whether Google might violate the European competition law (§102, TFEU, 2007); the abuse of a dominant position. Just one month ago, Joaquín Almunia, the EUs competition commis- sioner rejected Google’s second offer to settle the investigation (FT). If the accusations prove to be correct the firm might face fines up to 5 billion $ (Brian 2013). It is going to be especially interesting to find out the importance of Google’s first of a kind ad- vantage, as it was the first search engine to introduce the so-called ‘Page Rank’ algo- rithm, which increased the search results accurateness.

Nonetheless, according to §102 TFEU, there are two conditions that a have to be analyzed before: The relevant market and the existence of a dominant position. Therefore I am going to present the economic theory and methods which are relevant and then apply these to Google. After that I will examine whether Google is abusing its dominant position in order to draw an answer to the question.

2 Relevant Market

Concerning the relevant market, it is important to consider the relevant product and ge- ographical market (Horowitz 1981, p. 3). As explained in the introduction I will first introduce several methods how to define a market and then apply these to the case of Google.

2.1 Theory

Product Market

Demand side substitution is the most common used method, especially of the competi- tion Agencies, to define a relevant product market (Chiriţă 2011, p. 104). We take a certain product and pretend that there is only one supplier, a hypothetical monopolist

(HM). Now, we check whether the HM would find it profitable to increase his price for this product by a small but significant and non-transitory price (SSNIP). Usually this is an increase of 5% on the current price. If he can increase his profits this product itself is considered a market and then all its suppliers can be considered.

However, if the SSNIP does not increase the profits, the considered market is defined to narrow and therefore has to be expanded. The next best set of substitutes has to be in- cluded. These are products that consumers would switch to after a price increase. After including these, the process is repeated and again a HM is increasing the price for the new set of products, until he finds it profitable. This process is called SSNIP Test (Cameron et al. 2012, p. 723-724). The necessary parameter is demand elasticity of each consumer in order to find out how many consumer actually will substitute after a SSNIP.

Geographical Market

Besides the relevant product market, the geographical dimension plays an important role when defining the market. It is defined as the areas “[…] in which a firm is in- volved in the supply of products or services [...]“ and “[…] in which the conditions of competition are sufficiently homogeneous [...]” (Van Loon 2012, p. 12). Important is to distinguish between the countries/areas a firm operates in and the actual same condi- tions of competition: Therefore it is possible that a firm acts in different geographical markets, providing one identical product/service. It can also be measured throughout demand substitutability by applying a SSNIP test in different regions.

Moreover, the accessibility of the product/service is essential. Especially transport costs can change prices in specific areas. For instance, if it is more expensive for a supplier to provide his product/service in Asia than in Europe, they will have to be looked at as different geographical markets (Van Loon 2012, p. 12; Horowitz 1981, p. 6; Harris and Jorde 1983, pp. 472-475).

2.2 Google

To first define the relevant product market, I will apply the concept of demand substi- tutability, pretending that there is a HM for search engines. For the moment, I assume that Google is the sole provider of Internet searches and Yahoo, Bing, etc., are faded out.

However, the main problem is that it is fairly difficult to apply the SSNIP test when the price for a product/service is zero, as it is in this case. The normal scale of an increase of 5% is not possible. But if Google would now hypothetically increase its price for each search by a small amount, what would happen? To answer this question, knowledge of the search engine users demand elasticity is mandatory.

Nevertheless, we can look at possible substitutes consumers might find suitable. To start with, users searching for general information might directly enter websites, such as wik- ipedia.com or scholarpedia.com, to look up information. Other substitutes for infor- mation are books, but also friends who could be asked. Alternatives to find specific ser- vices, such as mechanics, painter, etc., are the Yellow Pages or websites such as Craiglist.com. Moreover, directory assistance (e.g. 11833 in Germany) can be consid- ered a strong substitute since they try to combine information. On the one hand, all of these steps will very likely lead to the same results as a search engine does. On the other hand, they will clearly take more time, therefore aren’t that efficient and compact as a search engine. These substitutes are more alike the vertical search engines explained in the end of this chapter.

That is why I believe that even if there was an increase in price, consumers would stick to the search engines seeing that no substitute could provide information as efficient, convenient and low priced. For that reason these substitutes should not be included in the relevant product market.

With regard to the geographical dimension, it looks like a worldwide market because of the accessibility of websites wherever there is an Internet access. There are linguistic differences though; nevertheless I believe that if Google Germany started charging a price for searches, consumers would immediately switch to e.g. Google France, since the same search results can be found just ranked differently. Anyhow, the conditions of competition are different. Particularly in countries such as China, where Google is not supported, but other websites as Baidu are sponsored by the government, the conditions of competition are obviously different. Consequently the geographical dimension has to be narrowed down to the European Union where they are identical.

Yet, the market can still be narrowed down on the grounds of so-called vertical search engines. These have to be distinguished from the market of general search engines. Ba- sically, users do not search horizontally throughout all kind of different information but vertically within a type of certain topics. They focus on a niche market, e.g. price com- parison between flights/hotels (Curran and Glinchey 2007, p. 22). As already explained in the introduction, three of them complained because of the possible abuse of Google’s dominance. They are experiencing a boost in popularity due to their benefit of better comparison and better filter in their niche. For that reason I consider them having a spe- cific consumer demand and hence a different product market. In addition, in the geo- graphical dimension, vertical search engines are stronger nationally oriented. Often they focus on country related topics, such as property (e.g. Funda: a vertical search engine for house sales in the Netherlands) (Van Loon 2012, p. 25).

As the relevant market is defined, it can be examined. The market for Internet searches is a two-sided market with two different user groups with cross-side network effects (Rochet and Tirole 2005, pp. 4-5). On the one side there are consumers, who look for information/services. On the other side, advertisers place their Ads on Google’s search engine. Especially this group has a cross-side network effect since it benefits from a higher search engine user rate who might click on the sponsored links.

Additionally, the market can be considered a new economy market. These kinds of markets characterize in standardization, technically very complex products with a short life circle and especially network effects and dynamic competition. Dynamic competition is a swap from price competition (static) towards competition around innovation (Van Loon 2012, pp. 13-14; Ahlborn et al. 2006, p. 15). This is going to be particularly important when assessing Google’s dominance.

illustration not visible in this excerpt

diagram 1: Search engine market share in Germany (Web-Stats 2013)

In diagram 1 the market shares of all search engines in Germany in 2012 are demonstrated. Noticeable is Google’s huge market share. Other players, such as Yahoo and Bing, are more present in North America. The development and interpretation of the market shares will be analyzed in the next chapter.

[...]