A Merger of the Top Two Food Delivery Apps in Korea: An Analytical Review

By Gene-Oh Kim¹ & Youngsik Lim²

Introduction

Over the past few years, we have observed a significant number of (two-sided) platform operators’ business combinations globally. As the scale of non-face-to-face transactions increases rapidly in the online era, traditional offline transactions and businesses are reorganizing into online transactions and businesses in various fields. As the number of attempts to enter new markets (conglomerate merger) or to increase market presence through the expansion of scale in existing markets (horizontal or vertical merger) steadily grows, the size and frequency of consolidation actions among platform operators are expected to increase.

Since such two-sided platforms are clearly distinguished from the traditional “brick-and-mortar” manufacturing industry, the assessment of anti-competitiveness also requires a new perspective and methodology.  From an economic analysis standpoint, which is an important pillar of competitive assessment, a merger between two-sided platforms poses significant challenges, so variations on well-established methodologies in traditional antitrust economics and/or the introduction of new methodologies appears inevitable. It is understood that many recent studies in academia on various analytical issues regarding merger cases between platforms reflect such practical trends and needs.

In this article, we would like to briefly introduce the Korea Fair Trade Commission’s (“KFTC”) recent decision on the merger between “Baemin” and “Yogiyo,” the first and second-largest players in the Korean food delivery app market, and the economic analysis discussed in the process of said decision. The purpose of this article is not to address the appropriateness of the KFTC’s decision. Rather, it aims to share the implications for future merger cases in the platform industry by introducing the logic and basis underlying the KFTC’s decision.

Details of the KFTC Decision

Regarding the 2020 business combination where Delivery Hero Korea LLC, which operates “Yogiyo,” the second-largest player in the Korean delivery app market, was to acquire about 88 percent of the voting shares of Woowa Brothers Corp. which operates “Baemin,” the number one player in the same market, the KFTC determined that the anti-competitiveness of the transaction would be significantly high, and decided on a structural remedy whereby Delivery Hero Korea would sell all of its shares to a third party within three months from the receipt of the corrective order.

The KFTC found that the merger would raise significant anti-competitive concerns in the food order delivery platform (“delivery app”) market on the following grounds: After defining the relevant product market as the market consisting of only Baemin and Yogiyo, or if expanded, only five delivery apps including Baemin and Yogiyo,³ and the relevant geographic market as the nation-wide market, the KFTC stated that (i) the combined market share of the parties meets the requirements for a presumption of anti-competitiveness, (ii) there is a high possibility of demand diversion between Baemin and Yogiyo, (iii) there is no competitor that can counteract the unilateral conduct of the parties in a timely manner, (iv) there is a possibility of impeding the growth of competitors based on overwhelming information assets, and (v) there is a possibility of a reduction in promotions on the consumer side and an increase in commissions on the restaurant side.

Key Issues in the Economic Analysis

Product Market Definition

The merging parties, based on the results of a consumer survey, conducted a critical loss analysis and an aggregate diversion ratio analysis limited to the consumer side.⁴ They argued that phone orders and orders through a franchises’ own apps/websites should be defined as the same market as delivery apps because consumers recognize them as close substitutes (a high diversion ratio between them is estimated).

However, the KFTC viewed that, in terms of function and utility, phone orders and orders through franchises’ own apps/websites are significantly different from delivery apps based on the following points: (i) according to various survey results, there is a high possibility of “lock-in” to delivery apps from a consumer and restaurant perspective, and (ii) the KFTC’s own economic analysis shows that substitutability between delivery apps and other methods of placing orders is not high, and thus, it is reasonable to define them as separate product markets.

What is notable in the KFTC’s economic analysis is that it conducted both a consumer-side and restaurant-side critical loss analysis⁵, and then examined the market definition results that would be derived with the SSNIP test of 5 or 10 percent based on the total price on both sides. The KFTC’s economic analysis led to the conclusion that a candidate market consisting of Baemin and Yogiyo does not need to be expanded because actual price elasticity is much lower than critical price elasticity on the consumer side, and the actual margin is lower than the critical margin on the restaurant side.

Assessment of Unilateral Effects

Presenting the results of economic analyses such as UPP (upward pricing pressure) analysis and promotion analysis, the parties argued that the diversion ratio between Baemin and Yogiyo is not high and that they lack the ability and incentive to unilaterally increase prices post-merger. However, the KFTC rejected the argument for the following reasons and concluded that anti-competitive concerns are significant:

First, the combined market share of the merging parties (the top three players in the market⁶) is almost 99 percent, and there is no actual competitor capable of exercising effective competitive pressure on them. Second, there is a high degree of multi-homing between Baemin and Yogiyo on both consumer and restaurant sides, so substitutability between them is high, but the possibility of switching to other competing delivery apps is low. Third, as the information asset and business capabilities held by the merging parties are overwhelming compared to those of their competitors, it would be difficult for competitors to secure a user base beyond the critical mass needed to compete effectively. Fourth, the economic analysis conducted by the KFTC shows that the post-merger price is expected to be significantly increased (reduction of promotions for consumers).

One distinctive feature of the KFTC’s economic analysis is that, considering that the delivery apps are two-sided platforms from the consumer’s point of view, it modified and applied the GUPPI (gross upward pricing pressure index) formula, well-established in the traditional single-sided analysis. According to the results, both Baemin’s and Yogiyo’s GUPPI were estimated to be well above 10 percent, leading to the conclusion that there is a high possibility of increasing prices by reducing coupon discounts on the consumer side after the merger.

Another salient feature of the KFTC’s economic analysis is that it showed through a simulation study that most restaurants depend a lot on Baemin and Yogiyo (about 30~50 percent of their sales come through both delivery apps) while the restaurants’ departure rate upon an increase in commission by 5 or 10 percent appears to be insignificant.  These findings lead to the conclusion that the incentives for the post-merger parties to increase commissions for the restaurants are significant.

Key Issues Related to Procompetitive Factors

Possibility of New Entry

As we are well aware, anti-competitive concerns can be mitigated to a significant extent if new entry can be made sufficiently (sufficiency) and easily (ease) in the near future (timeliness). The merging parties argued that, given the active market entry and exit and the continued innovation of business models in the delivery app market at home and abroad, such characteristics should be fully considered in assessing the merger’s anti-competitiveness, and that static factors such as current market shares do not sufficiently predict future competition⁷.

In response, the KFTC held that, even if new entrants can enter the market easily in the near future, this would hardly exert sufficient competitive pressure on the merging parties because new entrants must invest large amounts of money over a considerable period of time to obtain as-efficient competitiveness and technological capabilities as the merging parties in order to surpass the vast pool of loyal customers and restaurants attached to the merging parties.⁸

Dynamic Characteristics of the Delivery App Market

The parties argued that the possibility of future dynamic changes, rather than the current static market shares, should be considered because the delivery app market is a dynamic one, where entry and exit is free and there are constant competitive threats from growing adjacent industries and last-mile delivery competition between platforms.

The KFTC rejected this argument, however, for the following reasons: First, since Yogiyo entered the market in 2012, the oligopolistic market structure led by the merging parties has become solidified, and hardly any meaningful changes have been made.  Second, the delivery app market is characterized as a “winner-takes-all” phenomenon due to indirect network effects, meaning it is not easy for new entrants to exert meaningful competitive pressure on the incumbents. Third, although it is acknowledged that the business model can be expanded and remodeled through linkage with other industries, such as distribution and logistics, it is unclear how this will alleviate the anticompetitive concerns resulting from this merger.

Competitive Pressure from Competitors

The merging parties argued that CoupangEats and Naver are already exerting significant competitive pressure on them, which can offset the anti-competitive concerns raised by the merger to a significant extent. Specifically, in the case of CoupangEats, using the aggressive marketing strategy that Coupang, parent company of CoupangEats, had helped it succeed in the logistics and distribution sector, the company has grown explosively within one year of its launch, and is expected to be able to provide nationwide services within one or two years. In the case of Naver, as the largest portal company in Korea, it has a strong financial position and is already providing delivery app services such as Naver Easy Order, etc.

However, the KFTC determined that it is not clear whether CoupangEats or Naver could exert meaningful competitive pressure on the merging parties because, in the case of CoupangEats, it is uncertain whether it will grow enough to change the competitive landscape of the delivery app market in the near feature by expanding and establishing the “One-Order-One-Delivery” model, which entails huge variable costs in a deficit operating situation. As for the case of Naver, the delivery app business has a structure that must create revenue through securing as many restaurants as possible, but it is difficult to secure stable revenue through a free service model such as Naver Easy Order, and in fact, the actual transaction record has been relatively insignificant for more than three years since the launch of Naver Easy Order in 2017.

Efficiency-Enhancing Effects

The parties asserted that the merger would allow the “own delivery (“OD”)” system to be established in the Korean market by combining Delivery Hero’s technology and know-how in establishing an efficient delivery system with Baemin’s business experience and knowledge of the Korean market. Also, the OD system would bring efficiency-enhancing effects, significantly raising consumer satisfaction with faster and better delivery services and increasing the income of restaurants and deliverymen through increased delivery orders based on said consumer satisfaction.

However, the KFTC held that the “effects of integrating the ‘OD model’” as presented by the respondent is not a merger-specific effect that can only be achieved through the merger, and refused to accept the respondent’s argument that delivery times can be shortened only when “order density” increases, as no significant correlation was found between the number of restaurants available for delivery and the reduction of delivery time.

Conclusion

A “retroactive analysis” of what actually happens in the market after a certain period of time post-merger is quite important. It is even more important, in a dynamic market, to look into the effects of the competition authority’s decisions, as it provides meaningful lessons on how to view future merger cases.

For reference, in 2009 the KFTC gave conditional clearance to the merger between Gmarket and Auction, which were the No. 1 and No. 2 online marketplaces at that time. The KFTC judged that three-year behavioral remedies alone could sufficiently resolve any anti-competitive concerns in consideration of the dynamic nature of the market, despite the combined market share of the two companies being very high, exceeding 85 percent.

As of 2019, more than a decade later, the market share of Gmarket-Auction was estimated to be around 30 percent, and it was confirmed that the actual market was transformed into a more competitive and dynamic market, as predicted by the KFTC. At the time, as the grounds for its decision on conditional clearance, the KFTC claimed that new entries and exits were free, the business model was able to continuously change, which could easily change the competitive landscape of market participants, and the expansion of business was possible through linking with other industries. Yet, as we have described above, the KFTC’s position is that such arguments are not applicable to this merger.

One of the key arguments made by the merging parties is that new entry and dynamic changes are more likely to occur in the current competitive environment of the food delivery app market than in the online commerce platform market back in 2009. The KFTC’s judgment is, to the contrary, that it is difficult to expect such dynamics to emerge from this merger. It is interesting, from an antitrust perspective, to see which position and prediction will eventually prevail. Only time will tell.

Click here for a PDF version of the article

1 Attorney, Kim & Chang. The views expressed in this article are those of the author and do not represent the views of any organizations or clients with which they are or have been associated.

2 Senior Economic Advisor, Kim & Chang. The views expressed in this article are those of the author and do not represent the views of any organizations or clients with which they are or have been associated.

3 Woowa Brother’s two platforms (Baemin, Baemin Riders) and Delivery Hero Korea’s three platforms (Yogiyo, Baedaltong, Foodfly).

4 The main reason for applying the analyses to the consumer-side only is that, due to the nature of the two-sided transactional platforms, an analysis on the consumer-side alone can provide sufficient implications for market definition.

5 Considering the available scope of data, the KFTC conducted critical elasticity analysis on the consumer side and critical margin analysis on the restaurant side.

6 Baedaltong, the third largest player, also belongs to Delivery Hero along with Yogiyo.

7 The merging parties also argued that the Competition and Markets Authority (“CMA”), the British competition authority, unconditionally cleared the merger between Just Eat and Hungryhouse, which were the first and the second largest delivery app providers in the UK market in 2017, despite their high combined share, in consideration of the competitive pressure of (potential) competitors given the dynamics of the delivery app sector.

8 In other words, the KFTC recognized “timeliness” and “ease” of new entry, but judged that “sufficiency” was not satisfied.