In January 2020, The Competition Commission of Pakistan ("the Commission") approved the acquisition of Careem by Uber, transforming the duopolistic market of application-based ridesharing services into a monopolistic market. The Commission was established, in October 2007, with the objective to "maintain and enhance competition." This simple and clear mandate enshrined in the preamble of the Competition Act, 2010 ("the Act") serves as a beacon light for the functioning of the Commission.
Section 11 of the Act authorizes the Commission to review mergers and acquisitions, with a view to prohibiting any merger which "substantially lessens competition by creating or strengthening a dominant position in the relevant market" ("SLC by CSDPRM").
Clearly, §11 has very little tolerance for the creation or strengthening of a dominant position in the "relevant market". A firm is presumed to be dominant if it has 40 per cent or more of the market share. In a duopoly, two firms subsist with more than 40 per cent each of market share as was the case in the of application-based ridesharing services in Pakistan where both Uber and Careem had more than 40 per cent of the market share. Once determined that the merging parties meet the dominance threshold, the Commission proceeded to the second step to assess whether the merger would SLC by CSDPRM. The merger of two dominant firms would obviously result in the substantial lessening of competition, and therefore the Commission invited the merging parties to show if there were any efficiencies that may meet the criteria laid down in §11(10).
The merging parties mentioned seven different "efficiencies" to flow from the acquisition (Para 156 of the Order), none of which meets the criteria specified §11(10).; and the Order does not contain any analysis to demonstrate that criteria was met with respect to any of these proclaimed efficiencies. The foremost efficiency mentioned by the parties -- which could possibly have some relevance -- was that "by merging their supply and demand via integration of their applications and cross-dispatching drivers in order to increase network density and lower wait times." However, in our view, this is not an efficiency which "could not reasonably have been achieved by a less restrictive means of competition." Drivers and riders do multi-homing, i.e., be part of substitutable platforms. The Commission noted that "Uber and Careem users, both riders and drivers, are able to multi-home." (para 43). There are applications, like Ridester.com or Mystro, which allow drivers to receive demands from various applications, and collate those demands for drivers, without them having to switch between the platforms. Expedia, Tripadvisor and many other apps provide integrated search on the platforms of airlines for air travelers. The integration of applications touted as a merger-specific efficiency is already being done without the need of the parties to merge. On the contrary, the integration of platforms poses a serious threat to present and future competition in the market, as the "two-sided network effects may enable a large platform to become dominant and insulated from competition from smaller platforms with fewer participants.".
The Commission, however, without adverting to any cogent analysis, noted in the Order that the efficiencies proclaimed by the merging parties "could not be achieved through a less restrictive means of competition". In our view, the benefits of the purported efficiencies that the transaction is expected to bring about are hardly likely to outweigh the "adverse effects of the absence or lessening of competition." (Para 159). This aspect has not really been addressed in the Order. Nonetheless, the Commission was convinced that the claimed efficiencies meet the three pronged test of §11(10). This being the case, it is actually estopped from invoking §11(11), and thus 11(11)(b) -- the last resort saving option for a merger transaction. The opening words of §11(11) are: "In case the Commission determines that the transaction under review does not qualify the criteria specified in sub-section (10)." The Commission went against the qualifier of §11(11) and imposed conditions to its approval under §11(11)(b).
While §11 is silent as to the types of conditions the Commission may impose, the conditions must further the Commission's sole objective for its establishment, i.e., maintaining or enhancing competition. The Commission approved the merger with the following eight behavioral conditions on Uber: (i) No Contractual Exclusivity; (ii) Service Fee Cap; (iii) Annual Fare Increase; (iv) Surge Price Cap; (v) Innovation and Quality; (vi) Access to Map Data; (vii) User Data; and (viii) Personalized Pricing.
A bare perusal of the so-called conditions shows that they are useless, ineffective, and unlawful and it is difficult to perceive how these will be able to maintain or enhance competition, rather these conditions are likely to stifle any future competition. Let's briefly analyze them.
a. No contractual Exclusivity: exclusive vertical agreements by a dominant player, which precludes supply for new entrants, are always condemned as abuse of dominant position under section 3 of the Act. Uber is at present not concluding exclusive agreements, and cannot do so in future in view of the prohibition of section 3. Thus, the condition will achieve nothing, and is, therefore, useless.
b. Service Fee Cap: the Commission prohibited Uber from lowering its service fee cap, in case a new entrant enters the market, which it charges at the rate of 22.5 % to 27.5% from drivers for every ride. The Commission presumptively imposed this condition to prevent future predatory pricing. Predatory pricing occurs when a seller sells his products at a price which is below its marginal cost. In markets based on digital platforms, the marginal cost is close to zero if not zero. This condition encourages the new entrants to charge service fee closer to that charged by Uber, and will therefore stifle future price competition, and condemns the drivers to ever pay a service fee of at least 22.5%.
c. Annual Fare Increase: The Commission prohibited Uber not to raise its annual fare increase beyond 12.5% per year above inflationary cost increases. The cost of inflation hovers around 8% in Pakistan. The Commission in essence authorized Uber to raise its fare annually up to 12.5% plus 8%, which amounts to 20.5%. Is this a condition or an allowance granted to Uber under the guise of a condition? Price regulation is beyond the mandate of the Commission - the condition is ultra vires the power of the Commission.
d. Surge Price Cap: Commission allowed Uber to use up to 2.5 times multiplier as surge price cap. Surge price is employed by Uber's algorithm when there is high demand and low supply. Surge price of 2.5 times the normal fare would mean that if the normal fare is Rs. 100, Uber can now charge Rs. 250 for the same ride whenever there is high demand. By way of comparison, Egypt's competition Authority has imposed a surge fee cap of 10 per cent so that the fare can only increase from Rs. 100 to Rs. 110 and not beyond. Again, is this a condition or an allowance granted to Uber? Is the Commission working to safeguard the interest of Pakistani consumers or Uber?
e. Innovation and Quality: The Commission directed Uber to dedicate 10 engineers who will primarily work on R&D activities focused on bringing innovations to wider region, including Pakistan. Again, is this a condition? Uber is engaged in R&D activities anyway. Why is the Commission so worried about the wider region and not just Pakistan? How is this condition going to maintain or enhance competition in Pakistan?
f. Access to Map Data: "Uber shall grant access to a Ridesharing Services Provider upon the latter's request to Careem's points of interest map data." Why did the Commission not require Uber to provide its own points of interest map data, and why just that of Careem? Why the Commission is so much concerned that the data be provided on "one-time basis", and that too against a license fee and how the licensee will use the data? How will this condition enhance competition in the market?
g. User Data: For riders to port their data to alternative ridesharing suppliers, Uber shall continue to grant riders access to their data. User data is comprised of date, time and place the rider hailed the ride and where did he go, and how many times? Why would a rider share his travelling history with a new service-provider, and how is this condition going to enhance competition in the relevant market?
h. Personalized Pricing: The Commission banned Uber from giving any concession to frequent travellers based on the number of rides they took. Loyalty discounts are not anticompetitive, unless those discounts put "other parties at a competitive disadvantage." (§3(3)(e) of the Act). This is usually not the case when loyalty discounts are given and the logic of this condition is not obvious. And insofar as riders are concerned, what possible disadvantage could occur to other riders by giving discount to one rider? Again, one wonders whether this a condition or an allowance to Uber? And how is this condition going to maintain or enhance competition?
A merger resulting in "absence of competition" could only be saved if it meets the efficiencies test stipulated in §11(10); §11(11)(b) conditions cannot save such a merger, therefore the Order runs afoul the letter and spirit of §11. That said, some conditions are in fact allowances/licenses granted to Uber, such as the rate of surge price, service charge, and annual fare increase, and it is questionable whether the Commission is acting within its legal mandate by imposing these conditions.
The Commission also seems to have erred in technical analysis by clubbing all geographic markets as one, and by failing to consider the supply-side of the platform, the drivers, and the impact the acquisition would have on the drivers by elimination of choices. There were inconsistencies in the order, for example, it was effectively ordained that the efficiencies criteria specified in §11(10) was satisfied, yet it invoked §11(11) ; it is stated that the ridesharing market will "fetch a significant FDI" (paragraph 242) while in paragraph 151 it is noted "that Ridesharing Market is not capital intensive." As a matter of fact, after the acquisition there will be at least 22% of revenues from the application-based ride-sharing industry that will be repatriated from Pakistan, causing a significant monetary drain in the economy of Pakistan.
The Commission exhibited poor understanding of the concept of "competition." It noted in paragraph 32 of its the Order that in the app-based ride-sharing, the rider gets an estimate of the fare before the ride which "saves consumers from the hassle of haggling as with a taxi, as they don't get any estimates and which can vary with different cabbies offering different rates for the same route." Indeed, price competition is a hallmark of a competitive market. Yet, the Commission views the variation in price for the same route as something problematic, and a fare estimate given by a monopolist service provider, who says "take it or leave it" as a useful thing! The comprehension of the Commission as to what constitutes a competitive market gives rise to concerns regarding its ability to properly comprehend and apply recognized competition norms as embodied in the law and protect the interest of Pakistani consumers.
It is also worth noting that the Commission has removed the Order from its website, despite section 29(b) of the Act, suggests the possibility of mala fides. A judicial review of the Order seems warranted to protect competition and to safe guard the interest of Pakistani consumers, who were taken for ride by the Commission before Uber!
(Prof. Khalid Mirza served as the Founder Chairman of the Competition Commission of Pakistan, and at present teaches at the Lahore University of Management Sciences. Dr Joseph Wilson served as a founder member, and later as Chairman, of the Competition Commission of Pakistan and at present teaches as adjunct Professor at McGill University's Faculty of Law. The views expressed in this article, however, are not necessarily those of the newspaper).