Elsevier

Knowledge-Based Systems

Volume 204, 27 September 2020, 106093
Knowledge-Based Systems

Optimal business model for the monopolistic ride-hailing platform: Pooling, premier, or hybrid?

https://doi.org/10.1016/j.knosys.2020.106093Get rights and content

Highlights

  • We compare pooling, premier, and hybrid models employed by ride-hailing platforms.

  • The optimal model depends on heterogeneous passengers’ time-sensitive cost.

  • It also depends on the operating cost of the platform’s self-operating vehicles.

Abstract

Pooling, premier, and hybrid are three business models employed by ride-hailing platforms. We establish an analytical framework to examine these three models for addressing the platform’s optimal business decision. Our results reveal that both the time-sensitive cost of heterogeneous passengers and the operating cost of the platform’s self-operating vehicles play critical roles in the platform’s choice of optimal model. If the operating cost of the platform’s self-operating vehicles is relatively high, the platform should choose the pooling service model when passengers have a low ratio of time-sensitive cost between using the pooling service and using the premier service. The premier service model should be implemented if this ratio is in the middle range and the operating cost is sufficiently low. Otherwise, the hybrid service model is optimal. We characterize the conditions under which the pooling service model and the premier service model can achieve Pareto improvement for the platform and passengers. Furthermore, if the ratio is in the middle range, the pooling service model is more beneficial for passengers, while the premier service model is more profitable for the platform.

Introduction

Ride-hailing platforms have attracted much attention and achieved extraordinary growth over the past 5 years due to the dramatic advances in technology and economy. These platforms, such as Grab, Uber, and Didi, have brought substantial convenience to people’s lives, enabling them to request rides using their smartphones rather than hailing taxis on the street [1]. In June 2013, Grab claimed that it could receive an order every 8 s, and its total order volume had reached 10,000 per day [2]. In May 2014, Grab said it had 1.2 million users in Southeast Asia. In 2015, Uber completed its 1-billionth ride (Tracxn Blog, 2016), which was below the 1.43 billion rides completed by Didi at that time (Marketing Interactive, 2016). By June 2016, Uber had completed its 2-billionth ride [3]. In 2017, Didi provided nearly 8.48 million rides on the Chinese New Year holiday,1 and this grew rapidly to nearly 30.67 million rides on the 2018 holiday.2

In fact, one platform usually occupies most of the ride-hailing market of an individual country or region. For instance, in China, Didi is the dominant player in the ride-hailing market, with a market share exceeding 60% and services extending to 32 cities.3 Didi controls 80% of the private-car market and 99% of the taxi market.4 In the United States, at the beginning of 2017, Uber’s share of the ride-hailing market was 84%. In Southeast Asia, the leader of the ride-hailing platform is Grab, occupying a 97% market share in the third-party ride-hailing market and 72% in the private-vehicle market. In November 2017, Grab notched up 1 billion rides, with 66 concurrent rides in 1s across seven countries.

Unlike the traditional taxi companies that provide only unified taxi service, a ride-hailing platform offers both premium service (premier) and economy service (pooling), which makes the ride-hailing market extraordinarily complicated. On the one hand, the premier service is professional and personalized, because it is usually provided to passengers with customized demands. Examples of a premier service include Didi Premier, Didi Luxe, Uber Black, Uber XL, and Uber Black SUV. On the other hand, the economy service, such as Didi Express, Didi Express Pool, Uber Pool, and Uber X, is offered to alleviate traffic pressure by sharing idle resources, which is in line with the theme of global sustainable development. A significant difference between the economy service and the premier service is that the former divides travel expenses among all occupants of a vehicle (drivers and passengers) while the latter serves only one passenger who pays the full travel fee. Note that some ride-hailing platforms allow the traditional taxis to join the service in order to provide a convenient and conventional taxi service. For instance, Didi cooperates with 500 taxi companies to offer a traditional taxi service in China and Brazil,5 helping the traditional taxi drivers respond more quickly to passengers. However, our study does not consider this situation, since it is extremely uncommon in most regions or countries. Based on whether the service is premium or economy, ride-hailing platforms usually adopt three kinds of business models, namely, the premier service model, the pooling service model, and the hybrid service model.

Platforms adopting the premier service model, such as Grab, UCAR Inc., and Lyft Premier, have three common characteristics. First, drivers who have joined the platform are divided into hired drivers (self-operating employees) and cooperating drivers (private car owners). Second, each passenger uses one car for her personal demand and pays the full travel fee. Third, the platform needs to bear total operating costs of self-operating vehicles. Such a service is available not only to high-end business people but also to those who require professional and customized service.

Pooling service first appeared in the late 1990s and became a prevalent travel mode in the 2000s [4]. According to US Census statistics, pooling has risen stably from 10.1% to 10.7% of the total transportation market [5], [6]. These ride-hailing platforms, such as Sidecar, Via, and Kuaidi Carpooling, play a crucial role as an intermediary in rationally allocating the remaining capacity of private cars and providing information to passengers for alleviating traffic congestion and solving the difficulties associated with hailing taxis. The salient features of the platform under the pooling service model are summarized as follows. First, most of its drivers are private car owners.6 Second, a passenger always shares a car with others. Third, the platform does not need to own its self-operating vehicles, and thus, it has virtually no operating costs—a marked difference between this model and the premier service strategy.

The hybrid service model, which combines the features of the premier and pooling service models, is gaining popularity among ride-hailing platforms, such as Uber and Didi. Under this strategy, passengers can use either the premier service with paying the full fare or the pooling service by splitting the bill.

In fact, the business model decision of a ride-hailing platform depends on the passengers’ heterogeneous cost sensitivity and the cost of the platform, rather than the platform’s size or financing quota. For example, Didi and UCAR Inc. have similar scales and financing quotas, but they adopt different service strategies. Didi adopts the hybrid service strategy, while UCAR Inc. focuses on the premier service. Thus, the goal of this study is to explore which business model a ride-hailing platform should adopt in such a monopoly market. In particular, we seek to deal with the following research issues. First, what is the optimal price-setting and business model decision for the ride-hailing platform? Second, how does passengers’ heterogeneous cost sensitivity of waiting time affect market outcomes under different models of the ride-hailing platform? Third, how do differences in passengers’ heterogeneity ratio between the pooling service and premier service affect passenger surplus under different strategies of the platform? Finally, is there an absolutely dominant strategy choice for the platform?

We develop an analytical model to answer the above research questions. We consider that a ride-hailing platform can offer differentiated services with different intrinsic values in a monopoly market. A principal–agent relationship is formed between the ride-hailing platform and the driver. The platform serves as the principal and shares benefits with drivers, while the driver serves as the agent and completes various services. This paper analyzes market outcomes and price-setting by considering the time-sensitive cost of heterogeneous passengers under different models adopted by the platform. In addition, we consider passengers of different types including long distance vs. short distance and single order vs. multiple orders in the extension section.

The theoretical contribution and practical significance of this paper are as follows. Theoretically, this paper constructs the optimal pricing model of the ride-hailing platform based on game theory. Considering the pricing decision issues under the single service model and the hybrid service model, this paper fills the gaps in the related research on sharing economy and two-sided platforms. Practically, based on the model analysis, this paper provides the specific conditions and suggestions to the platform for selecting pricing and business strategy. Especially for the existing ride-hailing platform giants Didi and Uber, which use hybrid service strategy, they need to completely distinguish the inherent value of the two services and reasonably implement price discrimination strategies in order to gain more profits.

The remainder of the paper is organized as follows. Section 2 surveys the related literature. Section 3 describes our basic model. In Section 4, we analyze the passenger surplus and social welfare when the platform chooses different optimal strategies. Section 5 extends our model to consider the passengers with different types, including long distance vs. short distance and single order vs. multiple orders. Conclusions and future directions are presented in Section 6.

Section snippets

Literature review

Our work is primarily connected to two streams of literature: the research on ride-hailing platforms, and the research on pricing strategy with network effects in a two-sided market.

First, the literature on ride-hailing platform considers its market equilibrium in the transportation field. Yang et al. [7] establish a network model to describe the demand and supply equilibrium of taxi services under fare structure and fleet size regulation in both a competitive and a monopoly market. The authors

Basic model

Consider that a two-sided ride-hailing platform provides either a pooling service or a premier service for passengers (denoted as “she” in this paper). Based on the two services, the platform has three alternative strategies in our model setup. First, it can employ the pooling service model, in which case all passengers share rides with other people (Case POS). Second, the platform can choose to adopt the premier service model, which offers more professional and personalized services to

Consumer surplus and social welfare

In this section, we explore the implications of the optimal results on consumer surplus and social welfare. Consumer surplus is the benefit that a customer gains from participating in the platform, which equals the area to the right of the price under the demand curve (see, e.g., [45], [46]). To facilitate understanding, we use the term “passenger surplus” to represent the benefit obtained by the passenger.

For each model, passenger surplus can be calculated as follows: PSj=0tUidt(j=POS,PRS,HYS)

Model extensions

In this section, by focusing on different behaviors of passengers and analyzing two model variations we show that results are robust under the pooling service strategy and the premier service strategy.

Conclusions and future work

The development of the sharing economy has promoted the progress of ride-hailing platforms. These platforms have brought more convenient travel experiences to humankind. In this study, we provided an analytical model that captures some important market characteristics of a monopolistic ride-hailing platform and examined the optimal business model decision for the platform in the presence of time-sensitive cost for heterogeneous passengers.

Our results yielded several main findings. First, the

CRediT authorship contribution statement

Xin Wei: Conceptualization, Methodology, Formal analysis, Writing - original draft. Guofang Nan: Writing - original draft, Formal analysis, Funding acquisition. Runliang Dou: Funding acquisition, Writing - original draft, Writing - review & editing. Minqiang Li: Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research is partially supported by research grant from the National Social Science Foundation of China (Grant No. 18BGL095) and the Key Program of National Natural Science Foundation of China (No. 71631003).

References (50)

  • LiS. et al.

    Network effects in online two-sided market platform: A research note

    Decis. Support Syst.

    (2010)
  • KungL.C. et al.

    The optimal pricing strategy for two-sided platform delivery in the sharing economy

    Transp. Res. E

    (2017)
  • ChoW.Y. et al.

    Versioning of information goods under the threat of piracy

    Inf. Econ. Policy

    (2010)
  • G. Feng, G. Kong, Z. Wang, We are on the way: analysis of on-demand ride-hailing systems, 2017....
  • Lee Terence, GrabTaxi is one of Southeast Asia’s hottest startups. Here’s how it’s kicking ass. Tech in Asia. Archived...
  • A.J. Hawkins, Uber just completed its two-billionth trip @Verge, 2016....
  • ChanN.D. et al.

    Ridesharing in North America: past, present, and future

    Transp. Rev.

    (2012)
  • US Census Bureau, Sex of Workers by Means of Transportation. 2004 American Community Survey(Washington, DC: US Census...
  • US Census Bureau, Commuting Characteristics by Sex. 2005 American Community Survey 1-YearEstimates (Washington, DC: US...
  • S. Benjaafar, H. Bernhard, C. Courcoubetis, Drivers, riders and service providers: the impact of the sharing economy on...
  • C. Riquelme, S. Banerjee, R. Johari, Pricing in ride-share platforms: a queueing-theoretic approach, 2015....
  • M. Hu, Y. Zhou, Price, wage and fixed commission in on-demand matching, 2017....
  • GudaH. et al.

    Your Uber is arriving: Managing on-demand workers through surge pricing, forecast communication, and worker incentives

    Manage. Sci.

    (2019)
  • C.T. Lam, M. Liu, More than taxis with an app: how ride-hailing platforms promote market efficiency, 2018....
  • BaiJ. et al.

    Coordinating supply and demand on an on-demand service platform with impatient customers

    Manuf. Serv. Oper. Manag.

    (2018)
  • Cited by (11)

    • A Systematic Literature Review on Machine Learning in Shared Mobility

      2023, IEEE Open Journal of Intelligent Transportation Systems
    View all citing articles on Scopus
    View full text