Open or Not? Operation Strategies of Competitive eCommerce Platforms from an Ecosystem Perspective

Abstract

According to the traditional theory of industrial organizations, differentiated competition leads to higher returns. However, Chinese eCommerce platforms tend to homogenize in the choice of operational strategies, which deviates from the principle of differentiated competition. Some competitive eCommerce platforms were described based on the hotelling model, and a sequential game model for the competition of ecosystems was built to analyze the choice of operation strategies. From the perspective of a business ecosystem, this paper studied the impact of the strategy choice of the core platform on the ecological profit. The results showed that an open strategy would always improve the performance of business ecosystems and could even benefit the competitive ecosystem because of its spillover effect. The equilibrium of the competitive platforms was related to the game strategies, independent of the distribution contracts of the suppliers. We found that the platforms’ positions deviated from the maximum difference principle due to network externalities, and differentiation did not improve the profits of platforms.

1. Introduction

Taobao (NYSE:BABA) and Jingdong (NASDQ:JD), which are two eCommerce companies, account for more than 90% of China’s online retail market shares. According to the Chinese ECommerce Research Centre, China’s eCommerce transactions in 2021 reached RMB 42.13 trillion, up to 10.4% year-on-year. The Gross Merchandise Volume (GMV) of Jingdong Mall reached RMB 3.30 trillion, while Alibaba’s platforms achieved a GMV of RMB 7.49 trillion. ECommerce platforms (e.g., Amazon, JD, and Taobao) often have both platform-owned (first-party) and third-party stores, although they may initially focus on one type of store [1]. The self-operating strategy is the first-party strategy, and the open strategy is the strategy of introducing third-party merchants. A hybrid strategy refers to a strategy in which the platform not only operates on its own but also introduces third-party merchants. Amazon, for example, did not have third-party stores until 2000, and JD launched an open platform plan in 2010 to bring in third-party sellers. In contrast, Taobao, which has been a third-party marketplace since its inception, launched Tmall in 2017, a first-party store that sells directly to customers. Which strategy is better, the direct operation of Jingdong or the hybrid strategy that introduces third-party merchants?

Referring to differentiated competition theory, does the competitive advantage of eCommerce platforms come from differentiated positioning? Both Jingdong and Taobao have adopted a homogeneously hybrid operating strategy, with competition including financial payment, logistics distribution, and other businesses. Why do competitive eCommerce platforms deviate from the differentiation principle, and why does differentiated competition not necessarily lead to profit maximization?

The business ecosystem is an economic community supported by a foundation of interacting organizations and individuals—the organisms of the business world. This economic community produces goods and services of value to customers who are themselves members of the ecosystem [2,3]. The business ecosystem created by eCommerce platforms such as Taobao and JD.com refers to an economic community composed of platform enterprises, merchants, suppliers, and consumers. The competition between the business ecosystems is essentially the competition of the core firms, which construct a network platform of value sharing [4]. In particular, the competition between eCommerce platforms is quite different from traditional competition; scholars regard the platform as a bilateral market based on the business ecosystem [5], leading the whole system to participate in market competition. The strategic decision of the platform determines the competition and the structure of the ecosystem [6]. Moreover, an eCommerce platform undertakes the coordination and governance functions of the ecosystem [7]. Therefore, how to deal with the competitive relationship between platforms is an important issue, which concerns ecosystems’ competition. The competition between business ecosystems is the competition between different platforms [8,9,10]. Currently, China’s eCommerce market is dominated by Taobao and Jingdong, which determine their respective ecosystem structures. For example, when a platform decides to sell its own products, the ecosystem structure is relatively vertical and centralized, as with Jingdong; alternatively, when a platform decides to open to merchants, the ecosystem structure is relatively flat and scattered, as with Taobao. Both dominate their industrial structure and have become the shapers, rule designers, and managers of their business ecosystem, locking the leadership of the ecosystem, thus ensuring sustainable competitive advantages.

Most studies on platforms focus on the pricing and searching mechanism [4,11,12,13], and the reputation [14,15,16]. Recently, studies have appeared at the micro level based on two-sided market theory, focusing on the impact of network externality [17], platform strategy [18], platform non-paid complementarity [19], and other factors impacting the competition of eCommerce platforms [13]. The existing literature has achieved a lot in the field of platform competition [20]. The study of intra-platform competition includes the study of the complementarity between modules and the competition between platform owners and complementors [21]. In the study of competition between platforms, the competition between potential entrants and existing platforms has received great attention [22]. Moreover, in the digital era, the platform market structure [23], platform size, platform identity and other dimensions [24] will have an impact on the competition among platforms. According to the differentiation strategy, enterprises should adopt differentiated positioning to achieve maximum competitive advantage [25]. Several recent studies seem to demonstrate the applicability of differentiation strategies in platform competition [26]. However, it has also been pointed out that differentiation strategies improve the performance of a platform only when the platform system is extremely unique in comparison to its competitors [27]. However, both JD and Taobao have adopted a hybrid management strategy of homogenization, which deviates from the principle of differentiation. Some studies have already recognized the platform ecosystem but most of these have focused on core firms [4,28,29,30]; lacking is the in-depth study of the operation mechanism of intra-ecosystems, the competition mechanism, and the inter-ecosystem symbiosis [31,32,33]. Therefore, looking at eCommerce from the perspective of the eCommerce ecosystem, this paper focuses on two questions: (1) Why do eCommerce platforms deviate from the principle of differentiated competition? (2) Why does differentiated competition not bring maximum profit?

2. Game Model

2.1. Model Description and Assumptions

The equilibrium of the game reveals the strategy selection or the influence mechanism of the platform as the core enterprise of the ecosystem. Based on the literature above, the decisions made by both parties are observable, as the parties collaborate closely. However, they are not verifiable. Meanwhile, the agent is assumed to be risk averse in the eCommerce business ecosystem. The sequence of dynamic game decisions within the ecosystem is summarized as follows. First of all, the business decision of the platform in the ecosystem is selected; the platform and merchants sign a profit distribution contract and set the optimal wholesale price w under the opening strategy (OP) of the platform; if a direct operation strategy (SO) is selected by the platform, then the product on each platform is priced as p_i; if the open strategy is selected, the proportion of the profit distribution contract between the platform and the merchants in the ecosystem is η.

Game theory is often used in the research of platforms and ecosystems. Li J studied the competition between potential entrants and existing platforms with the method of game theory [22]. Nariaki Nishino built a game model among various players in the ecosystem [34]. Yusuke Zennyo examines strategic contracting between a monopoly platform and suppliers that sell their goods through the platform [35]. Han Y constructed a three-party game among the government, platform organizations and enterprises [36]. This paper constructs a Hoteling game model between two ecosystems. The Hotelling model describes the competition between two sellers selling the same product in a linearly bounded market, leading to a discussion of spatial competition [37] and product differentiation [38]. Modified by many scholars, the Hotelling model has evolved from simple positional games to strategic price games, political games, etc. [39]. We illustrate the competitive process between two eCommerce ecosystems based on the differentiation positioning of the Hotelling model [40,41]. The model described in this paper is one of cooperation between the core platform enterprise and the affiliated third-party merchant, where the two partners supply products and services to consumers. In this way, the ecosystem in the following model is simplified. Ecosystems in this paper refer to economic coalitions that are on the supply side and are in the same group competing with other platform ecosystems. Suppose two competitive eCommerce platforms i = (A, B) in the market are game players, and both are core firms in the business ecosystem. According to the Hotelling model, it is assumed that the products of the two eCommerce platforms are located at the line x_i ∈ [0, 1]. Here, we assume that the position of platform i is x_A < x_B; in this case, the location shows the product differentiation between the two eCommerce platforms, including the brand, logistics, and distribution costs, and the distance between the locations would represent the switch cost. To simplify the analysis, the product price of the two platforms is p_i, and the marginal cost of each platform is 0.

It is assumed that consumers follow uniform distribution on the Hotelling line. The consumer surplus is CS, which maintains sufficient advantages to ensure the shopping motivation of consumers. In addition, the disutility of consumers from an online shopping experience is extremely common [42]. In this case, $-t {\left(x-x_i\right)}^2$ represents the gap between consumers’ utility and their expectations, t is the utility coefficient, and the switch cost function is f(u) = u(x_A − x_B)², where u is an external variable and u′ > 0 as the switch cost would increase with the distance. The utility function of consumers at position x is

$$U_i=CS-p_i-t{\left(x-x_i\right)}^2-u{(x_i-x_B)}^2$$

(1)

To maximize the CS, consumers choose online platforms and assume that they are at the location $\tilde{x}$ ∈ [0, 1]; the equilibrium $\tilde{x}$ must satisfy the condition U_A = U_B; then,

$$\tilde{x}=\frac{x_B+x_A}{2t}+\frac{p_B-p_A}{2t\left(x_B-x_A\right)}-\frac{u\left(x_B-x_A\right)}{2t}$$

(2)

When the eCommerce platforms in the ecosystem select the open strategy (OP), the platform and the merchants on it within the business ecosystem execute the profit distribution contract. Third parties and the platform are required to determine the optimal wholesale price first to maximize the overall profit of the ecosystem. The profit distribution proportion is determined by the bargaining power of firms in the ecosystem. Assuming that the platform dominates the operation of the ecosystem, the redistribution of system benefits will be transferred to the platform based on a specific proportion.

Finally, assuming that the game players in the market are symmetrical, the game strategy set of platform i = (A, B) is {(SO, SO); (SO, OP); (OP, SO); (OP, OP)}.

2.2. Game Equilibrium between Platforms

2.2.1. Strategy (SO, SO)

If platforms A and B selected the (SO, SO) strategy, neither of them would open the platform to merchants. The sequential game equilibrium was solved with backward induction. First, it was assumed that all the players were rational, so that the pricing of both platforms could cover all online markets. Then, the profit function ${\mathsf{\prod }}_{\mathrm{i}}$ of the platform i = (A, B) was

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{\mathrm{A}}=p_A\tilde{x}\\ {\mathsf{\prod }}_{\mathrm{B}}=p_B(1-\tilde{x})\end{array}\right.$$

(3)

To maximize the profit of the platforms, the equilibrium prices of platform A and platform B were

$$\left\{\begin{array}{c}{p}_{A1}=\left[\left(x_B-x_A\right)\left(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u\right)\right]/4\\ p_{B1}=\left[\left(x_B-x_A\right)\left(4t-x_{A}t+x_{A}u-x_{B}t-x_{B}u\right)\right]/2\end{array}\right.$$

(4)

If we substitute p_A1 and p_B1 into equation $\tilde{x}$ to obtain the equilibrium point of the eCommerce platform on the Hotelling line, it gives

$${\tilde{x}}_1=(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)/8t$$

(5)

Then, the profit ${\mathsf{\prod }}_{\mathrm{i}}$ of platforms A and B was

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{A1}=\left[\left(x_B-x_A\right){(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2\right]/32t\\ {\mathsf{\prod }}_{B1}=\left[\left(x_B-x_A\right){(x_{A}t-4t-x_{A}u+x_{B}t+x_{B}u)}^2\right]/16t\end{array}\right.$$

(6)

2.2.2. Strategy (OP, SO)

If the platforms A and B selected the strategy (OP, SO), assuming that platform A selected the OP strategy, the merchants and platform A would then work closely in the eCommerce ecosystem, while platform B selected the SO strategy to directly supply products to the market. It was assumed that the production costs were the same between the opening platform and the direct operation platform. The merchants no longer had a production cost advantage to distinguish the influence of the production cost advantage and the choice of eCommerce platform strategy on the platforms. The game was constructed according to the decision sequence above, platform A had an open strategy, and the merchants on the platform coordinated with each other on the profit distribution contract (w_A2, η_A), where the three-staged sequential game was proposed. w_A2 was the optimal wholesale price of platform A, and η_A was the proportion of profit distribution contract between platform A and merchants. Correspondingly, the profit function $S_{A2}$ of the merchants on platform A was

$$S_{A2}=\left(1-{\eta }_A\right)w_{A2}\tilde{x}$$

(7)

According to the transfer payment contract within the eCommerce ecosystem, the profit functions ${\mathsf{\prod }}_{\mathrm{i}}$ of eCommerce platforms A and B were

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{A2}=(p_A-w_{A2})\tilde{x}+{\eta }_A{w}_{A2}\tilde{x}\\ {\mathsf{\prod }}_{B2}=p_B(1-\tilde{x})\end{array}\right.$$

(8)

The equilibrium was solved with backward induction. When eCommerce platform A adopted the open strategy, the platform and merchants established a profit distribution contract. ECommerce platform A and the merchants negotiated the internal price of the transfer payment and then set the price of the transferring of the payment w_i2 within the ecosystem; the product price of eCommerce A and B was obtained by maximizing their profit function. The price $p_{i2}$ was,

$$\left\{\begin{array}{l}{p}_{A2}=\left[3w_{A2}-4x_{A}t-3w_{A2}{\eta }_A+4x_{B}t-x_A^{2}t+x_A^{2}u+x_B^{2}t+x_B^{2}u-4x_A{x}_{B}u\right]/8\\ p_{B2}=\left[w_{A2}-4x_{A}t-w_{A2}{\eta }_A+4x_{B}t+x_A^{2}t-x_B^{2}t-x_B^{2}u+2x_A{x}_{B}u\right]/2\end{array}\right.$$

(9)

If we substitute p_A2 and p_B2 into equation $\tilde{x}$ to obtain the equilibrium of the platform on the Hotelling line, in this case, the equilibrium point ${\tilde{x}}_2$ was

$${\tilde{x}}_2=\frac{w_{A2}{\eta }_A-4x_{A}t-w_{A2}+4x_{B}t-x_A^{2}t+x_A^{2}u+x_B^{2}t+x_B^{2}u-2x_A{x}_{B}u}{8t\left(x_B-x_A\right)}$$

(10)

So, the total profit ${\mathsf{\prod }}_{EA2}$ of eCommerce ecosystem A was

$${\mathsf{\prod }}_{EA2}={\mathsf{\prod }}_{A2}+S_{A2}=p_{A2}{\tilde{x}}_2$$

(11)

Then, the total profit of ecosystem A could be maximized. Let $\frac{{\mathrm{d}\mathsf{\prod }}_{EA2}}{\mathrm{d}{w}_{A2}}=0$, and we solve the first order equilibrium condition for w_A2 as follows.

$$w_{A2}=\frac{\left(x_B-x_A\right)(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}{3(1-{\eta }_A)}$$

(12)

If we substitute w_A2 into the above equations p_A2, p_B2, and equation $\tilde{x}$, then the refined equilibrium of the subgame would be

$$\left\{\begin{array}{l}{p}_{A2}=\left[(x_B-x_A)(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)\right]/2\\ p_{B2}=\left[(x_B-x_A)(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)\right]/2\\ {\tilde{x}}_2=(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)/12t\end{array}\right.$$

(13)

In the case of the strategy (OP, SO), the equilibrium profit of the platform and merchants in the eCommerce ecosystem was

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{A2}=\left[\left(x_B-x_A\right){(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2\right]/72t\\ {\mathsf{\prod }}_{B2}=\left[\left(x_B-x_A\right){(x_{A}t-8t-x_{A}u+x_{B}t+x_{B}u)}^2\right]/36t\\ S_{A2}=\left[\left(x_B-x_A\right){(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2\right]/36t\end{array}\right.$$

(14)

2.2.3. Strategy (OP, OP)

If both eCommerce platforms A and B selected the open strategy, that is (OP, OP), the products of the platform came from merchants. It was assumed that the production cost of the merchants on the platform in the eCommerce ecosystem was c, which was equal to the cost of the self-operated products on the platform. The merchants enjoyed no production cost advantage. Then, the platform signed a profit distribution contract (w_i3, η_i) with the merchants, according to the decision-making sequence in the eCommerce ecosystem. Firstly, the platform and merchants gave the price w_i3; secondly, the eCommerce platform decided the product price p_i, at the same time; finally, the profit distribution proportion η_i was determined. The profit functions of the merchants on platforms A and B were

$$\left\{\begin{array}{l}{S}_{A3}=\left(1-{\eta }_A\right)w_{A3}\tilde{x}\\ S_{B3}=\left(1-{\eta }_B\right)w_{B3}\left(1-\tilde{x}\right)\end{array}\right.$$

(15)

The profit functions of platforms A and B were

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{A3}=(p_A-w_{A3})\tilde{x}+{\eta }_A{w}_{A3}\tilde{x}\\ {\mathsf{\prod }}_{B3}=(p_B-w_{B3})\left(1-\tilde{x}\right)+{\eta }_B{w}_{B3}\left(1-\tilde{x}\right)\end{array}\right.$$

(16)

When both eCommerce platforms A and B selected an open strategy, both eCommerce platforms and merchants on the platform first signed a profit distribution contract to determine the price of the transfer payment w_A3 and w_B3. The product price of the eCommerce platforms A and B were

$$\left\{\begin{array}{l}{p}_{A3}=3w_{A3}+w_{B3}-4x_{A}t-3w_{A3}{\eta }_A+4x_{B}t-w_{B3}{\eta }_B-x_A^{2}t+x_A^{2}u+x_B^{2}t+x_B^{2}u-2x_A{x}_{B}u\\ p_{B3}=w_{A3}+w_{B3}-4x_{A}t-w_{A3}{\eta }_A+4x_{B}t-w_{B3}{\eta }_B+x_A^{2}t-x_A^{2}u-x_B^{2}t-x_B^{2}u+2x_A{x}_{B}u\end{array}\right.$$

(17)

If we substitute p_A3 and p_B3 into equation $\tilde{x}$, we obtain the equilibrium point ${\tilde{x}}_3$

$${\tilde{x}}_3=\frac{w_{B3}(1-{\eta }_B)-w_{A3}(1-{\eta }_A)+t\left(x_B^2-x_A^2\right)+4t\left(x_B-x_A\right)+u(x_B^2+x_A^2)-2x_A{x}_{B}u}{8t\left(x_B-x_A\right)}$$

(18)

Now, we consider the total profit of the eCommerce ecosystem A and B

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{EA3}={\mathsf{\prod }}_{A3}+S_{A3}=p_{A3}{\tilde{x}}_3\\ {\mathsf{\prod }}_{EB3}={\mathsf{\prod }}_{B3}+S_{B3}=p_{B3}\left(1-{\tilde{x}}_3\right)\end{array}\right.$$

(19)

Assuming that system A and B simultaneously determined the internal price of the transfer payment of the benefit distribution contract w_A3 and w_B3, then the overall profit of the eCommerce ecosystem can be calculated by substitution. Then, let $\frac{{\mathrm{d}\mathsf{\prod }}_{EA3}}{\mathrm{d}{w}_{A3}}=0,\frac{{\mathrm{d}\mathsf{\prod }}_{EB3}}{\mathrm{d}{w}_{B3}}=0$ the first-order condition of the equilibrium was

$$\left\{\begin{array}{l}{w}_{A3}=\frac{[(x_B-x_A)(8t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)]}{4(1-{\eta }_A)}\\ w_{B3}=\frac{[(x_B-x_A)(8t-x_{A}t-x_{B}t-x_{B}u+x_{A}u)]}{4(1-{\eta }_B)}\end{array}\right.$$

(20)

If we substitute w_A3 and w_B3 into p_A1 and p_B1 and ${\tilde{x}}_2$, we obtain the refined equilibrium

$$\left\{\begin{array}{l}{p}_{A3}=[3(x_B-x_A)(8t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)]/8\\ p_{B3}=[(x_B-x_A)(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)]/2\\ {\tilde{x}}_3=(8t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)/16t\end{array}\right.$$

(21)

When the game strategy of the eCommerce platform was (OP, OP), and the game was balanced, the profits of the platform and merchants in the ecosystem were

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{A3}=(x_B-x_A){(8t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2/128t\\ {\mathsf{\prod }}_{B3}=(x_B-x_A){(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)}^2/64t\\ S_{A3}=(x_B-x_A){(8t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2/64t\\ S_{B3}=(x_B-x_A){(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)}^2/64t\end{array}\right.$$

(22)

3. Competition between eCommerce Platforms

3.1. Platform A Selects an Open Strategy

When an eCommerce platform selected the open strategy, the game equilibrium price was higher than the equilibrium price under the direct operation strategy of the platform due to the entrance of suppliers and the emergence of the price of transfer payment w. Based on the assumptions of the Hotelling model, when comparing ${\tilde{x}}_2$ and ${\tilde{x}}_1$, more users would choose rival online shopping platforms, resulting in an unfavorable position for eCommerce platforms that chose an open strategy. We compared the profit of the platforms with an (OP, SO) strategy and the platforms with an (SO, SO) strategy of an eCommerce ecosystem.

A comparison was made between the total profits of the eCommerce ecosystem when platform A selected the direct operation and open strategies

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{EA1}={\mathsf{\prod }}_{A1}=(x_B-x_A){(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2/32t\\ {\mathsf{\prod }}_{EA2}={\mathsf{\prod }}_{A2}+S_{A2}=(x_B-x_A){(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2/24t\end{array}\right.$$

(23)

Obviously, ${\mathsf{\prod }}_{EA2}>{\mathsf{\prod }}_{EA1}$.

A comparison was made between the profits of the eCommerce ecosystem with different strategies when eCommerce platform B selected a direct operation strategy,

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{EB1}={\mathsf{\prod }}_{B1}=(x_B-x_A){(4t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)}^2/16t\\ {\mathsf{\prod }}_{EB2}={\mathsf{\prod }}_{B2}=(x_B-x_A){(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)}^2/36t\end{array}\right.$$

(24)

When x_i ∈ [0, 1], then

$$\frac{{\mathsf{\prod }}_{EB2}}{{\mathsf{\prod }}_{EB1}}>1$$

(25)

Then, we obtain ${\mathsf{\prod }}_{EB2}>{\mathsf{\prod }}_{EB1}$.

Therefore, by comparing the total profit of eCommerce ecosystem A, it can be found that when the platforms selected the (OP, SO) strategy, the eCommerce platform that selected the open strategy would improve the total profit of the ecosystem as well as the profit of the rival eCommerce platform, which means ${\mathsf{\prod }}_{A1}<{\mathsf{\prod }}_{A2}+S_{A2}$, ${\mathsf{\prod }}_{B1}<{\mathsf{\prod }}_{B2}$, which demonstrates that although eCommerce platforms competed in the market, the selection of an open platform strategy led to a win–win profit for the eCommerce ecosystem.

Considering the profit change of player A who selected the open strategy, the profit function needed to meet the condition ${\mathsf{\prod }}_{A1}\le {\mathsf{\prod }}_{A2}$. This required substituting equations into the condition to obtain ${\eta }_A$ ≥ 7/9. This revealed that only when the allocation ratio was higher than the critical value in the profit distribution contract would the eCommerce platform A select the open strategy. In addition, with the increase in the negotiation ratio η of the profit distribution contract, the eCommerce platform would be more willing to choose an open strategy. It was assumed that an eCommerce platform acted as the core firm to lead the development of the ecosystem; it had a higher allocation of control rights and a relatively larger proportion of the profit distribution in the negotiation. By squeezing the profits of the platform merchants, an eCommerce platform could offset the loss of platform profits caused by the open strategy. When the platform merchants had stronger bargaining power, the profit distribution proportion would be relatively small. Once the control right of an eCommerce platform was weakened, and the distribution parameter was lower than the critical value, eCommerce platform A would choose the direct operation strategy. Meanwhile, the game equilibrium would converge to (SO, SO).

3.2. Both Platforms A and B Select Open Strategies

When platform B selected the open strategy sequentially, the game players selected the (OP, OP) strategy. By comparing equations ${\tilde{x}}_2$ and ${\tilde{x}}_3$, the customers of eCommerce platform B would turn to platform A. When eCommerce platforms in the market both executed an open strategy, we compared the changes in the total profit of the ecosystem formed around platforms A and B.

3.2.1. Profit of Ecosystem B

If platform B selected the open strategy, the overall profit of the ecosystem was ∏_EB3, then

$${\mathsf{\prod }}_{EB3}={\mathsf{\prod }}_{B3}+S_{B3}=(x_B-x_A){(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)}^2/32t$$

(26)

From equation ∏_A2, ∏_B2, S_A2,

$${\mathsf{\prod }}_{EB2}={\mathsf{\prod }}_{B2}=(x_B-x_A){(8t-x_{A}t+x_{A}u-x_{B}t-x_{B}u)}^2/36t$$

(27)

Then, we compared the profit of ecosystem B in the case of the strategy (OP, OP) and the case of the strategy (OP, SO).

$$\frac{{\mathsf{\prod }}_{EB3}}{{\mathsf{\prod }}_{EB2}}>1$$

(28)

Obviously, ${\mathsf{\prod }}_{EB3}>{\mathsf{\prod }}_{EB2}$.

When the equilibrium profit of platform B, which selected the open strategy, was considered, ${\mathsf{\prod }}_{B2}\le {\mathsf{\prod }}_{B3}$. Then, equations ∏_A2, ∏_B2, S_A2 and ∏_A3, ∏_B3 were substituted, and S_A3 was converted into the equation to obtain ${\eta }_B$≥ 9/16. Only when the allocation proportion in the profit distribution contract was higher than the value on the equilibrium would platform B select the open strategy. As long as the proportion of the profit distribution with the merchants was higher than the equilibrium value, the open strategy would be chosen by platform B, which would not only improve its own profit, but also improve the overall profit of both eCommerce ecosystems.

3.2.2. Profit of Ecosystem A

According to the sequence of the sequential game mentioned above, the profit of eCommerce ecosystem A with the strategy (OP, OP) was

$$\left\{\begin{array}{l}{\mathsf{\prod }}_{EA2}={\mathsf{\prod }}_{A2}+S_{A2}=(x_B-x_A){(4t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2/24t\\ {\mathsf{\prod }}_{EA3}={\mathsf{\prod }}_{A3}+S_{A3}=3(x_B-x_A){(8t+x_{A}t-x_{A}u+x_{B}t+x_{B}u)}^2/128t\end{array}\right.$$

(29)

Compare

$$\frac{{\mathsf{\prod }}_{EA3}}{{\mathsf{\prod }}_{EA2}}>1$$

(30)

where ${\mathsf{\prod }}_{EA3}>{\mathsf{\prod }}_{EA2}$. According to the above, when the strategy (OP, SO) was selected, the overall profit of eCommerce ecosystem A was higher than that of platform A when it executed the direct operation strategy. However, when the strategy (OP, OP) was selected, the overall profit of the ecosystem B was also higher than the overall profit of the system when platform B executed the direct operation strategy.

In summary, the study demonstrates that an open platform strategy is more conducive to improving the overall profit of the eCommerce ecosystem and overall social welfare, while the selection of a direct operation strategy leads to higher opportunity costs for the eCommerce platform. Meanwhile, given that the eCommerce platform A always selected the open strategy, due to x_i ∈ [0, 1], ${\mathsf{\prod }}_{A2}<{\mathsf{\prod }}_{A3}$, platform A either executed the open strategy or the direct operation strategy.

4. Equilibriums

According to the process of the sequential game and strategy analysis above, the game strategy equilibrium between competitive eCommerce ecosystems was further analyzed.

Proposition 1.

Platforms with an open strategy always improve the overall profit of the ecosystem.

Generally, the reason why a platform selects the open strategy is that it is believed that the merchants on the platform enjoy a production cost advantage. However, we assumed that eCommerce platforms and merchants had the same production costs but with different logistics costs and other differences. Meanwhile, the impact of different ecosystem structures on the overall profits was analyzed with the open or direct operation strategies. The equilibrium was refined by sequential subgames. Whatever strategy the eCommerce platform selected, the price of transfer payment and platform product price in the ecosystem were set by maximizing the profit, based on the production cost c. Therefore, under the profit distribution contract, whether the competitors selected the open strategy or not, the open strategy adopted by eCommerce platforms would help to improve the overall profit of the ecosystem where the platform was located, as well as the overall profit of the ecosystem where the competitive platform was located.

It has been shown from the research that the open strategy of an eCommerce platform improved the overall profit of the ecosystem where the platform was located. The increase in the overall profit of the system did not come from the production cost advantage of the platform merchants. It is likely that the open platform and ecosystem weakened the price competition between similar eCommerce platforms in the competitive market. By comparing the equilibrium prices with different strategy sets, due to the opening of the platform, price competition weakened. The equilibrium pricing p_A and p_B of products on the platforms was improved, which promoted the increase in the profits of the overall eCommerce ecosystem. Meanwhile, the internal transfer price w and profit distribution ratio η were negotiated with the platform merchants; as a result, the profit distribution contract became an important equilibrium coordination tool of the game, which was also conducive to realizing the competitive equilibrium of the eCommerce ecosystem and improving the overall profit of the ecosystem.

Proposition 2.

The profit distribution contract structure within the eCommerce ecosystem has no effect on the equilibrium of the competitive ecosystems.

There is no relationship between the strategies of the platforms and the competitive structure within each ecosystem. The profit distribution of the platforms and merchants did not affect the choice of the strategy by platforms; this indicates that the competition structure of the ecosystem was stable and balanced. Accordingly, the equilibrium resulted in the eCommerce platform changing with the profit distribution proportion of the internal contract in the eCommerce ecosystem. The result showed that pure strategic equilibrium (SO, SO) or (OP, OP) would be reached in most cases. Only when the platform cooperated with the merchants in the ecosystem and was not strictly superior or inferior would the game have a mixed strategy equilibrium (SO, SO) and (OP, OP). There is a trigger strategy that can be applied in which either platform could change strategy to obtain higher profits when the platform selected inconsistent strategies; either side changing the game strategy would lose its platform profits while in the state of equilibrium.

Proposition 3.

Differentiated competition between eCommerce platforms would not directly cause higher profits for platforms.

Generally, we believe that manufacturers can obtain more profits by adopting the differentiation strategy. Firms can conduct differentiated competition through innovative strategies such as logistics distribution, brand promotion, and product shopping experience [43]. However, the study also found that the overall profit function of eCommerce platforms and the ecosystem they constructed was not a strict increasing function of product differentiation under the internet economy; the transfer payment and profit distribution ratio of the internal contract in the ecosystem were the key factors impacting the profit of eCommerce platforms. ECommerce platforms internalized the network effect in the ecosystem and thus further reduced the product price and provided higher profits for eCommerce platforms. For eCommerce platforms, expanding product demand and network effect led to more intense price competition and a larger market effect, which brought more profits. When the network effect was greater than the competition effect, the product positioning of platforms would also deviate from the principle of maximizing differentiation.

The value of an eCommerce ecosystem relies on the network effect and the number of network nodes in the system under a network economy. The larger the market share of an eCommerce platform is, the greater the network effect is. Meanwhile, what promoted a positive feedback effect and improved the market share would be increased further. As a result, eCommerce platforms are devoted to expanding their market share, and they prefer to be less differentiated from their competitors in product positioning. The less differentiated the products are, the stronger the competition effect will be. Long-term competitive equilibrium is realized through the pricing mechanism of eCommerce platforms.

5. Discussion and Conclusions

5.1. Discussion

In this study, the Hotelling model was used to investigate the strategy selection of the two competing platforms from the perspective of ecosystems. In recent years, more and more attention has been paid to the study of the ecosystem. Scholars have studied the framework construction of ecosystems [44], the value creation of ecosystem [45], the interdependence of the main body of ecosystems [46], and the health of ecosystems [47]. There are more researches on competition and cooperation within ecosystems [34], but less on competition among platform ecosystems. In addition, most of the previous researches on platform competition focus on the core enterprises. However, the platform competition is not only related to the core enterprises, but also related to other participants in the ecosystem [47]. Therefore, this study considers the impact of platform strategy selection on the overall ecosystem returns.

Through the above analysis of game results, we find that the selection of open strategy will improve the overall profit of the eCommerce ecosystem and social welfare. Traditional industrial organization theory holds that companies can achieve higher profits through differentiated competition [25]. In this paper, we found that rational eCommerce platforms always choose open operation strategy, and different ecosystems tend to be homogeneous. In other words, differentiated competition did not necessarily lead to higher profits in an open and shared internet environment. We argued that the network effect among eCommerce platforms was larger than the competition effect, which made the competitive positioning of eCommerce platforms deviate from the principle of differentiation maximization. Therefore, the research complements the theoretical studies of platform externality in network organization economics. Then, this paper combined platform competition and business ecosystem theory, providing a new theoretical perspective for the study of competition between internet platforms, and we used the Hotelling game model to verify the homogeneity of different eCommerce platforms and their ecological strategy in the internet economic environment. That is, they will eventually move toward a hybrid strategy equilibrium, contradicting the traditional strategy of differentiated competition. Finally, the study also found that a certain monopoly power of the core enterprise in a competitive business ecosystem can determine strategy choice and improve Pareto efficiency, which may contribute to the research field on the boundary of an open economy and a business ecosystem.

5.2. Conclusions

This paper extended the analysis of the competition of eCommerce platforms from price competition and differentiated competition to strategic ecosphere building, transferring the analysis from micro firms to macrosystems. We analyzed the competitive eCommerce ecosystem equilibrium structure by constructing a sequential game of differentiated competition based on the Hotelling model and attempted to determine the influence of alternative game strategies on the profits of eCommerce platforms and ecosystems.

First, platforms selecting an open strategy decreased the transaction cost, which indicates that eCommerce platforms would choose the opening operation strategy, even though the suppliers would have a certain production cost disadvantage in the system; in addition, any eCommerce platform can improve the overall profit of the ecosystem and social welfare in the sharing economy. Second, the operation strategy of the platform determined the competitive equilibrium structure of ecosystems; an open strategy was conducive to improving the overall profit of ecosystem. However, platform profits may also be affected by profit redistribution within the platform-dominated ecosystem. Game strategies of eCommerce platforms would not depend on the profit distribution contract with merchants, that is, the proportion of profit distribution is independent of the equilibrium structure of competition within the ecosystem. Third, the equilibrium of the ecosystem with competitive eCommerce platforms was a hybrid strategy, which combined an open and self-operation strategy. In the process of building an eCommerce business ecosystem, the eCommerce platform gradually transitions from a pure strategy selection to a mixed strategy and would finally maximize the overall profit of the ecosystem. This also proves that the competition between platform ecosystems is not the same as the competition between platforms, while the equilibrium of competitive eCommerce systems deviates from the maximization difference of traditional platform competition studies. The conclusions of this paper contribute to the interdisciplinary field of competition between eCommerce platforms with contextual and theoretical extensions of the literature on business ecosystems.

5.3. Managerial Implications

The main conclusions of this paper also have some explanatory power and significant implications for the competitive strategy selection of eCommerce platform ecosystems.

First, this study provided ideas for the strategic selection of emerging platforms and the transformation of established platforms. The opening of eCommerce platforms and the building of ecological circles increases the number of products sold on platforms (SKUs), effectively meeting consumers’ diversified and personalized needs and improving the shopping experience of consumers. The eCommerce ecosystem can greatly improve the profit level of core platforms with monopoly power that would draw a certain proportion of contract distribution. Therefore, creating an ecosystem that shares logistics, information flow, capital flow, and alternative links with platform merchants is an area that eCommerce platforms are focusing on.

Second, the differentiation of eCommerce platforms is no longer the main direction of competition, and it is increasingly important to build an eCommerce ecosystem. Platforms should pay more attention to cooperation and strive to create an integrated, sharing, symbiotic and win–win ecosystem, so that eCommerce platforms can not only provide sharing services but also meet the common needs of merchants. In the case of China’s Taobao and JD, competition in the Internet environment is based on the implementation of platform ecosystem strategies.

Third, from the perspective of an ecosystem, eCommerce platforms should be granted certain discourse rights. The distribution ratio of contracts with merchants is the key for the platform to select an opening strategy. The choice of an open strategy not only improves the profits of the platform ecosystem but also helps to improve the overall profits of the competing platforms in the ecosystem. For eCommerce platforms with monopoly power, the market is sufficient to guarantee their revenue, but antimonopoly and industry regulations in the internet environment need to be strengthened.

5.4. Limitations and Future Research

First, the simplified Hotelling model in this paper had some limitations in the context of an ecosystem. The ecosystem discussed in our Hotelling model was a simplified supply-side ecosystem consisting only of platforms and merchants. However, an ecosystem with platforms as its core requires more association with system stakeholders, and competition and cooperation with network structures will be formed. In future research, we may consider more relevant stakeholders and investigate competition and cooperation in a more comprehensive business ecosystem. Meanwhile, social network theory considers social network systems to explain social behavior. Therefore, social network theory and methods from network dynamics research methods could help to explain competition between business ecosystems, and we may explore these theories to extend our research in the future.