The impact of capitalist profit-seeking behavior by online food delivery platforms on food safety risks and government regulation strategies

The paper examines how capitalist profit-seeking by online food delivery platforms affects food safety risks and government regulation strategies within the rapidly expanding platform economy driven by Internet technologies. With online food delivery becoming a necessity—especially during COVID-19—platforms like Meituan and Ele.me have achieved dominant positions through data-driven monopolistic advantages, intensifying information asymmetry and complicating food safety regulation. The research aims to model the strategic interactions between platforms and restaurants—focusing on commission and promotion fee strategies—and to analyze how government regulation influences restaurant behavior and overall food safety outcomes. The study further validates model insights using Chinese market data, arguing for universal implications given similar platform behaviors globally.

Prior work highlights the platform economy’s efficiency gains through data-driven multi-sided markets but also underscores heightened information asymmetry, especially salient for food products with credence attributes where consumers cannot easily verify safety. Platforms can leverage dominant positions and data control to extract profits and impose costs on restaurants, potentially reducing quality and fostering a 'market for lemons.' Studies note that commissions and additional burdens (e.g., refunds) can harm restaurant profitability, encouraging cost-cutting via ghost kitchens that may elevate food safety risks. Traditional regulatory approaches struggle amid fragmentation and real-time challenges; proposed solutions include platform self-regulation, co-regulatory models, information sharing, and government intervention. However, much of the literature is descriptive; quantitative modeling of food safety risks under platform profit-seeking and the platform–restaurant interest game is limited. This study addresses these gaps using an evolutionary game framework centered on Meituan.

The study builds an evolutionary game model of the platform–restaurant interaction grounded in principal–agent theory. It identifies a four-tier principal–agent structure among consumers, platforms, restaurants, and government, emphasizing information asymmetry and moral hazard. The model focuses on the platform (strategies: low vs. high commission) and restaurants (strategies: safe vs. illegal production), with key parameters: commission rate β, promotion fee M, restaurant revenues without/with promotion S0/S1, platform operating cost C0, restaurant costs for safe/illegal production C1/C2 (C1 > C2), probability α that restaurants adopt promotion under low commission, and government detection probability ω (0 ≤ ω ≤ 1). Penalties for detected illegal production differ by promotion status: platform losses F0 (no promotion) and F1 (with promotion), restaurant losses F3 (no promotion) and F2 (with promotion), with F1 > F0 and F2 > F3. A payoff matrix is specified for platform and restaurant combinations (commission level × production strategy), incorporating promotion adoption probabilities under low commissions and zero promotion under high commissions. Expected payoffs for the platform under low/high commissions (U11, U12) and for restaurants under safe/illegal production (U21, U22) are derived, leading to average payoffs U1 and U2. Replicator dynamics describe strategy evolution: - ẋ = x(1−x)[y(αω(F1−F0)) + αβ(δ1−δ0) + αM − αω(F1−F0)] (as presented, capturing differences via promotion and penalties) - ẏ = y(1−y)[x(αω(F2−F3)) + (C2 − C1) + ωF3] Setting ẋ = 0 and ẏ = 0 yields five equilibrium points: (0,0), (0,1), (1,0), (1,1), and the interior (x0, y0) with x0 = (C2 − C1 − ωF3)/[αω(F2 − F3)] and y0 = [αω(F1 − F0) − αβ(δ1 − δ0) − αM]/[αω(F1 − F0)]. Stability is analyzed via the Jacobian A with entries a, b, c, d; ESS requires |A| > 0 and trace D < 0. The interior point does not satisfy ESS (D = 0). Four parameter-regime-dependent ESS outcomes are analyzed, mapping government regulation strength (ω) and promotion fee level (M) to platform and restaurant strategies. Model insights are validated using Chinese market data (2017–2021) for market size, Meituan share, and complaint volumes, including estimated complaint volumes attributable to Meituan’s online food delivery activities.

- Across all stable equilibria, the platform remains profit-seeking, selecting strategies (high commissions or, alternatively, low commissions with high promotion fees) to maximize overall profit. - Government regulation shifts restaurant behavior but not the platform’s profit-seeking stance: • Situation (1) Moderate regulation, low promotion fees: ESS (platform high commission; restaurants illegal production). • Situation (2) Strong regulation, low promotion fees: ESS (platform high commission; restaurants safe production). • Situation (3) Weak regulation, high promotion fees: ESS (platform low commission; restaurants illegal production). • Situation (4) Very strong regulation, high promotion fees: ESS (platform low commission; restaurants safe production). - Higher promotion fees increase restaurants’ loss aversion, raising incentives for illegal production unless regulation is very strong; thus, a high share of promotion fees in platform revenue necessitates stronger regulation to achieve safe production. - Empirical indicators from China support the model’s implications: • Online food delivery users in China reached 544 million in 2021; market size rose 61.62% from 2019 to 2021. • Meituan’s market share of online food delivery transactions increased from 62.42% (2017) to 75.17% (2021). • National restaurant industry complaints grew from 8,595 (2012) to 37,204 (2021), average annual growth 36.98% (higher than many other sectors). Estimated Meituan complaint volumes increased from 222 (2018 Q3) to 1,193 (2021 Q4), broadly tracking revenue growth, suggesting a positive relation between platform profit expansion and food safety risk signals (complaints). - Regulatory focus solely on restaurants is insufficient; monopolistic platforms can pass through regulation-induced costs to restaurants without materially reducing their own payoff. Adjusting the commission–promotion fee mix can reduce government regulation costs while maintaining platform overall payoff.

The evolutionary game results address the research question by showing that monopolized platforms, driven by profit maximization, set pricing and promotion structures that pressure restaurants and can elevate food safety risks, especially under weak or moderate regulation. Stronger regulation can pivot restaurants from illegal to safe production, but does not alter the platform’s profit-seeking strategy. Therefore, effective policy must directly target platform incentives and structures—particularly the balance between commissions and promotion fees—to mitigate restaurants’ incentives for opportunism. Empirical trends in China (market growth, Meituan dominance, rising complaint volumes) are consistent with model predictions of heightened risk under aggressive profit extraction. The findings suggest that co-regulation and platform-focused interventions can reduce regulatory costs and improve compliance outcomes without diminishing platform profitability, aligning market mechanisms with public health goals.

The paper contributes a quantitative evolutionary game framework linking platform profit strategies, restaurant production choices, and government regulation to food safety risks in online food delivery. It shows that capital-monopolized platforms persistently pursue profit maximization, often squeezing restaurant margins and heightening incentives for illegal production unless regulation is sufficiently strong. Policy implications for China include: respecting market roles while curbing disorderly platform capital expansion; instituting a negative list to ensure fair competition; enhancing food safety risk management and co-governance (government, platforms, industry associations, consumers); leveraging platform technologies to reduce regulatory costs; and optimizing the commission–promotion fee mix to compel safe production without reducing platform overall payoff or raising consumer costs. Future research should test the model with data from other countries and platforms, explore alternative regulatory regimes (e.g., scientifically informed laissez-faire), and develop coordinated regulation frameworks among government, platforms, and restaurants.

The study primarily focuses on China and the Meituan platform, which may limit generalizability despite argued universal tendencies of platform capital. The evolutionary model relies on parameter assumptions (e.g., promotion adoption probability, penalty differentials, detection probability), and the interior equilibrium is not an ESS, narrowing dynamic outcomes discussed. Empirical validation uses estimated complaint volumes derived from aggregate industry data and market shares, which may introduce approximation errors. Causality between platform profit-seeking and food safety risks is inferred from patterns rather than established via causal identification.