1. Introduction

The global platform economy exceeds $7 trillion, yet the gap in bargaining power between labor and capital is widening (OECD 2024) [1]. Existing research focuses on monopoly pricing [2], but pays insufficient attention to the dynamic mechanisms of bargaining power transfer. The bargaining power of workers and merchants continues to be weakened, with frequent occurrences such as algorithms controlling delivery riders' pay and merchants passively accepting platform terms. This paper addresses the core question: "Where does the diminished bargaining power go in the platform economy?" aiming to clarify the mechanisms and impacts of its erosion.

2. Three mechanisms weakening bargaining power in the platform economy

According to relevant research and data, traditional bargaining power is reflected in the bilateral negotiation capability based on supply and demand elasticity [3]. In the platform economy, this bargaining power has been altered, specifically manifested through its quantification by digital metrics like algorithm scores and order volume, leading to a gradual loss of voice for laborers [4]. This weakened state of bargaining power is corroborated by specific data: the order rejection rate for Chinese food delivery riders is below 3% [5], nd ride-hailing drivers' incomes have decreased by 30% [6].

2.1. Network effects strengthen platform monopoly: from "convenient choice" to "no alternative"

Network effects are a core feature of the platform economy – the more users, the stronger the attraction for both supply and demand sides, creating a "winner-takes-all" monopoly landscape.

For workers: Leading platforms control the vast majority of market orders. Didi holds over 83% of China's ride-hailing market, and Meituan holds a 70% share in food delivery. If individuals reject the commission rates or work rules set by the platform, such as ride-hailing platforms increasing drivers' commissions from 15% to 25%, they face the dilemma of "exit means unemployment".

For SMEs: Platforms consolidate their monopoly position through traffic tilting, such as Taobao's "Zhitong" bidding ranking system. Merchants who do not pay high promotion fees are pushed out of users' sight. This "traffic monopoly" forces merchants to accept the platform's commission rates (some e-commerce platforms take 20%-30%), losing their ability to negotiate transaction terms. As noted by Rochet& Tirole's [7] in their "two-sided market theory," after locking in both supply and demand sides to form a monopoly, bargaining power tilts unidirectionally towards the platform, ultimately evolving into a pattern of "platform sets prices, individuals passively accept".

2.2. Unilateral rule-making

Platforms master the power to interpret and enforce rules through establishing "unilateral rules," while the opacity of algorithms and the compulsory nature of rating systems further exacerbate the imbalance.

Oppressive Algorithmic Black Box: Platforms use algorithms to set core transaction parameters, such as delivery times and dispatch priority, without disclosing the logic. For example, a 2023 investigation by Renwu Magazine [8] revealed that food delivery platforms compressed average delivery times per order from 45 minutes to 30 minutes via algorithms, even implementing extreme rules like "3-minute overtime deductions." Riders, unable to "negotiate" the reasonableness of time with the algorithm, resort to violations like running red lights or speeding to meet targets, essentially losing their ability to bargain over working conditions.

Controlling Rating Systems: Platforms directly link user ratings to individual rights, creating "rating hostage situations." For instance, Didi drivers with scores below 4.8 receive fewer orders, and food delivery riders face fines or even account suspension due to "bad reviews." Under this rule design, individuals must unconditionally satisfy user demands, even unreasonable ones, to maintain their ratings, losing the possibility of negotiating the boundaries of their rights with users or the platform. These rules appear "neutral" but are entirely formulated and revised unilaterally by the platform (e.g., Meituan adjusted rider reward and penalty rules multiple times in 2024 without consulting riders), reducing individual participants to passive rule acceptors.

2.3. Information and technological barriers

Platforms leverage their data advantage and technological tools to construct information and technological barriers that are difficult for individuals to breach, further weakening their bargaining power.

Overwhelming Data Asymmetry: Platforms possess real-time comprehensive supply and demand data (e.g., regional order density, user price sensitivity, worker online duration) and can maximize their own profits through dynamic pricing (e.g., ride-hailing "surge pricing," food delivery "peak scheduling fees"). Individual participants only have access to fragmented information (e.g., drivers cannot see overall order distribution, merchants are unaware of platform traffic allocation logic), making it difficult to judge the reasonableness of transaction terms and even harder to bargain effectively.

Monopoly on Technological Tools: Platforms control transaction processes through algorithmic tools (e.g., automatic dispatch systems, smart pricing models), while individuals lack counter-tools. For example, SMEs cannot use big data analytics like platforms to optimize pricing or inventory and must passively accept the platform's "recommended prices." Ride-hailing drivers struggle to decipher the priority rules of dispatch algorithms and can only strive for orders through passive means like "staying online for long hours," losing the possibility of using technological means to enhance their bargaining position.

3. The triple detriment of bargaining power imbalance

The severe imbalance in bargaining power between platforms and individuals is not an isolated issue of interest distribution but triggers systemic risks along the transmission chain of "micro individual – meso market – macro society," causing deep impacts on the economic ecosystem and social fairness.

3.1. Micro level: dual erosion of labor dignity and rights

To avoid deductions for late delivery (platforms penalize 20% of earnings for being 10 minutes late), food delivery riders have a 40% higher traffic accident rate compared to traditional industries [9], but the success rate for safeguarding their rights is less than 15% [10]. Confronted with platforms' "dynamic commissions" (reaching up to 30% during peak hours), ride-hailing drivers can only sustain their income by extending working hours (logging over 12 hours online daily), trapping themselves in a passive predicament of "trading time for survival". Shrinkage of SME Rights: E-commerce platforms transfer operational risks to merchants through rules like "refund without return" and "penalties for false advertising". For instance, Pinduoduo's 2023 "0-second refund" policy resulted in a 58% surge in merchant disputes, while the platform's response rate to merchant appeals stood at less than 30% (. Lacking bargaining power, merchants passively bear the losses. This imbalance essentially negates "labor value" – individuals' time, safety, and operational investment must yield to the platform's efficiency metrics and profit demands.

3.2. Meso level: adverse selection and innovation suppression in the market ecosystem

The concentration of bargaining power towards platforms distorts market competition logic, leading to a vicious cycle of "bad money driving out good money":

Low-price Competition Squeezes Quality Supply: Platforms force cost compression through "subsidy wars" and "low-price traffic generation"(e.g., food delivery platforms force merchants to participate in"spend 20 minus 15"promotions). A 2024 survey indicated that 62% of food and beverage (F&B) SMEs reduced investments in ingredient quality due to platforms' low-price demands, while 38% of high-quality artisan merchants withdrew from the market as they could not afford traffic costs.

Systemic Weakening of Innovation Incentives: Platforms restrict operators' channel choices through terms like "pick one of two" and "exclusive cooperation" (e.g., e-commerce platforms requiring merchants not to join competing platforms). This monopoly forces SMEs to devote over 70% of their energy to coping with platform rules rather than product innovation and service upgrade, which crowds out resources allocated to public services such as education and healthcare.

3.3. Macro level: soaring social governance costs and erosion of trust

Excessive Consumption of Regulatory Resources: Platform disputes have become a major area of consumer rights protection – 68% of complaints accepted by the National Consumer Association in 2024 involved platform rule disputes (e.g., delivery responsibility disputes between riders and platforms, commission disputes between merchants and platforms). To address these conflicts, local governments must establish "Platform Economy Mediation Committees," with related administrative expenses growing annually by 23% [11], crowding out resources for public services like education and healthcare.

Continuous Depletion of Social Trust: Platforms utilize information advantages for "algorithmic discrimination" (e.g., hiding high-income orders from new drivers, pushing high-priced products to elderly users), causing public trust in the digital economy to drop from 75% in 2020 to 48% in 2024 [12]. This trust depletion not only constrains the platform economy's development but also weakens societal tolerance for technological progress, increasing resistance to institutional reforms.

4. Collaborative pathways for reconstructing bargaining power

The imbalance in bargaining power stems from multiple defects in technology, institutions, and governance. Its reconstruction requires multi-dimensional collaborative efforts through a triple path of "technical correction – institutional guarantee – pluralistic co-governance" to break the platform's unilateral monopoly on bargaining power and achieve a dynamic balance of power redistribution.

4.1. Technical governance correction: using transparency to break algorithmic hegemony

Algorithmic Transparency and Explainability: The EU's Digital Services Act (DSA) [13] requires platforms to disclose the core logic of recommendation algorithms (e.g., food delivery platforms must explain delivery time calculations, ride-hailing platforms must publish surge pricing rules), allowing individual participants to understand the "technical boundaries" of bargaining power.

Data Empowerment and Tool Sharing: T The United Kingdom's (UK) "Digital Workers' Union" has collaborated with technical teams to develop order optimization tools for riders. By aggregating real-time data from riders, these tools calculate optimal routes for accepting orders and reasonable delivery times, helping riders counteract the oppressive algorithms used by platforms. In some regions of China, pilots of "Merchant Data Midpoints" are underway, which provide SMEs with open, desensitized data—including industry average pricing and user preferences—to alleviate information asymmetry.

4.2. Institutional design improvement: using legal rigidity to define power boundaries

Clarifying Platform Obligations and Individual Rights: Article 35 of China’s newly revised E-Commerce Law explicitly prohibits platforms from exploiting their dominant market position to coerce merchants into "pick-one-from-two" exclusivity arrangements, thereby granting merchants the bargaining power to independently choose their business channels. California's AB5 bill [14] classifies gig workers as employees, requiring platforms to provide protections such as minimum wage guarantees and health insurance coverage, legally enhancing workers' bargaining position vis-à-vis platforms.

Establishing Bargaining Power Remediation Mechanisms: Germany's Platform Economy Ordinance stipulates that platforms must consult with worker and merchant representatives before modifying rules, and changes require regulatory approval. Many regions in China have established "Platform Dispute Arbitration Committees" to simplify rights protection procedures for riders and merchants, improving remediation efficiency after bargaining power is harmed.

4.3. Pluralistic collaborative governance: using collective action to consolidate bargaining power

Industry Self-Governance and Organized Bargaining: The Korean Delivery Workers' Union forced a platform to reduce delivery commissions from 20% to 15% and promise not to unilaterally compress delivery times through "collective order refusal" actions. Chinese e-commerce merchants formed an "Anti-Unfair Competition Alliance" to collectively resist unreasonable platform commissions and traffic monopolies, creating a countervailing negotiating position).

Social Co-governance and Multi-stakeholder Coordination: Germany has established a "Platform Economy Tripartite Consultation Committee" composed of government representatives, platform enterprises, and worker/merchant representatives to regularly negotiate issues like commission rates and working conditions, integrating bargaining power dynamics into an institutionalized communication framework. The EU established a "Digital Markets Supervisor," collaborating with consumer organizations and industry associations to monitor platform rules and promptly correct actions that harm bargaining power.

5. Conclusion

In the platform economy, bargaining power lost by both laborers and merchants has become concentrated within platforms. Leveraging network effects to establish monopolies (e.g., Didi dominates over 83% of the ride-hailing market), imposing unilateral rules (such as algorithm-driven compression of delivery times), and erecting technological barriers, platforms significantly undermine the bargaining power of other actors. This leads to issues including erosion of micro-level rights, distortion of meso-level market structures, and a surge in macro-level governance costs. Addressing these challenges requires rebalancing through technological remediation, institutional safeguards, and polycentric governance.

Limitations of current research include narrow data coverage (e.g., lack of samples from small and medium-sized platforms), insufficient dynamic tracking, and superficial analysis of mechanistic interactions. Future efforts should expand the scope of empirical investigation, establish quarterly dynamic monitoring mechanisms, and quantitatively assess interactive effects among institutional frameworks. Further research may explore bargaining power disparities across demographic segments, impacts of emerging technologies, adaptability of transnational policies, or develop dynamic game-theoretic models to forecast optimal rebalancing pathways.