关键词: 政府治理, 在线劳动, 算法管理

Abstract: Under the trend of deep integration of gig economy and digital technology, online labor platforms, which provide services on demand through precise, efficient, and large-scale operation with algorithms as the core means, have emerged, and gradually formed an “Algorithmic Management” (AM) model. The automatic and efficient, data-driven characteristics of AM have significantly improved the operational efficiency of the platform. However, while optimizing efficiency through algorithms, online labor platforms often undermine the autonomy and rights protection of workers, triggering a contradiction between the concept of flexible labor and algorithmic control.
Relevant studies have mainly analyzed the characteristics and impacts of algorithmic management. However, few have discussed how to govern the “paradox” from the government’s perspective. To answer the question of how governments can solve the online labor “paradox” by designing and combining different governance policies, this paper establishes an evolutionary game model between online labor platforms and workers based on two governance scenarios: regulation and incentives. Through model analysis, the governance effects of regulatory and incentive policies on platforms and workers’strategy selection are compared. Additionally, through numerical simulations, we analyze the evolution paths of platforms and workers under different model parameter settings to further indicate the effectiveness of regulatory and incentive policies.
The study finds that government regulation alone cannot solve the “paradox” and inappropriate regulation intensity will make the game evolve toward lower overall welfare. In contrast, a synergistic mechanism combining regulation with incentive policies is more effective which indicates that an appropriate level of incentives is needed to lead the game to evolve to a stable point with higher overall welfare. Furthermore,the simulation results show that a well-combined governance mechanism of both regulation and incentives is significantly better than the effect of only adopting one of the two governance measures. Finally,the simulation results also indicate that the extent of how well online labor platforms can effectively manage the arbitrage behaviors of workers will influence the final state of evolutionary system.
The conclusions of this paper suggest that the government should (1)keep regulatory intensity in an appropriate range to avoid excessive regulatory pressure on platforms, (2)provide differentiated support for platforms that adopt flexible algorithmic management and (3)actively integrate regulatory and incentive governance mechanisms to build a multidimensional and complementary governance matrix to promote the platform economy’s standardized, healthy, and sustainable development and effectively protect the rights and interests of workers in this new employment pattern. Finally, the platform should actively promote management model innovation and technology iteration, explore new management methods, and promote “algorithms for good” while optimizing algorithm efficiency. This paper expands the research in the field of algorithmic management and provides some theoretical support and decision-making basis for the government to carry out governance over online labor platforms.
The next step of this research is, firstly, to broaden the government governance scenarios; secondly, to further refine the government governance behaviors; and thirdly, to deepen the assessment of the effect of government governance by combining quantitative methodologies to carry out more empirical research.

Key words: government governance, online labor platform, algorithmic management