Dynamic value sharing based on employee contribution as a competitiveness-enhancing device

The knowledge economy emphasizes attracting, motivating, and retaining talent for sustainable value creation. This paper investigates the effectiveness of a dynamic value-sharing mechanism (DVS) compared to static value sharing (SVS). The current landscape reveals that while value sharing is increasingly adopted (e.g., 16% of Chinese A-share listed companies in 2020 implemented various forms of value sharing), its effectiveness is inconsistent. Some studies show no impact or even a negative relationship between employee stock options (ESOPs) and firm performance. This inconsistency often stems from static value-sharing mechanisms that fix the value-sharing proportion based on initial contributions, potentially leading to free-riding and weakened incentives. These mechanisms lack dynamism and fairness, failing to adapt to the changing contributions of employees over time. In contrast, Huawei's success is highlighted as an example of a successful DVS model, where value-sharing proportions are adjusted based on individual contributions. This paper aims to use evolutionary game theory to analyze the dynamic process of DVS, considering bounded rationality and learning mechanisms among employees, and to compare its efficacy with SVS in enhancing human capital, profit, and firm value. The study seeks to understand how DVS facilitates the transition to an optimal state and identify the contributing internal and external conditions.

Existing literature on value sharing presents mixed findings. While principal-agent theory and human capital theory suggest positive impacts on reducing agency costs and improving productivity, attracting and retaining high-capacity workers, and enhancing innovation and performance, empirical evidence is inconsistent. Studies on employee stock ownership plans (ESOPs) show variable results, with some indicating no clear effect or even negative relationships with productivity and financial performance, particularly due to free-riding issues in static systems. The literature largely focuses on static value-sharing mechanisms, neglecting the dynamic nature of employee contributions. While studies examine value sharing based on contribution, they often assume constant contributions and perfect rationality among participants. This paper addresses these gaps by considering the dynamic adjustments of value-sharing proportions according to changing contributions and incorporating bounded rationality and learning mechanisms into the model.

The study uses a two-stage game-theoretic model. The first stage employs the Hotelling model to represent competition in the consumer market between firms with and without value sharing (VS and NVS, respectively). This model considers factors such as product base value, price, innovation utility brought about by human capital transformation, and consumer mismatch costs. The equilibrium profit for both types of firms is derived, revealing its dependence on the total amount of human capital, setting the stage for the second-stage game. The second stage utilizes an evolutionary game model to analyze competition in the talent market between firms (VS and NVS) and employees (high-capability HE and low-capability LE). The model incorporates parameters representing incentive capacity (r), transformation capacity (λ), profit retention ratio (θ), value-sharing ratio (δ), and wages (w). The authors consider two distinct value-sharing mechanisms: static value sharing (SVS), where the value-sharing proportion is fixed, and dynamic value sharing (DVS), where it adjusts based on contributions. The replicator dynamic equations are derived to analyze the evolution of the proportion of firms using VS and the proportion of high-capability employees in the market. The analysis involves finding evolutionarily stable strategies (ESS) and exploring how different parameter values affect the evolution of the system. The study employs numerical simulations using parameters based on Huawei's historical data from 2008 to 2012 (when they used a system similar to SVS) to verify and validate the models. Sensitivity analysis is conducted to examine the effects of initial proportion of high-capability employees, incentive capacity, transformation capacity, and overall profit-sharing ratio on the speed and outcome of system evolution.

The simulation results reveal several key findings. First, under SVS, there is no evolutionarily stable strategy (ESS), indicating instability in the proportion of firms using SVS and the proportion of high-capability employees. This is consistent with the observed brain drain experienced by Huawei during their period of Saturated Rights Issues (similar to SVS), where static value-sharing proportions led to apathy and high turnover among talented employees. Second, under DVS, the system evolves toward an ESS where high-capability employees dominate value sharing. The proportion of high-capability employees increases over time and stabilizes at a higher level compared to SVS. This demonstrates the effectiveness of DVS in attracting and retaining high-performing employees. Third, the profit and value of firms implementing DVS are significantly higher than those implementing SVS or NVS, and these outcomes are more stable under DVS. This highlights the positive effect of DVS on firm growth. Fourth, sensitivity analysis shows that a higher initial proportion of high-capability employees accelerates the evolution of the system towards the ESS and increases the speed of profit maximization. Similarly, a higher incentive capacity (r) and, to a certain extent, a higher transformation capacity (λ) lead to faster evolution and higher profits. However, excessively high transformation capacity can delay the achievement of equilibrium. The effect of the overall profit-sharing ratio (1-θ) on evolution is non-monotonic; moderately increasing this ratio initially speeds up evolution but excessively high ratios delay it. This suggests a need for balance between employee rewards and firm reinvestment.

The findings address the research question by demonstrating the superiority of DVS over SVS in terms of stability, attracting and retaining talent, and enhancing firm performance. The results show that dynamically adjusting value-sharing proportions based on employee contribution effectively counters free-riding and motivates higher performance. The success of Huawei after shifting to a DVS-like system supports the model's findings. The study's contribution lies in its integration of evolutionary game theory and the consideration of bounded rationality, providing a more realistic portrayal of human behavior and firm dynamics. The findings are relevant to the field of organizational behavior, human resource management, and strategic management, offering valuable insights into the design and implementation of effective compensation systems in the knowledge economy.

This study contributes to the ongoing debate on value sharing by demonstrating the benefits of dynamic value sharing (DVS) over static value sharing (SVS). DVS fosters a stable evolution towards an equilibrium where high-capability employees dominate, leading to higher and more stable firm profit and value. The study highlights the importance of considering the dynamic nature of employee contributions and the role of bounded rationality in the design of effective value-sharing mechanisms. Future research could explore the impact of other social preferences and risk preferences, investigate specific industry differences in the effectiveness of DVS, and conduct large-scale empirical studies to validate the model's findings further.

The study utilizes a simplified Hotelling model for consumer market competition, assuming full market coverage and not accounting for the complexities of real-world market dynamics. It also does not incorporate the impact of other social preferences (e.g., altruism, reciprocity) or risk preferences (e.g., risk aversion) on employee decision-making. The simulation uses data from Huawei, limiting the generalizability to other firms. Future research should address these limitations by incorporating a broader set of preferences and testing the model across various industries and company sizes.