China's massive rural-urban migration, totaling 376 million in 2020, presents significant housing challenges. High housing prices and low wages create a gap that government housing security programs fail to bridge for migrant workers, many of whom lack urban household registration (*hukou*). This has led to a large reliance on low-cost housing options like enterprise dormitories and urban villages. This paper specifically examines the role of enterprise dormitories, which often serve as both living and working spaces, in influencing migrant workers' urban integration. The research hypothesizes that enterprise dormitories serve a cost-cutting purpose for businesses in the initial individual migration stage, while hindering family reunification in the subsequent family migration stage due to space constraints. This study aims to explore this dual role of enterprise dormitories in shaping the urban integration process of migrant workers in China, thereby enriching existing literature and contributing to the optimization of housing policies.

Existing research on China's internal migration emphasizes the impact of the *hukou* system on migrant workers' access to urban resources and their housing choices. Studies have shown that many migrant workers reside in sub-standard housing, including rented houses, enterprise dormitories, and urban villages. While several studies explore the urban integration of migrant workers, the influence of living patterns, specifically enterprise dormitories, on this integration remains under-researched. Previous works often consider living arrangements as a consequence of urban integration rather than a factor influencing it. Research focusing on enterprise dormitories tends to portray them negatively, neglecting their potential positive roles, especially in the initial stages of migration. This study addresses these gaps by examining the effects of enterprise dormitories on urban integration across different migration phases.

The study uses two datasets: 1) survey data from 2010 on migrant workers in the Pearl River Delta and Yangtze River Delta regions, allowing for matching of enterprise and employee data; and 2) national-level data from the 2017 China Migrants Dynamic Survey (CMDS), providing broader geographical coverage and information on family characteristics. The key variable is whether the migrant worker lives in an enterprise dormitory. To examine Hypothesis 1 (enterprise dormitories as a means of cost reduction), a logit model is used for the dichotomous variable 'wage system' (hourly vs. basic salary), and OLS regression for continuous variables like working hours, hourly wage, and monthly wage. To test Hypothesis 2 (the differential impact of dormitories across migration stages), a logit model is used for dichotomous dependent variables: whether the migrant worker chose an enterprise dormitory, spouse reunification, and family reunification. Independent variables include marital status, initial employment status, and job stability. The CMDS data is used for robustness checks employing both logit and linear probability models.

The analysis supports Hypothesis 1, demonstrating that enterprise dormitories are indeed linked to cost-reduction strategies. Enterprises providing dormitories are less likely to use hourly wage systems, their employees work longer hours, and their hourly wages are significantly lower compared to those in enterprises without dormitories. However, there is no significant difference in monthly wages. Hypothesis 2 is also supported. In the individual migration stage, unmarried migrant workers and those new to urban employment are more likely to live in enterprise dormitories. Job stability is negatively correlated with choosing an enterprise dormitory, suggesting that workers aiming for long-term employment are less likely to choose this option. Conversely, in the family migration stage, residing in an enterprise dormitory significantly reduces the likelihood of spouse and family reunification. The robustness test using the CMDS data corroborates these findings. Specific statistical results (regression coefficients, p-values) are detailed in tables within the original paper.

The findings highlight the dual nature of enterprise dormitories. While they function as low-cost initial accommodation aiding economic integration for single migrants, their inadequacy in providing sufficient family living space hinders family reunification and urban integration at a later stage. This reveals a trade-off between cost-effective adaptation and family life. The study's findings contribute to a deeper understanding of China's low-cost urbanisation model, challenging the simplistic narrative of dormitories as solely negative spaces. The results suggest that addressing the housing needs of migrant workers, especially those aiming for family reunification, requires strategies beyond relying solely on enterprise-provided dormitories.

This study contributes to the understanding of enterprise dormitories' impact on the urban integration of migrant workers in China, revealing their dual role across different migration stages. For policy implications, increasing the supply of public housing that considers job-housing balance is crucial for addressing family reunification needs. Future research should examine children's development in the context of these various living arrangements and the impact of recent changes in China's labor and housing markets.

The study uses two datasets from different years (2010 and 2017), potentially affecting the direct comparability of findings. The 2010 dataset is geographically limited to the Pearl River Delta and Yangtze River Delta regions, while the CMDS data, while nationally representative, lacks detailed enterprise-level information. The absence of data post-2018 due to the COVID-19 pandemic prevents an analysis of recent market changes. The study focuses on China's context and may not fully generalize to other countries with different migration patterns.