Learning Long-Term Educational Investment Policies under Residential Sorting

2026-08-07

AI Summary

This study addresses how residential segregation and school investment drive up housing prices, exacerbate student stratification, and constrain educational access for low-income households, while existing approaches struggle to capture the long-term dynamic interplay among education, housing, and population mobility. To this end, the paper introduces a novel dynamic multi-agent framework that integrates government investment decisions, household location choices, housing price dynamics, demographic turnover, and school quality evolution. By employing reinforcement learning to optimize multi-year educational investment strategies, the model explicitly captures the feedback loop between education and housing markets. Simulation results demonstrate that the proposed strategy achieves the highest enrollment rate (0.4780) while maintaining a low enrollment Gini coefficient (0.0164), effectively balancing efficiency and equity and substantially mitigating the stratifying impact of socioeconomic status on educational opportunity.

Abstract

Allocating public-school investment effectively and fairly is difficult when school access depends on residence. School improvements can raise nearby housing demand and prices, reshape enrollment, and potentially limit access for lower-income households. These effects evolve as residential sorting changes school composition, quality, and future investment needs. Existing approaches often study school funding, household choice, and housing markets separately, while static models can miss their interconnected, long-term effects. We address this gap with a dynamic multi-agent framework that links government investment, household sorting, housing prices, population turnover, enrollment, and evolving school quality. A government planner uses reinforcement learning (RL) to identify multiyear allocation policies that account for household responses while balancing aggregate educational access and equity. In simulations, our RL-based policy attains the highest access level (0.4780) and second-lowest access Gini coefficient (0.0164) among representative baselines, demonstrating a favorable effectiveness-equity balance. The results also indicate reduced socioeconomic stratification in educational access. By making education-housing feedback explicit, our framework supports long-term analysis of how school investment shapes educational opportunity over time.

Problem

• educational investment
• residential sorting
• school access
• housing market
• equity

Innovation

• reinforcement learning
• dynamic multi-agent framework
• residential sorting
• educational equity
• housing-education feedback