Restrictions on stakeholders’ utilization of ecosystems and associated subsidies are commonly used to simultaneously enhance ecosystem conservation and poverty reduction. However, methodologies identifying adjustments to subsidy policies that produce more effective conservation approaches are lacking. In this study, we integrated cost–benefit analysis and land-use modeling to simulate changes in grassland use under different subsidy scenarios. Subsequently, we analyzed the effects of changes in subsidy policies on ecosystem services and household income using two new indices. The results showed that annually planted and mowed grassland was 21 times more economically profitable than natural grazed grassland, although its ecosystem services were of lower value. Fencing grassland resulted in economic and ecological profits, although their extent depended on the types of subsidies provided. Average capital-use efficiency under the interest-free loan scenario was considerably higher than that under the cash subsidy scenario, and thus the former was economically advantageous. From the perspective of ecosystem services benefits, capital efficiency was positive under the cash subsidy scenario but negative under the interest-free loan scenario. The results suggest that on the one hand, cash subsidies are effective in providing economic compensation for poor households, reducing poverty and supporting ecosystem conservation while releasing household labor for part-time urban employment. On the other hand, interest-free loans provide households with the initial capital necessary to commence forage production and subsequently provide employment. Therefore, simultaneously providing cash subsidies and interest-free loans could achieve a win–win outcome, whereby the ecosystem is conserved and household income is increased while limiting the government’s financial burden.