Agroforestry has been practiced for decades and is undoubtedly an important source of income for Indonesian households living near forests. However, there are still many cases of poverty among farmers due to a lack of ability to adopt advanced technology. This literature review aims to identify the characteristics and factors causing the occurrence of agricultural subsistence and analyze its implications for the level of farmer welfare and the regional forestry industry. The literature analysis conducted reveals that small land tenure, low literacy rates, and lack of forest maintenance are the main causes of the subsistence of small agroforestry farmers. Another reason is that subsistence-oriented agroforestry practices are considered a strong form of smallholder resilience. All of these limitations have implications for low land productivity and high-sawn timber waste from community forests. To reduce the subsistence level of farmers, government intervention is needed, especially in providing managerial assistance packages, capital assistance, and the marketing of forest products. Various agroforestry technologies are available but have not been implemented consistently by farmers. Therefore, it is necessary to develop an integrated collaboration between researchers, farmers, and regionally owned enterprises (BUMD) to increase access to technology and markets. Although it is still difficult to realize, forest services, such as upstream–downstream compensation and carbon capture, have the potential to increase farmer income.

Limited agricultural land areas combined with increasing demands for food require breakthroughs in land use development using agroforestry systems. Intercropping root crops with trees could be an alternative for food production in forest areas. This study aimed to assess the feasibility of cocoyam (Xanthosoma sagittifolium (L.) Schott) farming on dry land within 12- and 42-year-old teak (Tectona grandis L.f.) forests to support local food security. The feasibility assessment took into account both productivity and socio-economic aspects. The agroforestry land productivity was measured using the land equivalent ratio (LER), and our analysis of the cocoyam farming within the teak stands was carried out using the revenue/cost ratio (R/C) at the demonstration plot scale. Furthermore, we also surveyed farmers’ perceptions of the production of cocoyam for food security. The results showed that the R/C values of cocoyam tuber production in agroforestry systems were lower than 1. However, the production rates of cocoyam tubers in the 12-year-old teak stand (48.3% light intensity) and the 42-year-old teak stand (62.5% light intensity) were 2.64 and 2.76 tons/ha, respectively. The overall yields from the teak and cocoyam agroforestry systems were more profitable than those of the monoculture system, as indicated by the LER values of 1.61 and 1.85. Cocoyam production was socially acceptable (77% of respondents) as a smallholder subsistence agroforestry practice to meet food demand. Increasing cocoyam productivity in teak forests requires the adoption of agroforestry silvicultural technology to achieve food security for rural communities. To increase their farming production and income, farmers could apply intensive silvicultural practices. Governmental support that could be provided includes encouraging product diversification and providing assistance for the processing and marketing of cocoyam products.