Quantitative evidence on the impact of social protection such as livelihood programs (LPs) on reducing vulnerability to poverty (VtP) is scant. Therefore, this paper aims to estimate VtP and the ability of LPs to reduce it. The analysis is based on the primary data collected in 2018–19 from 479 households in a tribal region of eastern India. First, household VtP was estimated using the feasible generalized least squares (FGLS) method. Households were categorized into four groups based on their current and predicted future poverty status: chronic poor, transient poor, escaped poverty, and non-poor. Second, factors influencing this movement of households from current to future poverty states were determined using multinomial logistic regression. Third, the impact of the LP on VtP was investigated using the endogenous switching regression (ESR) approach. The FGLS estimates show that 34.7% of households are vulnerable to future poverty, which exceeds the actual poverty rate of 28.6%. The major factors responsible for exacerbating chronic poverty and causing transient poverty are chronic illness, climate shocks, and the sale of assets. The ESR results indicate that a household’s VtP is reduced by 3% for those that participate in LPs. Although the LP evaluated in this study has an average positive impact in reducing vulnerability, there is wide heterogeneity among the beneficiaries as the geographically disadvantaged communities lag behind the others. Therefore, the study suggests strengthening such LPs with a particular focus on disadvantaged and vulnerable communities.