Cocoa is key in Liberia's employment, wealth creation, and foreign exchange earnings. However, smallholder cocoa farmers face significant productivity challenges. This study evaluated total factor productivity (TFP) determinants among smallholder cocoa farmers in Nimba County, Liberia. Primary data from 331 farmers was collected using a stratified random sampling method using a semi-structured questionnaire. The non-parametric Fare-Primont superlative index method was used to compute farmers’ TFP. In contrast, Data Envelopment Analysis (DEA) was used to decompose TFP into different efficiency measures among cocoa farmers. Moreover, the determinants of TFP were obtained using a Fractional Logit Regression model. The study found that while farmers achieved a technical efficiency score (ITE) of 0.66, indicating good use of existing resources, the output-oriented technical efficiency (OTE) score was lower at 0.56. This suggests room for improvement in maximizing output. The Fractional Logit model identified credit access, gender, household size, and access to extension services as significant determinants of TFP and its components. Furthermore, the performance of these indices improves the quality of access to extension services, access to credit, age, land ownership, gender, and household size. They are the major factors influencing these indices in the study. Based on these findings, the study recommends further research on the TFP of other cash crops and suggests improving access to credit and extension services for cocoa farmers. This could involve providing credit at affordable rates and training more extension agents to disseminate knowledge on improved technologies and agricultural practices.

Keywords: Färe-Primont TFP index, Technical efficiency, Total Factor Productivity, Liberia