Abstract

The prolonged yield stagnation of rice, the main staple food for most Indonesians, threatens Indonesia’s food security. One of the measures pursued by the government of Indonesia is the intensification of agricultural mechanisation. This study aims to evaluate the impact of government assistance in the form of agricultural machinery on the rice yield of farmers in Java Island. The region contributed to almost 60 percent of the country's total rice production. To avoid bias in impact evaluation, we conducted a counterfactual analysis by implementing Propensity Score Matching (PSM) on the microdata of the National Crop-cutting Survey for the years 2018 and 2021. We found that in the two years, the government’s agricultural machinery assistance had a significant impact in increasing the rice yield. The estimate for the average yield increase from farmers having received agricultural machinery assistance from the government was around 0.5 tons per hectare. When we look at it separately, we found that the impact in 2018 was much higher (around 0.59 tons per hectare of yield increase) than in 2021 (only 0.32 tons per hectare). It may be explained by the impact of the pandemic, which was the worst in 2021, and the massive improvement in mechanization adoption from 2018 to 2021. Moreover, by implementing an Ordinary Least Squares Regression (OLS) with robust standard errors on the same microdata, our findings pointed out that the impact of the government’s machinery assistance on yield improvement came with a condition. The assistance resulted in a higher yield if it was used and fit with farmers’ specification needs. Our findings even confirmed that if this condition was not met, the beneficiaries of the assistance could have a lower yield than the non-beneficiaries. Therefore, our study insisted on the importance of a farmers-based need approach in the implementation of the Indonesian government’s machinery assistance program to improve its effectiveness in increasing the paddy yield.