Food security remains a challenge to a number of developing countries despite numerous efforts taken by governments and other national and international bodies to improve it. Approximately 27 percent of the global population experienced some level of food insecurity during 2018. Moreover, the prevalence of food insecurity is higher in countries with a larger percentage of the population living in rural areas (FAO, 2019). The agriculture sector, as the main source of employment in rural areas, has not been a sufficient vehicle for ensuring household-level food security. Therefore, many rural households maintain a portfolio of income sources, with nonfarm employment being a major component. However, there is a lack of universal agreement among researchers pertaining to the impact of nonfarm employment on rural household food security. While rural household food security could be improved through better access to food due to increased nonfarm income, food production could be negatively affected due to the increased competition of farm household labour for nonfarm employment. Surprisingly little is known about the impact of nonfarm employment on rural household food security. In this context, this thesis attempts to focus on five major research objectives, considering Bangladesh as a case study. Firstly, this thesis determines the factors that influence an individual’s participation in nonfarm employment. In doing so, the determinants of nonfarm employment are investigated at three levels (see Figure i below): (1) overall nonfarm employment, (2) nonfarm employment disaggregated into low-return and high-return nonfarm employment, and (3) low-return and high-return nonfarm employment both further disaggregated into wage employment, self-employment, and business employment. The estimation results using a binary logit model indicate that, in general, characteristics such as gender, age, the headship of the household, marital status and education influence individuals to participate in nonfarm employment at the overall level. However, by disaggregating nonfarm employment into different categories such as high-return or low-return employment and further disaggregating the low-return and high-return nonfarm employment into wage, self, and business employment, and by applying the multinomial logit model, this study finds a more complex and heterogeneous landscape with respect to the types of determinants that affect each category of nonfarm employment. Secondly, this thesis analyses the potential impact of earnings from rural nonfarm employment on agricultural production. By applying the instrumental variable Tobit model, this thesis finds that when nonfarm income increases, the amount spent on total rural farm production, as well as the expenditure on major purchase rural farm inputs, also both increase. In addition, the results of a stochastic frontier production model analysis indicate a decrease in the technical inefficiency of agricultural production due to an increase in nonfarm income. Overall, the findings suggest that rising rural nonfarm income exerts an income effect and could be a means to increase food production, ultimately leading to food availability as well as food security. Thirdly, this thesis investigates the impact of nonfarm employment on household food security based on calorie consumption per day per adult equivalent. The results show that household calorie consumption per day per adult equivalent can be significantly improved by generating additional income via participation in rural nonfarm employment. Moreover, the analysis reveals that average nonfarm income is significantly higher than average farm income. Fourthly, the investigation into food security looked at the possible impact of nonfarm employment on child nutritional security. Utilising the well-accepted measure of child anthropometric indicators, this study finds that additional earnings from nonfarm income have a significant and positive impact on children’s nutritional security in rural households. Finally, this thesis compares the level of food security between rural and urban populations. By applying logistic estimation, this study finds that rural households are more likely to be food secure than urban households in Bangladesh. In addition, it identifies a significant difference in a limited number of determinants of food security between rural and urban households. The findings of this study suggest that a common national-level policy that excludes rural–urban disparities might be ineffective in improving the overall food security of Bangladesh. In summary, the findings of this thesis indicate that an improvement in rural nonfarm employment opportunities could help to foster growth in rural household food security based on a range of dimensions. Thus, policies should be directed toward providing incentives to rural households and individuals to participate in specific and targeted rural nonfarm employment, as well as increasing their capacity to take advantage of such opportunities without reducing the required labour force for rural farming. To do so, policy initiatives could encapsulate measures that boost female participation in nonfarm employment, improve educational facilities and infrastructure facilities, and ease access to credit facilities in rural areas. The findings of this thesis have important policy implications for the effective targeting and improvement of household food security in relation to nonfarm employment in rural Bangladesh, and in other countries with similar socioeconomic characteristics.