The major objective of this study is to analyze the effect of non-farm employment on multidimensional poverty in rural Ethiopia using balanced panel data from ESS2013/14 and 2015/16. The study employed Alkire and Foster poverty model and probit regression model. Ten indicators of multidimensional poverty are categorized in three dimensions of multidimensional poverty. Results reveal that the incidence of multidimensional poverty among the rural households of Ethiopia isreducing from 94.9percent in 2013/14 to 92.5% in 2015/16,and average deprivations of indicators and the adjusted multidimensional poverty reduced from 50.2% and 49.5%in 2013/14 to 49.7% and 47.4% in 2015/16 respectively. From the three dimensions, standard of livingwas the highest contributor to the multidimensional poverty index of rural Ethiopia with share of 43.8 percent followed by health dimension at 34.2 percent. On the other hand,thepercentage of rural households thatengaged innon-farm employment increase from 28.9% in 2013/14 to 31.3% in 2015/16. The result of the probit regression shows that non-farm employment has asignificant effect on multidimensional poverty status.Non-farm employment activities reduce the likelihood of the rural household being multidimensionally poor on average by 0.032.It is recommended that the government should implement policies that promote non-farm employment, and increase credit access for rural households.