The use of biomass for energy production is considered a promising way to reduce net carbon emissions and mitigate climate change. However, land-use change to bioenergy crops can result in carbon emissions from soil and vegetation in amounts that could take decades to compensate. Perennial grasses such as Miscanthus offer a possible solution to this problem as measurements on experimental plots planted with Miscanthus have shown significant carbon sequestration in the soil. It can, however, be expected that sequestration potentials in commercial use might differ from those measured in experimental plots due to different farming practices and soil characteristics. For this study, Miscanthus plantations on 16 farms in SE Ireland as well as on-farm controls representing the former land-use (grassland and tillage) have been examined. The Miscanthus plantations were 2–3 years old. Soil organic carbon (SOC) content and a number of soil properties were measured and the amount of Miscanthus-derived carbon was determined using the 13C natural abundance method. On both former tillage fields and grasslands, although there were no significant differences in SOC contents between Miscanthus and control sites, it was shown that 2–3 years after Miscanthus establishment, 1.82 ± 1.69 and 2.17 ± 1.73 Mg ha−1 of the SOC under former-tilled and former grassland respectively were Miscanthus-derived. Mixed-effects models were used to link the total SOC concentrations and Miscanthus-derived carbon to the land-use parameters as well as to soil properties. It was shown that on control sites, pH had an effect on total SOC. In the case of Miscanthus-derived carbon, the initial SOC content, pH, former land-use and crop age had significant effects.