Crop residue incorporation (RI) is recommended to increase soil organic carbon (SOC) stocks. However, the positive effect on SOC is often reported to be relatively low and alternative use of crop residues, e.g. as a bioenergy source, may be more climate smart. In this context, it is important to understand: (i) the response of SOC stocks to long-term crop residue incorporation; and (ii) the qualitative SOC change, in order to judge the sustainability of this measure. We investigated the effect of 40 years of RI combined with five different nitrogen (N) fertilisation levels on SOC stocks and five SOC fractions differing in turnover times on a clay loam soil in Padua, Italy. The average increase in SOC stock in the 0–30 cm soil layer was 3.1 Mg /ha or 6.8%, with no difference between N fertilisation rates. Retention coefficients of residues did not exceed 4% and decreased significantly with increasing N rate (R2 = 0.49). The effect of RI was higher after 20 years (4.6 Mg/ha) than after 40 years, indicating that a new equilibrium has been reached and no further gains in SOC can be expected. Most (92%) of the total SOC was stored in the silt and clay fraction and 93% of the accumulated carbon was also found in this fraction, showing the importance of fine mineral particles for SOC storage, stabilisation and sequestration in arable soils. No change was detected in more labile fractions, indicating complete turnover of the annual residue-derived C in these fractions under a warm humid climate and in a highly base-saturated soil. The applied fractionation was thus useful to elucidate drivers and mechanisms of SOC formation and stabilisation. We conclude that residue incorporation is not a significant management practice affecting soil C storage in warm temperate climatic regions.