The objective of this study was to investigate the design and application of a carbon baseline for commercial timber harvest activities involving conventional timber harvest activities (CNV), relative to reduced-impact logging (RIL) in Sabah, Malaysia. As only RIL is eligible to be practiced in production forests, a baseline of CNV was estimated from the literature. The principle of net present value was applied to the post-harvest accumulation of carbon stocks after RIL to model a conservative ‘crediting’ baseline. Two areas representing opposite ends of a range of anthropogenic disturbance were sampled, with an old growth lower montane forest, and a lowland severely logged-over dipterocarp forest investigated before-, and two- and three-years after harvest, respectively. Areas impacted by CNV were estimated to contain 12–39% of pre-harvest carbon stock, relative to 57–63% under RIL and estimated to accumulate carbon in the range of 0.68–1.25 tC ha^-1 yr¹, averaging 14–55 years for recovery; in-line with body of knowledge. While the main limitation was our inability measure CNV directly, a balance of understanding is required for the development of a ‘best estimate’ using the literature.