Fossil‐based emissions can be avoided by using wood in place of non‐renewable raw materials as energy and materials. However, wood harvest influences forest carbon stocks. Increased harvest may reduce the overall climate benefit of wood use significantly, but is widely overlooked. We reviewed selected simulation studies and compared differences in forest carbon and amount of wood harvested between harvest scenarios of different intensities for three different time perspectives: short‐ (1–30 years), mid‐ (31–70 years), and long‐term (71–100 years). Out of more than 450 reviewed studies 45 provided adequate data. Our results show that increased harvest reduces carbon stocks over 100 years in temperate and boreal forests by about 1.6 (stdev 0.9) tC per tC harvested (referred to as carbon balance indicator (CBI)). CBI proved to be robust when outliers explicitly influenced by factors other than changes in the harvest rate, such as fertilization or increase in forest area, were removed. The carbon impacts tend to be greatest in the mid‐term, but no significant difference in was found for average values between short and long time‐horizons. CBI can be interpreted as carbon opportunity costs of wood harvest in forests. Our results indicate that even after 100 years, CBI is significant compared to the typical GHG credits expected in the technosphere by avoiding fossil emissions in substitution and increasing carbon stocks in harvested wood products. Our estimates provide typical values that can directly be included in GHG balances of products or assessments of mitigation policies and measures related to wood use. However, more systematic scenarios with transparent information on influencing factors for forest carbon stocks are required to provide better constrained estimates for specific forest types.

Vauhkonen (2023) raised concerns regarding the applicability of the carbon balance indicator (CBI) applied in our paper (Soimakallio et al. 2022) to characterize the results in various contexts, the way the scenario pairs were compared, and the way the results were filtered. Here we respond to each of the above‐mentioned concerns Vauhkonen raised and finally to the conclusions that he drew.