Even-aged stands can regenerate with many thousand seedlings per hectare before the density declines to just a few hundred trees per hectare 100 years later; management practices can lead to even lower tree numbers due to quality selection and thinning. In other words, during the development of unmanaged stands, the majority of individuals die naturally due to competition. Despite the far-reaching consequences for structural and genetic diversity, dead wood and fuel wood accumulation, we have only limited quantitative knowledge about the continuous mortality of trees and the wood volume loss over longer timespans. For this study, we used a unique set of 476 unmanaged, monospecific experimental plots of Norway spruce (Picea abies (L.) H. Karst.), silver fir (Abies alba Mill.), Scots pine (Pinus sylvestris L.), European larch (Larix decidua Mill.), Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco)), European beech (Fagus sylvatica L.), and oak (Quercus robur L. and Quercus petraea (Matt.) Liebl.) throughout Europe to analyze the competition-based mortality of trees and its dependency on age and site conditions. First, we show that the total stem volume production, standing stock, and mortality were continuously increasing until an age of 100–150 years. The accumulated competition-caused stem volume loss at that age amounted to 500–1000 m³ ha⁻¹. Second, the net growth of the stands (share of the growth that is accumulated in the standing stock) strongly decreased with increasing age even when the gross growth was still high. The proportion of the net growth versus gross growth continuously decreased with increasing age regardless of site quality. Third, we show a degressive decrease of the annual relative tree number mortality rates from 0.05 to 0.20 in young down to 0.01–0.02 in mature stands. For some species, we found these rates to be site dependent with different directions of the site effect. The interplay of decreasing mortality rates and increasing average volume of the dead trees resulted in unimodal mortality curves over time of the annual mortality, peaking at 3–12 m³ ha⁻¹ yr⁻¹ at ages of about 75–150 years. Over the whole rotation, the average annual biomass loss from mortality ranged between 0.8 and 2.1 t ha⁻¹ yr⁻¹ with a carbon content of 0.4–1.1 t C ha⁻¹ yr⁻¹. We discuss the relevance of the results for measuring, understanding, modelling, and managing forest stands. Our results reveal that the withdrawal of forest management and setting aside (previously managed) forests over a rotation time of 100–150 years means that about one third of the total production in monospecific stands would flow to the debris pool rather than being exploited for carbon sequestration and related emission savings in harvested wood products. The mortality related loss fractions of above ground biomass we quantified in this study indicate the trade-off between wood production and setting aside forest to allow deadwood accumulation and associate changes in biodiversity.