Influential research in terrestrial habitats indicates that several ecosystem processes are related to plant biodiversity, yet these links remain poorly studied in marine ecosystems. We conducted one field and one mesocosm experiment to quantify the relative effects of macroalgal species identity and richness on primary production in coral reef macroalgal communities off the north coast of Jamaica. We measured production as the net accumulation of algal biomass in the absence of consumers and as photosynthetic rate using oxygen probes in sealed aquaria. We used two recently developed techniques to attribute deviations in expected relative yield to components associated with species identity or diversity and then to further partition diversity effects into mechanistic components based on dominance, trait-dependent complementarity, and trait-independent complementarity. Our results indicate that algal identity had far greater effects on absolute net growth and photosynthesis than richness. The most diverse mixture of macroalgae did not outperform the most productive monoculture or the average monoculture in either measure of primary production (i.e. we did not find evidence of either transgressive or non-transgressive overyielding). Trait-independent complementarity effects were positive but dominance and trait-dependent complementarity were both negative and became stronger when richness was increased. Thus the potentially positive influence of species interactions and niche partitioning on production were negated by dominance and other negative selection effects. These results demonstrate that the counteracting influence of component effects can diminish the net richness effects on production. This could explain frequently observed weak net richness effects in other aquatic and terrestrial systems and suggests that life history tradeoffs greatly reduce the potential for ecologically relevant plant biodiversity effects on ecosystem properties.