AbstractAimTo assess spatio‐temporal and taxonomic patterns of available information on the costs of invasive freshwater bivalves, as well as to identify knowledge gaps.LocationGlobal.Time period1980–2020.Taxon studiedBivalvia.MethodsWe synthesize published global economic costs of impacts from freshwater bivalves using the InvaCost database and associated R package, explicitly considering the reliability of estimation methodologies, cost types, economic sectors and impacted regions.ResultsCumulative total global costs of invasive macrofouling bivalves were $ 63.7 billion (2017 US$) across all regions and socio‐economic sectors between 1980 and 2020. Costs were heavily biased taxonomically and spatially, dominated by two families, Dreissenidae and Cyrenidae (Corbiculidae), and largely reported in North America. The greatest share of reported costs ($ 31.5 billion) did not make the distinction between damage and management. However, of those that did, damages and resource losses were one order of magnitude higher ($ 30.5 billion) than control or preventative measures ($ 1.7 billion). Moreover, although many impacted socio‐economic sectors lacked specification, the largest shares of costs were incurred by authorities and stakeholders ($ 27.7 billion, e.g., public and private sector interventions) and through impacts on public and social welfare ($ 10.1 billion, e.g., via power/drinking water plant and irrigation system damage) in North America. Average cost estimates over the entire period amounted to approximately $ 1.6 billion per year, most of which was incurred in North America.Main conclusionsOur results highlight the burgeoning economic threat caused by invasive freshwater bivalves, offering a strong economic incentive to invest in preventative management such as biosecurity and rapid response eradications. Even if the damages and resource losses are severely understated because economic impacts are lacking for most invaded countries and invasive bivalve species, these impacts are substantial and likely growing.