An important problem in measuring the impacts of business subsidies is their separation from deadweight, which refers to changes that would have occurred even in the absence of intervention. Both public and private assessments have been used previously to study deadweight, but so far little is known about how they correspond to each other. To address this issue, we conducted a joint evaluation of the private and public assessments of deadweight for Finnish business projects. A unique data set combines large register data with both public and private information on projects financed in 2000–2003. First, our results suggest that the different measures for deadweight are greatly uncorrelated, and thus cannot be used as substitutes. Second, characteristics affecting the public and private measures of deadweight are identified using ordered probit models. We find that the public and private sectors emphasize different factors in their assessment of deadweight. Third, an upper bound for the level of deadweight spending is estimated at 73.8%.