Funding for a typical research project is in a scale of $10^5~$10^6. And private funding like $5000 or so doesn't do much in most "real" research but may be a part of the whole project and if in this case, it is definitely duplicated (http://www.nature.com/news/funding-agencies-urged-to-check-f...).
Regarding the notion of duplication in grants, there is still a need for both repeating research and following projects with repeated aims, though in a slightly different fashion as promoted through recent reproducibility initiatives (http://www.nature.com/news/independent-labs-to-verify-high-p...).
