It depends on what is meant by duplication. If two groups are researching the same thing using the same means regarding the same factors, then that is doing something in parallel. I can see that as a (possible) waste of money that could be used to research something else concurrently. Only slight related, when it is different agencies funding the same party, then you have fraud: http://www.nature.com/news/duplicate-grant-case-puts-funders-under-pressure-1.9984
Replication is different and would not fall under duplication as it is done serially. First, one group does research into the topic followed by a separate group that tries to reproduce the results. Trying to replicate the results at the same time as another group that is unfinished with their research is potentially wasteful.
If the results are useful, then I am sure some entity will try to reproduce it without government funding. If the project was politically-motivated, then I am almost certain another party will fund research into that topic without need for government funding.
Personally, I wish the news would do a little research into past projects that were duplicated to either prove or disprove the issue with duplication. They just want a fight between the two parties to get more readers. I guess this was too hard for them to find: http://www.coburn.senate.gov/public/index.cfm?a=Files.serve&File_id=2dccf06d-65fe-4087-b58d-b43ff68987fa Page 20 talks about duplication between the various agencies. Skimming through that report really makes me want to have the NSF cleaned. For example, "An Indiana University (IU) professor received a $263,281 grant from the NSF to study the social impact of tourism in the country of Norway." Funding that over cancer research?!?