But scientists who do fraud are not a problem of the system, but it's mostly a problem of the scientists themselves. It occurred always and is going to occur in future too. Nowadays it seems more, but this is IMO a result of the tools we have now to detect it. Who'd have done in the 30s-80s of the last century a by-hand examination of all the quoted papers and relevant theses and compared it with the type-writer written thesis in suspicion? Now programs do it for you and also programs and algorithms can check your data if they're manipulated. Paper producing is surely one major cause doing fraud, but this need was always existent, but also the need to finishing your thesis, being successful for your boss, etc. This are inherent causes you can never avoid and only you'd get only less fraud if you change the character of the scientists.
Also, the risk of funding race and publishing is that unscrupulous researchers may falsify data to publish and have funding or position. There is a lot of examples, alas, including people who published in Science and Nature! Google it, you'll find a lot.
Perhaps at the given sponsor you'd have problems then, but there are other sources of money and if you can justify your missing results then it will be much easier. A related problem is surely that you only can publish positive results and not negative ones, and it would be helpful because it would avoid a lot of useless studies
Imagine that you work on a given research project but you didn't obtain results, what is next in the light of materialistic rating basis?
With rating and ranking system, this means that you won't, or hardly, obtain funding for other research projects because you didn't get results and publication from your other projects (so you are bad according to this flaw system). You won't get funding unless you are known and well supported by a strong network. This is unfair.
Anyway another topic, related to the system how money is given to scientists (by merits, networks, connections, applications, whatever) :
IMO a bigger problem is, that money is more and more given to mainstream projects and that good ideas or exceptionally ideas that might not work or are a bit or more than a bit off the mainstream ideas, are not supported. Funding usually supports only "streamlined", conformist research and scientists only submit quite safe research plans then to have a better chance. So unusual ideas and new paths are rarely supported and have a comparative low chance to be done. Finally science is quite homogenised and "fashions" of research topics are done by most people, which is an impoverishment for science.