Tender business

We are currently working to prepare some replies to government tenders. Looking at the way the tenders are structured I'm amazed how much -to put it mildly- room for improvement is left.
All tenders overlap in hardware requirements, infrastructure, middleware and identity management. Since it is very unlikely, that one party will win all of them, they will end up with incompatible, divergent solutions. The implementation time frame also will make sure, that competing teams, once the tender is awarded, will be on the ground at the same time making each others life as miserable as they can.
It is quite amazing how the 'law of unintended consequences' has hijacked the tender process. While intended to get the best deal for the government, it actually locks out specialist and performers and leaves only very big contenders who, by nature of their size, have high overhead costs. The best team to define, assemble and run a portal will most likely have no interest in hardware sales. So they are out. 10% of the project value needs to be deposited upfront, so the smaller teams are out. The hardware guys won't understand implementation to make it sound.
At the end they get so so service and a lot of bureaucracy instead of performance.
Luckily I don't pay this taxes since it's overseas.