Ugh. Being able to delay social policies was the only thing that made them interesting. They were decently balanced before - early policies are weaker but make impacts earlier. Taking policies later makes them cost more (because you will likely have more cities) or you avoid making new cities - an associated cost of later policies. The balance might be a little off, but that can be changed by buffing earlier policies.
Now the only way to plan policy taking is to heavily micromanage techs and culture, delaying policy acquisition (culture buildings) until good ones open up. This seems like an even bigger deviation from how the game is supposed to be played than delayed policies!
FAIL.
I think the point is to enforce a 'natural' playstyle. In the same way that some shooters disallow firing sniper rifles that aren't aimed or make freejumping impossible they are attempting to make 'double ninja backflipping' impossible.
To clarify, 'double ninja backflipping' is playing in a counter-intuitive way in order to pump a mechanic way past its intended role. 'Natural' skillful playing would involve making a series of solid, rational decisions with an overarching goal. 'Backflipping' is usually very mechanical and flies in the face of what the game mechanics are intended to express.
Examples of what I mean by Double Ninja Backflipping:
- Inducing anarchy to avoid paying city upkeep in Civ4(I read this on a Deity strategy somewhere, I've since forgotten)
- Purposefully maximizing overflow on walls to get extra money with stone+protective(Also Civ4)
- Deliberately maximizing overflow on a tech with a high research modifier(Civ4 again)
- Staying level 1 on purpose to exploit the level scaling(Oblivion, others.)
- Severe ICS(Civ5)
- Ignore happiness(Civ5)
- Blowing yourself up with a grenade to 'jump'(various)
- Building nothing but builder units and upgrading them to fight(Company of Heroes)
- Getting a large number of your men killed so that medics replace them with more powerful men(Also Company of Heroes)
These are things that, if you take the game at face value, appear to be blatantly irrational behavior. However, due to unforeseen blind-spots in the game mechanics, they are actually quite effective.
I know this is a controversial opinion, but I like it when game designers target these tricks and make them suboptimal. I know most 'hardcore' gamers don't agree with me, but games should reward solid understanding of the mechanics and skillful, conservative play. They should not reward the person who is the best at exploiting whatever the most powerful exploit happens to be.
TL;DR: Mechanics shouldn't be able to be used in ways that circumvent/defy what the mechanic is intended to express.