But your threat level going forward is likely not just an individual, but rather the big data collection of corporations such as Facebook and Google which track every damn thing we do via cookies, etc.. And the likelihood these behemoths will be forced to cooperate with the NSA in the coming 666 world order that is developing. Have you all not seen the proclamations of the G20 to cooperate on sharing information about tax evaders? Have you all not seen that China has 10 cameras on every rooftop and London apparently the same. And even if one argues these behemoths won't target you as a nobody, don't forget their employees or hackers could obtain the data and blackmail you or what ever.
Zerocash could possibly ameliorate that horrible future (because all incriminating data is entirely encrypted into a featureless blob before it leaves your computer). So I say paradigmatic distinction.
The problem is, at that threat level you can't know that any of those 10 cameras on every rooftop along with other such techniques known and unknown aren't shoulder-surfing you before that blob leaves your computer or using some other techniques. It is even possible to technology to see through walls, so closing the window or staying away from windows may not help you. (Maybe they can't see your screen through walls, but they can see your fingers typing, or eavesdrop as you use a voice interface.)
Guarding against surveillance of all kinds is really extremely difficult. To make any progress you have to narrow the problem, and attack in pieces. Zerocash is no panacea.
I will argue that you are equating threats which have a paradigmatic distinction, in that one is globally unbounded and the other is locally bounded (at least at the non-quantum mechanics model). And just as the difference between proof-of-stake and proof-of-proof is fundamental on the distinction between unbounded entropy (and implications thereof such as the ability to defraud with a combination of shorting and stake control as I explained in some comments last month or so responding to older jl777 quotes ... no time to go digging for that now...), I argue this distinction on privacy threats is fundamental.
Obfuscating your IP address can never be provably controlled by you, because the threat is unbounded in entropy. Whereas, the ability to encrypt your blob without someone having access to your private key is a local perimeter threat (and you are incorrect to imply that merely observing whom is encrypting is any form of a threat if you did mean to imply that). As long as you do your encryption inside of a lead and RF shielded case and your private key has not been compromised, then you can be 100% sure that the activity in that encrypted blob is secure (up to the limits of number theory for encryption and the Star Trek-like quantum mechanics threats that science doesn't know about or fully understand yet such as tunneling, many worlds hypothesis, etc..).
You could retort that the number theory of the encryption can also be cracked with unbounded computation, but it is intractable else crypto is intractable. Whereas, IP unmasking is not intractable. Many white papers have been written on various tractable attack threats. My point in saying the threat level is unbounded for IP address, is we can't prove the unmasking is intractable because in one example we can't even know the level of Sybil attack on the anonymity sets. And there are numerous other reasons. I have spun my head around and around on this many times.