I don't think blanket statements like this are true, and I don't think they're a healthy way to frame the debate. To take a non-trivial example, requiring cars to bear license plates - something almost everyone accepts, at least in the United States - has a lot of upsides like enforcement of traffic laws. Sure, perhaps traffic cameras aid totalitarian states in tracking the movements of citizens, but I think most people agree that license plates are a good societal tradeoff in terms of privacy.

This is a long way of saying that, even as a strong privacy advocate, it's worth noting that the tradeoff is real and the terms of debate ought to be over whether the tradeoff is a good one or not. In the case of encryption, this is a difficult question for us to answer because we're essentially asking how many crimes might have been prevented or might have gone unpunished had the perpetrators used encrypted communications. That being said, I think the security state would be hitting us over the head with these statistics if they could actually make a coherent case - the fact that Reid Blackman is instead making a fallacious comparison between encryption and the nuclear launch codes suggests that the figures don't add up.