Welcome to the Kali Yuga! The age in which down is up.
However, power is a transferable abstract (like currency). If you frighten a population into paralysis, then you may appear to lose creativity, motivation, economic wealth, etc. However, those losses simply become the "cost" of "buying" different forms of power.
I don't think we can assume those are equal (or even of the same magnitude).
[1] I don't think people have the correct perception of how oppression works in a democracy. The line between the FBI trying to get MLK to commit suicide and the FBI imprisoning random middle-class white Americans is a lot bolder than people assume. I think the misunderstanding stems from a failure to appreciate the nature of oppressive government actions within a democracy. It's not like oppression of civil rights leaders in the 1950's and 1960's was the result of a self-contained entity within government, who could turn its attention to ordinary people at a whim. Instead, it was a classic case of the majority oppressing the minority, acting through government. Oppression of the majority requires an inversion of that power dynamic, and I think is harder to achieve than people assume.
This is all, of course, not to justify oppression of the minority. Rather, it's to point out that your typical person is quite justified in assuming the government has little reason to oppress him.
My concern is that excessive surveillance leads inevitably to excessive perceptions of culpability.
One does not need to be formally found guilty for a run-in with the law to lead to "temporary" detention and legal bills; and if you can't afford the legal bills, maybe you might be forced into plea bargaining and end up with a criminal record. Even if this scenario does not happen, a genuine harm to one's reputation as a result of "helping the police with their enquiries" often occurs, especially if one is a middle class corporate employee.
If you can't post bail you could be in jail for months on end, with no compensation if eventually found innocent.
Corporations can, and increasingly do, pursue actions against individuals which result in prison time being imposed, including for purely financial transgressions which bear an exceptionally strong resemblance to debtor's prison.
Corporations per se do not have the ability to imprison people, they petition the government to do so. Corporations may sue people in the courts, but if the suit does not result in a criminal conviction (as opposed to a civil one), it does not result in imprisonment.
The same law made violating a Website's TOU/TOS a felony -- which in the case of 17 Magazine meant that visiting its website if you were in fact 17 was a Federal crime.[2]
The reform to the Computer Fraud and Abuse Act which would have closed the loopholes under which Aaron Swartz was blocked by Larry Ellison, CEO of Oracle Corporation.[3]
______________________________
Notes:
1. http://www.documentcloud.org/documents/217117-united-states-...
2. https://www.eff.org/deeplinks/2013/04/until-today-if-you-wer...
3. http://o.canada.com/technology/swart z-doc-director-oracle-and-larry-ellison-killed-aarons-law
The other thing I'd add to this discussion is the question of what you're sharing with whom. The who of it all matters a great deal. Amazon collects a lot of data about my purchase history, for instance, and while the ramifications make me somewhat uncomfortable, I accept them. It's part of the shopping experience, and it's part of Amazon's business model. But take that purchase history out of context, and put it in the hands of someone looking for a very different set of inferences about me. A healthcare provider, for example, who can make startlingly precise inferences about my lifestyle and my likely state of health from my Amazon purchases. This scenario seems farfetched today. And I trust Amazon to be a reasonable steward of my information, and to maintain some sanctity over the context in which it's aggregated and analyzed. But I'm scattering plenty more breadcrumbs around the internet in my digital travels. I can only imagine the novel and ingenious ways they'll be used in the future -- for me, hopefully, but just as easily against me. It all depends on who's looking at my data, and for what purposes. Those people, and those purposes, might be very far removed from the people and purposes I'd originally envisioned when I'd shared my data. In most cases I can't really control that. That's what scares me.
Data context matters a great deal. So does the presumed aggregator of the data, and so do the presumed consumer(s) of that aggregation.
I think it's important to make this point to the general public, because the typical bogeymen (the NSA, Google, Facebook, etc.) don't really mean much to most people. When you conjure up the NSA, your average layperson will shrug and say "I have nothing to hide from them." Just mentioning the NSA brings up a fantastical frame of reference: a frame involving terrorists, leaks, Jack Bauer, and Edward Snowden. Most people, in that frame of reference, find it so far removed from their daily lives as to be almost comical. This breeds apathy. That apathy gets you the shrug and the dismissal. But conjure up a different frame of reference -- being denied health coverage because of purchase histories, or being turned down for a job because of image searches -- and people seem to get the idea.
We need to do a better job educating people about a) closer-to-home implications of privacy erosion, and b) examples of data, seemingly mundane, that can be aggregated to powerful effect. Yes, the big stuff matters. We should be making a big fuss about the NSA. But that's not the rhetorical starting point that's going to get Joe Sixpack to care. In some cases, it might even be counterproductive.