When does an investigation start and end? What about when you're close to filling your limit but a few new great leads come in. Now the agents have to go close a ongoing investigation to follow up on a new one. Sounds like a lot of overhead. Not only that, but the fact of modern intel work is that investigations don't operate in a clear, linear progression.
I think that anyone that's had to deal with arbitrary limits imposed from above understands the perverse incentives that they create. This would be another example of administrators trying to fix something that they don't understand.
>When does an investigation start and end? What about when you're close to filling your limit but a few new great leads come in.
What about that? They could have a limit in the tens of thousands or so. As long as they don't waste it on BS, they could follow all the promising cases they want.
The idea is not to let them follow everybody, just for the fun of it.
You're focused on lots of hypotheticals whereas I'm talking about what they do. Currently, the NSA and FBI are looking at hundreds of millions of connections trying to find needles in stacks of needles. They admit, outside these political debates, that they're drowning in data. FBI's metadata investigations are producing jack and missing major attacks while targeted investigations following actual evidence are resulting in convictions. Speaking of perverse incentives, the current model is so ineffective and broken that they're actually creating terrorists to convict while claiming the arrests are due to successful, post-9/11 power. Even have a number on the incentive: $100,000 per terrorist created and arrested.
So, I'm pushing for more than an arbitrary limit on number of investigations. I'm pushing for them to switch their focus back to pre-9/11 where they leave most of America alone, keep good red flags, and focus their investigators on those. The barriers to intelligence sharing have been knocked down and local LEO's are integrated so dots should be connected faster than ever. With this focus, the limit will be the number of cases they're working simultaneously rather than on the books in general. It can be a multiple of the number of agents they have. Because it's hard for me to imagine one investigator effectively working ten, time-critical cases simultaneously. That's just overstretched although exceptions can be made for elite, multitasking, savant investigators if and where they exist.
So, we start with non-hypothetical stuff: the current surveillance dragnet that treats a whole country as guilty and watches every communication/association is an utter failure, a waste of money, and imposes chilling effects on democracy. The alternative is the original model of presumption of innocence, trained investigators following leads, prioritizing for most serious, and following through with good policework. Interestingly, the alternative already exists and runs in parallel with the surveillance state. Most convictions come from the alternative method. So, I don't have to justify it so much as point out that one is field-proven, it harms fewer innocents, and its cheaper. Ditching the failed method for that one is a no-brainer.
Unless, it's about money for defense sector and power over dissidents/competition for the nation that wields it. In that case, they'd try to keep the surveillance tech while continuing to lie about its purpose of protecting citizens from stuff it can't catch. Like we're seeing. ;)
I think that anyone that's had to deal with arbitrary limits imposed from above understands the perverse incentives that they create. This would be another example of administrators trying to fix something that they don't understand.