In Germany it's not the job of a prosecutor to get the maximum possible sentence, so no the problem doesn't exist in the same way in Germany. If someone in Germany commits a crime that gets 3 years and another that's 4 years that doesn't simply add up to 7 years. Our system is much more well designed.

Being tough on crime happens to be a politicized topic in the US and there are many people who do hold the opinion that the US incarnates a percentage of it's population that's significantly to high. It's no Republican vs. Democratic issues but that doesn't mean it's not political in nature.

In this case I'm not sure whether blackmail is really the right term. Wikipedia defines blackmail as "Blackmail is an act, often a crime, involving unjustified threats to make a gain or cause loss to another unless a demand is met." Is using a valid law for charge inflation an unjustified thread? That depends a lot on your political beliefs.

political implications

No. We do have research on how people get mindkilled. It's not about implications. Certain classes of claims for example lead to most people stop being able to use Bayes Rule when you ask them to analyse a problem. I'm claiming that the example is in that class.

Are you saying that you disagree with it personally?

A core question in this case is: "What do we gain from defining blackmail in a way that this example is covered? What do we gain from defining it in a way that this isn't covered?"

Given that most people here disapprove of the action of those persecutors politically they feel the desire to punish them by using a negative label. That makes it harder to have the discussion based on the merits.

Because "don't talk about this general fact because someone else might think it has (weak) political implications" seems a heuristic to be avoided.

That's not the heuristic brought forward in "Politics is the mindkiller". The heuristic is: You have a set A of examples (X_1, Y_1. X_2, Y_2, Y_3 ...). The X examples are political and fire up a bunch of mental biases in your reader. Most of your readers will suddenly start to flunk Bayesian calculation if you make one of those X_i examples. The Y_i examples on the other hand allow your readers to reason normally. If you want to choose an example for A, to speak about A don't choose one of the X_i but one of the Y_i.

That example seems significantly more politicized. It definitely is, in a US context; the "Death Panels" political meme grew out of attempts to deal with that problem.