No, it doesn't work. The US has been actively trying this in many states, so we have data on the relationship between deterrence effect and the likelihood*harshness of punishments - it turns out likelihood really controls the strength of the effect of deterrence. If you have extremely harsh punishments you DO incentivize doing a lot of damage to get away with it, and also incentivize repeat or exacerbated offending. (See: "might as well be hung for a sheep as a lamb")
Uh you don't have data on that for white collar crime. If you do, please share.
Petty theft is done for different reasons to white collar crime. I agree with you that the punishment is not a deterrence for small offences.
I think there's a lot of deterrence coming from the Enron case where people actually went to jail. That's just too rare because you quickly get into questions of intent. And to prove intent is really hard.
Your note about 'proving intent is really hard' is kind of the underlying reason - white collar criminals tend to believe they will never be caught/convicted. So making sentences more severe for those who are caught and convicted doesn't actually impact those future criminals, because their calculations say there's no risk of it being applied to them.
