[patrec]

> As a society, we believe in the deterrent value of punishment (which is proven not to work).

Do you genuinely believe that high-IQ white collar criminals are exactly the type of person unable to do even the most basic risk-reward analysis?

[metabagel]

It doesn't seem to have worked for Elizabeth Holmes or Bernie Madoff.

According to this fact sheet, "the certainty of being caught is a vastly more powerful deterrent than punishment". Elizabeth Holmes and Bernie Madoff probably didn't think they would be caught.

https://www.ojp.gov/pdffiles1/nij/247350.pdf

NIJ is the research, development and evaluation agency of the U.S. Department of Justice.

[patrec]

> It doesn't seem to have worked for Elizabeth Holmes or Bernie Madoff.

Penalizing mass murder also doesn't seem to have worked to prevent 9/11. The question is how representative the perpetrators are in these cases. I'd say not very -- do you disagree?

> According to this fact sheet, "the certainty of being caught is a vastly more powerful deterrent than punishment".

Just because you read it in some fact sheet doesn't mean it's true or even makes any sense. Probabilities and punishments are not commensurable so what does this even mean (without some attempt to describe a subjective loss function)?

Taken literally it's obviously untrue. Examples abound where people don't care much about being caught because the punishment close to non-existent or very light in comparison to the rewards. This is why why gangs like to groom minors, the likely reason why San Francisco has a shoplifting problem, or to bring it back to white-collar crime, why certain types of corporate and individual malfeasance occur over and over again, because the risks, even when caught, are minor but the rewards are not.

Your original claim is tantamount to saying that people don't react to incentives. Which beggars belief. That is not to say that criminals in particular respond to incentives in a way that maximizes their life outcomes or is intuitively obvious, or that for many crimes lowering penalties and investing in other things instead (like detection or prevention) wouldn't produce better societal results. But on the whole I'd expect high-flying financial criminals to respond to incentives set by justice system in a way that is more aligned with their actual utility function than say crack addicts committing robberies to finance their next fix.

[candiodari]

What Elizabeth Holmes tried to do is possible. So we can be a bit kinder than that: she thought with enough help she’d figure it out and make her lies the truth before being found out.

She didn’t of course, but it wasn’t impossible to do. She just wasn’t good enough.

After all most cells in your body do a few thousand blood tests on millilitres of blood every second. Our technology has a long way to go but it can be done.