If you reduce the chance of going to jail for "non-violent" crimes such as burglary, you will get more such crimes, because burglars are neither deterred nor incarcerated.
For some reason my comment was downvoted. Here is a quote from the end of the article showing that burglars are not being punished severely enough to deter them:
When Martin recalled the June break-in at the Pine Tar Grill, what sticks in his mind wasn’t losing the computers or bobbleheads. Instead, it was a conversation he had with a police officer who viewed the footage of the break-in, who told him that he recognized the person in the video and that he had been arrested multiple times, Martin said.
“As a business owner,” Martin said, “that’s just really hard to hear.”
It's downvoted because it's a horrible simplification of a very complex problem and it isn't supported by evidence. If it were true then we would see a very clear correlation in the data between sentencing policy and crime rates, but we basically don't. In fact as sentencing laws in almost all developed countries have reduced, we've seen reductions in crime.
We shouldn't mix up "chance of going to jail" with "severity of punishment". Potential criminals easily dismiss severe punishment, but they don't dismiss near-certain punishment. Quick and reliable punishment is highly effective. Increasing the severity is weakly effective.
There is also the simple fact that people in custody are physically separated from potential victims. This very obviously works.
