Simple Justice raises a point I both want to agree with and disagree with. He posits that increasing maximum penalties is an ineffective deterrent to those who tend to break the law.
There are two reasons Scott's correct. The first is that most of the time nobody, outside of some legislators and lawyers, knows when laws are changed. There won't be multiple news stories splashed across TV, Drudge, and the local paper if the first of July comes around and Virginia changes larceny of a dog from a class 6 felony (up to 5 years) to a class 5 felony (up to 10 years) - along with the 500 other major and minor changes made to the law. If no one knows that the law has changed it can't be factored into a cost-benefit analysis and shan't be a deterrent.
The second reason is that often knowledge works against the intended effect. Those with familiarity with the system will know that changing stealing a cat from a misdemeanor to a felony will probably result in less actual jail time. A district court judge would probably give a little time in jail for stealing Tigger. In circuit court this case will be seen as less important than drug dealers, murderers, bank robbers, etc. and, at least in Virginia, the judge will have sentencing guidelines telling him the defendant should get probation. I'm not saying that the offenders know all the intricacies of sentencing, just that they know that low level felonies usually end up with probation (at least the first and maybe second conviction) and aren't concerned if a misdemeanor changes to a low level felony.
On the other hand, Scott is wrong in that changes in the law do make their ways into the consciousness of the community. Sometimes this rise in awareness can be fast; sometimes it can be slow. In Virginia an example of the fast rise would be Exile, gun control laws. The government saturated the market with well done, well placed, effective commercials and continues to follow up with reminder commercials. This effort penetrated the minds of those whom it was supposed to effect and they all know that felon+firearm or drugs+firearm is bad and will result in mandatory prison time. Slower penetration happens when the law changes and the community comes into contact with that law fairly regularly. An example of this was when Virginia overhauled its DUI statutes a few years back. Offenders did not know the law had changed, but they went back and told other members of the community; it took about two years, but the changes finally penetrated to the point I didn't have to argue over the state of DUI law with people anymore.
Personally, I think that offenders do a cost-benefit analysis. They wouldn't phrase it like that, but they do consider what they know and what they think they know. Scott's correct in his assertion that they sometimes have the wrong idea about what the law actually is: "I can't be convicted of shoplifting. He put the stuff in my purse; I didn't touch anything." That statement gets the law wrong, but it implies a cost-benefit analysis: I can't go to jail if he's the only one who touches the stuff so I'll let him put stuff in my purse because (multiple choice) 1. I luv him, or 2. I'll get my cut later, or 3. He'll give me drugs if I let him.
Of course, there will always be entirely irrational actors; however, I think these are few. And, of course, Scott is correct in his assertion that crimes committed in the heat of the moment will not be effected by the costs, no matter how great (nothing will effect these until pre-crime units are deployed). In the end, we must act to influence rational actors and be prepared to deal with those so far outside the norm as to be uneffectable.
How do we do this? Well, moving things up the maximum potential punishment scale isn't really going to accomplish anything. Two things need to happen. To begin with, any increase should be in the actual punishment scale. In other words changing something from a misdemeanor to a felony doesn't accomplish much. However, changing a misdemeanor from 0-12 months to 3-12 months does something which will impact the sensibilities of the community: it mandates more actual punishment. Personally, I'd be in favor of returning a lot of felonies to misdemeanors as long as the misdemeanors are misdemeanors with teeth. I think that 3 months in jail on a misdemeanor has more of a deterrent effect than felony probation.
Then, punishment needs to be consistent. Yes, yes, I have read some Emerson, but I'd remind you that he is talking about "foolish consistency." I'm talking about a considered consistency. Punishment needs to be consistent over a long period of time, otherwise it does not have the desired impact. If the punishment does not remain consistent it doesn't sink in to the community. For instance, if felony petit larceny (3d conviction) were changed back to a misdemeanor with 6-12 months after a couple years to sink in it could impact the thoughts and acts of the community. However, if the typical disposition were to reduce every 3d to a plain old petit larceny and give a weekend in jail it wouldn't work. This, IMO, is the reason that the massive federal drug sentences haven't made a dent in drug trade. 98% of offenders were getting State sentences of a year, maybe two, perhaps four if they were really bad (yes I know your State had much worse punishment - this is just an example) while 2% were getting 10 years or 20 in the federal system. Let's imagine the equation here (all numbers pulled out of my imagination):
$3,000 per week * 52 weeks [<=>] 25% chance of being caught (2% * 10 years in prison)
$3,000 per week * 52 weeks [<=>] 25% chance of being caught (98% * 2 years in prison)
Now, let's imagine a world in which the punishment in both was equal and set at the higher level.
$3,000 per week * 52 weeks [<=>] 25% chance of being caught (100% * 10 years in prison)
The risk is higher when the punishment is consistently higher and this will sink into the community. Will higher consistent minimums stop all crime? No. The best you can ever hope to do is raise the entry threshold. If 40% of the community is willing to enter into illegal activities under the first set of equations and 15% are willing to enter into illegal activities under the last equation that's moving things in the right direction.