Nobel prize winner Gary Becker postulated a utility maximization framework to describe an individual's participation in a crime. Certain crimes have clear economic rewards, but most criminal behaviors have costs.

Here is his general function:

y = f ( x1 , x2 , x3 , x4 , x5 , x6 , x7 )

y is hours spent in criminal activities,
x1 is 'wage' for an hour spent in criminal activity
x2 is hourly wage in legal employment
x3 is income other than from crime or employment
x4 is probability of getting caught
x5 is probability of being convicted if caught
x6 is expected sentence if convicted, and
x7 is age.

Other factors generally affect a person's decision to participate in a crime, but the list above is representative of what might result from a formal economic analysis. The function, denoting y, depends on an underlying utility function, which is rarely known. — Jeffrey M. Wooldridge, Introductory Economics : A Modern Approach, 4e, page 3.

I would like to address the italicized text. It is important that we not only crunch numbers; but, also try and understand why their values are the way they are, along with knowing how to assign them values. How would you explain the underlying utility function for crime and how would you address it? I'll go first.

Addressing each variable:

x1 is dependent on the socio-political climate and monetary reward for such activity. If the 'wage' is high enough, deterrence is the only solution, assuming that certain people are lured by criminal activity over other factors.

x2 is dependent on socio-economic factors that are often beyond the control of the individual.

x3 depends on prosperity and social spending and amount of effort put into the attainment of money via legal means by an individual.

x4, dependent on spending on policing.

x5 is dependent on laws and politico-ideological motivations in the formulation of certain laws.

x6 is for all purposes the same or similar to x5

x7 is to be assumed that we are only addressing young people who are prone to crime for monetary gain (drug dealing, prostitution, etc.)

How I would try and minimize y:

Maximize x2. This is assuming that we are again talking about x7 being people who are prone to commit a crime for the first time due to poverty.

Maximize x3. In my mind, this is entirely dependent on factors that are both micro and macroscopic. If a person is plain lazy and not motivated to make money the legal way, which is obviously less lucrative than crime, then a social program seems to be required here or some form of public education of the consequences of engaging in criminal activity. Social spending in one word would likely bring the greatest means to increase the value of x3.

x4 or the probability of getting caught is probably the most important and effective means in combating crime. This premise is a strong motivating factor for the inspiration of many books, movies, and politico-philosophical theories of governing.

As for x5, should we follow in the footsteps of China or be more like Sweden in the severity and type of punishment? It seems the rationale in preventing crime according to left wingers and right wingers is tremendous. As a side question, although a very important one and philosophically interesting, why the profound variation between such groups of people?

x6 and x5 are for all intents and purposes similar in light.

x7 has already been addressed.