Criminal Careers and "Career Criminals," Volume II (1986) / Chapter Skim
Currently Skimming:
9. Dynamic Models of Criminal Careers
9. Dynamic Models of Criminal Careers
Pages 356-379
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 356... ...
Second, all models recognize that risk is an essential component of the decision to engage in criminal activity. In contrast to the purchase of a can of soup, which has a virtually certain level of ultimate satisfaction associated with consumption of the procluct, the eventual level of satisfaction associated with the decision to undertake criminal activity can only be clescribecI probabilistically.
|
From page 357... ...
The converse is also obviously true; current empirical knowledge regarding the dynamics of criminal careers must be used as a guicle in the construction and evaluation of any theoretical model that purports to describe the criminal activity clecision over time. Structural models of decision making also serve a relatecl purpose.
|
From page 358... ...
Following the second option, we wouIcI CRIMINAL CAREERS AND CAREER CRIMINALS estimate a function of the form x = able, b; R
|
From page 359... ...
The correct point is that current attempts at behavioral modeling of criminal behavior using the expected-utility-maximization principle are unquestionably overly simplistic. Realistically, to capture the dynamics of criminal behavior adequately, structural models will have to evolve substantially.
|
From page 360... ...
This func CRIMINAL CAREERS AND CAREER CRIMINALS tional relationship is expresser]
|
From page 361... ...
1, .... We will also begin by assuming that conditional on apprehension, sentence length is the same for all inclividuals, regardless of criminal history, so (Oh _ ~)
|
From page 362... ...
. CRIMINAL CAREERS AND CAREER CRIMINALS When we combine all the elements discussed above, the maximum value of the individual's time allocation problem in all periods when he is not incarcerated as of the beginning of the period is given by V= max :~1 - P(~)
|
From page 363... ...
result in decreases in crime rates. An increase in the direct opportunity cost of crime, the wage rate in legitimate work (w)
|
From page 364... ...
In the model in the next section, criminal activity decreases, on average, as individuals age. Accumulation of Human Capital in Legitimate Activity Using the constant-wage model, the proportion of time nonincarcerated indi CRIMINAL CAREERS AND CAREER CRIMINALS viduals devote to criminal activity remains constant as they age.
|
From page 365... ...
Then, the total amount of market experience the individual has at the beginning of period t is ~ (1 - ski k = ~ Market-related human capital is assumed to be a simple transformation of market experience, kit= g: -1 ~ (1 - Ski k= 1 , (12) where g is a monotonically increasing function; human capital is increasing in labor market experience.
|
From page 366... ...
~ i = 0 hi + p7V(w') 11 Now the inclivicluaT's time-allocation problem depends on the set of parame CRIMINAL CAREERS AND CAREER CRIMINALS ters in the constant-wage model, plus the wage-growth parameter a.
|
From page 367... ...
An individual arrested in period 20, however, may be classified with virtual certainty as a high , , , 367 TABLE 1 Time Allocation to Criminal Activity Given Wage Growth with Constant Sentence Length Wage Level Low-Crime Types High-Crime Types 0.5 o.ss 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1.0 1.05 1.1 1.15 1.2 1.25 .4305 .3726 .3082 .2370 .1590 .0747 0.0 .6007 .5774 .5524 .5258 .4972 .4663 .4330 .3969 .3579 .3156 .2698 .2205 .1675 .1111 .0515 0.0 crime type. An even more accurate classification can be macle if the wage rate of the arrested inclividual is available.
|
From page 368... ...
As with the constant-wage model, we assume that legitimate market wages w are constant over time so that we can isolate the sentencing effect. Previously, we assumed that sentence lengths Were constant, which is obviously not the case in practice.
|
From page 369... ...
An example similar to the one used above illustrates the characteristics of criminal careers generated by this model. All parameter values are exactly the same with the exception of the sentence length a.
|
From page 370... ...
Let us consider the variable-sentence CRIMINAL CAREERS AND CAREER CRIMINALS length model presented above as an example. The model win increasing sentence lengths is described by the following set of parameters: 71, w, 3, A, c*
|
From page 371... ...
, even if we assume that criminal rewards are not measurable or even operationally cleanable. ECONOMETRIC MODELS OF CRIMINAL CAREERS The dynamic moclels of the criminal career developed above are based on optimizing behavior.
|
From page 372... ...
Or the fact that an attractive "mark" appears may induce a crime before we would normally expect one. Unanticipated or anticipated changes in the choice sets of individuals will cause variations in criminal behavior over the life ~i`7~n hat =` ~ CRIMINAL CAREERS AND CAREER CRIMINALS cycle, as the preceding section demonstrated.
|
From page 373... ...
In the case of the pure, unconditional renewal process first described, the den The hazard function is the conditional density of duration times given the individual has not committed a crime for a period of length w. The hazard function h is used in the econometric model formulated below.
|
From page 374... ...
controls for observed and unobserved CRIMINAL CAREERS AND CAREER CRIMINALS heterogeneity in the population by parameterizing the hazard function in a general way. If we assume that spell lengths for an incliviclual are i.i.~.
|
From page 375... ...
The one-state renewal moclel can be generalized in several ways that may prove useful in the study of criminal careers. The assumption that the criminal career is a conditional renewal process (i.e., conditional on other exogenous stochastic processes)
|
From page 376... ...
Up to this point we have assumect tnat only one type of crime is committed in CRIMINAL CAREERS AND CAREER CRIMINALS r.l l l the population or, at the least, that each individual commits only one type of crime, although different individuals may specialize in different crimes. It is rela tively straightforward to generalize the econometric model presented above to cover the possibility of crime switching when each individual may commit any one of a number of types of crimes.
|
From page 377... ...
.., K, which correspond to the latent duration densities from initial entry into the population at risk of committing a crime, which we will denote by state 0, until a crime of type j is committed.) For each latent density Gil, i = 0, 1, .
|
From page 379... ...
Thus,.the fact that an incliviclual facing a 1-year sentence if caught committing a crime will face stiffer sentences in the future if caught committing acIditional crimes will, in general, affect criminal behavior at all points over the life cycle. The static models usually employed in empirical research are not capable of capturing these dynamic deterrence effects.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.