<< News Scan | Main | News Scan >>

Prisoner Stats


The Bureau of Justice Statistics has released its annual report on prison populations. As usual, the report emphasizes the number of people in prison and what they call the "incarceration rate," i.e., the number in prison relative to the total population. The latter invariably prompts much hand-wringing every time it is announced, even though by itself it is a nearly useless number that tells us almost nothing about policy.

This "incarceration rate" is actually made up of at least two factors which must be determined and considered separately to see anything meaningful. The number of prisoners per capita is the product of prisoners per criminal times criminals per capita. For the algebraically inclined, Pr/Pop = Pr/Cr * Cr/Pop. The first factor is the percentage of criminals society chooses to lock up; the second is the percentage of people who choose to commit crimes. Because the two factors represent different choices by different people, it makes little sense to lump them together, and the undifferentiated product of the two tells us very little. A high "incarceration rate" could mean a society has strict sentencing policies, or it could mean the society is plagued with a high crime rate, or it could be a combination of the two.

The first factor could actually be broken down further as prisoners per criminal we catch times the proportion of criminals we catch. That separation would further separate our ability and determination to catch criminals from our determination to punish the ones we catch.

The situation is further complicated by interrelation of the factors. The probabilities of being caught and of being punished if caught are factors that go into a rational actor's decision to commit a crime. Were the low sentencing rates of the 60s and 70s a cause of the high crime rates of the 80s and 90s and the subsequent high sentencing rates? Quite possibly. The increase in prison population in California from the Three Strikes Law was much less than projected. That may be in part because the law contributed to California's rapidly declining crime rate. A dated but possibly still interesting article on these topics by yours truly and Michael Rushford is available here.

Stand by for a raft of simplistic denunciations that ignore these issues and cite the "incarceration rate" as proof that America is a cruel and heartless society. Bonus points to any reader who finds a single mention of how many people have not been robbed, raped, or murdered because we toughened up sentencing in the 80s and 90s.


One thing that always made intuitive sense to me was the idea that many criminals do what they do "for a living". I would think that most muggings are committed by people who have done it before. Thus, if you lock up a mugger, you prevent a bunch of muggings. Ditto for car thieves etc. I think when you do that, you prevent some people from "graduating" to more serious crimes and you deter people from going down that path in the first place.

The Justice Department also found: The rate of every major violent and property crime measured by the survey fell significantly between 1993 and 2005. The violent crime rate fell 58 percent during that period, and the property crime rate declined by 52 percent. The number of violent crimes decreased from an estimated 11 million in 1993 to 5.2 million in 2005." (Focuses on since 1993 though the crime rate has fallen since its peak in the 70s, but if imprisonment was responsible for 25% of the decline this suggests that 75% of the decline was not due to higher rates of imprisonment.)

Mr. Scheidegger suggests this fall is due to the "subsequent high sentencing rates." However,

In North Carolina, the state’s sentencing commission created a system to use scarce prison space for the most violent and frequent offenders, and to invest in non-prison sanctions for others. Between 1993 and 1997, prison admissions decreased more than half (52%), and from 1994 to 2000, North Carolina’s crime rate fell 12.5%.

This present study on incarceration also finds:

"In absolute numbers an estimated 650,400 inmates in State prison at yearend 2003 (the latest available offense data) were held for violent offenses: 151,500 for murder, 176,600 for robbery, 124,200 for assault, and 148,800 for rape and other sexual assaults (table 12). In addition, 262,000 inmates were held for property offenses, 250,900 for drug offenses, and 86,400 for public-order offenses."

That of course means the other half, about 1.1 million people, are not held for violent crimes.

Perhaps the argument is that the US has more criminals per capita, if so, why and what can we do about it? That liberal question is taboo. Or that we catch and convict more. Still, half of the imprisoned are not sentenced for violent offices.

Evidently the hand-wringing leads to some pocket-digging and more effective cost/benefit solutions.

Hand wringing and pocket digging might reduce incarceration rates of those persons who choose criminal behavior, but are also unwilling to resort to violence or to threaten the physical saftey of their victims.Does ProzacNation contend that hand wringing and pocket digging can quell predatory behavior in the violent offender?

Middleamerican asks an interesting question. Can anything quel "predatory" behavior in the violent offender? Take for example, snipers who shoot abortion doctors. Were there any social forces at work? Another example might be the death threats against SCOTUS and even actual attempts. Were there inciting forces at work? Likewise, can poverty, lack of health care, racism and injustice incite crime? All deep questions that may never have an answer. We are learning though that prison is not the only answer. States are, in fact, finding WHAT WORKS AND WHAT DOES NOT. Rather than digging into their pockets for more prisons, they can reduce crime by spending the money elsewhere. As long as these liberal programs do not become as extreme and unscientific as they were in the 60s and 70s, and has long as they do not become as rigid as the Right's punitopia, they will likely reduce crime, including violent crime, while saving money, and do so more humanely. Polls find the people overwhelmingly want other solutions. Those solutions should be based on what works and what doesn't.

In the study referenced by ProzacNation, it looks as if hand wringing and pocket digging might keep 1.2 out of 10 sex offenders from recidivism. Does that mean that no amount of hand wringing and pocket digging would keep 8.8 out of 10 sex offenders from recidivism? If that's the best stat in the study, does that prove the solution to criminal predatory behavior lies somewhere besides hand wringing and pocket digging?

Leave a comment