Do more convicts = less crime?

This is a follow on to the prison population question/column.

Do more convicts = less crime?

Rates of crimes following enactment of “Three strikes” laws seem to indicate so, especially in California.

Just providing a link to the colum in question. Carry on.

Does the United States lead the world in prison population?

Here’s an article that starts off with some basic stats.

In short, in the 1980s the prison population doubled and the crime rate went up. In the 1990s the prison population nearly doubled again and the crime rate went down. It may be counter-intuitive but throwing a lot of people in prison has very little effect on the crime rate. That’s because the crime has already happened.

If you want to change the crime rate you have to prevent crime from happening in the first place. There are various champions of different methods for this. E.g., do a “Guiliani” and throw anyone in jail for minor offences. Do “something” about illegal drugs (legalize, provide treatment, whatever). Reduce the poor economic opportunities that results in helplessness that leads to crime. (The later having been declared an “undiscussable topic” for over 20 years, despite the evidence of the last 10 years.)

What is this? Some sort of “lump of crime” fallacy?

All the data indicates that crimes are not committed by random member of the population. Rather, a small percentage of the population is responsible for a hugely disproportionate fraction of crimes committed, especially violent crimes.

So if you lock up people who commit violent crimes and never let them out, one of two things will happen. Either fewer crimes will be committed in the future (since habitual offenders are permantently banged up), or previously law-abiding citizens, sensing the opportunity for a lucrative career change, will rush in to fill the gap and keep the crime rate up.

This is not an endorsement of California’s three-strikes law, which is absurdly draconian in far too many cases.

The key thing is that crime has been trending downward over the last decade almost everywhere in the US, not just in places with 3 strikes laws. And most of the US’s peer countries have a much lower incarceration rate and a much lower crime rate. Ergo, the incarceration rate is not the sole determinant of the crime rate.

Truth Seeker, you’re committing a “lump of crime” fallacy yourself. If convicted criminals are taken out of the population, it’s not at all necessary for previously law-abiding citizens to start breaking the law to keep the crime rate constant: all that’s needed is for the criminals not behind bars to expand their operations. (In the case of things like drug trafficking and illegal gambling this is obviously exactly what would happen.)

The dubious smiley isn’t working but this’d be a perfect place to use it.

Most crime doesn’t model like an ordinary economic activity. You’re assuming that there is some unfilled consumer need out there and that locking up some “producers” simply stimulates others to enter the market.

This might work, partially, for drug dealing. (Maybe illegal gambling, too, though I don’t believe that prisons are exactly overflowing with bookies and people running numbers games.) But it doesn’t work at all for violent crime. There is no consumer demand for rape, murder, robbery, etc. Anyone who wants to become a mugger can become a mugger. The number of active muggers is not limited by competition for potential victims, nor is their “return” influenced by the number of other people who have taken up mugging.

To put it another way, no one decides not to become a mugger because they think there is too much competition in the mugging field.

I think all the good mugging jobs are going overseas.

I’d think the real question is whether we’re jailing the RIGHT PEOPLE. You can’t blindly link incarceration rates with crime rates, since both can be completely arbitrary. Police are throwing pot-smokers in jail by the hundreds of thousands, yet far less burglers are jailed. (especially compared to the number of burglaries committed) Would any reasonable person say that crime is going down, when non-violent personal-use offenders are locked up and burglers are left free?

The FBI’s Uniform Crime Reports ( are an excellent place to go for getting better numbers.

I have no doubt that jailing the right people would cause the crime rate to diminish. Robbers, rapists, murders, etc. But our system, currently, tends to encourage both police and prosecutors to go for the “easy kill” rather than the useful one. And the current directives from Ashcroft encouraging DAs and judges to, essentially, jail anyone who comes before them is little help. So that’s what we need; smarter convictions rather than more convictions.

The damn Free Trade Agreement outsourced my career as a pickpocket to Thailand. But the weak US dollar is making it much more profitable to fence stolen BMW’s these days.

Who knew that Alan Greenspan had more influence on the crime rate than John Ashcroft?

Then there is the theory, that since abortion was legalized in the U.S. some 30 years ago, fewer unwanted babies are born, so fewer children are neglected, and fewer raised in a style that leads to a criminal life. Ergo, the crime rate drops, independently of the rate of incarceration or other factors.

At the risk of stating the obvious, it depends on who you lock up. If you are talking about violent crime, and locking up people who have committed violent crime then it will probably have an effect on any repeat violent crime they might commit.

On the other hand, if you are talking about violent crime and you are locking up a bunch of people for smoking pot, then you probably haven’t done much about violent crime. That is the case in California, where the most frequent cause for a “third strike” is a non-violent marijuana conviction.

Also, if you look at crime rates within California and compare jurisdictions that have exercised three strikes with great regularity, versus those that haven’t, you will find that jurisdictions that don’t use it much have equal or better drops in crime as the ones that do.

As others have suggested, if you really have a beneficial effect on the crime rate, the thing to do first is change the drug prohibition laws. But, of course, that’s the same lesson we should have learned for all time because of the huge increase in homicides during alcohol prohibition.

Exactly. Roe vs Wade happened in 1973; US crime started dropping in 1991, 18 years later, just when you would expect that cohort to enter the criminal milieu. The increase in abortions was mostly in the social classes from which small-time criminals who commit the numerical majority of crime come. QED.

[INDENT]The prosecutors don’t seem to care who is guilty or not, just so long as they convict somebody Even when they have exculpatory evidence, their 99.9% conviction rate is all that matters.

When the prosecutor has no real evidence of the charges, that’s not a problem. They rack up a whole bunch of additional charges against the defendent and tell him if they get a conviction on these charges, the poor dude is looking at a mandatory 20-30 years of hard time. So the innocent schmuck cops a plea to lessor charges rather than run the risk of a life long jail sentence. He does a nickle even though he didn’t do the crime and the DA gets another notch on his belt.

If the bogus charges don’t work, the prosecutors have no shortage of people in prison willing to testify in court and commit perjury in exchange for 24 months being dropped from their ten years sentence.

The prosecutors don’t need to worry about being held accountable for their reprehensible actions. Even if they acted in a criminal fashion, they have “qualified immunity” so long as they can claim they were acting within the scope of their job. No criminal charges or civil action can be filed against them. That really sucks!

What are the estimates now for the percentage of death row inmates that are innocent? 10%? Our justice system appears to be anything but…

I taught my two kids that we have a civic and moral duty to “Question Authority”. Not any more. Now I’d tell 'em "FK Authority"** instead.[/INDENT]

This is under the assumption that abortion was illegal before Roe vs. Wade. Hypothesis fails.


“Assumption”? Vaz you dere, Sharlie?

That’s the least problematic assumption:

#1 - People somehow hold off on committing crimes when they are 18. (Try 13)
#2 - People who commit crimes and were born more than 18 or so years ago somehow stop committing crimes.
#3 - The poor/unwanted children are commiting the majority of the crimes. (On all three of these assumptions, just hang around in my office for any length of time to show the problems underlying those assumptions).
#4 - Abortions are primarily used by the poor or those unable to care for the child. (IIRC, statistics show most abortions are a product of middle class, suburban homes, and the parent undergoing the abortion often have children who become well-adjusted, productive, non-criminal citizens).
#5 - Crime rates have not fallen uniformly across the country. Somehow, non-aborted children were somehow disporportionately spread in some, mostly urban areas. (IIRC - abortions were most often legal pre-Roe in Urban areas of the northeast, some of the places that had the biggest crime rate drops. If the hypothesis held, their crime rate drops should have occurred years earlier).
#6 - It doesn’t explain the dramatic rise in crime rates in the earlier decades with no ties to changes in abortion laws.

There is also a hugely problematic ethical conclusion that may be drawn from the hypothesis - that criminals, particularly habitual criminals, are somehow missed abortion opportunities. Mucho disturbing.

Perfect example here. If you check the crime rates for 14-17 year olds, the first group of teens born after the legalization of abortion were the most criminal on record -BOJ Stats notice the spike during the early '90s.

Another good example - the mid 90’s crime drop started to occur among older groups first, the reverse of what you would expect if abortion were reducing crime. And a lot of the murders that occured during the crack wars were between rival gangs; this snuffed out the life of many a young criminal before they could do anything know.

Another note is that the idea that abortion increased the “wantedness” level of children, looks absurd when you look at the illegitimate rates after Roe v Wade. Among African Americans, the illegitimatcy rate grew from 30% in 1970 to 50% a decade later. Seems like there were a lot of fathers who didn’t want their children (or at least, couldn’t have a shotgun marriage performed.

I can’t give any statistics immediately, but I will after I go to the library. The library is a place where they have books. It’s not a “cite” where you click on something. You have to go there physically (“what a drag!”), and check out a book, or two, or three. I know it seems strange to many, but these books show you that locking up minor drug criminals in a state that has one of the hightest rates of incarceration in the world, California has done little to reduce crime, and especially crime related to drugs.

I apologize if you actually have to go to the library. I’m very sorry. I know books are a drag. You actually have to carry them. And read them. Sorry.

Well attributing the drop in crime to Roe is a bit of a stretch, although I haven’t read the book (Freakonomics) so I won’t pass judgement.

The main reason crime rates went down in the 90s was that there were simply fewer people in the age group responsible for the most crimes, 18-35. This could have to do with Roe v. Wade or any number of factors. The crime rates went up in the 60s (whether due to the baby boomers’ numbers or innate mendacity is a different issue.)

The book that saoirse is referring to is discussed here.

Let me tackle some of the points made here already, although I don’t have the book in front of me:

Huh? It was legal in some jurisdictions, illegal in others.

I have no idea where you came up with assumptions 1 and 2. I have no idea where your office is. I do know that when discussing statistics, anecdotes you might have are worthless. Look, on a statistical basis it is quite safe to argue that criminals can start before age 18, but they tend to start with property crimes; only later do they move into violent crime. The book is clear to distinguish between the two.

I’d love to see where you saw this. No, check that. I don’t care. What the book (and an earlier study partly written by the book’s coauthor) says is that Roe vs Wade essentially legalized abortion for the lower class; middle- and upper-class women could more easily afford the travel to jurisdictions that allowed abortion, or at least they could afford illegal abortions.

But aside from the general statement that “Crime rates have not fallen uniformly across the country” (which is true), you don’t recall the particulars correctly. In the five states where abortion was legal at least three years before Roe v Wade, the crime rate began to drop three years before it began to drop nationwide, and in roughly the same proportions. Furthermore, the authors found a correlation between crime rates and the availability of abortion in the period 1985-1997–even after Roe vs Wade it was virtually impossible in some states like North Dakota and some areas in the Deep South to get an abortion.

Nope, it doesn’t try to, because the causes of crime are so myriad. I’ll leave it to the statisticians, sociologists, and criminologists to understand the factors that cause crime. But I think I’m safe in assuming that there are economic factors to crime rates, political factors, sociological, meteorological (it’s a truism in police work that crime rates rise in the summer), and demographic.

In fact, critics of Levitt and Donohue are quick to jump on one phenomenon to argue against their hypothesis: the crack cocaine epidemic of the late 1980s and early 90s. The cite posted by RandomLetters is a good example of the spike in crime, especially violent crime, during that time period. But you have to look closer before dismissing the Levitt/Donohue hypothesis. They do an analysis on a state-by-state level and still see their correlation, despite the obvious surge in crime caused by an apparently idiosyncratic event.

In other words, Roe vs Wade couldn’t have prevented the crack epidemic, but the authors are still able to correct for its effect. Hypothesis still stands, because it’s been replicated among different states and even in different countries.

Look, sorry I’m so pissy, and I can’t really go any further with cites, and I’m not even a trained statistician, but I am at work and I’ve wasted an hour typing this response out and I want to go home for the weekend. At this point, I’d simply suggest that you read the book. For a quick version of the crime and abortion story, go to the authors’ blog. And whatever you do, please try not to view it through your political/ethical/moral prism; the authors express their distinct unease at using their findings to argue the pro-choice side, and I (a pro-choicer) sure feel similar.