More on Porn and Rape: Does Internet Access Increase Sex Crimes?

(iStockphoto)

Last week we wrote about a new Scientific American Mind cover story that makes the case for a link between internet pornography and lower cases of rape – something we’ve been skeptical of in the past, and remain so today.

A new study from researchers in Norway and the Netherlands offers evidence that suggests the opposite effect, that higher levels of broadband access actually increase the rate of sex crimes.

The study is titled,”Broadband Internet: An Information Superhighway to Sex Crime?” Here’s a full version. And here’s the abstract:

Does internet use trigger sex crime? We use unique Norwegian data on crime and internet adoption to shed light on this question. A public program with limited funding rolled out broadband access points in 2000-2008, and provides plausibly exogenous variation in internet use. Our instrumental variables and fixed effect estimates show that internet use is associated with a substantial increase in reported incidences of rape and other sex crimes. We present a theoretical framework that highlights three mechanisms for how internet use may affect reported sex crime, namely a reporting effect, a matching effect on potential offenders and victims, and a direct effect on sex crime propensity. Our results suggest that the direct effect is non-negligible and positive, possibly as a result of increased consumption of pornography.

As the authors point out, there’s little (if any) causal evidence of the link between internet use and sex crime, and much of what we know is either circumstantial or anecdotal. But this study may finally have some solid evidence. It uses a unique Norwegian data set to study one particular kind of consumption externality: how internet use affects the prevalence of sex crime in general, and rape and child sex abuse in particular. The data come from Norway’s government rollout of broadband across the country in the late 1990s. This gave the authors the chance to conduct a natural experiment, one with a nice before and after picture. Here’s what they found:

Our IV estimates show that internet use is associated with a substantial and statistically significant increase in reported sex crime: Overall, the estimates suggest that about 3.5% of the total number of sex crimes, rapes and child sex abuses that occurred between 2000 and 2008 would have been avoided if broadband internet had not been introduced. The difference peaks in 2006, when we estimate that roughly three out of 50 sex crimes per 100,000 inhabitants would have been avoided if broadband internet had not been introduced. Our analysis also suggests that internet use in 2006 explains about one out of eleven rapes and one out of 22 child sex abuses, per 100,000 inhabitants.

 

Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.

 

COMMENTS: 25


  1. Scott says:

    Very possible that the different cultures US/Norway have different reaction to the same stimuli. Could the more structured Norwegian society react differently to something as expressive as porn than the less structured US society, I would be surprised if the reactions were the same.

    Hot debate. What do you think? Thumb up 12 Thumb down 8
  2. Missinlnk says:

    How do they account for unreported sex crimes? I can see a situation where having access to the internet would let a victim find out that they were not alone and would give them the strength to report the sex crime. Are they sure that the problem wasn’t that sex crime was previously underreported due to a local culture that discouraged reporting?

    Did they address this somewhere in their study?

    Well-loved. Like or Dislike: Thumb up 32 Thumb down 3
    • Shadist says:

      That was the very thing I was thinking. A potential connection between a greater connection to the world at large allowing for more people to report the crimes.

      Thumb up 3 Thumb down 2
    • Tarjei Havnes says:

      We estimate the effect on reporting directly, looking at 1) the ratio of charges and convictions to reports and 2) the time between alleged crime and report, finding no evidence of changes in reporting behavior.

      Well-loved. Like or Dislike: Thumb up 9 Thumb down 1
  3. Ian M says:

    I’m sure broadband was the only significant change during that time period.

    Do the recent events in Norway not indicate that there were other significant social changes in the country?

    Thumb up 4 Thumb down 3
  4. J. Smith says:

    Sounds exciting, but when you look at the real numbers of cases under discussion in the full article and the graphs in the appendix, it turns out the increase of such crimes in all of Norway (with a population of roughly 4,8 million) if broadband internet access existed everywhere would amount to 20-25 a year. However, broadband internet doesn’t exist everywhere. So in real life terms, the impact of this stunning phenomenon amounts to maybe 15-20 cases more Norway-wide, which the authors blame on maybe (note… they don’t know what the perps were doing on the internet, just that they had access), maybe, maybe a connection between porn and sex crime. Stunning science.

    Well-loved. Like or Dislike: Thumb up 16 Thumb down 7
    • Tarjei Havnes says:

      Wherever did you get that number? Even from the excerpt above and your own post, “three out of 50 sex crimes per 100,000 inhabitants” in 2006 for a population of 4.8 million suggests an impact of 3*48 = 144 sex crimes in 2006. More importantly, this suggests that the increase in internet use from new broadband access points explains about 5% of sex crimes, including 10% of rapes and 5% of child sex abuses in 2006. That seems rather important to me.

      We have no strong opinions on what is driving this directly — reporting, matching or sex crime propensity — but we provide evidence suggesting that 1) reporting does not change, 2) effects are weaker in areas with better access to pornography offline, and 3) there are no effects on other types of crime that typically covary with sex crime. This does indeed lend support to the hypothesized link between pornography consumption and the risk factor for sex crime.

      Thumb up 5 Thumb down 1
  5. Neil (SM) says:

    This reminds me of the “Global average temperature vs. Number of Pirates” graph.

    Well-loved. Like or Dislike: Thumb up 18 Thumb down 6
  6. Matthew whitehead says:

    I’m wondering if the evidence is based on an increase of cases per capita or an association with Internet porn in the criminal investigation?

    Thumb up 2 Thumb down 1
  7. Alexandre Padilla says:

    All these studies focus on Internet access and sex crimes, none of them actually measures the effect of increased access to pornography on sex crimes. Increased access to Internet increases anonymous access to porn but saying that therefore increased access to porn leads to an increase or decrease of sex crimes is jumping the gun. I don’t see in that paper or Kendall’s paper an actual dataset discussing porn consumption by states. There is no data set or survey discussing whether rapists consumed porn (and even then, one could argue that rapists would use such defense in court so you can’t trust them. You would therefore need to have forensic evidence that rapists consumed porn and what type of porn. The implied conclusion of the paper that we should regulate porn to reduce sex crimes shows a complete ignorance of basic cost/benefit analysis, economic efficiency, and the possible unintended consequences that such regulation would create. Ever heard of prohibition or war on drugs? This is in my opinion an evidence that the others let their personal morality guide their research. They should have stop at Internet & sex crimes and let porn out of the picture.
    Last thought, increased Internet access has also let to an increased in online chatting, dating sites, why not imply that we should also regulate those sites? Why just porn?

    Thumb up 6 Thumb down 4
    • Tarjei Havnes says:

      I think it is controversial, to say the least, that one should not try to explore the potential mechanisms behind reduced form evidence. A link between porn and sex crimes has been proposed, both negative and positive, and we try to take it to the test by estimating separate effects in areas with better access to pornography offline (border to porn-liberal Sweden). Weaker effects in these areas compared to other areas suggest that a positive link is more likely.

      Thumb up 4 Thumb down 2
  8. Sarah says:

    Hidden due to low comment rating. Click here to see.

    Disliked! Like or Dislike: Thumb up 3 Thumb down 19
  9. Enter your name says:

    It would be interesting to run a different study: what’s the likelihood that a man who’s charged with a sex crime has used a credit card to pay for online porn vs a similar (age, race, income) man?

    I’m sure that the credit card companies could get that data, and this would be a more direct connection than access to the internet in general.

    Thumb up 4 Thumb down 2
    • Michael says:

      That wouldn’t really work, because there is a lot of free pornography on the internet, so that would miss that point, plus I am pretty sure most of these pedophile rings don’t charge for “membership”.

      Saying that makes me think that it’s interesting to note the validity of this report in that what correlation does pornography to say YouTube showing clips of Hannah Montana shows produces. It’s not entirely accurate to suggest that internet “porn” is the cause, unless you have the IP of every site and every person that connects to it in the area you are recording.

      Sounds to me the new version of “does TV make you violent”? Didn’t we go through this in the 80′s? Michel Heneke has done films relating to this very subject.

      And don’t forget about video games – did the study suggest some games online has the same effect.

      Oh my, the list can go on and on and on and on…

      Thumb up 3 Thumb down 2
  10. Santosh Anand says:

    Does the study take into account the racial makeups of the areas where the crime occurrences were measured.
    Something tells me broadband access would not have had any effect in homogeneous Norwegian areas.

    Thumb up 2 Thumb down 5
  11. Sarah says:

    I agree with the ‘enter your name’ comment tying the charges to the person, seems the most balanced comment, however, how would they guage potential fraud or using others’ identity info? Obviously to me some would try to be annonymous.

    Thumb up 0 Thumb down 3
  12. Consumer says:

    Last week’s article drew a connection between Internet PORN and decrease in sex-related crime. This article is about a connection between Internet ACCESS and increase in sex-crime. But what is the relationship between ACCESS to internet and PORN usage? Is there any?

    Thumb up 2 Thumb down 2
    • Tarjei Havnes says:

      I think you’re misunderstanding: the study considers the link between internet use and sex crimes (similar the article discussed last week). But we are concerned that internet use depends on some (unobserved) factors that can also be related to sex crimes. We therefore use a staggered expansion in internet access created by a public program to extract variation that should not be correlated with sex crimes via other things than internet use. That the internet makes pornography more readily available is uncontroversial. In particular since explicit pornography was banned in Norway over this period.

      Thumb up 4 Thumb down 1
  13. Sarah says:

    Or simply using others internet, people who share accounts or if someone uses a connection unbeknownst?? Skewed, questionable results?

    Thumb up 0 Thumb down 4
  14. Eric M. Jones. says:

    Can I predict that we’ll be discussing this or something quite similar 10,000 years from now.

    Oh, those Victorians…they took all the bad words and sexual jokes out of the plays of Aristophanes. How’d THAT work out…? And the British Museum has a dark storehouse for pornographic ancient drawings and statues, even Egyptian smut….even topless Minoan babes.

    If it weren’t for porn there’d be no VCRs, DVDs, and who knows what else.

    They made us all safer.

    It reminds me of my favorite Bible passage (Ezekiel 23: 21):

    “..Yet she became more and more promiscuous as she recalled the days of her youth, when she was a prostitute in Egypt. So she longed for the lewdness of her youth, her body was caressed and her young breasts fondled. She lusted after her lovers there, whose Johnsons were like those of donkeys, and whose hot spunk exploded like that of mighty stallions.” Amen….

    Thumb up 1 Thumb down 2
  15. Trent says:

    I didn’t get a chance to read the entire study, but did they differentiate between sex crimes against children and child pornorgraphy? Please don’t get me wrong, both are child abuse and are beyond horrible, but I don’t think this study is offering anything overly surprising if it is showing that an increase in internet access led to a proliferation I’m child pornography.

    Thumb up 2 Thumb down 2
    • Tarjei Havnes says:

      Child sexual abuse includes only actual offenses against children. All crimes related to pornography consumption and distribution is taken out.

      Thumb up 3 Thumb down 0
  16. JustSayin says:

    I think it’s from all the pirating of porn movies. PIPA and SOPA will fix this as well. Downloading free content is a gateway crime to more serious crimes.

    http://www.freakonomics.com/2012/01/12/how-much-do-music-and-movie-piracy-really-hurt-the-u-s-economy/

    Thumb up 0 Thumb down 0
  17. Bob says:

    Two things are interesting to note.

    1) During this time frame prostitution was legal in Norway. This possibly eliminates the primary mechanism through which the availability of pornography may decrease rape in that an alternative outlet for sexual release already existed in Norway.

    2) There are very few reported rapes in Norway. At the beginning of their sample there was less than one rape per municipality. When you remove municipalities that have no rape, their results on rape completely disappear. Does that mean the interpretation should be: pornography makes it less likely to have zero reported rapes in a municipality; but pornography doesn’t increase the amount of reported rape in a municipalities having nonzero reported rape.

    You can see this again in that their results get even stronger when they exclude large cities. This suggests large cities face no increase in rape with an increased availability of pornography.

    Due to this I feel like their paper actually suggests that we should not expect pornography to cause an increase in rape in most areas in the United States.

    Thumb up 0 Thumb down 0