Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Violent Crime Rose in 2006, F.B.I. Reports

  1. Dec 18, 2006 #1

    Astronuc

    User Avatar

    Staff: Mentor

    http://www.nytimes.com/aponline/us/AP-Crime.html

    Well this is worrisome. I suppose it's a reflection of the economy.
    http://www.fbi.gov/ucr/prelim06/index.html
     
  2. jcsd
  3. Dec 18, 2006 #2

    russ_watters

    User Avatar

    Staff: Mentor

    It is worriesome precisely because it is not a reflection of the economy.
     
  4. Dec 18, 2006 #3

    jim mcnamara

    User Avatar
    Science Advisor
    Gold Member

    What does it reflect then? Don't leave us hanging.....

    The last dictate (two-five steps lower than theory) on the subject showed correlations between crime and:

    1. increased enforcement of petty infractions (less crime)
    2. shrinking labor supply (less crime)

    This was in the 1990's that the wags liked these ideas.
     
  5. Dec 18, 2006 #4

    russ_watters

    User Avatar

    Staff: Mentor

    I don't know what it correlates to this time. A rise in crime generally does correlate to a stall in the economy (particularly unemployment), but the economy is strong right now. I don't know why it would rise. Philadelphia, though, has been particularly bad this year.
     
  6. Dec 18, 2006 #5
    I don't know about Russ' statement that it's not a reflection of economy or unemployment. There's been a large unemployment rise in 2000-2003, maybe there is a delayed effect on violent crime?

    http://en.wikipedia.org/wiki/Image:Us_unemployment_rates_1950_2005.png

    Certainly 2000-2006 has an overall higher unemployment from 1994-2000 (from the index used in that graph), and you'd think that violent crime would lag behind movement in unemployment (especially robberies). What matters are indexes like - for how long are they unemployed, and how much of the unemployment is due to high turnover or whatnot. And reading all the statistics to make a correlation with crime - and showing that the statistics are real and robust - is not a trivial task.

    We need to find some peer-reviewed work in statistics looking for this correlation. Otherwise, we're all just speculating and blowing hot air. Myself included.
     
  7. Dec 18, 2006 #6
    A lazy start would be here:

    http://scholar.google.com/scholar?q=U.S.+economy+violent+crime&hl=en&lr=&btnG=Search
    Now the problem is to figure out which journals are reputable, and which of the 45,400 papers have useful conclusions.
     
    Last edited: Dec 18, 2006
  8. Dec 18, 2006 #7
    (That's the problem with newspapers reporting the results of studies; the journalists don't really understand them, so they just throw together a bunch of quotes from the "sides" of the issue, and call it an article. Honestly, it's not that helpful.)

    e.g.

    http://www.nytimes.com/aponline/us/AP-Crime.html?_r=1&oref=slogin
     
    Last edited: Dec 18, 2006
  9. Dec 19, 2006 #8

    russ_watters

    User Avatar

    Staff: Mentor

    I wouldn't call 4-6% a large increase in unemployment. By your graph, it is smaller than any of the previous up-cycles, with the exception of a weird blip in 1958.

    I also wouldn't expect it to be that much of a lagging indicator. The logic behind the correlation is that people who are in dire straits commit crime. Based on that, you'd expect crime rates to go up within months of unemployment going up, not years. And it certainly wouldn't rise after 4 straight years of falling unemployment and rising wages.
    Well, year-to-year correlations are probably difficult anyway, but the overall trend over the past few decades is that low unemployment has correlated with low crime. Here's crime for the past few decades (no need to look for articles - the data typically correlates very nicely):

    http://en.wikipedia.org/wiki/Image:Ncsucr2.gif

    Notice the '93 peak in crime correlates very nicely with the '93 peak in unemployment and the y2k valleys corellate as well. The past few years are a little weird, but that may be due to the effects of 9/11, both on crime and the little double-dip recession around that time (which doesn't show up in the unemployment stats). Crime should have gone up more from 2001-2003, but may not have for that reason. So we may just be seeing the settling-out of the 9/11 effect.
     
    Last edited: Dec 19, 2006
  10. Dec 19, 2006 #9
    From that graph, one might conclude that the correlations are too weak to matter, and that other factors are more important. Look at the 1980's, almost flat crime rates while unemployment does some major acrobatics. The crime decrease in the 90's is much bigger than all variation in the 80's, even while the unemployment changes are much more moderate.
     
    Last edited: Dec 19, 2006
  11. Dec 19, 2006 #10

    turbo

    User Avatar
    Gold Member

    Please keep in mind that the government does not tally all unemployed people. It tallies only unemployed people who are trying to collect unemployment benefits and who say they are trying to find work. People who are unemployed and who want to find a job but are ineligible for unemployment benefits are not counted. Unemployment in Maine is a huge problem, but it is under-reported because so many people have overrun the 13-week benefit period and are no longer eligible for benefits and no longer counted.
     
  12. Dec 19, 2006 #11

    Astronuc

    User Avatar

    Staff: Mentor

    And then there is 'underemployment' where people have a job, but do not earn sufficient money to meet all expenses. Many people are just one major illness from losing what little they have.

    If one looks at people who bounce from job to job, or live in subsidized housing, or have go to 'check cashing' businesses (which charge fees of several % of one's check), one get's an entirely different picture of the economy.

    A lot of the economy is built on credit, and that is inherently unstable since the growth rate is barely ahead of inflation. And it seems that inflation rate is low because wages increase at a rate less than the increase in cost of living.
     
  13. Dec 19, 2006 #12

    turbo

    User Avatar
    Gold Member

    And then there are regional variations that affect parts of the population while others are unaware of a problem. For instance, Maine has lost much of its manufacturing job base in the last 10 years. When a big mill closes in a mill town, real estate drops like a rock, making it harder for people to sell their homes and move to another area for work without incurring more debt. Some of these people manage to "get by" by patching together a series of seasonal jobs. The weather has not been cooperative for the last couple of years, though. The weather has been unseasonably warm the last couple of winters, so people who cater to snowmobilers (including people who normally staff motels and restaurants in the winter) have gone unemployed. This last summer, high fuel prices depressed tourism. leaving more people unemployed. Tourism to Baxter State Park and Acadia National Park was down sharply, and the once-popular CAT high-speed catamaran car ferry service from Bar Harbor to Nova Scotia had to cut down on the number of runs due to insufficient volume. These may seem like little things to an outsider, but to people living here, they are important, not only because seasonal employment is a critical part of our economy, but because we residents have to make up any shortfall in state tax revenues, and much of those revenues are derived from taxes on meals, lodging, sales, fuel, etc, which all fall when tourism declines.

    Another reason why crime is on the increase might be the increased popularity of methamphetamine and prescription pain-killers like Oxycontin. These are a bigger problem in rural areas than crack, cocaine, and heroin.
     
  14. Dec 19, 2006 #13
    Yeap, the UK economy is driven by consumer debt, its amazing how much debt people are in in the UK its was at 120% of disposable cash at 2003

    http://www.samuelbrittan.co.uk/spee30_p.html
     
  15. Dec 19, 2006 #14

    arildno

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    Dearly Missed

    I don't think you can ascribe a single source to an increase in murder.

    There are so many reasons why people kill each other, so that a statistics just lumping together all murders is of little use in analyzing societal trends.
     
  16. Dec 21, 2006 #15
    I know wife-beating does.
     
  17. Dec 21, 2006 #16

    russ_watters

    User Avatar

    Staff: Mentor

    I basically agree, though there may be other factors that put limits on it or provide critical levels. Ie, perhaps above 6% crime starts to rise and below it crime starts to fall. I don't know of any logical reason for that, though, so it is just speculation.
     
  18. Dec 21, 2006 #17

    russ_watters

    User Avatar

    Staff: Mentor

    That's true, but it doesn't really affect what we are talking about because if it is a shortcoming in the statistics, it is a consistent one.
    Those factors would show up in household income stats, so that isn't an issue here either.
    True, but irrelevant because:
    Absent numbers to back that up, "barely ahead of inflation" is pretty much meaningless. However, it is flat-out untrue (over timeperiods longer than about 5 years) that wages increase at a rate less than the cost of living. In fact, those two statements are direct contradictions of each other: since inflation is a measure of cost of living, if wages are growing faster than inflation, they are growing faster than the cost of living. Again, I'll post the income stats: http://www.census.gov/hhes/www/income/histinc/h03ar.html

    Even for the lowest income bracket, inflation adjusted household income has risen by slightly less than 1% a year on average over the past 40 years.
     
    Last edited: Dec 21, 2006
  19. Dec 21, 2006 #18

    russ_watters

    User Avatar

    Staff: Mentor

    I'm not certain, but I suspect that most murders are committed during the commission of other crimes. That, plus the shock value of the news/stats is why they are used.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Violent Crime Rose in 2006, F.B.I. Reports
  1. Statistical Crime (Replies: 7)

Loading...