4 ms·

Beware of cherry picked data. It's not statista doing it, but whoever they sourced their data from choosing to start at 1990. Here [1] are the same data (from t

Beware of cherry picked data. It's not statista doing it, but whoever they sourced their data from choosing to start at 1990. Here [1] are the same data (from the FBI) from 1960, instead of 1990. 1990 was not a historical rate, but the result of an exceptionally sharp spike in crime that began in 1960. In 1960 motor vehicle theft rates were 183. The statista table (which goes until 2021 instead of 2019 from my source) shows today they're at 268 and trending upwards. I really wish I could find data prior to 1960, but I guess you probably end up with a poor signal:noise ratio at some point. Those data are all from the same source, so have meaningful comparability. In any case, the long term trends are anything but clear. Crime rates currently look something like an eccentric sine wave. People in 200 years sure will have it easier, data wise. [1] - https://www.disastercenter.com/crime/uscrime.htm https://www.disastercenter.com/crime/uscrime.htm

Retric 3y ago
Obviously the rate was 0 in 1850 because cars didn’t exist. You need a steady state of car ownership for theft comparisons by population to make sense. More recently the number of cellphone thefts spiked as more people owned cellphones, its true but not an accurate reflection of overall crime. Historically you can do numbers of thefts vs numbers of cars but that’s not a great. There was 61.7m passenger cars + 12.7m commercials vehicles in the 1960’s vs 140.7m passenger cars + 42.8m commercials vehicles by 1988. An increase dramatically faster than population. https://www.statista.com/statistics/1246890/vehicles-use-united-states-historical/ https://www.statista.com/statistics/1246890/vehicles-use-uni... From 1990, the numbers increase roughly with population which is why many comparisons start in 1990: https://www.statista.com/statistics/183505/number-of-vehicles-in-the-united-states-since-1990/ https://www.statista.com/statistics/183505/number-of-vehicle... 328,200 / (61.7m + 12.7m) = 0.44% of vehicles stolen in 1960. 1,635,900 / 193.1m = 0.85% in 1990 is a big jump. But that then falls all the way to 2019’s 721,885 / 276.5m = 0.26% in 2019 sitting just over half of 1960’s rates. PS: It’s perfectly reasonably to argue that what matters is rate over the population, but including non owners in the population is IMO misleading.
- somenameforme 3y ago
  By 1960 the overwhelming majority of households in America already had cars. Increasing number of cars beyond that (to modern times where many households have 2 or even more cars) is unlikely to have played a significant role. This is indirectly corroborated by the fact that near to every other form of crime was much lower in 1960 than modern times, as well. The two exceptions being burglary, whose levels we fell below and have stayed below since 2014 (well at least as of 2019 data), and murder which was just statistical noise. Historically what happened is that crime started rising in the mid 60s and then reached catastrophic levels in 80s-90s to the point that many were predicting the complete collapse of society. This is what drove all of the 'tough on crime' rhetoric and legislation of that era, or a certain politician's choice of referring to a certain demographic group as 'super-predators.' Crime rates then began plummeting over the next 30 years into modern times. And now we're back to seeing a rise in many forms of crime. What remains to be seen is whether it's noise, or whether we're headed 'into the 80s' again. Or maybe 30 years from now we'll learn that we've all poisoned ourselves inadvertently, again.. The lead-crime hypothesis [1] is one explanation for the absurdly high rates of crime in the 80s-90s, and then it's subsequent collapse. [1] - https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis
  - Retric 3y ago
    22% of households still didn’t have cars in 1960. We can get into why households is a poor metric but 1960 was simply a low point not a long term average. Murder per 100k. 1960 5.1, 2014: 4.5, 2019 5.0. Go back to 1950: 5.3, 1940: 6.3 and it hit 9.5 in 1934. PS: Lead is one factor post 1960’s, but changing demographics is another. The post war baby boom adds a lot of kids who aren’t killing anyone as preteens/young teens but average into population statistics. The low point of births during WWII is especially important when those years line up with older teens more likely to commit crimes. Add in a good economy and it’s a perfect storm for low crime.
    - somenameforme 3y ago
      I would not be entirely surprised if the homicide rate was higher during the Great Depression and WW2, but I'm not really sure what that means, if anything. Please cite the data you're referencing, and stop cherry picking. I don't expect you to get decadal averages or whatever, but at least get something pretty averagish looking for the 'group' of time we're talking about. Comparing local mins to local maxes is something we ought leave to lying politicians. Yeah the vehicular ownership rate only increased from 80% or whatever in the 60s to about 90% now a days. But I'm not saying that's why the increase in vehicles probably isn't a big factor. What I'm saying is that vehicular theft is a very intentional crime. Crimes like phone theft are highly opportunistic - somebody sees a fancy phone somebody forgot, and just decides to pocket it. So more people have more phones means more crime due to more random opportunity. But car theft is overwhelmingly a targeted crime, so changes in the number of cars is going to have a pretty negligible effect above some baseline point. It's like bank robbery - you're not going to just inherently have more bank robberies per capita because there are more banks.

Retric 3y ago

Obviously the rate was 0 in 1850 because cars didn’t exist. You need a steady state of car ownership for theft comparisons by population to make sense. More recently the number of cellphone thefts spiked as more people owned cellphones, its true but not an accurate reflection of overall crime. Historically you can do numbers of thefts vs numbers of cars but that’s not a great. There was 61.7m passenger cars + 12.7m commercials vehicles in the 1960’s vs 140.7m passenger cars + 42.8m commercials vehicles by 1988. An increase dramatically faster than population. https://www.statista.com/statistics/1246890/vehicles-use-united-states-historical/ https://www.statista.com/statistics/1246890/vehicles-use-uni... From 1990, the numbers increase roughly with population which is why many comparisons start in 1990: https://www.statista.com/statistics/183505/number-of-vehicles-in-the-united-states-since-1990/ https://www.statista.com/statistics/183505/number-of-vehicle... 328,200 / (61.7m + 12.7m) = 0.44% of vehicles stolen in 1960. 1,635,900 / 193.1m = 0.85% in 1990 is a big jump. But that then falls all the way to 2019’s 721,885 / 276.5m = 0.26% in 2019 sitting just over half of 1960’s rates. PS: It’s perfectly reasonably to argue that what matters is rate over the population, but including non owners in the population is IMO misleading.

somenameforme 3y ago
By 1960 the overwhelming majority of households in America already had cars. Increasing number of cars beyond that (to modern times where many households have 2 or even more cars) is unlikely to have played a significant role. This is indirectly corroborated by the fact that near to every other form of crime was much lower in 1960 than modern times, as well. The two exceptions being burglary, whose levels we fell below and have stayed below since 2014 (well at least as of 2019 data), and murder which was just statistical noise. Historically what happened is that crime started rising in the mid 60s and then reached catastrophic levels in 80s-90s to the point that many were predicting the complete collapse of society. This is what drove all of the 'tough on crime' rhetoric and legislation of that era, or a certain politician's choice of referring to a certain demographic group as 'super-predators.' Crime rates then began plummeting over the next 30 years into modern times. And now we're back to seeing a rise in many forms of crime. What remains to be seen is whether it's noise, or whether we're headed 'into the 80s' again. Or maybe 30 years from now we'll learn that we've all poisoned ourselves inadvertently, again.. The lead-crime hypothesis [1] is one explanation for the absurdly high rates of crime in the 80s-90s, and then it's subsequent collapse. [1] - https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis
- Retric 3y ago
  22% of households still didn’t have cars in 1960. We can get into why households is a poor metric but 1960 was simply a low point not a long term average. Murder per 100k. 1960 5.1, 2014: 4.5, 2019 5.0. Go back to 1950: 5.3, 1940: 6.3 and it hit 9.5 in 1934. PS: Lead is one factor post 1960’s, but changing demographics is another. The post war baby boom adds a lot of kids who aren’t killing anyone as preteens/young teens but average into population statistics. The low point of births during WWII is especially important when those years line up with older teens more likely to commit crimes. Add in a good economy and it’s a perfect storm for low crime.
  - somenameforme 3y ago
    I would not be entirely surprised if the homicide rate was higher during the Great Depression and WW2, but I'm not really sure what that means, if anything. Please cite the data you're referencing, and stop cherry picking. I don't expect you to get decadal averages or whatever, but at least get something pretty averagish looking for the 'group' of time we're talking about. Comparing local mins to local maxes is something we ought leave to lying politicians. Yeah the vehicular ownership rate only increased from 80% or whatever in the 60s to about 90% now a days. But I'm not saying that's why the increase in vehicles probably isn't a big factor. What I'm saying is that vehicular theft is a very intentional crime. Crimes like phone theft are highly opportunistic - somebody sees a fancy phone somebody forgot, and just decides to pocket it. So more people have more phones means more crime due to more random opportunity. But car theft is overwhelmingly a targeted crime, so changes in the number of cars is going to have a pretty negligible effect above some baseline point. It's like bank robbery - you're not going to just inherently have more bank robberies per capita because there are more banks.