Most people think the U.S. crime rate is rising. They're wrong

So much knowledge of various places (downtown of my city, for example) is not first hand, but from (social) media.

I'm reminded of this video from Bowling for Columbine, where they stroll around South Central LA, and realize its pleasant and nothing crazy is happening.

https://vimeo.com/39131010

Even without any nefarious motivations, simple survivorship bias will mean any form of media will report on the exceptions to the rule, not the rule itself. Don't substitute media input for first hand knowledge.

Since Covid+WfH, I've felt this acutely and had to remind myself online is not real life.

I dunno, I see a pretty large spike in homicide rate: https://www.wolframalpha.com/input?i=us+homicide+rate

or https://i.imgur.com/v38RdZU.png https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/explorer/crim... (not per-capita)

Much like inflation, people aren't great at estimating the precise year-over-year change, but the last 5 or 10 years shows some real reasons for concern.

Maybe because in the places where most news are written (DC, NY, SF) crime has spiked drastically?

EG:

"D.C. Police have responded to 227 homicides so far in 2023. That's up 34% from the same time last year. It also tops 2021's total for the year, which was the highest since 2003. That means the homicide rate in the District is the highest it's been in 20 years, and climbing, with two months left in the year."

https://www.wusa9.com/article/news/crime/dc-homicide-rate-20...

https://www.axios.com/local/washington-dc/2023/10/03/dc-crim...

Nobody cares that it dropped in 2023. We care that it's worse than a decade ago, which it is. We hit a local minimum of 4.4 murders per 100k in 2014. It was 6.3 in 2022. That's a big difference.

Property crime goes unreported. It's not worth the trouble. I've had a dozen packages stolen, for example. My apartment building had 30 bikes stolen and hundreds more packages stolen. None of these will show up in the statistics. I have friends that have been carjacked or witnessed carjacking. Carjacking numbers are way up in my city compared to 10 years ago. Murders hit a low of 246 for us in 2013. The last 3 years were 499, 562, and 516, and they now cook the books by not counting suspicious deaths. We're down this year, which is great, but the past 5 years have been bloody as hell.

Major crimes dropping 8% from recent local maxima is not an achievement and doesn't mean people's perceptions of rising crime are incorrect. The past single year's delta alone is just not super relevant.

I don't trust the stats. Things that would have been reported as crimes 15 years ago aren't being called in now.

* Leaders inside the police force have incentives to make it look like their departments are effective and nudge the numbers.

* Elected officials, such as the mayor, have incentives to make it look like the police department is effective and reinforce the number nudgers.

* This has a reinforcing cycle where number nudgers are promoted and in turn promote other number nudgers until leadership is saturated. More creative accounting practices are encouraged as the environment becomes more competitive.

* Many District Attorney's are happy to prosecute lesser charges that are easier to convict on instead of more serious charges that more accurately reflect the reality of what happened.

The question isn't whether it is or isn't rising, but what voters consider acceptable. If voters find the current level of crime unacceptable, it doesn't matter what the broader trend is.

I personally find the concept of open air drug use, retail theft, and shit on the street unacceptable, I could care less what the numbers say and will vote based off of what I see with my own eyes. And whether or not that's "statistically sound", I could care less. Crime is decreasing? Okay, I guess it isn't decreasing quickly enough

I've noticed a general pattern on HN, and I think this thread exemplifies it: If the research confirms someone's personal experience, then they don't question the research methodology. On the other hand, if the research contradicts their experience, they look deeper into the methodology to find flaws in it. It's just a more intellectual form of confirmation bias.

Crime is such an immensely local and personal phenomenon. Statistics are abstract until you're the one getting carjacked by 5 guys or getting punched in the face during a home burglary.

This would make sense since the Gallup survey is based on people's perceptions based on social & traditional media that they see regularly. People remember notable events or shocking things, no one remembers the regular days where everything just went as normal - and the media doesn't cover such things since they are looking to grab attention with headlines.

The article mentions this at the end as outlier incidents, but just giving people the overall statistics may not help change their mind either because of what they perceive from other media sources.

They probably think it's rising because it rose substantially in 2020/21 and is nowhere close to returning to the typical levels of the 20 years prior.

St Louis should have all-time low total homicides, it's experiencing terminal population collapse. There are American suburbs similar in population to St Louis. So much of this article is just contrarian clickbait.

On the other hand, Detroit feels significantly safer than it did 10 years ago. It's still desperately poor, but it's a pleasant place to visit now. It's pretty, and it has clung onto its culture and history. For better and worse, these things tend to come back around. That doesn't happen if we pretend nothing is wrong.

I think "news" entertainment has painted this situation as a return to '70s-'80s violence, and that's not even close to reality. I'm not fleeing to the burbs, or going outside with a bulletproof vest. There's a problem, but it's solvable, and it won't take decades to fix.

I had to get someone to unlock a case the other day to buy underwear at Target. Other times it's been toothpaste or deodorant. Someone want to explain to me why these things are locked up now, but they used to not be? Someone want to explain to me why I see videos of people stealing without consequence while security looks on powerless to do anything? Someone want to explain to me why there are open air markets for stolen goods on the streets? Yeah, I'm sure crime is down. That sounds right.

Every time I see a similar headline, I'm remembering The Wire and how Baltimor police was working with stats to support elections promises.

"Who are you going to trust, me or your lying eyes?" The issue is that the stats don't tell the whole story. Most states/cities have simply decriminalized behavior that was previously considered 'crime' so that's why people think crime rates are rising when they see more open drug use, brazen theft like shoplifting or car break ins, and prostitutes walking around in public in their underwear--so people rightfully feel like crime is increasing because they see all this happening around them. Alas, the media and politicians say 'nothing is happening' because now prostitution, theft, and drugs are decriminalized. I am only in my mid 30s and have lived in California my entire life and I've never seen more open drug use, more theft/shoplifting, and prostitution is now more flagrant than it's ever been. Things really seem to accelerate into a free fall after prop 47 was passed.

Fewer people reporting crimes

anger and "concern" about rising crime is a great way to pump the impressions numbers.

i love these "research shows" crime articles. their research methodology is hilarious:

https://www.foxnews.com/us/most-unsafe-cities-holiday-season...

> Rockville, Maryland, ranked No. 1 for having the largest percentage of Airbnb properties without security cameras

isn't that... a good thing? i feel like not having security cameras usually means you don't feel like you need to have an eye on everything... and privacy at an airbnb is quite nice. might find a reason to visit rockville!

> Newark, Delaware, ranked No. 1 for Google searches related to holiday security

i feel like the only relation to actual safety here is the word "security" in this sentence.

A US crime rate is as useful to individual conceptions of potential victimization as US GDP is to employment.

I have a some personal anecdotes from the past few years living in Los Angeles that might help explain the discrepancy between people's perceptions and the crime data.

A couple years ago two cars pulled up in front of the house across the street from mine, four people got out and started trying to break in by throwing a rock through the front window. The noise was loud enough to alert myself and a couple other neighbors on the street who came out to see the burglars fleeing back to their cars after the homeowner scared them away with a gun.

I called 911 and waited on hold for 2 hours for someone to answer only to tell me that the only the people who had their house broken into could file a report and to have them call back. When I went to tell this to the neighbor it turned out that he did try to report it but gave up after waiting on hold for an hour. In the end this home break in was never reported and did not make it into the crime data.

Last year, living in a different part of Los Angeles, I woke up at 4am to the sound of people yelling and bottles breaking outside. There was a group of very noticeably drunk people sitting on the hoods of their cars drinking in the street. I watched them long enough to determine they were all wasted, at least one of them would be driving, and if any of them drove they'd be a danger to themselves and others. Again I called 911, waited a half hour, and was told they'd dispatch someone. A half hour later they all got into separate cars and did indeed drive away, leaving empty beer bottles next to where their cars had been. It's hard to know if LAPD actually came to check because this all happened on the street adjacent to the back gate of the Hollywood police department and around 6am it gets a lot of traffic from police cars heading back to the station at the end of the shift. As far as I can tell this crime was never reported and did not get included in the data.

Also within the last year, on the same street adjacent to the police station, a homeless man seemingly in a manic episode was vandalizing parked cars while screaming about killing people of a certain protected group. This time I called 911, waited 20 minutes and told the operator what I saw. Because the homeless man had wandered to a different street and I couldn't see him anymore I was told it's not an ongoing crime and to call the LAPD non-emergency line so they can take the report. I called the non-emergency line and reached their voicemail. This is another crime that wasn't reported and didn't get included in the data.

The crime statistics are only as good as the data being collected. From personal experience I've witnessed 3 crimes which should have contributed to crime statistics but didn't. My suspicion is that cities across America are having similar problems simply taking in and documenting reports of crime. It could very well be that crime is actually rising, people are correctly perceiving that crime is rising, yet the data to substantiate the rising crime rate isn't getting recorded.

Fell back to pre pandemic levels? I guess people are right that the last few years (covid era) have quite concerning crime rates.

There is a growing consensus that there is no point in reporting a crime to the police in most instances since it will take up your time doing so and either the police either won't even try to catch the perp or will fail to do so--and even if they do he will not receive a significant sanction.

For a while now, there has been complete suppression/derision of people who use their own life anecdotes to come to conclusions about society. The church of only using rigorous statistics to draw conclusions is a method of thought control.