Why San Francisco? #
May 8, 2024
Happy Wok
I traveled to San Francisco late in December of 2023. Going in, I think the only thing I heard spoken of on the six-hour drive from Los Angeles was crime, both from everybody in the car with me and from messages from friends. It felt almost like a knee jerk reaction to bring up crime whenever you hear speak of San Francisco.
Around the time I had been to San Francisco, an opinion piece was published in the National Review, “Requiem for San Francisco”, where the author laments about the crime and unaffordability in the city (more particularly about crime, of course). Another article from the National Review (“The Left Gaslights Us on San Francisco’s Problems”) discusses similar crime related issues within San Francisco and began with a discussion of Elon Musk’s words that SF felt “post-apocalyptic”. I will say that a lot of these comments, both in articles and from people I’ve talked with, don’t give a timeframe for when “SF was safe/healthy", it’s mostly just “SF is so dangerous/bad now”. Everything I’d heard kept turning around in my mind, so I want to understand why people feel this way, because when I was there, I hadn’t seen or had many issues to speak of.
Firstly, a lot of what I hear about San Francisco (or really any American city) is hearsay without much information or numbers to back it. We can level this with data from the FBI’s Crime Data Explorer targeted towards the San Francisco Police Department ending at the year 2022.
This is what all violent crime looks like in San Francisco between 2012-2022. The quantity of reported crimes in 2022 is 5,323 and we can overall see that the quantity of crimes has remained fairly level over time, even going down a bit in 2020, which is particularly interesting when compared to the entirety of California when it comes to all violent crime.
In California in general, we can see a general increase in crime over time, most prominently around and after the pandemic and after 2014. This itself contrasts quite heavily with San Francisco. It is interesting, firstly, that violent crime increased at a greater rate of change in California in general than in San Francisco.
Of course, cases of all property crime are higher than that of violent crime, San Francisco’s property crime looks as follows.
The fluctuations are less intense than with the other graphs, but we can notice overall that property crime in the city is much the same as it has been for the entire decade.
Let’s take these datasets back quite a bit.
Property crime in San Francisco has been the lowest overall in the 2000s (in this timeline), whereas the periods of time before and after have higher levels of property crime, though it is somewhat lower now than it was in the late 80s and early 90s.
Violent crime in SF has been overall trending downward. Noticeably, it was quite high in the 80s and 90s, but has been falling reasonably consistently ever since. The same is true with California in general. The primary distinction between California and SF on this metric is property crime. Where California has been overall trending lower over time, whereas SF’s property crime has gone up from the 00s.
What if we were to consider homelessness? This is another issue that brings a lot of ire to people with claims that homelessness has gone up a lot in San Francisco.
From this graph from the SF government, we can see that homelessness has indeed gone up in SF since 2005. I cannot find much quantitative data that goes further back than 2005, but other sources suggest that homelessness in SF has been the same (or just generally, “a lot”) for quite a long time [1], [2], [3]. I did manage to find that, according to the CDC, an estimated 6,000 to 18,000 people were homeless in San Francisco in 1990 [4]. In this case, even the lower bound of 6,000 people is about equivalent to what was found from the SF government between 2005 and 2022. And even with that, unhoused homeless persons in SF are down from 2019, and is about equivalent to the mid to early 2010s.
Overall, homelessness and property crime are about the same or even a bit better than in the 80s and 90s (homelessness could very well even be significantly better, if the CDC estimates are accurate), and violent crime is down. In that case, it could be that people are more concerned about the increase in homelessness and property crime to pre-2000 levels than they used to be. However, that doesn’t necessarily strike me very much as necessary for a “requiem”, unless you considered it necessary to also give a “requiem” to SF in the 80s and 90s. Most charitably, it can be said that many of these commentators want a return to the mid-2000s, where property crime and homelessness were lower (though violent crime was about the same or even a bit higher).
Note, however, that a lot of this commentary on SF is recent, within say the last half a decade to a decade. The commentary I was looking at and referenced earlier was from the National Review and came from the last few years.
In considering just the last few years then, we can first consider homelessness. The homelessness discussion I pointed out earlier has a fair caveat that homelessness is down post-2019, and the unhoused homeless population (what these commentators appear to be complaining about most prominently) is about equivalent to mid-2010s to early-2010s levels. Property crime, as well, is about the same as the mid 2010s, though higher than the early 2010s. Once again, violent crime is down, and continues to be down.
To me, what this really suggests is that a visitor to SF in the mid to early 2010s would have an equivalent likelihood of seeing some number of homeless persons as they would now, but there appears to be, in this commentary, a suggestion that SF used to be better.
My hypothesis, then, is that commentators decrying SF’s problems are truly concerned about property crime, but may be conflating it with other types of crimes or social ills. I want to preface this with that, in my opinion, telling a person that some city or place is safer than they think and that they should feel safer there is certainly not helpful, or even actively unhelpful, to understanding and resolving crime and its perceptions. If someone feels unsafe somewhere, it’s best to understand why they feel unsafe, as opposed to telling them why they should feel safe. Now, I want to give two to three reasons to back my assertion.
Firstly, violent crime being down means that you are safer from active, physical harm against yourself. However, it is useful to know that the plurality of violent crime is committed against poor and low-income persons [5]. In that case, we can hypothesize that a majority of the violent crime that had been committed in SF, when the violent crime rate was higher, was against poor and low-income persons. My assumption is that most of the travelers to SF at that time, and especially now, were more likely of the middle class or higher classes, which are people less likely to be victims of violent crime. This assumption that the people traveling to SF are middle class or higher class is based upon research that income correlates highly with international travel [6], which I would hypothesize would hold for within-USA travel as well, though granted probably to a less prominent degree. In that case, violent crime, while still certainly a concern for travelers, is not as prevalent among the types of populations that travel to SF or among the commentariat of SF’s woes, as among other groups.
Secondly, violent crime is not necessarily as visible as property crime to travelers and general people. It can be seen in the charts above that violent crime is numerically significantly lower than property crime, which suggests that you have a higher likelihood of seeing property crime (or being a victim of property crime). Being a victim of property crime surely makes any place that the crime was committed in feel unsafe, this is also surely the case for people who see property crime (either in action or after the fact) in any particular location. If you feel unsafe about any particular location, you wouldn’t think about if you are going to be a victim of violent crime or property crime, you would be concerned about being the victim of any kind of crime. Putting it in a box would feel irrelevant in the moment.
For my third point, I used a dataset from the SF police department titled “Police Department Incident Reports: Historical 2003 to May 2018” [7]. I split this data up amongst the time-period from 2003 to 2010 and from 2011 to 2018 and analyzed it from there. I additionally used the assumption that people are more likely to spend time in the more touristy parts of the cities, which I presume are the police districts of Northern, Central, and Southern.
Comparing the two periods of time together, we can see that crime has fluctuated throughout the different police districts in curious ways. In this way, it’s hard to analyze much about what people (travelers into the city) see or predict what they might see. In this case, I would have presumed that there was in general a greater quantity of crime in touristy districts of the city, which would explain why people feel more in danger there, because they are statistically more likely to see more crime. But in this case, in Northern the crime is down, whereas in Central and Southern, crime is up. If you spend all your time in Central and Southern, then you are more likely to see more crime now than you would a decade ago, thus making you feel like you are in danger. However, a very large portion of people visit areas in Northern, where quantitative crime is down. It could very well be the case that crime in Central and Southern crowd out any positive effects that may have happened in Northern, or it could just be that because people are statistically quantitatively more likely to see crime overall, they feel unsafe regardless of the part of the city they would be in.
Interestingly, crime in the Tenderloin district is significantly down, which is not something I’d ever known or heard of until I looked at the data. Still, I digress.
From these three points, it strikes me as though people are feeling more crime now than there very well might be in the city. Certainly, compared to the 1980s and 1990s, SF has less crime, but in comparison to the mid-2000s, that becomes more debatable. Violent crimes have decreased in the city overall, and we see that property crime has fluctuated, increasing since the 2000s, but not more than the 80s and 90s. Visitors to the city in the 2000s coming again now, then, may very well feel less safe as they are statistically more likely to see property crime. Violent crime, however, they are unlikely to see in general (though it’s not impossible), given the average economic standing of a traveler.
In that case, it strikes me that the primary concern of these travelers, from what they have heard of from friends that have traveled to SF, or from their own experiences in SF, would be property crime. Then, it is possible that this property crime could be seen by them as a reflection of all crime, or all social ills present in the city. This conflation of property crime with all social ills is certainly something that can be reflected already in society itself. Consider how crimes against middle class or upper class “normal families” are much more televised than crimes against poor families. Regardless, whether these visitors and commentators feel safe or not in San Francisco does not change based upon that assertion, but it hopefully gives a potential explanation to why that fear exists and how it compares to SF’s current state of affairs.
Citations
[1] https://projects.sfchronicle.com/sf-homeless/overview/
[4] https://www.cdc.gov/mmwr/preview/mmwrhtml/00015783.htm