If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:

Or download the MP3


Last week I decided to order some pizza for the family to eat while we watched Touching the Void. Before going any further I should say that Touching the Void is far and away my favorite documentary, and I couldn’t recommend it more highly, though there is a significant amount of swearing, albeit in contexts where swearing is entirely appropriate. In any case the pizza…

I ordered pizza from Papa John’s (not sure in this day and age if that’s important or not.) When it arrived I took the pizzas from the delivery driver and handed them to my son. From there he handed me the credit card slip so that I could sign it. As I was signing it I noticed that a car had pulled up next to the car of the delivery driver (who had parked across the end of my driveway). I didn’t think much of it, there was another car parked across the street and I assumed the recently arrived car was just squeezing in between the two before parking in the driveway across from me.

When I next glanced up the car of the delivery driver was in motion, at which point I figured something weird had to be happening and I said, motioning towards the car, “What the heck is happening there!” The delivery driver turned and said something along the lines of “Hey bro! Don’t steal my car!” and began running after him. A second earlier and he might have got in front of him because the thief had to turn the car around to get out of my neighborhood, but by the time I pointed it out the thief had already backed into my driveway and from there he roared off. (I wonder what would have happened if I’d paid with cash? Maybe the delivery guy could have stopped it, or maybe he just would have gotten run over?)

I assume the delivery driver left his keys in the car or left it running, I honestly don’t remember if it was the latter. He also left his phone in the car, along with another order of pizzas he was supposed to deliver after mine. Anyway, I lent him my phone and he called the police. He unfortunately couldn’t remember his license plate number, and I assume one of the first things the thief did was take off the Papa John’s topper, since if he had left that on he would have been pretty easy to catch.

A police officer showed up pretty quickly, fast enough that the driver was still on the phone with Papa John’s explaining to them that they were going to have to make some more pizzas. But once he got off the call and gave me my phone back the policeman told me I was good and I went back inside to eat. I’m a little bit annoyed that I’m not in the loop on things. I’d like to know how it ends up getting resolved, though I’ll definitely be asking the next Papa John’s driver who shows up about it. (Assuming they don’t put me on some kind of blacklist.)

In any event all of this got me to wondering about the state of crime and other social indicators, like the number of homeless people. This particular crime seemed fairly brazen and unusual, and also the delivery driver assured me that he had been doing pizza delivery for a long time (though to be fair he looked like he was at most in his late 20s) and had never heard of this sort of thing happening. And one assumes that if he had, he wouldn’t have left himself in a position for it to happen to him. I had certainly never heard of it happening, nor do I know anyone who’s even had their car stolen, period, that I can think of, at least not anyone I know well.

In other words you have an unusually brazen crime, happening right in front of my house. Is it just an exceptionally rare thing that I just happened to witness by chance? Or is it part of some larger trend of increasing lawlessness. You probably already know where my biases lay: towards it being part of some larger trend. And, of course being biased, I immediately started looking for something else that might be an example of that trend. The thing that immediately came to mind was homelessness. Which seems to be getting worse and worse despite the New York Times apparently running out of words to describe how good the jobs numbers are.

Most of the time when I start one of these posts I have a pretty good idea of what my conclusion is. Either because I’ve already come across some piece of evidence which represents a smoking gun, or I have some point of my own that I’m hoping to arrive at. That is not the case with this post. Having just read Pinker’s Enlightenment Now, I’m predisposed to think that crime, like most things has been getting better. (I’m guessing that the hapless delivery driver would disagree with me.) Is crime getting better along with everything else? If so, then what’s up with the increasing number of homeless people? And shouldn’t homelessness and crime track pretty closely? These are the questions I’m setting out to answer, and at the moment I’m not sure how it’s going to fall.

Speaking of Pinker, I thought I might as well start with him, and I would have thought that among his 75 figures that there would be one on the decrease in crime, though I had no specific memory of one. And it turns out the reason that I have no specific memory of one is because he didn’t include one. He covers murder, but he doesn’t get into property crime. In fact there is no entry for crime in the index at all. I’m not sure if I should be surprised by this or not. The graphs one does find in this area definitely show a decrease in all sorts of property crime, though since Pinker likes to credit “enlightenment values” with the decrease I think he prefers to be able to show that the decrease started at around the time of the enlightenment. As such many of his graphs start in the 1700 and 1800’s, with some going back much further (with one graph on GDP going all the way back to 1 AD).

Of course not all of the things he wants to talk about have data available going that far back so oftentimes his graphs start much later, but only because he doesn’t have the data to go back any further. In this case he has the data, but it doesn’t show the nice smooth downward slope of most of his charts, a chart of property crimes starts off really low in 1960, then rises steadily before reaching and staying at a peak through the 80’s and then starting to decrease around 1990, though even after a decrease of over two and a half decades things are still not to the level they were in 1960. In other words, this graph does not quite fit the narrative Pinker is going for in his book, so perhaps it’s not surprising that he didn’t include it.

Besides not fitting his narrative, one additional reason for not including it might be that no one is entirely sure why crime has been falling since 1990. Certainly Pinker can’t easily map “enlightenment values” to this data as an explanation, though that may be only explanation that hasn’t been offered. Vox.com ran an article listing 16 possible reasons for the decline in crime everything from an aging population, to video games, to abortion, and lead are suggested. I know that lead has been a favorite of many people, though just recently someone pointed out that despite horrible lead pollution in Eastern Europe under communism there appears to be no evidence of increased criminality there.

In any case, whatever the cause for the decrease in crime, the graph doesn’t fit Pinker’s narrative, but it also doesn’t fit my narrative very well either. I find no evidence that there has been a recent surge in car thefts, Car thefts have in fact been falling, though at this point it’s important to talk about the role of technology in preventing car thefts. Cars are, in general, much more difficult to steal these days than they have been in the past. With stuff like locking steering wheels, immobilizers, GPS tracking, and similar, the only way they were able to steal the delivery drivers’ car is that they had the keys. So technology has made car theft much more difficult, but that may have nothing to do with the “base rate” of criminality in society.

Okay, so it appears, despite the dramatic nature of my own experience, that there hasn’t been any increase in property crimes, or car theft. Which takes us to the next questions, has there been an increase in the number of homeless people and if so why has there been no corresponding increase in the amount of crime?

Here again, I’ll once again start out by describing my own experience. My memory is that homeless individuals and panhandlers in general were pretty rare when I was growing up. In 2000 I moved into my current house, and I don’t remember seeing any homeless panhandlers in the area, that is until the financial crisis of 2007-2008, at which point I started seeing them everywhere, especially on a particular corner near my house. My initial assumption was that the sudden increase was due to the housing crisis and the economy cratering. This would make sense of course, if unemployment shoots up and people find themselves suddenly unable to make their mortgage payments, then it’s only to be expected that the number of homeless would increase. But when the economy improved, there didn’t appear to be any corresponding decrease in the number of homeless. Just this month the current period of economic expansion hit nine years, and on top of that I recently saw that the number of job openings exceeds the number of people looking for employment for the first time since 2000. And yet, despite all this, there appear to be as many homeless people and panhandlers as ever, if not more. Why is that?

Of course the first step is to see if my observations match reality. It’s entirely possible that I’m just suffering from confirmation bias, that I’ve developed this theory of an increase in the number of homeless people and consequently I pay particular attention to them. Or maybe Salt Lake City just has it especially bad for some reason. Obviously we need some hard numbers, but as it turns out even when we look at the data the picture is mixed.

First up we have a report from HUD which says that the number of homeless, after falling from 650,000 in 2007 to 550,000 in 2016, rose for the first time in 2017. (Good summary here from BBC.) And apparently much of the gain is in LA because a booming economy has increased the cost of housing. Unfortunately the numbers only go back to 2007, so it’s impossible to say if the numbers are still historically high, or if we’re back to the level of homelessness which existed in say the mid 90s…

On the other side there are reports of homelessness increasing among children and students. We could certainly reconcile the decrease mentioned above with these numbers, but only by assuming that children and students are becoming a greater percentage of the total homeless population, while the number of homeless adults is declining, which isn’t exactly great news.

New York City appears to have the best data on homelessness of any source and here the situation is unambiguous. The rate of homelessness in New York is skyrocketing. There were 12,000 homeless people in NYC in 1984 and now there’s 63,000. Five times as many, even though 1984 was in the middle of New York’s crime and murder epidemic. Additionally much of that increase has come just since 2012, when we were already three years into the recovery. Now of course when speaking of New York (or anywhere really) you can have an argument about to what extent the leadership at the time was responsible. Many people feel that Bloomberg was horrible for the homeless and De Blasio has done much better. But you’re still looking at a huge increase in the numbers no matter how you slice it.

Moving farther afield there are numerous stories about the increasing problems with homelessness from all over the country:

  • Starting in my own backyard, we have conflicting reports:
    • Here’s an article saying that Salt Lake City has reached a critical mass of homeless people and wonders how we got into this state. This was written in 2016.
    • Here’s another article written 10 months before the first one, claiming that SLC had reduced the population of the chronically homeless by 91 percent. (Certainly that’s not my impression, though the chronically homeless are only 20% of total homeless.)
  • Then we turn to an article about Anchorage’s homeless problem. One feels like Anchorage could just buy all their homeless people a bus ticket, and have that problem solved, who wants to be homeless in Alaska, particularly in the winter?
  • Next, here’s an article from just this week about a neighborhood in Las Vegas that’s overrun by homeless people, despite numerous attempts to deal with the problem.
  • Speaking of stories from this week I found the following stories about Seattle. First Seattle voters are fed up with homeless spending, homelessness in Seattle is increasing and it has reached a horrific tipping point. Perhaps voters are fed up with spending because it doesn’t appear to be doing anything to solve the problem?
  • Finally there’s the situation in LA. The report I mentioned earlier, about homelessness increasing for the first time in 2017, placed much of the blame on LA. The number of homeless in LA jumped a staggering 23% in just the last year, and this is despite billions in taxes which have been earmarked to fight the problem.

After working through all of the above, it would appear that the only real outlier to a picture of increasing homelessness is the HUD report, and even it shows a recent uptick. What would have been really useful is if their data went back farther than 2007. Is the 2016 number of 550,000 still really high from a historical perspective, or is that number actually as low as it has ever been? If so homelessness should be added to Steven Pinker’s list of things that are perpetually improving, but all of the other evidence suggests that this is probably not the case. That whatever else can be said about the number of homeless people in 2016, one can not say that it represents some sort of historical nadir.

This seems particularly borne out by the NYC data, which has the advantage of going all the way back to the end of 1984. Having this additional historic data, allows us to see not only that the current rate is five time the rate back then, but also that the “Great Recession” didn’t seem to have much of an effect on homeless rates. That rather than going up during that time and then falling back down once the economy improve that instead, homeless rates, for some reason have skyrocketed during the last few years even though the economy has been expanding.

Pulling this all together it appears that it’s going to be difficult to say what the true number of homeless people is, and whether that number has recently increased slightly, increased dramatically, or decreased slightly (I see no reason to believe it’s decreased dramatically). That said, it does seem safe to conclude that it’s extremely high. Certainly higher than it’s been for many decades. (I could certainly imagine that it was higher during the Great Depression.) And this is all happening during a time with very low unemployment, an expanding economy, and presumably, more money being spent directly on the problem than at any point in history (and this does include the Great Depression.) How do we explain this apparent paradox?

Honestly I don’t know. Though I can certainly speculate. Here are some possibilities:

Maybe homelessness is inversely correlated with the economy. Homelessness isn’t bad in spite of a good economy it’s bad because of a good economy. As I mentioned with respect to LA, some people think that this is the problem. That robust economic growth has increased the cost of housing. So, perhaps homelessness always increases when the economy is booming because “homes” are more expensive. Certainly this could be a contributing factor, but if you look at the New York numbers, I don’t see any obvious correlation between, for instance, the annual increase in GDP and the homelessness rate.

Alternatively, perhaps this is just the far left end of the advancing spread in inequality. That, yes the economy is doing well, and there are lots of job openings, but that most of the economic gains and most of the jobs have gone to the top 5% and things just keep getting worse for the bottom 5%, which is reflected by the increase in the number of homeless people. That not only are the poor getting poorer, but also, despite more jobs than job seekers, there are no entry level jobs. There is some marginal evidence for a decline in entry level jobs, but as far as I can tell McDonald’s is almost always hiring. At least in my neck of the woods.

Of course, we shouldn’t overlook the opiate crisis. Perhaps the increase in the homeless rate is just an increase in the number of people who are addicted to heroin or meth and consequently can’t hold down a job. Once again I’m sure it’s a factor, but I’m also not sure how big of a factor it is. This page claims that only 26% of homeless people abuse a drug other than alcohol. (To be honest that sounds low.) If the number of homeless people had only increased by 26% over the last few years and previously no homeless people were addicted to drugs other than alcohol, then this might make sense, but neither of those things are very likely to be true. The homeless rate in NYC more than doubled from 2006 to today, and I’m reasonably certain that the homeless have have been abusing drugs as long as there have been drugs and homeless people.

Finally, it may just be that people don’t have the support structure they used to. Families are smaller, single mothers are more common. All of this means that there are fewer people to catch you “on the way down.” I saw this process play out with my college roommate. He had bad health and couldn’t keep a job. This was unfortunate, but on top of that he was an orphan with no siblings. He stayed with some friends for awhile, and he stayed with his uncle for awhile, he even got the Mormon Church to pay for his apartment for a while (despite not being Mormon). Somewhat tragically, he actually died of alcoholic hepatitis before actually becoming homeless, but my guess is that at that point it was only a few months away, and even if it had taken longer, I think he surely would have eventually ended up homeless if he hadn’t died. I think if his parents had still been alive, or if he had had any brothers and sisters, the story would have been quite a bit different.

To conclude, I’ll repeat again, I don’t know why homelessness is so high despite an economy that by all appearances is doing great. Of the four things I mentioned, I suspect that inequality, the opiate crisis, and lack of support all contribute, but I don’t think they’re sufficient. Also I’m going to say it doesn’t have much, if anything, to do with the strength of the economy and high housing costs. And I’m willing to predict that if the economy does start tanking, that the situation will only get much worse.


I am not in any danger of becoming homeless, but if you want to help me avoid that danger anyway, no matter how tiny it is, consider donating.