more on policing expenses in a diverse society

the other day i posted about some research on the costs of policing in the different swiss cantons and how the more diverse a canton’s population, the more money was spent on policing. the researcher thought this fits pretty nicely with genetic relatedness and inclusive fitness-related behaviors — and so do i.

olave wondered, tho, if more policing might be required in those cantons with greater numbers of immigrants with impulse control issues — like africans, for example. well, because i got a screw loose curious, i thought i’d check the numbers out.

population stats for switzerland are available from the swiss government here (you didn’t know i was fluent in swissese, did ya? (~_^) ). now, afaict, the numbers aren’t broken down by race, so i used nation-of-origin as a proxy. that’s not exactly right, of course, ’cause a lot of people of african descent immigrating to switzerland might be coming from places like france and germany. but what can a gal do? consider this a rough guide.

so, having said that, i found that there was a correlation of 0.48 between number of african or african-descent immigrants in a canton and the amount of money spent on policing in a canton. that’s not 0, but that’s not that high either:


illus – x-axis = percentage of african immigrant in canton; y-axis = policing costs in millions of swiss francs.

the highest correlations were between total population and total monies spent (0.93) and total foreign population and total monies spend (0.94). but those two things — total population and total foreign population — have a correlation with each other of 0.93, so who knows what’s what here. maybe you just have to spend more money on policing the bigger your population gets and, of course, immigrants usually go to places with high population numbers (i.e. cities). coincidence. or, maybe your high population centers (i.e. cities) have higher crime rates because of all the immigrants there. or, maybe there’s something about population density that requires greater policing (i didn’t check that out … yet). so, who knows?

i wondered if a lot of diversity might impact on policing costs — i.e. if a community has 50 different ethnic groups in it versus 5. do the policing costs go up then?

so, because i really got a screw loose i counted how many different countries the immigrants in each canton came from. here are the figures i got (don’t ask me why they’re listed in this order — this is the way they popped out of the swiss database!):

Uri = 74
Schwyz = 123
Obwalden = 87
Nidwalden = 87
Glarus = 88
Zug = 132
Solothurn = 145
Schaffhausen = 117
Appenzell Ausserrhoden = 93
Appenzell Innerrhoden = 63
St. Gallen = 151
Graubünden / Grigioni / Grischun = 131
Aargau = 161
Thurgau = 138
Ticino = 159
Vaud = 178
Valais / Wallis = 157
Jura = 117
Zürich = 179
Bern / Berne = 180
Luzern = 150
Fribourg / Freiburg = 160
Basel-Stadt = 159
Basel-Landschaft = 151
Genève = 190

and i get a correlation of 0.65 between number of different countries from which immigrants in a canton hail and amount of money spent on policing in each canton. that’s a stronger coefficient than the researcher’s -0.541 for his similarity index (“number of citizens and proportion of foreigners”) and amount spent on policing in 2009.

here are a couple of nifty charts (i arranged the data on the x-axis backwards so that you could compare these with the researcher’s, rolf kümmerli’s original charts):


illus – x-axis = number of different countries immigrants in cantons come from; y-axis = policing costs in millions of swiss francs.

that crazy outlier is zurich. (presumably you have to spend a lot of money on policing to protect — all that money!) if i take zurich out, the chart looks like this:


illus – x-axis = number of different countries immigrants in cantons (minus zurich) come from; y-axis = policing costs in millions of swiss francs.

the more immigrants you have, the more money you have to spend on policing (or so it seems). the more different types of immigrants you have makes it even more likely you will have to spend more money on policing (or so it seems).

previously: “can we all get along?”

(note: comments do not require an email. more swiss chicks!)

Advertisements

11 Comments

  1. I think at root it’s a variation on Putnam’s research on social capital. The less of an “us” there is the less social restraint there is in all areas including all sorts of crime. This can be seen in everything from littering upwards.

    On top of that there are specific added bonuses from having particular population groups but that does depend somewhat on which end of their national spectrum they come from.

    African immigrant groups for example are often very noticeably different depending on whether they are from the brighter end or the dumber end of the source nation usually depending on whether they’ve entered as (legitimate) students or refugees.

    One aspect of the recent flash mob type violence is (imo) the cognitive dissonance created in black people who’ve been told by white liberals that their problems are all the fault of white racism seeing equally black African students leapfrog right over them.

    On top of all the above there’s a very straightforward aspect which is the huge additional cost of translation services and related language issues.

    Reply

  2. @g.w. – “I think at root it’s a variation on Putnam’s research on social capital.”

    exactly!

    @g.w. – “On top of all the above there’s a very straightforward aspect which is the huge additional cost of translation services and related language issues.”

    oh yeah! i didn’t think of that. especially when cases come to court and all that. *facepalm*

    @g.w. – “One aspect of the recent flash mob type violence is (imo) the cognitive dissonance created in black people who’ve been told by white liberals that their problems are all the fault of white racism seeing equally black African students leapfrog right over them.”

    i wonder how much of a problem that is here in the states? i mean, yeah there are blacks from africa that do well here, but i wonder how noticeable they are to african-americans yet. there’s not that many in the population … yet — except maybe in places like new york. not so much in milwaukee, wi, tho. i think a lot of it here in the states has to do with finally getting a (half-)black president — feeling cocky now, you know? not to mention all the years of pc indoctrination that people under 20 (maybe even 30) have simply grown up with.

    Reply

  3. “i wonder how much of a problem that is here in the states?”

    more of a uk thing

    “tho. i think a lot of it here in the states has to do with finally getting a (half-)black president”

    yes a related but opposite effect – it’s our turn now

    Reply

  4. i think i’ve part-figured this out.

    inner-city areas of western cities over the last 30 years or so. 100s of different ethnic groups in their own enclaves. each group has their own crime pattern like a signature. i used to think it was just culture but when you start reading about HBD stuff you wonder maybe it isn’t. this is especially so when you consider the different groups cluster into a few patterns and groups from places far apart geographically can have the same pattern.

    anyway my first model was influenced by reading about IQ and latitude so my first thought was latitudunal with people divided into tropics, sub-tropics, mid-latitudes (muslim and non-muslim version), europe and east asia.

    (i now think the muslim vs non-muslim divide is fbd / cousin-marriage etc as they’re mostly the same as other mid-latitudes but a bit more violent.)

    in terms of crime the tropicals seemed to major in violent crime, sub-tropics in bluecollar crime, mid-latitudes and east asia in whitecollar crime. white people were quite jekyll and hyde both in exemplifying the bluecollar and whitecollar division in non-violent crime and also clustering with south asians in violent crime when sober and with tropicals when drunk.

    this model is actually a pretty good fit but with a lot of exceptions at the two extremes. some african groups are more like south asians and some are… somalis. i used to think it was maybe a class thing, that some African immigrant groups were coming from the top 10% of their country and some from the bottom 20% for whatever reason. i think there’s some truth in that especially vis a vis immigrants versus refugees but it’s not the main thing.

    i think the reason the latitudinal model is a pretty good fit is because it mostly mirrors the actual model which is

    centuries since forager

    More precisely i think if you took each population / region and divide up the last 10,000 years into units of centuries and then add up
    – number of centuries as forager * k1
    – number of centuries as pastoralists * k2
    – number of centuries as farmer * (minus) k3
    – number of centuries as urban farmer * (minus) k4
    where k1 to k4 are constants and urban farmer simply means high density and you figured out the right constants you could (roughly) match the current murder rate for each population group.

    so for example Chinese near Peking might be
    – 50 * forager – 30 * farmer – 20 * urban

    Africans from the interior 400 years ago might be
    – 90 * forager – 10 * farmer
    or
    – 100 * forager

    Africans from the settled coast might be
    – 70 * forager – 30 * farmer

    Average Europeans might be
    – 60 * forager – 30 * farmer – 10 * urban

    This model explains European outliers like Lithuanians and Finns.

    The settled coastal African tribes captured most of the slave-descended Africans. I think they captured them on raids into the interior and that’s why slave-descended Africans have much higher rates of violent crime than many other African groups.

    Reply

  5. forgot, i guess the reason the latitudinal model was a reasonable fit was because farming spread faster east-west than north-south.

    Reply

  6. @g.w. – “in terms of crime the tropicals seemed to major in violent crime, sub-tropics in bluecollar crime, mid-latitudes and east asia in whitecollar crime. white people were quite jekyll and hyde both in exemplifying the bluecollar and whitecollar division in non-violent crime and also clustering with south asians in violent crime when sober and with tropicals when drunk.”

    the latitude thing is very, very right i think. the crime thing is clearly a frequency thing — violent crime much more frequent in the tropics, but whites and east asians still capable of it and do commit violent crimes, just not so frequently.

    farming + large, settled populations. you’d think living in large, settled populations would just select for less violent individuals, otherwise it would just never work. you can’t have everyone hitting each other over the head all the time in ancient athens or rome or even new york city.

    santa brought me pinker’s new book on violence. (^_^) one of these days i might actually read it. (~_^) i can’t help thinking before i read it (which i shouldn’t do, i know!) that he must be talking about eurasians — yes, we’ve gotten less violent, but have all populations? really? but, i’ll give him a chance to present his data, first.

    Reply

  7. “i can’t help thinking before i read it (which i shouldn’t do, i know!) that he must be talking about eurasians — yes, we’ve gotten less violent, but have all populations? really?”

    i think he’s broadly right but the change hasn’t been uniform – if it’s a function of selection pressure from agriculture and higher density living then it’s effected different populations in proportion to how much time they’ve spent under those selection pressures. Those populations where farming and urban living came late – south and to an extent north of the main eurasian axis – have been less effected and that explains a lot.

    Also i think it can go backwards. I think the 3000 years the cucuteni people spent in high density living would have made them less violent but if they went onto the steppe when the climate changed and became nomads then i think there would have been competitive pressure to get more violent again. Later still after settling in Hungary (or Iran or India) as farmers again then the pressure goes into reverse again. In the context of propensity to violence peoples would be a composite of those centuries of differential pressures.

    So i do think there’ll be a lot of variation in Africa even if the average is still higher than elsewhere. I think black Americans and West Indians are distinctive in this from most African groups (but not all). I’m guessing what slavery did was reach back a 1000 years or more into the past and grab a bunch of people who hadn’t gone through much of that process or who were still part of extremely violent cultures at the time who still had very high frequencies of these violent traits.

    The worst bit is i think the inner-city underclass environment is selecting for it – breeding barbarism. When i was involved in those areas i thought it was getting worse but assumed it was caused by the culture getting nastier – rap, MTV etc – but now i think the culture was getting nastier because the frequency of things like MAOA (or its equivalents) were increasing and things were getting worse.

    (If 8% of males in a population had those traits and that 8% were responsible for most of the violent crime then the frequency only needs to go up 1% to 9% for a ~16% increase in violent crime.)

    Reply

  8. @g.w. – “Also i think it can go backwards. I think the 3000 years the cucuteni people spent in high density living would have made them less violent but if they went onto the steppe when the climate changed and became nomads then i think there would have been competitive pressure to get more violent again.”

    absolutely! people still like to assume that evolution is unidirectional and generally means what we like to think of as progress. wrong! greg cochran talked about this a bit over here. (this is why i’ve realized that’s it’s really important for me not to assume that, say, arabs have always been inbreeding the way that they do even in the distant past. they might not have. mating patterns change. everybody: remind me of that if you see me screwing up about it!)

    @g.w. – “I’m guessing what slavery did was reach back a 1000 years or more into the past and grab a bunch of people who hadn’t gone through much of that process or who were still part of extremely violent cultures at the time who still had very high frequencies of these violent traits.”

    that’s a really interesting thought. haven’t seen anyone say that before.

    @g.w. – “The worst bit is i think the inner-city underclass environment is selecting for it – breeding barbarism.”

    unfortunately, yes. =/

    @g.w. – “When i was involved in those areas i thought it was getting worse but assumed it was caused by the culture getting nastier – rap, MTV etc – but now i think the culture was getting nastier because the frequency of things like MAOA (or its equivalents) were increasing and things were getting worse.

    well, this gets back to my perennial question — where does culture from? culture is just learned behavior, right? at least, that’s what the anthropologists keep telling us. well lots of animals learn behaviors — this is a good solution that Mother Nature has come up with — if you’re too rigid in your behaviors, you might quickly go extinct if the environment changes. so bear cubs, for instance, learn from their mothers to eat berries and fish … or maybe human garbage out of garbage cans if that’s available. but they don’t learn how to open cans of spaghetti-o’s and zap them in the microwave ’cause they don’t have the intelligence genes (or opposable thumbs) to enable them to do that. bear culture is dependant on their nature.

    same with humans, i think. different cultures depend upon the different natures of the peoples we’re talking about. it’s a flexible system — that’s the whole point of it — so you could never precisely predict gangster rap from the frequency of maoa alleles in a population — but you could probably predict that the culture will have violent tendencies (gangster rap, hakka dances, etc.).

    Reply

  9. “that’s a really interesting thought. haven’t seen anyone say that before.”

    it suddenly hit me when i was reading about the yanomani because the similarities between them and urban black gang culture was so overwhelming.

    it’s a pretty big deal if true because most people don’t realise how insanely violent those areas are and at the same time don’t realise the relatively small percentage who drive it.

    .
    “so you could never precisely predict gangster rap from the frequency of maoa alleles in a population — but you could probably predict that the culture will have violent tendencies (gangster rap, hakka dances, etc.).”

    yes, without the rule of law then above a certain (quite small) percentage they’ll dominate the environment and then by extension gradually come to dominate the local culture.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s