March 26, 2009

Lower crime is not associated with more materialistic values

In Irrational Exuberance, Robert Shiller reviews many possible reasons why people may have been so enthusiastic about the stock market in the 1990s. One of them is that with the fall in violent crime, it was no longer dangerous to flaunt your wealth, lest it attract hordes of criminals. Now that more basic survival needs had been taken care of, you could indulge your materialistic side -- say, by jumping on the stock market bandwagon.

To be fair, he doesn't dwell on this point; there's only a single, off-the-cuff paragraph about it. Still, level of materialism is not predicted by the change in the crime rate. The reason for this oversight is that no one knows -- consciously -- that there have actually been two decade-long crime waves in the 20th C: the familiar one that began with Elvis and peaked with Dr. Dre, and the nearly-down-the-memory-hole wave of the 1910s and the Roaring Twenties, peaking in 1933 -- y'know, those flashback parts from The Godfather Part II, the Saint Valentine's Day Massacre, etc. Here is a graph of the homicide rate for the entire century (data here):

If the downward trend in crime during the 1990s aided the rising materialism, why didn't a similar decline in crime from 1934 - 1956 do the same? If anything, materialism was evaporating during that time. Conversely, the ostentatious 1920s saw a skyrocketing crime rate, not a decline. So, while the relation between lower crime and higher materialism may hold for the recent crime wave, it was just the opposite for the earlier crime wave.

Probably crime and materialism follow independent courses. If they have slightly different periods, superimposing one trend on the other will produce beats -- at one point, they'll be completely in sync (mid-1920s), and at another point they'll be completely out of sync (mid-1990s). This is a key reason that we need to look at as long of a stretch of data as possible -- otherwise, we'll see a strong moving-in-tandem pattern in what is really a coincidence.


  1. I think prohibition played a large part in the first crime wave and WWII may have temporarily stunted "materialism" in the aftermath of its repeal.

    What do you mean by materialism? Overall growth in wealth, conspicuous consumption, or something else?

  2. What's your metric for materialism? As a child growing up in the 1960s, I absorbed the conventional wisdom that the 1950s were a period of soulless materialism, while the 30s and 40s - if horrible in many ways - at least could boast the idealism of the New Deal and the war effort.

    Also, some criminologists suggest that homicide rates understate current violence, since medical advances now save the lives of many who would have died in earlier decades. (Which is not to say the recent fall in violent crime isn't real, just that a current homocide level of x indicates more violent crime than a rate of x 50 years ago.)

    intellectual pariah

  3. My metric for materialism is whatever people mean when they say the '80s and '90s were materialistic. Ostentation, upbeatness, obsession with things and possessions, etc.

    Prohibition did not cause the first crime wave -- it began in about 1905, and was rising faster during the 1910s than the 1920s.

  4. I don't know about this. Weren't the 50s & 60s the birth of marketing as we know it today? There was a building boom, cars were becoming hugely popular, so many new highways were built, the airline industry started it's rapid growth, and TV was taking over peoples life.

    Still it was before my time. I'm not sure it's related to crime rate either though. I think there are a lot of factors going on. The 80s murder rate was actually due to the video game crash. Kids had nothing to do anymore because they couldn't play their favorite games =p

  5. Using the murder rate as a stand-in for the crime rate over the course of the 20th century is doubly-dubious. Firstly, there is no evidence of any correlation between murder and crime rates at large and secondly, the murder rate has been pressed down by the quantum leap in trauma care enjoyed over the last several decades:

    "had it not been for improvements in surgical technique since 1960, the murder rate would be five times higher than it is"

  6. Firstly, there is no evidence of any correlation between murder and crime rates at large

    Ha, you're retarded. Maybe if you mean there's little correlation with jaywalking, but with violent crime -- what people actually mean by "crime" -- you're dead wrong.


You MUST enter a nickname with the "Name/URL" option if you're not signed in. We can't follow who is saying what if everyone is "Anonymous."