Mugged by reality

Progressives of the younger generation may wonder what led to the curious phenomenon of “neoliberalism.”  Why did some 1960s progressives gradually become more conservative over time?  There are many reasons, but crime is one issue that almost certainly played a role.  Take a look at the US murder rate from 1950 to 1962 (per 100,000):

1950 – 4.6

1951 – 4.4

1952 – 4.6

1953 – 4.5

1954 – 4.2

1955 – 4.1

1956 – 4.1

1957 – 4.0

1958 – 4.8

1959 – 4.9

1960 – 5.1

1961 – 4.8

1962 – 4.6

No net change, but pretty high.  Suppose that like all good liberals you believe that America’s unusually high murder rate is due to poverty and guns.  What would you expect to happen to that murder rate if the US succeeded in dramatically reducing the poverty rate over the next 12 years?  You’d obviously expect a sharp fall in the murder rate. First, here’s what actually happened to poverty:

Screen Shot 2015-03-18 at 9.21.16 PMTake a look at the poverty rate in 1962, about 21%.  That’s down slightly from previous decades, but still pretty high.  Then something dramatic happened, poverty started falling really fast.  In 1964 LBJ announced a “War on Poverty,” although to be honest the decline began even before those programs really kicked in a few years later.  I’d guess the civil rights movement also played a role, as did the economic boom of the 1960s. By 1974 the poverty rate was less than 12%.

And here’s what happened to the murder rate during the golden boom years of plentiful jobs for the working class and amazing poverty reduction:

1963 – 4.6

1964 – 4.9

1965 – 5.1

1966 – 5.6

1967 – 6.2

1968 – 6.9

1969 – 7.3

1970 – 7.9

1971 – 8.6

1972 – 9.0

1973 – 9.4

1974 – 9.8

OK, 1974 wasn’t so “golden,” but you get my point.  The murder rate stayed high until 1991 and then started falling.  In 2012 it’s back down to 4.7, having fallen from 5.9 to 5.0 just during the Great Recession of 2007-09 (as poverty soared). And the big crime surge of 1963-74 was not due to riots, which petered out after 1970.

The huge crime wave of the 1960s and 1970s was one of the things that finished off old liberalism.  Old liberalism is now being revived by a generation of progressives too young to remember why their parents or grandparents became disillusioned.

BTW, other forms of crime such as rape, robbery, burglary, and assault increased even more rapidly during the 1960s and 1970s.  However the murder rate data is regarded as more reliable.  I was very young back then, but I do recall that people were pretty shocked by the surge in crime.  Nixon was elected president on that issue. All the liberal east coast states except Massachusetts voted for Nixon in 1972. Imagine someone running today (for federal office) on the crime issue. Weird.

I don’t have any answers here; just that poverty reduction is not a cure for crime. Poor people are more likely to commit violent crimes, but we don’t really know why.

The New York Review of Books has an interesting piece on the War on Poverty. I chuckled when they said that neither liberals nor conservatives could make up their minds as to whether poverty has declined sharply (the war worked!) or whether poverty has declined sharply (the war is no longer needed!)  As soon as the liberals and conservatives figure out what a decline means, they’ll all line up predictably on each side the debate over empirical studies.  The article also contains this gem, as if we needed any more reasons to ignore all income inequality data:

Cohabiting couples. Imagine two twenty-five-year-olds who are romantically involved, live together, and each earned $12,000 in 2013. If they were unmarried, the Census Bureau would have classified them as unrelated individuals, with poverty thresholds of $12,119 each. Since their incomes were only $12,000, the bureau would have counted them both as poor. They would each have needed at least $12,199, bringing their total household income to at least $24,238, for the bureau to stop counting either of them as poor.

Had they been married, however, the bureau would have taken a more upbeat view of their economic situation, classifying them as a family of two with a poverty threshold of $15,600. As a result of this change they would both have been above their poverty threshold instead of below it. According to the Census Bureau’s most recent data, 11 percent of all opposite-sex couples who lived together in 2012 were unmarried.2 We don’t have such a figure for 1964, but it was probably only 1 or 2 percent. The assumption that cohabiting couples need more income than married couples has therefore raised the official poverty rate. This increase in the poverty rate would make sense only if the absence of a marriage license increased a couple’s expenses by 55 percent (from $15,600 to $24,238). The Census Bureau has never tried to defend that assumption, presumably because it is a byproduct of rules set by the Office of Management and Budget, which the Census must follow whether it likes them or not.




36 Responses to “Mugged by reality”

  1. Gravatar of Bob Murphy Bob Murphy
    21. March 2015 at 17:47

    The answer is obvious, Scott.

  2. Gravatar of John Voorheis John Voorheis
    21. March 2015 at 18:04

    In re: the last bit, note that left unmentioned in the article is that the Census Bureau has developed a much better poverty measure (the SPM) that addresses all of the issues mentioned.

  3. Gravatar of Jason Smith Jason Smith
    21. March 2015 at 18:32

    I’m fond of the leaded gasoline theory …

  4. Gravatar of E. Harding E. Harding
    21. March 2015 at 18:37

    I’m fond of looking at changes in the age structure.

  5. Gravatar of Don Geddis Don Geddis
    21. March 2015 at 18:45

    Freakonomics credited legalized abortion (starting in 1973), for the observed reduction in violent criminals (young males with poor impulse control) two decades later.

  6. Gravatar of Matt Waters Matt Waters
    21. March 2015 at 19:28

    The main paper referenced for the Kevin Drum article also has legalized abortion as a state-level factor. She estimates most of the decrease in violent crime came from lead, but a significant portion also came from legalized abortion.

    Lead was not significant for murder rates, but it had a p-value very close to significant and murder rates have the additional complication of varying trauma care. Much of the decrease in murder rates has been due to better trauma care, which isn’t really a decrease in crime.

    Krugman linked to an article a couple of weeks ago that poverty can cause poor social behavior rather than the other way around. I remember not being able to find the underlying data in what he linked to.

    That said, most of the poverty reduction was rural poverty. Blacks moved up north in the 40’s for industrial jobs which were later cut by competition or automation. Although the inner city blacks had access to the same safety net program, having many poor people without jobs in high density could have been a contributing factor. A factor which was not there for the rural poverty.

  7. Gravatar of Lorenzo from Oz Lorenzo from Oz
    21. March 2015 at 19:34

    I am surprised that a film buff like yourself did not note the surge in vigilantes-as-heroes-films. Though the crime angst shows up in all sorts of ways in the films of the period: “Escape from New York” springs to mind.

  8. Gravatar of Steve Steve
    21. March 2015 at 21:29

    This is a fascinating topic — I’d be interested in hearing more thoughts.

    Folks on this board have proposed abortion and lead as big factors in the 60s and 70s.

    I also wonder if demographics, i.e., baby boomers reaching angry young adulthood circa 1964, has played a role. Oh, and psychedelic drugs.

    I’ve also believed that the large fall in crime in recent times has correlated to technology, especially surveillance cameras. This happens partly by reducing recidivism among repeat criminals, but also by serving as a powerful deterrent to would-be-criminals.

    Nowadays, there’s a shift underway from isolated violent criminals, to mass terror events, driven by a new technology: social media.

  9. Gravatar of Ray Lopez Ray Lopez
    21. March 2015 at 21:30

    Racist and outdated theories trotted out… this blog is so predictable. Bonus: not only is over 40% of the Fed balance sheet in mortgages since 2008, a record high (which I choose to call ‘toxic’) but Sumner says that the percentage is zero! You can Google yourself that I am right. Pied Piper.

  10. Gravatar of Brett Brett
    21. March 2015 at 21:56

    @Jason Smith

    Same here, especially since the pattern seems to hold with crime waves in other countries as well upon the introduction of increased environmental lead contamination. I’d probably go with a combination of lead exposure and demographic change (abortion helped, but the crime demographic age-group was shrinking in size already and abortion rates were already quite high even before Roe). Poverty doesn’t appear to correlate much at all, and even the unemployment correlation is rather weak although popular.

  11. Gravatar of Greg Ransom Greg Ransom
    22. March 2015 at 00:57

    Reduced crime punishment and lowered likelihood of crime punishment + urban renewal cultural destruction = more crime

  12. Gravatar of W. Peden W. Peden
    22. March 2015 at 02:15

    Bob Murphy,

    That explains the fall in US poverty (GDP per head was raised significantly once you were able to hear our best musical acts) but what explains the rise in crime?

    I blame the decline of epic cinema, which meant that young people had more free time doing things other than sitting all the way through Ben Hur, and the devil finds work for idle hands. Once video games began to get really good, youths had an alternative way of spending their time enjoying violence without standing up, and the crime rate fell. In other words, I credit violent movies and violent video games with causing falls in crime.

  13. Gravatar of W. Peden W. Peden
    22. March 2015 at 02:16

    Also, compare crime rates in South Korea and countries of comparable economic status. If anything, kids playing violent video games should be subsidised.

  14. Gravatar of W. Peden W. Peden
    22. March 2015 at 02:20

    Matt Waters,

    I seem to recall that a problem with the abortion explanation is that the timing of the decline in crime is very similar across developed countries, but different countries liberalised abortion at different times, e.g. the big milestone in the UK was the Abortion Act 1967, six years before Roe vs. Wade, but the beginning of the fall in our crime rate was about precisely the same time as in the US.

    Of course, we know that the actual reason was the release of Street Fighter II in 1991.

  15. Gravatar of W. Peden W. Peden
    22. March 2015 at 02:22

    In fact, I checked on Wikipedia and 1991 also saw Super Mario World, Sonic the Hedgehog, A Link to the Past, Alien Breed, and Civilization. The net virtual bodycount in those games is in the trillions. This proves irrefutably that violent video games are the cure for crime.

  16. Gravatar of W. Peden W. Peden
    22. March 2015 at 02:25


    As far as I know, there’s not a very strong link between most psychedelic drugs and crime, e.g. LSD and crime. Heroin and crack cocaine are obviously related to crime.

    However, there is a strong link between the declining fortunes of violent epic Westerns and crime. It was in 1992 that Unforgiven revived the epic violent Western movie, contributing towards the decline in crime.

  17. Gravatar of Daniel Daniel
    22. March 2015 at 03:55

    The day when social scientist (I’m using the term “scientist” very loosely here) actually grasp the meaning of “correlation does not imply causation” is the day they’ll stop being clowns.

  18. Gravatar of Nick Nick
    22. March 2015 at 04:14

    This whole topic is a big face / vase situation. Prof sumner notices this with regards to the war on poverty .. Did we win … Does that prove it worked and so we need to quit or it that it works and we need more?
    Unfortunately, his view of violence is not as nuanced. The decline in violence in the 90s is just the same as the big reduction in poverty in the late 60s … does it prove the war on crime worked? And would that mean we need more or less of it?
    It’s just the same, and both drops (and the rise in crime) have a ton of confounding factors. The new ‘old liberals’ haven’t forgotten that liberalism was discredited by violence in the 70s and 80s … They think the last 20 years have discredited the discrediting.

  19. Gravatar of Nick Nick
    22. March 2015 at 04:54

    Face / Vase
    let’s say it’s the year 2001 and you thought AFDC reform turned out great in the 90s … Would that prove that people who want to spend more on welfare are just dumb or that our welfare system had become more efficient and our next step should be to increase funding elsewhere?
    It’s a continual wonder to me that conservatives have yet to take this approach on healthcare. It is very possible in politics to win in the big picture by ‘losing’ a debate. Maybe when they retake the White House. But the effort expended on fighting the ACA these last few years could have much more easily retrenched a different part of the welfare state, with larger implications for taxes and growth.
    ‘Now that the safety net includes healthcare, do we really need X or Y inefficient programs?’
    but I guess if they focused on getting rid of just the worst programs, all they would do is strengthen the case for increasing funding to the remaining ones in the long run…

  20. Gravatar of ssumner ssumner
    22. March 2015 at 05:42

    Bob, Bingo.

    John, Yes, I also prefer that measure. I’m pretty sure it would also show a big drop in the 1960s and early 1970s.

    Jason, I’m not convinced, but it probably had some effect. I did mention that hypothesis in an earlier post on the millennials.

    Don, I’m also skeptical of that theory, at least the amount of reduction it explains.

    Matt, You said;

    “Although the inner city blacks had access to the same safety net program, having many poor people without jobs in high density could have been a contributing factor.”

    You missed the point. Crime surged when the availability of jobs increased rapidly.

    Lorenzo, Good point.

    Steve, I’m pretty sure that technology has played a role. For instance, car thefts fell very sharply with better anti-theft controls.

    Ray, I think you must have breathed in too much lead pollution. The Fed’s MBSs are backed by the Treasury, there is zero risk. Toxic bonds? ROFL.

    Greg, I think that was the most important factor. Punishment was reduced very sharply during this period. That’s the most straightforward application of EC101, I wonder why it took so long for anyone to mention it.

    W. Peden, Isn’t (or wasn’t) there a lot of lead pollution in East Asian countries? What sort of murder rate do they have?

    And yes, the drug war is certainly a factor.

    Nick, Please don’t lump me in with these stupid “conservatives.” Those aren’t my views.

  21. Gravatar of benjamin cole benjamin cole
    22. March 2015 at 07:05

    Incarceration rates soared; crime went down.
    And I think a message was sent when they put Nixon behind bars.

  22. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    22. March 2015 at 07:38

    The famous sociologist Jerry Lewis even checked in;

  23. Gravatar of Don Geddis Don Geddis
    22. March 2015 at 09:15

    @Daniel: “correlation does not imply causation

    But if it weren’t for spurious correlations, how would we ever have discovered the close connection between Nick Cage movies and swimming pool drownings?

  24. Gravatar of Jose Romeu Robazzi Jose Romeu Robazzi
    22. March 2015 at 10:03

    @Greg, @Sumner
    I am not familiar with most of the historical context in the US, but I live in Brazil, we have very lenient laws, our police system is a mess, according to the UN Brazil had 56k muders in the last year we have data on, we have 25+ muders for every 100k people, only 5% of these muders get “solved” by the police, yes, I would say that lack of punishment plays a very important role indeed … (Brazil is certainly the most violent among those countries with population above 100 million). We easily beat Mexico and Nigeria …

  25. Gravatar of collin collin
    22. March 2015 at 10:58

    Growing up in Socal in the early 1990s, the fall in crime is absolutely astonishing as the gang violence was incredible back in the day. Nobody was predicting this huge drop. Why?

    1) Abortion and long term increased birth control
    2) Cashless society
    3) Lower use of hard drugs (This one I can’t explain either)
    4) Increased incarnation
    5) Increased liberation of minor vices to decrease petty criminals
    6) I will give Kevin Drum his worth: Lead

  26. Gravatar of Browsing Catharsis – 03.23.15 | Increasing Marginal Utility Browsing Catharsis – 03.23.15 | Increasing Marginal Utility
    23. March 2015 at 04:00

    […] “Mugged by Reality,” by Scott Sumner. […]

  27. Gravatar of ssumner ssumner
    23. March 2015 at 05:20

    Jose, Thanks for that info.

    Collin, Good post.

  28. Gravatar of Matt Waters Matt Waters
    23. March 2015 at 08:19

    W. Pedan,

    These time-series have a lot of spurious regressions, such as Street Fighter II, because 5% of all data-sets will have a p-value less than 5% by pure luck. So you can only really make causal inferences with many studies, all showing low p-values, as well as directly causal studies such as animal studies.

    For lead and abortion, the regression for the paper was modeling crime by year AND BY STATE. So you have a collection of 50 crime-years for each year. Some states legalized abortion before Roe, at different times, and some states removed lead at different times. So the regression is not on a general time trend, like Street Fighter II, but on how the trends vary between different states. Street Fighter II went to all 50 states at the same time.


    The unemployment rate did up significantly in the 70’s and early-80’s. The unemployment effect may have also disproportionately affected blacks who moved to urban inner-cities. I only see the general poverty rate in the post, but like I said, poverty could go down with safety net programs while unemployment itself leads to crime.

    Still, one has to disentangle all of the orthogonal effects to possibly get at how unemployment may have lead to even more crime, on top of lead in gasoline.

  29. Gravatar of Floccina Floccina
    23. March 2015 at 10:18

    Great post Scott.
    Interestingly, if you can get along well with others you can live well on less that poverty line income. The last time that I looked at it the University of Florida said incoming students planning to live off campus should budget $11,000/year for living expenses, but the poverty is below $12,199.
    I am rich and I spent about $50k per year for my family of 4 a few years back when both my children lived with me. So that is 12,500/person and I spent more than $10,000/year on vacations and we ate out much more than typical middle class families did the 1970’s.

    Poor people are those unable to earn much income and who cannot get allow with others well (or often work well with others). Just low income can be overcome.

  30. Gravatar of Floccina Floccina
    23. March 2015 at 10:27

    I once read a theory that blacks deaths due to homicide (and crimes where the victim was black in general) where under reported prior to 1965.

  31. Gravatar of Charlie Jamieson Charlie Jamieson
    23. March 2015 at 10:35

    The drop in the murder rate from 1990 is driven by the falling rates in the black community.
    The murder rate among blacks in the early 1990s was 39.4, now it’s 19.6.
    The murder rate by whites, 3.3 today, is largely unchanged.
    Just guessing here, but I’d say the drug trade is more settled than it was in the 1990s when it was truly the Wild West out there.
    And probably the three strikes laws, etc., are putting potential murderers in jail.

  32. Gravatar of ssumner ssumner
    23. March 2015 at 16:10

    Matt, You said:

    “The unemployment rate did up significantly in the 70’s and early-80’s.”

    My point is that the crime rate soared BEFORE the rise in unemployment, indeed during the best job market in our history. Anyone who could fog a mirror could get a job in the late 1960s.

    BTW, I was “poor” in the 1970s, but didn’t murder anyone. (Now that’s a softball that anyone should be able to hit out of the park.)

    Charlie, Maybe, but those numbers don’t seem to add up. If the white rate stays at 3.3, and the overall rate falls from 10 to 5, then the black rate would have to fall by far more than 50%. Are your sure those numbers are accurate?

  33. Gravatar of Jim Glass Jim Glass
    23. March 2015 at 18:22

    Here’s the real deal on what happened to crime, in detail:

    The City That Became Safe: New York’s Lessons for Urban Crime and Its Control (Studies in Crime and Public Policy)

  34. Gravatar of Benny Lava Benny Lava
    25. March 2015 at 04:49

    This is similar to the thesis of the book Nixonland. Have you read it?

  35. Gravatar of ssumner ssumner
    25. March 2015 at 05:10

    Benny, No.

  36. Gravatar of TallDave TallDave
    27. March 2015 at 11:42

    The problem is that “poverty” as used above only represents the relative income distribution, not the living standards of the poorest deciles.

    I actually have a lot of sympathy for the progressive notion that reductions in poverty can reduce crime (though it’s trivially obvious that reductions in crime also reduce poverty); when one looks at absolute poverty this seems to be true, at least for very low levels of consumption.

    But of course looking at absolute poverty is really inconvenient for progressives, because it’s always declining.

Leave a Reply