Crime and our common life

Don't you know the crime rate's going up! up! up! up! up!

--"SHATTERED," The Rolling Stones (1978)

Republican presidential candidate Donald J. Trump is asserting a dark, dystopian portrayal of America in 2016 as overwhelmed by predatory criminals and terrorists, in spite of data that show the national violent crime rate has steadily fallen for 25 years. This sharp and sustained turn in the crime rate has created space for the rejuvenation of American central cities, many of which have seen population increases above the national average for a decade or more. For those of us who lived through the three decades before 1990, during which the violent crime rate had steadily risen, the new era has come as something of a miracle, one scarcely-to-be-believed (which may be why many people don't believe it).

Trend in U.S. violent crime, from newgeography.com

According to the Uniform Crime Reports produced by the U.S. Federal Bureau of Investigation, the national violent crime rate in 1970 was 363.5 per 100,000 population. In 1990, it was 731.8, slightly more than double the 1970 rate. Ten years later, it had fallen nearly a third, to 506.5, at the time the lowest since 1978; and in 2014, the last year for which a report has been produced, it was back to 375.7, roughly back to the 1970 level. A preliminary report for the first half of 2015 shows a 1.7 percent increase in violent crime over the same period in 2014--probably statistically significant, but hardly a reversal of the post-1991 trend (Comey 2016). Criminologist James Alan Fox told PolitiFact: "There are some spikes in homicide and shootings in certain cities, yet other cities continue to experience low rates. As a nation, we are far better off than anytime for the past several decades. Crime rates are low, and there is no consistent and reliable indication that things are getting worse" (Jacobson 2016a).

While this scenario played out over much of the country, it has not been playing out uniformly in all areas. Nearly all states experienced the rough doubling of violent crime (2.013x) from 1970-1990. However, Michigan (1.37 times their 1970 rate), Virginia (1.16x) and West Virginia (1.37x) had much lower increases, while the District of Columbia saw its already sky-high rate hold nearly steady (1.10x). On the other hand, a number of states saw much higher rates of increase during this period, including Iowa (3.77x), Massachusetts (3.63x), South Carolina (3.41x), Connecticut (3.25x) and Wisconsin (3.09x).

There is more divergence among states' experiences after 1990. In the chart below, Iowa's violent crime is always lower than that of Illinois, but increased faster before 1990 and its decline thereafter was slight. Iowa never exceeded half of Illinois' rate before 2003, after which it always did. (Source: F.B.I. Uniform Crime Reports, aggregated at disastercenter.com)

STATE  1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984
Illinois    468    477   508   556   627   670   626   631   677   744   808   793   774   728   725
Iowa         79      99     87   102   121   141   133   144   161   181   200   204   173   181   199

STATE 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Illinois   715   800   796   810   846   967   1039  977   960   961   996   886   861   808   690
Iowa      212   235   231   257   266   300     303  278   326   315   354   273   310   312   280

STATE 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Illinois   657   886   602   556   546   552   542   533   528   497   445   424   416   403   370
Iowa      266   268   286   278   288   293   284   295   289   282   269   257   266   273   274

Nationally, violent crime in 2014 was 51.3 percent of the 1990 rate. New York and California, notably, had 2014 rates less than 40 percent of 1990 crime. Seven states, however, saw crime rates actually increase in this period: Montana (203.2%), South Dakota (200.6%), West Virginia (178.4%), New Hampshire (149.1%), Alaska (121.2%), Wisconsin (109.7%), and Nevada (105.8%). Another seven, including my home state of Iowa, had crime rates between 80 and 100 percent of the 1990 level, declines that are unlikely to be perceptible at the level of personal experience. Incidentally...

• the largest metro area in these fourteen states is Las Vegas, Nevada (pop. 2,114,801, rank 29th); most of these states had no metros close to this size. I wonder if the surge and decline in violent crime since 1960 was largely an urban phenomenon?
• The 14 states' 2012 presidential votes were roughly split, but more Republican than the nation as a whole: Mitt Romney won eight of the 14 for an electoral vote advantage of 40-34. This offers mild-at-best support to the thought that the Republican Party's 2016 theme might resonate with their core voters' experiences.

Even at the national level where data are joyous, explanations vary widely, which is hardly going to give people confidence that we know how we got here or how to sustain the decline. Washington Post reporter Max Ehrenfreund (2015) lists five common explanations for the decline in homicides, all of which can be extended to violent crime in general:

1. Larger police forces, funded by the federal Violent Crime and Law Enforcement Act of 1994 as well as state and local initiatives. The ratio of police officers to the general population increased in the mid-1990s, though it had declined to its previous level by 2010. Some analysts credit this with reducing the incidence of crime (Levitt 2004), while others do not (Kleck and Barnes 2010).
2. Use of big data by law enforcement agencies enables better targeting of resources. Specifically identifying certain "hot spots" means police officers can be where the action is, or is likely to be, and their being there can prevent some of it (Roeder, Eisen and Bowling 2015: 67-73).
3. Alcohol consumption per capita, as estimated by the National Institutes of Health, shows a remarkably similar surge and decline with violent crime, rising from around 2 gallons per person through 1958 to a peak of 2.76 (1980-81), and then declining to 2.15 in the late 1990s (Haughwout et al. 2016, Table 1). Since 1999 it has risen back to around 2.3. Should we be alarmed yet? It makes sense that alcohol consumption beyond a certain level is associated with violence, but who knows why national alcohol consumption rises and falls? 
4. The banning of leaded gasoline under the Clean Air Act of 1970 removed a significant hazard to developing young brains, which may have made people less likely to turn violent. However, criminologist Phil Cook notes that the decline in violent crime occurred among all age cohorts, not just those benefiting from the lead ban, so he is skeptical (Kleiman 2014). Meanwhile, many children in older homes--including, in the 1990s, the unfortunate Freddie Gray--still face the hazard of lead paint (Bendix 2016).
5. Sustained economic growth beginning in 1982 preceded the sharp bend in the violent crime curve, and could have accounted for increased confidence and greater actual economic opportunity as well as more money to spend on security. However, it doesn't explain why the declining crime rate has continued past 2000, which years have seen two recessions including one of epic proportions, and when even periods of economic growth have been paradoxically unhelpful to poverty rates and median incomes. And if people in power knew how to fix that paradox, they'd do it, instead of rounding up the same old partisan policy proposals.

Other explanations include (6) the zero-tolerance approach by police to nuisance behavior known as "broken windows" (Wilson and Kelling 1982); (7) "community policing," which is more direct interaction by police with citizens, emphasizing a cooperative approach to problem solving; (8) increased incarceration, thought to decrease crime by incapacitating the most likely offenders; (9) declining proportion of the population in the most crime-prone age group i.e. those between 15 and 24 years of age; (10) widespread repeal of gun regulations and adoption of "right-to-carry" measures; and (11) the end of the surge in use of crack cocaine by the mid-1990s. The Brennan Center for Justice produced in 2015 a comprehensive review of the various theories including results of their own empirical analyses, What Caused the Crime Decline?, which is available online. They find support for theories 1, 2, 3, 4, 5 (broken out various ways) and 9 (for the 1990s only). Needless to say, the other theories have their advocates, too. The point is that the degree of uncertainty about what caused the positive outcomes of the last 25 years can lead to anxiety about whether they can or will be sustained. Or that an equally unexplainable short-term uptick isn't the first sign of Armageddon (Ehrenfreund and Lu 2016), especially if opportunistic political candidates are pushing that story (Jacobson 2016b, "Fact Checker" 2016).

Whatever the reason(s) for the long-term decline in violent crime, public reaction has often been muted-to-disbelieving. In most of the 25 years that the incidence of crime has been reduced, large majorities of the American population have told polls crime is actually on the increase.

Most years since 1990 vast majorities think crime is on the increase. Source: Gallup

Policy analyst Wesley K. Skogan of Northwestern University points out that, of course, beliefs about the nationwide incidence of violent crime is not the same as personal fear of being victimized. In a 2011 paper he reported ten years of survey data from Chicago that showed declining fear of crime, which he attributed to declining incidence of both crime and disorderly behavior, improvement in the physical appearance of neighborhoods and increased confidence in the police. And there's all that movement back to cities to back him up. However, it seems to me that if most people believe overall crime to be on the increase, feelings of personal safety are more vulnerable to change. And confidence in the police has been shaken by a number of widely-publicized police shootings of black men and in some cases their own ham-handed responses (See also Ruud 2014).

A sudden increase in incidents involving firearms--shootings in Chicago in the first half of 2016 were half again as many as the same period in 2015 (Sweeney and Gorner 2016), and on a much smaller scale Cedar Rapids saw a spate of shootings (KCRG 2016)--focus attention on crime, although it's hard to account for the consistency of public perception as depicted above. A small number of spectacular, widely-publicized terror attacks in America and elsewhere in the West can also fuel the impression that danger is increasing. And, as I've shown earlier, the nationwide improvement in violent crime rates has been experienced to varying degrees in particular areas.

Many people blame news media for this misimpression by pushing sensational "if it bleeds, it leads" reporting (See about a zillion sources, such as Matthew Robinson, Media Coverage of Crime and Criminal Justice, Carolina Academic, 2010).  David Rothenberg of the Fortune Society wrote The New York Times after the 2016 Republican convention: "Perception is everything at election time. Any casual surfing of the television channels tonight will show murder after murder. Sadly, some of the most dramatic 'crime stories' never make prime time — the stories of men and women being released from prison, facing barriers that limit housing and jobs, fighting personal demons while navigating societal restrictions, struggling just to get through the day. Such stories are the ones that should be surfacing at political conventions. The fiction lives while the truth remains in the shadows." For this reason, violent crime gets far more attention per incident than do fatal auto crashes (Shultis 2016).

Both the reality and perception of crime are important to the future of cities. Urban criminals most frequently victimize the most vulnerable of our citizens; as Dreier et al. point out (Place Matters: Metropolitics for the Twenty-First Century, Kansas University Press, 2001, 202-203) most residents of high-crime urban neighborhoods are hard-working and law-abiding, but suffer from predation by a few of their neighbors. To compete successfully with suburbs for residents and businesses, cities need at minimum to be clean and safe (Duany et al., Suburban Nation: The Rise of Sprawl and the Decline of the American Dream, North Point, 2000, 157-158). The dangerous reputation of cities makes many people fearful of the population density we need to be environmentally and fiscally sustainable (Hester, Design for Ecological Democracy, MIT Press, 2006, 205).

Crime and the fear of it threaten to undermine our efforts to build a common life. A common life in the 21st century depends on the success of cities: It's where a large percentage of our citizens live and more are moving, urban design is the most ecologically and fiscally resilient, and diversity and the potential for inclusion are greatest. To the extent Trump's dystopian rhetoric resonates with a significant part of the American population, cities and the people who care about them must strive to improve their reputation.

ON THE SAME TOPIC: Josh Stephens, "Trump to Cities: You're Dead to Me," Planetizen, 26 July 2016, http://www.planetizen.com/node/87620/trump-cities-youre-dead-me

NOTE: The Department of Justice compiles data from the FBI and other agencies at the Bureau of Justice statistics page. However, I find other aggregators easier to use; for this piece I used the same data as found at Disaster Center.