Contents

• Violent Crime and Homicide Trends in the United States
• Select Theories About Why Violent Crime is Increasing in Some Cities
• The "Ferguson Effect" and Issues of Police Legitimacy
• Increases in Gang Violence
• Repeat Violent Offenders
• Is There a Need for "Real Time" Crime Data?

Summary

On August 31, 2015, the New York Times ran an article with the headline "Murder Rates Rising Sharply in Many U.S. Cities." The story highlighted double-digit percentage increases in homicide rates in several cities, and came on the heels of reports from other media outlets of recent spikes in violent crime in cities across the country. Accounts of rising violent crime rates in some cities have generated speculation about whether the United States is in the midst of a new crime wave.

Overall, homicide and violent crime rates have been trending downward for more than two decades, and both rates are at historic lows. An analysis comparing 2014 and 2015 homicide data from the nation's 60 most populous cities suggests that violent crime is not increasing. Overall, reported homicides were up 16% in 2015, but a majority of cities (44 of 60) have not seen a statistically significant increase in homicides. The general consensus is that it is too early to draw any conclusions about the reversal of long-term trends. Also, even if homicide and violent crime rates do increase this year, it may not portend a break in the long-term trend. Even though both rates have been on a downward trend since 1990, there were years where either the homicide rate or violent crime rate increased.

There are several short-term factors that might help explain some of the reported upticks in violent crime across the country.

• Year-to-year changes in crime rates can be subject to random fluctuations.
• Crime rates are subject to seasonal effects.
• Many cities are experiencing increases from historically low levels of crime.
• Percentage change in reported crimes is a relative measure and is sensitive to magnitude.

While it might be too early to make any definitive conclusions about whether violent crime is on the rise, several commentators have speculated as to why some cities are experiencing spikes in violent crimes. Suggested explanations include the following:

• The "Ferguson effect" (i.e., in the wake of a spate of high-profile officer-involved deaths, police have become reluctant to engage in proactive policing, thereby emboldening criminals).
• Law enforcement is facing a legitimacy problem in some communities where residents feel that they are not treated fairly by the police, and this may mean that people are more likely to take matters into their own hands when conflicts arise.
• The increase in violence can be attributed to battles between gangs for control of drug turf or released violent offenders committing new crimes.

The recent discussion about the increases in violent crime in some cities might raise the question of whether there is a need for more "real time" nationwide crime statistics. More frequent and consistent crime data might be able to provide greater insight into crime trends. However, there are logistical issues involved with collecting and reporting timely and accurate crime statistics from the nation's approximately 18,000 law enforcement agencies.

---

Is Violent Crime in the United States Increasing?

On August 31, 2015, the New York Times (Times) ran a story with the headline "Murder Rates Rising Sharply in Many U.S. Cities."¹ The Times reported that at least 35 cities had seen increases in violent crime² compared to 2014, but the story specifically highlighted increases in reported homicides in 10 cities. According to the Times, homicides have increased

• 76% in Milwaukee, WI;
• 60% in St. Louis, MO;
• 56% in Baltimore, MD;
• 44% in Washington, DC;
• 22% in New Orleans, LA;
• 20% in both Chicago, IL, and Kansas City, MO;
• 17% in Dallas, TX;
• 9% in New York City; and
• 4% in Philadelphia, PA.³

The Times story followed reports from other media outlets about a growing number of violent crimes in some cities.⁴

The recent increases in violent crime have grabbed the attention of law enforcement officials. The Major Cities Chiefs Association held a summit in Washington, DC, in August to discuss why it was increasing in some cities and what could be done to reduce it.⁵ The Department of Justice (DOJ) held a meeting on October 7 with officials from 15 cities to discuss how DOJ's resources could be utilized to help combat increases in violent crime.⁶

After years of declining crime rates, policymakers might consider legislation to reduce or eliminate some mandatory minimum sentences and reduce the number of people held in prison.⁷ The changes policymakers might consider could roll back some of the changes Congress made to criminal justice policy during the 1980s and early 1990s in response to long-term rising crime rates.⁸ In general, these changes increased sentences, especially for violent and drug offenses, and ensured that inmates served a greater proportion of their sentences in prisons. However, some observers are now concerned that recent reports of an increasing number of violent crimes, especially homicides, in some cities might stymie criminal justice reform in Congress.

There has also been a spate of high-profile police-involved deaths over the past year.⁹ This has led to calls for legislation to reform police policies and tactics. Some observers are concerned that reports of rising numbers of violent crimes might thwart efforts to increase police accountability if those efforts are viewed as hampering law enforcement's ability to control crime.

Has the United States reached the end of, as one criminologist characterized it, the "great American crime decline?"¹⁰ The general consensus is that it is too early to draw any conclusions about the reversal of long-term trends. Also, there are several explanations for why some cities might be experiencing an increase in violent crime other than that the recent era of diminishing violent crime is coming to a close.

Violent Crime and Homicide Trends in the United States

Reports of an increasing number of violent crimes in some cities do not necessarily mean that the United States is in the midst of a crime wave. Violent crime and homicide rates have been trending downward for more than two decades (Figure 1). The nation's violent crime rate¹¹ in 2014 was the lowest it has been since 1970, while the homicide rate was the lowest since 1960. Even if homicide and violent crime rates are on the upswing in 2015, the homicide rate would have to increase by 5.7 homicides per 100,000 people to reach the post-1960 high of 10.2 per 100,000, while the violent crime rate would have to increase by 392.7 violent crimes per 100,000 people to equal the post-1960 high-point of 758.2 per 100,000 in 1991. Both of these increases would represent more than a doubling of the 2014 homicide and violent crime rates.

As the figures below indicate, Uniform Crime Report (UCR) data from the Federal Bureau of Investigation (FBI) lags by a year.¹² Data on a rising number of homicides or other violent crimes generally comes either from a law enforcement agency or a state UCR program.

Figure 1. National Violent Crime and Homicide Rates, 1960-2014 Rate per 100,000 people

Source: Sourcebook of Criminal Justice Statistics, Table 3.106.2012; Federal Bureau of Investigation, Crime in the United States 2014, Table 1. Notes: The violent crime rate is presented on the left vertical axis, while the homicide rate is presented on the right vertical axis. The violent crime rate includes the total number of reported homicides, rapes, robberies, and aggravated assaults per 100,000.

Data show that since 1990, homicide and violent crime rates in large and medium-sized cities¹³ have mirrored national trends (Figure 2 and Figure 3). All cities with populations of 50,000 or more had lower homicide and violent crime rates in 2014 compared to 1990, with the largest cities (i.e., cities with populations of 250,000 or more) experiencing the greatest decreases between those years.

Figure 2. Violent Crime Rates, by Population of Jurisdiction, 1990-2014 Rate per 100,000 people

Source: Federal Bureau of Investigation, Crime in the United States 1990-2014. Notes: The data presented in Figure 2 start in 1990 to show how violent crime rates changed in jurisdictions of different sizes after the nation's violent crime rate started to decrease.

Figure 3. Homicide Rate, by Population of Jurisdiction, 1990-2014 Rate per 100,000 people

Source: Federal Bureau of Investigation, Crime in the United States 1990-2014. Notes: The data presented in Figure 3 start in 1990 to show how homicide rates changed in jurisdictions of different sizes after the nation's homicide rate started to decrease.

In general, crime data should be viewed over longer time periods in order to determine trends.¹⁴ For example, even though violent crime and homicide rates have been on a downward trend since the early 1990s, there were years where one or both increased, but those year-to-year increases did not portend a break in the overall trend. Evaluating crime data in short intervals (quarterly or monthly, for example) can amplify the "noise"¹⁵ in the data and make it harder to distinguish the underlying trend.¹⁶

There are several factors that might help explain some of the reported upticks in violent crime across the country:

• Year-to-year changes in crime rates can be subject to random fluctuations and not related to how the police do their job.¹⁷ For example, a short-term but intense dispute between rival gangs might lead to an increase in reported violent crimes.
• Crime is subject to seasonal effects.¹⁸ It tends to increase in the summer and decrease as the year goes on. So a year with a flood of homicides in July may still end up being roughly in-line with the previous year's total by December.
• Increases in crime might also be indicative of past successes (i.e., many of the reported increases in homicides are from cities where homicide rates are at historic lows, therefore it is possible that, at some point, homicides rates were going to increase compared to the previous year).¹⁹
• Percentage change in reported crimes is a relative measure and is sensitive to magnitude.²⁰ For example, an increase in 10 homicides in a city with 20 homicides the previous year would represent a 50% increase year-over-year, but an additional 10 homicides in a city with 100 homicides the previous year would only represent a 10% increase. In addition, percentage change can sometimes be misleading depending on the time period chosen.²¹ For example, if a city had 20 homicides by the end of March 2014 but recorded 25 homicides by the end of March 2015, that would represent a 25% year-to-year increase in the number of homicides. However, if that gap persisted and the city ended the year with 105 homicides compared to 100 homicides in 2014, it would represent a 5% increase in the number of homicides.

Also, it is possible that increases in violent crime in some cities do not portend a nationwide increase. An analysis of homicide data from a broader array of cities than those cited in the Times article provides a more nuanced insight into the issue. Data from the nation's 60 most populous cities show that reported homicides were up 16% overall and up by 20% or more in 26 cities.²² On the other hand, homicides were down in 20 other cities. The majority of cities (44 of 60) have not seen statistically significant increases in homicides, but the overall increase in homicides is statistically significant.²³ Also, it is not rare for there to be large fluctuations in the reported number of homicides from year to year. For example, 17 of today's 60 most populous cities had statistically significant decreases in the number of reported homicides in 2009, and in 2005 there were 15 cities that had statistically significant increases in the number of homicides.²⁴

It is probably too early to draw any definitive conclusions about whether the country is experiencing a reversal of decreasing violent crime rates. Even if final crime data show an increase in homicides or violent crime in 2015, if history is any indicator a one-year increase does not necessarily suggest a reversal of the nation's two-plus-decade crime decline.

Select Theories About Why Violent Crime is Increasing in Some Cities

While it might be too early to make any definitive conclusions about whether there has been a reversal of the decades-long decrease in violent crime, several commentators have speculated about why some cities are experiencing spikes in violent crime. This section of the report provides a brief overview of some of the more frequently discussed possibilities.

The "Ferguson Effect" and Issues of Police Legitimacy

The "Ferguson effect" is one of the more widely discussed, and controversial, explanations for the recent increases in violent crime in some cities. It posits that protests over police-involved shootings and attempts to reform how police use force are to blame for the recent increases in some areas.²⁵ Specifically, this theory suggests that in the wake of recent high-profile officer-involved deaths, the police have become reluctant to engage in proactive policing, thereby emboldening criminals.

Proponents of the "Ferguson effect" cite what is largely anecdotal and correlational evidence to support their claim. For example, it has been argued that the police in some cities have limited their use of proactive policing techniques, leading to more homicides.²⁶ It was reported that by November 2014, arrests were down one-third in St. Louis city and county after the August 2014 shooting of Michael Brown in Ferguson, MO.²⁷ By early November, homicides in the city were up 42%.²⁸ Baltimore recorded 43 homicides in May 2015 after rioting resulting from anger over the death of Freddie Gray while he was in police custody.²⁹ At the same time, arrests were down 57% compared to the previous year.³⁰

On the other hand, data show that the increase in homicides in St. Louis started before Michael Brown's shooting, and arrests were down and homicides were increasing in Baltimore before Freddy Gray died.³¹ Also, arrests might have decreased for reasons other than police officers believing they could not do their jobs due to public criticism of how they use force. For example, the reduced number of arrests in Baltimore can be attributed to staffing shortages and the implementation of a new patrol strategy that preceded Freddy Gray's death.³²

In addition, one criminologist has argued that efforts to reform police and reign in the most egregious abuses have not prevented police from being effective crime fighters.³³ He points to the New York City Police Department (NYPD) as an example. He writes,

There is one longer-term test of whether [the "Ferguson effect"] is fact, fiction or something in between. In New York City, concern about police aggressiveness has been a cottage industry for more than two years, as the number of stops has plummeted—yet crime levels that were historically low three years ago have stayed just as low. The police department has been under legal scrutiny and political pressure, but there is no evidence that operational efficiency has suffered. And New York City crime statistics remain astonishingly good compared to other cities.³⁴

Another theory is that law enforcement is facing a legitimacy problem in communities where residents feel that they are not treated fairly by the police. This might be considered the other side of the argument with respect to the "Ferguson effect": law enforcement's effectiveness is not hampered because police are reluctant to engage in proactive policing, it is hampered because the community does not want to work with law enforcement to prevent or solve crimes. Two criminologists argue that when people lose trust in the police they are more likely to take matters into their own hands when conflicts arise.³⁵ Specifically, they write,

Research finds that law is most effective when it is perceived as having high levels of legitimacy. According to studies, when people have trust and confidence in the criminal justice system, they are more likely to act in accordance with the law. When citizens question whether police and other criminal justice officials act in a fair and impartial manner, when they fear that interactions with the police will result in unwarranted levels of bodily harm, and when they doubt that police will thoroughly investigate questionable actions by fellow officers, legal cynicism, in which people perceive the law as illegitimate, unresponsive, and ill equipped to ensure public safety, is likely to result.³⁶

These researchers also note that one of the implications of the "Ferguson effect" is that critics of law enforcement practices must support the police or violence will be allowed to grip the city. This approach, however, might reduce police legitimacy is the eyes of some communities because it can make members of those communities, who may believe that they have legitimate complaints about their relationship with the police, feel that their concerns are not valid.

Increases in Gang Violence

A rise in gang violence may be to blame for the increases in violent crime in select urban areas. Half of the 35 cities surveyed by the Major Cities Chiefs Association identified "gang-related activity and retaliatory violence" as a reason for violent crime increases.³⁷ Gang violence is episodic. For example, a shift in drug markets might trigger fights between rival gangs for control of turf, the murder of one gang member might result in retaliatory violence, or the arrest of a gang leader might lead to a fight for control of the gang.³⁸

Largely because of its episodic nature, this explanation for increasing violent crime is likely highly localized, meaning that it is unlikely that rising gang violence provides an explanation for any perceived national violent crime wave.³⁹ It is possible that an increase in gang crime might be driving an increase in violent crime in one city, but it does not necessarily mean that other cities are experiencing a concomitant increase in gang crime, thereby explaining why violent crime might have increased in those cities. For example, it was reported that prescription narcotics were stolen from stores and clinics during riots in Baltimore.⁴⁰ This might have led to gangs fighting over turf for distribution of the stolen drugs.⁴¹ If this is true, it might provide an explanation for the increased violent crime in Baltimore, but it might not explain why violent crime increased in another city.

Repeat Violent Offenders

Both Washington DC Police Chief Cathy Lanier and Chicago Police Superintendent Garry McCarthy have identified repeat violent offenders as possible sources of the growing number of homicides in their respective cities. Police Chief Lanier was quoted as saying, "[w]e are seeing far too many of our repeat violent offenders out here being reckless with firearms over and over again. There is a push to release a lot of people."⁴² Superintendent McCarthy noted, "[a]cross the country, we've all found it's not the individual who never committed a crime before suddenly killing somebody. It's the repeat offenders. It's the same people over and over again."⁴³

Recidivism amongst violent offenders is a stubborn problem. The most recent data from the Bureau of Justice Statistics (BJS) indicates that approximately 71% of violent offenders released in 2005 were rearrested within five years;⁴⁴ though only about one-third (33%) of these offenders were rearrested for a violent offense.⁴⁵ Data also suggest that a small proportion of released offenders accounted for a disproportionate number of post-release arrests: 16.1% of released inmates accounted for approximately half (48.4%) of all of post-release arrests of the 2005 release cohort.⁴⁶

On the other hand, hundreds of thousands of released inmates have been returning to communities across the country for more than two decades. On average, approximately 595,000 inmates were released from state and federal prisons each year between 1990 and 2014.⁴⁷ The number of released inmates increased every year between 1990 and 2008 before decreasing five of the six years between 2009 and 2014. As discussed above, the nation's homicide and violent crime rates were generally decreasing at the same time the number of inmates released from prison was increasing. It is possible that recent cohorts of released inmates included a disproportionate number of ex-offenders with violent proclivities, but currently available data does not provide any insight into this possibility.

Is There a Need for "Real Time" Crime Data?

There is discussion about whether increasing violent crime rates in some cities highlights the lack of nationwide "real time" crime data. As previously mentioned, Uniform Crime Report (UCR) data published by the FBI lags by nearly a year. Research on homicides in the 60 most populous cities in the United States was collected from a variety of sources.⁴⁸ Data for the analysis were taken from police departments' websites, state UCR programs, media reports, and the survey conducted by the Major Cities Chiefs Association. In some instances, police departments had to be contacted directly. Also, the data available were not always consistent. Some cities only reported data on the number of homicides that had occurred as of May 31, 2015. Others had data available as of September 7, 2015.

More frequent and consistent crime data might be able to provide more insight into crime trends. It has been reported that the FBI is in the process of redesigning and redeveloping the UCR program, with one goal being the more frequent publishing of crime data.⁴⁹ Up-to-date national crime data could show whether the increases in homicides in some cities are a part of a national trend or a more localized issue. It could also show whether homicides are increasing as part of an overall increase in violent crime or if the increase in homicides is its own phenomenon.

It has been argued that the federal government publishes economic and labor data more frequently than it does crime data. The FBI notes that collecting and reporting data is labor intensive: there are approximately 18,000 law enforcement agencies in the United States and the FBI checks crime data before it is published.⁵⁰ It has also been argued that preliminary crime data might not be useful or it could be misleading, because the data can be, and usually is to some extent, revised (e.g., an aggravated assault might be changed to a homicide if the victim later dies from injuries suffered during the assault). However, economic and labor data can also be revised, and a variety of people find value in these preliminary data.

Author Contact Information

Footnotes

1.	Monica Davey and Mitch Smith, "Murder Rates Rising Sharply in Many U.S. Cities," New York Times, August 31, 2015.
2.	"Violent crimes" are widely considered to be homicide, rape, robbery, and aggravated assault.
3.	Ibid.
4.	See, for example, Martin Kaste, "Nationwide Crime Spike Has Law Enforcement Retooling Its Approach," National Public Radio, July 1, 2015; Ashley Gold, "Why Has the Murder Rate in Some U.S. Cities Suddenly Spiked?," British Broadcasting Corporation News, June 5, 2015; Aamer Madhani, "Several Big U.S. Cities See Homicide Rates Surge," USA Today, July 10, 2015.
5.	Will Greenberg, "Police Chiefs From Across the Country Meet in D.C. to Discuss Violent Summer," Washington Post, August 3, 2015.
6.	Christine Byers, "St. Louis Officials to Attend Crime Summit in Washington, D.C.," St. Louis Post-Dispatch, October 5, 2015.
7.	See, for example, S. 2123 and H.R. 3713. For more information on these bills see, CRS Report R44226, Sentence Reform Acts: S. 2123 and H.R. 3713, by [author name scrubbed].
8.	See, for example, the Crime Control Act of 1984 (P.L. 98-473); the Anti-Drug Abuse Act of 1986 (P.L. 99-570); the Anti-Drug Abuse Act of 1988 (P.L. 100-690); the Crime Control Act of 1990 (P.L. 101-647); and the Violent Crime Control and Law Enforcement Act of 1994 (P.L. 103-322).
9.	CRS Report R43904, Public Trust and Law Enforcement—A Brief Discussion for Policy Makers, coordinated by [author name scrubbed].
10.	Franklin E. Zimring, The Great American Crime Decline (New York: Oxford University Press, 2007).
11.	The violent crime rate is the total number of reported violent crimes per 100,000 people.
12.	The FBI traditionally releases annual UCR data in the fall (late September or early October) of the following year. For example, 2014 UCR data was released on September 28, 2015.
13.	For the purposes of this report, "large and medium-sized cities" is defined as cities with populations of 50,000 or more people.
14.	Tracey L Meares, "Crime Statistics Don't Show That the Sky Is Falling," New York Times, June 4, 2015 (updated).
15.	"Noise" can be defined as transient increases and decreases attributable to happenstance or short-run shocks.
16.	Matthew Friedman's blog post on recent violent crime trends provides an example of how the interval used (i.e., monthly, quarterly, or annually) can affect the visual presentation of homicide data and how this can influence the conclusions drawn from the data. Matthew Friedman, "Just Facts: America's Non-Existent 'Spike in Crime'," in a blog by the Brennan Center for Justice, August 10, 2015.
17.	Ibid.
18.	Tracey L Meares, "Crime Statistics Don't Show That the Sky Is Falling," New York Times, June 4, 2015 (updated).
19.	James Alan Fox, "Don't Fear Monger Over Narrow Measures of Crime," New York Times, September 1, 2015 (updated).
20.	Matthew Friedman, "Just Facts: America's Non-Existent 'Spike in Crime'," in a blog by the Brennan Center for Justice, August 10, 2015.
21.	Ibid.
22.	Carl Bialik, "Scared Headlines Exaggerated the U.S. Crime Wave," FiveThirtyEight, September 11, 2015.
23.	When calculating whether an increase in a city's homicide rate was statistically significant, it was assumed that the 2014 homicide rate was the true rate, which was used to calculate how much of a statistical outlier the 2015 rate was. If the 2015 rate would have arisen by chance alone less than 5% of the time, it was deemed statistically significant. Specifically, the researchers calculated a rate of homicides per second using each city's 2014 data. They then calculated the p-value for the number of homicides in 2015. If it was lower than 0.05 they considered it a statistically significant decrease. They also looked at whether cities had statistically significant increases by seeing if the p-value for that test was below 0.05. Their methodology ignores changes in population, which usually increases over time. The researchers admit that not considering population increases could lead them to find more cities with significant increases than decreases in homicides. However, they note that other factors tend to drive homicide rates down over time, such as medical advances that have reduced the percentage of people who die after being stabbed or shot. Ibid.
24.	Ibid.
25.	Heather MacDonald, "The New Nationwide Crime Wave," Wall Street Journal, May 29, 2015.
26.	Peter Moskos, "Crime in Baltimore Rises as Support for Police Declines," New York Times, June 4, 2015 (updated).
27.	Heather MacDonald, "The New Nationwide Crime Wave," Wall Street Journal, May 29, 2015.
28.	Ibid.
29.	Patrik Jonsson, "Just What Is the 'Ferguson Effect'? It Depends on How You View Police," Christian Science Monitor, June 12, 2015.
30.	Ibid.
31.	German Lopez, "Why Murder Rates are Up in St. Louis, Baltimore, and Other Cities," Vox, September 8, 2015.
32.	Justin George, "Police Commissioner Defends Patrol Strategy Amid Violence," Baltimore Sun, June 3, 2015.
33.	Franklin E. Zimring, "What National Crime Wave? The 'Ferguson Effect' is Fiction," New York Daily News, June 7, 2015.
34.	Ibid.
35.	Tracy Sohoni and Charis Kubrin, "Is the Ferguson Effect a Myth?," Baltimore Sun, September 10, 2015.
36.	Ibid.
37.	Will Greenberg, "Police Chiefs From Across the Country Meet in D.C. to Discuss Violent Summer," Washington Post, August 3, 2015.
38.	German Lopez, "Why Murder Rates are Up in St. Louis, Baltimore, and Other Cities," Vox, September 8, 2015.
39.	Ibid.
40.	Wesley Bruer and Evan Perez, "DEA: Prescription Drugs Stolen in Baltimore Flooding the Street," CNN, June 25, 2015.
41.	German Lopez, "Why Murder Rates are Up in St. Louis, Baltimore, and Other Cities," Vox, September 8, 2015.
42.	Peter Herman, "Police Say Repeat Offenders Driving Crime Rise in District and Elsewhere," Washington Post, August 22, 2015.
43.	Monica Davey and Mitch Smith, "Murder Rates Rising Sharply in Many U.S. Cities," New York Times, August 31, 2015.
44.	Matthew R. Durose, Alexia D. Cooper, and Howard N. Snyder, Recidivism of Prisoners Released in 30 States in 2005: Patterns from 2005 to 2010, NCJ 244205, April 2014, p. 8.
45.	Ibid., p. 9.
46.	Ibid., pp. 7-8.
47.	Data on prison releases downloaded from the Bureau of Justice Statistics' website. See http://www.bjs.gov/index.cfm?ty=nps.
48.	Carl Bialik, "Scared Headlines Exaggerated the U.S. Crime Wave," FiveThirtyEight, September 11, 2015.
49.	Ibid.
50.	Ibid.