Violent crime is a key issue for U.S. midterm election

Crime has become a big issue in the U.S. midterm elections. According to a recent Gallup poll, crime ranked as the second most important issue among voters, with 71% of registered voters saying it was either “extremely” or “very” important to their vote. This came second after the economy, which 85% of voters said would be extremely or very important to their vote. According to a different Gallup survey released a few weeks ago, more than half of Americans (56%) said there was more crime in their area compared to a year ago. According to Gallup, “That’s up a whopping 11 points from 2021 and 18 points from 2020, and is the highest mark ever measured.” This issue was a concern across party lines, with 73% of Republicans, 51% of Independents, and 42% of Democrats all agreeing that crime is getting worse (in 2020, these percentages were 38%, 38%, and 37%, respectively).

Other surveys conducted with voters about the midterm elections have also revealed concerns about crime. According to a recent poll conducted by Politico, 64% of voters said crime would play a “major role” in they decision. Similarly, a recent survey by the Pew Research Center found that 61% of voters said that violent crime was a “very important” consideration for their vote.

That makes sense, considering the recent spike in homicide, the most serious and most reliably measured crime. From 2019 to 2020, the nation saw its largest ever recorded spike in homicides with an increase of nearly 30%. That same year, aggravated assaults also increased by 12%, and total violent crime increased by 5.2%. In 2021, murders increased by an additional 4.3%, despite FBI statistics showing no increase in overall violent crime (however, it is important to note that 2021 statistics are questionable due to the FBI’s transition to a new data system). 2022 federal statistics on crime are not currently available, though results from a recent survey conducted by the Major Cities Chiefs Association (MCCA), an organization of police executives representing major cities, showed that violent crime seems to be up again this year. Comparing the first six months of 2022 to the first six months of 2021, MCCA’s survey of their members (i.e., 70 large police agencies) found that violent crime is up about 4.4% this year, mostly driven by increases in robberies (+13%) and aggravated assaults (+2.6%). That same survey found that homicide decreased by an average of 2% among the 70 responding agencies, but 28 agencies (40.6%) still saw increases in homicides.

For some reason though, many policymakers, researchers, and the media oversell the fact that crime rates are lower than the 1990s crime peak. For example, the Pew report seems to think it is odd that Americans are concerned about crime, citing results from the Bureau of Justice Statistics’ (BJS) National Crime Victimization Survey (NCVS): “…annual government surveys from the Bureau of Justice Statistics show no recent increase in the U.S. violent crime rate.” But what they don’t mention is that the same exact NCVS survey showed that the violent victimization rate increased significantly in urban areas, from 19.4 to 24.5 per 1,000. The NCVS also doesn’t measure murder, which (as stated above) increased by about 30% in 2020 and an additional 4% in 2021. Further, the Pew report cites the most recently published NIBRS data to provide support for the fact that there was “no rise in the national violent crime rate between 2020 and 2021.” Yet, as we know, there is considerable uncertainty regarding the national crime rates for 2021 due to the FBI’s transition from summary-based reporting to incident-based reporting. Despite this uncertainty though, murders still appeared to go up (though this is not mentioned in the Pew report).

Indeed, crime rates are lower than they were in the 1990s. But keep in mind that the 1990s saw a historic crime peak. If “success” means having crime rates that are lower than the historic peak, then that’s a pretty low bar for success. In addition, murders exprienced a historic increase in 2020, a pattern that has lessened but nonetheless continued in 2021. While the spike was less drastic in 2021 in comparison to 2020, a 4% increase still represents an additional increase on top of 2020 rates. In short, the homicide rate remains very high despite being “the lowest since the 1990s crime spike.” Further, it is also important to look at multiple years of crime data to truly understand whether crime is trending upward or downward. Hyperfocusing on only one year does not provide enough data to examine change over time, nor does it allow for consideration of contextual factors that varied from year to year that may have affected crime rates (e.g., the transition to incident-based reporting).

Problems with the violent crime index

One thing that is important to understand is that there are a lot of official crime numbers, and they don’t all move together. Crime trends look different depending on what crime indicators one is studying. For example, in 2021, certain types of crime (e.g., homicides) increased but the total violent crime index fell. This is because the total violent crime index is an aggregate measure that includes four different crime types: murder (and voluntary manslaughter), rape, robbery, and aggravated assault. But one major problem with the total violent crime index is that it tends to be dominated by the high-frequency offenses, which are typically of lower severity. Then, because crimes are simply counted to compute the index, the less serious crimes dominate. In the case of the FBI’s violent crime index, murder and rape are the most serious offeses, but they do not occur very often. In contrast, robbery and aggravated assault are less serious crimes, but occur in very high frequency. As a result, the FBI’s violent crime index is largely a measure of robbery and aggravated assault and is relatively insensitive to changes in the rates of murder and rape.

This method reflects the simple and most common method of measuring crime — counting the number of crimes and calculating the rates per 100,000 people. But because not all crimes are created equal, these numbers don’t provide a reliable measure of the harm caused to victims across communities. Fewer people are directly affected by low-volume crimes like homicide, yet these crimes are life-changing for the victims and their families and cause irreparable harm.

An alternative strategy is to weight crimes according to their severity, rather than counting all crimes as if they were created equal. Otherwise referred to as a “crime harm index,” this approach weights offenses based on how much harm they cause, with larger weight given to more harmful crimes. This approach is currently used in several countries and a handful of jurisdictions in the United States, though not for the United States as a whole. To address the latter, CJLF’s own Kent Scheidegger computed a preliminary severity-weighted violent crime index for the United States. That first attempt showed that the most severe crimes have been increasing the most in recent years. Comparing the weighted index and the traditional index, beginning in 2015, the rise in the weighted index is significantly greater than in the unweighted. The weighted index is up 21% since 2011, while the unweighted index is unchanged from that year.

Conclusion

Crime is a major issue in the upcoming midterm election, with several polls indicating that crime is a “very” or “extremely” important issue for voters. But it is common to hear politicians and media outlets claim that the official statistics say otherwise and that the people who are worried about crime are just ignorant. But there are many ways to crunch the numbers, and the official violent crime index can be misleading. To lump all crime into one measure presents an incomplete and inaccurate view of reality. For a more nuanced understanding, it is important to distinguish different types of crime based on the harm that they cause. In addition, it is important to look beyond just one year of data when determining whether crime is on an upward or downward trajectory.