Big U.S. cities2015 population estimates.
">1 saw another increase in murders in the first half of 2017, likely putting them on track for a third straight year of rising totals after murder rates reached historic lows in 2014. So far, however, this year’s increase is considerably smaller than it was in each of the past two years; the big-city numbers are consistent with only a modest rise in murders nationwide. Overall, if recent numbers hold, the nation’s murder rate will likely rise but remain low relative to where it was from the late 1960s through the 1990s.Murder in 2017
The FBI collects national data on murdersFBI defines murder as “the willful (nonnegligent) killing of one human being by another.” This does not include deaths caused by negligence, accidents or homicides deemed justifiable, such as a killing in self-defense.
">2 and other major crimes, but it releases them after a significant lag. The most recent full year for which official data is available is 2015, when murders rose at their fastest pace in a quarter century. Official 2016 data won’t be available until the fall, but murder almost certainly rose last year too; in January, I found that big cities experienced a roughly 11 percent increase in murders in 2016, which past patterns suggest is consistent with about an 8 percent rise in murder overall.pointed out last January. We won’t know 2016’s national murder count for a few months, but the murder rate in big cities rose about 3 percentage points more than the the national trend the last three times murder rose nationally, which suggests that an 11 percent rise in murder in big cities will probably correspond with an increase of roughly 8 percent nationally. ">3In order to gauge changes in the prevalence of murder in big cities in 2017, I collected year-to-date murder counts for 2017 and 2016 in 68 of the country’s big cities, using a mixture of data from the cities themselves and from media reports.4 Data from 63 of the cities included murders committed through at least the end of May, and 50 cities provided data covering the month of June. These big cities have had roughly 4 percent more murders so far in 2017 than they did at the same point in 2016.5
Only a handful of cities are seeing large increases or decreases in murder this year, which is what we would expect to see given a small overall rise in the sample.6
U.S. cities with biggest changes in murder rates at midyear
| INCREASE IN MURDERS BY … | ||
|---|---|---|
| CITY | NUMBER | PERCENTAGE |
| Baltimore | +34 | +26% |
| New Orleans | +31 | +46 |
| Philadelphia | +26 | +20 |
| Charlotte, NC | +24 | +104 |
| Kansas City, MO | +21 | +41 |
| DECREASE IN MURDERS BY … | ||
| CITY | NUMBER | PERCENTAGE |
| New York | -34 | -22% |
| Houston | -31 | -20 |
| Atlanta | -21 | -40 |
| Newark, NJ | -18 | -40 |
| Wichita, KS | -15 | -44 |
St. Louis, Baltimore, New Orleans and Detroit have ranked among the top four cities in murder rate per 100,000 residents every year since 2014, and so far, 2017 does not appear to be any different. The table below shows the cities with the highest approximate murder rates so far in 2017, prorated over a full year.2015.
">7Big cities with highest murder rates in early 2017
Murder rates among cities with populations over 250,000 for which data was available
| MURDER RATE PER 100,000 | |||||
|---|---|---|---|---|---|
| CITY | 2016 UNOFFICIAL | 2017 PROJECTED | PROJECTED CHANGE | ||
| 1 | St. Louis | 59.3 | 57.9 | -1.4 | |
| 2 | Baltimore | 51.2 | 55.5 | +4.3 | |
| 3 | New Orleans | 44.5 | 49.7 | +5.2 | |
| 4 | Detroit | 44.9 | 40.7 | -4.2 | |
| 5 | Kansas City, MO | 26.4 | 30.3 | +3.9 | |
| 6 | Memphis, TN | 31.9 | 28.1 | -3.8 | |
| 7 | Cleveland | 35.0 | 27.9 | -7.1 | |
| 8 | Chicago | 27.9 | 24.1 | -3.8 | |
| 9 | Cincinnati | 20.8 | 23.1 | +2.3 | |
| 10 | Philadelphia | 17.7 | 19.8 | +2.1 | |
What midyear stats tell us
What can we really learn from a sample of big-city murder counts from midyear? The FBI has provided preliminary crime counts for the first six months of the year for each year since 2011. That isn’t enough to find any long-term trends, but the patterns do offer some hints about what the big-city sample suggests about the rest of the year.
First, there tend to be more murders in the second half of the year, when it’s warmer, especially in northern cities. Between 52 and 54 percent of big-city murders occurred in the second half of the year in every year between 2010 and 2015, according to the FBI’s data.8 So murder rates in those cities will likely ultimately be higher than the midyear statistics suggest.
Second, recent history suggests that not only does the absolute number of murders increase in the second half of the year, but the rate of increase also accelerates. The year-over-year rate of change in murder in big cities increased in the second half of each year from 2011 to 2015, regardless of whether murder was increasing or decreasing nationally. In other words, if the overall murder rate was rising, it rose faster in the second half of the year, and if the rate was falling, it fell slower in the second half of the year. In 2015, for example, in the 63 big cities that reported midyear data to the FBI, murder was up 11.8 percent at midyear and 15.1 percent at the end of the year. We can see this second-half acceleration in the FBI’s data for each year from 2011 to 2015 in the table below.
Murders accelerate in the second half of the year
The murder rate as measured at midyear tends to underestimate the actual annual rate among cities with populations over 250,000
| CHANGE IN MURDER RATE FROM PRIOR YEAR | |||
|---|---|---|---|
| YEAR | AS OF MIDYEAR | AT END OF YEAR | DIFFERENCE |
| 2011 | -4.1% | -2.9% | +1.2 |
| 2012 | +3.2 | +5.3 | +2.0 |
| 2013 | -11.7 | -9.3 | +2.4 |
| 2014 | -4.4 | -1.9 | +2.4 |
| 2015 | +11.8 | +15.1 | +3.2 |
The national picture: We’re still near historic lows
Big cities tend to exaggerate national murder trends, both up and down — so a large rise in big-city murder usually corresponds with a slightly smaller national increase. If murder rose roughly 8 percent nationally in 2016 (as my January estimate suggests) and is set to rise a few percentage points in 2017, then the nation’s murder rate in 2017 will be roughly the same as it was in 2008. That’s still more than 40 percent lower than the country’s murder rate in the early 1990s (but roughly 27 percent higher than it was in 2014).
Ultimately, this year’s trend is similar to last year’s in that more big cities are seeing a rise in the number of murders than are seeing a decline. There are still six months left in 2017, and while anything could happen, the most likely outcome is that — although this year’s rise will likely be smaller than last year’s — the country will see murders increase for a third straight year.