There have been several posts in this space which use Calls for Service to predict the city’s Uniform Crime Report (UCR) Part I crime figures. UCR is the country’s main official source of crime totals, but it often takes weeks or months for the city’s official quarterly tallies to be posts. Calls for Service provides an unofficial means of estimating UCR counts which is imperfect but fairly accurate.

Which brings us to the question: is crime up, down or about the same in 2016?

And the answer is:  it’s complicated.

First things first, there is no evidence that UCR Part I crime as a whole is up this year compared to where it has been on average the last few years. But crime will almost certainly be up in 2016 relative to last year.

UCR Part I consists of person crimes (murder, assault, rape & robbery) and property crimes (theft, auto theft, arson & burglary), so while murder is rising in 2016, the total number of Part I crimes does not appear to be following suit. The table below shows UCR Part I crimes in New Orleans through the 3rd quarter of every year since 2013 (source: NOPD).

Year UCR Part I through Q3 Change
2013 12,959
2014 14,899 +15%
2015 13,964 -6.3%
2016 14,677 +5.1%

The table shows a 5 percent rise from 2015 to 2016 through the third quarter. We won’t know the official annual totals until February 2017, but tracking Calls for Service of UCR Part I crimes (below chart) shows nothing has changed in the last few months to change the assessment that crime will likely rise about 5 percent from 2015 to 2016. The black line shows the pace of 2016 UCR Part I Calls for Service compared to 2015 (red line) and 2014 (purple line). ucr-cfs

The problem with the official crime figures is that they’re heavily influenced by response times. In fact, I previously argued that nearly all of last year’s drop in crime was attributable to worse response times in 2015. Response times have gotten marginally better in 2016, but the city’s focus has been on improving emergency responses which does not do much for the 80 percent of UCR Part I crimes which are non-emergencies.

average-response

The below chart should help with visualizing the effect of response times on crime deflation. This chart tracks the percent of property crime Calls for Service that are given a disposition of ‘Report to Follow’ (RTF, Y-axis) based on the amount of time it takes for a dispatch to occur (X-axis).

response-deflation

About 73.8 percent of property crimes receive a RTF disposition. There’s a fairly sharp disparity between property crimes which receive a dispatch within an hour of the initial call and those that wait longer than an hour. The former are marked RTF 90 percent of the time while the latter receive reports only 59 percent of the time.

What happens if we measure UCR Part I Calls for Service marked ‘Gone on Arrival’ (GOA) and ‘Unfounded’ (UNF) along with RTF? It turns out that including all of those calls causes the annualized pace of crime to drop below 2015’s YTD count as seen in the below chart.

ucr-cfs-goa

Measuring crime in this way suggests UCR Part I crime was down last year though it was only down about 2.5 percent (rather than 6 percent) relative to 2014. The official figures show crime is up this year relative to 2015, but this metric suggests a 1.5 percent decrease YTD 2016 relative to a year ago. This way of tracking crime attempts to figure out how much crime there has been regardless of whether a victim was at the scene when the officer arrived.

The lesson for me is that crime has been relatively steady since 2014 at a level considerably above where it was in 2012 and 2013.

One final way of thinking about crime is more hypothetical: how many crimes would we expect if NOPD reached every property crime in 2 hours or less and every person crime in 30 minutes or less. It’s not a fully realistic expectation, but it does allow for a comparison across years as to how much crime is likely taking place in New Orleans.

Deflation.jpg

According to this table we are on pace for slightly less deflation this year than last year and should have virtually the same number of UCR Part I crimes in 2014, 2015 and 2016 (bottom row shows pace through early December).

So there you have it. Crime will almost certainly be up relative to 2015 but down relative to 2014 when the official numbers are released. There’s some evidence, however, that crime has been relatively steady in New Orleans since making a big leap in 2014. Finally, response times may have improved but they’re still longer than they were a few years ago and that means crime totals continue to be impacted by victims being forced to wait longer.