NY arrests, pre-trial release, and rearrests

Far be it for me to say if these changes are because of changes in law, police, bail reform, or prosecution,but things have changed in NY. And blaming “bail reform” may be simplistic. But crime AND recidivism have gotten worse, and “changes in law, policing, bail reform, discovery reform, the diaphragm law, raise the age, Less is More Act, and prosecution” doesn’t quite roll off the tongue nicely. And of course it’s nearly impossible in the real world to separate variables that change at the same time. But things have changed.

There’s something very disingenuous when advocates say, “it’s not bail reform,” as if the common person is supposed to know the name of every bit of reform that has happened in the past few years. The onus should be on reform advocates to ensure public safety, not the public to know which part of the reform movement screwed things up. Of course other double down and say the the entire “crime” rise is, and I quote more than one person, “a hoax.”

Now back to a recidivism figure — and it’s hard to figure out the best one to use — usually recidivism is defined as 3 or 5 years after release from incarceration. but here we’re talking about similar but different: the odds of reoffending in 180 days is a low threshold. And the data is even more limited because many cases don’t last that long. I think having an open pending case when arrested again is one fair measure of recidivism. In 2021 it was 36% in New York State (34% in NYC and 41% upstate) and was just 29%(?) in 2019. The increase was greatest upstate, from 31% to 41%. Just to be clear. This means that in Upstate New York (Upstate is defined in a very NYC way as everything but NYC and surrounding counties) 31% of those arrested in 2019 had an open pending criminal case. In 2021 the figure jumped to 41%. People notice. People vote. And as some point you can’t fool all of the people all of the time.

For instance, it’s been reported, and more than once, that “less than 2% re-offend!” If comes from this, which if nothing else, really pushes the limits of acceptable truth telling. https://www.nycja.org/pretrial-release-dashboard.

To put it gently, never is the history of criminal justice data presentation has “felony rearrests in August 2022” been presented as a “key fact and insight.” (Nor have I ever heard of the “NYC community,” but I’ll pick my battles.) I’m not saying monthly re-arrest is not a useful figure in terms of comparing trends in felony pre-trial recidivism over time. That’s a novel concept. Kudos. But to highlight 1.6% as a “key” fact? Doesn’t rub me right in the ol’ data-presentation integrity department. But at least it is clearly stating what it means. What really bothers me are those who know better and still take anything on a screen and then say “less than 2% re-offend!” Professors and institute people and even journalists who know the data isn’t reflecting the truth, but are happy to use it because they can cite it to something. Yet they know it’s not true and do so anyway. Yes, I’m talking to you. (I hope that causes a lot of people to react.)

What matters, as most people understand it, is how many people under criminal justice control hurt somebody else while “in the system,” pending trial or post conviction and released. Not how many do so in August! 95% charged with a new offense in August 2022? Honestly, I’d say that 1 in 20 being charged with a new offense in one month(!) is pretty significant.

The percentage of people rearrested in NY State while there case was still pending and within 180 days of arraignment in 2021 was 21%. This is up from 18% in 2019. 10% were arrested for a felony in 2021, up from 7.3% 2019. For all these figure 2020 is even higher, but I’m leaving it out because Covid really did affect everything, including the number of cases that were not finishing within 180 days. I have a lot of issues with “180 days” because people will mistake it for a true measure of recidivism, which it is, sort of, but only in a comparative sense. And even then one must assume (incorrectly) that courts have moved at constant speed, and we know that haven’t. Since fewer cases are adjudicated within 180 days in 2020, more people will be counted as rearrested. The true re-arrest number is higher, as cases linger, but “180 days” is good because we can use 2021 data and compare it to 2020 and 2019. (Since we have the first 180 days for 2022 data.) Another (perhaps better) measure of pre-trial recidivism is the percent of people arrested who have another open case still pending. (I’ll get to that later.)

FWIW, 1,251 people were arrested in NYC in 2021 and then, and within 180 days, were rearrested for both a violent felony AND again for a non-violent felony. 20% of this group had a previous firearm arrest. 9% were re-arrested for a firearm. Now in a city of 8,500,000 with 76,534 people in this sample of arrests in 2021, 1,251 is pretty small, 1.6% to be precise. But that’s the thing. Of this small group 18% (n=221) also have 10 or more misdemeanor convictions. These people ARE recidivism. Who do you think is pulling the trigger? Really. Guns don’t shoot themselves. People pull the trigger. And a full people pull the trigger a lot.

Data Prep, feel free to skip this part, but it’s here both for transparency, replication, and fact checking. The raw data can (should) be downloaded here: https://ww2.nycourts.gov/pretrial-release-data-33136 It’s a CVS data file. It is not novice friendly. I used SPSS for analysis. The actual file you’re looking for is Supplemental_Pretrial_Release_Data.csv. “This file contains arraignments resulting from arrests requiring fingerprints from Jan. 1, 2019, through Dec. 31, 2021.” And do not forget the data dictionary, as you’ll need it.

File contains 562,646 cases. Here are some very basic numbers and demographics:

Roughly just under half the people arrested in New York State are convicted (2019-2021). 0.1 are acquitted (423 in three years, to be precise). Most of the rest of the cases are dismissed. Of those convicted, 18% percent are sentenced to jail or prison.

Of those arrested: 79.6% Male. 49.6% Black, 43.4% White, 4.1% Asian. 26.3% Hispanic. For non-Hispanics: 60% Black, 33% White, 5.6% Asian, <1% some other categories.

Total arrests arraigned in NY State 2019: 229,090; 2020: 156,610; 2021: 166,012.

After all the cutting that follows below (25% of the sample) none of the percentages above changed to any degree (which is generally a good sign).

I selected for arrests between 2019 and 2021, which leaves 551,712 cases. (many arrests go back a bit from DATs and a few go back decades, as I guess a warrant was issued.) Then I excluded 68,392 cases which were bumped up to arraignment in Upper Courts, because these cases no longer appear in lower court data (best I understand). That leaves 483,320. I dropped 47 non-NY-State arrests just because they were in my way. n= 482,273.

I cut cases (61,204) in which Lower Court-First Pretrial Release Decision at Arraignment was “Disposed at arraign,” because these cases are not followed. Same for 6,035 “unknown” in that category. Not followed. n = 422,069

So this is the sample I’m going to use. 172,526 arrests from 2019. 121,212 arrests from 2020. 128,331 arrests from 2021. Now 2020 is kind of an outlier because of Covid plus protests and riots. Things shut down. Generally everything spiked in 2020 and then came down a bit in 2021. 2019 and 2021 are good years to compare pre- and post- everything that changed (and not just bail reform).

Of those cases: 245,756 are in New York City. 72,470 are “Suburban NYC” (Nassau, Suffolk, and Westchester Counties). 109,843 are the rest of NY State.

I created a few new variables. I combined violent felony (VFO), felony, OR misdemeanors into one category for pending charge and for re-arrests. Seems important.

So that was fun, but now what? What can we learn? Well given the data we have.

Other tidbids This is for un-trimmed state data (I didn’t want to exclude cases kicked up to upper court for this). Chances of conviction and dismissal are roughly equal. There are (almost) no acquittals. Most of the “pending” will end up as convictions.

Most of those pending cases will end up as convictions. For instance in 2019 the number still pending is just(?) 5.5%. And conviction in 2019 was nearly 50%. 42% of cases in 2019 were dismissed.

What else, for those convicted, 27,000 get jail (less than a year) or jail and probation. Another 11,600 prison. So fewer than 40,000 get some form of incarceration as punishment Given the total of over 500,000 arrests in three years, it’s not that many people sentenced to incarceration. I mention this, in part, because any reform house of cards movement built on the idea that we’re locking everybody up and for nothing is going to fall. Because it’s not true.

(Now I’m back to the data set I was using where n = 422,069) Citywide 9.6% of arrests in 2021 had a pending open violent felony case. This is up from 5.3% in 2019. In numbers we’re talking 7,322 people arrested in 2021 with open violent felony cases. There’s possibly some low-hanging crime prevention possibilities here, were we willing to detain people arrested for violent felonies.

In 2021: 33.8% in NYC arrests had an open case (29% in Queens and 39% in Manhattan). In 2019 28.7% (24% in Queens and 29% in Manhattan).

In 2021 New York City, 20.8% of people arrested and arraigned and were released and arrested again (n=15,893) in 180 days. Compared to 19.9% in 2019. 14% rearrested for a felony in 180 days (up from 10.8%). 4.5% (n=3,436) for violent felony in 180 days (up from 3% in 2019).

Here’s a chart title you’ve never seen before, and perhaps for good reason. But I’m trying to show that there are a few people who keep committing crime over and over again. And we can identify them… because they keep getting arrested over and over again.

Some more data. NY State, 2019-2021. Release decision at arraignment (lower court arraignment = “larg”). ~12% are detained. And then the release outcomes. And then for the 12.5% for whom bail is set, what happens. In 2021 79% didn’t post bail (were still jailed) within 5 days, up from 66% in 2019.

Now just selected for Violent Felony charges (VFO): 6% remanded in 2021. 38% bail set (of those 2/3rd remained jailed after 5 days). 30% RORd (released). 27% released with “supervision,” which has been the big push. (I’d love to see more data and outcomes in this category.)

(All data unchecked. Corrections welcome. Sorry for the typos.)