John Pfaff Profile picture
Sep 27 4 tweets 2 min read
That Black incarceration rates have been declining for a while is something that is not well known but getting there. But I was surprised at the magnitude of the decline for Black women: total #s have fallen by over 60% (!) from 1999-2020.... while total #s for white women rose. ImageImage
Black women are still incarcerated at a higher rate than white women, but in 1999 the rates where 212 vs. 27; in 2020, 65 vs. 38.

It's a really stunning drop. Tho the reversal in actual trends strikes me as even more interesting. ImageImage
We often hear that women are the fastest growing population in prisons. Put aside the baserate issue with that claim--big % change bc small baserate--the reality is not that WOMEN are the fastest growing, but WHITE women.
I don't have any profound takes on that empirical point. But feel like there's a lot of really important things wrapped up in it.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with John Pfaff

John Pfaff Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @JohnFPfaff

Sep 30
So, some prelim results abt bail reform and rearrest from the new NYS data (here, looking at NYC).

You're going to hear a lot soon abt how rearrest-on-release rose in 2020 vs. 2019, and it's true. But if you look at it by month, a clear story emerges: Covid.
Here is the prob someone is rearrested within 180 days of arraignment, for those who would never be bail eligible in 2020 (and equivalent people/cases in 2019).

Bail reform goes into effect at the red line, Covid peaks in NYC at the blue (April). Image
The black line is 180 days from the end of 2020--the new DCJS data stops at 2020, so those arraigned after the black line are in the data for under 180 days--some declines there may be due to truncation. But we see the big drop before that.
Read 10 tweets
Sep 27
There's an interesting empirical question here, though, about politics.

If those arguing for "defund" had used a far more Centrist Think Tank appropriate term, would we still be debating it?

It got under ppl's skin. It stuck. It's long-run impact is still unclear.
If someone had proposed "smarter funding," I bet everyone would have nodded, said some nice things, and quietly buried the idea at the next budget cycle.

But "defund" has been an irritant that has not gone away.
Don't get me wrong: the long-run implications of "defund" as a political approach is unclear. It's kept the idea alive, maybe moved the Overton Window, and created a lot of backlash that could be a net negative in the end.

But that's what its impact is: unclear.
Read 4 tweets
Sep 27
Good thing no one (seriously) arguing for scaling back the police is making this point.

Simon's hours-long debate over defunding is a peculiar mix of valid points and flagrant strawmen, with a surprising dose of self-contradiction tossed in. Image
Simon is right that a substantial amount of violence is committed by a small number of people, and that targeting those people could have outsized returns--but that police focus on sweeps, etc.

But that's... sort of an argument for CUTTING funding, right?
Simon's OWN ARGUMENT supports cutting back on conventional patrol officers and funding more investigatory approaches (with the key caveat that you cannot turn the Hercs into Freamons: it'll take time).

Which likely means we could shift $ from police to other stuff.
Read 7 tweets
Sep 25
How many claims about what bail reform did or didn't do using the new 2019-2020 DCJS data actually coded for whether the offense's bail-eligibility status changed over the jump?

Can't just compare rearrest rates for ROR-in-2019 vs ROR-in-2020, or bail-2019 vs bail-2020.
For ex: A person who could have gotten bail in 2020 but got RORed is not a bail reform "case," since that bail choice wasn't determined by bail reform.

So measuring across 2019-2020 is tricky; even within 2020 is hard, since some 2019 changes got reversed in July 2020.
Also, it's almost impossible to compare 2019 without being really careful careful to pay attention to what happened in March-April of 2020--some of you might recall something that went around that upended both the courts and crime.

They are... fundamentally different years.
Read 8 tweets
Sep 23
Remember: the ppl EXPERIENCING THE CRIME are the ones voting FOR KRASNER.

This is the primary election map for Krasner v Vega in 2021. Krasner won the red/orange precincts, lost the yellow/green.

The dots? Gun crime.

Look at the correlation. Image
I’m seeing this in city after city after city.

The question isn’t “can reform DAs survive violence?” The ppl experiencing the violence ALSO WANT REFORM.

The q is whether reformers can survive a backlash by those LEAST affected.
Here's Pittsburgh, where the reformer lost... but won where the violence was greatest (and got crushed in suburban Allegheny County):

Read 7 tweets
Sep 20
The 2021 National Crime Victimization Survey just came out today. More important than usual, given the shambles our reported-crime data is in.

The top-line: victimization rates are almost all at 5-year lows, and all below 2017-2019 levels (2020 was a ... strange year).
Reporting to the police, interestingly, rose, by about 5 percentage points, or over 12%, and seems to have risen pretty much everywhere.

Given 2020 was strange, also compared to 2019. Result holds: 2021 reporting rates higher than that too.
One might think, given the pushback against policing, that reporting might have fallen. But it actually rose fairly significantly, and not just some sort of recovery to pre-pandemic levels, but above the last pre-pandemic year as well.
Read 8 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(