Crime problems are capacity problems

My home invasion and the slow response to it reveal what's wrong with how we talk about crime and what we need to do for public safety.

People keep asking me if I’m okay. An intruder came into my house in Oakland while I was home recently. 911 took six minutes to answer (well longer than that — you can read the full story here), and the Oakland Police Department took two days to show up. I’m okay about the intruder, but less so about what this means for the city I’ve lived in for thirty years (but don’t really live in anymore, as my husband and I have mostly moved to the DC area.) Thank you to all the kind souls who’ve checked in on me.

Here, I want to talk about people asking me if I’m okay not because I was a crime victim, but because of Twitter comments. I tweeted about what happened shortly before I went to sleep the night of the incident, and when I woke up, friends on one of my group chats had noticed it and reached out. My post had garnered a few thousand likes and hundreds of replies while I’d been sleeping, and while they were concerned about what had happened to me the previous day, they were also concerned about how the day might go for me if I made the classic mistake of reading the comments.

My friends were worried about tweets like this one: 

“You get what you fucking voted for.”

The group chat explained that various factions of the right had piled on. They say as a liberal you asked for this. That you are a victim of your own ideology. That you should move to Texas and get a gun. 

I ignored my friends’ advice and read the comments – or rather, some of them, since there are now over a thousand replies to that thread. Clearly, not all of them conveyed the authors’ views with tact and maturity. But the vast majority were sympathetic, just sorry this happened to me. There were dozens, maybe hundreds of comments to the same effect as the insensitive one above, but said more kindly, many of them displaying true concern for me, and offering to help me leave the state, get a gun, or both. Most, they just really thought I’ve been voting wrong. 

“Honest question. Did you vote for the people who implemented the policies that resulted in your unfortunate situation? I live somewhere that the police would have been here in minutes and the invader would have been met with the business end of a shotgun. 

“Glad you’re ok, but the next person might not be so lucky. People who continue to vote in the politicians that allow this to happen are insane.”

Other commenters had obviously dug into my background, and, noting that I worked for the Obama Administration, assumed that I have indeed voted for the people who promoted “the policies that resulted in my unfortunate situation” and deemed me, as Samsational above suggested, insane. 

Of course, if they’d really done their research on me, these twitterers would never have suggested that the politicians had implemented the policies — they may have written and/or authorized them, but the implementation is a whole ‘nother ball of wax. That’s Pahlka 101. And they would have noted what I wrote two posts ago:

“…when I come back to California, I think of it as the place where you can’t blame Republicans for what’s gone wrong. And a lot has gone wrong here. We’re 40th among states in educational outcomes, we have sky-high housing prices and inequality, heartbreaking levels of homelessness, and only 9 other states spend more per capita than we do.”

What they didn’t get is that if I’m a liberal, I’m an Ezra Klein liberal. I thought it was irresponsible to run Joe Biden as a candidate, I think Kamala Harris should run as a prosecutor, and I like my bagels with no more than three toppings.¹ The comments didn’t hurt me because I see a lot of messy truths in that messy pile-on. That doesn’t mean I agree with all the remedies my commenters were proposing. If anything, I think the left and the right are often both missing the point – or at least ignoring the levers that might get us somewhere. 

Crime is not the issue I usually think about when interrogating the track record of the left — I’m generally much more critical of the inability of Democrat-led jurisdictions to build housing and green infrastructure, and Dems’ overall lack of gumption to challenge the kludgeocracy that’s strangling our government. But here I am digging into why it took two full days for the Oakland Police Department to respond to my 911 call. And I’m finding myself in eerily familiar territory.

I’ve long railed, for example, against the accretion of policy in archeological layers that ossify systems and constrain the people in them. Right around the time of my incident, it turns out, Gagan Biyani of Empower Oakland went on a 12-hour ride-along with Oakland PD, and observed: 

Nearly every time an officer does something, they must consider the multitude of rules that govern how they do their job. Many of these rules are sensible, but in recent years they have made it incredibly difficult to do the mere basics. For example, an officer must consider 19 different factors to determine if they can chase a criminal. 19! Most of the time, they either don’t pursue or it’s already too late.

This sounds remarkably similar to many stories in my book, most notably the 800 page training manual for unemployment insurance claims processors, which references 7,119 pages of regulations created over decades by both Democratic and Republican administrations, or to something I wrote last year about the volume of policy at the Department of Defense:

Navigating the complexity and sheer volume of Pentagon policy — equivalent to 100 copies of “War and Peace” — slows everything from acquisitions to hiring to logistics to combat operations. According to Stuart Wagner, “Warfighters each day are required to assess laws of war against planned operations. If the time to understand the policies exceeds the time window, the operation gets canceled.” There are plenty of reasons an operation should be canceled, but failure to understand our own rules should not be one of them.

There are plenty of reasons why an officer might not pursue a criminal, but the cognitive load of 19 governing policies should not be one of them. 

Also similar to things I have written is Gagan’s observation that “The patrol job is 80% paperwork.”  

We visited 9 crimes during our 12-hour shift. Let that sink in. While most calls took just 5-10 minutes to handle, the rest of the time was spent on the computer writing up notes… We have created a “police state” to monitor our police, hamstringing an already beleaguered police force. 

It sounds like they didn’t handle anything very violent that night if everything took 5 minutes. I’d wager that in more serious cases, the job is more like 90% paperwork. 

A different Oaklander, Tim Gardner, has been reporting on that paperwork:

Take the Oakland Police Department’s crime-reporting system, which is more than twenty years old. It’s so degraded that the department can no longer use it to centrally track police investigation documents. Instead, employees use Microsoft Word, saving files on desktop folders and sharing by email.

That system is supposed to automatically transfer uniform crime reporting data to the state for its annual crime compilation. But that too is broken. The state recently released 2023 data showing 11,169 aggravated assaults in Oakland in 2023—a 3.4-fold increase over 2022. But that’s an error. OPD’s own report shows about 3,500 aggravated assaults.¹ The state data is flawed due to an unresolved bug in data transmission discovered a month ago.

Police sources tell us that OPD is now “beta testing” a new $12 million computer-aided dispatch system for 911 operations, funded by a state grant. The new system has a records module that could replace OPD’s current system, but police can’t use it because it won’t communicate with the myriad of archaic OPD data and reporting systems required by local, federal and state oversight policies.

Let me be very clear that I am not suggesting we do away with OPD oversight. In the realm of police department scandals, which can be pretty bad, Oakland’s are…I don’t know how to say it…really bad. Ever heard of the Riders? They were four cops who routinely planted evidence, beat and even kidnapped Oaklanders, and falsified reports. They got away with it for years. It’s hard to look at that history and not conclude that we need a very high level of oversight.

But the oversight we have gotten seems to have given us the worst of both worlds. The Riders scandal was followed by several others…all while under court-ordered oversight. Perhaps there were scandals averted, and the net result on abuses has been positive. But there are clearly costs here. I don’t know enough about the nature of the oversight to understand mechanically how it is creating this drag on capacity, but there are patterns here from my work with government that feel very familiar. So here’s a guess at what’s going on:

High-level officials are appointed and charged with coming up with the metrics that will help them determine whether the department is on track. They throw those reporting requirements over the fence without understanding the existing data management infrastructure and practices. Things like data structures are way too low level for the officials appointed to this task to bother with, and the people who understand the data infrastructure, already disempowered by our waterfall practices, are even less likely to speak up because when your department is in deep trouble, you don’t “push back” on reporting requirements. It looks like you’re resisting the oversight. The other oversight bodies are doing the same thing, and rarely coordinating with each other. I asked Claude to tell me what bodies have oversight authority over OPD, and this is what I got back (caveats apply, of course):

1. Oakland Police Commission: An independent civilian body that oversees the OPD's policies and practices.

2. Community Police Review Agency (CPRA): Investigates complaints against police officers and makes disciplinary recommendations.

3. Oakland City Council: Has authority over the police department's budget and can influence policies.

4. Mayor of Oakland: Appoints the police chief and has executive oversight of the department.

5. Federal Monitor: As part of a Negotiated Settlement Agreement stemming from a lawsuit, a court-appointed monitor oversees OPD's compliance with required reforms.

6. U.S. District Court: Maintains oversight through the Negotiated Settlement Agreement.

7. Alameda County District Attorney's Office: Investigates and prosecutes criminal misconduct by police officers.

8. California Department of Justice: Can investigate patterns of misconduct and enforce state laws.

9. US Department of Justice: The DOJ was involved in the original lawsuit that led to the NSA in 2003. While the day-to-day monitoring is done by the court-appointed monitor, the DOJ retains the ability to intervene if there are significant issues with compliance. The DOJ’s Pattern or Practice Investigations, part of the Civil Rights Division, has the authority to conduct investigations into patterns or practices of constitutional violations by law enforcement agencies, including OPD, under the Violent Crime Control and Law Enforcement Act of 1994. Grant Oversight: If OPD receives federal grants, the DOJ may have some oversight related to the use of those funds and compliance with grant requirements.

10. UPDATE: Apparently I missed one, the Office of the Inspector General. According to the City of Oakland website, “In November 2020, 81% of Oakland voters approved of Measure S1, establishing the OIG as an independent civilian oversight agency tasked with auditing OPD's operations and procedures.”²

I talk in my book about concrete boats, technology projects that are doomed to sink upon launch because they are the sum total of hundreds or thousands of undifferentiated, unprioritized, conflicting requirements without any vision of what the resulting product should act or do and how people will actually use it. Let's say you’re responsible for data systems at OPD. Do you think you could tame the reporting requirements from nine different oversight bodies into a clear product vision under the constraints not only of enormous risk aversion but also equally complex and unhelpful procurement frameworks? I sure couldn’t. 

So the data systems become a huge mess. Years ago, I heard that OPD officers reporting a gang-related crime have to click through some ridiculous number of screens in their software to get to the separate software module that was added to their records management system to comply with state requirements. It’s a good idea to collect specific data on gang violence, I’m sure, but if no one pays any attention to how it’s collected, the costs over time, as officers sit in the squad car for an hour or more navigating crappy software, add up. The job is 80% paperwork on an easy night. 

Our 911 system has a capacity problem too, which is why I found myself on hold with 911 for six minutes during my home invasion. The call center is understaffed, and some of the smarter reporters are starting to look upstream at the challenges of fixing that. The hiring process stalled for a while, and I liked that the press picked up on that as a lever, though no reporter ever pulled the thread far enough to explain why it stalled. But there are other questions. One reporter shared with me that it takes many months to make a hire, and once on board, a 911 operator goes through nine months of training before being able to take calls. Nine months! That right there is probably your biggest leverage. A nine month training process means you’ve got massive policy and process accretion. Reducing the clutter should accrue a variety of benefits across operations. Tackling that means going way upstream — but that’s the work, folks! 

My three calls to 911: The first, at the bottom, dropped after 5 minutes. The second one must have picked up around 6.5 minutes, since the actual conversation was incredibly brief. The third call, at the top, was to tell the operator that the perpetrator was still around, and if the cops came soon, they would be able to recover my house and car keys, which the intruder had taken. Oakland’s stated average hold time is under one minute.

Back to my concerned Twitter followers, those who wanted me to fix Oakland’s problem by voting for Republicans.³ As I’ve said before, I’m not sure that California’s one party rule has done it any favors, so I’ll give them that. But for the most part, I see neither Republicans nor Democrats engaging with the actual problem. They both seem to accept that the job is to fight over budgets and tough on crime policies without regard to the way either is implemented. As Charles Fain Lehman said on Slow Boring in regard to DC’s (quite similar) policing capacity problems: In the long run, no matter how kind or cruel the criminal justice system’s normative framework, its efficacy is ultimately determined by its capacity to actually enforce the law.

Regardless of party, I would vote for someone who engaged with the actual levers of change: the complex governance, the implementation issues, the defeatism that sets in when your officers and leaders realize that while they would like to protect the public, their actual job is compliance. I would vote for someone who promised to work with all nine oversight bodies to reduce the training time for 911 operators. 

My friends on the right (my new Twitter friends and actual real world friends) have a valid point about progressive lack of self-reflection and willingness to take responsibility for our status quo. And there are a few Republican leaders who think about capacity issues in ways that excite me, including Indiana Senator Todd Young, though he displays his capacity savvy on issues of national security and competitiveness more than on crime. But where I do see this thinking is in the Abundance movement, and more generally in the fight to define what comes after neoliberalism. Noah Smith nails it when he says: 

Mobilizing society’s resources for these concrete goals (cheap high-quality housing, low and stable inflation, cheaper medical care, a military-industrial complex capable of resisting America’s powerful new enemies, public safety and public order, etc) is not very neoliberal at all. Neoliberalism is agnostic about what people want; the abundance paradigm identifies goals and goes after them. 

Noah does a great job in that post explaining how neoliberalism wasn’t quite as terrible as many are claiming now, but that it constrained our thinking and our sense of what was possible. That feels right to me when I look at local level problems like capacity in a police department. We’re in a moment where we can fundamentally think differently about how to get the public safety we want, and lots else. To do that, we need to address narrow interest capture, and we need government with real capacity– at all levels. 

There is no Abundance party. As Steve Teles points out, Abundance is mostly a faction of the Democratic party at the moment. It is a nascent faction, but one suddenly on the rise, after several oblique shout-outs at the Democratic National Convention. For Oaklanders, Abundant Oakland launched recently. From its website: 

As one example of many, Oakland schools will be asked to track 200 performance metrics this year. While we all agree OUSD needs to improve student outcomes, no organization…is capable of working towards 200 priorities simultaneously.

This is the kind of thinking that counters our “learned helplessness,” to borrow a phrase from the Abundance Network’s Misha Chellam. It fights the Democrats’ unfortunate tendency to indulge in tradeoff denial. It finds the levers that can tackle actual brokenness. If Republicans want my vote, that’s what I’d want to hear from them, too. There’s no reason there shouldn’t be an Abundance faction of the Republican party, the same way both parties had reform-minded, state capacity focused Progressive factions at the turn of the century. 

I’m very much okay, and grateful for everyone who's checked in on me. It’s disturbing to know (as all Oaklanders should) that you really can’t count on emergency services at the moment, but I’m okay because I see a path forward in all this mess. I’ve been observing the abyss of bureaucratic dysfunction for a decade, but now I’m finally seeing the loop start to close, as an activism interested in a different set of levers is starting to build political power from the local level up. My Twitter trollers are right. I may not agree with all of their politics, but we do get what we f***ing vote for. The “coke or pepsi” of left and right have been lousy options, so the work is developing new ones. I’m okay because of the people doing that work. 

1. Everything bagel liberalism is a phrase Ezra used to describe the dysfunction of loading every policy with every goal. Matt Yglesias explains it here.

2. More from the city website:
What is the difference between the Community Police Review Agency (CPRA) and the Office of the Inspector General (OIG)?

The OIG audits OPD's policies, practices, and procedures. We produce reports that include recommendations on how OPD can ensure departmental compliance with a given policy.

The CPRA investigates community members' complaints of OPD misconduct. They provide recommendations on disciplinary actions when there are sustained findings of misconduct.

3. The other thing Twitter really wanted me to do was buy a gun. I am not entirely anti-gun. Most of the summer, I live in a tiny town (population 137) in rural California. I stopped by a neighbor’s the other day and mentioned that we’d just had a visit from a mama bear and her cub. “Oh yeah, they’re fine, but there’s a big aggressive bear that’s been around here lately trying to get into the house. He tries that again, we’re gonna take him out with our [insert model of gun that people know about but totally went over my head here.]” My neighbors assume we have a gun. I have not disabused them of that notion. We should probably have one. I got a lot of offers on Twitter to teach me to shoot, and I appreciate them. Thank you. I intend to learn some day. 

But in Oakland, I didn’t need a gun and it wouldn’t have helped. Perhaps I could have held him in place til the cops came, but that would not be a short amount of time, given OPD’s challenges, and I would have found that incredibly stressful. Or I could have shot him, which would have been awful for everyone (including me) and totally out of proportion to the incident. The intruder was not violent, just mentally ill (and oddly generous, since he brought my car key back the next day). If the current crime stats are correct, that’s the case for the vast majority of crime in Oakland right now. Residential robberies are up a shocking 118%, but violent crime is actually down by 33%. A gun in this situation would likely have turned a non-violent crime into a tragedy. What we need in Oakland is a functioning police department, not more guns.