The Los Angeles Police Department routinely violates the civil rights of Angelenos. They kill, beat, and maim, of course, but also conduct countless racist pretextual stops of drivers and bicyclists. They’re allowed to do this by the LA City Council, and without good information about what they’re up to it’s not easy for Angelenos to control them.

But they also routinely violate the California Public Records Act. The details range from egregiously obvious to subtly technical but in every case the goal, and for the most part the actual result, is to keep public records out of the hands of the public. They have been sued repeatedly and successfully for this over the last five years.¹ To my knowledge they’ve never prevailed in a public records case.

LAPD’s violations are expensive.² Since 2016 the City of Los Angeles has paid out at least $1,377,224 to settle CPRA suits against the police department only.³ Given the number of pending cases this figure is likely to top $1.5M by the end of 2021. The fact that settlement payments come from the City’s general fund clearly facilitates LAPD’s strategy of denying access to records until a suit is filed.

One of the most common reasons LAPD gives for denying the public access to records is that to produce them would be “burdensome.” There’s no such exemption in the CPRA, but courts have found, in some cases, that a public agency’s use of its resources, including employee salaries, to fill a request serves the public interest less than the production of the records would do.⁴

There are many, many problems with LAPD’s use of this theory to deny access to records. One is the fact that the burdensomeness of a given request isn’t only a function of its intrinsic difficulty, but also depends on the processing capacity of the CPRA Unit. This capacity directly depends on the Unit’s staffing level. That is, requests which might plausibly be burdensome if they had only one staffer would very likely not be burdensome if they had a hundred.

Which brings us to this very interesting set of records about the LAPD CPRA Unit, obtained from the LAPD CPRA Unit via CPRA request by Twitter user @CarpendiCacoeth. In particular, these four “Discovery Weekly Reports” show, among other things, that LAPD has cut its CPRA Unit staff by 50%, from eight to four, over the 30 months from January 2019 to June 2021:⁵

• December 26, 2018 — January 01, 2019
• December 24, 2019 — December 30, 2019
• December 22, 2020 — December 28, 2020
• June 01, 2021 — June 07, 2021

The lack of staff superficially supports LAPD’s claims that requests are too burdensome to fill, but it’s intentional and therefore no excuse. There’s no set required level of staffing in the CPRA, but the law requires agencies to fill requests promptly, which implies that they’re required to hire sufficient staff to be able to do so.

The reports also include the numbers of requests completed in the prior week. These numbers have, interestingly, gone up over time even as the staff has dwindled:⁶

Dates	Staff	Requests Completed
12/26/18 — 01/01/19	8	27
12/24/19 — 12/30/19	6	29
12/22/20 — 12/28/20	6	50
06/01/21 — 06/07/21	4	67

It’s impossible to tell from publicly available data which requests LAPD counts as completed during a given time span.⁷ It is possible to search uploaded documents by date, but LAPD closes most of its requests without producing anything, so again this doesn’t give a complete picture of what’s going on.

It’s also not possible to filter uploaded documents by responding department, so I had to sort through them by hand. I found that LAPD uploaded 120 SB1421 records. These mostly aren’t responsive to particular requests and I don’t know how they counted them. They uploaded records responsive to 11 requests for calls for service and 8 other requests.

They surely closed a bunch without uploading anything because they always do, but it’s not possible to filter or even sort requests by date closed, so I don’t know how to figure out how they came up with 67 closed requests in that week. But what they did upload was not onerous to produce.

First let’s talk about calls for service (“CFS”) requests. LAPD will gladly generate a spreadsheet showing all calls for service in a date range to a given address. They can pop these things out in seconds and nothing is redactable. There’s no work involved at all so they fill them really fast, usually within a week or two.⁸ There’s zero chance that the records will reflect badly on the LAPD, so they have no qualms about releasing them. Here, see for yourself:

CFS CPRA Requests closed June 1 — June 7, 2021
21-3551	21-4007	21-4110	21-4115
21-4128	21-4129	21-4130	21-4171
21-4172	21-4343	21-4399

The non-CFS requests that had records uploaded during this week are more interesting. They comprise a total of 1254 PDF pages and 3 spreadsheets. Since LAPD refuses to redact spreadsheets we can conclude that those took essentially no time to produce. Also, LAPD Constitutional Copper Lizabeth Rhodes is on record as claiming LAPD CPRA analysts can process 150 PDF pages per hour, so that’s about 8.5 hours of work for the non-CFS requests filled during the week in question.

Non-CFS CPRA Requests closed June 1 — June 7, 2021
20-6768	20-8162	21-1282	21-2728
21-2798	21-3882	21-3947	21-4275

We can round up to 10 hours if we include the CFS requests since, although they take very little time, there’s still some uploading and sending of templated messages to account for. There were only two analysts responsible for all 8 of these non-CFS requests. These two, Vanessa and Anna,⁹ mostly did the CFS ones too.

There were two other analysts working on CFS during that week. These two, Jennifer and Brittany, had nothing to do with numbered non-CFS requests. It seems reasonable therefore to assume that they spent the rest of their time on the SB1421 requests, which would mean that as of June 2021 LAPD has only two analysts working on non-SB1421 requests, and they’re only spending a total of 10 hours per week working on them.

This calls LAPD’s claims of burdensomeness even more into question. They’re understaffed, the staff they have aren’t working at even close to their capacity, and they’ve latched onto SB1421’s mandated releases as an excuse for basically failing to fill any other CPRA requests. Whatever the legislature had in mind in passing SB1421 they certainly didn’t mean it to undermine the fundamental right of citizens to see records quickly.

And finally, these reports reveal that LAPD tracks “Media Requests” as a separate category. It has been clear for a while that they greatly favor requests from mainstream media outlets over all others¹⁰ and this finding supports that story. This is certainly a violation of both the US and the California Constitutions, both of which preclude the government from favoring one set of speakers over another without a legitimate reason serving a legitimate purpose.

The public’s right to have City officials, including LAPD, comply with the Public Records Act is a fundamental right, guaranteed by the California Constitution. LAPD’s ongoing refusal to comply is a serious violation of our civil rights. And yet, as the law is structured, it’s going to be a lot of work to fix.

LAPD is nominally overseen by the Police Commission, which is not willing to hold LAPD to any standards at all. Only a Charter amendment, like a municipal sunshine ordinance, or changes in state law can solve the problem. But fixing this problem is essential.

Angelenos can’t be safe without reliable and consistent access to every possible scrap of information about this most powerful of municipal institutions, the only one with the power to kill people at will.

1. And almost certainly repeatedly and successfully before that, but according to the LA City Attorney it’s not possible to search for CPRA cases specificially before 2016.
2. The cost of LAPD’s CPRA violations is trivial compared to the cost of its civil rights violations, but not objectively low.
3. This figure is based on a spreadsheet I obtained from the Los Angeles City Attorney via the California Public Records Act.
4. This is distantly related to judicial interpretations of the so-called catch-all exemption, found at §6255(a) of the CPRA. Courts have found “needle in haystack” requests to be deniable on these grounds and the LAPD has proceeded to warp the holding beyond recognition to serve their illicit purposes. Just for instance, such a determination must be based on the actual public interest served by releasing the records, an element which LAPD never addresses.
5. These four reports are the only ones produced by LAPD. I’m not sure why this is. Probably the rest of them can be obtained.
6. The report for December 24, 2019 through December 30, 2019 shows 8 total staff with 2 dedicated to SB1421 work. It’s not clear how LAPD counts SB1421 material with respect to completed requests, since they’re just producing all of it, not necessarily in response to a particular request. I’m looking at ordinary requests here, though, so I listed 6 staff for that time period.
7. This is due to LAPD’s use of NextRequest, which is abysmally useless software, whose main purpose seems to be to allow government agencies to create the appearance of CPRA compliance while exploiting every possible loophole in the law to obstruct access. It’s not possible to search for requests closed on a certain date, for instance. Also it’s not possible to search unpublished requests, and LAPD doesn’t publish all of them.
8. You can try it yourself! Just make a new Gmail account and ask for CFS to some random address for the last 10 years. They’ll fill it in mere days. Just for instance take a look at Request 20-7318. On October 14, 2020 the requester asked for over a year’s worth of service calls to a long list of addresses in XLSX format. On October 30, 2020 LAPD produced everything requested in the requested format without arguing at all. This counts as a completed request even though it took no more than a minute or two of the analyst’s time.
9. LAPD CPRA analysts are identified by first name only on NextRequest.
10. It’s likely that they favor some reporters over others even from mainstream outlets, but there’s no practical way to prove that that I can see.