Open data on public expenditure — datasets, categories and a SPARQL endpoint for research and analysis

Chicago police overtime and special event staffing expenditure

The Chicago Police Department's payroll line for overtime has long been the line item that budget nerds, reform advocates and council members alike reach for first. It is large, it swings wildly from year to year, and it captures something the public can actually see on a street: extra patrol cars idling at a protest, mounted officers moving through a music festival, or detectives working a homicide investigation well past their shift end. publicspending.net has aggregated and normalised those payments from raw City of Chicago records so the figures can be compared with equivalent disclosures from other jurisdictions.

For an Australian audience, the value of having a comparable overseas dataset is that it puts local spending in context. State police forces in Melbourne, Sydney and Brisbane publish annual reports that break out some of these costs, but they rarely expose vendor-level line items. Pulling Chicago's payroll records into a normalised table lets a researcher in Carlton or Surry Hills ask the same sort of question they might ask of NSW Police Force disclosures on the state open data portal: how much extra is being spent, on what kind of deployment, and to whose pay packet does it ultimately land?

Where the Chicago overtime figures come from

The City of Chicago publishes payroll and overtime through three overlapping channels. The Chicago Data Portal hosts the "Employee Salary" dataset, updated twice yearly, and that includes overtime and other pay components for every named employee. The Office of Budget and Management produces an annual budget book, with the Police Department broken into dozens of sub-lines including "Overtime", "Holiday Premium Pay", and "Contractual Services for Special Events". The Inspector General, finally, issues quarterly and thematic reports that scrutinise a subset of payments and often flag abuse.

publicspending.net reconciles these into a single normalised table. Each row carries the officer identifier (anonymised), the duty code, the deployment description, the date range, the hours logged and the dollar value. The result is a flat dataset that can be sorted, summed or fed directly into a visualisation tool. For Australians used to looking at a single PDF annual report and trying to extract comparable overtime cost strings, that level of granularity is something of a novelty.

The summary below compares how Chicago's overtime and special event staffing disclosures stack up against three Australian state police forces, using figures converted to Australian dollars. Numbers are drawn from the most recent publicly available annual reports and from the Chicago Data Portal snapshot held on publicspending.net.

Item Chicago Police Department Victoria Police NSW Police Force Queensland Police Service
Disclosed overtime, most recent full year (AUD) ~$185m ~$92m ~$125m ~$78m
Source of disclosure Itemised payroll dataset Annual report summary Annual report summary Annual report summary
Granularity Per officer, per pay period Aggregate by division Aggregate by division Aggregate by division
Special event staffing, per officer (AUD/hour) $95–$150 $95–$120 $95–$125 $85–$115
Independent oversight body Office of Inspector General IBAC Law Enforcement Conduct Commission Crime and Corruption Commission

The takeaway is less about which force spends the most, and more about who is allowed to see how it is spent. Victoria's Independent Broad-based Anti-corruption Commission (IBAC) and Queensland's Crime and Corruption Commission both publish casework summaries, but only Chicago publishes the line-by-line data in machine-readable form on an open portal. publicspending.net's role is to widen that gap by republishing Chicago's records in a way that is jurisdiction-blind.

What counts as a special event at CPD

The Chicago Police Department classifies deployments in roughly four buckets: routine patrol overtime, investigative overtime, court time, and special event staffing. Special event staffing is the bucket that draws the most public attention, because it covers everything from the Lollapalooza music festival in Grant Park to the NASCAR race, the Chicago Marathon, the St Patrick's Day parade and Democratic National Convention security.

Each of these events triggers a deployment order issued by the Office of Emergency Management and Communications, then handed to CPD's Special Events Unit. Officers are assigned in waves, with the bulk drawn from patrol divisions and a smaller number from specialised units such as SWAT, K-9, marine and air support. The cost is split between the City of Chicago's general fund, the event organiser where one exists, and mutual aid agreements with neighbouring jurisdictions such as the Illinois State Police and the Cook County Sheriff.

Australian readers will recognise the pattern. The Sydney New Year's Eve fireworks operation, the Melbourne Cup carnival at Flemington, and the Big Bash League fixtures at the MCG all generate temporary surges in officer numbers that get folded into a single special event line in annual reports. The difference is disclosure: Victoria Police's annual report will name the operation and the gross figure, while Chicago's records will identify every individual officer paid for every hour of every shift.

Pay rates, allowances and the union bargaining backdrop

Overtime at CPD is not a single rate. Base overtime is paid at one-and-a-half times the officer's regular hourly rate, but the rates climb sharply for court appearances, holiday shifts, hazardous duty and special event deployments that require long hours without a break. Officers who hold specialised ranks or qualifications can earn considerably more per hour. The Fraternal Order of Police Lodge 7 negotiates the underlying contract on behalf of sworn officers, and the agreement sets out the multipliers, the caps on consecutive hours, and the conditions under which compensatory time can be substituted for cash.

For an Australian, the closest analogue is the Police Association of New South Wales or the Police Association Victoria, both of which bargain with the state government over enterprise agreements. The bargaining outcomes matter because overtime multipliers and minimum shift guarantees shape how much an event deployment actually costs. If a captain in Chicago earns a base hourly rate of $60 (roughly A$92 at current exchange rates), a 14-hour shift on a protest detail at one-and-a-half-times would add about $1,260 (A$1,930) in overtime alone, before pension contributions, FICA and other on-costs. Multiplied across hundreds of officers over a week, the headline figure quickly balloons.

Fringe benefits and pension contributions also distort the visible overtime figure. A dollar of overtime pay can mean up to forty cents in additional pension liability, which is why budget officers in Chicago scrutinise it as closely as the cash outlay. Victoria Police and NSW Police disclose similar on-costs in their long-term financial statements, but rarely in the same line as the overtime figure itself.

Year-on-year movement and the 2024–25 spike

The Chicago Inspector General's most recent thematic report flagged that overtime spending in fiscal year 2024 ran well ahead of forecast, driven by a combination of factors: major events including the DNC security perimeter, the city's response to a spike in retail theft that pulled officers off routine assignments, and the ongoing toll of investigations in violence hotspots. The total ran to roughly $185 million in Australian dollar terms, against a budget that had assumed around $155 million. Special event staffing alone accounted for close to a third of that overrun.

Looking back over the longer arc, the trend is upward, with a few notable dips. The 2020 pandemic year saw overtime collapse as gatherings were cancelled; 2022 saw it bounce back sharply as marches and demonstrations followed the killing of George Floyd. Each cycle provides a fresh reminder that overtime is as much a political signal as it is a budget line. When a force is stretched, overtime rises; when overtime rises, budget pressure forces trade-offs elsewhere.

Australians watching this from Brisbane or Adelaide in the quiet of a Sunday arvo might be tempted to read it as a uniquely American phenomenon. It is not, fair dinkum. NSW Police Force's annual report for the same period showed similar upward pressure, with the Sydney NYE operation and the response to multiple public order incidents contributing to an above-budget overtime outcome. The difference is that Chicago's data lets you trace the connection between specific events and specific overtime dollars in a way that the NSW open data portal currently does not.

Querying the dataset yourself

For readers who want to move beyond summary tables, publicspending.net exposes the underlying records through a SPARQL endpoint. Each payment row is represented as an RDF triple, with classes drawn from a spending ontology that maps CPD line items to equivalent terms in the Australian Government Information Service classification. A simple query returns the total overtime paid in any given quarter, broken down by division or by event type, in a way that is comparable with equivalent Victorian or NSW data.

Bulk downloads are available in CSV, JSON and Parquet formats. For a researcher in Parramatta or a journalist in Pyrmont, that means the same analysis a Chicago-based academic might run can be reproduced locally without scraping, without licensing restrictions and without waiting on a freedom-of-information request. The site's ontology tools also let users map local event categories to Chicago's deployment codes, which is useful when trying to compare, say, the cost of a State of Origin match at Lang Park with the cost of securing a football game at Soldier Field.

The same SPARQL endpoint can be queried for related jurisdictions covered by the project, including Illinois state payroll and other US cities. That cross-jurisdictional reach is what turns a single force's overtime records into something resembling a public spending atlas, in which any one deployment can be benchmarked against a handful of plausible comparators.

Readers keen to start exploring the Chicago dataset can head to publicspending.net, download the full Chicago payroll snapshot, and run their first query against the SPARQL endpoint described above. The same download includes the Illinois state and other US jurisdictions that the project covers, so a single Python notebook is enough to start producing cross-jurisdictional comparisons.

The team behind publicspending.net welcomes ontology mapping suggestions and pull requests from researchers who want to add an Australian jurisdiction to the comparative table. For anyone in Adelaide, Hobart or Darwin who has been waiting for a reason to dig into Chicago's spending data, this is as good a starting point as any.