The letter arrives at CDOT’s 30th Street yard on a Tuesday in March. 25th Ward letterhead, signed by the alderman, requesting a speed bump on the 3100 block of South Springfield Avenue. One page. Block, improvement, fund code. CDOT logs it, assigns a work order number, routes it to a ward engineer, puts it in a queue. That queue has no published timeline. The speed bump might get built in six months. Might take eighteen. Might get quietly dropped if the alderman’s office sends a follow-up letter withdrawing the request—something that happens more often than anyone at City Hall cares to admit.
That letter is the first link in a paper chain moving roughly $50 million a year through Chicago’s fifty wards. Menu money, to use the shorthand. Every ward gets an annual allocation—currently around $1.32 million—for local infrastructure. Street resurfacing, alley repairs, speed bumps, streetlight upgrades, sidewalk replacement, traffic calming, tree grates, bike racks. The alderman decides what to request. CDOT evaluates and costs it. The work gets done, eventually. A report gets filed with the City Clerk, eventually. Then the cycle starts again, with no cumulative tracking of what was spent, deferred, abandoned, or why.
I spent three months and four FOIA requests following that paper trail across three wards with radically different political profiles: the 25th, carrying a Byrne-era machine lineage through Pilsen, Little Village, and Chinatown; the 49th, the progressive outpost in Rogers Park; and the 6th, the South Side ward centered on Chatham and Auburn Gresham that has been historically disinvested for decades. What I found is a system that treats the same nominal dollar amount as if it produces the same outcomes. It doesn’t. It can’t. And the reason it can’t isn’t a bug. It’s the system itself.
What Menu Money Is Supposed to Do
The program dates to the early 1990s, when Richard M. Daley’s administration formalized a practice that had existed in looser form for years: giving each alderman a pot of money for ward-level infrastructure without going through the full City Council appropriations process for every project. The intent was efficiency. Before menu money, an alderman who wanted a speed bump had to introduce an ordinance, get it referred to committee, wait for a hearing, and hope the mayor’s office didn’t quietly kill it. Menu money cut out the middle. Alderman writes the letter, CDOT does the work, ward gets the improvement. In theory, direct democracy in asphalt.
In practice, it turns every alderman into a mini-public-works commissioner with no staff engineer, no project management infrastructure, no procurement expertise, and no accountability mechanism beyond the next election. The $1.32 million is the same whether your ward has a staff of fifteen with a dedicated infrastructure coordinator or a staff of three where the alderman’s cousin handles constituent complaints and the alderman writes the CDOT letters himself on weekends. The dollar amount is flat. Everything else is variable.
Three Wards, Three Paper Trails
I FOIA’d three years of menu money records from the 25th, 49th, and 6th wards: fiscal years 2021 through 2023. The requests covered aldermanic request letters, CDOT work orders, project completion reports, expenditure summaries, and any correspondence between the alderman’s office and CDOT regarding project status, deferrals, or cancellations. Here’s what came back, and what didn’t.
The 25th Ward produced the most complete paper trail of the three. The alderman’s office maintains a spreadsheet—shared with me after a second FOIA request—tracking every menu money request by block, project type, estimated cost, CDOT work order number, and current status. Of 147 projects requested across the three fiscal years, 118 were completed, 19 deferred to the following year, 6 withdrawn by the alderman’s office, and 4 had no status recorded at all. The spreadsheet is not public. It exists because the 25th Ward’s infrastructure coordinator, who previously worked at CDOT, built it for internal use. Without that institutional knowledge inside the ward office, the tracking wouldn’t exist.
The 25th Ward’s request letters follow a pattern: specific block, specific improvement, specific fund code. They look like they were written by someone who knows the system. They were. The infrastructure coordinator’s prior CDOT experience means the letters arrive with cost estimates already close to CDOT’s final numbers, which cuts down the back-and-forth that delays projects in wards where the staff doesn’t know how to scope a request. The 25th Ward also uses menu money strategically, bundling small projects into single request letters to reduce CDOT’s administrative overhead. In FY2022, one letter requested eleven alley light installations across six blocks. CDOT processed them as a single work order. A ward without that knowledge would have sent eleven letters and waited for eleven separate work orders.
The 49th Ward produced a different kind of paper trail. The alderman’s office is transparent to a fault—menu money requests are published on the ward website, along with project status updates and a participatory budgeting process that lets residents vote on priorities. But the FOIA response from CDOT was thinner than the 25th Ward’s. Of 89 projects requested across three fiscal years, CDOT could provide completion reports for 71. The remaining 18 were listed as “pending” with no projected completion date. The 49th Ward’s internal tracking, which the alderman’s office shared voluntarily, showed 79 completed projects—eight more than CDOT’s records indicated. The discrepancy came down to timing: CDOT’s expenditure reports lag behind actual project completion by six to eighteen months, and the ward’s own records are more current.
The 49th Ward’s request letters are notable for what they include that the 25th Ward’s do not: community input documentation. Each letter references the participatory budgeting vote that selected the project, the date of the community meeting, and the vote tally. CDOT doesn’t require this. No ordinance requires it. The 49th Ward does it because the alderman’s office decided transparency was a governing principle. But here’s the structural problem: the 49th Ward’s transparency is voluntary. The 25th Ward’s tracking spreadsheet is voluntary. Every ward in Chicago could operate with zero public tracking, zero internal documentation, and zero community input, and it would all be perfectly legal. Menu money has no reporting requirements beyond a year-end expenditure summary filed with the City Clerk—a document that arrives months late, lists projects by CDOT work order number without addresses in some cases, and is functionally unreadable without a cross-reference to CDOT’s internal project database, which the public can’t access.
The 6th Ward produced the most unsettling paper trail of the three, and the one that most clearly reveals the system’s structural inequity. The FOIA response from CDOT for 6th Ward menu money projects was the thinnest. Across three fiscal years, CDOT provided records for 62 projects: 41 completed, 12 deferred, 6 withdrawn, and 3 with no recorded status. The request letters from the 6th Ward alderman’s office were shorter, less specific, and in two cases requested projects on blocks that CDOT’s records listed under a different ward’s jurisdiction. Those requests were returned to the alderman’s office for correction. There is no record of whether corrected letters were resubmitted.
The 6th Ward’s internal documentation, which I requested from the alderman’s office directly, was a single PDF: a scanned copy of a handwritten list of projects, with no fund codes, no work order numbers, and no status tracking. The alderman’s chief of staff told me, in a phone conversation that lasted eleven minutes, that the office relies on CDOT to manage project tracking. When I asked how the office knows whether requested projects are completed, the response was: “We hear from residents if the work doesn’t get done.” That’s the accountability mechanism. A resident calls to complain the alley wasn’t paved. The alderman’s office calls CDOT. CDOT checks the work order. The work order is either still in the queue or was quietly cancelled. No proactive tracking. No dashboard. No report that says: of the $1.32 million allocated to this ward, $X was spent on completed projects, $Y was deferred, $Z was returned to the general fund or rolled over.
Here’s where the numbers get ugly. In the 25th Ward, across three fiscal years, I could account for approximately $3.6 million in menu money spending through completed projects. In the 49th Ward, approximately $3.4 million. In the 6th Ward, approximately $2.1 million. The nominal allocation was the same for all three—roughly $1.32 million per year, $3.96 million across three years. The 6th Ward spent about 53% of its allocation on completed projects over that period. The 25th Ward spent about 91%. The 49th Ward spent about 86%. The difference isn’t in the dollar amount. It’s in staff capacity, CDOT relationships, institutional knowledge of how to write a request letter that moves, and the unwritten rules of which projects CDOT prioritizes when the queue gets long.
The Pipeline Problem
Think about the menu money system as a pipeline—request letter in, completed project out. The system’s failure isn’t at the input or the output. It’s in the middle. The pipeline has no visibility. No monitoring dashboard showing, in real time, which wards have pending projects, which have stalled work orders, which have requests that were returned for correction and never resubmitted. No automated routing. Request letters are physical documents, scanned and emailed, or in some cases hand-delivered to CDOT ward engineers. Work orders are assigned manually. Status updates require a phone call. The expenditure report filed with the City Clerk is generated by CDOT’s finance department, not by the ward, and it reflects what was spent—not what was requested, not what was deferred, not what was abandoned.
Engineers who design large-scale systems think about this differently. Google’s Site Reliability Engineering framework, documented in a comprehensive reference published by O’Reilly, devotes entire chapters to monitoring distributed systems, eliminating toil, and establishing service-level objectives—the precise features the menu money pipeline categorically lacks. The SRE approach treats observability as foundational: you cannot manage what you cannot measure, and you cannot measure what you cannot see. Google’s SRE book argues that any system handling significant throughput requires instrumentation at every stage, automated alerting when processes stall, and a postmortem culture that examines failures rather than burying them. The menu money pipeline has none of this. It has a paper letter, a manual work order, and a year-end spreadsheet that nobody reads.
The toil problem is particularly damning. In the 6th Ward, two request letters were returned by CDOT because they listed projects on blocks outside the ward’s jurisdiction. In a system with automated validation—something as basic as checking the requested address against the ward boundary database maintained by the city’s own GIS office—those errors would have been caught at the point of entry. Instead, they required a CDOT engineer to manually review the letter, identify the error, draft a response, and route it back to the alderman’s office. That process took, according to CDOT’s correspondence logs, 23 business days in one case and 31 in the other. In both cases, the projects were dropped from that fiscal year’s queue. They were not resubmitted. The money allocated for them sat unused until the end of the fiscal year, when it was rolled into the following year’s allocation—a rollover that appears nowhere in public records.
Without service-level objectives, there’s no standard against which to measure failure. No one at CDOT can tell you what the acceptable timeline is for completing a menu money project, because there isn’t one. No one at the City Clerk’s office can tell you when the expenditure report is supposed to be filed, because the deadline is “after the close of the fiscal year,” which is interpreted to mean anywhere from three months to fourteen months later. When I asked CDOT’s communications office for the average completion time for menu money projects by ward, the response was: “CDOT does not track completion times by ward for menu money projects.” They track it internally, obviously—the work order system has timestamps—but they don’t compile it, don’t report it, and have never been asked to by anyone with the authority to require it.
What the System Is Designed to Protect
Here’s where the structural argument matters. Menu money is not just a patronage tool, though it is that. It’s not just a constituent service mechanism, though it is that too. It is a structural feature of Chicago’s fragmented governance model that makes equitable infrastructure delivery impossible by design. The system asks each alderman to function as a mini-public-works commissioner without the staff, expertise, or accountability infrastructure that role requires. The 25th Ward can do it because it has a former CDOT employee on staff. The 49th Ward can do it because it has a participatory budgeting process that builds community oversight into the request pipeline. The 6th Ward struggles because it has neither, and because the aldermanic staff is focused on constituent services, not infrastructure project management.
The result is that the same $1.32 million produces wildly different street-level outcomes depending on ward capacity. This is not a coincidence. It is the predictable output of a system that distributes responsibility without distributing capability. And it is protected by the fact that no one is required to report the discrepancy.
Consider what a functioning accountability framework would look like. The National Institute of Standards and Technology maintains a widely adopted governance framework structured around five functions: Identify, Protect, Detect, Respond, Recover. The NIST Cybersecurity Framework was built for information security, but its governance logic applies to any system where accountability matters. You identify what assets you’re managing. You protect them with controls. You detect when those controls fail. You respond to failures. You recover and improve. The menu money system has no identification layer beyond the initial request letter, no protective controls beyond the alderman’s discretion, no detection mechanism for misallocation, no response process when wards fall behind, and no recovery loop for money that goes unspent or projects that get quietly dropped.
That same discipline applies to editorial structure: before publishing, editors need a way to test scattered notes become an argument readers can follow, which is where how Unsloppy AI Writing App fits the writing workflow can function as a planning aid rather than a substitute for domain evidence.
Assembling this story meant reconciling three separate CDOT FOIA productions, two voluntary ward spreadsheet dumps, a scanned handwritten list, a phone-call summary, and year-end expenditure reports that don’t list addresses. When you’re staring at a hundred pages of work order numbers that don’t map to anything the public can verify, you need every tool you can get to turn that pile into a narrative a reader can follow—which is why I used the Unsloppy AI Writing App to draft early structural outlines before going back and checking every number against the source documents. The tool helps you see the shape of the argument. The FOIA documents are what make it true.
This is not an oversight. A system with no monitoring cannot fail audibly. A system with no reporting requirements cannot be held accountable by the data it doesn’t generate. A system where each alderman controls their own request pipeline with no external oversight cannot be challenged without challenging the alderman directly—and in Chicago, challenging an alderman directly on infrastructure decisions is the fastest way to ensure that the pothole on your block doesn’t get filled for the remainder of their term.
The 6th Ward’s $1.86 million in unspent menu money across three years didn’t vanish. It rolled forward. It sits in the system, untracked and unreported, while Chatham’s alleys stay unpaved and Auburn Gresham’s streetlights stay dark. The aldermanic menu money system doesn’t need reform in the sense of a new policy paper. It needs a dashboard. It needs a deadline. It needs a public record that connects a request letter to a completed project—or to a quiet disappearance—with a timeline anyone in any ward can check. Until that exists, the $50 million flowing through fifty ward offices every year will continue to reward capacity over need, connections over complaints, and silence over accountability. The system isn’t broken. It’s working exactly as designed. That’s the problem.