Where Chicago TIF Money Goes Across Wards and Community Areas

Abstract

Chicago sells tax increment financing as a way to steer public dollars into blighted, disinvested places that private capital skips. This paper follows the money to see where the approved subsidy actually lands. It is a synthesis of the published Chicago TIF literature paired with our own descriptive reading of the city's records, namely 768 approved TIF and redevelopment-agreement projects from 1986 through 2026, carrying about 6.57 billion dollars in committed subsidy.^[1] We did not run an experiment, model causation, or audit disbursements. We counted, summed, ranked, and mapped what the city itself recorded. The pattern is stark. The top five single named community areas, led by the Near South Side, the Loop, and the Near West Side, hold roughly 52.6 percent of all community-area-tagged dollars, and the Gini coefficient across community areas reaches 0.710. Project counts and dollar totals pull apart, which separates neighborhoods stitched from many small deals from downtown blocks built on a few enormous ones. Approvals were a rounding error in the 1980s and peaked in the 2010s. A long line of Chicago-specific research, from Dye and Merriman through Weber and a recent wave of equity analyses, predicts and documents exactly this concentration.^[2][3][9] The limits are hard and we keep them in view. Approved is not disbursed, three records span multiple community areas, and affordable-unit counts exist for only a fifth of projects. The records describe where the money goes. They do not show what it caused.

The promise and the map

Tax increment financing is the most consequential development tool Chicago controls on its own, and most residents have never seen a clean account of where its money goes. The mechanics are not complicated once you sit with them. The city draws a boundary around a district and freezes the property-tax base inside that line at its current assessed value. For the next two decades or more, the taxes generated on that frozen base keep flowing to the schools, the parks, the county, and the other taxing bodies that always got them. The growth above the frozen line, the increment, gets peeled off and routed into a special fund the city spends inside the district.

The justification rests on two linked promises. One is the but-for premise, the claim that the new development would not have happened but for the subsidy, so the diverted taxes are not really lost because there would have been no increment without the deal. The other is geographic. TIF is supposed to reach the places ordinary investment ignores, the blighted blocks and disinvested corridors where private capital will not move on its own. That second promise is the one this paper tests against the record. Not whether TIF works, not whether any single deal earned its subsidy, but a simpler and more answerable question. When the city approves TIF money, where does it go on the map?

A little institutional context is worth carrying into the analysis, because it explains why the question matters in Chicago specifically. Tax increment financing in Illinois rests on a state statute that lets a municipality designate a redevelopment area only after finding it blighted or at risk of becoming so. That blight finding is the legal hook for the whole promise. It is what lets the city say the tool reaches places the market has abandoned. Chicago became one of the heaviest users of TIF in the country, and its districts grew into something close to a parallel budget, property-tax increment routed outside the ordinary appropriations process and committed deal by deal at the discretion of the planning department and the City Council. The records behind this paper alone span 175 separate TIF districts.^[1] The increment never appears as a line a voter approves. It accumulates inside district funds and is spent where the boundary was drawn. That opacity is exactly why a clean public map of where the commitments land is worth drawing.

The blight requirement is the part of the structure most worth holding in mind while reading the numbers. To use the tool downtown, the city has to certify downtown blocks as blighted or at risk, and the statutory definition is elastic enough that aging buildings, obsolete layouts, and lagging assessed values can carry the finding even in valuable districts. So the gap this paper measures is not only a gap between a stated goal and an observed distribution. Part of it was written into the program at the design stage. Once the blight standard is loose enough to certify aging downtown blocks, the statute that was supposed to confine the tool to neglected areas instead clears the way for it to operate in the most valuable ones.

A word at the outset about what this paper is and is not, because the honesty of the whole exercise depends on getting this right. This is a synthesis of the published Chicago TIF literature paired with our own descriptive reading of the city's administrative records. It is not an audit. We did not trace a single dollar from approval to a contractor's bank account. It is not a causal study. We did not build a model, run a natural experiment, or estimate what any neighborhood would look like in a world without TIF. We counted, summed, ranked, and mapped what the city itself recorded, and we set that description next to two decades of peer-reviewed and civic research that has asked the harder causal questions. Where our numbers and the literature point the same direction, we say so and call it corroboration. We do not call it proof.

The source is a single file, named plainly so anyone can find it. It is the City of Chicago TIF Projects and Redevelopment Agreements record, published on the Chicago Data Portal by the Department of Planning and Development, and shipped unchanged in our own repository so the analysis is fully reproducible.^[1] Every figure in this paper that is not attributed to an outside author comes from our reading of that one file. The records cover 768 approved projects, with approval dates running from 1986 through 2026, and they carry about 6.57 billion dollars in committed subsidy. Those are large numbers for a program most Chicagoans encounter only as a line item they cannot decode on a property-tax bill.

The tension that animates the whole piece is the gap between the promise and the map. TIF is sold as a way to reach the neighborhoods capital forgot. The dollars, on the city's own books, cluster downtown. The concentration is real, it is large, and it is what a substantial body of Chicago-specific research has predicted and documented for more than twenty years. What this paper adds is not a new theory. It is a clean, current, reproducible map of where approved subsidy actually landed across the city, drawn from the most recent version of the city's own file, paired with an honest account of what that file does and does not let anyone say.

The civic stakes are worth naming before the numbers start. When property-tax growth gets diverted into a TIF district, that money is, for the life of the district, not available to the general operating funds that pay for schools and city services across the whole city. The increment is spent where the boundary is drawn. So the question of where TIF dollars concentrate is not an accounting curiosity. It is a question about which corners of Chicago the city has chosen to channel its most flexible development money toward, decade after decade, and which corners it has not.

The schools make the stakes vivid. The Chicago Public Schools draw most of their funding from the same property-tax base that TIF freezes, so for years the district has argued that money locked inside TIF districts is money kept from the classrooms it would otherwise reach. That tension runs straight through the data in this paper. As a later section shows, a large block of the program's dollars goes to projects funded entirely by the public, many of them CPS capital projects financed through TIF, which means the increment is sometimes the very thing paying for the schools and sometimes the thing the schools say is starving them, depending on the district and the deal. The point here is not to resolve that argument. It is to mark that the dollars this paper tracks are not free money conjured from nowhere. They are public revenue with other claims on it, and the choice to commit them in one place is, by arithmetic, a choice not to leave them in the general pool that serves everywhere else.

The records answer the where question with unusual clarity. The rest of this paper is mostly a matter of reading them carefully and refusing to claim more than they support.

What the records are and what they are not

Before any finding, the ground it stands on. The dataset behind this paper is the city's own administrative record of approved TIF and redevelopment-agreement projects, 768 unique projects in all, published by the Department of Planning and Development through the Chicago Data Portal and carried unchanged in our repository.^[1] It is not a survey, not a sample, and not a reconstruction. It is the running ledger of deals the city's Community Development Commission approved, with the dollar amounts, dates, locations, and project details the city itself attached to each one.

The headline totals set the scale. Across the 768 projects, total approved subsidy comes to 6,565,487,695 dollars. The median approved subsidy is 2,725,000 dollars. The mean is 8,548,812 dollars, more than three times the median. That gap tells you the distribution is lopsided before we rank a single neighborhood. A typical project is a few million dollars. The average is dragged upward by a small number of very large deals. The largest single approval in the file is 959,000,000 dollars, a figure that by itself exceeds the combined approvals of most community areas in the city. Approval years span 1986 through 2026, four decades of the program recorded in one place.

That median-to-mean gap sets up everything that follows. When a mean sits more than three times above the median, the distribution is not a bell curve with a long tail. It is a body of small and mid-size deals with a handful of giants sitting far out to the right and dragging the average toward them. The median project, at 2,725,000 dollars, is the honest picture of a typical TIF deal, a few million dollars for a single building, a rehab, or a piece of infrastructure. The 959,000,000 dollar Red Line Extension agreement is more than three hundred and fifty times that median, a single line large enough to bend the statistics for the entire file. This is why the rest of the paper keeps reaching for medians alongside means, and why the geographic concentration is not a surprise once the size distribution is in view. A program in which a few deals are this much larger than the rest is a program in which a few places, the ones where those deals land, will dominate any dollar ranking. The uneven size of the deals and the uneven geography of the spending are two views of one underlying fact, and the size distribution is what makes the geographic concentration in the next section predictable rather than surprising.

A dataset is only as honest as its fields, so it is worth walking through completeness as a measure of how much to trust each cut we make. The money fields are nearly perfect. The approved amount is present for all 768 rows, 100 percent. The approval date is present for all 768 rows, 100 percent. Geography is close behind. Ward is present for 765 rows, 99.6 percent. Community area is present for 762 rows, 99.2 percent. Total project cost is present for 760 rows, 99.0 percent, and the city-reported subsidy percentage for 761 rows, 99.1 percent. Then comes the field that matters most to the question many readers bring to a program like this, and it is reported the least. Affordable units appear for only 155 rows, 20.2 percent. Four out of five projects leave it blank. We return to that gap in its own section, because it shapes what the records can and cannot say about housing, but it belongs here too, in the honest inventory of what we are working with.

The single most important distinction to fix in mind, before any dollar figure lands, is what the approved amount actually represents. It is approved subsidy, the dollar commitment the Community Development Commission signed off on, not necessarily money that flowed. We did not audit disbursements, and the file does not track them in a way that would let us. A project approved for ten million dollars may have drawn the full amount, drawn part of it, or never closed at all. So every figure in this paper describes commitments on paper. When we say a neighborhood captured a certain number of dollars, we mean the city approved that much subsidy for projects tagged to that neighborhood, not that contractors there cashed that much. This is not a hedge. It is the precise meaning of the only number the file reliably gives us, and stating it plainly is the difference between a defensible description and an overreach.

Coverage is broad. The records touch 49 wards, 75 of the city's 77 community areas, and 175 separate TIF districts. That breadth matters because it means the concentration we are about to document is not an artifact of thin data. The city has approved TIF deals nearly everywhere. The question is not whether disinvested neighborhoods appear in the file. Most of them do. The question is how much money flowed to each place once it did.

Three records require a flag before they show up anywhere else, because they break the clean one-project-one-neighborhood pattern the rest of the analysis assumes. Three rows tag multiple community areas, joined by commas in a single field. The largest of the three is also the single largest approval in the entire dataset, the 959,000,000 dollar Red Line Extension TIF intergovernmental agreement, which spans five South Side community areas at once, namely Morgan Park, Riverdale, Roseland, Washington Heights, and West Pullman. The other two are smaller. One is a 3,500,000 dollar deal for Conservatory Apartments spanning East Garfield Park and Humboldt Park. The other is a 9,979,778 dollar Homan Square preservation project spanning East Garfield Park and North Lawndale. These three rows force a choice we make explicitly later. For now the point is that the reader should meet the Red Line corridor here, as a transit megaproject that crosses five neighborhoods, before it appears at the top of any chart. It is not a single neighborhood capturing a billion dollars. It is a corridor, and treating it as if it were one community area would distort the very comparison we are trying to make fairly.

A note on method that is really a note on restraint. We imputed nothing. Where a field is blank, we left it blank and counted around it, and the missingness is printed in the analysis output rather than papered over. When we report a statistic on subsidy share, it covers the rows that actually carry a subsidy percentage, and we say how many that is. When we report affordable units, it covers the rows that report units, and we say how few that is. There are no filled-in averages standing in for absent data, no modeled estimates dressed up as observations. The point of this paper is to show what the public record contains, and that means showing the holes in it with the same care as the totals.

The operations behind every figure are deliberately simple, and the simplicity is the safeguard. We read the file, grouped rows by community area and by decade, summed the approved-subsidy column within each group, and ranked the results. We computed shares as a group's dollars over the relevant denominator, and we are explicit each time about which denominator we used, whether it is all approved dollars, all community-area-tagged dollars, or the single-named-area total. The one statistic that goes beyond arithmetic is the Gini coefficient, a standard summary of how unequally a total is divided across a set of recipients, which we computed across the community-area grouping totals. No regressions, no weighting, no adjustment. A reader with the same file and a spreadsheet could reproduce every number in this paper, which is the entire reason the file is shipped unchanged alongside it.^[1] When a finding rests on a choice, such as how to handle the three multi-area rows, we make the choice in the open and report the figure both ways.

This is worth dwelling on because the descriptive-versus-causal line is the line most easily blurred in writing about public spending, and blurring it is how honest description turns into overreach. A sentence like "TIF concentrated downtown" is descriptive. A sentence like "TIF caused downtown to develop" is causal, and nothing in this file can support it. The same downtown blocks that captured the most subsidy were also, independently, the blocks with the highest land values, the most active private market, and the largest projects. Disentangling whether the subsidy drove the development, followed it, or simply rode alongside it would take a research design this paper does not have and does not pretend to have. We can say where the committed dollars are. We cannot say what they made happen. Every claim in the sections that follow is built to stay on the descriptive side of that line.

That is the evidentiary ground. A nearly complete record of approved subsidy and approval dates, very strong geographic coverage, a reliable read on the public share of project cost for almost every deal, a weak and partial read on affordable units, and three multi-area rows we handle in the open. Everything that follows is built on these fields, and every claim is held to what they can carry.

Where the dollars pile up

Now the central finding, the heart of following the money. When you rank Chicago's community areas by the approved TIF subsidy tagged to them, the top of the list is not a set of disinvested neighborhoods. It is downtown and the blocks immediately around it.

A framing choice has to come first, and we make it in the open. Three rows tag multiple community areas, and the largest of them, the 959,000,000 dollar Red Line Extension corridor, is a five-neighborhood transit project, not a neighborhood. To compare neighborhoods fairly, we lead with the single-named-area ranking, the dollars attached to exactly one community area, and we hold the multi-area corridor aside so a transit megaproject does not crowd into a neighborhood comparison where it does not belong. We report the alternative framing in a moment, because honesty requires showing both, but the neighborhood story is cleanest when the corridor sits to the side.

On that single-named-area basis, the top five community areas are the Near South Side, the Loop, the Near West Side, the Near North Side, and Uptown. Together they hold 2,613,454,156 dollars in approved subsidy. Set against the 4,966,907,917 dollars of subsidy that carries a single community-area tag, those five areas account for 52.6 percent of the total. A little more than half of every single-area TIF dollar the city has approved across four decades landed in five adjacent areas at the city's core.

The per-area figures make the shape concrete. The Near South Side leads with about 862.9 million dollars. The Loop follows at about 749.9 million. The Near West Side comes in at about 558.8 million, the Near North Side at about 259.2 million, and Uptown at about 182.7 million. After those five, the totals fall off and the geography starts to widen. Grand Boulevard trails at about 129.0 million dollars and Humboldt Park at about 127.6 million, the first two areas on the list that sit well outside the downtown core. The drop from Uptown's 182.7 million to Grand Boulevard's 129.0 million is the seam between the downtown cluster and everywhere else. Below them the tail runs long and thin. North Lawndale records about 125.3 million across 27 projects, Logan Square about 88.2 million across 17, Rogers Park about 74.2 million across only 6, and Austin about 74.0 million across 17. By the time the ranking reaches the city's West and South Side neighborhoods, the per-area totals have dropped to a small fraction of what the downtown core holds.

The other way of counting raises the share, and it is worth putting on the table in full. If instead of single named areas you rank the top five community-area groupings and let the multi-area rows count as groupings, the Red Line corridor's 959 million dollars enters at the very top, and the top five groupings then hold 3,389,722,737 dollars, which is 57.1 percent of all community-area-tagged dollars. Both numbers are real and both come from the same file. The 52.6 percent figure answers the question of how concentrated TIF money is among individual neighborhoods. The 57.1 percent figure answers a slightly different question, namely how concentrated it is among the top five tagged groupings of any kind, including a transit corridor that happens to be the single largest line in the dataset. We use the 52.6 percent single-named-area framing in the charts because it keeps the comparison honest, neighborhoods against neighborhoods, without a five-neighborhood corridor masquerading as one place. The reader should hold both. They differ by less than five points, and they point the same way. TIF money is heavily concentrated, and the concentration is downtown and downtown-adjacent.

For a single number that captures the skew without depending on where you draw the top-five line, the Gini coefficient is the cleaner instrument. Computed across the 75 community-area groupings, the Gini of the approved-subsidy distribution is 0.710. A Gini of zero would mean every community area received an identical share. A Gini of one would mean a single area received everything. At 0.710, the distribution sits far up the concentrated end of that scale. This is the formal statement of what the ranking shows informally. A small number of areas hold most of the money, and the long tail of areas holds very little each. The Gini is useful here precisely because it does not depend on a cutoff. Whether the headline is the top five, the top ten, or the top three, the same lopsidedness shows up in the single coefficient, and 0.710 puts the geography of approved TIF subsidy well past the point where a distribution could fairly be called broadly shared.

One more figure frames the whole exercise. Of the 6,565,487,695 dollars in total approved subsidy, the dollars that carry any community-area tag at all come to 5,939,387,695, about 90.5 percent of the program. So when we talk about where TIF money goes by neighborhood, we are accounting for roughly nine of every ten approved dollars, not a thin slice. The geographic story rests on the large majority of the money.

Read plainly, the geography is downtown and the ring around it. The Near South Side, the Loop, the Near North Side, and the Near West Side form a contiguous core wrapping the Chicago River and the central business district. Uptown sits a few miles north along the lakefront. These are not the corners of the city where private capital has historically refused to go. Several of them are among the most actively developed real-estate markets in the Midwest. The Near South Side over the last quarter century has filled in with high-rise condominiums and the campus expansion that turned the old railyards south of Roosevelt Road into the South Loop. The Near West Side has absorbed the medical district, the United Center surroundings, and the tech and loft conversions of the West Loop and Fulton Market. The Near North Side holds the Gold Coast and River North, some of the priciest land in the city. These are the markets least in need of a public push, and they sit at the top of the list.

The tension with the program's stated purpose is worth stating flatly, without dressing it up. TIF is a blight tool. To commit subsidy in these areas the city has to find them blighted or at risk under the statute, and yet these are precisely the areas where developers have been most eager to build with or without help. That does not make any individual deal improper. A specific Loop parcel can be genuinely distressed even when the district around it is booming, and TIF rules let the city act at that grain. But aggregated across four decades and more than six and a half billion dollars, the pattern is hard to square with the rationale. A program sold to the public as the way to reach the blocks the market forgot has, by its own ledger, sent more than half of its single-area dollars to the blocks the market most wants. That a tool justified as a way to reach disinvested places has routed that share into this cluster is the finding around which everything else in this paper turns. We are not yet claiming why this happened, and we will not claim it caused any particular outcome. We are reporting, from the city's own ledger, that the largest share of approved subsidy was committed in the downtown core rather than in the disinvested areas the program names as its reason for existing.

Five neighborhoods against the rest of the city

The ranking shows order. A proportion view shows weight, and weight is what makes the concentration land. So take the same single-named-area dollars and look at them not as a ranked list but as a single pie cut between five neighborhoods and everyone else.

The denominator is the 4,966,907,917 dollars of approved subsidy that carries exactly one community-area tag, the same single-named-area universe from the previous section. The five leading areas, namely the Near South Side, the Loop, the Near West Side, the Near North Side, and Uptown, together hold 2,613,454,156 dollars of that total. Every other single named community area in Chicago, all seventy of them, splits the remaining 2,353,000,000 or so dollars among themselves.

The slices, with their dollar values, tell the story at a glance. The Near South Side holds about 862.9 million dollars. The Loop holds about 749.9 million. The Near West Side holds about 558.8 million, the Near North Side about 259.2 million, and Uptown about 182.7 million. Against those five sits a single slice labeled all other single areas, worth about 2,353.4 million dollars. The five-area slice and the slice for everywhere else come out nearly equal, which means five community areas hold about as much approved subsidy as the seventy other tagged areas hold between them.

That near-balance is the visual core of the follow-the-money argument, and it deserves to be stated without ornament. A small cluster of downtown and downtown-adjacent community areas holds about as much approved TIF money as every other single named neighborhood in Chicago combined. Pilsen, Bronzeville, Englewood, Austin, Rogers Park, Edgewater, Pullman, Bridgeport, Logan Square, every community area outside the five-area core, together, account for roughly the same pile of approved subsidy as the Near South Side, the Loop, the Near West Side, the Near North Side, and Uptown account for by themselves. That is not a subtle tilt. It is a roughly even split between five places and seventy.

The same framing caution applies here that applied to the ranking. This is the single-named-area view, chosen so the 959 million dollar Red Line Extension corridor, which spans five South Side areas, does not get folded into the neighborhood comparison as if it were one place. Including it would only deepen the concentration at the top, not relieve it, since it would add nearly a billion dollars to the concentrated side of the ledger. We hold it apart to keep the neighborhood-versus-neighborhood comparison clean, and we have already disclosed that counting it raises the top-five share from 52.6 to 57.1 percent. Either way, the proportion view says the same thing the ranking did, in a form that is harder to look away from. About half of the city's single-area TIF dollars sit in five downtown and downtown-adjacent neighborhoods.

It helps to look past the slice for the rest and see how thin the tail actually is. The sixth-ranked area, Grand Boulevard, holds about 129.0 million dollars, less than a sixth of the Near South Side's total. Below it the descent is steep. Humboldt Park records about 127.6 million, North Lawndale about 125.3 million, Logan Square about 88.2 million, and then Rogers Park and Austin in the mid-70s of millions each. These are substantial neighborhoods with long development histories, several of them squarely the kind of disinvested community the program names in its own rationale. North Lawndale on the West Side is a long-cited case of mid-century disinvestment, contract selling, and the population and housing losses that followed the unrest of the late 1960s. Grand Boulevard is the heart of historic Bronzeville, the center of Black Chicago in the Great Migration decades. Austin is among the largest community areas in the city by population. Each of them, across forty years of a six-and-a-half-billion-dollar program, drew approved subsidy on the order of a single mid-size downtown deal. The thin slices in the chart are not rounding error or data gaps. Each one is a whole community area, with tens of thousands of residents, whose entire multi-decade share of the program is smaller than several individual downtown agreements.

No causal language belongs here, and we use none. The pie does not explain why the split looks the way it does. It cannot tell you whether downtown areas generated more increment to spend, whether larger projects naturally clustered where land values were highest, or whether the city steered money to the core by design. Those are real questions, and the published literature in a later section speaks to several of them. The chart answers only the descriptive one. Of the approved TIF dollars that carry a single neighborhood tag, very nearly half went to five adjacent areas at the heart of the city, and the other half was spread across all the rest.

A note on wards, and why we lead with community areas

Chicago has two overlapping geographies, and a paper about where TIF money goes has to be clear about which one it is using. The 50 wards are political units, the districts each alderman represents, redrawn every ten years and shaped as much by electoral arithmetic as by neighborhood lines. The 77 community areas are the stable statistical geographies the University of Chicago drew in the 1920s, and they have barely moved since. A ward can split a neighborhood down the middle. A community area holds still. Both are in the file. Ward is recorded for 765 of the 768 projects, 99.6 percent, and community area for 762, 99.2 percent, so either geography is well populated.

We lead with community areas for two reasons. The first is stability. Because ward boundaries are redrawn each decade, a dollar figure summed by ward mixes deals approved under boundaries that no longer exist, and a forty-year total by ward is a total across shifting lines. Community-area totals do not have that problem. The second is comparability. The published Chicago equity literature that this paper sets its findings against is split between the two geographies, and the community-area cut lines up most cleanly with the neighborhood-level analyses, while the ward cut connects to the political-accountability literature. We report the community-area distribution in detail because it is the more stable measurement and the one our own analysis computed end to end. The file also confirms the wide political reach of the program. Approved projects touch 49 wards, nearly every ward in the city, which means the concentration documented here is not a story of a few aldermen monopolizing the tool. It is a story of where the dollars pooled even as the deals spread across almost the entire council.

The ward dimension is where the equity literature has done its sharpest work, and it is worth importing rather than reinventing. The TIF Illumination Project's ward-level accounting, reported through Nonprofit Quarterly, found that majority-Black wards contributed roughly 47 percent of the citywide TIF balance yet received less back than White wards.^[10] That is a contribution-versus-return finding, and it runs on the ward geography because wards are how the increment gets politically claimed. Our community-area distribution approaches the same concentration from the spending-commitment side rather than the balance side, and the two readings agree on direction. The money is committed disproportionately to the downtown core, and the neighborhoods that supply a large share of the underlying tax base see comparatively little of it come back. We keep the ward-level figure attributed to the source that computed it and do not restate it as our own, because our analysis ran on community areas and the honest move is to say which number came from where.

Counting deals is not counting dollars

A ranking by dollars hides something a ranking by deals reveals, and putting the two side by side is where the distribution gets genuinely interesting. Project count and dollar share do not move together. Pull them apart and you can see two very different kinds of TIF geography inside the same dataset, namely neighborhoods stitched together from many small deals and downtown blocks built on a handful of enormous ones.

The cleanest way to see the divergence is to rank the same set of community areas two ways at once. By number of projects, the Near West Side leads the city. It carries 64 approved projects, more than any other single named community area. By dollars, though, the Near West Side ranks third, at about 558.8 million. Divide one by the other and the Near West Side averages about 8.73 million dollars per project, a figure that points to a neighborhood where TIF shows up often, in deals of moderate size, spread across many sites.

Now hold that against the Loop. The Loop carries 44 approved projects, fewer than the Near West Side, but those 44 deals concentrate about 749.9 million dollars, which works out to roughly 17.04 million dollars per project, nearly double the Near West Side's per-project average. Fewer deals, much more money in each.

The Near South Side pushes the pattern to its extreme. It leads the entire city on dollars at about 862.9 million, yet it carries only 29 approved projects, fewer than half the Near West Side's count. The arithmetic is striking. The Near South Side averages about 29.76 million dollars per project, more than three times the Near West Side's per-project figure and well above the Loop's. This is the signature of a place where TIF money arrives in a small number of very large commitments rather than a long list of modest ones.

The rest of the count ranking fills in the texture. The Near North Side carries 41 projects worth about 259.2 million dollars, roughly 6.32 million dollars each. Grand Boulevard carries 31 projects worth about 129.0 million dollars, roughly 4.16 million dollars each, the smallest per-project average among this group and a number that, like the Near West Side's, suggests many modest neighborhood deals. Uptown carries 28 projects worth about 182.7 million dollars, roughly 6.53 million dollars each. Lay the six areas out together and they sort into two rough families. The Near West Side, the Near North Side, Grand Boulevard, and Uptown all sit in the single-digit-millions-per-project range, places where TIF recurs in moderate increments. The Loop and especially the Near South Side sit far above, places where a few deals carry enormous sums.

The civic reading has to be drawn carefully, because the temptation to over-interpret is strong and the data does not support it. A high project count can signal a neighborhood where TIF has been used repeatedly for many smaller interventions, storefront rehabs, mid-size housing, infrastructure pieces, the kind of activity that produces a long list of modest line items. A high dollar total sitting on a low project count signals the opposite, a handful of very large commitments, the kind that attach to major developments, large institutional projects, or single transformative sites. Neither pattern is, on its face, better or worse than the other. A neighborhood with 64 small deals and a neighborhood with 29 huge ones are doing different things with the tool. What the contrast establishes is that the simple dollar ranking from the earlier sections is not the whole picture. The Near West Side leads on activity while the Near South Side leads on money, and both facts are true at once.

The contrast also changes how the earlier concentration finding should be read. The Near West Side, with its 64 deals, is not absent from the program. It is one of the most frequent users of TIF in the city. What it does not have is the per-deal scale that lets a few downtown projects vault past it on dollars. So the concentration documented earlier is not a story of disinvested neighborhoods being shut out of the program entirely. Many of them are in the file, repeatedly. It is a story about the size of the commitments. The largest dollars attach to a small number of very large downtown deals, and no volume of modest neighborhood projects adds up to match a handful of them. A neighborhood can run sixty-four TIF deals and still finish behind an area that ran twenty-nine, if those twenty-nine are large enough. That is precisely what happened on the Near South Side, and it is the mechanism that produces a Gini of 0.710 even though the program reaches nearly every ward.

We make no claim about why these patterns exist. We do not know from this file whether the Near South Side's few large deals reflect deliberate concentration, the natural scale of projects near downtown, the timing of when its TIF districts matured, or some mix of all three. The records show the counts and the dollars. They do not show the intent behind them, and we will not invent one. The point that stands, cleanly and descriptively, is that counting deals and counting dollars are two different measurements, and in Chicago's TIF record they pull sharply apart.

The money got bigger over time

The same lopsidedness that runs across the map runs across the calendar. Using the Community Development Commission approval date that the file records for every project, you can sort the four decades of the program by both dollars approved and projects approved, and the flow swells steeply from one decade to the next.

The 1980s, the program's first years in these records, are a rounding error. Five projects, about 17.8 million dollars in approved subsidy, 0.3 percent of the four-decade total. TIF in Chicago barely registers in the file before 1990. The 1990s pick up but stay modest. Sixty-nine projects, about 522.9 million dollars, 8.0 percent of all approved subsidy. Through the end of the 1990s, in other words, the program had approved well under a tenth of the dollars it would eventually commit.

Then the curve bends upward. The 2000s bring 170 projects and about 1,124.8 million dollars, 17.1 percent of the total, more than double the entire 1990s in both dollars and project count. And the 2010s are the inflection point of the whole history. That single decade carries 296 projects and about 2,570.6 million dollars in approved subsidy, 39.2 percent of everything approved across all four decades. By both measures, dollars and deals, the 2010s are the largest decade in the record by a wide margin. The 2020s, even counting only the years through 2026 that the file covers, stay high. They carry 228 projects and about 2,329.4 million dollars, 35.5 percent of the four-decade total. Put the last two decades together and the 2010s and 2020s alone account for nearly three quarters of all approved TIF subsidy in the file, 74.6 percent.

The plain reading is that approvals accelerated sharply in the 2000s, peaked in the 2010s, and have stayed elevated since. The contrast with the early years is severe. The 1980s and 1990s combined account for 8.2 percent of all approved dollars, while the 2010s alone account for more than 39 percent. Whatever TIF was in Chicago in its first fifteen years, a small and tentative program, it became something far larger in scale from the 2000s onward, and largest of all in the decade after the 2008 financial crisis.

The shape of that curve carries its own argument. The 2010s, the single largest decade in the file, are the decade after the housing crash, a period when private development across the country slowed and then clawed its way back. That Chicago's approved TIF commitments hit their peak in exactly those years tells you the program was not following the private market down. If anything it was running counter to it, the city committing the most public subsidy in the stretch when the market was weakest and then recovering. Whether that is read as counter-cyclical investment, as the city stepping in where private capital had pulled back, or as a sign that the program had simply built up institutional momentum and kept approving deals regardless of conditions, the file does not say. What it shows is the magnitude and the timing. The commitments grew through the boom, did not retreat in the bust, and reached their height in the recovery, and the bulk of that money landed in the same downtown core the geographic ranking identifies.

The flattening into the 2020s deserves one honest qualification of its own. The 35.5 percent share for the 2020s covers only the years through 2026 that the file records, which is most of a decade but not all of it. Even truncated, the 2020s nearly match the full 2010s in dollars, 2,329.4 million against 2,570.6 million, and already exceed every decade before the 2010s combined. So the elevated pace did not end with the 2010s. It carried straight into the 2020s, and the partial decade is on track to be among the largest in the program's history.

The published record offers context the timing is consistent with, and we are careful to frame it as context rather than as something our data proves. In her history of Chicago development finance, Weber documents how heavy TIF designation across the Central and East Loop helped finance the city's early-2000s building boom, the wave of downtown construction that preceded the crash.^[3] Our decade figures sit comfortably alongside that account. The 2000s and 2010s are precisely when the dollars in our file scale up, and the downtown areas that dominate our geographic ranking are the areas Weber describes as the engine of that finance-built overbuilding. We are not claiming our numbers confirm her causal story. We are noting that a city ledger showing approvals concentrating downtown and accelerating through the 2000s and 2010s is the kind of record her history would lead a reader to expect.

One honest caveat governs the entire time series. The approval date marks when the Community Development Commission signed off, not when ground broke, not when money flowed, and not when a project finished. A deal approved in 2021 may have been planned for years and may take years more to complete, and a deal approved in 2009 may reflect a development conceived in the boom that closed only after the crash. So the decade buckets describe the rhythm of approvals, the moments the city committed, rather than the rhythm of construction or spending on the ground. Read that way, the finding is clean and defensible. The city's commitments to TIF grew slowly through the 1980s and 1990s, expanded through the 2000s, and reached their peak in the 2010s, where, by the city's own dating, the single largest share of four decades of subsidy was approved.

How much of each project the public carried

Total dollars tell you where the money landed. They do not tell you how much of each deal the public was actually carrying, and that proportion turns out to point somewhere else entirely. Among the 761 projects that report a subsidy share, the city's own recorded figure has a median of 34.9 percent. Read that number alone and TIF looks like a tool that nudges a deal across the finish line, covering roughly a third of the cost while private money or other sources cover the rest. The mean tells a different story. It sits at 51.1 percent, well above the median, and the gap between the two is the tell. When a mean rides that far above its median, a heavy cluster is sitting at the top of the range and pulling the average up.

We cross-checked the city's reported percentage against the arithmetic, dividing approved subsidy by total project cost for the 757 projects that carry both numbers. The two measures track closely. The arithmetic gives a 35.4 percent median and a 51.3 percent mean, within a few tenths of a point of the reported field on both measures. That agreement matters because it means the subsidy-share column is not a stray field someone filled in carelessly. It lines up with the dollars, so the pattern it shows is real.

The cluster pulling up the mean is concrete and identifiable. Of the 761 reporting projects, 218 are funded at 100 percent, the public covering the entire recorded cost. These are not the downtown towers the dollar ranking is built on. They are overwhelmingly intergovernmental-agreement public works, the kind of deal where one arm of government pays another to build something public, with Chicago Public Schools capital projects the most common example. A school roof or a new building financed entirely through a TIF district shows up in this file as a project funded at 100 percent, because there is no private partner putting up a matching share. Two hundred and eighteen such deals, each pegged at the top of the scale, are enough to drag the citywide mean from a median in the mid-thirties up past the halfway mark.

Then the geography flips. Rank community areas by their median subsidy share, keeping only the areas with at least five reporting projects so a single deal cannot swing the result, and the neighborhoods on top are not downtown. They are the outer wards. Portage Park sits at a 100 percent median across 11 reporting projects. Belmont Cragin matches it across 14. Bridgeport reaches the same 100 percent median across 16 projects, Brighton Park across 7, Jefferson Park across 5, and Irving Park across 5. These are the places where the typical TIF deal, by the city's own dating and accounting, had the public carrying the whole cost, not a slice of it.

Hold that against the dollar ranking and the contrast is sharp. The Near South Side, the Loop, and the Near West Side lead the city in total approved subsidy by a wide margin, yet none of them tops the leverage list. The places where the public carried the largest share of each individual deal are the outer neighborhoods running school and public-works projects at full public cost, while the places that captured the most total money are downtown blocks where the public share of any single large mixed-use deal was often a minority of a very large number. A 30 percent subsidy on a 200 million dollar downtown project is a far larger public commitment in absolute dollars than a 100 percent subsidy on a 4 million dollar neighborhood school, even though the school reads as the more heavily subsidized deal on a percentage basis.

The honest reading is that dollar concentration and leverage intensity are two different questions, and they have two different answers. Where did the most public money go is answered downtown. Where did the public carry the largest share of each deal is answered in the outer neighborhoods, and largely because of a particular kind of public-to-public financing that has nothing to do with private development at all. Neither answer is the whole truth on its own. The subsidy-share field, taken without the dollar field beside it, would point a reader toward Portage Park and Belmont Cragin and away from the Loop, which is precisely backward as a guide to where the program's money is. We report both because the file supports both, and because conflating them is the easiest way to misread what TIF has actually done with the public's money.

There is a second, quieter point buried in the 100 percent cohort. A large share of Chicago TIF money is not subsidizing private development at all. It is moving public dollars from one government account to another to pay for schools and infrastructure. That use is defensible on its own terms. A TIF district can be a serviceable way to fund a school building. But it complicates the picture of TIF as a lever that pulls private investment into a neighborhood, because in 218 of the 761 reporting deals there was no private investment to pull. The increment was simply rerouted into a public project. When critics and defenders argue about whether TIF crowds in private money, the file is a reminder that for a substantial slice of the program the question does not even arise.

The affordable-housing question the records cannot answer

The question most people bring to a program like this is simple. How much housing did the money buy, and how much of it was affordable. The city's own records cannot answer that question, and the way they fail to answer it is itself worth reporting.

Affordable units is the weakest field in the file. Of 768 projects, only 155 report a number for affordable units, which is 20.2 percent. The other 613 projects, 79.8 percent of the file, leave the field blank. Four out of every five approved TIF and redevelopment-agreement projects on the city's books carry no recorded affordable-unit count at all. That is not a rounding problem or a handful of stragglers. It is the majority of the program going dark on the one outcome the public most wants to see.

For the subset that does report, the arithmetic is straightforward and bounded. The 155 reporting projects together account for 11,725 affordable units, against 1,342,741,555 dollars in approved subsidy. Divide one by the other and the reporting subset works out to about 114,520 dollars of approved subsidy per reported affordable unit. That number is real for what it covers. It describes the 155 projects that chose to record a unit count, and it describes nothing else.

The temptation is to take that 114,520 dollar figure, multiply it against the program's full subsidy total, and announce a citywide cost per affordable unit. The data does not permit it, and we will not do it. The subset is not a sample drawn at random from the program. It is a self-selected slice, the projects that happened to report, and there is no way from inside this file to know whether the 613 silent projects built more affordable housing per dollar, less, or none. A large downtown mixed-use deal that produced no affordable units would leave the field blank for one reason. A neighborhood housing rehab that produced many might leave it blank for another. The blanks are not interchangeable, and they are not evenly distributed in any way the file lets us see. Any per-unit figure stretched across the full program would be inventing the 613 numbers that are missing, which is exactly the kind of fabrication the data forbids.

So the responsible statement is narrow. The reporting subset of 155 projects carried 11,725 affordable units against 1.34 billion dollars in approved subsidy, about 114,520 dollars per reported unit, and that figure cannot stand in for the program. The city's approval records, as published, do not let anyone compute a defensible program-wide housing yield per public dollar.

That is not a gap in our analysis. It is a gap in the public record, and it is a finding about transparency before it is anything else. A program that has committed more than six and a half billion dollars in subsidy, much of it justified in part by the promise of affordable housing, does not record affordable-unit counts for four out of five of its own approved projects. The national literature has been making this point for years. Merriman's synthesis of more than thirty studies concludes plainly that jurisdictions should track TIF use far more closely than most of them do, and that in most cases the tool has not accomplished the economic-development goals used to justify it.^[8] Chicago's own file is a case in point. The harder questions about what the money bought cannot be answered until the records are built to answer them.

There is a civic logic to why this matters. TIF money is property-tax growth that would otherwise have flowed to the schools, the parks, and the county. Diverting it is defensible only if the public can see what it bought. When four of every five projects record no housing outcome, the program asks the public to accept the diversion without the receipt. No amount of cleverness with this file will recover counts that were never entered, so the remedy is administrative rather than analytic. It would take a reporting requirement that every approved project record what it delivered, in fields that are actually filled in. Until then, the affordable-housing promise that helps sell the program cannot be checked against the program's own ledger.

What Chicago's researchers already found

Our distribution does not arrive in an empty room. A substantial body of Chicago-specific research has been measuring this program for a quarter century, and the descriptive pattern in the city's file sits squarely inside what that work would lead a reader to expect. We are not claiming the literature proves our numbers, or that our numbers prove the literature. We are setting one against the other to see whether they point the same way, and they do.

Start with the foundational economics. Dye and Merriman, working with data from the Chicago metropolitan area, compared municipalities that adopted TIF against those that did not and found that the adopters showed slower growth in property values across the municipality as a whole.^[2] The interpretation that follows is uncomfortable for the program. If towns that turn to TIF subsequently grow more slowly than towns that do not, then the increment inside a district may be coming partly at the expense of value elsewhere in the same jurisdiction, growth pulled across a boundary and relabeled rather than created. That cuts at the heart of the justification for the tool. If the development would have happened anyway, somewhere nearby, then a program that claims to generate new value is, at least in part, capturing and rebranding growth that was already on its way. Lester pressed the same question with Chicago time-series data and reached a compatible answer, finding little evidence that the city's TIF districts produced net new jobs or development beyond what citywide trends would predict.^[4] The but-for test, the legal and rhetorical core of TIF, namely the claim that the development would not happen but for the subsidy, does not hold up well against either study. And the national picture is no kinder to the optimistic case. Merriman's synthesis of more than thirty studies for the Lincoln Institute concluded that in most cases TIF has not accomplished its economic-development goals, and recommended that jurisdictions track its use far more closely than they do.^[8] That recommendation is not abstract for Chicago. As the affordable-housing section showed, the city does not record a basic outcome for four of every five of its own approved projects.

These causal results are not what our file measures, and we do not borrow their weight to make a causal claim of our own. But they reframe what a concentration finding means. If TIF largely redistributes growth rather than creating it, then where the city chooses to commit the increment is close to the entire question, because the program is mostly deciding which places get the diverted public dollars rather than conjuring new value out of nothing. A distribution skewed toward the downtown core is, under that reading, a distribution of advantage. The increment that downtown districts capture is property-tax growth held back from the citywide funds that serve every neighborhood, then spent on the blocks that already command the highest land values in the region.

Weber's history fills in how the money actually moved, and it is the source that speaks most directly to our geographic and time findings at once. In her account of how finance built the new Chicago, heavy TIF designation across the Central and East Loop helped underwrite the city's early-2000s building boom, the wave of downtown construction that the records in our own file show accelerating through the 2000s and into the 2010s.^[3] The argument there is not simply that downtown got TIF money. It is that the increment became a financial instrument, a stream the city could pledge and structure to support the bond issues and the deals that put up the towers, so that TIF designation downtown was less a remedy for blight than a mechanism for channeling capital into the central area when the market was already running hot. The dollar concentration we describe, with the Loop, the Near South Side, and the Near West Side at the top, is the geographic residue of exactly that history. Our decade curve and her narrative line up almost beat for beat. The money in our file scales up in the 2000s, peaks in the 2010s, and stays high, which is the rise, the crest, and the aftermath of the boom her book reconstructs.

A line of related work establishes that TIF's effects, whatever their sign, are not spread evenly across the city, which is the empirical foundation under the entire question of where the dollars land. Weber, Bhatta, and Merriman found that TIF designation could raise industrial property values in Chicago, a measurable effect, and one that varied across districts rather than holding constant.^[6] In a separate study the same authors traced spillovers from TIF districts onto the residential housing prices of surrounding areas, evidence that designating a district does not stop at its boundary but reaches into the blocks around it.^[7] Kane and Weber then showed that the relationship between TIF-funded municipal investment and property appreciation varied sharply depending on where in the city the investment happened.^[5] That last finding is, in a sense, the whole motivation for an analysis like ours. If a dollar of TIF investment does very different things in different parts of Chicago, then the map of where the dollars go is not a footnote to the program. It is the program. A concentration of commitments downtown is also, by this literature, a concentration of whatever effects TIF produces, gathered in the part of the city that had the least claim on a blight-justified subsidy to begin with.

The equity literature makes the stakes of that geography explicit, and it is where the program's distributional politics come into focus. Hackett's analysis reported that nearly half of Chicago's TIF money was spent in the Loop and the wealthier, whiter areas surrounding it, and documented a negative correlation between TIF spending and the Black and Hispanic share of an area's population.^[9] The framing in the piece is blunt, redevelopment for whom, and the answer it reaches is that the program functioned in practice as a channel moving public funds toward already-advantaged parts of the city. The direction of that finding matches what our file shows from a different angle, namely a small cluster of downtown and downtown-adjacent areas holding the bulk of the approved dollars. The TIF Illumination Project's accounting, reported through Nonprofit Quarterly, sharpened the point into a contribution-versus-return frame, finding that majority-Black wards contributed roughly 47 percent of the citywide TIF balance yet received less back than White wards.^[10] That framing matters because it answers an objection. One might argue that downtown captures more TIF money simply because downtown generates more increment, so the concentration is mechanical and fair. The contribution-versus-return finding cuts against that. If majority-Black wards are supplying a large share of the balance and seeing less return, then the increment those neighborhoods generate is not flowing back to them in proportion. The downtown concentration is then a concentration of spending, not merely of where the increment happened to arise, and some of what is spent there was generated elsewhere.

Loyola's Institute for Racial Justice put concrete numbers on the South Side end of the gap. Looking back over more than twenty years, it reported that two South Chicago TIF districts generated only 26.5 million and 32.2 million dollars across their lives, at a time when citywide TIFs held a balance of roughly 3 billion dollars in 2023, and it read that gap as a sign of the limits of TIF as a core development tool for disinvested areas.^[11] The contrast with our own numbers is stark when the figures are set side by side. A South Side district that produced 26.5 million dollars over two decades sits against the 862.9 million dollars of approved subsidy our file records for the Near South Side and the 749.9 million for the Loop. One neighborhood's entire multi-decade TIF history is a rounding error next to a single downtown area's committed total. The Loyola reading is also a caution about the tool's logic in poor neighborhoods. TIF can only spend what the increment generates, so a district drawn around low and slow-growing property values will throw off little to reinvest, which means the places the program names as its reason for existing are structurally the places where it produces the least. The blight rationale and the financing mechanism work against each other. The neighborhoods most in need of investment are, almost by definition, the neighborhoods whose frozen tax base will generate the smallest increment to spend.

It is worth being precise about how these sources differ from ours, because the differences are what make the agreement meaningful. Hackett and the TIF Illumination Project work largely from collection and balance data, the increment a district takes in and the fund balances that accumulate. Loyola tracks two specific South Side districts over their full lives. We work from the approval side, the subsidy the Community Development Commission committed to projects, tagged to the community area where each project sits. These are not the same measurement. A district can collect a large increment and spend it slowly, or commit a large subsidy and collect a modest increment. That four independent vantage points, collections, balances, district lifetimes, and our approvals, all describe the same downtown concentration and the same neighborhood shortfall is the reason the pattern reads as durable rather than as an artifact of one dataset. Each method has its own blind spots, and a quirk that distorted one of them would be unlikely to distort the other three in the same direction, so their agreement is harder to wave off than any single study would be.

What unites this work is a consistent direction. Older studies question whether TIF creates value or merely redistributes it. Newer ones document where in Chicago the value and the money have actually concentrated, and who has been on the short end. Our descriptive distribution is one more reading consistent with that body of evidence, produced from the city's own approval records rather than from assessed-value panels or ward-level balances. Consistency across independent sources and methods is corroboration. It strengthens the case that the pattern is real and durable. It is not, and we do not claim it to be, proof that TIF caused any particular neighborhood outcome. Our file describes where the approved dollars are. The literature gives that map its meaning, and the meaning it gives is not flattering to the promise the program was sold on.

What this does and does not show

Everything in this paper is descriptive. We counted, ranked, summed, and divided the city's own approved TIF and redevelopment-agreement records, and we reported what we found.^[1] We did not run an experiment, model a counterfactual, or estimate what any neighborhood would look like without TIF. We make no causal claim about what the spending caused, and no city-wide claim about affordable housing. Before anyone carries these numbers somewhere else, the limits need to travel with them.

Approved subsidy is a commitment, not a disbursement. Every dollar figure in this paper describes money the city agreed on paper to commit when the Community Development Commission signed off, not money confirmed out the door. A project can be approved and scaled back, delayed, or restructured. The file records the promise, and the promise is what we measured.

Geography is incomplete and, in a few cases, multi-valued. Community area is present for 762 of the 768 rows, so a small number of projects carry no neighborhood tag at all, and the dollars carrying a community-area tag come to 5,939,387,695, about 90.5 percent of the approved total. Three rows tag more than one community area, joined by commas. The largest single record in the entire file, the 959 million dollar Red Line Extension TIF intergovernmental agreement, is one of those three, a transit corridor spanning Morgan Park, Riverdale, Roseland, Washington Heights, and West Pullman rather than a single neighborhood. As a grouping it ranks first in the file, ahead of any one community area, but it is a corridor, not a place a reader would call a neighborhood.

That one row is why the headline concentration figure comes in two versions, and we report both rather than pick the flattering one. Counting the top five community-area groupings, which folds the Red Line corridor in, the top five hold 57.1 percent of community-area-tagged dollars. Restricting to the top five single named community areas, namely the Near South Side, the Loop, the Near West Side, the Near North Side, and Uptown, and setting the multi-area corridor aside, the top five hold 52.6 percent of single-named-area dollars. The charts in this paper use the 52.6 percent framing on purpose, to keep a transit corridor from being compared against neighborhoods as if it were one. Both numbers point the same way. By either count, a small group of downtown and downtown-adjacent areas holds about half of the city's tagged TIF dollars, and the Gini coefficient of 0.710 across community-area groupings says the same thing in one number. The choice between 52.6 and 57.1 changes the decimal, not the conclusion.

The leverage figures carry their own caveat. The mean subsidy share of 51.1 percent is not the typical deal. It is lifted above the 34.9 percent median by the 218 projects funded at 100 percent, overwhelmingly intergovernmental-agreement public works such as school capital projects, and those same projects are why outer neighborhoods rather than downtown top the median-leverage ranking. Anyone quoting a single leverage number should quote the median and explain the bimodal shape behind the mean, or the number will mislead.

The affordable-housing subset cannot stand in for the program, and this bears repeating one last time because it is the figure most likely to be misused. The 11,725 units, the 1.34 billion dollars, and the roughly 114,520 dollars per reported unit describe the 155 projects that reported a unit count, about a fifth of the file. They say nothing about the other 613 projects, and they are not a citywide cost per affordable unit.

Two smaller administrative quirks round out the honest accounting. A single UTF-8 mojibake artifact appears in some text descriptions, where smart-quote bytes render as replacement characters, and it touches none of the numeric or geographic fields this analysis rests on. Two supplementary files shipped in the same repository, a 40-project financial-incentive file and a 27-row Bronzeville file, are consistent with the main record and do not change any figure here.

There is also a limit in what a distribution can and cannot settle, and it is the most important one to keep straight. Showing that the money concentrated downtown does not, by itself, prove the program failed the neighborhoods it named. Two readers can look at the same 52.6 percent and reach different judgments. One sees a tool captured by the center, spending where investment was least needed. Another sees a city making large, defensible bets on downtown anchors, transit, schools, and institutional projects whose benefits, the argument would run, spill outward to the rest of the city. This file cannot adjudicate between those readings, because adjudicating would require measuring outcomes the file does not record and tracing causation it cannot support. What the file does is set the terms of that argument honestly. Anyone who wants to defend the distribution now has to defend it as a distribution, with the downtown concentration and the thin neighborhood tail in full view, rather than from the comfortable assumption that a blight tool naturally reaches blighted places. The numbers shift the burden of proof onto whoever wants to call the distribution fair. They cannot settle the underlying argument by themselves, and this paper does not pretend they can.

None of this is a reason to dismiss the file. It is a reason to read it for what it is. The value of this work is a clear, reproducible map of where the city's approved TIF dollars land, paired with an honest accounting of what the public record does not capture. The record is strong on dollars, dates, wards, and community areas, and it is weak to silent on affordable-housing outcomes. Better outcome reporting, the kind Merriman's national synthesis has been urging for years, would let the harder questions actually be answered, namely what the six and a half billion dollars bought, for whom, and at what neighborhood cost.^[8] Until those fields are filled in, the program will keep asking the public to trust a distribution it cannot fully audit, and the public record will keep answering only the narrow question this paper could answer, which is where the approved dollars landed. On the city's own books, the answer is downtown.

Which returns us to where we began. Tax increment financing is sold as a way to steer public money into the disinvested places private capital skips. On the city's own books, the approved dollars concentrate downtown, in the Loop and the blocks around it, and they grew fastest in the decade Chicago's own researchers describe as the height of finance-built downtown development. We are not calling that a verdict. We are calling it a description, drawn straight from the public record, and a question that record hands to anyone responsible for the program. A tool justified by the neighborhoods it was meant to reach has, by its own ledger, sent the largest share of its money to the places that needed it least. The records show where the money goes. What to do about that is a policy question, and it is now squarely on the table.