The 606 and Displacement Pressure in Humboldt Park

Abstract

On June 6, 2015, Chicago opened the Bloomingdale Trail, a 2.7-mile elevated greenway known as the 606, on a disused rail embankment that runs west through Humboldt Park, West Town, Logan Square, and Avondale. This paper joins an honest synthesis of published research to our own descriptive analysis of two real public datasets. It makes no causal claim and computes no price effect of its own. A peer-reviewed and institutional literature establishes that home prices on the trail's lower-income western segment rose sharply after the opening, with the DePaul Institute for Housing Studies reporting western-segment median sale prices climbing from $97,000 in 2012 to $462,000 in 2018, close to a quadrupling, while the eastern segment changed little ^[6]. Against that documented pressure, we count what stands on the other side of the ledger. Using the City of Chicago affordable rental developments list and its TIF and RDA project records, both already mirrored in our repository, we census the subsidized stock and committed public subsidy in the four corridor community areas ^[1]. The corridor holds 119 affordable rental developments totaling 2,391 units, and 20 residential TIF and RDA projects committing 1,068 affordable units backed by roughly $145.9 million in approved subsidy ^[1]. Of those affordable units, 88.4 percent sit west of Western Avenue, where the literature locates the steepest appreciation ^[1]. We report the public counterweight at its real, modest scale, and we keep the borrowed price figures and our own counts strictly apart, because they come from different measurement efforts and cannot be fused into a quantity nobody measured.

June 6, 2015, and the ribbon over the embankment

For most of a century the embankment was just there, a low ridge of earth and ballast and old steel running east and west across the near Northwest Side, carrying freight above the cross streets so the trains and the children would not meet at grade. Chicago built thousands of such ridges in the railroad century, raising the lines onto berms after too many people died at the crossings. The Bloomingdale Line was one of them, a corridor that climbed up over Bloomingdale Avenue and stayed up, hauling lumber and coal and boxcars past the back fences of Humboldt Park and West Town and Logan Square. Then the freight thinned, and then it stopped, and the trains left the embankment to the weeds. Volunteer ash trees took the ballast. Foxes and possums took the ash trees. For two decades the ridge was a green secret, a wild seam stitched through dense brick neighborhoods, visible from the street only as a line of treetops where no trees should be.

On June 6, 2015, the city put a ribbon across it and opened it to the public. The disused rail embankment had become the Bloomingdale Trail, the spine of a larger park system marketed under the name of the area code that covers the Northwest Side, the 606. The path runs 2.7 miles, a continuous elevated greenway lifted a story or so above the street grid, with access ramps and small ground-level parks set at intervals along the way. You can get on near Ashland on the east end and walk or ride west, the city falling away below the parapet, until the trail sets you down again near its western terminus. It crosses four community areas as it goes, namely Humboldt Park, West Town, Logan Square, and Avondale, four of the seventy-seven areas the city uses to carve itself into legible pieces. The opening drew crowds. People who had grown up beneath the embankment, who had thrown balls over it and wondered what was up there, climbed the ramps and found out.

A greenway is an easy thing to love and a hard thing to read. The path was, and is, a genuine public good. It gave a dense and historically park-poor stretch of the city a continuous ribbon of open space, a safe route for a kid on a bike, a place to run that was not a sidewalk and not a street. Nothing in this paper disputes that. Hold the path in one hand. In the other, hold the ground it crosses, because the ground beneath and beside a new amenity does not stay still once the amenity arrives. Land near a thing people want becomes a thing people want. The trail cast a shadow, and the shadow was a revaluation, slow at first and then not slow, of the lots and two-flats and corner buildings that the embankment had run behind for a hundred years. The visible good of the path and the quieter repricing of the land are the two halves of this account, and they do not cancel. They sit side by side.

They also sit unevenly, because the corridor is not one place. A rough seam runs north and south through it, and the seam falls near Western Avenue, a major arterial that has long marked a social as well as a cartographic boundary on this part of the Northwest Side. East of Western, toward the lake, the neighborhoods had already gentrified or were well along, wealthier and whiter, the trail arriving as one more amenity in places that were already expensive. West of Western, the corridor ran through a lower-income and heavily Latino stretch, the part of Humboldt Park and the western reach of Logan Square and Avondale where the housing was older, the incomes lower, the cushion against rising costs thinner. The same elevated path threaded both, and it did not land on both the same way. A new bench in a rich neighborhood is a bench. A new bench in a poor neighborhood, if it lifts the value of every parcel around it, can be the leading edge of a wave.

It did not take long for people to start naming the wave. Within a few years of the opening, residents, organizers, journalists, and academic researchers were describing the western segment of the 606 in nearly the same terms, as a textbook case of a green amenity meeting displacement pressure, of greening that arrived without the protections that might have let longtime residents stay to enjoy it. The qualitative work captured the feeling directly. In one interview and observation study, Latinx residents along the western segment described experiencing exclusion on and around the new path and read the trail itself as a vehicle of the gentrification reshaping their blocks ^[7]. Not everyone agreed on the mechanism. But the pattern was legible enough, and fast enough, that it became part of how the corridor talked about itself.

The geography of that wave was not symmetric, and the asymmetry runs through everything that follows. Picture the corridor as a line drawn west from Ashland, and picture Western Avenue crossing it perpendicular, roughly a third of the way along. East of that crossing the trail runs through blocks that had already turned over, where a continuous elevated park was a welcome addition to a market that had long since stopped being affordable to the families who once filled it. West of the crossing the same trail runs through blocks where the turnover was still in motion, where the older two-flats and corner buildings still housed working families at rents the open market had not yet caught up to. A single piece of infrastructure, built and opened all at once, landed on two markets at two different stages, and the literature that grew up around the 606 spent most of its energy on the western stage, because that was the stage where something could still be lost. This paper keeps returning to Western Avenue for the same reason the researchers did. It is the line on which the story changes character.

Several named actors built the thing. The Trust for Public Land led the capital project and the land assembly. A local group, Friends of the Bloomingdale Trail, had carried the vision for years and lent it grassroots legitimacy. The City of Chicago, through its parks and planning apparatus, was the public partner that made the embankment a public space. How that coalition was structured, and what it did and did not take responsibility for, matters a great deal to the displacement story, and the published case study of the 606 puts the governance arrangement at the center of its argument ^[2]. That analysis belongs to the literature section. The question this paper actually asks is narrower than governance and narrower than blame.

Here is the question. Set aside, for now, whether the trail caused anything, because we are not going to try to prove that it did. Take as given, from the published record, that the western segment saw steep appreciation after 2015. Then ask a different and more answerable thing. What did the public sector put on the other side of the ledger? While the land beneath the path was being revalued upward, what subsidized and affordable housing already stood in these four community areas, and how much public money had been committed to keep some of it affordable? Those are countable things. They live in real city files. This paper counts them, and reports the count at its true scale, which is modest. A trail, and its shadow, and the small, real weight the public set against the shadow. That is the whole of it.

What this account can and cannot claim

Let me be plain, early and in the same voice as the rest, about what kind of document this is, because the subject invites overreach and the house style here refuses it. This paper is two things bound into one argument. One half is a synthesis of published research, an honest gathering of what peer-reviewed and institutional studies have already established about the 606 specifically and about green gentrification generally. The other half is our own descriptive census, a count drawn from two real City of Chicago datasets, of the affordable housing and the public subsidy that exist along the corridor today. The synthesis supplies the pressure. The census supplies the counterweight. Neither half pretends to be the other.

So the bright line. We did not run an experiment. We did not interview a single resident. We did not build a hedonic price model, and we did not estimate a treatment effect. We did not measure any change in price or rent ourselves, not for a block, not for a tract, not for the corridor as a whole. Every price figure, every premium, every percentage of appreciation that appears anywhere in this paper is borrowed from the cited literature and attributed to it in the text with a footnote marker. When you read that western-segment median sale prices rose from $97,000 to $462,000, that number is the DePaul Institute for Housing Studies speaking, not us ^[6]. When you read that a comparable greenway lifted nearby values by some double-digit percentage, that is the published Atlanta study speaking, not us ^[4]. We are the messengers for those numbers. We did not generate them.

What we did generate is the count. Every unit of affordable housing, every dollar of approved subsidy, every share and ratio that describes the corridor's affordable stock, is our own tally from two files, and it carries its own footnote ^[1]. Those counts are the boundary of our original contribution, and we mean to stay inside it.

The distinction does real work in the argument, because it changes what each kind of number is allowed to do. A borrowed price figure can establish that pressure existed and roughly how large it was, but it cannot be recombined with our counts into some new quantity we never measured, because the moment two numbers from two unrelated measurement efforts get multiplied or subtracted against each other, the result is a fabrication wearing the costume of arithmetic. So the price figures stay in the literature section and the limits section, where they are clearly labeled as someone else's findings, and the counts stay in the analysis sections, where they are clearly labeled as ours. When the two finally appear on the same page near the end, they appear side by side and unjoined, two facts about the same stretch of city that we decline to fuse. That separation is the spine of the whole document, and most of the discipline in what follows is the work of holding it.

Why lean on the literature for the prices at all, rather than computing them ourselves? Because the data that would let us price the corridor honestly are not in our repository, and we will not fake what we do not have. We looked. The rent series we hold for Chicago covers only a set of South Side ZIP codes and does not reach the corridor's 60622, 60639, and 60647, so it cannot speak to the 606 ^[1]. No Cook County parcel sales file and no American Community Survey table by census tract was ever shipped to the corridor in our data. To compute a local price effect from data we do not have would mean inventing rows, and inventing rows is the one thing this site does not do. Honesty therefore requires that appreciation be sourced from outside, from researchers who held the sales records and did the modeling, and that our own contribution stay limited strictly to what we can actually count. We do not apologize for that limitation, because a tally confined to what the files actually record is exactly what makes the count trustworthy, immune to inflation by a borrowed price.

The two sources we do hold are easy to name. One is the City of Chicago affordable rental developments list, maintained by the Department of Housing, a record of subsidized and affordability-restricted rental properties with their addresses, unit counts, and coordinates ^[1]. The other is the City's TIF and RDA project records, maintained through its planning and development apparatus, a log of projects financed by tax increment financing and redevelopment agreements, including approved subsidy amounts, total project costs, and committed affordable units ^[1]. Both files are public. Both are already mirrored in our repository, and the corridor subset of each is reproducible from the files we publish alongside this paper. We did not collect them in the field. We received them, filtered them to the four corridor community areas, and counted.

The governing stance follows directly. Our numbers describe a pattern. They do not prove a cause. We can tell you, with confidence, how many affordable units stand in these four areas and where they sit relative to Western Avenue, because we counted them in a file. We cannot tell you, and will not pretend to tell you, whether those units offset displacement, or whether the trail displaced anyone, or by how much, because we did not measure any of that. Where we are unsure that a figure is real, the figure does not appear. The pre-analysis plan for this work carried a rough early guess at the TIF numbers, and when the real data contradicted the guess we threw the guess out and reported the data ^[1]. That is the rule applied to ourselves.

What you are holding, then, is a descriptive ledger, not a verdict. It lays the documented pressure and the countable response side by side and lets the reader see the distance between them. It declines to rule on whether the response was adequate, because adequacy is a measurement we did not make. A ledger can be honest without being a conclusion. This one tries to be exactly that, and no more.

The case the researchers have already made

Begin upstream of Chicago, with the idea itself, because the 606 did not surprise the people who study green space. There was already a name for what they expected. A foundational synthesis in the environmental-justice literature established that new urban green space, for all its public-health value, tends to flow toward affluent and predominantly white residents, and that absent deliberate protection it can drive lower-income communities out of the very neighborhoods being greened ^[8]. The authors proposed a deceptively modest corrective, the idea that a neighborhood might be kept "just green enough," improved sufficiently to share the health benefits of nature without being polished into unaffordability ^[8]. The phrase came with a case attached. It had been coined in Greenpoint, Brooklyn, where researchers watched environmental cleanup and greening proceed without any accompanying anti-displacement policy, and where the predictable result was environmental gentrification, the neighborhood improving its way out from under the people who had lived through its dirty years ^[9]. That Greenpoint study is the policy counterfactual hanging over every greenway built since. Green the place, but protect the people, or watch the improvement become an eviction notice.

The mechanism is not abstract, and it has been measured on infrastructure closely comparable to the 606. The clearest analog is the Atlanta BeltLine, another rails-to-trails conversion that turned a defunct rail corridor into a linear park and path. Studying it, Immergluck and Balan found that proximity to the BeltLine raised nearby home values well above what the rest of the metropolitan market was doing, reporting a relative premium on the order of 17.9 to 26.6 percent for the 2011 to 2015 window ^[4]. That is the greenway-gentrification mechanism stated in dollars. A new linear amenity does not lift all of a region's prices equally. It lifts the prices next to itself, and it lifts them by enough to matter to anyone living there on a fixed or modest income. The BeltLine work is the empirical bridge the 606 researchers stand on when they argue that a Chicago greenway would do something similar.

What makes the BeltLine comparison more than an analogy is the shape of the premium, which matters more here than its size. A relative premium means the homes near the trail rose faster than comparable homes elsewhere in the same metropolitan market, which is the part that bites. If every house in a region appreciates together, a longtime owner is no worse off relative to her neighbors, and a renter faces a rising tide she at least shares with everyone. A localized premium is different. It singles out the blocks next to the amenity, lifting them away from the surrounding market, so that the people living closest to the new green space are precisely the people whose housing costs climb fastest. The closer you live to the thing everyone now wants, the more the wanting shows up in your rent or your tax bill. That is the geometry the 606 researchers expected to find in Chicago, and it is why they went looking on the blocks nearest the embankment rather than across the corridor as a whole.

Bring it home to the embankment. The most direct scholarly treatment of the 606 itself is a case study by Rigolon and Németh, and its argument is as much about governance as about land ^[2]. The trail was built by a coalition that was fragmented and substantially nonprofit-led, with the Trust for Public Land and Friends of the Bloomingdale Trail carrying much of the project, and that arrangement had a consequence. Affordable-housing protection was nobody's clear job. The coalition was, in the authors' framing, not in the business of housing, so the housing question went unaddressed while the park advanced, which is precisely the condition under which environmental gentrification takes hold ^[2]. The authors call this the nonprofitization of green infrastructure, the handing of a public project to a constellation of mission-driven organizations each responsible for its own piece and none responsible for the housing consequences the whole produces ^[2]. The trail got built, and built well, because the coalition was good at building trails. The displacement went unmanaged because managing it sat outside every member's mandate. That governance lesson is worth holding next to our own count, because the affordable units we tally were placed by a different set of public hands than the ones that built the path, and nothing in either file we hold says those two efforts were ever coordinated.

The same authors widened the lens in a later study across many Chicago parks, asking what distinguishes the green spaces that trigger gentrification from those that do not ^[3]. Their answer cut against the intuition that bigger parks matter more. Size did not predict gentrification. Function and location did. New active-transportation greenways, paths built for walking and cycling, placed in historically disinvested areas, were the type most associated with subsequent gentrification, which describes the 606 with uncomfortable precision ^[3]. The finding matters for how alarmed to be in advance. A city cannot easily know which of its many green investments will set off a revaluation, but this work says the riskiest profile is a specific one, a brand-new path for walking and biking dropped into a neighborhood that had been starved of investment, which is to say the exact profile of the western 606. The trail was not a large park, and on the older intuition that would have made it relatively safe. On the function-and-location reading it was close to the worst case a city could pick.

The numbers that anchor the price story come from the DePaul Institute for Housing Studies, and they deserve careful handling because they are the figures this whole paper borrows rather than computes. The Institute built a hedonic house-price model, the standard econometric tool for isolating the effect of a single attribute on sale price while holding other characteristics constant. The idea is to compare houses that are alike in the ways that ordinarily drive price, namely size, age, lot, condition, and the rest, so that whatever difference remains can be attributed to the one feature under study, here nearness to the trail. The technical paper rested on single-family sales recorded across Cook County, a long run of transactions from 1997 to 2016, covering 281 census tracts within three miles of the trail ^[5]. That is a deliberately wide net in both time and space. Reaching back to 1997 gives the model years of price behavior from well before the trail was even proposed, a baseline against which any post-opening jump can be seen. Drawing the boundary at three miles and 281 tracts gives it a large surrounding market to serve as the comparison, so that a premium near the trail is measured against the ordinary movement of the wider area rather than asserted in a vacuum.

The finding split exactly along the seam this paper keeps returning to. West of Western Avenue, in the lower and moderate market, the model found large and highly statistically significant price premiums associated with proximity to the trail. East of Western, the effect was stable and minimal, close to nothing ^[5]. Same trail, two outcomes, divided by an arterial. The asymmetry is the part that should not be glossed. It is not that the trail raised prices everywhere and merely raised them more in the west. East of Western the model found essentially no trail effect at all, which fits the intuition that an amenity adds little to a market that has already priced in everything desirable about a location. The action was entirely on the lower and moderate side, where there was still room for a new desirable feature to move the price. The methodology is laid out in full in the public document, which is part of why these numbers are safe to lean on. We did not have to take the headline on faith. The model that produced it is described step by step, the data window is stated, and the geographic split is the authors' own, not a frame we imposed.

The Institute's companion release put a headline trajectory on the western premium, and it is the single most cited number in any honest account of the 606. On the western segment, median sale price rose from $97,000 in 2012 to $462,000 in 2018, close to a quadrupling, across the window that brackets the trail's opening ^[6]. That dollar move, the price of a home on the lower-income side of the corridor multiplying by close to four in six years, is the appreciation every other account of the 606 cites. The same release sounded a more specific alarm about what that appreciation threatened. The corridor's stock of unsubsidized two-to-four-flat buildings, the small multifamily housing that has long served as a quiet engine of naturally affordable rental units in Chicago neighborhoods, was eroding as prices climbed, the buildings converted or sold or priced into a different market entirely ^[6]. That stock is the most exposed when land revalues, because it is affordable by circumstance rather than by covenant, and circumstance is exactly what rising prices erase. Keep that exposed two-to-four-flat stock in mind, because it is the backdrop against which our own modest counterweight has to be read.

Numbers describe the pressure from above. The lived experience describes it from inside, and the qualitative record fills that in. The interview and observation study of the western segment documented residents, many of them Latinx and long-rooted on these blocks, who experienced the new trail less as an amenity they had been given than as a boundary that excluded them, and who understood it explicitly as a mechanism of the gentrification remaking their neighborhood ^[7]. The price data does not correct that reading so much as confirm it from another angle, the same phenomenon felt rather than measured, the human side of the documented appreciation. People know when the ground is shifting under them, and they named what they were standing on.

Step back from the corridor and the whole sequence has a name and a shape in the theoretical literature. Rigolon and Collins describe a green gentrification cycle, a self-reinforcing loop in which a new green amenity raises surrounding property values, the higher values attract wealthier residents, the wealthier residents displace incumbents, and the now-more-affluent neighborhood becomes the constituency for still more green investment, which lifts values again ^[10]. The 606 fits as one turn of that wheel, a stage in a cycle rather than a one-time event, which is part of why the appreciation did not simply plateau. The loop, once started, tends to keep going.

End the literature honestly, with its own internal caution, because the most rigorous version of this story refuses to be universal. Comparative work across multiple urban trails found that trail-linked gentrification is real but context-dependent, not an automatic consequence of building a path ^[11]. Some trails in some places coincided with little neighborhood change. What makes the comparison sting is where the 606 lands within that range. In the cross-trail comparison, Chicago's 606 sits at the high-displacement end, among the clearest cases of a trail accompanied by the kind of change that pushes lower-income residents out ^[11]. So the literature does not say every greenway gentrifies. It says this one is close to the worst case observed, which is the opposite of an exoneration. That is the documented pressure. The rest of this paper turns to the far smaller thing we can count against it.

Counting what stands on the other side

Set the literature down for a moment and walk the ground, because this is where our own contribution begins, and it begins as an act of plain counting. A trail running west across four community areas passes through four distinct housing markets, and each of those markets already held some amount of subsidized and affordable rental supply before the embankment ever became a greenway. That supply did not arrive as a single program or a single act of council. It accumulated over decades through the City's Department of Housing, through the Affordable Requirements Ordinance, through senior and supportive-housing developers, through the slow institutional work that rarely makes a headline. We set out to count it, plainly, in the four areas the trail touches, and to report the total at whatever scale it turned out to be.

The City of Chicago maintains a public list of affordable rental developments, naming each property, its address, its community area, its unit count, and its coordinates ^[1]. We restricted that list to the four corridor community areas by exact match on each row's own community-area field, and we counted. The four areas hold 119 affordable rental developments totaling 2,391 units, and the count is clean in a specific and reassuring way ^[1]. Within the corridor subset, not one development is missing its unit count, zero rows out of 119, so the 2,391 figure is a full tally rather than a partial one padded by guesses ^[1]. When we report 2,391 units, we are not estimating around holes in the data. We are reading every row. That is the standing inventory, the visible institutional answer, accumulated over years, to the kind of pressure the researchers describe.

A word about why the exact-match rule, plain as it sounds, is the honest choice rather than simply the easy one. There are looser ways to decide whether a building belongs to the corridor. We could have drawn a buffer of some chosen width around the trail and swept in every development inside it, or hand-picked properties that felt close enough on a map. Both introduce a judgment call the reader cannot check, and both invite the quiet temptation to nudge the boundary until the count comes out where one wants it. Matching on the file's own community-area field removes that discretion. A development is in the corridor if the City's own record says it sits in Humboldt Park, West Town, Logan Square, or Avondale, and not otherwise. The rule is reproducible by anyone with the same file, it does not bend to the result, and its one real cost, the dropping of a few ambiguously labeled rows, falls in the direction of undercounting rather than inflation. We would rather be reproducibly conservative than impressively vague.

The distribution across the four areas is uneven in a way that maps onto the geography the literature cares about. Humboldt Park, the namesake western neighborhood and the one most often named in accounts of 606 displacement, holds the largest single share, 45 developments and 1,018 units, nearly the western half of the entire corridor inventory concentrated in one neighborhood ^[1]. Logan Square comes next by units, 32 developments and 657 units ^[1]. West Town sits close behind, 36 developments and 640 units, slightly more developments than Logan Square but slightly fewer units, which tells you its developments run a touch smaller on average ^[1]. Avondale, the northern end of the corridor and the area the trail reaches last, trails well back, 6 developments and 76 units, a thin presence at the corridor's northwest end ^[1]. Add the four and you return to 119 developments and 2,391 units, the whole standing inventory accounted for area by area ^[1].

Read that bar chart slowly, because the shape of it carries the point. Humboldt Park alone holds more affordable units than Logan Square, and it nearly equals West Town and Avondale together. The standing affordable stock is heaviest at the western, lower-income end of the line, the same end where the DePaul work locates the steepest post-2015 appreciation ^[6]. We are not yet making anything of that alignment. We are only noting that the inventory is not spread evenly along the 2.7 miles, that it clusters where the neighborhoods were poorer and the later price pressure was sharpest, and that Humboldt Park carries the load.

The composition of that inventory has a texture worth describing, since 2,391 units is not a single kind of housing, though we describe it carefully and without reading more into it than the file supports. The City's affordable rental list sorts the corridor developments by property type, and the picture is mostly conventional family rental with a smaller band of specialized housing layered in. Multifamily developments make up the largest category at 61 properties, and developments built under the Affordable Requirements Ordinance, the city rule that obliges certain market-rate projects to set aside affordable units or pay into a fund, account for 42 ^[1]. Together those two categories are the bulk of the count. The remainder is specialized stock. Senior developments number 5, supportive-housing developments another 5, with a scattering of properties for veterans, for young men aging out of other systems, and for people with disabilities ^[1]. The mix is the ordinary accumulated portfolio of a city rather than a monoculture of one program, family buildings and inclusionary set-asides doing most of the work, with senior and supportive housing filling in at the edges.

The specialized tail rewards a slower look, because it is where the portfolio stops being generic and starts naming particular kinds of need. Past the senior and supportive-housing developments, the property-type breakdown shows a thin scattering of properties built for narrower populations, a pair flagged for veterans, a single one for young men in the eighteen-to-twenty-four range aging out of other systems, one tied to people experiencing both disability and homelessness, one for people with disabilities ^[1]. These are small numbers, one or two developments apiece, and we build nothing on them. They matter only because they show how a city's affordable stock actually accumulates, not as a single block of identical units but as a sediment of separate decisions, each program answering a particular constituency, layered over years into the inventory a corridor happens to hold when you finally go to count it. The 606 corridor's affordable housing was never designed as a response to the 606 at all, but is rather the residue of decades of unrelated commitments that happened to land in these four community areas.

A note on what the property-type list does and does not tell us, because the categories invite more reading than they can bear. The Affordable Requirements Ordinance share is large, 42 of 119 developments, and it is tempting to treat that as evidence of a particular policy response keyed to new market-rate construction along a hot corridor ^[1]. We resist that reading. The list records a property's type, not the year it entered the portfolio, not the depth of its income restriction, not when its affordability covenant expires. A development built under the ordinance in 2003 and one built in 2019 sit in the same row of the same category, indistinguishable on this file. So we report the mix as a mix and leave the chronology alone. What the composition establishes is narrow and real. The 2,391 units are mostly family rental and inclusionary set-asides, the everyday instruments of affordable supply, with a modest specialized tail. We do not grade that mix, because the file does not tell us how deep the affordability runs in any of these buildings, and we return to that silence in the limits.

Hold the meaning of this section to exactly what it is. We have counted the subsidized and affordable rental units that the public sector and its development partners placed in the four community areas the 606 crosses, and we found 119 developments and 2,391 units, weighted toward the western neighborhoods and toward Humboldt Park above all ^[1]. That much is simply true, and we state it without hedging. A real institutional response stands in the corridor, the visible answer to the pressure the researchers documented, and it is heaviest where the pressure is heaviest, at least at the level of the community area. That is the standing counterweight, the part of the public response you can walk to and point at, the buildings that were already there when the trail opened and the ones that have opened since.

There is a strong temptation at this point to say more than counting allows, and we are not going to give in to it. That 2,391 units stand in the corridor is a fact. That they concentrate in the hardest-pressed neighborhood is a fact. Whether 2,391 units is enough, whether the response was adequate to the scale of displacement pressure, whether the affordable stock offset what the rising market took away, none of that is a fact we possess, because none of it is a thing we measured. Adequacy is a judgment that requires a denominator we do not have, the full count of households at risk, and we declined to fabricate that denominator. So this section counts and stops. Whether the counterweight was big enough belongs to data we do not hold, and we leave it there rather than pretend.

The public money behind the counterweight

Behind a standing affordable building there is almost always a financing decision, and in Chicago that decision very often runs through tax increment financing or a redevelopment agreement, so the next thing to count is the subsidy itself. These are the city's workhorse subsidy tools, and they deserve a plain explanation for a reader who has heard the acronyms without ever seeing the mechanics. Tax increment financing, or TIF, lets the city carve out a district, freeze the property-tax base inside it at its current level, and capture the growth in tax revenue above that frozen line to spend on development within the same district rather than route it into the general budget. A redevelopment agreement, or RDA, is the contract through which the city commits that captured money, or other subsidy, to a specific project, often in exchange for obligations such as a count of affordable units. Together, TIF and RDA are the city's main levers for putting public dollars behind private development, including the affordable kind. The City's Department of Planning and Development keeps a public record of these projects, naming the developer, the approved subsidy, the total project cost, the committed affordable units, the ward, and the community area ^[1]. We took the same four-area corridor cut we used for the affordable list and counted the money.

Filtered to the four corridor community areas, that file holds 65 project records, and the first thing the data forced on us is a distinction we did not get to choose ^[1]. Of those 65 records, only 20 are residential projects that commit affordable units ^[1]. The other 45, which is 69.2 percent of the corridor records, carry no affordable-units value at all, because they are not housing but schools, infrastructure work, and commercial reimbursements, the ordinary non-residential business of a TIF district ^[1]. We do not fold those 45 into the housing story. We also do not treat their blank affordable-units field as a zero, which would be a different and misleading move, quietly diluting the corridor's per-project averages with items that were never meant to produce a single unit. We set the 45 non-residential records aside entirely and count only the 20 that actually commit affordable housing ^[1]. That exclusion is deliberate, and it is the honest way to read a field that is blank because the question did not apply. The 69.2 percent share is itself a fact to sit with, since it means most of what the city's redevelopment machinery did along this corridor was not housing at all ^[1].

So the headline subsidy figures, drawn only from those 20 residential projects. They commit a combined 1,068 affordable units ^[1]. Standing behind that commitment is $145,939,883 in approved public subsidy, set against $547,095,393 in total project cost ^[1]. Put the subsidy over the cost and public money accounts for 26.7 percent of the total invested in these projects, with private and other financing carrying the remaining roughly three-quarters ^[1]. That ratio is a useful piece of texture rather than a verdict on efficiency or adequacy. It says the public dollar in the corridor was a partial dollar, real and substantial without being the whole of any project's cost, buying affordable units inside larger deals rather than funding them outright, which is the ordinary shape of subsidized affordable development under TIF and RDA.

It helps to translate that 26.7 percent into what it means on the ground. For roughly every four dollars that went into building these twenty projects, the public put in one and private and other sources put in three. That is the design intent of the tools rather than an accident of these particular deals, public money used to make private money show up where it otherwise might not. A redevelopment agreement is meant to be the smallest public contribution that still closes the financing gap on a project the market would not build unsubsidized, and a subsidy share well under half is roughly what one would expect from instruments built to stretch a public dollar. We say all of this descriptively. Whether a quarter on the dollar is generous or stingy, whether it bought enough affordability or too little, whether the same public money might have produced more units deployed some other way, are exactly the efficiency questions this paper does not answer, because answering them would require a comparison we cannot make from two files that simply record what was approved.

The subsidy concentrates by community area in a way that closely tracks the affordable stock, and the breakdown rewards a careful read. Humboldt Park again leads, drawing roughly $74.1 million in approved subsidy across 10 projects committing 570 affordable units, more than half of the corridor's committed units sitting in the single western neighborhood ^[1]. Logan Square is next in dollars, about $54.6 million across 5 projects and 332 units ^[1]. West Town, despite holding a large share of the standing affordable stock, drew far less TIF and RDA money for new commitments, about $9.1 million across 3 projects and 109 units ^[1]. Avondale drew about $8.2 million across 2 projects and 57 units ^[1]. Add the four and the totals reconcile to the 20 projects, 1,068 units, and roughly $145.9 million already reported ^[1]. The money, like the stock, pooled in Humboldt Park and Logan Square, the corridor's more western and more pressured reaches.

The bars make plain that two neighborhoods absorb almost all of the corridor's affordable-housing subsidy. Humboldt Park and Logan Square together account for roughly $128.7 million of the $145.9 million approved, leaving West Town and Avondale to split the remainder ^[1]. Humboldt Park alone commits more than half of the corridor's subsidized units, 570 of 1,068, and its 10 projects are half of the residential project count ^[1]. Whatever else the public money did along this line, it was aimed first and hardest at the western neighborhood the literature flags as most exposed, and the dollar figures concentrate even more sharply than the building counts did.

Candor requires a word about how this count came to differ from the plan that preceded it, because the difference is itself a small lesson in doing this honestly. The pre-analysis sketch for this work carried a rough early guess, roughly 12 TIF projects committing on the order of 615 affordable units in the corridor, a back-of-the-envelope figure formed before anyone matched the records exactly ^[1]. That guess did not survive contact with the file. When we matched the records exactly to the four community-area names and counted, the real corridor held 20 residential projects committing 1,068 affordable units, not 12 and not 615 ^[1]. We did not split the difference, and we did not quietly keep the rounder, smaller pre-estimate because it had been written down first. We replaced the guess with the computed figures and report the computed figures, and we flag the revision rather than bury it ^[1]. A plan is a hypothesis. The data overruled it, and the data wins.

One caveat belongs here in passing, with its fuller treatment saved for the limits section. Our corridor membership rests on exact matches to each file's own community-area field, and a few TIF rows label themselves with comma-joined multiple areas, such as a record tagged for both East Garfield Park and Humboldt Park at once ^[1]. An exact match for "Humboldt Park" does not catch a cell that reads "East Garfield Park,Humboldt Park," so a handful of genuinely corridor-adjacent projects fall outside our net ^[1]. The practical consequence is that 20 residential projects is a conservative floor rather than a ceiling. The true count of corridor-touching TIF affordable projects is at least 20 and plausibly a little more, and we would rather undercount honestly than pad the total by guessing which multi-area rows to claim. The money we report is money we can stand behind, every dollar of it tied to a project that names one of the four corridor areas as its own.

How the response tracks the pressure, west of Western

There is a meridian running through this story, and it should be named before the map of public money is laid over the map of private revaluation. Western Avenue divides the corridor into its two halves, the wealthier eastern stretch and the lower-income, heavily Latino western stretch, and the entire empirical literature on the 606 turns on that divide. The DePaul hedonic work finds its large, significant price premiums west of Western and minimal, stable effects to the east ^[5]. The displacement release tracks its steep western-segment price climb on the same side of the same street ^[6]. If the public counterweight we have been counting is aimed where the pressure is greatest, it should pile up west of Western too. We can check that directly, because both city files carry coordinates, and so we did.

The standing affordable stock lands hard on the western side. Of the 2,391 corridor affordable units, 2,113 sit west of Western Avenue and 278 sit east of it, which puts 88.4 percent of the total on the western side ^[1]. By development count the split runs the same way, 93 of 119 developments west of the line and 26 to the east ^[1]. The pie below shows the share.

Almost nine in ten of the corridor's affordable units stand on the western, lower-income segment, the same segment where DePaul locates the steepest appreciation after the 2015 opening ^[1][6]. We are placing two findings side by side here, not deriving one from the other. The literature's price geography and our supply geography point at the same half of the same corridor, and the alignment is close. That is a description of where two different things are located. It is not a claim that one produced the other, and it is not a claim that the supply was a response calibrated to the appreciation, since we have measured neither the timing nor the price.

The subsidized projects concentrate even more completely than the standing stock. Every one of the 20 residential TIF and RDA projects sits west of Western Avenue, all 1,068 committed affordable units, a full 100 percent ^[1]. Not one of the corridor's subsidized residential projects landed on the eastern side of the line. A round number like 100 percent invites suspicion that it is an artifact of where the cut was drawn, so we checked the boundary case directly. The logic is simple. If a 100 percent figure were an accident of where we placed the line, the easternmost project would be sitting just barely west of it, close enough that a small shift in the boundary would flip it east and break the clean number. So we found the easternmost of the twenty and measured its distance from the cut. The John Pennycuff Apartments, the nearest of the subsidized projects to the line, sit at longitude -87.6897, which is still west of the -87.6877 meridian we used for the split ^[1]. That is not a hair's breadth. The project that comes closest to the boundary clears it with room to spare, which means the boundary could be nudged in either direction by a meaningful amount and the count would not change ^[1]. The 100 percent figure is a genuine concentration rather than a result balanced on the edge of an arbitrary boundary ^[1]. The subsidized money went west entirely.

Read together, the standing stock at 88.4 percent west and the subsidized projects at 100 percent west say the same thing twice. The public counterweight along this corridor, both the buildings already standing and the dollars committed through redevelopment agreements, is overwhelmingly placed on the western segment, the segment the researchers identify as most exposed to post-opening price pressure ^[1][5][6]. That is consistent with a public sector directing its affordable resources toward the lower-income, higher-pressure half of the corridor, and it is fair to say the geography of the response and the geography of the documented pressure overlap closely.

It would be a mistake to let that overlap say more than it can. Concentration of supply on the western segment is not evidence that the supply was sufficient, and it is not evidence that it offset displacement, because we have measured neither sufficiency nor offset. We have counted units and located them. We know that 2,113 affordable units and 1,068 committed units sit west of Western, and we know the literature documents severe price appreciation on that same side, but we do not know, and have not tried to compute, whether those units housed the people the appreciation pushed, or held rents down on any block, or arrived in time to matter. The supply is concentrated where the pressure is. Whether it was enough is a different question, and not one these two files can answer.

A method note belongs in plain sight rather than tucked into an appendix. The west-versus-east split uses a single fixed Western Avenue longitude of -87.6877, a straight north-south meridian, not the actual street centerline, which jogs east and west as it runs the length of the corridor ^[1]. A straight line is a simplification of a real street, and a few developments near the meridian could fall on the other side of the true centerline than they fall on our cut. We checked the consequential case, the easternmost subsidized project, and it clears the line with room to spare, so the headline 100 percent holds ^[1]. We develop the boundary question further where the other limits are gathered. The honest short version is that the meridian is an approximation, the result survives it for the figure that most invites doubt, and the reader should hold the precise shares as close approximations rather than surveyed truth.

A gap measured honestly, and what these counts leave out

Now the two halves of the ledger can sit on the same page. On one side the literature documents a western-segment median sale price climbing from $97,000 in 2012 to $462,000 in 2018, close to a quadrupling, with the same DePaul work warning that this appreciation ate into the unsubsidized two-to-four-flat stock that had quietly housed lower-income families for generations ^[6]. On the other side stand the public counts we have assembled, 2,391 listed affordable units and 1,068 TIF and RDA committed units in the four corridor community areas ^[1]. We place those next to each other deliberately, as a juxtaposition rather than a subtraction. The chart below shows the two public counts side by side.

The reader will want to add the two bars together, and we will not, for a reason that matters. The 2,391 listed units and the 1,068 committed units are not disjoint sets. Several of the buildings financed through TIF and redevelopment agreements also appear on the affordable rental list, the Rosa Parks Apartments and the Zapata Apartments among them, so a single building can be counted once in each total ^[1]. The overlap reflects no flaw in either file. It is simply the ordinary way affordable housing gets made. A redevelopment agreement is the financing that brings a building into being, and the affordable rental list is the registry of buildings that exist, so a development that was subsidized through an RDA and then opened its doors will naturally appear on both, once as a commitment and once as standing stock. Adding the columns yields 3,459, and that figure is an explicit upper bound, the most units the two files could describe if every entry were distinct, not a count of net new supply ^[1]. We lead with the two numbers apart and name 3,459 only to bound it from above, because summing overlapping sets into a headline would manufacture supply that does not exist. The committed units are, in part, the listed stock seen at an earlier stage of its life, not a separate population to be stacked on top of it.

What the gap framing can carry, then, is modest. We can say that the documented appreciation on the western segment was steep, that the DePaul work describes real erosion of the unsubsidized affordable stock under that pressure, and that the public affordable response we can count in the same four areas runs to a few thousand units at most, concentrated on the same western side ^[1][6]. We cannot say the response closed the gap, widened against it, or held even, because closing a gap is a quantity we have not measured. The juxtaposition shows a large documented price movement and a modest countable public response in the same geography. It does not weigh one against the other on a common scale, and we decline to pretend it does.

From here the honest course is to walk through what these two files cannot tell you, as a continuous accounting rather than a checklist of disclaimers. Start with what the affordable list omits. It records a property's name, location, and unit count, and nothing about the rent charged, the depth of the income restriction, or the date the affordability covenant expires ^[1]. So our 2,391 units measure subsidized stock, the number of doors, not affordability burden, not how far below market those rents sit, not how long any of it stays affordable ^[1]. A unit whose covenant lapses in 2027 and one restricted for forty more years count identically in our total. The file does not see duration or depth, and a count of units is therefore a count of units, not a measure of how much affordability the corridor actually holds or how long it will hold it.

That omission matters more than it might seem, because the literature's central worry is precisely about depth and duration. The DePaul work flags the erosion of unsubsidized two-to-four-flat housing, units affordable because the market has not yet reached them, not because any covenant holds them down ^[6]. The protection a covenanted unit offers against exactly that erosion depends entirely on how deep its restriction runs and how long it lasts, and those are the two things our file cannot see. A development restricted to deeply low incomes for forty years is a very different counterweight than one with a shallow restriction expiring next year, yet both register as the same single line in our tally. We can count the doors. We cannot tell you how firmly any of them are held open, or for how long, and a reader weighing our count against the documented pressure should keep that blindness firmly in view. The number is real. It is also thinner than it looks, because it is silent on the two dimensions that would tell you what the units are worth as protection.

Corridor membership carries its own limit, already flagged and now stated fully. Every count in this paper rests on an exact string match to each file's own community-area field for the four area names ^[1]. That rule is clean and reproducible, and it has a cost. TIF rows labeled with comma-joined multiple areas, an entry tagged to two neighborhoods such as a label reading East Garfield Park and Humboldt Park together, do not match a single area name and are dropped ^[1]. The consequence is that our residential TIF count of 20 projects is a conservative floor, a number that could only rise if those ambiguous rows were resolved and folded in, never fall ^[1]. We chose the floor over a padded total, and the reader should hold 20 as a minimum.

The geographic split is an approximation, as the previous section said and this one confirms. The west-versus-east division uses a fixed Western Avenue meridian at longitude -87.6877, a straight line standing in for a street that does not run perfectly straight ^[1]. We verified that the easternmost subsidized project still falls west of that line, so the 100 percent figure is not an artifact, but the cut remains a meridian and not the true centerline, and the precise shares should be read as close estimates ^[1].

The largest omission is the one we chose on purpose. This paper began with a plan to compute a census-tract trend in median sale price, gross rent, and owner-versus-renter share across the corridor from 2012 forward, the kind of descriptive series that would let a reader watch the numbers move year by year. We dropped that analysis in full. No American Community Survey table by tract and no Cook County sales file for these neighborhoods was ever shipped into the repository, and the one rent dataset present covers South Side ZIP codes that do not touch the corridor ^[1]. The rent series we do hold stops at a set of South Side ZIPs and never reaches the corridor's 60622, 60639, and 60647, so it is silent on exactly the blocks this paper is about ^[1]. To produce a local price or rent trend we would have had to fabricate the underlying data, and the house rule on this site is absolute that we do not. There is no honest halfway version of that analysis. A price trend assembled from data that does not cover the place is not a rough estimate. It is an invention with a chart attached, and a reader would have no way to tell it from the real thing, which is precisely why it cannot appear. So we declined, and we sourced every price and premium figure in this paper from the published literature instead, attributed in text, while limiting our own contribution to the units, dollars, and shares we could actually count ^[1].

That decision sets the firewall this whole account has tried to hold, and it should be stated once more in the plainest terms. We describe a standing public response and its geography. We do not prove what the trail did to anyone. Every appreciation figure here is borrowed from the cited researchers and marked as theirs, and every unit, dollar, and share is our own count from two City of Chicago files ^[1]. The 88.4 percent of affordable units west of Western and the steep western price rise the Institute reports are findings from two different efforts that happen to point at the same half of the same corridor ^[1][6]. They are not links in a causal chain we have forged, and we have built no model that would let us forge one. This is a descriptive ledger of what the public sector put on its side of the corridor, reported at real scale, and bounded by everything it cannot see.

The longer Chicago history the corridor sits on

A trail does not produce displacement on its own, and neither does any single amenity. It does so because of the ground it lands on, and the ground of the Northwest Side was prepared by a century of public and private decisions about who could live where and who could borrow what. This history is context, not new data, and none of it is something we measured. But reading the 606 without it makes the trail look like a cause when it is closer to a trigger, and that distinction shapes what one concludes about parks.

The deep structure runs under most of Chicago, the structure of redlining and the racialized geography of credit. When the federal Home Owners' Loan Corporation and, after it, the Federal Housing Administration drew their security maps in the 1930s and beyond, they coded neighborhoods by race and graded them for mortgage risk, marking Black and immigrant areas in red and steering federally backed credit away from them. Far from a passive description, the map was an instrument that channeled investment toward white neighborhoods and starved everywhere else, and its effects compounded across generations because housing equity is how American families build and pass on wealth. The result was a metropolitan map on which some neighborhoods accumulated value and others were locked out of the machine that produces it. That this was government policy, not market accident, is the central historical fact, and it is why an amenity arriving decades later lands on terrain already steeply unequal by design.

Denying credit to its segregated neighborhoods was only half of what Chicago did. It also built a parallel and predatory market to extract wealth from the families locked out of the normal one, and that chapter belongs in any honest account of how the city's land came to be priced as it is. Through the 1950s and into the 1960s, when Black and other excluded buyers could not get conventional mortgages, speculators sold them homes on contract, an arrangement in which the buyer made payments toward a deed that did not transfer until the very last payment and could be voided, with everything already paid forfeited, on a single missed installment. The contract sale stripped the protections of ownership while charging more than ownership should have cost, and it moved enormous sums from excluded families to speculators across the city's changing blocks. Alongside it ran blockbusting, in which the same speculators stoked white panic about racial change, bought low from fleeing white owners, and sold high to the Black and Latino families with nowhere else to go. Neither practice was a market failure. Both were engineered extractions, and they shaped the West and Northwest Sides as surely as the redlining maps did, so that the cost of housing for nonwhite Chicagoans had long been inflated by exploitation even as the official market called their neighborhoods worthless.

On Chicago's Northwest Side that history took a particular form, because the neighborhoods around the Bloomingdale Line were the receiving ground for one displacement before they became the staging ground for another. The Puerto Rican community that made Humboldt Park its center did not start there. It assembled there after being pushed out of earlier Chicago neighborhoods near the Near North Side and the lakefront, moved along by urban renewal, rising rents, and outright hostility through the 1950s and 1960s. The industry along the rail embankment, the tanneries and metal shops that clustered to take freight deliveries, made the surrounding blocks working-class housing for the people who worked those jobs, and as the industry left, the housing stayed cheap, and cheap housing in a major city is, eventually, an invitation. The community that anchored itself around Division Street, the stretch later marked by two great steel flags arching over the roadway, was the product of one removal already. The flags went up in the 1990s, as gentrification was advancing through Wicker Park and into West Town, an assertion of permanence by people who had reason to fear they might be moved again. The trail did not create the desirability of the Northwest Side. It capitalized a desirability the neighborhood's affordability and its position near the gentrified Near Northwest had been building for years, and it handed the resulting gain to whoever owned the land.

The geography of that change moved like a tide with a direction. It started near downtown and rolled northwest. Wicker Park and the eastern stretches of West Town turned over first, in the 1980s and 1990s, when artists and then professionals discovered the cheap, well-built housing near the Loop. Bucktown followed. The wave pushed steadily outward along the same corridors, raising prices block by block, so that by the 2010s the frontier of reinvestment had reached Logan Square and was pressing into Humboldt Park. The 606 opened just as that wave reached the western blocks. The trail did not summon the tide. It accelerated a wave that redlining, urban renewal, and three decades of northwest-rolling gentrification had already set in motion, and it accelerated it most where the wave was just arriving, at the affordable western end. That sequencing is why the literature's western-versus-eastern split owes nothing to coincidence in the trail's route. The trail met a neighborhood at the precise stage when an affordable, majority-Latino place was, by every prior pattern in the city, the next to turn.

The mechanism by which the abstract pressure becomes a changed streetscape deserves a concrete description, because it connects directly to the affordable stock this paper counts. The classic Northwest Side building is a two-flat or three-flat, a modest masonry structure with two or three rental units, often owner-occupied with the rent from the other units helping pay the mortgage. That building type is the backbone of naturally affordable housing in Chicago, affordable not because of any subsidy but because it is older, smaller, and plentiful, and it is exactly the stock the DePaul work describes eroding near the trail ^[6]. An investor buys the two-flat, declines to renew the tenants, guts it, and converts it into a single high-end house or two luxury condominiums, removing affordable units from the market and replacing them with stock priced for the new arrivals. Each conversion subtracts cheap rentals and adds expensive ownership, and the cumulative effect along a corridor is a wholesale change in who can afford to live there. The covenanted units this paper tallies are the deliberate, subsidized answer to that quiet attrition, which is part of why their depth and duration, invisible in our file, matter so much to whether they hold against it.

The city saw the mechanism clearly enough to design a tool against it. The demolition-fee idea floated for the 606 corridor was an attempt to make the conversion math less attractive by adding a cost to the demolition that usually precedes a teardown-and-flip ^[1]. Much of the corridor's affordable stock sits on lots zoned to permit two or three units, and the financial incentive after 2015 ran toward converting those multi-unit buildings into single luxury homes or a smaller number of high-end condominiums, reducing the unit count while raising the value. Deconversion and the slow loss of two-flats to single-family rehab are the fine-grained moves through which a corridor loses its naturally affordable housing, one parcel at a time, and they are largely invisible in any single year. That the city reached for a demolition fee tells you it understood the problem. That the tool remained limited tells you how hard it is to arrest the mechanism once the land market has set every owner's incentive against the public interest. We can name the tool because the policy record around the corridor names it; we cannot quantify its effect, and we do not try ^[1].

Set the 606 inside this history and its lesson sharpens. The trail is the latest event in a century-long story rather than its villain. Redlining built the unequal ground. The contract-sale market extracted wealth from the families penned onto it. Urban renewal moved the Puerto Rican community onto part of it. Deindustrialization left the housing cheap. The northwest-rolling tide of gentrification built demand right up to the neighborhood's edge. And then a beautifully designed public amenity landed on top of all of it. To blame the park is to mistake the trigger for the gun. To absolve it is to ignore that the people who study these projects had named the pattern in advance, in Atlanta and in the green-gentrification literature, before the 606's western prices ever moved ^[4][8]. The useful conclusion is neither that parks are bad nor that this one was innocent. It is that a public investment in an unequal city, made without a serious mechanism to keep nearby residents in place, predictably enriches owners and squeezes renters, and that the size of the public counterweight, which this paper has counted, is the measure of how seriously a city took that predictable problem.

There is one more piece of the Chicago machinery worth naming, because it sharpens what the modest counterweight reflects. The idea behind value capture, that the public should recover some of the land-value increase its own investments create and steer it toward chosen ends, is not foreign to this city. Chicago has been one of the most aggressive users of tax increment financing in the country, and TIF is value capture by another name. A district freezes its tax base and devotes the increment, the growth in tax revenue above that frozen base, to development inside the district. The 606 generated exactly the kind of land-value growth that machinery is built to capture. The tools existed. What was missing was not the machinery but the decision to point it, at a scale matched to the threat, at affordable housing for the residents the appreciation endangered. A city that can capture increment to subsidize commercial redevelopment can capture it to keep longtime renters housed. The 1,068 committed units and roughly $145.9 million this paper counts are what that machinery produced along the corridor ^[1]. Whether that was the most it could have produced, or the least, is a judgment the two files do not support and we do not make.

The shadow and the path, ten years on

Stand on the embankment now, a decade past that June morning in 2015, and the trail is simply part of the city. Strollers and bike commuters, runners at dusk, the planted slope greening over each spring above four neighborhoods that were never quite one neighborhood. The 2.7 miles of the Bloomingdale Trail are a genuine public good, well used and well loved, the disused rail line reborn as the thing its builders promised. Nothing this paper has counted takes any of that away. The path is real and it is good.

So is the shadow it casts, and the researchers have documented its shape with care. The ground the trail crosses was revalued after it opened, sharply and unevenly, with the steepest climb on the lower-income western stretch where the people most exposed to that climb already lived ^[6]. A trail and its shadow, the visible amenity above and the quieter revaluation of the land beneath and beside it, have run the length of this corridor together for ten years now. Holding both in view at once, the good path and the pressure it walks alongside, is the only honest way to see the 606.

The temptation, looking back over ten years, is to resolve the two into a single verdict, to decide the trail was either a gift or a betrayal and file it under one heading. The corridor does not cooperate with that tidiness, and neither does the evidence. A child who learned to ride a bike on the embankment and a family priced off the block beneath it are both true, and they are often true a few doors apart. The literature does not let us pretend the pressure was imagined, and our own count does not let us pretend the public did nothing in response. What the record will not supply is the single number that would tell us whether the response was equal to the pressure, and rather than manufacture that number we have left the two halves of the picture standing as they are, unreconciled, because that is the honest shape of what is known.

Our contribution to that seeing is small, and we will not inflate it. We counted the public counterweight in the four corridor community areas and found it real, 119 affordable developments and 2,391 units standing, 20 subsidized projects committing 1,068 more, concentrated, like the documented pressure, west of Western ^[1]. The response sits where the pressure sits, and it can be counted, which is worth something. But a count is not a measure of sufficiency, and we have nowhere claimed the response was sufficient, because the measure that would test it is precisely the measure we do not have.

An honest next step is therefore easy to name and hard to take. It would require the very data we declined to invent, a real Cook County sales series or a Census series by tract, shipped to the corridor and held to the same standard as everything else on this site, before anyone computes a local price effect or weighs the public response against the private revaluation on a common scale. That is the gap a future contribution could fill. We left it open rather than paper over it, and naming it plainly is part of the work.

For now the lasting image is the place itself, not the policy. A ribbon of green over an old freight line, running west through Humboldt Park and West Town and Logan Square and Avondale, carrying its walkers and its shadow together into a second decade. What we can offer beside it is a ledger, honest about what it counted and quiet about what it cannot say, a record of the modest public response standing on the western side of a line that has come to mean a great deal in this part of Chicago.

^[1]: Author analysis of Chicago Affordable Rental Housing Developments (606-corridor subset) plus TIF and RDA project records. City of Chicago (Department of Housing affordable rental developments list; Department of Planning and Development TIF and RDA projects), already mirrored in the Rooted Forward repository. Reproducible files at rooted-forward.org/research/data/the-606-trail-displacement.

^[2]: Rigolon, A., & Nemeth, J. (2018). "We're not in the business of housing": Environmental gentrification and the nonprofitization of green infrastructure projects. Cities, 81, 71-80.

^[3]: Rigolon, A., & Nemeth, J. (2020). Green gentrification or "just green enough": Do park location, size and function affect whether a place gentrifies or not? Urban Studies, 57(2), 402-420.

^[4]: Immergluck, D., & Balan, T. (2018). Sustainable for whom? Green urban development, environmental gentrification, and the Atlanta Beltline. Urban Geography, 39(4), 546-562.

^[5]: Institute for Housing Studies at DePaul University (2016). Measuring the Impact of The 606, Technical Paper, Hedonic House Price Models for Small Geographical Areas.

^[6]: Institute for Housing Studies at DePaul University (2020). Displacement Pressure in Context, Examining Recent Housing Market Changes Near The 606.

^[7]: Harris, B., Schmalz, D., Larson, L., Fernandez, M., & Griffin, S. (2020). Contested Spaces, Intimate Segregation and Environmental Gentrification on Chicago's 606 Trail. City & Community, 19(4), 933-962.

^[8]: Wolch, J. R., Byrne, J., & Newell, J. P. (2014). Urban green space, public health, and environmental justice, the challenge of making cities "just green enough". Landscape and Urban Planning, 125, 234-244.

^[9]: Curran, W., & Hamilton, T. (2012). Just green enough, contesting environmental gentrification in Greenpoint, Brooklyn. Local Environment, 17(9), 1027-1042.

^[10]: Rigolon, A., & Collins, T. (2023). The green gentrification cycle. Urban Studies, 60(4), 770-785.

^[11]: Lindsey, G., Qi, Y., Bhattacharya, T., & Loh, T. (2021). Neighborhood Change and Gentrification Near Three Urban Trails. Findings, October.