Our Zoning is Racist: a multi-family housing story

On Saturday, June 6th, I participated in the Kirkwood Teachers of Color-organized, Black Lives Matter Peace Walk. Early in the week leading up to the event, it seemed like the walk would be a somewhat controversial and divisive event for a pretty quiet, wealthy suburb; one likely attended by no more than a couple hundred of people who sought to push that town out of its comfort zone. As the week wore on however, and protests across the country continued to gain steam, the expectations changed. Sometime between Thursday afternoon and Friday morning planning to participate in the walk shifted from a fringe political act to a litmus test for which side of history you stood on. What a few days earlier would have marked you a soft-core radical now positioned you in the ideological center.

There is a clear difference between marches and protests, even when both are “peaceful”, non-destructive. While marches demonstrate solidarity, protests have the goal of provoking a response. Sometimes that response is anger and violence on the part of those seeking to maintain the status quo: police use tear gas or beat protestors, or counter protestors take it upon themselves to try and intimidate those pushing for change. This violent response is to some degree the desired one on the part of the protestors as it reveals the extent of the malice that had previously been obscured and builds public support in favor of those pushing for change. None of this is to say that a Kirkwood protest would have been Birmingham in 1963, but it did seem like there was a chance that a vocal demonstration that closed some streets might piss some people off. But Kirkwood’s was a march (even if it did still close some streets). Marches work to demonstrate that public opinion already supports the change that is being called for. That the powers that be have fallen behind the eight ball and those not yet on board with the change need to figure it out because the train is leaving the station whether they like it or not. The only qualification for marches being effective in this form is that the demands for what it is that needs to be changed have to be clear. Otherwise everyone marches, they all feel good about themselves for signaling where they stand, and we all go home with the status quo still in place. Kirkwood’s march had more participants than I ever dreamed of; I’m still not sure many of them had a real sense of what needed to change in our own backyard. Most of the signs at the march called on us to do very little: some were value statements like “Black Lives Matter” and others called for pretty simple action like “Vote!”. I don’t want to discredit either of these things; I explicitly agree with both of those statements. But for these to be the only statements that are made disregards the role citizens of Kirkwood play in perpetuating racial oppression and the concrete steps we can take as a town of 27,000 to effect real change.

So what can we do? Well there’s some police reform we could probably undertake. Kirkwood’s police department has stated that all eight of the 8 Can’t Wait policies have already been implemented for a couple years, so thats good. Ending qualified immunity is, unfortunately, not really something we’re going to be able to tackle at this local of a level. We could probably introduce more stringent policies in the hiring and firing process, namely hiring more diverse officers, firing officers with multiple complaints filed against them and refusing to hire officers who have been fired from other departments (some of those policies may already be in place, I’m not sure). At the more radical end of the spectrum we could theoretically defund or abolish the police but I think the weight of the evidence still falls on the side of maintaining a police department. But racism obviously runs much deeper than just how the police department is ran. The single biggest step we could take to address the depths of this systematic racism as Kirkwood experiences and perpetuates it, is, simply enough, to build more homes, or rather, at an individual level, to allow more homes to be built.

Our zoning is explicitly designed to keep the total number of homes in Kirkwood limited by allowing, almost exclusively, single-family homes. That limited supply of homes makes buying those homes and living in Kirkwood more expensive. Kirkwood being more expensive to live in, in praxis, winds up keeping Black people out. So in the end, our zoning, regardless of our intentions, has the effect of keeping little Black boys and girls out of our schools and young Black people from working and shopping in our businesses and eating in our restaurants. These decisions we make to keep the number of homes in Kirkwood limited (the zoning rules we implement, the votes we cast and the opposition to development that we offer) are bad (and the situation on the ground has only gotten worse over the last few decades as your about to see), but before we get there I want to say this: We have real power, as individuals, to make better decisions on each of those fronts and to address some of the root causes of the racism we’ve seen bubble up to our collective consciousness recently.

This is the structural racism you’ve been reading about. Right here in our own town. All we have to do to make that change, to dismantle that structural racism, is to let new homes be built. It might not sound like a heroic action for us to undertake —unless you wanted to really go the extra mile and speak in favor of new homes it quite literally requires you to do nothing— but I promise you it is heroic. Simply not opposing more homes being built in Kirkwood, (not opposing them because of traffic, not opposing them because they’re too tall, not opposing them because they’re rentals, not opposing them because you disagree with the architecture, etc, etc) is likely the single biggest cause that you can undertake to bring about a more just world. It is likely more important, in terms of impact on Black people’s lives, than reading any book or marching down any street. But before we get to fixing it, first we have to understand the scope of the problem. Strap in.

Part 1: Mapping Our Multi-Family Housing

These days, if there’s a proposal to build multi-family housing, (everything from duplexes to condo and apartment complexes), it means automatically that there is going to be a fight in the approval process. Residents will show up to oppose it and city council member will have to be won over. Some projects get approved anyway, other get approved after being drastically reduced in scope and other projects still get pulled from the docket in the face of opposition or get rejected by the council outright. But if you’ve ever spent much time walking or driving around Kirkwood, you have definitely seen lots of random, usually older looking apartment and condo buildings scattered around. They’re usually brick and usually assembled in tight clusters of two or three story buildings. Clearly at some point, it wasn’t all that difficult to build new homes for families who wanted to live in Kirkwood. So when did we stop building these homes? When did we handcuff the market’s invisible hand? I did some work to find out.

“Data”

The first thing I did was to plot every single example of multi-family housing within Kirkwood’s boundaries could find. Everything from duplexes to the Station Plaza building was counted. I’m sure I missed one or two (I just found another apartment complex yesterday) but in general I looked really hard and found way more examples than I was expecting.

After finding these buildings I then used a map the Post-Dispatch published in 2014 to look up when each of these developments were constructed and labeled them by that year. (If buildings were built after 2014 I had to search around a little bit more but as we’re about to see, that unfortunately not very many).

Finally I divided each of these developments into layers by decade (each color representing a different decade) running the gambit from 60’s all the way to projects currently under construction in the 2020’s. Three developments can be dated even older than the 60’s (1936, ’47 and ’55 respectively) and these were included in a catch-all layer titled “older”.

The Map

What follows is the map I came up with. Click on the box-arrow symbol on the top left to play around with it, to toggle decades on and off, to switch from the generic map view to the satellite view to tell exactly where each building is, etc. This map will go under the maps tab of the website and I plan on continuing to add to it as (hopefully) new multi-family developments get added. The basic purpose of it though, is to show when we used to build homes and when we stopped.

Potential Shortfalls (skip if ya want)

This sort of ad-hoc assemblage of data certainly is not perfect. I looked hard and found the vats majority, but I certainly did not find every example of multi-family housing in Kirkwood. I particularly fell short in accounting for older mixed-use buildings in downtown Kirkwood. While I know people live above some of the shops and restaurants, it was difficult to figure out if those were single family residences or multi-family ones, or just offices, so I left them out. I also left out retirement homes. I’m not sure if it was the right call or not but I was looking at the housing supply that effected the price of living in Kirkwood and all the advantages that offers including our school district and proximity to jobs, things that don’t really effect the retired.

Another potential hole in the data is the fact that I counted the number of projects, not the number of units those projects contained (duplexes would be counted the same as 180 unit apartment complexes) and thus the data doesn’t do as good of a job measuring the actual housing capacity that was added year over year. This was largely because projects are way easier to find than the number of units they contain. I can look for multi-family structures on google maps, to find the number of units I’d almost certainly have to visit each one of those structures in person and hand count them. I don’t know if that is the worst thing in the world though. While it doesn’t measure Kirkwood’s housing capacity directly, it does measure how many individual multi-family housing projects the council was willing to approve and thus their attitudes on housing in general. Additionally, anecdotally, there doesn’t seem to be a clear pattern on multi-family housing projects getting larger or smaller in scope over the years and that the number of projects approved is a good proxy for figuring out the number of units that got built.

The final potential shortcoming of the data is that it undercounts older approvals. It is much less likely that buildings from the 60’s are still around today compared to those built in the 2000’s. As buildings age, they get replaced by new ones. It’s politically much easier to put new multi-family homes where old ones were before. Lots get re-zoned and the public is already used to it being there, the fake argument that “traffic will increase” is no longer viable. This means that the real work of adding homes to Kirkwood occurs whenever a lot is first converted to multi-family housing rather than whenever the latest example of building multi-family has occurred on that lot. In my data, 1984 would get credit for building an apartment complex even if that apartment complex was just replacing another one built in say 1944. The number of units added or lost in these replacements would be helpful information to have in evaluating whether those replacements were moving things in the right (more homes) or wrong (fewer homes) direction.

Part 2: When We Stopped Building Housing

When we look at this map and then convert it to an excel data set what we find is that despite a spike in the early 2000’s the amount of multi-family housing we’ve approved has been drastically curtailed. The obvious pushback to this assertion is that Kirkwood is simply following national booms and busts in housing, that the recent dearth of multi-family projects in Kirkwood is merely the legacy of the 2008 housing bubble and the slow recovery. I wanted to try to disprove this possibility as completely explaining our shift towards NIMBYism (even if it partially does so) as well as ground the observation in data so I attempted to compare the Kirkwood’s multi-family project approval to national housing environment writ-large.

Methodology (skip down to the chart and upshot if you don’t care)

To do this we take the data recorded in the map (how many multi-family housing structures were built in any given year) and record it on a spreadsheet. Because Kirkwood is small and there are only so many properties/opportunities available for development in any given year, I then took those numbers and converted it into four year rolling totals. Picking which four years —whether it should be the four previous year’s completions or the four subsequent year’s completions— was hard to decide. On the one hand previous developments make more developments less likely (we weren’t going to do a second Kirkwood Station the year after we approved the first one, people would have lost their minds). On the other hand there’s a delay between approval and construction so to credit city council approvals for the correct year you have to look at buildings that were completed in the years that follow (remember, my only data source is when buildings were completed). What I ended up deciding is that I would sum the projects from the preceding year, the year itself and two years after the fact. 1971’s figure, for example, counts 1970, 1971, 1972 and 1973 in its total. It’s not perfect but I think it might be about as well as we can do. It’s not like we’re doing a lot of three year long builds here.

Now to measure the national housing environment that I mentioned earlier, I took the seasonally adjusted Private Residential Fixed Investment annual rate (a measure of the investment in housing’s contribution to GDP on a year by year basis). Because those numbers are given in nominal terms (inflation isn’t accounted for) I then divided it by an annual deflator to convert the numbers to “real” figures. This number takes into account everything from putting an addition on your home to building a 60 floor tower overlooking Central Park but in general it gives us a pretty good idea as to whether or not people are spending money on homes.

The Upshot

From 1960 until 1985 Kirkwood built, on average 7.5 multi-family housing structures in each four year span. from 1985 until today, we build on average, 2.6 multi-family structures every four years, a third of that. The assertion that Kirkwood is beholden to national trends is obviously partially true (we see Kirkwood closely mirror the National steep decline in housing investment after 2008/9), but more can be gained in looking at how Kirkwood has differed from the national housing market. From at least 1960 until around 1990, Kirkwood’s pace of multi family housing construction has remained above the national trend line. After 1990, and for twenty seven of the thirty years since Kirkwood has fallen below that line. We just stopped building multi-family housing. And when most of the land in Kirkwood is already owned and built on, to stop building multi-family housing is to stop adding homes writ large. To do this as demand rises is to make Kirkwood a more expensive and whiter town.

Qualifications

While this method should account for wholesale booms and busts in national housing market (e.g. 2008), it fails to account for shifts in trends in the types of housing investments being made. The housing market can simultaneously be hot and still be shifting away from multi-family housing and towards investments in new and improved single-family housing. This is, at least anecdotally, very true to some extent in Kirkwood. If we had a graph of Kirkwood’s investment in housing like we do of national investment (the blue line), I have no doubt that it would have a strong upward trajectory despite the fact that apartment construction has fallen off. But I also don’t think this completely disqualifies the usefulness of the above graph. Sure perhaps Kirkwood is simply matching national trends rather than truly being an outlier, but 1) just because a trend is broad rather than local doesn’t make it any better or worse of a trend. It also doesn’t mean we have any less control regarding reversing course and pushing back against that trend. 2) If the housing market gets hot enough,multi-family housing eventually is the result. Even if you tear down all the single family homes and build new bigger single family homes in Kirkwood, that will only get you to a certain level of residential investment. Eventually, if demand is strong enough, especially locally where the amount of land is fixed and limited, you have to start stacking those homes on top of one another and smashing them together to reach higher levels of potential investments in developments. There’s a reason Manhattan doesn’t look like Ladue or N Taylor.

Why We Stopped

This evolution from openness towards multi-family development (YIMBYism) to full throated NIMBYism has no single origin. In our case it is likely a combination of the culmination of historical trends towards restrictive zoning and idiosyncratic factors specific to Kirkwood. On the idiosyncratic front, of course each cohort of Kirkwood city councilors may have different attitudes towards housing, especially since Kirkwood’s term limits ensure turnover. But even if there are a few idiosyncratic NIMBY councilors here and there, councilors are elected and govern in response to public sentiment. A larger viewpoint would clearly have to be at play to get them elected in the first place. At the national level, NIMBYism and the restrictive zoning that facilitated it peaked in the early 90’s as a way to continue the de facto segregation even after the laws that enforced segregation-writ-large were ended. As redlining (which was ended in 1968 with the Fair Housing Act, which itself was largely the result of MLK’s efforts in the Chicago Freedom Movement) and the resulting school segregation that redlining brought about (which was drastically curtailed with bussing which began in earnest with a Supreme Court go-ahead in 1971 and ran nationally until 1999 and locally until 2019) it was replaced in earnest by zoning that drastically limited the construction of more homes. Locally these trends were playing out explicitly. The St. Louis bussing program was the largest and longest running in the country. The desire of whites to insulate themselves from the “other” (i.e. Black people) gained further salience in Kirkwood with the annexation of Meachem Park in 1991 and the further integration of Kirkwood schools with the closing of the predominately if not exclusively black Turner School in Meachem Park that followed in the late 90’s. So Kirkwood started to change pretty fast. Our schools got less white, our electorate got less white our town got less white. From 1990 to 2000 our black population increased by 22.8% while our white population actually fell by 2.04%. And in anticipation of these changes and in response to them, we stopped allowing homes to be built. Stopped allowing them via our zoning, our votes and our public opposition. If people were going to move to Kirkwood, we were going to make sure they were the “right kind” of people.

This legacy is not a good one. The good news is it has a really simple solution. All we have to do is allow more homes to be built. Allow the Kirkwood Flats and the commerce bank proposal, allow duplexes on any lot zoned residential, reduce setbacks. Simply building more homes will increase supply, put downward pressure on price and, eventually, allow the demographics of Kirkwood to come into line with the demographics of the larger St. Louis area in which we are located. We each have tremendous power to correct a legacy of discrimination, literally all we have to do is not stand in the way.

Thanks for reading.