Housing in New York City (photo: Edwin J. Torres/Mayor's Office)
New York City has approved a requirement for racial-equity studies to inform the city’s land-use rezoning efforts. Public Advocate Jumaane Williams, one of the proponents of the requirement, has expressed that “the more we have tools like this one, the more we can sift through what’s just noise and what’s actually serious and needs to be addressed. People have told us that these rezonings are going to be for the betterment of the city but, collectively, they have sped up gentrification, and they’ve been a fundamental failure.”
This requirement comes in response to long-voiced concerns over rezoning-induced displacement during both the Bloomberg and de Blasio eras — concerns that have effectively derailed a significant component of Mayor de Blasio’s rezoning agenda.
Producing and preserving 200,000 affordable homes by the end of the administration and 300,000 by 2026 has been one of Mayor de Blasio’s signature issues. His administration has pointed to its success toward meeting these targets. Most of this plan is supported by government-assisted housing programs but one key component of the de Blasio housing agenda was a goal of 15 neighborhood-wide rezonings accompanied by a Mandatory Inclusionary Housing program (MIH) that requires between 20% and 30% (depending on the applicable income levels) of the units to be permanently income-restricted.
As a matter of policy, MIH is also attached to private applications that substantially increase residential zoning density. Only six of the administration’s proposed rezonings, concentrated in lower-income communities and communities of color, have been adopted. Two others, in the comparably higher-income and whiter neighborhoods of Gowanus and SoHo/Noho are currently in the public review process. The others are effectively dead.
For example, a proposed area-wide rezoning of Bushwick, released in 2019 after a planning process that sought a consensus between the city government and neighborhood residents, failed to garner adequate public or political support. The community came up with its own plan in 2018 and both City Council members that represent portions of the Bushwick community (Council Members Reynoso and Espinal) came out against the city’s version of the proposed rezoning.
Under the New York City Charter, the final vote on proposed zoning changes is up to the New York City Council and the Council has near-uniformly deferred to local Council members in approving, denying, or modifying proposed zoning changes in their districts. Council members often used their effective veto power to get the mayoral administration to include changes to a proposed rezoning or to achieve other commitments from the city that would benefit the affected communities. Without the potential of an agreement between the city and the local Council members, there was no path forward for the city’s proposed Bushwick zoning plan.
In the Bronx, Council Member Salamanca effectively ended another of the administration’s proposed 15 area-wide rezonings— a proposed Southern Boulevard rezoning – when he publicly came out against it. Salamanca cited a study by the Churches United for Fair Housing (CUFFH), which he said showed that “neighborhood rezonings are directly responsible for displacing thousands of residents of color in primarily low-income communities.”
The CUFFH study has received plenty of press coverage, which, for the most part, has accepted its findings as gospel. It provided intellectual support for the requirement for racial-impact studies that culminated in the City Council legislation. Requiring these studies makes perfect sense given the implication of the CUFFH study in a city that is predominantly Black and Hispanic undergoing significant local area demographic changes (more on this later).
The CUFFH study looked at two rezonings, Park Slope in 2003 and Greenpoint/Williamsburg in 2005, and, with regard to its examination of racial change, concluded that:
“While displacement trends existed before the zoning changes, they were significantly accelerated during this time, implying the zoning changes exacerbated previous displacement pressures rather than alleviating them.”
A quick read of the CUFFH study will show that it concludes that the data is “implying” the neighborhood rezonings worsened racial displacement but, in the all-too-common way of the world, this has been interpreted by the press and many readers as shown to be “directly responsible.” But does the study show or even imply that the rezonings are responsible for displacement?
In summing up its findings, the CUFFH report notes that what it refers to as the Latinx population (and the Census describes as Hispanic) declined by 15,000 between 2000 and 2015 in the Greenpoint and Williamsburg neighborhoods and a 5,000 person decline in the Black and Latinx population of Park Slope between 2000 and 2013, while the overall populations of both communities were increasing – by over 20,000 in Greenpoint and Williamsburg and over 6,000 in Park Slope.
Racial disparities have a long, and way too often, horrendous history in the United States beginning with slavery, one of America’s original sins that are, sadly, still expressed in New York City’s housing market. The racial change described by many as zoning-induced is real, as we will see later, but is it induced by zoning changes?
When the CUFFH study appears to provide intellectual heft to the notion that public zoning policies are exacerbating racial disparities, it is not surprising that those policies are being called into question and challenged.
The discussion that follows looks at Census data over several decades in both communities that were the subject of city-sponsored areawide rezonings and those that were not. The discussion that follows necessarily includes lots of numbers as it looks at four different demographic groups in multiple neighborhoods over several decades. For those that don’t want to wade through all those numbers, the short summary is that:
- Racial displacement and replacement is real in the communities studied in the CUFFH study and elsewhere;
- The data does not support a conclusion that zoning changes induce racial displacement and replacement;
- In Park Slope, there was greater racial change in the years prior to rezoning than in any period covered in the CUFFH study, and the white population declined numerically and as a percentage of the population since 2000;
- In Greenpoint/Williamsburg, the white share of the population remained constant even as thousands of units of market-rate housing were added;
- There were dramatic racial changes in other areas such as Bushwick that were not rezoned;
- Other areas that were rezoned, such as Bay Ridge, did not have similar racial change;
- There is at least one alternative explanation of the drivers of displacement and replacement that is supported by Census data; and
- Critics of rezonings have not considered where the households occupying new development facilitated by rezonings would live in the absence of such housing and what displacement and replacement impacts might result.
I’m a 46-year resident of North Park Slope and a former urban planner with the city’s Department of City Planning, and have seen the enormous economic and demographic changes that have altered this neighborhood over the years.
It has been my observation over the years that market-rate development follows demographic shifts toward more affluent residents rather than development leading the market.
After reading the CUFFH report, I decided to spend some time exploring demographic change in Park Slope. There is no precise date by which what is now called gentrification began in Park Slope. One potential starting date would be 60 years ago when the Park Slope Civic Council began its House Tour showcasing the neighborhood’s brownstones. Another might be 1963 when Evelyn and Everett Ortner bought their brownstone and began proselytizing about the neighborhood. It was certainly well under way in 1975 when I scraped together every dime I could and purchased my home from the couple that did the original renovation. No rezoning was needed to trigger gentrification.
The CUFFH study looked at three time periods and concluded that the faster rate of change in the second time period (in which the rezoning occurred) implied that the acceleration of change was caused by the rezoning. The CUFFH study examines three periods that it refers to as “before the rezoning” (1990-2000), “the decade the zoning was passed” (2000-2010), and “after the rezoning” (2010-2015).
By CUFFH’s own analysis , the Latinx population continued to decline but the decline was much, much smaller in the “after rezoning” period after two decades of steeper decline. During the decade in which the zoning was passed, the Black population moved in the opposite direction from the Latinx population, increasing slightly, followed by a very small decline in the “after rezoning” period. Essentially, the CUFFH-compiled data shows that within a short period after rezonings are adopted, the previously rapid decline of Black and Hispanic people comes to a virtual halt. This is exactly the opposite of the conclusion that the CUFFH study reaches. If these changes imply anything about rezoning and racial change it is that racial change does not closely track zoning changes.
I looked back another decade prior to the CUFFH study period and found that racial change was well underway even then and that some of that change was at least as fast as any found in the CUFFH study. For this exercise, I looked at all ten census tracts (shown outlined in red in Figure 1) that were entirely or mostly within the boundaries of the 2003 Park Slope rezoning (shown outlined in gold). It does not include census tracts in which the rezoned area accounts for half or less of the census tract. For the periods examined by CUFFH, the Census data for my study area roughly match those of the CUFFH study.
Figure 1: Park Slope Study Area
But in the prior decade (1980 and 1990), the white population was already growing as a share of the population, reaching, at 76%, its highest share of the area’s total population. In the recently-released 2020 Census, it stood at 70.8%. Seventeen years after the Park Slope rezoning in a neighborhood that could be a poster child for gentrification, the white share of the population was 5% lower than it had been in 1990.
In spite of an overall population gain of 2,758 over 30 years, the white population declined by 237. By 1990, the Hispanic population was already declining and experienced its most dramatic decline (30%) in the prior decade, well before the city contemplated any areawide zoning changes*.
The Black population declined by 11.5%. The Asian population, beginning from a small base, showed an absolute and percentage gain as it will in every tabulation of the city’s population up to and including the most recent Census.
*This author excepted, there was no speculation about future rezoning in the years of greatest racial change. In 1982, I approached the then-director of the Department of City Planning’s (DCP) Brooklyn office and suggested that the rising values in Park Slope could support or would soon support new construction and that the DCP should rezone Fourth Avenue to accommodate new development. The director did not believe the zoning was an obstacle to new development and no rezoning was proposed. In the very late 1980s, years before the Park Slope rezoning, construction began on a few market-rate apartment buildings, many or all of which ran into financial trouble during the city’s 1989-1993 recession.
The Department of City Planning (DCP) has conducted its own follow-up to the Greenpoint/Williamsburg rezonings of 2005 and 2009 comparing data from the 1990 and 2000 Censuses, the 2006-2010 ACS and the 2013-2017 ACS. The two ACS data points are roughly a before-and-after measure of the rezonings. The population changes in both rezoning areas do not support the conclusion of zoning-induced racial change.
In the area of the 2005 waterfront rezoning, the population jumped by over 41% following the rezoning, adding over 13,000 residents between the two ACS periods. Each population group increased in absolute numbers while many of their relative shares changed. The white population gained the most people but its share remained static with 62%. The share of Hispanics went down from 27% to 23% post-rezoning but, contrary to all the recent publicity, actually went up in absolute numbers. This is more striking because the Hispanic population declined in the two periods preceding the rezoning: between 1990 and 2000 and again between 2000 and the first ACS average for 2006-2010.
The Black population also increased post-rezoning with absolute numbers and its share of population rising slightly from 2% to 3%. The Black population had been relatively static in the two periods preceding the rezoning. The Asian population increased with each new Census total and it increased its share from 6% to 8% post rezoning. For each of these population groups, the increase in the post-rezoning period is considered statistically significant, which is to say that it is not attributable to chance.
In 2009, the city approved an areawide rezoning inland of the 2005 waterfront rezoning. This rezoning was largely about imposing contextual zoning regulations (which permit shorter and bulkier buildings as compared with tower-type buildings) although some areas within the rezoning were upzoned (that is, rezoned to permit greater density). In this contextual rezoning area, all four population groups registered numerical growth although only the Black population’s growth was statistically significant.
What is striking about this area is that the Black population had declined modestly in the two periods prior to the rezoning and the Hispanic population had declined dramatically in the period mostly preceding the rezoning. As in Park Slope, racial change existed in the period before the rezoning, accelerated during the decade in which the rezoning occurred, and stopped after the rezoning. The data does not imply rezoning accelerates racial turnover. Indeed, the data presented in the CUFFH study would also be consistent with a conclusion that, in the period following the rezoning, the pace of racial change slows once rezonings are able to lead to increases in the supply of housing. (Truth be told, there isn’t sufficient data in the CUFFH study to draw this, or any, conclusion.)
One need not be a data junkie to understand why rezonings that increase the allowable amount of building do not have the effect of creating the racial change ascribed to it.
The DCP reports that the housing supply in the study areas increased by 17,100 apartments (after accounting for 650 lost through demolition). Of these, about 2,100 were income-restricted (another 1,070 income-restricted apartments were preserved but those do not increase the housing supply). So, 15,000 market-rate units were added. Opponents of rezonings assert that rezonings generate change when, in reality, it is the demands for housing created by these and other extra households that create change.
Anyone looking at the effect of rezonings should ask what would have happened if the rezonings did not occur.
In the case of the Greenpoint/Williamsburg rezonings, there would be roughly 15,000 fewer market-rate apartments. Almost all of the households that today occupy these new apartments would remain in the housing marketplace. A small share might conclude that New York doesn’t offer what they want and choose to go elsewhere but most just join the rest of us in competing for the existing housing stock. More competitors for a static supply drive up prices in all the locations they seek to live resulting in higher housing prices in Greenpoint/Williamsburg and in other communities where these households may seek to live.
If there is a valid criticism relating zoning changes to racial displacement or displacement in general, one could certainly be that past zoning changes were too modest to promote adequate housing production. Another could be that the voluntary Inclusionary Housing Programs were not as aggressive as the more recently adopted Mandatory Inclusionary Housing Program.
Most areawide rezonings, dating back to the Koch era, were a mixture of upzonings, often on a modest number of underbuilt commercial streets, and downzonings or contextual rezonings where the permitted density was lowered or held constant. The downzonings and contextual rezonings were usually on the more numerous, mostly built-up, residential streets.
In 2010, NYU’s Furman Center for Real Estate and Urban Policy published a paper that looked at early Bloomberg–era rezonings. For the 188,000 lots that were rezoned between 2003 and 2007, almost 63% were subject to a contextual-only rezoning (no significant change in permitted density) while 23% were downzoned and 14% were upzoned.
The Furman Center calculated the net effect of these rezonings as increasing the city’s residential development capacity by 80,000 apartments but that represented an increase of only about 1.7% in the potential development capacity of the city. In short, the city’s timidity in creating additional zoning capacity minimizes the beneficial effects of creating opportunities to increase the supply of housing in the city. If, as an example, Greenpoint/Williamsburg, the Lower East Side, or other neighborhoods serving this more affluent and perhaps whiter demographic had been rezoned to create even more capacity, either by permitting increased density on more lots or by permitting modestly higher densities on upzoned lots, fewer people who could not find housing in Greenpoint/Williamsburg would have moved on to the next available neighborhood.
In the case of Greenpoint/Williamsburg, the next available neighborhood is Bushwick, which is just to the east and served by two of the three same subway lines as Williamsburg. With opposition from the community and both local City Council members, there has been no areawide rezoning in Bushwick so it is a good area to examine as a comparison for control purposes.
Between 2000 and 2010 Censuses, Bushwick’s total population increased by over 8,000. The Hispanic population rose by 3,500 even as the Hispanic share of the population declined, but the previously small white population more than tripled, increasing by over 6,500. In the most recent decade, the Latino population declined over 15,000 while the white population almost tripled again, increasing by over 14,000. In contrast to Park Slope, which was rezoned and saw a modest decline in the white population, Bushwick – without the city’s proposed rezoning – added close to 23,000 white residents. The Black population in Bushwick declined between the 2000 and 2010 Censuses by about 2,000 and by almost 5,000 in the last decade.
The driving force behind racial change in Bushwick is not a zoning change but the spread into the next community of the Greenpoint/Williamsburg more affluent and whiter demographic that cannot be accommodated there. Greenpoint/Williamsburg’s change was similarly triggered, in large part, by the overflow of this unaccommodated demographic on the Lower East Side. Similar demographic changes have pushed through brownstone Brooklyn from Brooklyn Heights in the 1950s, to Park Slope, Carroll Gardens, Cobble and Boerum Hills and Fort Greene, and later to Prospect Heights, Crown Heights, and Bedford Stuyvesant. The common denominator is a generally capped supply of housing in one neighborhood pushing up prices there and shifting some of the housing demand to nearby communities.
East New York
Another way to look at the effect of rezoning on neighborhood racial makeup is to look at communities rezoned that were not already undergoing significant gentrification pressure. The East New York rezoning, approved in April 2016, was the first of the de Blasio era neighborhood rezonings.
Comparing the 2010 and 2020 Censuses for the core** of the neighborhood shows population loss of 1,535 (5.6%). During this period, the Hispanic population grew by 3.7%; the Black population declined by 26.3%, and the much smaller Asian population grew by 71.5% but still accounted for only 4.2% of the total population. The even smaller white population grew by a dramatic 38.6% but still only accounts for 2.7% of the total population.
A 2020 paper from the Manhattan Institute on the de Blasio era Mandatory Inclusionary Housing (MIH) program provides an indication about what development-related changes may be happening in East New York. It found that 30 of the 36 MIH projects citywide on which construction had started or been completed are 100% affordable projects. All the projects that appear to be in East New York are 100% affordable, meaning that there will be no market-rate apartments.
Income distribution is not the same as racial makeup but there is a correlation between race and income. It would be helpful to know the racial makeup of those occupying apartments produced under the Inclusionary Housing Programs. Nevertheless, 100% affordable housing is hardly the stuff on which gentrification and racial change are likely to be built.
**Census tracts east of Pennsylvania Avenue that are entirely within the rezoning area.
Contextual zoning regulations were developed at the Department of City Planning in the 1980s in recognition that many neighborhoods in the city have distinct characters and that often residents have chosen to live in these communities, at least in part, because of those characteristics.
Contextual zoning provided an alternative to the 1961 regulations that sought to promote often context-busting tower-in-the-park developments. The values that inspired the creation of contextual zoning are still widely held today.
But contextual zoning was developed at a time when the city had over a million fewer residents than it does now and more than a million fewer jobs than it had pre-pandemic. Since many of the earlier contextual rezonings were prompted by a concern about randomly located out-of-scale buildings, and those concerns were expressed primarily in neighborhoods where the market was producing new buildings, the earlier contextual rezonings were disproportionately concentrated in relatively affluent areas. This has left a legacy of comparatively protective zoning in affluent areas that are centrally located and well-served by transit. The balance struck in the past between preserving neighborhood character and accommodating change needs to be reevaluated in light of current needs.
The odd thing about racial displacement is that, at least to date, it does not represent an expansion of the city’s white population. That population peaked as a percentage of the population decades ago and has been declining since. It stood at 63% of the population in 1970 when the concept of “white flight” was very much in vogue. The 1980 Census put the non-Hispanic white population at 51.9%, the last Census in which it was a majority. By 2010, the white share of the city’s population had declined to one-third and the 2020 Census shows a further decline, dipping to 30.8%, representing a modest numerical decrease of 3,000 (0.1%) from the 2010 count.***
Even as the city’s overall enumerated population has grown, the city’s Black population has been in modest decline since 2000, dropping by 26,000 by 2010 and another 84,000 in the last decade from 22.8% to 20.2% of the total. The Hispanic population increased by 52,000 between 2000 and 2010 and a further 154,000 by the 2020 Census, increasing its share of the population by about 1.5%. The fastest growing of the four major groups is Asians, fastest growing of these four groups is Asians who increased in 20 years by 595,000 and increasing its share of the population by 9.9% to 15.68%.****
So, if the white population is not growing where is the fuel for displacement coming from?
Given that the citywide Hispanic population is both larger and growing faster than the Black population is declining while the white population has been in long-term decline, the increase in the white population in some neighborhoods necessarily means that it has declined elsewhere. This change can be loosely described as a declining white population in the outer ring of the city, a growing white population in the center of the city, and a growing Black and Hispanic population in the middle being pushed out by the center and pushing out against the outer ring.
It can be seen clearly in locating the blue (increasing white population) and red (decreasing white population) areas in Figure 2. The notable exception is a large white area on the map in southeast Queens that was 98% nonwhite in 2010.
Figure 2: Growth in White Population Pushing into New Areas (Source: DCP)
These changes can be seen in a look at the outer ring community of Bay Ridge, which was the subject of an area-wide rezoning in 2005.
The overall population declined slightly between 2000 and 2020. If the CUFFH theory were valid, the white and Asian populations should have increased and the Hispanic and Black populations should have declined between the 2000 and the 2020 Censuses. The Asian population did grow, as it did throughout most of the city and as would be expected for a community next to Sunset Park, a heavily Asian community. But the Hispanic share of the population rose from 11.7% to 18.9% and the Black share of the population rose from 0.8% to 2.1%. Only the white population declined, going from 70.6% to 59.3%.
The CUFFH study looks at changes in racial composition and the presence or absence of a zoning change and, in its view, finds correlation between the two and concludes that this implies causation. It does not look at any other possible factors. This is like saying that increases in the sale of ice cream increase the sale of sunglasses (or the incidence of murder) since all these increase in the warmer months.
An Alternative Explanation
I would posit that the gain of a million residents and a million jobs explains much of the population trends of the last two decades.
For example, the L-train connection to the growing tech and media employment center in Chelsea supports the changing demographics of the Lower East Side, Greenpoint/Williamsburg, and Bushwick.
So has changing tastes among different demographic groups. In 2010, New York magazine published an article about the best New York City neighborhoods to live in. Brownstone Brooklyn, Dumbo, Greenpoint, and Williamsburg accounted for half of the top 20. It ranked Park Slope number 1. Personally, I view any such ranking as arbitrary but these efforts are an indicator of forces shaping where people seek to live. The Beyond Times Square website currently lists Greenpoint as the city’s hippest neighborhood followed by Bushwick, which it describes as “another up-and-coming hipster hotspot, teeming with art, music, food, drinks, and more.” Zoning does not draw a mention for either neighborhood.
This is not to argue that race does not play a role in gentrifying neighborhoods. It is not accidental that gentrification moved from the Lower East Side through Greenpoint/Williamsburg to Bushwick or from Brooklyn Heights through much of Brownstone Brooklyn to Bed-Stuy in that order and not in reverse even though less expensive housing was available in places like Bed-Stuy when Park Slope or Prospect Heights were gentrifying. Gentrification moved out from neighborhoods where the “gentry” would feel comfortable, whether that’s because they want to be close to amenity-rich neighborhoods or people similar to themselves.
Perhaps the current ground zero for the conflict between an expanding white demographic and a shrinking Black demographic encompasses the communities in and around Bedford-Stuyvesant, Crown Heights, Bushwick, and into East New York (roughly highlighted in the yellow ovals on Figures 3 and 4 below) where there has been a particularly striking racial turnover.
The younger white demographic typical of gentrifying neighborhoods is joined in parts of the area by a white, Ultra-orthodox Jewish community that has large families which, when the children reach adulthood, results in high levels of new household formation. Both groups have been expanding locally for at least the past two decades and created significant housing pressure on what have historically been communities of color.
Figure 3: Changes in White Non-Hispanic Population
Figure 4: Changes in Black Non-Hispanic Population (Source: DCP)
Although I do not have data on the racial makeup of occupants of new construction in these areas, based on the average incomes of the various demographic groups and the Census-documented demographic changes, it seems reasonable to assume, to at least some extent, the expanding white demographics have been accommodated by newly-built housing, some of which was made possible by both city-sponsored and privately-initiated rezonings and the state-facilitated Pacific Park development (formerly Atlantic Yards), while the remainder of the white demographic soaks up an increasing portion of the existing housing stock, replacing or displacing the prior, predominantly Black and, to some extent, Latino population. It is important to ask again, where would the households now occupying the new housing be if this new housing had not been accommodated.
While earlier generations of New Yorkers lived in a constantly growing, constantly redeveloping city, since the mid-twentieth century, more recent generations had become accustomed to a physically more static city. Through the ups and downs of half a century, the city’s population grew by just over 100,000 between the years 1950 and 2000. Nevertheless, even when the city’s population was declining in the 1970s – and in some areas housing was abandoned – permits were issued for more than 150,000 new dwellings and 22,000 more dwellings were added than lost.
New York City is now coming off 20 years of remarkable population growth. Since 2000, the population has grown by almost 800,000. This growth has created enormous pressures in the housing market citywide, but perhaps most acutely in the areas where the expanding gentrifying core has been pushing into communities historically home to largely Black and Latino populations.
The City Council has now mandated that zoning changes to accommodate the city’s housing needs must examine the anticipated racial consequences of these changes. This essay shows that racial change mostly takes place independent of zoning changes: sometimes there was more racial change prior to rezoning; sometimes there were racial changes in areas that were not rezoned; and sometimes areas that were rezoned did not have significant racial change.
What might be a prototype of the mandated racial impact studies, for the proposed Gowanus rezoning, has now been completed. A collaboration among the Fifth Avenue Committee, Columbia University Professor Lance Freeman, and City Council Land Use division staff, it is an extensive look at the historical and current racial divisions in housing – both citywide and locally — and how the proposed rezoning would alter them within the community.
I am not going to attempt a complete summary of the report but just touch on a few points it makes and one it does not. The report avoids debating whether new development helps to address or exacerbates the city’s housing problems. Instead, it observes that evidence of new development causing indirect displacement “is sparse” and that evidence that increased supply would lower prices locally “rather than regionally” is “murky.”
It concludes that, while in this specific market the risk of displacement by rising market-rate rents for Black and Latino families is low because most such households are already living in protected affordable housing, the market-rate housing would “perpetuate” existing racial patterns where high-income white households predominate. It also found that adding nearly 3,000 “new low-income affordable housing…within a high-cost, high-opportunity neighborhood, both in regards to the number of units and percentage of the overall projected development (35%), is unprecedented in recent decades in New York City” and that “the added population from the Gowanus Neighborhood Plan is likely to meaningfully reduce segregation in Community Board 6.”
The Council-mandated studies are not required to examine any racial impacts of not rezoning to accommodate more housing, so, unsurprisingly, the Gowanus Racial Equity Study did not consider what, if any, impact not adding over 5,000 market-rate units has on the overall housing market. Those households remain in the regional housing market, even if there were to be no Gowanus rezoning.
Where would they go? One of the basic arguments of this paper is, if Gowanus is not rezoned to accommodate more development, these households will create housing pressure in comparable markets, push up prices and threaten other unprotected lower-income households. The effects of the city’s tight housing market are felt more across existing buildings than new ones because so much more of the housing supply is in existing buildings. Everyone who is displaced or replaced was a resident of an existing building.
New York City needs more housing to accommodate its population and it needs zoning changes to facilitate housing production. It needs housing to be built across a wide variety of neighborhoods, even those that received relatively protective zoning when the city’s population was less than it is today. I don’t think it means the city must permit unlimited building, but, together with community residents, it should re-evaluate the current constraints on housing production.
While it needs housing throughout the city, it particularly needs it in the areas where the housing pressures are the greatest, which are often areas where a comparably affluent and predominantly white population is up against communities of color with lower incomes. Thanks to the de Blasio administration’s Mandatory Inclusionary Housing, every upzoning to permit more residential development comes with a requirement for 20-30% of the new apartments reserved as permanently affordable housing so that every new building — in gentrifying neighborhoods or not — can also contribute to maintaining diversity within communities in perpetuity. In some communities, the city has been able to increase the share of affordable housing by dedicating city-owned sites in or near the rezoning area.
Only a small share of the housing market can afford newly-constructed market-rate housing and the accompanying MIH units can never fully satisfy the need for affordable housing. Smart use of available publicly-owned land and public subsidies are essential but will still likely leave much undone. Rezoning is not a magic bullet that can solve the city’s myriad housing issues, but not rezoning is sure to make our current housing woes worse.
***The Bloomberg administration challenged the 2010 Census results, alleging undercounting in Brooklyn and Queens. The Census Bureau rejected the challenge and the population and housing unit counts for the City remained unchanged. I believe that it is reasonable to assume that some part of population changes reported in the 2020 Census represent shifts that occurred in the 2000-2010 period but were underreported in the 2010 Census.
****One common approach in examining racial change has been to lump the Asian population in with the white population to show gentrifying racial change in “high-opportunity” areas. There has certainly been a growing Asian population in the New York City areas where the white population is increasing. However, the Asian population growth is widespread with only a small share in the areas with an increasing white population.
The New York City Department of City Planning has created Neighborhood Tabulation Areas (NTAs) as a means of aggregating Census tracts into somewhat larger neighborhood areas for analytical purposes. The city is divided into 197 NTAs. Of these, the white population increased by 5% or more in 37 NTAs (18.8%) and by more than 500 people in 48 of them (24.4%). The white population decreased by more than 500 in 92 NTAs (46.8%) and by 5% or more in 62 of them (32.5%).
The Asian population increased by 500 in 136 NTAs (69%) and by 5% or more in 79 (40.1%) of them while decreasing in both number and percent in only 3 NTAs, all in and around Manhattan’s historic Chinatown. There is a high correlation between the growth of the Asian population in NTAs where there has been significant increase in the white population. By my count, 36 NTAs experienced an increase of both whites and Asians of 500 or more, but these 36 NTAs account for only 28.6% of the 136 NTAs where the Asian population increased by 500 or more people. Only 10 NTAs had an increase of 5% or more of both populations and these 10 account for only 16.2% of the 62 NTAs where the Asian population had this significant growth.
Sandy Hornick is a land use and zoning consultant and a former urban planner (Director of Strategic Planning) for the New York City Department of City Planning.