Gentrification & Socioeconomic Diversity

Lisa Brewster

Lei Ding et al. have posted a Federal Reserve Bank of Philadelphia Working Paper, Gentrification and Residential Mobility in Philadelphia, to SSRN. The abstract reads,

Gentrification has provoked considerable debate and controversy about its effects on neighborhoods and the people residing in them. This paper draws on a unique large-scale consumer credit database to examine the mobility patterns of residents in gentrifying neighborhoods in the city of Philadelphia from 2002 to 2014. We find significant heterogeneity in the effects of gentrification across neighborhoods and subpopulations. Residents in gentrifying neighborhoods have slightly higher mobility rates than those in nongentrifying neighborhoods, but they do not have a higher risk of moving to a lower-income neighborhood. Moreover, gentrification is associated with some positive changes in residents’ financial health as measured by individuals’ credit scores. However, when more vulnerable residents (low-score, longer-term residents, or residents without mortgages) move from gentrifying neighborhoods, they are more likely to move to lower-income neighborhoods and neighborhoods with lower values on quality-of-life indicators. The results reveal the nuances of mobility in gentrifying neighborhoods and demonstrate how the positive and negative consequences of gentrification are unevenly distributed.
I am not in a position to fully evaluate the methodology of this paper in this post. At first glance, however, it appears to be a well-constructed and large study, tracking “the residential location and financial health of a random sample of more than 50,000 adults.” (1) At the same time, useful household characteristics like income and race were not available to the authors, so the study has some significant limitations.

Given the intensive debates that gentrification engenders in NYC and elsewhere, it is still helpful that the authors offer up some facts and grounded interpretation of those facts. The authors specifically find that “gentrification is associated with positive changes in residents’ financial health: Residents in gentrifying neighborhoods experience an average increase of 11 points in their credit scores, compared with those who are not residents.” (1) At the same time, their results “suggest that less advantaged residents generally gained less from gentrification than others, and those who were unable to remain in a gentrifying neighborhood had negative residential and financial outcomes in the gentrification process.” (2)

Those who decry gentrification as well as those who promote it (quietly, more often than not) will find support for their positions in this paper. But those who are trying to understand just what we are talking about when we talk about gentrification and its effects will be left with a more textured understanding of how the demographics of gentrifying neighborhoods change. If cities are serious about promoting socioeconomic diversity, they must understand what is happening when neighborhoods are in flux.

Does Historic Preservation Destroy Affordable Housing?

Spencer Means

The Real Estate Board of New York released a report about Rent Regulated Units in Landmark Districts. The report opens,

This analysis was conducted to examine the frequent assertion that landmarking helps preserve existing affordable housing. It is based on data that recently became publicly available that provides a snapshot of the number of rent-stabilized units in 2007 and again in 2014.

Contrary to statements made by advocates, affordable housing is not preserved at higher levels in NYC’s historic districts. The data shows that properties located within New York City’s historic districts showed a greater net loss of rent regulated apartments than those located in non-landmarked parts of the City.


An analysis of the data found that, from 2007 to 2014, the decline in the number of rent regulated apartments located within New York City’s landmarked properties was four times higher than in non-landmarked parts of the City.

Citywide, landmarked properties showed a much greater decrease in the number of rent stabilized units (-22.5%) than non-landmarked properties (-5.1%). At the end of this seven year period, there was a net loss of nearly 10,000 rent-stabilized units in landmarked districts in the City.

The Manhattan and Brooklyn numbers are particularly startling. Manhattan landmarked properties lost 24.5% of their rent-stabilized units compared to a loss of 11.5% in nonlandmarked properties. And Brooklyn landmarked properties lost 27.1% of their rent-stabilized units compared to 3.4% in non-landmarked properties.

The historic districts that had the highest net loss of rent stabilized units were Greenwich Village (-1432 units) and the Upper West Side/Central Park West (-2730 units). Combined, these two historic districts showed a decrease of 30% in rent stabilized units during this seven-year period. (1, footnotes and references omitted)

This study has been criticized for conflating causation with correlation. I think the criticism is warranted. The relevant question appears to be whether landmarking causes an increase or decrease in the number of rent stabilized units. The REBNY study does nothing to demonstrate causation.

Intuitively, it would seem that residents of hot neighborhoods like Greenwich Village would both seek to keep out new, large developments (which landmarking would achieve) and see higher and higher rents over time (which would lead to a reduction in rent-regulated units through a variety of mechanisms). It is not obvious how landmarking itself would lead to a reduction in rent stabilized units.

It is a shame that the REBNY study is so flawed. It raises important questions, but just leaves us more confused than before. There are serious arguments that historic preservation reduces affordable housing overall. If REBNY wants to take a meaningful position in this debate, it should produce a serious study.

The (R)evolution of Single-Family Rental Securitization

Kroll Bond Rating Agency distributed its Single-Family Rental Securitization Methodology. Because this is a new asset class, it is interesting to watch how rating agency’s assess the risks inherent in it. And it will be interesting, of course, to evaluate down the road whether they got it right or not. The Methodology states that

Single-family Rental (SFR) securitizations are a new class of asset-backed securities with characteristics of both commercial mortgage-backed securities (CMBS) and residential mortgage-backed securities (RMBS). Like CMBS, the primary source of certificateholder distributions during the term of an SFR transaction are loan debt service payments that are generated by income producing real estate collateral. Also like CMBS, there is an element of balloon risk, as SFR loans do not fully amortize over their terms, and the repayment of ultimate principal on the certificates is dependent upon a successful refinance of the loan or loans that serve as trust collateral. However, there is a broader source of demand for the single-family homes underlying an SFR securitization, which can be sold into the vast market for owner-occupied homes, totaling approximately 79 million units. In the event that the pool of single-family homes backing an SFR securitization needs to be partially or entirely liquidated due to an event of default either during the loan’s term or at the loan’s maturity, the expected recovery from such a distressed sale of homes would be largely determined by the conditions in the larger market for single-family homes, which is a primary focus of RMBS analysis.

*     *     *

the SFR securitization market is currently characterized by large institutional sponsors that have engaged in purchasing and refurbishing large numbers of single-family homes in distressed markets over relatively short periods of time.

*     *     *

As this is an evolving asset class, we will modify or adjust our methodology to address new transaction features as they emerge. SFR securitizations to date have been collateralized by a single large loan that is in turn secured by mortgages on several thousand income producing single-family homes. While this methodology is designed for this structure, it is also applicable to securitizations secured by a few large loans. Structures featuring a larger number of loans to distinct borrowers, many of whom may be non-institutional in nature, pose additional credit considerations that are not addressed herein. (3)

This summary demonstrates that there are a lot of new characteristics for this asset-class that Kroll is trying to capture in its rating methodology. These include the hybrid nature of the security itself; the hybrid nature of the underlying collateral for the security; the innovative business model of institutional investors entering the single-family market in a big way; and the possible entry of new players in that market, such as non-institutional ones; and changes in the type of collateral underlying the securities.

The takeaway for readers: don’t mistake the apparent simplicity of a rating (AAA, Aaa) as a signal of the solidity of the reasoning that went into it. Ratings, particularly those for new types of securities, are constantly evolving. To think otherwise is to risk being left holding a bag filled with all of lemons that the market has to offer to unsuspecting investors.