Uncovering Neighborhood-level Portfolios of Corporate

Single-Family Rental Holdings and Equity Loss

Nicholas Polimeni*
School of Computer Science & School of Public Policy
Georgia Institute of Technology
nicholas.polimeni@gatech.edu

Brian Y. An†
School of Public Policy & School of City and Regional Planning
Georgia Institute of Technology
yan74@gatech.edu (corresponding author)

March 5, 2024

Abstract

The rise of institutional single-family rental (SFR) investors has raised concerns about the impact
on American home equity and wealth, particularly in markets like Atlanta, Georgia, a prime target
for these businesses. This study introduces a straightforward and efficient methodology for
classifying investor size, overcoming challenges related to obscured corporate ownership
structures. Utilizing an extensive dataset of parcel and sale records from 2010 to 2022, we provide
an in-depth analysis of Atlanta’s single-family (SF) housing market, extending the temporal scope
of previous studies. We propose a novel framework to assess community-level financial equity
impacts of corporate SFR investors. To test our assumptions, we use hedonic regressions,
demonstrating that corporate investors purchase SF properties for less and sell for more, when
transacting with an individual buyer or seller, compared to individual-to-individual transactions.
Our key findings reveal that Atlanta lost $1.25B in financial equity between 2011 and 2021, with
predominantly African American neighborhoods bearing more than half of the total loss. The most
affected neighborhood suffered a loss proportional to nearly 4% of their total household income.
This research has critical implications for those seeking to understand and remedy past, present,
and future harms of corporate SFR investments on local housing equity.

Keywords

Housing, Single-Family Rental (SFR), REITs, Parcel Records, Real Estate Ownership, Equity

*
Nicholas Polimeni is a senior undergraduate student in computer science, and he is currently serving as a
manager of Urban Research Lab at Georgia Tech.
†
Brian Y. An is an assistant professor in the School of Public Policy and an adjunct assistant professor in
the School of City & Regional Planning, both at Georgia Tech. He also serves as the Director of Master of
Science in Public Policy program.
 Introduction

In recent years, academic interest in the financialization of rental housing has surged, particularly
with the entrance of large corporate firms into the single-family (SF) housing market. In
America, these firms tend to heavily buy homes in Sunbelt metropolitan areas (Immergluck
2018; Goodman et al. 2023; Seymour et al. 2023), because of the states’ weak tenant protections
in landlord-tenant legal regime and related policies (Fields and Vergerio 2022), rising home
values and economic growth, and weaker access to mortgage markets for nonwhites (An 2023).
Considering a recent demographic trend in which many American families are moving to
Southern metro urban areas (US Census Bureau 2022), the infiltration and growth of corporate
investment into the SF housing market yields a variety of implications for cities like Atlanta that
are geographically divided between non-Hispanic whites and minorities.

Century-long racially segregationist housing policies and urban development, such as racial
covenants (Jones-Correa 2000; Gotham 2000), redlining (Rothstein 2017; Orlando 2021), urban
renewal (Hyra 2012), and discriminatory mortgage lending practices (Taylor 2019) have fueled
geographical racial segregation and low homeownership rates for African American households
long before the Great Recession. But Wall street-backed investors’ growing control of single-
family rental (SFR) homes only widens existing racial gaps in homeownership and wealth. For
instance, recent research investigating metro Atlanta shows that large corporate investors
intensively targeted majority-Black neighborhoods with strong rental market potential (Charles
2020), shrinking Black families’ homeownership rates (An 2023). Similar findings are also
observed in other metro areas (Harrison et al. 2024). Furthermore, evidence illustrates that,
compared with publicly traded real estate investment trusts (REITs), private equity firms tend to
concentrate their SFR holdings in relatively lower-opportunity neighborhoods with larger shares
of Black residents nationwide (Seymour et al. 2023).

In the wake of institutional investors as a new asset class in the SFR market which has been
traditionally dominated by mom-and-pop landlords (Colburn et al. 2021; Mills et al. 2019;
Lambie-Hanson et al. 2022; Demers and Eisfeldt 2022), some scholars worry that the
financialization of SFR market by corporate actors is essentially another manifestation of a
century-long racial capitalism in which institutions and corporations may extract wealth from
Black and other minorities in the form of home equity and rent (Fields and Raymond 2021;
1
Charles 2020). That is, in the absence of corporate investors, individual homebuyers should have
been able to secure a home as an asset base for their own equity- and generational wealth-
building, a reason why practitioners and scholars need to be sensitive about geographically
concentrated SFR portfolios of global investment firms in majority minority neighborhoods.

Concerns are not just about reduced homeownership for families of color. Corporations,
especially institutional investors, have an asymmetric advantage in information about the real
estate market trends, and they are equipped with financial liquidity via securitization. As we
show in this paper, such resources and leverage allow them to purchase a bulk of properties at
once, cut down prices during transactions with home sellers, and outbid individual buyers who
typically have mortgage constraints. A similar dynamic may also favor corporations when they
sell properties to individuals. Moreover, when a significant portion of properties are owned by
the same corporate landlords, they may exert market power in rental price determination (Gurun
et al. 2022; Watson and Ziv 2021) particularly in cities without rent control. While institutional
scale in property ownership can drive professional and innovative property management, it may
simultaneously be a source of vulnerability in adequate property management (An et al. 2024) if
a portfolio exceeds its capacity to manage, or the primary business model pursues rent extraction
at the expense of responsive service provision.

These concerns have pushed legislators in the US Congress to consider regulatory approaches. In
2022, both a Senate hearing (S. HRG. 117-696) and a House hearing (H. HRG. 117-90) ignited
efforts for national legislation limiting institutional landlords. Two notable bills have arisen from
this effort, H.R.9246 “Stop Wall Street Landlords Act of 2022” and S.3402 “End Hedge Fund
Control of American Homes Act.” The former denies benefits like mortgage interest deductions
and financial support from government-sponsored enterprises to large corporate investors,
classified as those with over $100 million dollars of SFR assets in any given year. It also imposes
an excise tax on the transfer of homes owned by large corporate investors. The latter bill bans
corporations and individuals from owning over 100 homes via a $20,000 tax penalty per home,
excluding nonprofits, public housing agencies, and home builders.

Both approaches, however, fail to account for small and medium-sized corporate investors who
may have a non-negligible ownership concentration in certain neighborhoods. As both bills
impose a limit on ownership, if they or their variants pass in the future, corporate owners will

2
respond by selling a large portion of their portfolio, either redistributing land-lording activity
back to small and medium-sized local investors or creating a supply-shock in neighborhoods
where they hold a large portion of SFRs. Even if corporate landlords are forced to halt their
activity, much of the damage has already been done. After all, neither regulatory approach takes
significant action to remedy the past harms. This paper is motivated by the perspective that
policy efforts should focus on returning previously lost equity to communities, rather than solely
preventing such type of corporate land-lording activity.

Furthermore, all national and local efforts implicitly rely on the existence of consistent and
robust rental registries. Yet, the lack of such registries in many states and localities pose serious
data challenges that require analysts to use sophisticated techniques to identify parcel ownership.
Without such registries in place, it is likely that corporations will adopt new models of obscurity
and nested ownership that will be even more difficult to track.

To address these real world concerns while contributing to the emerging body of literature on
institutional ownership of SFR housing, in this paper, we propose both a 1) methodology that
identifies and tracks corporate landlords’ SFR ownership portfolio change for more than a decade
at any geographic scale, but with an emphasis on neighborhoods and 2) a framework to quantify
community’s equity loss by focusing on three major points of time in corporate SFR business:
asset purchase, holding, and selling.

After applying the methodology to Atlanta City and Fulton County’s tax assessment and real estate
sales data, we find several patterns for corporate SFR investment trends in the area. Notably,
Atlanta’s SF market has grown rapidly from 2010 to 2021, with corporate actors exhibiting a
sizable 24% of purchases and 35% of sales. Within majority-Black neighborhoods, however, 70%
of total SF transactions involve a corporate entity, compared to 30-40% for other neighborhoods.
Collectively, corporate investors have grown their ownership share of total Atlanta SFRs from 30%
in 2010 to nearly 50% in 2022. These trends contextualize the unique market factors that
corporations may exploit to further their SFR portfolios and profits from home transactions.

Next, we investigate whether corporate investors’ pricing behaviors are different from individuals
in SF property transactions. Our hedonic regression analysis reveals that, in the 2011-2022 Atlanta
SF housing markets, corporate buyers purchased SF properties at a 37.6% lower price than

3
individuals. However, when corporate entities sold those properties back to individuals after
running a SFR rental business, they charged a 7.4% higher price premium than individual sellers
even when the model controls for property-, transaction-, and neighborhood-level characteristics.

We then apply this framework to 95 of Atlanta’s neighborhood statistical areas (NSAs) within the
greater context of Fulton County. Our analysis uncovers $1.25B of equity lost from Atlanta’s
neighborhoods during 2010-2022, with majority-Black neighborhoods bearing $681M of that loss.
At the neighborhood level, communities that lost the largest share of their wealth— amounting to
5% of the total income their households generated— were concentrated in Atlanta’s lower-income
and majority-Black neighborhoods. Moreover, we find that SFR rental activity only accounts for
about half of total equity loss, while corporate home buying and selling make up the rest of private
profit gains or communities’ equity loss. Hence, compared to previous studies, we examine the
dimension of damage caused by SFR investment more comprehensively.

Importantly, this framework can be used to quantify corporate-driven SFR equity loss in any
community, holding its value to comparing such financial losses between neighborhoods. In light
of the growing concerns regarding the presence and impact of SFR corporate investors in the local
housing markets, our methodology and framework can shed light on strategies for place-based
interventions. Understanding when and by what mechanisms corporate investors extract most of
the community equity can inform national, state, and local policy discussions that seek to limit
SFR activity and its adverse impact on American communities.

Context, Related Literature, and Research Questions

In the years since the Great Recession, when housing prices dramatically fell, Wall Street-backed
investors began acquiring tens of thousands of single-family homes for conversion into rental
properties. As of 2022, large corporate investment firms that own at least 100 single-family rentals
operate nearly 600,000 of such properties nationwide (Goodman et al. 2023). Large investment
firms own a tiny fraction of SFR housing across the country, amounting to less than 4% of the total
SFR stock (Goodman et al. 2023). But geographically, they are highly concentrated in major
metropolitan areas in the Sunbelt, including Phoenix (AZ), Dallas and Houston (TX), Nashville
and Memphis (TN), Birmingham (AL), Atlanta (GA), Charlotte (NC), and Jacksonville and Tampa
(FL).
4
Among these hotspot areas for large corporate investors, metro Atlanta is outstanding. In 2022,
rental housing units made up 35% of all housing units in metro Atlanta, while 15% of all single-
family detached units in the area were rental homes (American Community Survey 2018-2022).
According to Goodman et al. (2023)’s estimation, institutional investors owned one third of all
SFR properties in metro Atlanta in 2022. The geographical concentration of institutionally owned
SFRs is even more salient at the neighborhood level within the metro area (Charles 2020). An’s
(2023) research, for instance, indicates that corporate investors purchased up to 76 percent of all
single-family home sales in suburban, predominantly African American metro Atlanta census
tracts. In this paper, we investigate the neighborhood-level impacts of corporate SFR investors.

As metro Atlanta has been ground zero for the Wall Street-backed investment firms’ SFR business
for more than a decade, the area has drawn attention from several researchers (Raymond et al.
2018; Immergluck 2018; Charles 2020; Smith and Liu 2020; Austin 2022; An 2023; Coven 2023;
Billings and Soliman 2023; Immergluck and Law 2014). These studies generally find potentially
negative impacts on the affected communities, such as a higher eviction filing rate (Raymond et
al. 2018), shrunken homeownership for families of color (An 2023; also see Billings and Soliman
2023), discounted home purchase price (Smith and Liu 2020), rising housing prices in the
surrounding neighborhoods (Coven 2023), and a dampened property tax revenue due to higher
rates of serial appeal filings and subsequent tax assessment adjustments (Austin 2022).

While documenting downstream effects of institutional investors, such as those on crimes and
evictions, is gaining traction in the current literature (e.g. Gurun et al. 2022; Billings and Soliman
2023; Raymond et al. 2018), the efforts to quantify their direct impact in the housing market have
been mostly limited to rents and housing prices of nearby properties (Coven 2023; Smith and Liu
2020; Mills et al. 2019; Lambie-Hanson et al. 2022; Garriga et al. 2023; Ganduri et al. 2023).
These metrics themselves, however, do not directly tell us how much financial gain or loss has
been made in communities affected by corporate investors.

To our knowledge, while the notion of corporate investor-driven equity loss is often acknowledged
among practitioners in the field, no scholarly work has devoted the efforts to examine the
phenomenon. Part of the challenge is that doing so requires a comprehensive understanding of
corporate SFR ownership portfolio at the neighborhood level, but most studies provide a snapshot
of corporate ownership from either transaction data or a limited number of years of tax assessment

5
data. A more thorough understanding of the corporate SFR portfolio requires examination of both
datasets over a longer period. Using both types of data, especially tax assessment, however, is
costly due to significantly increased amounts of data.

Furthermore, researchers often lack a standardized methodology to identify corporate SFR

investors and landlords when they do not have a luxury of accessing proprietary datasets with
detailed and cleaned owner information (cf. An et al. 2024). Therefore, the first question we seek
to address is how researchers and analysts can use publicly available real estate data, namely a
combination of county tax assessment and transactions to identify corporate landlords while
uncovering nested corporate ownership issues in SFR property domain.

Next and relatedly, while corporate investors entered the SF market near the end of the Great
Recession and they have continued their business activities until now, virtually no study has
examined the entire period of their market entry and sustained business. Studying a specific period
is understandable, given that each study has unique constraints in data availability, but it does not
help us quantify how much financial gain or loss has been cumulatively triggered by corporate
investment in any given neighborhood in urban areas. Hence, we ask how decade-long changes in
neighborhood-level portfolios of corporate SFR ownership and transactions can better help us
understand the trajectory of their financial impact on neighborhoods.

Speaking of financial impact, since homeownership serves as the key wealth building asset for
most Americans (Goodman 2003; Rappaport 2010), the conversion of housing stocks from owner-
occupied to renter-occupied by corporate actors is effectively the same as foregone ownership
opportunities for would-be homebuyers. When a bulk of single-family housing stock is acquired
and held as rentals by corporations, for those properties, individual homebuyers are losing a chance
for owner-occupancy, and subsequently an opportunity to accumulate their assets either via
property value appreciation (i.e. wealth effect) or owner’s equity with mortgage payments. Said
differently, for corporate-owned SFRs, the rental yield does not stay in local communities, but they
get distributed to global (i.e. non-local) shareholders of corporations. Due to this corporate activity,
a proportion of wealth is not captured, or more pointedly, is extracted from residents of that
community. Yet, existing research and policy reports have not presented adequate methodologies
to measure the extent of communities’ home asset loss. Hence the third research question regards
the ways in which we can measure a community’s equity loss driven by corporate SFR business.

6
The paper proceeds as follows. In the next section, we first lay out a framework for estimating
neighborhoods’ equity loss due to corporate SFR business. We then introduce our data and
methodologies for identifying neighborhoods’ corporate-held SFR portfolios while addressing the
nested corporate ownership issue. The next section provides an overview of corporate SFR
portfolio trends in the city of Atlanta over a 13-year period (2010-2022). Next, to validate the
underlying assumptions in our proposed framework for equity loss estimation, we empirically test,
using hedonic models (both fixed-effects panel least squares and mixed linear regressions),
whether corporate investors acquire properties for less, and sell for more, when transacting with
an individual buyer or seller, compared to individual-to-individual transactions. Finally, we
present our analysis of corporate SFR-driven equity loss for 95 neighborhoods in Atlanta. We
conclude the paper with a discussion on how our methodology and framework can advance
existing policy discourse and future policymaking.

A Framework for Estimating Equity Loss at the Neighborhood Level

A traditional concept of home equity theft refers to the cases where existing homeowners lose
possession of their home due to missed property tax payments. In some states, even arrears of just
a few dollars can cause tax lien foreclosure (Pacific Legal Foundation, n.d.). Our concept of
communities’ equity loss is different from those cases, as it pertains to the activities of SFR
corporate investors. We use the term “loss” in place of “theft” given our interest in any negative
impact corporate investors may bring to the communities where they hold a non-negligible share
of SFR portfolios. To clarify, we are not zeroing in on home equity, which in technical terms, is
defined as the difference between how much a home is worth and how much a homeowner owes
on the mortgage. Instead, we are focusing on the financial loss of the affected community with
respect to the rental property business.

A conventional thinking for measuring such financial gain or loss on communities involves
tracking property price appreciation, namely how much corporate investors have benefited from
appreciated home values due to their investment and broader economic forces in the
macroeconomy and local housing markets. If corporate-owned SFR properties were sold at a
certain point, another relevant metric could be the price differential between an initial purchase

7
price and a later selling price; this information would include both price appreciation and any
profits corporate investors top off from the sale. Likewise, analysts could examine if corporations
purchased properties at a lower (or discounted) price from individual home sellers, compared to
other non-corporate actors buying similar homes. Typically, these approaches require property
transaction records, and researchers need to be keen on which parties between the sellers and
buyers are the corporate actors and individuals. The distinction between transaction activities and
rental business is important as Eisfeldt and Demers (2022) find that net rental yields and house
price appreciation (i.e. capital gains) each contributed half total returns in the estimation of total
returns to SFRs in the United States during 1986-2014,

To capture all these dynamics, our framework focuses on three important points of time in
corporate-run SFR home business for each property: 1) entry (purchase), 2) holding for rental, and
3) exit (selling). As detailed below, our methodology allows us to track if sellers and buyers are
corporations (i.e. LLC, LLLP, LP, INC) or individuals. For purchases, we capture corporate
buying from individuals for rental business. For selling, we also capture corporate sales of rental
properties to individual home buyers. Lastly, for property holding as rentals, we track how long
each SFR property has been owned by corporate landlords who do not reside on those premises.

Figure 1 shows our framework for capturing purchasing, rental holding, and selling activities of
SFR corporate investors. Several points are worth discussing to apply this framework. First, for
both purchasing and selling activities, we need a baseline to which a purchase or selling price by
SFR corporate investors will be compared. We consider a property’s fair market value as the
baseline value. If a corporate buyer’s purchase price is lower than the fair market value assessed
by a county appraiser in the same year, we interpret this as a case where the corporate entities are
financially gaining or squeezing more profits from the transaction itself. If a corporate seller’s sale
price is higher than the fair market value, we also consider this as financial gain at the expense of
the community where the transaction occurs.

Two underlying logics guide this interpretation. First, if any renovation was done on the property
in the prior years, the fair market value, determined by county appraisers on an annual basis, should
reflect such an improvement. Second, it is based on the viewpoint that corporations have
transaction leverages over individuals with their institutionalized power and resources. As such,
they could pay less on the purchased properties than individual buyers would (Smith and Liu 2020;

8
Brunson 2020) and sell for more. However, corporate selling of SF properties at higher prices has
not been examined by prior studies. Hence, we later seek to validate this assumption with a hedonic
regression, demonstrating that corporate investors pay less when they purchase SF homes from
individuals and that they later sell those properties to individuals for more, both compared to
individual-to-individual transactions after controlling for property, transaction, and neighborhood-
level characteristics.

While the loss triggered by property purchasing and selling activities is straightforward, the
procedure for property holding is not. It requires an understanding of annual/monthly rental value
of the SFR properties held by corporate landlords. When determining how much rent to charge a
tenant, many landlords use the 1% rule — which suggests charging 1% of the home’s value for a
monthly rent (Zillow 2021). According to this formula, a home valued at $300,000, for example,
would rent for $3,000 per month. However, the 1% rule does not hold in many places, and a more
conservative percent after considering the operating expenses—which typically ranges from 35 to
80 percent of the gross operating income (Zillow 2012)—would be 0.5%, meaning the net
operating income would be $1,500 per month for a home valued at $300,000.

We use the latter number (i.e. 0.5%) to estimate the annual net operating income of SFR held by
corporate landlords. Notably, in doing so, we track the number of years these corporate-owned
properties are held as rentals and use their annual assessed value (i.e. fair market value) as their
home values. In this way, we can partly account for any profits accrued to corporate landlords for
a price appreciation effect, though it should be a conservative approach since the fair market value
tends to lag a transaction price in the housing market. We do not consider additional price
appreciation from the purchase and sale of the home; this framework intends to quantify the unique
opportunity for wealth extraction that corporations have rather than general market trends that may
also benefit individuals in the community. Later, after calculating all three major components of
corporate SFR investor-driven equity loss, we aggregate all of them to neighborhoods for each
year.

Figure 1. Three Points of Time Capturing Corporate Profits in SFR business

9
 Data

Parcel Ownership and Property Sales Data

Tax assessment (also called tax parcel) and property sales data were obtained from the Fulton
County Board of Assessors. The tax parcel data covers all parcels in Fulton County from 2010 to
2022, and provides various information, including parcel identification (Parcel ID) number,
parcel address, land use code, property class, annual appraised value, fair market value, owner
4

name, and owner mailing address. The property sales data provides information related to
transactions, including Parcel ID, parcel address, land use code, property class, sale price, deed
type, transaction date, grantor name (i.e. seller), and grantee name (buyer). The sales data we
obtained covers 2011 through 2022 (equivalent to sales occurring in calendar year 2010 to 2021).
Both datasets have a unique key consisting jointly of Parcel ID and year; we merge on these two
common variables. 5

Our procedure of cleaning both datasets is presented in the Online Appendix. To briefly
summarize them, for tax parcel data, we drop non-SF properties, duplicate records, non-
geocodable ones, and keep only the most recently updated structure information on each parcel
when a parcel has several historical records for structure additions. After cleaning, we retain
2,739,296 SF parcel data of the initial 2,785,662 SF data entries (98.3%). For sales data, we
apply the same procedure, and additionally we drop non-arm’s length transactions and keep
unique transactions with the latest buyer and initial seller information when there are multiple
records for the same sales. For multi-parcel transactions, we adjust the sale price to reflect the
average price per parcel. Lastly, we drop government and bank sales. After applying all these

10
steps, our working sale data size is 150,284 from the initial 275,353. Further data cleaning details
are reported in Online Appendix A.

It should be noted that, throughout the analysis, we classify the sales data into two general
classes: typical market sales and atypical sales. Typical market sales are those which have been
classified as “valid” by Fulton County’s sale validity classification. Atypical sales include other
sales, most notably, foreclosure sales, multi-parcel sales, and remodeled after sale.

Atlanta Neighborhoods

The geometry of Atlanta’s Neighborhood Statistical Areas (NSAs), along with American
Community Survey (ACS) 2012-2021 data aggregated at the NSA level, was obtained from
Neighborhood Nexus. The NSAs are built from census blocks to match up with true
neighborhood boundaries as closely as possible using Census geography. For details about
NSAs, please see the Online Appendix A.

Because our data is from Fulton County, a small portion (6.5% in land area) of Atlanta, which
lies east of Fulton in DeKalb County, is excluded. The excluded portion contains eight of
Atlanta’s 103 NSAs, resulting in 95 NSAs as neighborhoods of interest in our study frame.
Precise parcel boundaries were sourced from the publicly available 2022 parcel map for Fulton
County. To determine which parcels are within the city of Atlanta, we join each parcel in the
historical tax data to its associated boundary from the 2022 parcel map using Parcel ID.

Unraveling Corporate Ownership in SFR Housing

Existing Challenges

Institutional investors often use different business owner names and, occasionally, multiple
mailing addresses for SF properties they acquire and operate. Besides, different business
structures such as LLC, LP, LLP, LLLP, and INC add another layer of complexity for tracking
the parent companies of various subsidiaries, often including shell companies registered as
separate property owners. However, in most states, including Georgia, there is no rental registry
mandate. As a result, researchers and analysts must bear significant burdens to navigate through
a county’s tax assessment data and a state’s business registry; what is worse, in most cases, the

11
formats of tax parcel data are not standardized between counties. Moreover, while publicly
traded REITs must report their business activities to the Securities and Exchange Commission
(SEC), including a list of all subsidiary companies under which they hold SFR portfolios, private
equity firms and private REITs are exempt from such SEC registration requirements. Regardless
of whether they are publicly traded REITs or not, the corporations’ business strategies for using
different names and mailing addresses greatly vary (Seymour et al. 2024; An et al. 2024),
rendering the tracking of true ownership exceedingly difficult by traditional approaches.

After developing a procedure to unravel ownership to a sufficient degree, we create two different
metrics— ownership and transaction scale—which will be used in the assessment of the Atlanta
SFR market over a 13-year period. Both metrics aim to characterize the concentration and
market power of corporate entities in local SF housing markets. They are used in the contexts
they best describe; for parcel ownership analysis, ownership scale is used, whereas transaction
scale is used for market and transaction analysis.

Approach for Parcel Data and Ownership Scale

While methodologies have been developed that address some of the challenges mentioned above
(e.g. An et al. 2024), our methodology uses a simplified approach that is relatively accurate with
low potential for false matches. We rely on a modified version of the owner address to group the
same owners together. Owner address has a lower potential for false positive matches than owner
name while also being easier to standardize and correctly match. Moreover, owner name has
greater potential for inconsistency between subsidiary corporations, but many subsidiaries or
shell SFR corporations use the same address. In this way, we account for many cases of nested
ownership structure, where different holding corporations may own properties, but report the
same business address for records purposes. We favor this approach as it narrows focus to our
research question without opaque or complex methodological underpinnings. To clarify, this
approach is best suited for quickly producing accurate ownership or transaction scale measures.
However, it is not optimized for robustly characterizing the portfolios of specific owners, as
discussed in the following sections. Practitioners and researchers may find this method more
approachable or feasible for their requirements.

Constructing an Address Key

12
For each parcel, we construct an owner address key by concatenating the address number,
address string (street name or PO Box), and zip code. Because the address string excludes
postfixes, we can avoid some missed matches due to inconsistencies between labeling, such as
“STREET" vs. “ST”. Before creating the key, we eliminate various other inconsistencies by
removing periods, commas, multiple spaces, and uppercasing all characters. Finally, to resolve
inconsistencies for PO boxes, we can take any address string with at least one number, extract
the numbers, and postpend them to the string literal “PO BOX.” In our Fulton County SF parcel
data, after completing this procedure, all PO Boxes were standardized except 37 (of 63K)
malformed address strings which were dropped accordingly. Table 1 showcases examples of
address key generation.

Table 1. Examples of Owner Address Key Procedure

Address Number Address String Zip Code Address Key

52 CREEKSIDE PARK 30022 52 CREEKSIDE PARK 30022

NA (filled with 0) P.O. BOX 370049 30037 0 PO BOX 370049 30037

NA (filled with 0) P O Box 370049 30037 0 PO BOX 370049 30037

Ownership Scale

Utilizing our constructed address key, we aggregate owners on this column to understand the
scale each owner operates on, and the names associated with their operations, each year. In Table
2 shows the five largest SFR corporate landlords in Fulton County in 2020 and compares our
calculations to An et al. (2024)’s OpenRefine methodology and their additional extensive manual
review approach that includes business registry data. As our method is not intended to
automatically describe specific corporate portfolios, one simple additional step is taken for an
accurate comparison. For each investor, we query for address keys where any of the associated
names contain the investor’s known names (ex: “PROGRESS” for Progress Residential), and
aggregate ownership totals for each record with 50+ properties. Manual inspection also

13
confirmed that records meeting these criteria were related or subsidiary corporations. Query
keywords, associated names, and addresses matched by our method can be found in the Online
Appendix C. These records may be useful for other academics or practitioners compiling names
of SFR subsidiaries. Because these different-address subsidiaries would have been classified as
large investors regardless, this comparison remains an accurate demonstration of our method's
ability to determine ownership scale, as well as to quantify specific corporate portfolios if
necessary.

Overall, our methodology captures significantly more corporate subsidiaries compared to An et

al. (2024)’s baseline. Compared with their extensive manual review approach, which they argue
represents a complete picture of ownership, we capture somewhat fewer properties, as expected
for an automated approach. Therefore, while this method was developed with the intention of
offering a simple but robust way to categorize owner scale, it may also be leveraged to describe
the portfolios of specific corporations. Most importantly, our methodology has the unique
advantage of classifying ownership scale for small (under 10 properties), medium (10-49
properties), and large (50+ properties) landlords, whereas existing methods focusing on owner
names are limited to identifying larger actors. Hence, we can paint a neighborhood’ SFR
portfolio of corporations at any scale, not just institutional, global firms.

Table 2. SFR Ownership Scale of Top 7 Corporate Landlords in Fulton County in 2020

Name Parcels Owned in OpenRefine Net Diff. OR method + Net Diff.

Fulton (Address Key (OR) method manual review
Query) (An et al. 2024) (An et al. 2024)

Amherst
720 103 +617 750 -30
Residential

Invitation
677 524 +153 719 -42
Homes

Progress
619 457 +159 760 -141
Residential

Sylvan Realty
422 250 +172 433 -11
(RNTR)

Tricon
219 222 -3 280 -61
Residential

14
Cerberus
256 340 -86 349 -93
Capital

Starwood
122 361 -239 450 -328
Capital

Approach for Sales Data and Transaction Scale

As most county’s sales data are structured similarly, within our Fulton County sales original
records, we do not have address information for either the grantee (buyer) or grantor (seller). In
these cases, we choose to use a conditional-based methodology to match the grantee and grantor
to their associated records in parcel data, which contains the cleaned address key used for
aggregation in the previous step. It is important to note that the grantee (buyer) is the owner of
the parcel in the tax year of the sale (tax year is one year after the sale calendar year), while the
grantor (seller) is the owner of the parcel in the tax year preceding the sale.

There are three potential methods to create a match, each of which has less specificity than the
previous. First, we attempt to match the grantee and grantor name to their associated parcel
records exactly. For this first pass, we merge sales and parcel data on parcel ID and sale/tax year
with the addition of grantee or grantor name to the merge key. For grantee matches, we use the
tax year columns from both datasets for matching, whereas for grantor matches, we use sale year
from sales data and match with the tax year from parcel data. This logic matches how ownership
is recorded in the parcel data after a sale. We label this match as exact match.

For the next pass, we complete the same merge but only on parcel ID and sale/tax year keys. For
grantee, this is the parcel record from the current tax year, for grantor, this is the parcel record
from the previous tax year (e.g. sale year). However, this matching is only effective when the
parcel has only one transaction in a given year. If a parcel was transacted upon multiple times in
that year, then we cannot tell which grantor or grantee will be reflected in the parcel data as only
one owner name is retained (in most cases, this is the last recorded owner for that tax year).
Therefore, we only take such matches where the parcel was involved in only one sale in the
given tax year. We label this pass as single sale only.

15
In the final pass, we separately take the grantee and grantor name and attempt to match it with
any exact owner name match in the parcel record. If there are multiple matches, we keep the last
matching row to reflect the owner’s most recent address. We label this final pass as exact name
only. Figure 2 illustrates all three matching methods: exact matching, single sale only matching,
and exact name matching.

Figure 2. Transaction Entity Matching Procedure

After completing all passes, we take the best match for each grantee and grantor name. We
consider the best match as the first pass that found an associated entry. Online Appendix B
tabulates the success of each pass. After completion of all three passes, 96% of grantees and 88%
of grantors were matched to an owner address from the parcel data. To ensure a high degree of
accuracy, we took a random sample of 100 records and manually confirmed that their matched
grantee and grantor information was accurate. Every record with manual verification produced
the desired, correct result or, in a small subset, no match. Matching grantor information was
noticeably less successful than matching grantee information. It is hard to speculate with
confidence on the exact cause of this divergence; some patterns that may produce this result
include sellers who have held their property for many years and record their name differently
now than for records in years prior, or that there are more cases of multiple sellers (for example,
divorces) than multiple buyers in one transaction.

After acquiring grantee and grantor addresses from parcel data, we use the matched address key
to aggregate transactions for each sale year. This produces a tabulation of the entities and a count

16
of their SF transactions in a given year. This metric can shed light on the concentration of
potential market power a given firm could have for transactions. All sales, rather than just typical
market sales, are aggregated to create the transaction scale table as any transaction, including
atypical, should increase the firm’s overall influence in the SFR market in the sale year. For the
same reason, when calculating transaction scale, we consider sale records where the grantor were
government entities or banks, despite excluding them from the dataset we use for further
analysis.

Assessment of Fulton County’s Single-Family Housing Market

In this section, we first unveil a trend of corporate SFR activities in Atlanta. We find that
Atlanta’s SF market grew substantially from 2010 to 2021. Overall corporate transactions
remained steady despite significant yearly fluctuations, with corporate SF sales consistently
outpacing corporate purchases. Yet, small and large corporations grew their share of the total
SFR housing stock in Atlanta. We then zoom into neighborhood-level corporate SFR portfolios,
contextualizing our study area, and demonstrating a disparate pattern of corporate activity in
Atlanta’s predominantly Black neighborhoods to the south and east.

Proportion of Purchases and Sales by Investors

Figure 3 shows that corporate sales have consistently outpaced corporate purchases on a yearly
basis in Atlanta. While this may run counter to a conventional thinking that corporate
homebuying characterizes the SFR market, the pattern may be explained by the temporal nature
of sales; a corporation can sell a property that it has bought in any previous year but can only buy
a property in the current year. In any case, it illustrates that looking at both sides of transactions,
not just buying but also selling, is important to understand the broader picture. Manual inspection
revealed that corporations also frequently transfer their properties or sell to other corporations,
contributing to total corporate sale counts.

Notably, corporations purchase the bulk of their holdings through atypical sales and sell a
majority of their holdings through typical sales. This trend may help uncover how corporations
use their scale and unique market advantages to purchase homes at lower prices and command
higher sale prices compared to fair market value; namely, a key mechanism is atypical sales,
especially foreclosure and multi-parcel sales, which consistently ranked as the top saleval codes
17
for these transactions. Atypical sales may be one tool which corporations use to purchase
properties for less than fair market value, extracting additional profit from a community.

Figure 3. Corporate Purchases Versus Corporate Sales in Atlanta

Total Sales by Investor Scale

Next, we track the SFR transaction trends by corporate investors’ scale. We categorize small,
medium, and large corporations as those with 1-10 (inclusive), 10-49 (inclusive), or 50+
transactions across Fulton County in the sale year, respectively. Figure 4 is consistent with the
previous in that corporate investors have participated in a growing number of SF transactions
from 2010 to 2021. But our analysis finds that the SF market is still dominated by smaller-scale
corporate investors. We observe a steady increase in small corporate transactions throughout the

18
period. But we also note a significant growth in medium and large corporate transactions after
2016. The proportion of corporate transactions compared to total transactions peaked in 2018,
when nearly 60% of sales involved a corporate entity; 2018 was the peak year for institutional
investors’ home purchasing activity in Atlanta. However, despite yearly fluctuations, neither
corporate purchases or sales increased significantly in proportion to total transactions from the
beginning to the end of the period, remaining around 24% and 35%, respectively.

Overall, large corporates’ proportion of total transactions increased slightly while medium-sized
corporates’ share decreased slightly throughout the period. Small corporations command the
majority of corporate transactions.

Figure 4. Corporate Purchases and Sales by Investor Size (Transaction Scale)

Share of SFR Ownership

19
While tracking the transaction trends is helpful to understand how the SF housing market has
shifted over the years in Atlanta city and Fulton County, it does not necessarily paint a picture of
corporate SFR ownership unless buying and selling are jointly tracked for the same corporations.
A more straightforward and complementary approach is directly dissecting the ownership
information. Figure 5 shows that corporate investors have rapidly increased their share of SFRs
in Atlanta throughout the study period 2010-2022, collectively growing from less than 30% to
almost 50%, meaning the presence of individual landlords has substantially shrunken by
corporate entities.

When it comes to the scale of corporate ownership, large investment firms have steadily grown
from owning less than 5% of SFRs in Atlanta to nearly 10%. Notably, small corporate investors
have experienced the greatest growth, from 15% to 27%. But medium-sized entities experienced
the slowest growth, from about 7% to 10%.

Figure 5. Share of SFR Ownership by Investor Size (Ownership Scale)

Contextualizing Atlanta’s Neighborhoods

20
Before we present neighborhood-level SFR portfolios held by corporate landlords, we visualize
Atlanta’s geospatial division along a historical boundary crossing southeast through the city, the
consequence of racially discriminatory planning decisions and economic development (Bayor
1988). Beneath this line and using Downtown (marked “1”) as a central point, in the south and
west of the city, neighborhoods have a high proportion of Non-Hispanic Black residents,
whereas neighborhoods to the northeast are predominantly Non-Hispanic whites. Household
income follows the same divide, with sections in the south and west having comparatively low
median incomes, and those in the north having high median incomes. In particular, the highest
income earners are highly concentrated in the north-most neighborhoods of Atlanta around
Buckhead (marked “2”). Understanding neighborhoods’ demographics is a crucial factor in
contextualizing the potentially disparate concentration of corporate SFR portfolios and their
financial impact on communities.

Figure 6. Demographics of Atlanta Neighborhoods

Neighborhood-level Corporate Single-Family Rental Portfolio

On average, individuals owned 86% of SF parcels across Atlanta. However, in some

neighborhoods to the south and east, individual ownership levels begin to fall under 70%, as
demonstrated by Figure 7. This pattern is reflected in corporate ownership; in particular, 15-20%

21
of the SF stock in many of these same neighborhoods are owned by small corporations. While
corporations of all sizes tend to concentrate on the south and east of the city, it is notable that
large corporations have a higher ownership share in neighborhoods further out from Downtown.
Since corporate owners own a large proportion of SF parcels in these neighborhoods, they can
extract larger sums of rent from these communities and may have disproportionate power in the
SF market. Due to our focus on estimating equity loss, we do not investigate or speculate the
ownership activity and concentration of large or institutional corporate investors.

Figure 7. Parcel Ownership in Atlanta Neighborhoods by Investor Size (Ownership Scale)

Next, we seek to understand the geography of SF transactions between corporations and

individuals, differentiating buying from selling for each type of actor. This results in four
possible cases as depicted in Figure 8. Transactions involving a corporate entity are concentrated

22
beneath the historical divide. In many of these neighborhoods, the share of corporate purchases is
between 15% and 25% whereas it is below 10% in the rest of neighborhoods. A similar and
stronger divide is observed for corporate-to-corporate sales, while the same but weaker pattern
can be seen in corporate sales to individuals. The inverse pattern is true of individual-to-
individual sales, with most sales being transactions between individuals in wealthier and
predominantly white neighborhoods. Only corporate purchases from individuals and corporate
sales to individuals are considered in our equity loss framework. Therefore, we should expect to
see equity loss follow a similar pattern, concentrating in south and east neighborhoods where
corporate transactions are a large share of the SF market. More concretely, Figure 8 reveals that
corporations are involved in a significant majority (about 70%) of all SF transactions in majority-
Black neighborhoods. Meanwhile, corporations are only involved in 30-40% of SF transactions
in other neighborhoods. Consequently, we expect to see disparate equity loss along racial lines
due to the differential concentrations of corporate SFR portfolios in these neighborhoods.

Figure 8. Proportion of SF Transaction Type in Atlanta Neighborhoods

23
 Corporate Investors’ Power in Single-Family Home Transactions

In proposing our framework of community’s equity loss, recall that we had a major assumption
that corporations have power and resources to extract profit during home buying and selling
transactions. We test this assumption using a set of hedonic regressions for SF home sales in
Atlanta during 2010-2021. We use two dependent variables. The first is sale price (logged) and
the second is sale price minus fair market value.

The notable feature of our model is that, based on our aforementioned methodology, we identify
whether buyers and sellers are corporate entities or individuals, resulting in four possible
categories in transaction dynamic, serving as focal independent variables: 1) corporate purchase
from individual; 2) corporate purchase from (or selling to) another corporation; 3) corporate
selling to individual; and 4) individual purchase from (or selling to) another individual. Each of
the four categories are mutually exclusive binary variables with the last being a reference (i.e.
omitted) category.

Our hedonic models have two units of analysis—property or transaction characteristics and
neighborhood features. For the former, we include 1) a binary indicator variable of whether a
transaction is a typical (i.e. valid) sale or not according to the Fulton County Board of Assessor’s
criteria, 2) log of living area square foot, log of lot size in square foot, total number of rooms,
number of baths, and effective age and its squared term to capture any non-linear relationship.

Further, the model controls for a grade factor, which is assigned to each property by the
appraiser. It represents the quality, type, style, or other judgment factors the appraiser believes
are encompassed in the property. The grades range from highest magnitude to lowest in the
following order: E+, E, E-, X+, X-, A+, A, A-, B+, B-, C+, C, C-, D+, D, and D-. They are
included in the model as a set of binary dummy variables (i.e. grade factor fixed effects). An
internal county code book writes “Grades are not affected by condition or depreciation, and E
grades are typically mansion and multi-million-dollar properties while D grades may include
small, square, short-gun style homes.” Hence, in addition to living area and lot size, both the
grade factor and effective age should capture quality, condition, and style of properties. Next, we
also include a categorical variable of heating, measuring if a property has central air,
central/forced heat, non-central heating and cooling, or no central air or heat. Lastly, the

24
neighborhood-level features are captured by neighborhood fixed-effects and the model controls
for yearly fixed-effects.

We first report two models in Table 3 where the dependent variable is sale price logged. The first
one is panel least square regressions with neighborhood and year fixed effects. The second one is
linear mixed models with neighborhood random effects and year fixed effects. The results, which
are identical between the two models, show that corporate investors purchase SF homes from
individuals on average at a 37.6 percent lower price than individuals buyers to from individual
sellers. Yet, when corporate investors sell those properties back to individuals, they on average
charge a price 7.4 percent higher than individual sellers do. Overall, all other control variables
show expected results consistent with the existing literature related to housing asset pricing and
hedonic regressions.

Table 3. Hedonic Regressions Testing Corporate Pricing Behavior in SF Transactions

Dependent var: Sale price of single-family detached houses logged

(1) (2)
Models Panel Least Squares Linear Mixed

Variables Coeff Robust SE Coeff Robust SE

Corporate buying from individuals -0.376*** 0.019 -0.376*** 0.019

Corporate selling to individuals 0.074*** 0.014 0.074*** 0.014

Corporate transaction with corporations -0.279*** 0.033 -0.279*** 0.033

Individual transaction with individuals Omitted Omitted

(Reference category) (Reference category)

Valid sale 0.219*** 0.018 0.219*** 0.018

Living area logged 0.266*** 0.031 0.266*** 0.031

Lot size logged 0.165*** 0.020 0.165*** 0.020

Number of rooms -0.001 0.003 -0.001 0.003

Number of baths 0.044*** 0.007 0.044*** 0.007

Effective age -0.011*** 0.001 -0.011*** 0.001

25
 Effective age squared 0.0001*** 0.00001 0.0001*** 0.00001

Grade FE Yes Yes

Heating/AC FE Yes Yes

Sale year FE Yes Yes

Neighborhood FE Yes No

Neighborhood RE No Yes

Observations 53,526 53,526

Adjusted R squared 0.827 -

Notes: FE = Fixed Effects. RE = Random Effects. Robust standard errors clustered by 95 NSAs
(Neighborhood Statistical Areas). Group variable for a mixed linear model is NSAs.
*** p<0.001, ** p<0.01, * p<0.05, + p<0.1.

One may question the results reported in Table 3, asking whether the type of properties corporate
investors target is systematically different from those purchased by individual buyers. This is a
reasonable question considering the existing knowledge that some corporate actors tend to target
properties that require structural repairs which are not preferred by individual buyers. While the
control variables in the model, including grading factor and effective age should address this
concern to some extent, it is still possible that the analysis may not be an apples-to-apples
comparison.

To address this concern, we next use a different dependent variable with the same model set ups;
here, for each property’s transaction, we compare its transaction price to its own fair market
value determined by county appraisers. That is, we ask if corporate buyers pay less than what a
property is worth, compared to what individuals pay than what it is worth. Since the fair market
value should capture any unobservable housing and neighborhood characteristics that are not
captured by hedonic features and neighborhood fixed/random effects in our model, this
dependent variable, operationalized as transaction price minus fair market value, gives us more
comfort in interpreting the results to measure whether corporate investors have more
institutionalized power and leverage for the pricing behaviors in SF property transactions. It
should be noted that this dependent variable has negative to positive values. As such, unlike the
previous models, we are unable to express the dependent variable in log. Instead, we leave them
in $ figures. Accordingly, we do not take logs for both living area and lot size.

26
The results, reported in Table 4, show that, on average, corporate buyers paid about $22,000 to
$23,000 less than an SF property’s fair market value when compared with what individual
buyers paid for what properties were worth. Also, consistent with the results in Table 3, when
corporate investors sold those SF properties back to individual buyers, they charged about
$7,600 to $8,000 more than the properties’ fair market value, compared to the prices individual
buyers paid for what they were worth. In general, the control variables show similar results to the
previous ones.

Table 4. Hedonic Regressions with an Alternative Dependent Variable

Dependent var: Differential between sale price and fair market value (price minus FMV in $)

(1) (2)
Models Panel Least Squares Linear Mixed

Variables Coeff Robust SE Coeff Robust SE

Corporate buying from individuals -22389.70*** 1640.15 -22666.67*** 1616.68

Corporate selling to individuals 7969.53*** 1237.46 7595.15*** 1230.45

Corporate transaction with corporations -15682.03*** 2445.21 -16121.1*** 2416.01

Individual transaction with individuals Omitted Omitted

(Reference category) (Reference category)

Valid sale 25928.2*** 1508.67 26138.1*** 1495.18

Living area in sqft 0.68 0.89 0.84 0.89

Lot size in sqft 0.06+ 0.03 0.07+ 0.03

Number of rooms -264.19 251.51 -296.03 250.91

Number of baths 2386.72*** 637.12 2473.74*** 632.11

Effective age -195.22+ 111.33 -173.88 110.04

Effective age squared 2.53** 0.92 2.37** 0.91

Grade FE Yes Yes

Heating/AC FE Yes Yes

Sale year FE Yes Yes

Neighborhood FE Yes No

27
 Neighborhood RE No Yes

Observations 53,526 53,526

Adjusted R squared 0.227 -

Notes: FE = Fixed Effects. RE = Random Effects. Robust standard errors clustered by 95 NSAs
(Neighborhood Statistical Areas). Group variable for a mixed linear model is NSAs.
*** p<0.001, ** p<0.01, * p<0.05, + p<0.1 .

Neighborhoods’ Equity Loss by Corporate SFR Investors, Atlanta 2011-2021

Our framework considers equity loss due to corporate purchases for less than fair market value,
corporate sales for greater than fair market value, and SFR net rental yields. We label these as
purchase loss, sale loss, and rental loss, respectively. Total loss is the sum of each loss, and
Figure 9 displays the loss across all years in the study period. We exclude 2022 because we do
not have sales data for that year. The total equity loss triggered by SFR corporate investors in
Atlanta between 2011 and 2021 amounts to a sum of $1.25B in 2022 dollars. Stratifying the
neighborhoods into majority-Black and those are not, these sums are $681M and $568M,
respectively. Across all neighborhoods, purchase-driven equity loss amounted to $77M (6%),
whereas sale loss amounted to $472M (37%), and rental loss stood at $700M (56%). For
majority-Black neighborhoods, the corresponding figures are $95M purchase loss (14%), $245M
sale loss (36%), and $340M rental loss (50%). For non-majority-Black neighborhoods, these
totals are $-18M (-3%), $226M (40%), and $359M (63%), respectively.

Equity loss from corporate purchases is concentrated south and east. In these neighborhoods, as
shown in our hedonic regressions above, corporate investors often paid well below fair market
value when acquiring SF properties from individuals. Meanwhile, sale loss appears to
concentrate heavily in a few neighborhoods, although there is no spatially discernable pattern
between them. It is worth noting that Grant Park and Ormewood Park, neighborhoods with some
of the highest sale loss, are gentrifying due to their significantly higher incomes and lower
percentage of minorities than bordering neighborhoods. Finally, rental business-driven equity
loss concentrates heavily in northern neighborhoods where property values are high. For
instance, the Kingswood neighborhood has multiple properties with an estimated yearly rent
value above $500K and they are owned by obscure holding companies.

28
Most of the loss within northern Atlanta neighborhoods (e.g. Buckhead) is due to rental loss.
Equity loss for neighborhoods to the south and east is driven disproportionately by purchase loss
and, to a lesser extent, sale loss. Total equity loss is greatest in Buckhead, followed closely by a
subset of neighborhoods to the south and east. Adjusted for inflation (2022 dollars), total equity
loss in Kingswood in Buckhead, the neighborhood with the greatest loss, was $44M, followed by
Morningside/Lenox Park ($38M), North Buckhead ($36M), Grant Park ($36M), Chastain Park
($36M), Adair Park/Pittsburgh ($35M), Ormewood Park ($34M), and Bush Mountain ($32M).
In contrast, Castleberry Hill/Downtown ($31K), Brookwood Hills ($547K), and Ardmore
($1.4M) had the least equity loss. The median equity loss across all neighborhoods is $11M and
the average equity loss is $13M.

Figure 9. Equity Loss in Atlanta Neighborhoods

29
Standardizing Equity Loss Between Neighborhoods

While the absolute levels of corporate SFR investor driven equity loss are helpful to understand
the scope of the issue, they themselves are not sufficient to derive meaningful implications for
place-based interventions. This is a consequence of various factors, namely that neighborhoods
have different numbers of parcels, median sales prices, and number of transactions. Each of these
variables can influence the total calculated equity loss, skewing equity loss calculations in favor
of large and wealthy neighborhoods where greater rents and sales prices mean that losses are
magnified even if they are a relatively small portion of the sales price or the median resident’s
income.

To account for these factors and place the emphasis back on the lived experience of those in a
neighborhood, we construct an equity loss share, which is arguably more salient to understand
the impact of corporate SFR investment and orient place-based policy discussions. This metric,
equity loss share, draws on previous literature related to rent burden (Goodman and Kawai 1984;
Sharma and Samarin 2022), which is the portion of an individual’s monthly or yearly income
spent on rent. While those who live in communities impacted by equity loss do not directly pay
with their incomes, we can calculate the percentage of total income generated by each
community that is extracted by corporations through SFR business activity. These community
losses, which materialize in residents’ lives when they purchase or sell a home, or decide to rent
a home instead, are effectively financial loss which can no longer be invested into their future, or
the future of the community. Instead, these assets or income are extracted by external corporate
owners and shareholders, who have exerted market power and likely pushed homebuyers out of
the market, as recent studies show (An 2023; Billings and Soliman 2023).

Therefore, we construct an equity loss share measure which is the total equity loss of a
neighborhood divided by the total income generated by households in the community in any
given year as follows.
𝑇𝑜𝑡𝑎𝑙 𝐸𝑞𝑢𝑖𝑡𝑦 𝐿𝑜𝑠𝑠 (𝑛)
𝐸𝑞𝑢𝑖𝑡𝑦 𝐿𝑜𝑠𝑠 𝑆ℎ𝑎𝑟𝑒 (𝑛) =
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝐻𝑜𝑢𝑠𝑒ℎ𝑜𝑙𝑑 𝐼𝑛𝑐𝑜𝑚𝑒 (𝑛) ∗ 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐻𝑜𝑢𝑠𝑒ℎ𝑜𝑙𝑑𝑠 (𝑛)
where n is neighborhood

For calculation over multiple years, the equation can be modified as follows.

30
 𝑦𝑒𝑎𝑟𝑠
∑𝑦 𝑇𝑜𝑡𝑎𝑙 𝐸𝑞𝑢𝑖𝑡𝑦 𝐿𝑜𝑠𝑠 (𝑛,𝑦)
𝐸𝑞𝑢𝑖𝑡𝑦 𝐿𝑜𝑠𝑠 𝑆ℎ𝑎𝑟𝑒 (𝑛, 𝑦𝑒𝑎𝑟𝑠) = 𝑦𝑒𝑎𝑟𝑠
∑𝑦 (𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝐻𝑜𝑢𝑠𝑒ℎ𝑜𝑙𝑑 𝐼𝑛𝑐𝑜𝑚𝑒 (𝑛,𝑦)∗𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐻𝑜𝑢𝑠𝑒ℎ𝑜𝑙𝑑𝑠 (𝑛,𝑦))

where n is neighborhood and years is the study period range

This measure places equity loss in the context of the average resident’s income in a
neighborhood. It quantifies the share, or a portion of average resident’s financial assets being
extracted equivalent to their income, and otherwise prevented from becoming equity, per year by
corporate SFR investment in their neighborhood. To clarify, by preventing equity, we mean that
these individual and community losses may have instead been invested into home equity or other
personal investments to grow and compound, or back into the community itself by providing
resources or support to family members, neighbors, local business, and other community and
economic development activities.

Figure 10. Relative Equity Loss in Atlanta Neighborhoods

When standardizing equity loss in terms of each neighborhood’s average household income, in
Figure 10, we see a clear pattern that predominantly non-Hispanic Black and low-income
neighborhoods to the south and east, near the Downtown, suffer the highest proportional impact

31
of corporate SFR investment. Between 2012 and 2021, the most burdened neighborhoods have
lost about 5% of the income they generated to corporate SFR investors every year as indicated in
Figure 10. While we cannot say that every or most households individually experienced a 5%
loss of their wealth, the community as a whole did, and certain households, especially residents
who were directly involved in SF property transactions with corporate investors, as well as those
who were likely pushed out of the homeownership market, should have been more impacted than
others. In particular, households who had to sell their properties to corporate investors at
discounted prices, such as those experiencing foreclosures, suddenly increasing tax bills,
unexpected medical debt, or economic shocks, likely bore the most significant loss as they are
more likely to be prime targets for corporate investors. Additionally, some renters in these
neighborhoods, who considered housing tenure mobility to homeownership, are also
disproportionately prevented from building future equity and wealth. Overall, the majority-Black
neighborhoods lost financial assets equivalent to 1.3% of the income they collectively generated
during the study period, whereas the remaining neighborhoods lost 0.28% of their income due to
corporate SFR business.

More concretely, take Bush Mountain (see Figure 10), a majority-Black neighborhood with the
highest loss relative to its average household income, as an example. In 2019, the year Bush
Mountain suffered the highest loss proportional to its income, a typical household’s share of loss
amounted to $4,341, out of an average household income of $40,524. Such a level of equity
extraction (10.7% of income) might disempower and prevent families from pathways to a higher
homeownership rate and generational wealth-building. Aggregated over all years, Bush
Mountain’s equity loss was approximately 4% of the total income it generated throughout the
study period.

Conclusions

The direct financial impact of the wholesale purchase and subsequent business of single-family
rental (SFR) homes by corporate investors on the communities has been difficult to discern for
various reasons, including challenges in identifying corporate property owners and a lack of
relevant framework for the impact measurement. Quantifying the value, or equity, extracted out
of a community and distributed to corporate external shareholders is a particularly salient

32
measure, considering that home equity is the single most important wealth-building asset for a
typical American family. This paper provides a comprehensive method to quantify the equity
loss of communities due to corporate SFR business activity with the application to 95
neighborhood statistical areas (NSAs) in the city of Atlanta for the years between 2010 and
2022.

To understand trends in the SF housing market and the role of corporate investors in each of the
Atlanta neighborhoods, we use Fulton County tax assessment and property sales data to track SF
home purchasing, holding, and selling activities by corporate entities. We create an easy, yet
robust method to aggregate parcel owners, facilitating a comprehensive ownership analysis that
unveils even nested corporate ownership structures. Our descriptive analysis shows that Atlanta’s
SF market grew dramatically between 2010 and 2021, with an overall consistent share of
corporate rental business activity despite yearly fluctuations.

Corporate purchases have been around 24% of all transactions, whereas corporate sales had a
35% share, but over the years, their shares have increased. Interestingly, corporate investors have
increasingly relied upon atypical transactions, including foreclosure sales and multi-parcel sales,
when acquiring properties but they sought regular market (labeled typical) transactions when
selling properties. Majority-Black neighborhoods have seen a much higher share of corporate
transactions in their neighborhoods, with about 70% of transactions each year involving a
corporate entity as either a buyer or seller, compared to 30-40% for the rest of the
neighborhoods. Such a pattern likely reveals the mechanisms by which corporate investors
extract wealth from communities, utilizing their institutional advantages and unique financing
resources to control a large share of SFR properties in target neighborhoods.

When it comes to the scale of corporate investors, overall, large corporates’ (those with 50+
transactions in any given year) proportion of total transactions increased slightly while medium
corporates’ (10-49 transactions) share decreased slightly throughout the period. Notably, small
corporate investors (1-10 transactions) command the majority of corporate transactions and their
share has substantially increased over the years. Collectively, corporate investors have rapidly
increased their share of SFRs in Atlanta throughout the study period, collectively growing from
less than 30% to almost 50%.

33
Zeroing in further on the impact SFR investment has on neighborhoods, we develop and apply an
equity loss methodology to each of Atlanta’s 95 neighborhoods. To quantify equity loss to the
affected communities, our framework estimates the net profits accumulated by corporate
investors through the acquisition of SF properties by paying under fair market value (FMV),
gaining the income from rental business, and the selling of such properties at higher prices over
FMV. While we consider price appreciation channels over the years via growth in annual rents
from increased property values, it should be a conservative approach since our estimate is based
on home values appraised by county appraisers that tend to lag true market values.

In absolute terms and adjusted for inflation (2022 dollars), our estimation shows that Atlanta’s
neighborhoods lost $1.25B in financial equity due to corporate SFR investors between 2011 and
2021. Stratifying the neighborhoods into majority-Black and the rest, these sums are $681M and
$568M, respectively. Neighborhoods like Kingswood, Grant Park, and Adair Park/Pittsburgh,
which are amongst the neighborhoods that suffered the most lost $35M to $44M. Across all
neighborhoods, equity loss due to corporate purchase only amounts to $77M (6%), while the loss
triggered by corporate sale and rental business accrue to $472M (38%) and $700M (56%),
respectively, making the most of the financial loss. Interestingly, we find that SFR rental
operation only accounts for about half of total equity loss triggered by corporate investors, while
the rest is mostly attributable to profits made by home selling businesses.

While these figures provide a broad overview of city-wide effects of corporate SFR investment,
they need to be contextualized to illustrate the direct impacts at the community level, to
demonstrate how corporate investment has influenced the development of specific
neighborhoods and the lives of their residents. To this end, we propose an equity loss share
metric. This metric shows equity loss relative to the neighborhood’s overall economic standing,
quantifying the community’s financial assets extracted by corporate investors in the annual
income percentage. Neighborhoods with the highest equity loss have experienced a financial
extraction equivalent to 5% of their income annually between 2011 and 2021 by corporate
investors. In total, majority-Black neighborhoods lost assets equivalent to 1.6% of their income
during the study period, whereas other neighborhoods lost about 0.4%.

Equity loss is a result of the unique market power that corporate SFR investors hold over
individuals in the local housing markets with minority majority neighborhoods being more

34
vulnerable due to the concentrated SFR portfolios of corporate landlords in the areas. Both panel
fixed effects and mixed linear models of hedonic regression analysis show that, on average,
corporate buyers paid 37.6% less prices than individual buyers when acquiring SF properties
from individual home sellers, yet corporate entities charged a 7.4% higher premium for selling
those properties back to individuals after operating the rental businesses.

Our equity loss framework could help expose a community-centric narrative about the impact of
corporate SFR investors. Our analysis illustrates that these metrics provide valuable insight into
the overall influence of corporate investors in the SF local housing markets, as well as its
community-level impacts that may disempower families from building financial assets and
generational wealth. While exploring the mechanisms of these impacts is beyond the scope of the
current study, they may include debt-stricken families taking all-cash offers for their homes from
corporate investors, well below fair market value, leading to a shrunken housing availability for
homeownership in the neighborhood. Future studies may uncover these and other mechanisms,
for instance, by using surveys and community stakeholder interviews.

Additionally, we acknowledge some limitations. First, we simplistically estimate the net rental
income based on each property’s appraised home value and the number of years and months they
were operated as rentals. A more precise, refined approach should holistically consider other
relevant factors, including property taxes (or appeal filing outcomes), property (management)
expense, and rents charged and collected. Next, while we examine the scales of corporate
ownership and transaction in the descriptive analysis of the overall Atlanta SF housing markets,
we do not examine its role in equity loss estimation for the neighborhoods. For instance, the
concentration of institutional investors could impact communities (in terms of transaction price
or rents) more strongly or differently than that of small corporate actors. Future studies could
examine such potentially dynamic roles of corporate SFR investors by their scales.

Lastly, our framework does not help discern whether a specific corporate actor is a good or bad
investor but instead treats them alike as those causing equity loss to communities. It is possible
that certain corporate (or non-profit) investors serve their residents as community quarterback
organizations to make rental housing more affordable after acquiring them from individuals; they
could make rents more affordable than individual landlords. Yet our framework can, at best,
measure which corporate landlords are causing more or less financial equity loss to communities,

35
especially for those whose homeownership opportunities are foregone, but not the rental
affordability they contribute to.

Notwithstanding these limitations, focused on equity loss estimation, our methodology and
framework can inform housing policies from a place-based intervention perspective. The existing
policy discourses focus on regulating the owning and purchasing activities of institutional
investors, incentivizing them to sell their properties back to communities. Such approaches,
however, would only address what communities face today, but not what they have experienced
until now. If anything, policymakers could come up with a more comprehensive framework that
considers cumulative impacts wrought by corporate investors. With an adequate framework to
quantify such financial equity loss and comprehensive data often publicly available, planners and
analysts can seek to measure the financial impacts for their local jurisdictions to inform place-
based intervention strategies or priorities.

An equally important consideration is having broader perspectives when quantifying such equity
loss by corporate investors. While institutional investors are the actors in the SFR market that
have received most attention in recent years, we should keep in mind that smaller-scale corporate
entities play a more prominent role in most neighborhoods. That noted, a framework should
capture a broad array of corporate actors for quantifying the financial impact on affected
neighborhoods, and within that framework, a differentiated impact by different scales of
corporate investors could be further measured and understood.

36
References

An, B. Y. (2023). The Influence of Institutional Single-Family Rental Investors on

Homeownership: Who Gets Targeted and Pushed Out of the Local Market? Journal of Planning
Education and Research, 0739456X231176072.
An, B. Y., Jakabovics, A., Orlando, A. W., Rodnyansky, S., & Son, E. (2024). Who Owns
America? A Methodology for Identifying Landlords’ Ownership Scale and the Implications for
Targeted Code Enforcement. Journal of the American Planning Association, 1-15.

Austin, N. (2022). Keeping up with the blackstones: Institutional investors and gentrification.
Working Paper. SSRN. https://dx.doi.org/10.2139/ssrn.4269561.
Bayor, R. H. (1988). Roads to Racial Segregation: Atlanta in the Twentieth Century. Journal of
Urban History, 15(1), 3–21.
Billings, S. B., & Soliman, A. (2023). The erosion of homeownership and minority wealth.
Working Paper. SSRN. https://dx.doi.org/10.2139/ssrn.4649479.

Brunson, S. K. (2020). “Wall Street, Main Street, Your Street: How Investors Impact The Single-
family Housing Market.” Three Essays in Housing. [Ph.D. dissertation]. The University of North
Carolina at Charlotte. https://search.proquest.com/dissertations-theses/three-essays-
housing/docview/2395652812/se-2.

Charles, S. L. (2020). The financialization of single-family rental housing: An examination of

real estate investment trusts’ ownership of single-family houses in the Atlanta metropolitan area.
Journal of Urban Affairs, 42(8), 1321-1341.
Colburn, G., Walter, R. J., & Pfeiffer, D. (2021). Capitalizing on collapse: An analysis of
institutional single-family rental investors. Urban Affairs Review, 57(6), 1590-1625.
Coven, J. (2023). The Impact of Institutional Investors on Homeownership and Neighborhood
Access. Working Paper. SSRN. https://dx.doi.org/10.2139/ssrn.4554831.

Demers, A., & Eisfeldt, A. L. (2022). Total returns to single‐family rentals. Real Estate
Economics, 50(1), 7-32.

Fields, D., & Raymond, E. L. (2021). Racialized geographies of housing financialization.

Progress in Human Geography, 45(6), 1625-1645.

Fields, D., & Vergerio, M. (2022). Corporate landlords and market power: What does the single-
family rental boom mean for our housing future? University of California, Berkeley.
https://escholarship.org/uc/item/07d6445s.

Ganduri, R., Xiao, S. C., & Xiao, S. W. (2023). Tracing the source of liquidity for distressed
housing markets. Real Estate Economics, 51(2), 408-440.

37
Garriga, C., Gete, P., & Tsouderou, A. (2023). The economic effects of real estate investors. Real
Estate Economics, 51(3), 655-685.

Goodman, A. C., & Kawai, M. (1984). Estimation and policy implications of rental housing
demand. Journal of Urban Economics, 16(1), 76–90.
Goodman, J. (2003). Homeownership and investment in real estate stocks. Journal of Real Estate
Portfolio Management, 9(2), 93-105.
Goodman, L., Zinn, A., Reynolds, K., & Noble, E. (2023). A Profile of Institutional Investor–
Owned Single-Family Rental Properties. Urban Institute, the Housing Finance Policy Center.
Gotham, K. F. (2000). Urban space, restrictive covenants and the origins of racial residential
segregation in a US city, 1900–50. International Journal of Urban and Regional
Research, 24(3), 616-633.
Gurun, U. G., Wu, J., Xiao, S. C., & Xiao, S. W. (2023). Do Wall Street Landlords Undermine
Renters’ Welfare? The Review of Financial Studies, 36(1), 70-121.

Harrison, A., Immergluck, D., & Walker, J. (2024). Single-family rental (SFR) investor types,
property conditions, and implications for urban neighbourhoods: evidence from Memphis,
Tennessee. Housing Studies, 1-21.

Hyra, D. S. (2012). Conceptualizing the new urban renewal: Comparing the past to the
present. Urban Affairs Review, 48(4), 498-527.

Immergluck, D. (2018). Renting the dream: The rise of single-family rentership in the Sunbelt
Metropolis. Housing Policy Debate, 28(5), 814-829.

Immergluck, D., & Law, J. (2014). Speculating in crisis: The intrametropolitan geography of
investing in foreclosed homes in Atlanta. Urban Geography, 35(1), 1-24.

Jones-Correa, M. (2000). The origins and diffusion of racial restrictive covenants. Political
Science Quarterly, 115(4), 541-568.

Lambie‐Hanson, L., Li, W., & Slonkosky, M. (2022). Real estate investors and the US housing
recovery. Real Estate Economics, 50(6), 1425-1461.
Mills, J., Molloy, R., & Zarutskie, R. (2019). Large‐scale buy‐to‐rent investors in the single‐
family housing market: The emergence of a new asset class. Real Estate Economics, 47(2), 399-
430.

Orlando, A. W. (2021). Keeping races in their places: The dividing lines that shaped the
American city. Routledge.

Pacific Legal Foundation. (n.d.). Ending Home Equity. https://pacificlegal.org/property-

rights/home-equity-theft/.

38
Rappaport, J. (2010). The effectiveness of homeownership in building household wealth.
Economic Review-Federal Reserve Bank of Kansas City, 35.
https://www.kansascityfed.org/Economic%20Review/documents/379/PDF-10q4Rappaport.pdf.

Raymond, E. L., Duckworth, R., Miller, B., Lucas, M., & Pokharel, S. (2018). From foreclosure
to eviction: Housing insecurity in corporate-owned single-family rentals. Cityscape, 20(3), 159-
188.

Rothstein, R. (2017). The color of law: A forgotten history of how our government segregated
America. Liveright Publishing.

Seymour, E., Shelton, T., Sherman, S. A., & Akers, J. (2023). The metropolitan and
neighborhood geographies of REIT-and private equity–owned single-family rentals. Journal of
Urban Affairs, 1-25.

Sharma, M., & Samarin, M. (2022). Rental tenure and rent burden: progress in interdisciplinary
scholarship and pathways for geographical research. GeoJournal, 87(4), 3403–3421.
Smith, P. S., & Liu, C. H. (2020). Institutional investment, asset illiquidity and post‐crash
housing market dynamics. Real Estate Economics, 48(3), 673-709.

Taylor, K. Y. (2019). Race for profit: How banks and the real estate industry undermined black
homeownership. UNC Press Books.

US Census Bureau. 2022. Growth in U.S. Population Shows Early Indication of Recovery Amid
COVID-19 Pandemic. Census.Gov. Retrieved February 24, 2024.
https://www.census.gov/newsroom/press-releases/2022/2022-population-estimates.html.

Watson, C. L., & Ziv, O. (2021). Is the Rent Too High? Land Ownership and Monopoly Power.
Working Paper. SSRN. https://dx.doi.org/10.2139/ssrn.3774672.

Zillow. 2012, August 24). Investing 101: Estimating Rental Property Expenses.
https://www.zillow.com/learn/investing-101-estimating-rental-property-expenses/.

Zillow. (2021, February 4). How Much Should I Charge for Rent?
https://www.zillow.com/rental-manager/resources/how-much-can-i-rent-my-house-for/.

39
 Online Appendix A. Data Processing Additional Details

NSA Additional Information

The Atlanta Regional Commission accomplished this NSA delineation with assistance of the
city’s planning department. Their research team looked at imagery and zoning to resolve
questions about assignment, and they consulted the Neighborhood Planning Units (NPUs), which
are community-based local government entities in Atlanta. The designated NSAs were required
to have a minimum population of 2,000; areas not meeting the threshold were combined with
neighboring areas within the same NPU to reach the minimum population size. Every census
block within the city of Atlanta was assigned to NSA on one-to-one basis.

Parcel Data Cleaning

Step 1 (parcel and sales data): drop non-SF and complete duplicates. As our interest lies in
single-family detached properties, we keep only SF detached properties by filtering the relevant
land use code in both datasets before merging. 6

Note. A small subset of parcels are missing tax records for at least one year throughout the 13
year period. The total number of missing entries is 18,722 out of 2,785,447. This data loss is not
a consequence of processing but rather of incomplete records from Fulton County. While
noteworthy, we believe this will only have a negligible impact on the results. At worst, the
calculations will undercount; we consider this acceptable.

Step 2 (parcel data): drop rows with essential values missing. To facilitate further analysis
steps, we drop records with empty owner address strings. Some additional string cleaning is later
applied to the parcel owner’s address, as detailed in the following section, but no additional rows
were dropped during that process.

Step 3 (parcel data) and Step 2 (sales data): geocoding.

Step 4 (parcel data): retain only the main structure on each parcel. Within tax assessor data,
we consider the Parcel ID and tax year columns collectively as the unique key. However, the
format of Fulton County’s assessor data allows for a single parcel to have multiple records in one
year if there are multiple structures on the parcel. For SF parcels, these additional structures are

40
ADUs. As ADUs are outside the scope of our research question, we retain only the record with
the largest living area square footage— the main structure— for each key.

Sales Data Cleaning

Steps 1-2: see equivalent steps for the parcel data.

Step 3 (sales data): drop non-arm’s length transactions. Additionally, we drop non-arms
length sales. Retaining such sales would inappropriately skew calculations because the sales
price is unlikely to reflect any true market condition. We consider a non-arms length sale as any
sale classified with one of the following Saleval codes: T (sale price <= 1,000), G (gift), 6F
(quitclaim), 4 (related individuals or corporations). Saleval codes, defined by the Fulton County
Board of Assessors, will be discussed further in the next section.

Step 4 (sales data): retain only one record per unique transaction. Sales data also contains
some irrelevant duplicate records on the sale date, Parcel ID, and sale price unique key. These
duplicates occur when the entire row is identical except for the Saleval code (i.e. the type of
transaction). We drop any duplicate rows where sale date, Parcel ID, and sales price are identical
(11,416 records), retaining only the first record.

Step 5 (sales data): replace sales price for multi-parcel transactions to reflect average price
per parcel. Multi-parcel sales, which are of special interest as a common tool for investors to
acquire a bulk of properties, require an additional cleaning step. As usual in other proprietary
real estate transaction data, in Fulton’s sales records, every parcel involved in a multi-parcel sale
has the total transaction amount recorded as the sales price, rather than the price paid on a per-
parcel basis. For such cases, we divide the transaction price by the total number of parcels
involved in the multi-parcel sale. While this method does not perfectly capture the value each
individual parcel brought to the sale, it effectively averages out these differences.

Step 6 (sales data): drop government and bank sales. Finally, we exclude transactions in
which the grantee or grantor is a government institution or bank. These sales do not share the
same implications as investors profiting from SFRs and thus are excluded from our framework.

41
 Online Appendix B. Uncovering Ownership

Count and Percent of Grantee Matches for Each Pass

Pass Count and Percent Matched

Pass 1: Exact Count: 142,531

Percent: 78.9%

Pass 2: Single-Sale Only Count: 20,342

Percent (cumulative): 90.1%

Pass 3: Exact Name Count: 12199

Percent (cumulative): 96.9%

Number and Percent of Grantee Matches for Each Pass

Pass Number and Percent Matched

Pass 1: Exact Count: 64062

Percent: 35.4%

Pass 2: Single-Sale Only Count: 80562

Percent (cumulative):80.0%

Pass 3: Exact Name Count: 21671

Percent (cumulative): 92.0%

42
 Online Appendix C. SFR Portfolios of Institutional Investors

Note: for keywords that can be substrings without representing the desired value (e.g. “IH” for Invitation
Homes may appear in a name like “IHOTEL”), an additional space is added to the beginning or end. This
measure ensures that the substring (often an acronym) is standalone, as typically written in institutional
investor subsidiary names.

AMHERST CAPITAL

KEYWORDS: AMHERST, ARVM

ADDRESSES: 5001 PLAZA ON THE 78746, 5001 PLAZA ON THE LAKE 78746, 8300 MOPAC
EXPRESSWAY 78759

NAMES

ALTO ASSET COMPANY 1 LLC FORMERLY KNOWN, VM PRONTO LLC, BAF 2 LLC, MESA
VERDE ASSETS LLC, TRANS AM SFE II LLC, LHF 4 ASSETTS LLC, JEFF I LLC, SRMZ 4 ASSET
COMPANY Z LLC, BAF ASSE TS 2 LLC, CPI/AMHERST SFR PROGRAM OWNER LLC, SRMZ 4
ASSET COMPANY 2 LLC, TCVM 6 LLC, AMNL ASSETS COMPANY 2 LLC, RH PARTNERS
OWNERCO I LLC, JEFF L LLC, BAF ASSETS 3 LLC, US SFE ASSET COMPANY 5 LLC, MUPR 3
ASSETS COMPANY, RH PARTNERS OWNERCO LLC, BAF ASSETS 2 LLC, SRMZ 1 LLC, CPI
AMHERST SFR PROGRAM II OWNER L L C, CPI AMHERST SFR PROGRAM OWNER LLC, JEFF
1 LLC, BROADTREE HOMES GEORGIA LLC, ARVM 5 LLC, HOME SFR BORROWE LLC,
GRAND AM 4 LLC, US SFE ASSET COMPANY 1 LLC, SRMZ 4 ASST COMPANY 2 LLC, AMNL
ASSET COMPANY 1 LLC, EPH 2 ASSETS LLC, BAF 3 LLC, ALTO ASSET COMPANY 1 LLC,
BSVM ASSETS LLC, MSR I ASSETS COMPANY LLC, SFR ASSETS OWNER LLC, BTRA V LLC,
BAF 2 TRS LLC, RPA4 LLC, HOME SFR BORROWER LLC, FIREBRD SFE I LLC, SRAM PACK I
C L L C, SUNBELT INVESTORS ASSET COMPANY LLC, AMNL ASSSET COMPANY 1 LLC,
SFRA III LLC, ARVVM 5 LLC, RAC 2 LLC, US SFE ASSET COMPANY 2 LLC, RH PARTNERS
OWNER CO LLC, BAF ASSETS LLC, MUPR 3 ASSETS LLC, MSR 1 ASSETS COMPANY LLC,
VICTRUM ANDREW, SAFARI TWO ASSET COMPANY LLC, SAFARI ONE ASSET COMPANY
LLC, RH EVERGREEN OWNERCO LLC, BAF I LLC, US SFE ASSET COMPANY 4 LLC, AMNI
ASSET COMPANY 1 LLC, LHF4 ASSETS LLC, LHF 4 ASSETS LLC, VM PRONTO LLC LLC,
ARMM ASSET COMPANY 1 LLC, BAF 1 LLC, SRMZ 3 LLC, EPH 2 ASSETSS LLC, EPH2 ASSETS
LLC, BAF 1LLC, PREP 6 LLC, HFS I ASSETS COMPANY LLC, FIREBIRD SFE I LLC, AMNL
ASSET COMPANY 2 LLC, ALTO ASSET COMPANY 2 LLC, CPI AMHERST SFR PROGRAM

43
OWNER L L C, CPI AMHERST SFR PROGRAM II OWNER LLC, SRMZ 2 LLC, MSR I ASSET
COMPANY LLC, SUNFIRE 3 LLC, SRMZ 4 ASSET COMPANY 1 LLC

CERBERUS CAPITAL

KEYWORDS: CERBERUS, FKH, RM1, RMI

ADDRESSES: 0 PO BOX 2249 30028, 1850 PARKWAY 30067

NAMES

FKH SFR C2 LP, RMI SFR PROPCO A LP, FKH SFR PROPCO I L P, FKH SFR PROPCO G LP,
CERBERUS SFR HOLDINGS III L P, ASHTON ATLANTA RESIDENTIAL LLC, FKH SFR
PROPCO H L P, FKH SFR C2 L P, NEWMAN ROHN E & ELISA F, CEREBUS SFR HOLDINGS II
LP, FKH SFR PROPCO B HLD LP, CDCG 3AW LP, RM1 SFR PROPCO A L P, FKH SFR C1 LP,
CERBERUS SFR HOLDINGS LP, CERBERUS SFR HOLDINGS III LP, CERBERUS SFR HOLDING
II L P, ASHTON ATLANTA RESIDENTIAL L L C, FKH SFR PROPCO D LP, RM1 SFR PROPCO A
LP, STARS & STRIPES 2G LLC, FKH SFR PROPCO D L P, FKH SFR PROPCO H LP, CERBERUS
SFR HOLDINGS V LP, RM1 SFR PROPCO B LP, RM1 SFR PROPCO B L P, CEREBRUS SFR
HOLDINGS II LP, SFR JV 1 PROPERTY LLC, CERBERUS SFR HOLDINGS L P, CSMA BLT LLC,
RMI SFR PROPCO A L P, FKH SFR PROPCO G L P, FKH SFR PROPCO I LP, FKH SFR PROPCO A
L P, CERBERUS SFR HOLDINGS II L P, CERBERUS SFR HOLDINGS II LP, CSMA SFR
HOLDINGS II LSE LLC, CERBERUS SFR HOLDINGS V L P, RM1SFR PROPCO B LP

INVITATION HOMES

KEYWORDS: INVITATION, IH

ADDRESSES: 1717 MAIN 75201, 8665 HARTFORD 85255, 901 MAIN 75202

NAMES

2018 1 IH BORROWER LP, 2018 3 IH BORROWER LP, 2015 3 IH2 BORROWER LP, 2017 2 IH
BORROWER LP, 2018 1 IH BORROWE LP, 2015-2 IH2 BORROWER TRS LLC, IH2 PROPERTY
TRS 2 L P, TARBERT LLC, 2019 1 IH BORROWER LP, COLFIN AI GA LLC, THR GEORGIA L P,
COLFIN AI GA 2 LLC, 2014 1 IH BORROWER L P, STARWOOD WAYPOINT TRS LLC, BEAULY
LLC, 2015 3 IH2 BORROWER TRS LLC, 2018 2 IH BORROWER LP, 2014 2 IH BORROWER L P,
CAH 2014 1 BORROWER LLC, 2017 1 IH BORROWER LP, COLFIN AI GA I LLC, 2014 3 IH

44
BORROWER L P, SFR JAVELIN BORROWER L P, SWH 2017 1 BORROWER LP, CSH PROPERTY
ONE LLC, SRP SUB LLC, 2017 1 IH BORROWER L P, CAH 2015 1 BORROWER LLC, 2018 4 IH
BORROWER LP, IH6 PROPERTY GEORGIA LP, SWAY 2014 1 BORROWER LLC, COLFIN AI
GA1 LLC, SFR JAVELIN BORROWER LP, IH3 PROPERTY GEORGIA, CSH 2016 2 BORROWER
LLC, COLFIN AL GA 1 LLC, MORVEN LLC, 2015 1 IH2 BORROWER L P, COLFIN AI GA 1 LLC,
CSH 2016 1 BORROWER LLC, CAH 2014 2 BORROWER LLC, SWAY 2014-1 BORROWER LLC,
IH3 PROPERTY GEORGIA L P, IH4 PROPERTY GEORGIA LP, 2013 1 IH BORROWER L P, 2015 2
IH2 BORROWER L P, THR GEORGIA LP, IH2 PROPERTY GEORGIA L P

STARWOOD CAPITAL

KEYWORDS: COLONY, STARWOOD, CSH, CAH

ADDRESSES: 8665 HARTFORD 85255

NAMES

CSH PROPERTY ONE LLC, SRP SUB LLC, TARBERT LLC, CAH 2015 1 BORROWER LLC,
SWAY 2014 1 BORROWER LLC, COLFIN AI GA LLC, COLFIN AI GA 2 LLC, COLFIN AI GA1
LLC, STARWOOD WAYPOINT TRS LLC, IH3 PROPERTY GEORGIA, BEAULY LLC, CSH 2016 2
BORROWER LLC, COLFIN AL GA 1 LLC, MORVEN LLC, COLFIN AI GA 1 LLC, 2018 2 IH
BORROWER LP, CSH 2016 1 BORROWER LLC, CAH 2014 2 BORROWER LLC, SWAY 2014-1
BORROWER LLC, CAH 2014 1 BORROWER LLC, IH3 PROPERTY GEORGIA L P, IH4
PROPERTY GEORGIA LP, THR GEORGIA LP, COLFIN AI GA I LLC, SWH 2017 1 BORROWER
LP

PROGRESS RESIDENTIAL

KEYWORDS: PROGRESS, FYR

ADDRESSES: 0 PO BOX 4090 85261, 3505 KOGER 30096, 3505 KOGER BLVD 30096, 5100
TAMARIND REEF 820

NAMES

TRUE NORTH TRS LLC, YAMASA CO LTD, RHA 1 TRS LLC, MJ RENTAL I LLC, PROPERTY
OWNER 6 LLC, TRUE NORTH BORROWER GEORGIA LLC, PROGRESS RESIDENTIAL
BORROWER 8 LLC, FYR SFR BORROWER LLC, PROPERTY OWNER 11 LLC, BEE SOCIAL

45
ADULT DAY PROGRAM CORP, PROPERTY OWNER 8, PROGRESS RESIDENTIAL BORROWER
11 LLC, PROPERTY OWNER 2 LLC, MILE HIGH BORROWER 1 CORE LLC, PROGRESS
RESIDENTIAL BORROWER 4 LLC, WHITEHAWK DELMAR LLC, PROGRESS RESIDENTIAL
BORROWER 3 LLC, HOME SFR BORROWER II LLC, HIDDEN MARKET PROPERTIES LLC,
PROGRESS RESIDENTIAL BORROWER 2 LLC, PMC SFR BORROWER LLC, PROGRESS
RESIDENTIAL BORROWER 17 LLC, WHITEHAWK DELMAN LLC, MILE HIGH BORROWER 1
CORE, ALMITAS GEORGIA RENTAL FUND LLC, GONSENHEIM IRREVOCABLE 1998 TR THE,
PROGRESS RESIDENTIAL BORROWER 18 LLC, HOME SFR BORROWER IV LLC, TRUE
NORTH PROPERTY OWNER A LLC, PROPERTY OWNER 8 LLC, PROGRESS ATLANTA LLC,
PROPERTY OWNER 5 LLC, RHA 1 LLC, RESI SFR SUB LLC, LYRA INVESTMENTS LLC, ARNS
INC, ROCKLYN HOMES INC, PROGRESS RESIDENTIAL 2014 1 BORROWER LLC, SFR
INVESTMENTS V BORROWER 1 LLC, HOME SFR BORROWER LLC, MILE HIGH BORROWER
1 INCOME LLC, RHA I LLC, GONSENHEIM IRREVOCABLE 1998 TRUST, PROGRESS
RESIDENTIAL BORROWER 10 LLC, PROPERTY OWNER 3 LLC, ARLP TRUST, PROGRESS
RESIDENTIAL BORROWER 15 LLC, YAMASA CO LTD A JAPANESE CORPORATION, SRC
STARS FUND LP, FREO PROGRESS LLC, OLYMPUS BORROWER 1 LLC, RESI TL1
BORROWER LLC, OLYMPUS BORROWER LLC, MIDWAY EXCHANGE BORROWER 8 LLC, P4
SFR PROPERTY OWNER 1 LLC, PROGRESS RESIDENTIAL 2016 1 BORROWER LLC,
PROGRESS RESIDENTIAL BORROWER 9 LLC, SFR INVESTMENTS V BORROWER I LLC,
PROPERTY OWNER 10 LLC, P5 2021 2 BORROWER LLC, RBTR 3 LLC, RNTR 3 LLC,
PROGRESS RESIDENTIAL BORROWER 13 LLC, RANTOR & RNTR 1 LLC, RESI REO SUB LLC,
RESI TL1 BORROWE LLC, PROPERTY OWNER 12 LLC, FYR SFE BORROWER LLC, PROGRESS
RESIDENTIAL BORROWER 7 LLC, ARLP REO 400 LLC, MILE HIGH BORROWER 1 VALUE
LLC, PMC SFR BORROWER 2 LLC, PROGRESS RESIDENTIAL BORROWER 5 LLC, PROGRESS
RESIDENTIAL BORROWER 14 LLC, PROGRESS RESIDENTIAL BORROWER 12 LLC, RNTR 1
LLC, PROGRESS RESIDENTIAL BRIDGE BORROWER LLC, PROGRESS RESIDENTIAL
BORROWER 6 LLC, PROPERTY OWNER 7 LLC, PRETIUM SFR HOLDING LLC, RESIDENTIAL
HOME BUYER ATLANTA LLC

SYLVAN REALTY

KEYWORDS: SYLVAN, RNTR

ADDRESSES: 3495 PIEDMONT 30305, 3505 KOGER 30096

NAMES

46
FF & G CAPITAL LLC, SFR ATLANTA LLC, HOME SFR BORROWER LLC, RHA 1 TRS LLC,
FUND IV HOMES 1 LLC, FYR SFR BORROWER LLC, RHA I LLC, MARDON CONSTRUCTION
INC, BEE SOCIAL ADULT DAY PROGRAM CORP, NCRC GROWTH SRC1 LLC, NCRC
GROWTH SRC 1 LLC, GONSENHEIM IRREVOCABLE 1998 TRUST, RNTR 2 LLC, SOUTHERN
STATE INVESTMENTS LLC, COOPER SANJA ET AL, MORGAN DAVID G, ATLANTA
NEIGHBORHOOD DEVELOPMENT PARTNER, SRC STARS FUND LP, RESI TL1 BORROWER
LLC, WHITEHAWK DELMAR LLC, BUTLER GLIDDEN COOPER LLC, CC ATL LLC,
DIVERSIFIED RESIDENTAL HOMES 1 LLC, VSP ATLANTA LLC, MIDWAY EXCHANGE
BORROWER 8 LLC, HOME SFR BORROWER II LLC, RNTR1 LLC, PEACH HOMES RENTAL
LLC, WHITEHAWK DELMAN LLC, RBTR 3 LLC, ALMITAS GEORGIA RENTAL FUND LLC,
RNTR 3 LLC, GONSENHEIM IRREVOCABLE 1998 TR THE, FRIEDMANS INVESTMENTS LLC,
RANTOR & RNTR 1 LLC, RESI REO SUB LLC, FSBBH LLC, RESI TL1 BORROWE LLC, FYR SFE
BORROWER LLC, ARLP REO 400 LLC, CC-ATL LLC, ATLANTA NEIGHBORHOOD
DEVELOPMENT, RNTR 1 LLC, CHO HOLDINGS LLC, RP HOMES 1 LLC, HILLS HOMES
ATLANTA LLC, RHA 1 LLC, RESI SFR SUB LLC, CITY LIVING DEVELOPMENT LLC, LYRA
INVESTMENTS LLC, DIVERSIFIED RESIDENTIAL HOMES 1 LLC, ROCKLYN HOMES INC,
SHV HOMES 4 LLC

TRICON RESIDENTIAL

KEYWORDS: TRICON, TAH

ADDRESSES: 1508 BROOKHOLLOW 92705

NAMES

TRICON SFR 2020 2 BORROWER LLC, SFR JV 1 2020 1 BORROWER LLC, SFR JV 1 2021 1
BORROWER LLC, TAH MS BORROWER LLC, SFR JV 1 PROPERTY LLC, TAH 2017 2
BORROWER LLC, 2016A PROPERTY OWNER LLC, TAH 2018 1 BORROWER LLC, TAH
PROPERTY HOLDINGS LLC, TAH HOLDING LP, TAH 2016 1 BORROWER LLC, 2015B
PROPERTY OWNER LLC, TAH MS2 BORROWER LLC, TAH 2017 1 BORROWER LLC

47