Understanding the Small and Medium Multifamily Housing Stock


Understanding the Small and Medium Multifamily Housing Stock

By Brian An, Raphael W. Bostic, Andrew Jakabovics, Anthony W. Orlando and Seva Rodnyansky


Residential buildings containing two to 49 units are a much more common and important source of housing in the United States than has generally been recognized, especially among those who rent. For many people, the terms "rental housing" and "apartments" mean large- scale, often high-rise, developments. The reality is that buildings with more than 50 units account for less than 10 percent of all rental units. Instead, the overlooked segment of the real estate mark which we term small and medium multifamily housing (SMMF), has served as a long-run provider of naturally occurring affordable housing.

Moreover, SMMF serves as an equilibrating force in the larger market, allowing people to continue living with a sustainable housing cost relative to income. Yet, for the past 25 years, construction of SMMF has declined, relative to historical trends, and old SMMF buildings are not being replaced with similar building types. Buildings in this segment are aging out, leading a decrease in physical quality.

This paper, based on our in-depth analyses of public and proprietary dataset provides an introduction to SMMF, highlighting four main features of these properties and three takeaways. We focus on SMMF’s prevalence in the overall housing stock, its affordability, its age, and its place in the rental and ownership stock. Because of SMMF’s critical role in providing affordable shelter, we suggest that policymakers support the development of financial tools to preserve existing SMMF as it ages and reduce barriers to production of new SMMF. We also identify unanswered questions and call upon the housing policy community to join us in further exploration of this critical segment of the housing stock.

Visit our Resource Center to download this report. Also available is a version of the paper that includes more extensive data about small- and medium-sized multifamily housing.

  1. SMMF's Prevalence in the Housing Stock

  2. SMMF's Affordability

  3. Age of the SMMF Stock

  4. Owner-Occupied SMMF Units

1: SMFF's Prevalence in the Housing Stock

Small and medium multifamily properties are ubiquitous throughout the United States, accounting for 21 percent of the national housing stock and providing homes to 22 percent of the total population. They are most commonly found in the central cities and suburbs of major metropolitan areas, where they contribute 34 and 22 percent, respectively, to the total stock of housing units. SMMF’s presence is most pronounced in the rental stock: SMMF makes up 54 percent of this housing segment.

Urban economic theory predicts higher density development, such as SMMF and "large"(50+-unit) multifamily buildings, in central cities than in suburbs or rural areas. Consistent with this prediction, we find that SMMF provides a higher share of all homes in central cities (34 percent) than in suburbs or rural areas, but even in the suburbs, SMMF accounts for more than 1 in 5 housing units. Overall, slightly less than half (48 percent) of all SMMF units are in central cities. Again, this speaks to the broad distribution of these homes across the landscape - more than one-third of all SMMF units are in the suburbs - and hints at why SMMF plays a critical role in offering affordable housing choices throughout the country.

Table 1. Percentage of Housing Stock by Building Size by Geography

Table 1
 Source: 2013 American Housing Survey 

Figure 1: Building Categories by Tenure Type

Figure 1
 Source: 2013 American Housing Survey 

SMMF is a crucial source of housing for renters, as indicated in Figure 1, which shows the distribution, as of 2013, of the national housing stock by building size and tenure. SMMF buildings contain a higher share (54 percent) of the rental housing stock than do single family houses (34 percent) or large multifamily buildings with 50 or more units (9 percent). As the next section of this paper shows, while rental affordability has largely been treated as a multifamily problem, our findings suggest affordability is a more nuanced issue and properly understanding it requires further segmenting of multifamily properties. Even within SMMF, distinctions may be helpful, as buildings with 2 - 19 units, which we consider to be smaller SMMF, account for 45 percent of the total rental housing stock. While the overwhelming majority of owner-occupied units are single-family (88 percent), with a modest share comprised of mobile homes, RVs, boats, etc. (7 percent), there are still many more owner occupants in 2 - 49-unit buildings than in large condominiums or cooperatives.

2: SMFF's Affordability

On average, SMMF provides homes to the lowest-income households. This is true for both units that are owned and rented, while single-family and large multifamily properties in both categories have higher-income residents. In particular, buildings with 2 - 19 units house the most low-income households. SMMF overall has an outsized concentration of subsidized units: Despite accounting for 21 percent of the national housing stock overall, it has more than
55 percent of all subsidized units.

Figure 2: Rent per Unit (weighted average out of total rental stock)

Figure 2
 Source: 2013 American Housing Survey 

Smaller SMMF forms the most affordable segment of the housing stock. Nationwide, the average rental unit’s monthly rent was $833, according to 2013 data. As shown by the distribution of rents in Figure 2, 40 - 49 and 50+-unit buildings had the most expensive average rents. Conversely, the most affordable rents are in buildings with 2 - 9 units. (On a per square foot basis, single-family homes are the least expensive of any building type, but the low cost per square foot is more than offset by the much larger size of the homes.)

Table 2: Household Income by Tenure Group by Building Types

Table 2
 Source: 2013 American Housing Survey 

Table 2 shows the average annual household income of those living in different building sizes. Income distribution by building size is basically U-shaped for both renters and owners, with the wealthiest households living in the 1- and 50+-unit buildings and the poorest living in SMMF. Within the SMMF category, structures with fewer than 10 rental units have the lowest-income residents.

Table 3: The Number and Percentage of Subsidized Housing by Building Type

Table 3
 Source: 2013 American Housing Survey 

Table 3 shows the number of housing units receiving government assistance in three mutually exclusive forms. Of these subsidized units, 56 percent are located in SMMF buildings, 26 percent in 1-unit buildings, and 18 percent in large 50+-unit buildings.

Table 4: Household Income Bands by Building Size

Table 4
 Source: 2013 American Housing Survey 

The median renter household income in 2013 was just shy of $33,0001. Table 4 shows that SMMF is a crucial source of housing for renters in the bottom half of the income distribution. Among renters with the lowest incomes - those making less than $10,000 annually - 60 percent live in SMMF. Similarly, the segment also houses the majority of renter households in the $10 - 25,000, $25 - 35,000 and $35 - 50,000 income bands. In contrast, SMMF units house a much smaller share of rental households making above $100,000 in annual income.

3: Age of the SMMF Stock

Because of the importance of SMMF for low- and moderate-income households, the long-term viability of the stock matters. It is therefore worrisome that the construction of SMMF has slowed in recent decades. While a boom in the construction of SMMF happened in 1970s and 1980s, its share of new construction has subsequently declined significantly, even as the retirement of older SMMF from the stock is accelerating. Consequently, SMMF is, on average, much older than large multifamily properties.

Figure 3: Percent of Existing Units Build by Decade by Building Category

Figure 3
 Source: 2013 American Housing Survey 

Figure 3 shows, by decade of construction, the distribution of existing single-family, SMMF and 50+-unit buildings. While the construction of existing SMMF and 50+-unit buildings rose over time until 1970s, starting in 1990, the new construction of SMMF has lagged significantly compared to that of single-unit and 50+-unit buildings. Of the existing units, SMMF accounted for more than a quarter of all units built in the 1970s and 1980s, but since 1990, it has only represented about 15 percent of new construction.

Figure 4: Distribution of SMMF Properties Built in Cook County, Ill., Over Time

Figure 4
 Source: DataQuick 

Turning from national survey data to a parcel-level dataset, we see that the national pattern holds true in Chicago. Figure 4 shows the number of existing SMMF properties in Cook County by year built. Similar to national trends, SMMF construction peaked in the mid-1920s and saw a modest resurgence in the 1960s and 1970s, but has substantially declined since 1980. As a result, many of the buildings from the height of the last SMMF construction boom are approaching 100 years old.

Figure 5: Comparison of Rents by Decade Built and Building Type (Weighted Average)

Figure 5
 Source: 2013 American Housing Survey 

As might be expected, we find that newer buildings command higher rents than older ones across the country. SMMF has the lowest rents compared to single-family or large buildings created in the same decade, going back as far as buildings built in the 1950s. A slight rent premium ($30 - 100 per month) exists for older SMMF buildings and for 50+-unit buildings (built in the 1930s or prior) relative to rents for the average-aged SMMF building. In contrast, older 1-unit rentals command significantly lower rents.

Table 5: Detailed Rents by Decadal Cohort (Weighted Average)

Table 5
 Source: 2013 American Housing Survey 

Generally, newer and larger buildings command higher rents. Many SMMF size categories follow a U-shaped rent pattern: older buildings (pre-1940) and newer buildings (1990s and later) have higher rents, while SMMF buildings 30 to 80 years of age are less expensive. This is broadly consistent with the filtering hypothesis, which posits that as buildings age, rents decrease, allowing them to transition to providing housing for lower-income households. It is likely that rents are high in very old buildings (older than 80 years) because of their historical or& architectural value and/or selective non-demolition. In other words, if very old buildings were in poor quality and lacked historical or other significance, they were more likely to be demolished and thus not show up in the data.

Table 6: Comparison of Percentage of Severely Inadequate Units by Building Size and Decade Built

Table 6
 Source: 2013 American Housing Survey 

Housing quality decreases with time for all three building categories, but multifamily housing shows a higher absolute degree of severe inadequacy than 1-unit buildings, as shown in Table 6. Single-family units have higher absolute quality on average than SMMF units in every time period. Over time, however, these single-family units decline in average quality at a faster rate than SMMF units. Viewed as a ratio of new vintages to old vintages, the proportion of severely inadequate new buildings (post-2000) to old buildings (pre-1919) is 1:5 for SMMF, and greater than 1:8 for 1-unit and 50+-unit buildings.

4: Owner-Occupied SMMF Units

4-owner-occupied.jpgMost of this analysis has focused on the important role that SMMF plays in the rental market. While SMMF accounts for only a small share (4 percent) of the overall owner-occupied housing stock, we do find interesting geographic concentrations of owner-occupied SMMF, with 2 – 4-unit buildings more common in the New England, Chicago and South Florida metro areas. South Florida is also an outlier with respect to the concentration of owner occupancy in 5 – 49-unit buildings.

Compared to rental units in similarly sized buildings, however, owner-occupied units have more rooms (usually by a half-bath, on average) and greater overall square footage. While rents per square foot tend to rise as buildings get larger, per-square-foot valuations for owner-occupied units do not follow the trend closely, with per-square-foot valuations for homes in 3 – 4-unit buildings roughly 20 percent below 2- and 5 – 9-unit buildings. Just as SMMF buildings tend to have lower-income renters than single-family or 50+-unit buildings, homeowners in SMMF buildings also have lower median incomes than single-family owners or owners of units in large buildings.

Key Policy Takeaways

The majority of America’s largest, most productive cities are experiencing rapidly rising rents, and the size of the population facing significant housing cost burdens is growing. Moreover, evidence indicates that the rise in cost-burdened households is not limited to coastal metropolitan areas or other "hot" markets. This paper indicates the extent to which SMMF matters for renters and homeowners. Because these buildings are the most affordable segment of the housing stock, they are playing, and will continue to play, a crucial role in providing reasonably priced housing. Policymakers must understand that role and begin to consider the best approaches to both preservation and production of SMMF. We must preserve and expand the affordable stock to meet current needs even as we encourage the market to produce more housing which, with time, will become affordable. Here, we present three policy implications of the trends in SMMF to start the much-needed conversation.

Takeaway #1: Production of SMMF Must Be Simplified and Expanded
Communities should have the flexibility to produce more SMMF units when needed. The current low rate of production suggests that the construction of affordable units is not keeping up with the demand of low-income renters. While there are many reasons why SMMF is not built in particular places, policymakers should not stand in the way of building this type of housing when it can help communities, and they should offer incentives for more production to overcome market failures that are clearly occurring in many areas. Streamlining regulations to expand the supply of smaller buildings can improve longer-term affordability. SMMF properties have exhibited a greater likelihood of becoming more affordable as they age than other building categories, so the relative lack of new construction over the past few decades may indicate future challenges to the supply of market-rate affordable units.

Takeaway #2: New Tools Must Be Created to Encourage Preservation and Financing of SMMF
Again, given the production gap, preserving existing SMMF is critical to broader affordability in the rental market. The loss of this crucial stock due to age and deterioration will increase pressure on rents and displace lower-income families. Communities should be given the necessary capital and financing tools to rehabilitate and preserve SMMF when they determine such units are the best way to maintain affordable housing options. Due to the age of this segment of the housing stock, preservation requires investment, but existing financing tools don’t easily work for smaller properties seeking to retain more affordable rents. Developing and funding tools that encourage preservation of affordable SMMF is critical to protecting these underappreciated sources of affordable housing.

Takeaway #3: Additional Research on SMMF Is Needed
We need more research on SMMF housing. Considering the importance of the segment of the housing stock we documented above, we remain surprised that researchers and policymakers have focused so little on the SMMF segment. Many questions remain. Where are these buildings located within metro areas? Why has their production declined in recent decades? What is the best way for communities to leverage their affordability to foster efficient and equitable market outcomes? We will continue studying these issues, and we encourage the rest of the housing policy community to join us in this pursuit. As we begin to answer these questions, we hope to be able to identify opportunities to preserve and expand the supply of SMMF buildings that offer greater affordability than other building types, both after they are newly built, and also over time.

These takeaways are based on our working paper "Small and Medium Multifamily Housing Units: Affordability, Distribution, and Trends." Please see this paper for more information on the data sources and analysis methodology. It is available in the Enterprise Resource Center.

About the Authors

Brian Y. An is a doctoral candidate at the USC Price School of Public Policy. His research examines urban and metropolitan governance and policies with a special emphasis on local government finance and structures.

Dr. Raphael W. Bostic is the Judith and John Bedrosian Chair in Governance and the Public Enterprise at the Sol Price School of Public Policy, University of Southern California.

Andrew Jakabovics is Vice President, Policy Development at Enterprise Community Partners. He researches the supply and demand of affordable housing, among other public policy issues related to housing affordability and community development.

Anthony W. Orlando is a PhD candidate in the Sol Price School of Public Policy at the University of Southern California. His research is primarily focused on risks, returns, and regulations in real estate.

Seva Rodnyansky is a PhD Candidate in Urban Planning and Development at the University of Southern California (USC) Price School of Public Policy. His research focuses on housing supply, transit-oriented development, mobility, and neighborhood change.


This research was funded in part by a generous grant from the JPMorgan Chase Foundation, and we gratefully acknowledge their support, as well as the University of Southern California’s Bedrosian Center on Governance and the Public Enterprise. All errors are ours alone.

About Enterprise Community Partners

Enterprise works with partners nationwide to build opportunity. We create and advocate for affordable homes in thriving communities linked to good schools, jobs, transit and health care. We lend funds, finance development and manage and build affordable housing, while shaping new strategies, solutions and policy. Over more than 30 years, Enterprise has created nearly 358,000 homes, invested $23.4 billion and touched millions of lives. 

About the USC Bedrosian Center on Governance

The USC Bedrosian Center on Governance, an applied research center at the USC Sol Price School of Public Policy, is dedicated to understanding and fostering effective democratic governance as an essential component in ensuring the betterment of communities. Visit us at bedrosian.usc.edu.

About the USC Sol Price School of Public Policy

The mission of the USC Sol Price School of Public Policy is to improve the quality of life for people and their communities, here and abroad. We achieve this mission through education and research that promote innovative solutions to the most critical issues facing society. As the 21st century unfolds, strands and textures of society intersect in unexpected ways. Much of this occurs across booming urban populations — in America and around the world — giving rise to new challenges or further complicating those that have not been adequately solved. Over the past 85 years, the USC Sol Price School of Public Policy has earned the public trust by creating path-breaking research and scholarship. We believe it is our ethical responsibility to continue serving our global community at the highest level of excellence. Visit us at priceschool.usc.edu.