If you have trouble reading this email, please go to the online version.
June 27, 2018

Did We Ever Build Apartments for the Middle Market?

It is well understood, within the apartment industry at least, that in most parts of the country it is not feasible to build new market-rate apartments with rents affordable to median-income households. As a result, new apartment construction is heavily skewed toward the high-end of the rent spectrum, except for those built using the Low-Income Housing Tax Credit.

Less well understood is whether it has ever been different. While the historical data is insufficient to provide a comprehensive answer, it is possible to shed some light on the issue. The data suggest that this hasn’t always been the case and that low levels of income growth have exacerbated the challenge. 

Data Limitations

Ideally to answer that question, we would like to have data on the rents for all new apartment completions, the range of rents in the markets in which those apartments are delivered and the income distribution of renter households (or all households). The best current source with those data for all metropolitan areas is the American Community Survey (ACS). Unfortunately, reliable data from the ACS goes back only to 2006.
The American Housing Survey (AHS) goes back much farther and has much useful information, but its survey sample is much smaller and the biannual national data do not include metro area-specific information. In addition, there are differences in what data were collected over the years which adds complications in using the data as a time series.

Even with these limitations, the AHS data provide some useful perspective and suggest that it may have been somewhat more feasible to build market-rate workforce apartments in past decades.

Asking Rents on New Apartments 

The Census Bureau has also tracked the distribution of asking rents on new apartments (apartment homes in buildings with at least five units) since 1970 in its Survey of Market Absorption (SOMA). 

The rent categories changed over time (and the microdata are not available), making this all but useless for gauging long-term changes, but median asking rent is available throughout the period, and that is of value.

Figure 1 compares the increase in median asking rent on new apartments with the overall increase in the consumer price index (CPI). Between 1970-2012, the periods in which the median rent grew more slowly than overall inflation were balanced by periods in which it grew faster. Since then, the median asking rent has risen much more rapidly than inflation.

While helpful, comparing rent changes to overall inflation doesn’t tell us much about how many people can afford these new rents. Figure 2 adds an income component, namely median household income—for all households and for renter households.

As the first wave of baby boomers entered the housing market in the 1970s, the median asking rent on new apartments fell relative to the median income of all households. In 1979, the median rent was just 20 percent of median household income. By 2016 this ratio had risen to 30 percent. Looking at the median income of just renter households (available only since 1979), this ratio reached a high of 46 percent in 2014 and still stood at 43 percent in 2016 (most recent data available), higher than at any time prior to 2013.

These data suggest that new apartments have gotten more skewed toward the higher end over recent decades. Looking only at medians is of limited value, however. More helpful is using microdata to assess affordability across all new apartment rents and across all households.

Rents on New Apartments and the Existing Stock

Even without sufficiently granular geographic information, the AHS data can bring the picture into sharper focus. In the analysis that follows, we looked at apartments built within the most recent five years (“new apartments”) and compared rents on those homes to rents of apartments built earlier.

Figure 3 shows the share of new apartments that have rents within the four quartiles, except that the top quartile (the top 25 percent of rents) is broken into two parts: rents in the 75th-89th percentiles, and rents at the 90th percentile or above.

Not surprisingly, the percentage of new apartment rents below the 25th percentile of rents has been relatively low. The share averaged just 21 percent from 1979-1985, a time of reduced construction as a result of recession. Since then, this share has declined to an average of 11 percent, even with the introduction of the Low-Income Housing Tax Credit in 1986. 
At the other end of the rent distribution, the share of new apartments with rents at or above the 90th percentile has averaged 21 percent over the entire period. It has been a little higher recently, but not as high as it was in the late 1980s and early 1990s. Similarly, the share of new apartment rents in the 75th-89th percentile range was highest between 1987-2001.
These tabulations suggest that while it may harder than ever to deliver apartments affordable to low-income households, delivering middle market apartments may be no more difficult that in the past. 

Our last tabulation adds household income back into the equation—and income may be a key part of the problem.

The share of all apartments that are affordable to median-income households and were no more than five years old ranged from about 10-20 percent in most of the 1970s and 1980s. It fell to an average of 3.1 percent in the 1990s and 2000s and has slipped even further to only 1.8 percent in the current decade (data through 2013). 


Putting this all together, these data suggest two conclusions. First, apartments delivered in the last decade or two have rents that are only somewhat more skewed toward the high end than the rents on the existing apartment stock. 

This jibes with the data showing that rents on newly constructed apartments have, until very recently, grown at a rate very much in line with the overall inflation rate. Second, therefore, the increasing difficulty in affording new apartments is mainly due to the fact that income growth has not kept pace—with either apartment rents or prices generally.

The key caveat to this analysis is that these data are national only.  They do not capture the variation in construction costs, rents and incomes by metro area.

About Research Notes
Published quarterly, Research Notes offers exclusive, in-depth analysis from NMHC's research team on topics of special interest to apartment industry professionals, from the demographics behind apartment demand to effect of changing economic conditions on the multifamily industry.
Questions or comments on Research Notes should be directed to Mark Obrinsky, NMHC's Senior Vice President of Research and Chief Economist, at mobrinsky@nmhc.org or 202/974-2329.
Print-Friendly Version RSS Search Back Issues
Contact Us
1775 Eye Street, N.W., Suite 1100,
Washington, D.C. 20006
202 974 2300 Office
Manage My Email / Unsubscribe
Please add ‘news@nmhc.org’ to your email address book or Safe Sender List. If you are still having problems receiving our communications, see our White List page for more details.