The curious connection between DC population growth and new Class A residential buildings

The new buildings are clearly related to DC’s population growth, yet population growth slowed as building increased

According to the US Census Bureau, DC’s population started growing in 2006, and in the 13 years from 2005 to 2018 DC added 135,319 people, a 23.9% increase. Over that same time, according to CoStar, there was a net increase of 303 multi-family residential buildings containing 40,388 multifamily housing units, a 27.6% increase.

 

table 1

 

Of the increase in multi-family housing, 74.3% of the buildings and 84.3% of the units occurred in what CoStar classifies as Class A buildings. These are newer, well-located apartments and condominiums with modern amenities. About 80% of the Class A units are apartments (see the appendix for more detail). Over the 13 years from 2005 to 2018 there was a 446% increase in units in Class A buildings compared to a 4.6% net increase in all other units.

 

graph 1

 

That there has been a significant increase in multi-family buildings and units is not surprising to anyone familiar with all the cranes that have dotted the city for more than a decade. What may be surprising, however, is the extent to which a relatively large proportion of the residential construction, especially that of Class A units, lagged population growth and occurred as population growth was actually slowing. The proportion of the entire 2005 to 2018 population growth that had occurred by 2013 was much greater than proportion of new housing units that had occurred by that date:

  • By 2013 DC had added 60.8% of the population growth that occurred from 2005 to 2018, but only 44.2% of the net increase in all multi-family housing units over the entire period—and only 39.0% of those in Class A structures.
  • From 2013 to 2018 DC added 38.4% of the population growth that occurred from 2005 to 2018, but a much greater share (55.8%) of the net increase in all housing units in multifamily structures occurred then. The share of Class A structures added after 2013 was 61%.
  • In 2013, the year of the largest annual gain in population, DC added 15,706 people and 2,078 net new multi-family housing units. In 2018 population growth was less than half (6,764) of what it was 5 years earlier while the net increase in housing units in multifamily buldings was more than twice as much ( 4,408).

 

table 2.PNG

 

graph 2

 

From the CoStar data on housing it is not possible to explain all of the dynamics that link population changes in DC to housing market developments. Clearly until 2013 most of the increase in population did not find housing in new Class A buildings. As we approach 2018, however, a much higher percentage of the growth could be housed in such units.  In 2013 the ratio of population growth to net increase in multifamily units of all classes was 7.6.  In 2018 that ratio has fallen to l.5. This means that the entire net increase in population in 2018 could have been housed in new (mostly Class A) housing if the average household size was 1.5.

 

Going forward, an interesting question is how much of a limiting factor the availability of new Class A housing may to population growth in DC.  There are quite a number of factors at play and so it is difficult to draw a firm conclusion here.  For example, not all Class A units that are occupied (or rented) are necessarily occupied or rented by residents who would be counted by Census as part of DC’s population. This could involve persons whose primary residence is in another state, units owned for temporary housing of corporate personnel, or units owned for short term rentals. Also if persons now sharing units in DC move to newly constructed ones to live by themselves, occupied units would increase without any increase in population. The full story linking population and housing market changes must, of course, take account of all housing units in the city, not just those in Class A units in multi-family structures.

 

As noted above, most of the Class A buildings are apartments. More details on Class A residential buildings are contained in the appendix.

About the data: 

This is the second of two blogs dealing with  population dynamics in the District of Columbia in relation to the latest  US Census Bureau estimates of DC population.

The population information reported here is from the DC population tables released in December 2018 by the US Bureau of the Census in connection with population estimates for the 50 states and the District of Columbia as of July 1, 2018.  The data include revisions to the years 2010 through 2017, and all information is subject to further revision next year.

Housing data is from CoStar, a real estate information firm that tracks all private sector apartment and condominium housing units in multifamily buildings with 5 or more units. Information for each year is for the second quarter, which corresponds closely to the July 1 date used by the Census Bureau for estimating annual population numbers. CoStar data is continuously updated and revised as more information becomes available.

A version of this blog appeared in the January/February District of Columbia Economic and Revenue Trends report, issued by the DC Office of Revenue Analysis.

Appendix

table 3

graph 3

graph 4.PNG

 

graph 5

 

graph 6

DC’s population topped 700,000 in 2018, but last year also saw the slowest annual increase in a decade

Natural increase, not net in-migration, has become the main source of DC’s population growth.

The US Bureau of the Census estimates DC’s population on July 1, 2018 was 702,455, an increase of 6,764 (1.0%) from the revised estimate for 2017. This is notable for several reasons:

  • 2018 is the 13th straight year of population growth. From 2005 to 2018 population grew by 135,319, a gain of 23.9%.
  • 2018 is also the slowest population growth in a decade. From 2008 to 2018 DC’s population grew an annual average of 12,222. Growth in 2018 was just 55.3% of the
  • decade’s average.
    graph 2

 

graph 1

 

  • 60.7% of the net increase in population from 2017 to 2018 was accounted for by natural increase of 4,104. (Natural increase is births minus deaths.) The rest of the net change was from international migration. For migration within the US, 936 more people left DC than moved here.

 

graph 3

 

  • DC’s population was larger than that of 2 states (Vermont and Wyoming), and last year’s growth was more than in 16 states (9 of which actually lost population).
  • In percentage terms, DC’s rate of growth was well above the US (0.6%) rate and above that for 36 states.

Additional details on recent growth, revisions to past years, population change since 2000, and comparison with the 50 states are included below.

 

Revision to the 2017 estimate. The July 1, 2018 DC population estimate of 702,455 is actually 8,463 higher than last year’s 2017 estimate. However, the increase over 2017 is now estimated to be 6,764 because revisions to population estimates in earlier years have resulted in a new estimate for 2017 that was 1,719 higher.   The current estimate for 2017, 695,691, is 1,719 less than the prior one (693,972).

As shown in the accompanying table and graph, the revision to prior years involved all the years from 2010 to 2017. The higher estimate for 2017 reflects the cumulative impact of increases and decreases since 2010. Although the level in 2017 was higher, growth in 2017 over 2016 was actually lower (by 520 persons).

table 1.PNG

graph 4

Components of population change last year compared to the 8 1/4 years since the April 1, 2010 Census. The 6,764 increase in DC’s population from 2017 to 2018 was just a little over half (55.4%) of the 12,205 annual average since the April 2010 decennial census count. Several features stand out:

  • Natural increase has become the major contributor to DC’s net population growth. This past year natural increase accounted for 60.6% of the increase; its average contribution since 2010 was 37.4%. Although births and deaths last year both exceeded their annual averages since the census, natural increase was less than the post-census average because deaths increased more than births.
  • Domestic migration is no longer a source of net growth for DC. Net domestic migration contributed 29.3% to growth since the census, but now more people are leaving DC for other parts of the US than are arriving from them.
  • International migration has become a more important source of growth for DC. It was the source of 32.0% of the net increase in DC’s population in the years since the census, and 53.1% of the growth last year.

table 2.PNG

 

Comparison with the 50 states. In 2018 DC’s population was greater than that of Wyoming and Vermont. (The next closest states to DC are Alaska (737,438) and North Dakota (760,077)), Also, from 2017 to 2018:

  • DC’s population increase exceeded that in 16 states (9 of which lost population).
  • DC’s natural increase was greater than in 11 states (2 of which were negative).
  • DC’s net domestic migration was less negative than in 26 states.
  • DC’s net international migration was greater than in 18 states.

 

table 3

Details on comparisons with the states for 2018 and for changes from 2017 to 2018 are shown in the appendix.

About the data: This is the first of two blogs dealing with aspects of population growth in the District of Columbia.

The information reported here is from the tables released in December 2018 by the US Bureau of the Census in connection with population estimates for the 50 states and the District of Columbia as of July 1, 2018. Those tables include (1) total population; (2) population as of April 1, 2010 in the decennial census and as of July 1 of each year from 2010 through 2018; (3) components of population change from July 1, 2017, to July 1,2018; and (4) components of population change from April 1, 2010, to July 1, 2018. The components of change are natural increase (with births and deaths shown separately) and net migration (with international and domestic migration shown separately. The data include revisions to the years 2010 through 2017.

A version of this blog appeared in the December 2018 District of Columbia Economic and Revenue Trends, issued by the D.C. Office of Revenue Analysis.

Appendix

graph 5.PNG

 

 

The District of Columbia Reforms its DCEITC for Childless Workers

In 2015, the city replaced its local Earned Income Tax Credit (DCEITC) policy for childless workers as a fixed percentage of the Federal Earned Income Tax Credit (EITC) with a formula that appreciably increased both the local credit amounts and maximum eligible income. (See full policy brief here.) In this new formula, the city applies the federal maximum credit amount to a much larger income range ($6,600 to $18,111) and the federal phase-out credit rate to higher income levels ($18,111 to $24,040). As shown in Figure 1, the DCEITC encompasses all the city’s childless workers eligible for the Federal EITC plus a substantially larger number of workers who earned between $14,800 and $24,040 in annual wage income in 2015.

Fig1

In 2014, 53,839 tax filers received the EITC and DCEITC. These were comprised of 41,391 filers with qualified children and 12,448 without qualified children (Figure 2).  With the expansion of the DCEITC for childless workers in 2015, there was a 26.8 percent increase in total DCEITC claimants with nearly all the increase being attributed to 12,490 new childless workers (2,983 who were also eligible for the Federal EITC and 9,507 who were still not eligible for the Federal EITC and not previously eligible for the DCEITC).

Fig2

Structure of the Childless Worker EITC: DC v. Federal

A comparison of the structure of the expanded DC childless worker EITC with the federal credit focuses on the five income ranges. In Figure 3, the first income range (wage and salary income less than $6,600), labelled Income Range 1, is the phase-in range for both the federal and District credit, where both the federal and local credits increase by 7.65 cents for every additional dollar of earnings. The second income range (income between $6,600 and $8,250), Income Range 2, is the plateau of the EITC structure, where the credit is $503 for both the federal and local credits, regardless of earnings. Income Range 3 (income between $8,250, and $14,800) is the phaseout range for the federal credit, where the federal credit is reduced by 7.65 cents per dollar of earnings, while the District credit remains at $503 regardless of earnings.  For Income Range 4 (income between $14,800 and $18,111) each childless worker receives a DC credit of $503 regardless of earnings but receives no federal credit. And for Income Range 5 (income between $18,111 and $24,040), a DC childless worker receives a DC credit that decreases by 7.65 cents for every additional dollar of earnings up to an income level of $24,040. Beyond $24,040 in income, a resident is ineligible for the DCEITC childless worker credit.

Fig3

The DCEITC Impact on Low-income Households

Table 1 shows how the DC childless worker credit varies across the identified income ranges and compares the DC childless worker credit to that of the federal’s for Tax Year 2015, the first year the new DC credit was in effect. In 2015, there were 2,983 new claimants who were eligible for both the Federal EITC and the DCEITC (Income Range 3), and 9,507 new DCEITC claimants who were not eligible for the Federal EITC (Income Ranges 4 and 5). Tabl1_After Fig3

 

In tax year 2015, the District expanded the eligibility for its local credit for childless workers by extending the credit to childless workers with income from $14,800 to $24,040; it also increased the local credit amounts. The maximum Federal EITC for a childless worker in 2014 was $496, and the maximum local credit amount a childless worker could receive was $198 (Figure 4). Under the 2014 policy, childless workers in a narrow range of income ($6,600 to $8,150), making up 12.8 percent of all DCEITC childless workers, claimed a combined maximum amount of $694. Workers that earned more than $14,550 that year were not eligible for either the federal nor local credit.

Fig4

Under the new 2015 policy, 12,632 claimants (50.7 percent of all DCEITC childless workers) that earned between $6,600 to $18,111 in income received a DCEITC of a maximum amount of $503 (Figure 5). And further, claimants with income between the narrow range of $6,600 and $8,250 received a combined EITC and DCEITC of $1,006. Stated differently, childless workers with income less than $8,250 received a DCEITC that was a 100 percent match of their federal credit, and claimants with income between $8,250 and $14,800 received a DCEITC that was an average of 192.3 percent more than their Federal EITC.

Fig5

The DCEITC in a Changing District

The expanded DCEITC for childless workers provides additional income support and security for the city’s lowest income earners, encourages labor market participation, and helps to facilitate poverty alleviation. But, it may also be incentivizing continued city residency for a growing number of the city’s lowest income earners through higher refundable tax credits that could be used to help counter the rapidly rising costs of living in the city.

 

Are growth of the labor force and resident jobs slowing in DC? Maybe—and maybe not.

Seasonally-adjusted and unadjusted data from BLS currently tell different stories

Each month the US Bureau of Labor Statistics (BLS) estimates labor market statistics for all states and the District of Columbia. Labor market statistics include the labor force, resident employment, unemployment, and the unemployment rate. The data is reported on both a seasonally adjusted and unadjusted basis. Seasonal adjustment takes account of recurrent events during a year such as holiday employment that can mask trends in the data.

Typically, comparing data from the same month of the prior year eliminates the need for seasonal adjustment. Accordingly, it would be expected that there should be little difference between seasonally adjusted and unadjusted estimates of the annual change in DC resident employment from September 2017 to September 2018. Currently, however, the two data sets give very different pictures of the change over this time, leaving unanswered the question as to whether DC’s resident employment is or is not slowing significantly.

  • The seasonally-unadjusted data say that from September 2017 to September 2018 resident employment (measured by the 3-month moving average) increased by 3,870 (1.0%). By contrast, the seasonally-adjusted data peg the increase at twice that (8,067, a 2.1% gain).
  • The unadjusted data show quite a sharp decline in the amount of year-over- year growth since May 2018, while the adjusted data show an increase.
  • The unadjusted data peg growth over the past year at about half the annual average increase over the past 5 years. Seasonally adjusted, the growth is very close to the average of the past 5 years.          

Details are shown in the tables and charts in the appendix. As indicated there, the story is similar for the seasonally adjusted and unadjusted estimates of DC’s labor force.

The current difference between the one year change in the seasonally adjusted and unadjusted resident employment and labor force data is an unusually clear example of the difficulty in spotting changes in the economy by closely monitoring data as it is released each month or each quarter. As with Personal Income, population, and other data produced by federal agencies, labor market data is revised as more information becomes available.

As the labor force data is revised the current differences in the story about changes over the past year told by the seasonally-adjusted and unadjusted data will be resolved. However, it will likely not be until March 2019 when major annual revisions typically occur that the matter will be cleared up.

It should be noted, however, that the seasonally-adjusted and unadjusted data both tell the same story about unemployment: the amount and rate of unemployment fell over the past year.

Appendix

 

 

 

 

 

 

 

 

 

About the data. The labor market information is from the statistics released each month for the District of Columbia (along with all states) based on a population survey. The data include resident employment (persons over 16 years of age who say they are working on a full or part time basis); unemployment (persons over 16 years of age who are not working but say they are looking for work); labor force (the sum of resident employment and unemployment); and the unemployment rate (unemployment as a percentage of the labor force).

The data are reported on both a seasonally adjusted and not seasonally adjusted basis. For the month of September 2018 the data reflect the revisions which were part of the October 2018 release. The annual comprehensive revision to the data will occur in March 2019. All calculations here are based on 3-month moving averages (e.g., September 2018 is the average of July, August, and September as reported by BLS).

Seasonal adjustment is a statistical method for removing the seasonal component of a time series that exhibits a seasonal pattern. It is usually done when wanting to analyze the trend of a time series during a year independently of the seasonal components. It is common, for example,  to report seasonally-adjusted data for unemployment rates to reveal the underlying trends in labor markets.

An earlier version of this blog was included in the October 2018 District of Columbia Economic and Revenue Trends report issued by the Office of Revenue Analysis of the District of Columbia Office of the Chief Financial Officer.

 

 

Estimating the Effects of the Tax Cuts and Jobs Act (TCJA) on the City’s 2018 Income Tax Burdens

This is an analysis of the effects of the TCJA on individual income tax liabilities of District of Columbia residents. We estimate that under the new legislation, and in the aggregate, income taxes paid to the federal government for 2018 will be lower but income taxes paid to the District of Columbia government will be higher with the onus of the increased local income tax burden being borne largely by the city’s low to middle-income tax payers. (See here for a more in-depth statistical analysis of the TCJA, and see here for an extended discussion of the results discussed in this blog.)

The TCJA will affect the income tax liability of different District residents in various ways, depending on a tax filers circumstances, such as income, marital status, number of children, etc.  Using the Office of Revenue Analysis Income Tax Microsimulation Model, we estimate that for tax year 2018 (for which taxes are due April 15, 2019) the TCJA will cause city residents to pay $617.3 million less in federal income taxes but $56.4 million more in city income taxes. Figure 1 shows an estimated 76 percent of city filers will see a decrease in their federal tax liability, and figure 2 shows that 43 percent of the same city filers will see an increase in their District income tax liability.

Fig 1Figure 2_New

Distributional Effects of the TCJA

Of the estimated 76 percent that are likely to see reduced federal taxes in 2018, an estimated 40 percent of the reduced tax liability will go to residents of Wards 2 and 3. The remaining 60 percent of the reduced tax liability will go to residents of the remaining 6 wards (figure 3). This 60 percent is estimated to be further distributed in dollar terms as follows: 14 percent to filers with income less than $50,000, 19 percent to filers with income between $50,000 and $100,000 and 27 percent to those with income greater than $100,000.

Of the estimated 15 percent that will likely see higher federal taxes (from figure 1), 64 percent of the higher tax burden will be paid by residents of Wards 2 and 3, and an additional 22 percent will be paid by residents of the remaining 6 wards whose income is greater than $100,000 (figure 4).

Figure 3Figure 4

The above figures indicate that, in the aggregate, residents of Wards 2 and 3 will be significantly impacted. But the ward distribution of the impact of the TCJA on the city’s federal income tax liability largely reflects the District’s distribution of income by wards. Though each ward has an average of about 13 percent of the city’s income tax filers, Wards 2 and 3 account for 46 percent of the city’s annual income and 51 percent of income tax liability (as shown in figure 5).

Figure 5

Turning to the distributional impact of the TCJA changes on the District’s income tax liability, of the 39 percent tax filers (figure 2) with decreased District income tax liability, 56 percent are in Wards 2 and 3 (figure 6).   Of the tax filers outside of Wards 2 and 3 with lower tax liability, an estimated 14 percent have income less than $50,000 (figure 6). On the other hand, Wards 2 and 3 will only shoulder 29 percent of the increased District income tax liability and residents in the other wards and with income less than $50,000 will shoulder 18 percent of the increased local tax burden.

Figure 6Figure 7

Why did TCJA lower District resident’s federal tax liability, but raise their District income tax liability?

The main reason for differential impact of TCJA on federal versus the District income tax liability is the TCJA’s suspension of the personal exemption (for tax years 2018-2025). As figure 8 shows, Wards 2 and 3 only accounted for 19 percent of all personal exemptions in tax year 2015, while filers with income less than $50,000 who lived in wards outside Wards 2 and 3, accounted for nearly 50 percent of all District personal exemptions. On average, head of household tax filers (unmarried tax filers with dependents) claimed 3.62 exemptions on their tax returns in tax year 2015 (figure 9). Figure 10 shows that Wards 4, 7 and 8, have the highest share of dependents, accounting for 53 percent of the District’s dependents.  As such, tax filers in these wards will likely be most negatively impacted by the loss of the personal exemption with overall higher District income tax liability because of the TCJA changes, even with the almost doubling of the standard deduction. Our analysis suggests that the lack of a child tax credit like that in the federal tax structure is one of the major causes for expected higher District income tax burdens for lower income filers with dependents.

figure 8

fig 10

 

Combined Federal and DC Effects of the TCJA

Our analysis estimates that tax filers with income between $25,000 and $75,000 will experience an average increase between $155 and $173 in their 2018 District of Columbia income taxes. This translates into about a 0.2 percentage point increase in their average effective tax rate. In contrast, it is anticipated that filers with income greater than $500,000 will experience a decrease in their District of Columbia income taxes of roughly $1,000, and this translates into practically no change in their average effective tax rate.

When we consider the effect of the TCJA on the combined federal and local income taxes on District residents’ tax liability, it is estimated that only 17 percent of filers will see a net increase in their combined income tax liability and 64 percent of filers will see a net decrease in their combined income taxes. This suggests that for many filers the expected decrease in their federal income taxes is likely to offset the expected increase in their city income taxes for 2018.

Conclusion

At the federal level, it appears that a combination of the higher standard deduction and the higher child credit allows eligible low-income earners with dependents to lower their federal income liability despite the elimination of personal exemptions. The absence of such a credit at the District level appears to be a principle reason why many low-income earners with dependents will pay more income taxes to the District of Columbia government in 2018 under the new TCJA.

We estimate that the TCJA will make the District’s income tax structure less progressive than it otherwise would have been in 2018. That is, beginning in 2018 the District of Columbia effective income tax rates for many low income filers with dependents are estimated to increase slightly more, on average, than the effective income tax rates for city’s highest income filers.

In the last 5 years DC added 180 new apartment and condominium buildings with 22,348 units

The new units are essential to DC’s ability to absorb a growing population, but they are not the whole story

From 2012 to 2017 DC’s inventory of apartment and condominium housing units grew by a net of 22,348 according to CoStar, a real estate information firm that tracks developments in the District and many other locations. This 13.7% increase in housing units is a major element in the growth of DC’s population over that time, a relationship that will be looked at shortly. But first, a few details on the recent changes to the District’s stock of apartment and condominium units.

graph 1

table 1

  • Of the 22,324 net increase in units, 20,128 (90.1%) were in 95 apartments and 2,196 were in 84 condominiums. The average size of condominium buildings was much smaller, 26 compared to 212 for apartments. Condominiums therefore accounted for 47% of the buildings, but just 10% of the units. (In 2017, condominiums accounted for about 19% of the combined total multifamily units— condominium, apartment, and co-op—in the city.)
  • The number of new units delivered over the five years was 23,099, but 775 units, about one-half percent of DC’s stock of multifamily housing in 2012, were demolished or otherwise went out of existence.
  • The number of vacant apartment units increased by 2,374 over the past 5 years, and the vacancy rate rose from 6.5% in 2012 to 7.3% in 2017.
  • New construction at the end of 2017—11,179 apartment units and 1,614 condominium units—essentially continues the pace of recent development activity. It will take two or more years for all of this existing construction to deliver new units to inventory, and the amount of this new construction is more than half of the net increase in units that occurred over the previous 5-year period of 2012 to 2017. For apartments current construction is 56% of the prior 5 year net increase in units, and for condominiums the percentage is 74%.

Detail for each of these points are in the tables in Appendix 1.

Apartments, condominiums, and DC population dynamics

The number of occupied apartment and condominium units grew by 19,930 from 2012 to 2017. This is almost 2,400 less than the 22,348 net increase in inventory, but in percentage terms the gain in occupied units—12.9%—was greater than that of population over that time (9.2%).

table 2

table 3

Newly occupied apartment and condominium units clearly represent a major element in the demographic changes occurring in DC. Available data does not, however, make it possible to know exactly how much of the 58,342 growth in DC’s population from 2012 to 2017 was accommodated by the net increase in occupied apartments and condominiums described in the CoStar data. A judgment on this depends on the additional number of households associated with the increase in population—and this can only be estimated.

Definitions further complicate the task of trying to compare Census and CoStar information. For the Census Bureau, a household is equivalent to a housing unit occupied by DC residents. The resident household can be a single person, a small family, a large family, or a group of unrelated persons who share the unit. To be classified as occupied by CoStar, however, a unit need not only be occupied by a DC resident household (as defined by Census). Occupied units from the point of view of a property owner can also involve second homes, short-term rentals, corporate accommodations for employees and guests, units in transition waiting for a new owner or tenant to move in, or ones undergoing repairs. The growth in occupied units as defined by CoStar can thus easily exceed the increase in households as defined by Census.

That said, a reasonable place to start in connecting population increase to change in apartment and condominium occupancy is with the Census. In 2000 the ratio of population to households was 2.30, a ratio which fell to 2.27 in 2010. In that decade, the number of households grew faster than population (7.4% compared to 5.8%), and the ratio of the increase in population to the increase in households was only 1.80, suggesting that one of the features of population growth at that time was small households.

What of the period between 2012 and 2017? Was the increase in small households so great that the number of households grew faster than population, and the population/household ratio continued to get smaller? Or did population grow faster, resulting in a rise in the ratio? Among the factors contributing to this latter result would be more children and more people doubling up due to affordability issues.

Assumptions about an appropriate population/household ratio makes a big difference in assessing how many new households population growth has brought—and how much of that growth was accommodated by occupancy growth in apartment and condominium buildings. As noted above, the 2010 average population/household ratio was 2.27. That ratio applied to population growth over the 2012 to 2017 period yields an estimated growth in households of 25,701, a number that exceeds CoStar’s estimate of the increase in occupied units by 5,771.

The 2012-2016 American Community Survey 5 year estimate pegs the population/household ratio at 2.38. Applying that higher ratio gives 4,583 as the amount by which household growth exceeded the increase in occupied units. With ratios a little higher or lower than these averages new household growth exceeds the growth in occupied units. Only if the population/household ratio rises to 2.93 does the new household estimate equal that for the growth in occupied units. Under a range of plausible assumptions it appears likely that the District of Columbia added more households over the past five years than could live in newly occupied units in new apartment and condominium buildings. If this is the case, this means that other components of the housing stock—structures with four or less units—must have been able to absorb some of the increase in population.

table 5

Taken as a whole, the DC housing stock does provide room to accommodate some of the growth in population in units other than those in new apartment or condominium buildings. The Census Bureau’s American Community Survey estimated that for the 5-years 2012 through 2016 there were 306,711 housing units in DC, and that 160,750 (52.4%) of those units were in structures with 5 or more units. In other words, almost half of DC’s housing stock is in smaller buildings. Over the past 5 years more than 1,000 permits have been issued for projects with less than 5 units, and many smaller buildings can be reconfigured to accommodate more units.

table 4.PNG

In this connection, in the prior 5 year period— 2007 to 2012—larger buildings appear to have played a smaller role in accommodating population growth than in the most recent five years. From 2007 to 2012 Co-Star’s estimate of occupied units grew more slowly than population (6.9% versus 10.7%). The ratio of new population to the estimated increase in occupied apartments and condominiums was 6.1. (See Appendix 2. for more details.)

table 6

With the current pace of construction for apartment and condominium buildings much like that of the past few years, multi-family apartment and condominium projects will undoubtedly provide the largest share of the additional housing needed for DC’s growing population—provided, of course, that a sufficient number of households can afford to live in them. But the connection between increases in population and changes in the housing stock is a complex one, and DC’s changing demographics involve more than simply building large new apartment and condominium buildings for new people.

Appendix 1

Gross and net changes to inventory.

table 7

Apartment share.

table 8

Vacant units.

table 9

New construction.

table 10

Appendix 2

table 11.PNG

About the data. Data on DC multi-family housing buildings and units are for apartments, condominiums, and co-operative apartments of all classes in structures with 5 or more units as reported by CoStar, a real estate information company. The data was accessed toward the end of April. Population data and data from the American Community Survey are from the US Bureau of the Census. The housing and population data are all subject to revision by the source as more information becomes available.

This analysis looks solely at the statistical relationships between changes in DC population, households, and housing units.  It does not address issues related to affordability or homelessness.

An earlier version of this blog was included in the April 2018 District of Columbia Economic and Revenue Trends report issued by the Office of Revenue Analysis of the District of Columbia Office of the Chief Financial Officer.

 

 

High-Income Residents: Are They the Driving Force Behind DC’s Premium Apartments?

In a recent post, we concluded that the premium apartment rental market is the more popular and ascendant segment of the city’s housing market in the context of the current trend in net population growth. To further elaborate on this topic, we profile the tenants in the city’s Class A and Class B apartment buildings built after 2000 based on income tax data characteristics. The full research paper can be found here.

Economic Profile of Tenants

Table 1 tells us that in 2015 half of the residents who were income tax filers in the 88 Class A and Class B large and mid-sized apartment buildings that were built after 2000 had annual reported income of less than $57,428 and were under the age of 31.5. And, the vast majority of these tenants were single tax filers (unmarried and no dependents) and were relatively new[1] to the city.

table111.PNG

[1] We classify a new resident as someone who existed in the city’s income tax data in either 2013, 2014, and/or 2015, but did not exist in 2012 or prior.

Who is more likely to live in new apartment units?

Our data shows that there was a tripling in the number of premium apartment units delivered in 2013 compared to 2012. To better evaluate the data, we divided the buildings into two groups. The first cohort is comprised of all 2015 tax filers found to be residents in multifamily buildings that delivered between January 2000 and December 2012 (relatively older premium multifamily buildings). The second cohort is comprised of all 2015 tax filers found to be residents in multifamily buildings that delivered between January 2013 and December 2015 (newer premium multifamily buildings).  We then fit a statistical model to the data to determine the characteristics of new buildings versus older buildings.

Using T-tests, we find that the newer buildings tended to have units that were an average of 88.3 square feet (10.5 percent) smaller and cost 17.5 percent more per square foot (Table 2). We also found that individual tenants in newer buildings tended to have income that was on average of $9,884 (12.3 percent) less and 1.3 years younger than renters in older buildings.

table222.PNG

Using a statistical model to differentiate the characteristics of tax filers living in a newer building in 2015 versus older buildings, we calculate the probability that certain factors affect the choice of residing in newer apartment buildings instead of older buildings.

While the tenants in new and older apartment buildings are generally very similar, we were able to again tease apart a few distinctions in the two populations as well as a few contributing factors for their housing choices.

chart114.PNG

We find that income has almost no influence on whether a resident chooses to live in a newer or older apartment building (for every $100,000 increase in income, the probability to choose a newer building increases only about 4 percent). Age is also an important factor in determining how likely a resident will choose newer or older apartment units. Younger residents are more likely to reside in newer apartment buildings. For each additional year in age, existing residents are 0.8 percent less likely to reside in newer buildings, while this percentage is 0.2 percent for new residents. We also find that tenants commonly supplement their traditional wage/salary income with additional business income from entrepreneurial or other self-employment endeavors.[2]

Given that 83 percent of all tenants in these buildings are single filers (as shown in Table 1), we find that long time city residents who are head of household tax filers (unmarried income earning adults with dependent children) are 23 percent more likely to live in newer buildings compared to married residents. This is possibly due to the city’s affordable housing efforts to place low-income households in these new buildings via affordable housing programs.  And finally, single residents are more likely to reside in newer buildings compared to married filers, especially when they are relatively long-time residents.

[2] On government tax forms, adjusted gross income is comprised of wages and salaries, business income, investment gains or losses and other income.

Several Ways DC is Changing

In sum, we find the following results. First, 64 percent of the tenants in all the apartment buildings in this study tended to be new to the city. Second, the newest apartment units are smaller and more expensive, and their residents tended to be slightly younger and have less income than residents in the relatively older buildings. Third, residents in the newest units are more likely to have business income as part of their total reported income, which suggests there is an increased tendency for these residents to supplement their traditional wage and salary income with additional income from entrepreneurial or other self-employment endeavors. Lastly and surprisingly, the analysis shows a relatively strong increase in probability for residents in the newer buildings to be head of household filers. This is possibly due to the city’s affordable housing efforts to place low-income households in these new buildings via inclusionary zoning and various housing subsidy programs.

Conventional wisdom assumes that these newer buildings are attracting primarily high-income residents; however, we find that compared to older buildings, the city’s newest and pricier apartment buildings built during the recent residential construction surge (2013 and after) tend to attract a higher percentage of new residents to the city, and also attract a higher percentage of single, young residents with income below the city average. It appears that both the city’s demographics and apartment rental market are continuing to evolve and change in significant ways. And, it is very likely these changes will have considerable implications on the residential and economic patterns of the city in the years to come.

 

The Data

Using data from CoStar, we identified 88 Class A and Class B large and mid-sized apartment buildings (containing 21,203 total residential units) from across the city that were built after 2000. The list can be found here. This study also uses 2015 individual income tax data for all DC tax filers who listed their home address as being in one of the 88 apartment buildings mentioned above.