Understanding the DC housing market: demand for urban living, not the construction of new housing, is the driving force

=========

Originally published Friday March 12th, but republished on Saturday March 13th, because of reordering of the list of items, edits and expansions including adding an item to the list, on public housing.

=========

There is an op-ed in the Washington Post, written by DC housing activists, on their belief that building more housing merely fosters gentrification.

-- "D.C.’s build-build-build mind-set results in more gentrification"

From the article:

The past 20 years have made clear that the build-build-build scenario just results in gentrification, the forcing out of Black and Brown residents from D.C., and increased land values that destroy existing affordability without producing replacement affordable housing. In addition, hundreds of acres of public land have been given away or leased at fire-sale prices with all manner of financial incentives thrown in, from gap funding to tax-increment financing to tax abatement, usually for more luxury studios and trophy office space. Far from correcting for these disproved policies, the mayor’s amended Comprehensive Plan provides more of the same.

I didn't do super well at economics in college, because I'm not great at higher math, although I did reasonably well at microeconomics ("economics of the firm").  In any case, I joke that DC housing activists make me out to be a great economist, because they argue that not adding to supply, when demand is greater than supply, leads to lower prices.

(My other joke is about politicians and how politics is often trying to do everything possible to avoid having to acknowledge micro- and macro- economics.)

-- "What people don't get about the DC housing market: supply is much less than demand, so prices keep rising (a/k/a basic economics)," 2017
-- "Will DC's Housing Ever Be Affordable Again? Answer: No," 2016

One of the biggest problems in understanding the housing market is that there are many different elements to it, all happening simultaneously, it's not just one thing.

Levittown suburban housing constructed on Long Island, New York, Single family detached houses, requiring a car to get places: work, school, and home.  Mostly for white people.

But the reality of the last 20 years of the DC housing market is pretty simple.  Since 2000, demand for housing in center cities has increased significantly, compared to the 50+ years after World War 2 when people preferred to live in the suburbs.

-- "Why "gentrification" is visible now... the Diffusion of Innovations Curve of Everett Rogers," 2018

In The Option of Urbanism: Investing in a New American Dream, Chris Leinberger writes that whereas before 2000, 70% of the housing market was focused on the suburbs, today 30% of the housing market is focused on cities, 30% on suburbs, and 40% is satisfied by either type of housing.  

Housing preference trends today mean that potentially 70% of households are interested in center city living, significantly increasing demand and driving up prices. In situations of high demand, people with more money dictate conditions and control the market.

Q Street NW, Georgetown, DC

Displacement did occur some in the last 20 years, as some rental housing was converted to owner occupied, and as prices rose in response to increased demand.  

-- "More about contested spaces--gentrification," 2008

But that's a different phenomenon from what has happened with the addition of new housing.  In DC, building more housing was a response to increased demand, and for the most part*, in DC, it wasn't produced through the replacement of existing occupied housing.

(* The rebuilding of public housing did occur in the 1990s and early 2000s and did result in a decreased number of units, both from making smaller complexes and adding some market rate housing.  That's a different segment of the market.

There are some instances where some apartment buildings, such as Clifton Terrace in Columbia Heights, were converted from low income to high income housing. There are a couple instances of a string of rowhouses being torn down and replaced by apartment buildings. )   

What I've learned in 20 years through close observation of the DC housing market

1.  Strong markets function differently than weak markets.  DC is a strong market.  You can also assess strength of the market at the neighborhood scale.  For the most part, all of NW DC is a strong market, and most of SW (it's so small). NE is a strong market along the Red Line Metrorail, and weaker with distance from the Red Line.  SE is a strong market west of the Anacostia River.

With a couple exceptions, the public safety cost of living east of the river is high and "gentrification pressures" won't fully build until the majority of places in higher demand areas are mostly built out.

(In weak markets, the problem is lack of demand, vacant housing, the need for neighborhood stabilization, and low prices.  Low prices can make it difficult to maintain older houses, because the cost of rehabilitation and maintenance isn't necessarily recaptured in higher value. New buildings are difficult to finance because they won't appraise high enough to cover the costs of new construction.)

2.  The markets for rental and owner occupied housing function differently.

3.  The markets function differently for higher income versus lower income housing segments, for both owner occupied and rental markets.  

For example, Washington Post columnist Catherine Rampell writes, "Rents for the rich are plummeting. Rents for the poor are rising. Why?."  

In short, I'd say, there's always demand for lower priced housing, so it will continually rise in price, while certain exogeneous shocks to the high income market, or overproduction can affect prices, at least in the short and intermediate run.

4.  Proximity to Metrorail transit increases housing demand and pricing.

5.  Increased demand for center city living vis a vis the suburbs has led to higher pricing.  (In 2003, you could still buy a house in the H Street NE neighborhood for under $125,000.  Now houses there cost more than $750,000.)  Into the early 2000s, a small studio in the Cairo Building in Dupont Circle cost less than $100,000. Today the cost is upwards of $400,000--a 373 s.f. unit sold for $270,000+ in 2014.

Because of the rise in demand for center city living, more people with more resources have entered the urban housing market, increasing pricing compared to the period when there was less demand for urban housing and a great number of vacant dwellings as a result of outmigration to the suburbs in the many decades after WWII.

6.  In keeping with residential choice preferences, the greatest demand is for single family housing, especially historic building stock.  Although the addition of multiunit housing stock supports demand for different market segments that hadn't been satisfied by the previously narrow range of housing types.

Blue is zoned for single family rowhouse; pink for single family detached housing.

7.  Most of DC's SFH was built before 1935, when the nation's population was 40% of today's. Based on national population growth, DC has less housing supply relative to demand.

• US population, 1930: 123.1 million
• US population, 2020: 331 million
• DC population, 1930: 486,000
• DC population, 2020: 693,000

(According to editions of the Statistical Abstract of the United States published in the 1940s, DC's peak population, during the War years, may have been as high as 900,000 although it dropped to 802,500 in the 1950 Census, which is the highest ever official population.)

8. The supply of SFH is relatively fixed.  Demand is greater than supply, and prices rise in response.  

The land dedicated to single family housing is mostly built out.  More housing can't be added, except in odd locations where institutional land is deaccessioned.  By definition the supply of "historic" building stock is fixed, it is comprised of housing built in the past.  Houses built before 1930 comprise the largest stock of historic buildings in DC.

In short, decisions made 90-120 years ago affect today's market.  

9.  As long as demand is greater than supply, even with new additions to supply, housing costs will continue to rise.

This point is not widely understood.  For example, from the article:

The principal failed policy is that building more market-rate residential density will result in adequate affordable replacement units. It hasn’t, and it won’t. Our city’s feeble inclusionary zoning program has produced only 1,000 units in the past decade because typically only 8 to 20 percent of new units are set aside as “affordable.”

The issue in DC's housing market isn't "affordable replacement units."  It's having enough supply to meet demand.  DC's population has increased by about 90,000 from 2010 to 2020--that's 15%.  And it could have increased more had there been available housing units, especially if available at a lower price.

Note that even if not enough, inclusionary zoning adds lower cost housing units to the market, over previously levels.  The problem is the overall rise in demand for housing in DC, which has led to a "precipitous" rise in pricing.  ("Precipitous rises" don't matter if you are well off.  They matter a lot if you aren't.)

10.  Zoning rules created relatively homogeneous zones of residential units, with a limited range of types of units, forcing all segments of the market towards SFH rather than a mix of single family and multiunit housing.  This narrowing of available housing types displaces segments of the market that can't afford but also don't need traditional SFH.

11.  Because until the Great Depression and the growth of the federal government, DC was a relatively small city without much of an industrial base, housing was not built to be particularly dense.  Compare housing built at the time to cities like New York and their tenements and rowhouses, or Pittsburgh, Philadelphia, and Boston, Montreal, etc.

Now that demand for housing is greater, DC's housing stock is small in density, relative to demand, and less able to meet demand, further leading to higher prices.

12.  Household size has decreased, increasing the number of units required to house the same number of people.  (See the concept of "overhousing" and Eric Klinenberg's book Going Solo: The Extraordinary Rise and Surprising Appeal of Living Alone.)

✷ Average household size (US), 1930: 4.11

✷ Average household size (US), 1950: 3.54

✷ Average household size (US), 2010: 2.58

✷ Average household size (DC), 1950: 3.2

✷ Average household size (DC), 2010: 2.11

Now vs. Then: renovated and rehabilitated apartment building at 2nd and T Streets NE, Eckington neighborhood, DC

13.  Because of the cost of land and the extreme difficulty of redeveloping SFH, most new housing constructed is multiunit, and located on commercially zoned land, which is easier to redevelop and can be built to a higher density.

14.  DC's Height Limit, by limiting height of (multiunit residential) buildings restricts the housing supply, increasing prices.

On wide streets buildings can be taller.  But there is no density bonus for buildings close to transit, regardless of the width of the street.

Iowa Apartment Building (condominiums now)

1325 13th Street NW (near Logan Circle)

five blocks to the Shaw or Dupont Circle Metrorail Stations

Fort Totten Square Apartments

5661 3rd Street NE (fronting Riggs Road NE)

three blocks from the Fort Totten Metrorail Station

15. Irrespective of the Height Limit, constructing housing at less than maximum allowable density decreases supply and raises prices, which is an opportunity cost (along with reduced commercial and personal property, income, and sales tax revenue to the city).

The same building was empty and dilapidated in 2007.

16.  New housing is constructed at today's cost of land, labor and materials so it is priced at the top of the market.  Only over long periods of time do "new" additions to supply put downward pressure on prices, as the housing ages and even newer and higher priced housing is constructed and added to supply.

17.  It costs more to rehabilitate "historic housing" to historic standards.  In strong markets like DC or neighborhoods like Georgetown, this increased cost comes back to the bottom line by increased property values.  In weak markets like Baltimore or Cleveland this isn't necessarily the case.

18.  Every generation or so, a house may need serious rehabilitation.  This leads to an upward repricing of housing because it has been "refreshed."  Hopefully, if done through "flipping," the rehabilitation has been executed to a high standard, frequently this isn't the case.

19.  In the rental market, because of the demand-supply mismatch, substandard housing (Class B, C and D) rents for higher prices than would prevail in a weaker market. (This phenomenon of higher prices for less well maintained housing bleeds across the DC border into Prince George's County.)

20.  New rental housing, unless subsidized, is priced at the top of the market because it is built at today's prices.  Volume constraints imposed by the Height Limit and the high cost of land, labor and materials, leads for profit housing developers to focus on the highest income segments of the market.

21.  The demand-supply mismatch also creates pressure to convert smaller and older apartment buildings and complexes to owner occupied housing, which reduces supply in the rental market, leading to increased prices and displacement.

22. Conversion of apartment buildings does extend the range of housing types available in a neighborhood, increasing the opportunity for segments of the market that can afford apartment units, but not traditional SFH units.

23.  There is a lot of talk about the "missing middle" of housing types such as "rowhouses," duplexes, carriage houses, etc.  

-- "Bring Back Missing Middle Housing," AARP
-- Missing Middle Housing website (and book)

DC does have significant amounts of this type of housing.  But because of the demand-supply imbalance, this housing is priced significantly higher than it would be in a community where the demand and supply for housing is closer to being balanced.

24.  Increasing the ability to create carriage houses/alley dwellings, English basements, and attic apartments in existing housing would increase the supply of housing, and make it easier for households to pay off expensive mortgages.  But the cost of creating this housing, in the short run--to do it legally will cost upwards of $200,000 per unit, even with zero cost for land--will add to housing supply, but it won't be "affordable housing."

(In the intermediate run, funding this housing is complicated because mortgage products haven't caught up to the phenomenon of an owned house including rental units.)

-- "Calculating the costs of building an ADU," Building an ADU website/Backyard Revolution book

25.  Current zoning allows small apartment buildings to be increased slightly in size, and rowhouses can be converted into two to three unit "flats."  High demand neighborhoods are seeing an increased number of conversions as a result.  

But the housing won't be cheap.  The acquisition cost of the property is the normal cost for a single family house, plus the cost of conversion, plus the profit percentage, carrying costs, etc.

-- "A short point about why eliminating single family zoning won't result in a rise in "affordable housing" (any time soon)," 2019

A sign in front of 3625 13th Street NW announces the conversion of this single family rowhouse into three separate units.

This four unit small apartment building has been converted into a larger four unit condominium, bythe addition of a third floor ("Two-bedroom, one-bathroom condo in D.C.'s Trinidad lists for $399,000," Washington Post).

26. A more recent phenomenon is the rise of temporary housing rental through Airbnb and other programs.  In strong markets this leads to an increase in rental prices because of the reduction of supply. The impact on the owner occupied market is probably minimal ("When Airbnb Listings in a City Increase, So Do Rent Prices," Harvard Business Review, "What Airbnb really does to a neighborhood," BBC News).

27.  Reducing the supply of public and social housing through rebuilding programs, increased demand for affordable housing, and led to higher prices.  The HUD HOPEVI program of rebuilding public housing but at the same time reducing the total number of units, reducing the size of units making it difficult to accommodate large families, and adding market rate housing to the mix in certain cases reduced the supply of low income/affordable housing, thereby increasing demand and pricing.  

Other low income housing rebuilding programs, such as Sursum Corda and DC's parallel program, New Communities, have similar effects.

This 1999 article from the Washington City Paper, "Dream City," discusses the rebuilding of the Ellen Wilson Dwellings project on Capitol Hill.  The original development had 129 units, the new development 134.  But only 33 units were set aside as affordable, a decrease of 96 units of affordable housing.

Resulting from these programs, low income housing residents have been displaced to Prince George's County, Maryland with especially deleterious effects on that community including a higher rate of murders and crimes, and bankruptcy of the community hospital ("Shouldering the Burden," Suburban Gazette Newspapers, 2003).

-----------------

I'm not being totally fair.  Yes, there is gentrification.   There is no question, using the classic definition, there is gentrification in DC.  Although again, it's a function of increased demand, and new housing construction is a function of increased demand.  

In the argot of urban sociology, neighborhoods are being "reproduced" for higher income segments of the market, as more people of higher income move into a neighborhood, attracted by its location within the center city and relative affordability.  

Neighborhood reproduction = gentrification.  

Yard signs posted in a yard on Peabody Street NW in the Brightwood neighborhood.  

Photo: Perry Stein, Washington Post

But the phenomenon is different in low income neighborhoods than it is in higher income neighborhoods.  In the latter neighborhoods, displacement isn't what's happening so much as people are choosing to cash out.

The velocity of neighborhood reproduction is accelerated through the process of the creation of "one over neighborhoods" -- aspiring homeowners prefer a high profile neighborhood, like Capitol Hill or Dupont Circle, but they can't afford it, so they move into the closest comparable nearby neighborhood that they can afford, in the process also changing that neighborhood's demographics and market for retail and other amenities, to make it more similar to the preferred neighborhood.

-- "Shaw (and Mid City East) as a one-over neighborhood: revitalization, displacement, gentrification as a function of critical mass and timing," 2013

For example, choosing H Street NE or Hill East because they can't afford a house on Capitol Hill, buying in Trinidad because they can't afford H Street NE, Columbia Heights or Shaw because they can't afford Dupont Circle, Petworth because they can't afford Columbia Heights, Brightwood because they can't afford Petworth and Manor Park because they can't afford Takoma DC, Takoma Park, Maryland or Brightwood, etc.

(The one over neighborhood concept was coined by the Live Baltimore resident recruitment program.)

In low income neighborhoods and buildings, people are displaced as the housing is rehabilitated and converted to appeal to higher income segments of the housing market.

But again, because of the significant rise in demand for center city living, prices rose significantly, and this created pressure on low income households, irrespective of the construction of new housing.

Affordability in the face of significant rising demand.  Dealing with affordability is a whole other issue and requires a suite of responses. Some the city has employed, many it has not.  But the biggest issue is that DC is comparatively small and mostly built out.  Supply is comparatively static.