View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

ISSUE 7 | DECEMBER 2017

F EDE RAL
R E S E RVE
BANK
ST. LOU IS

HOUSING MARKET
PERSPECTIVES
On the Level with Bill Emmons
Bill Emmons is an
assistant vice president and
economist at the Federal
Reserve Bank of St. Louis and
the senior economic adviser for
the Bank’s Center for Household
Financial Stability.

Is Homeownership Bad
for Wealth Accumulation?
H

ouse prices have risen significantly from their low point after
the Great Recession. Homeowners’
equity (HOE)—the market value of
U.S. households’ residential real estate
minus the value of home-secured
debt—has increased even more in
percentage terms; this is because
leverage in the form of mortgage borrowing amplifies price gains, resulting
in proportionately larger HOE wealth
accumulation.
Leverage is a double-edged sword,
however, as debt financing also
magnifies house price declines, causing even steeper percentage losses in
HOE. In fact, long-term data suggest
that debt-financed homeownership,
when compared with other sources

of wealth such as stocks, bonds and
savings accounts, may not generate
returns high enough to compensate
for the higher risk.1 Might concentrating a family’s wealth in homeownership that is financed with a large
mortgage be hazardous to the family’s
wealth? This seems especially likely
for less wealthy families who may be
vulnerable to income disruptions.

Why Recent Gains in House Values
Are Misleading
Total HOE increased by $6.9 trillion
between 2011 and 2016, according
to the Federal Reserve’s Financial
Accounts of the United States.2 After
adjusting for inflation, this gain was
$6.4 trillion, representing an average
annualized increase of 13.5 percent.

Adjusted for changes in the number
of households, inflation-adjusted HOE
rose by 12.5 percent annually during
2011-16, compared with an average
annualized increase of 4.4 percent for
all other types of household wealth
(Table 1, Column C). Both figures
are far higher than their respective
annual averages during the longest
prior period available (1947-2011) of
0.5 percent for HOE and 1.2 percent
for all other wealth.
Nonetheless, three important pieces
of long-run evidence suggest caution in advocating homeownership
as a primary means of wealth accumulation for many families. First,
HOE increased less than nonhousing
wealth during the longest period of

TABLE 1

Aggregate Housing and Nonhousing Wealth, All Races
A) Total amount
(in $ trillions)
Homeowners’ equity (HOE)

B) Average amount
per household
(in $ thousands)

C) Compounded average annualized
change in B) after inflation*

D) Standard deviation of
annual percent change

2016

2016

1947-2011

2011-16

1947-2016

1947-2016

$13.1

$104.4

0.5%

12.5%

1.4%

9.1%

All other household wealth

$79.0

$628.1

1.2%

4.4%

1.5%

4.9%

Total household wealth

$92.2**

$732.6**

1.1%

5.4%

1.5%

4.9%

SOURCES: Federal Reserve Board Financial Accounts of the United States; U.S. Bureau of Labor Statistics.
*Continuously compounded average annualized changes for holding periods indicated, adjusted for inflation
**Totals are not exact due to rounding.

1

“The high volatility of individual
house prices, together with high
transaction costs, lead to low Sharpe
ratios (defined as average excess
return on an asset, divided by its
volatility) on housing.”3
Thus, when properly measured,
investment in an individual house
may be riskier than investments in
nonhousing assets, as leverage amplifies the effects of house price fluctuations on HOE and virtually all families
own just one house.
Finally, another Federal Reserve
dataset, the Survey of Consumer
Finances (SCF), reveals that housing wealth is especially volatile for
black and Hispanic families (Table 2,
Column C). Despite very high volatility, the average increases in the value
of these families’ HOE are no higher
than the gains experienced by white

Cumulative Increases in Average Housing and Nonhousing Wealth
350
Real per household homeowners’ equity
300
Real per household nonhousing wealth
250

Index, 100=1947

available data, 1947-2016, albeit by a
small margin. Table 1 shows that nonhousing wealth increased by 1.5 percent annually, while HOE increased
by 1.4 percent even with the strong
increases in homeowners’ equity in
recent years.
Second, the volatility of HOE during
this long period far outstripped the
volatility of nonhousing wealth (Table
1, Column D). This has been especially
true in recent decades, as reflected in
the accompanying figure. While the
estimates in Table 1 include the volatility of HOE in excess of house price
variability induced by leverage, they
do not reflect the extra volatility experienced by a family holding an undiversified housing portfolio—namely, a
single house. Some research suggests
that individual house prices may be
twice as volatile as broad house price
indexes; this lack of diversification,
combined with leverage, can increase
a family’s risk significantly:

200
150
100
50
0
1950

1955

1960

1965

1970

1975

1980

1985

1990

1995 2000 2005 2010

2015

SOURCES: Federal Reserve Board Financial Accounts of the United States; U.S. Bureau of Labor Statistics;
U.S. Census Bureau.
FEDERAL RESERVE BANK OF ST. LOUIS

TABLE 2

Changes in Housing Wealth by Race or Ethnicity
A) Compounded average annualized
change in HOE after inflation*

B) Average threeyear change in HOE
after inflation**

C) Standard
deviation of threeyear change**

1989-2010

2010-16

1989-2016

1989-2016

1989-2016

1.6%

2.0%

1.7%

5.0%

18.8%

Black

1.5%

-0.3%

1.1%

3.3%

30.6%

Hispanic

0.6%

5.4%

1.7%

5.0%

41.5%

Asian or other

1.9%

5.7%

2.7%

8.2%

22.2%

White nonHispanic

SOURCE: Federal Reserve Board Survey of Consumer Finances (2016).
*Continuously compounded average annualized changes for holding periods indicated, adjusted for inflation
**Simple three-year average (standard deviation) calculated from all nine three-year periods between surveys

TABLE 3

Changes in Nonhousing Wealth by Race or Ethnicity
Compounded average annualized
change in non-HOE wealth
after inflation*

Average three-year
change in non-HOE
wealth after inflation**

Standard
deviation of threeyear change**

1989-2010

2010-16

1989-2016

1989-2016

1989-2016

2.6%

5.0%

3.1%

9.3%

15.3%

Black

1.7%

8.5%

3.2%

9.7%

28.2%

Hispanic

2.6%

7.9%

3.8%

11.4%

25.2%

Asian or other

3.4%

2.9%

3.3%

9.9%

19.7%

White nonHispanic

SOURCE: Federal Reserve Board Survey of Consumer Finances (2016).
*Continuously compounded average annualized changes for holding periods indicated, adjusted for inflation
**Simple three-year average (standard deviation) calculated from all nine three-year periods between surveys

2

families. In contrast to the aggregate
data described above, the SCF shows
that nonhousing wealth generally
increased much more than housing
wealth with no higher volatility.
(See comparative results in Table 2
and Table 3.)4

Avoiding Homeownership for
Building Wealth
To be sure, owner-occupied housing provides benefits to many families. It may be difficult to find rental
housing of the quality, location, style,
etc., comparable to owned housing.
Owning provides a hedge against the
risk of future rent increases; however,
owning also means forgoing potential declines in rental housing costs.
Owning also can make moving more
expensive due to associated buying
and selling costs.
There are portfolio diversification
benefits to owning a house, as house
price changes and returns on other
assets are less than perfectly correlated. Overall, however, our data
show that leveraged homeownership
is riskier than nonhousing wealth and
provides average increases that are
no higher (and may be lower) than
returns on nonhousing assets. This
suggests that caution is warranted
from a wealth-building perspective.

TABLE 4

Housing and Nonhousing Wealth by Race or Ethnicity in 2016
A) Average amount of
homeowners’ equity (HOE)

B) Average amount of
non-HOE wealth

C) HOE share of
total wealth

$205,145

$695,467

22.8%

$57,694

$82,251

41.2%

White nonHispanic
Black
Hispanic

$69,055

$113,144

37.9%

Asian or other

$218,937

$493,304

30.7%

SOURCE: Federal Reserve Board Survey of Consumer Finances (2016).

Black and Hispanic families have
long maintained higher concentrations of housing in their portfolios
than have white families, as shown
in Table 4.5 This is an important
reason why blacks’ long-run wealth
trajectories have been lower and
more volatile than those of whites.
High concentration in housing also
has made Hispanic families’ wealth
much more volatile, although their
average returns have been similar to
whites’. Homeownership can be part
of a financially sound household’s
portfolio, but evidence suggests that
it should constitute a limited share of
total assets.

ENDNOTES
1

2

3

4

5

3

The figures discussed in this publication are
not true rate-of-return calculations because
balance-sheet aggregates reflect both
changes in asset prices and net investment
flows in all asset categories. However, these
are appropriate measures of actual wealth
accumulation by asset type.
See the Federal Reserve statistical release,
dated Sept. 21, 2017, www.federalreserve.
gov/releases/z1/Current/z1.pdf.
See Monika Piazzesi and Martin Schneider, “Housing and Macroeconomics,”
National Bureau of Economic Research
(NBER) working paper 22354, July 2016,
p. 9, http://web.stanford.edu/~piazzesi/
housingandmacroeconomics.pdf.
The two Fed datasets collect data in very
different ways. The aggregate data in the
Financial Accounts reflect institutional data
sources, while the SCF data reflect individual survey responses. The SCF data include
estimates made by families of the value of
their nontraded assets, such as houses.
A previous issue of Housing Market
Perspectives (Issue 5, July 2017) showed
that black and Hispanic families historically
have had higher shares of housing in their
household portfolios when compared with
white and Asian or other families.