Multifamily Return Volatility and Why it Matters
The ebb and flow of multifamily returns has implications for portfolio structuring, risk management, and overall positioning in the market. In pursuit of a more exhaustive risk assessment, wise investors should examine both relative and absolute volatility, measured by beta (ß) and standard deviation (Std Dev), respectively. Why are these measures important and how do they differ? How have multifamily returns behaved both historically and recently? How do factors like GDP, inflation, household growth, single-family home (SFH) prices, the cap rate yield spread, and an overvalued stock market affect multifamily return volatility?
RETURNS
Absolute Volatility
The degree to which returns fluctuate over time.
TIME
1%
Relative Volatility
The degree to which MSA returns move given movements in national returns.
“Historically, if national returns
U.S.
PH X
β=1.00
β=1.17
increase by 1.00%, then the average Phoenix return increases by 1.17%”
-1%
What is Beta (ß)?
How has ß Changed across and within MSAs? HISTORICAL (4Q88 – PRESENT ) MOVEMENT IN MSA RE TURNS GIVEN 1.00% MOVEMENT IN U.S. RE TURNS
1%
U.S.
PHX
LA
SEA
DC
C HI
HOU
β=1.00
β=1.17
β=1.09
β=1.05
β=1.05
β=0.88
β=0.44
If the average annualized return in Los Angeles increases by 1.09% given a 1.00% increase in national returns, then Los Angeles provides more upside when overall returns rise but more downside when overall returns fall. Conversely, Houston has historically shielded investors from downside risk—for every 1.00% decline in national returns, average annualized Houston returns decline by 0.44%. In this sense, Los Angeles and Houston are said to have historical ß coefficients of 1.09 and 0.44, respectively.
-1%
FOUR-YE AR (2Q17 – PRESENT ) MOVEMENT IN MSA RE TURNS GIVEN 1.00% MOVEMENT IN U.S. RE TURNS
1%
U. S .
PHX
HOU
LA
SEA
C HI
DC
β=1.00
β=1.38
β=1.32
β=1.23
β=1.20
β=0.94
β=0.83
As useful as ß is in assessing risk, seldom are financial measurements constant, and ß is no exception. To ensure a more comprehensive risk assessment, view the diagrams on the left showing ß coefficients for each market both historically and over the past four years.
-1%
Switching from a historical to more recent perspective, Houston and DC experienced the most notable changes in their ß coefficients: Houston’s recent ß is nearly 200% greater than its historical ß; DC’s recent ß is over 20% lower than its historical ß. Therefore, compared to each market’s respective historical behavior, Houston is now a more aggressive market and DC is now a more defensive market.
The Nonconstant Nature of ß, Visualized ß Coefficients Over Time, By MSA
Sharp Rise In Money Supply
Declining SFH Prices
6.00
5.00
4.00
CA Begins Statewide Rent Control
Rolllliing 4-Year ß
3.00
2.00
1.00
Q1 2021
Q3 2020
Q3 2019
Q1 2020
Q1 2019
Q3 2018
Q1 2018
Q3 2017
Q1 2017
Q3 2016
Q1 2016
Q3 2015
Q1 2015
Q3 2014
Q1 2014
Q3 2013
Q1 2013
Q3 2012
Q1 2012
Q3 2011
Q1 2011
Q3 2010
Q1 2010
Q3 2009
Q1 2009
Q3 2008
Q1 2008
Q3 2007
Q1 2007
Q3 2006
Q1 2006
Q3 2005
Q1 2005
Q3 2004
Q1 2004
Q3 2003
Q1 2003
Q3 2002
Q1 2002
Q3 2001
Q1 2001
Q3 2000
Q3 1999
Q1 2000
Q1 1999
Q3 1998
Q1 1998
Q3 1997
Q1 1997
Q3 1996
Q1 1996
Q3 1995
Q1 1995
Q3 1994
Q1 1994
Q3 1993
Q1 1993
Q3 1992
0.00
-1.00
-2.00
Costa-Hawkins Rental Housing Act
-3.00
Rising SFH Prices PHX
LA
DC
SEA
HOU
CHI
N OTE : S HAD E D R EG IO N S I N D I C ATE R ECE S S IO N S . C ALLO UTS DO N OT I M PLY C AU SATIO N , B UT R ATH E R PROVI D E CO NTE X T.
From roughly 2008 to 2012, all ß coefficients above began approaching one, indicating returns in each market were moving in lockstep with national returns to a greater extent than before. Markets with ß coefficients less than one increased, and markets with ß coefficients greater than one declined, both approaching the relative volatility of the nation (1.00, by definition). Coinciding with these more homogeneous ß coefficients was a reduction in national SFH prices, measured by the Case Shiller HPI. Likewise, the rapid post-2011 rise in SFH prices coincided with the return of heterogeneous ß coefficients across these markets, with the exception of recent times. Additionally, the recent return of homogeneity in ß coincides with the COVIDinduced rise in the money supply. Clearly, there are innumerable factors which may or may not influence MSA ß coefficients.
Relationship between Absolute and Relative Volatility, and Likelihoods of Positive/Above-Average Returns
HISTORICAL ABSOLUTE (STD DEV)
RELATIVE (ß)
ABSOLUTE (STD DEV)
RELATIVE (ß)
PHX
13.08%
1.17
5.08%
1.38
LA
14.17%
1.09
3.31%
1.23
DC
10.29%
1.05
2.20%
0.83
SEA
11.51%
1.05
3.92%
1.20
HOU
8.87%
0.44
3.83%
1.32
CHI
8.94%
0.88
2.68%
0.94
U.S.
8.08%
1.00
2.30%
1.00
MSA
A powerful way to measure absolute return volatility in a market is via standard deviation, which measures the degree to which returns are spread out. The higher the standard deviation, the more extreme the fluctuation in returns over time. Historically, Phoenix and Los Angeles are the two most volatile markets out of the considered markets, in an absolute sense. Notably, Phoenix’s historical upside potential (ß) is greater than that of Los Angeles, but Phoenix’s historical absolute volatility is less than that of Los Angeles. As such, Phoenix has historically provided leveraged upside potential alongside less drastic fluctuations. Notice how the absolute volatility of each market over the past four years is significantly lower than its historical counterpart. Discussed in more detail later, the recent ß values are also more similar than they are historically.
2Q17 - PRESENT
Historical Likelihood of Positive/Above-Average Returns HOU
CHI
Generally, higher absolute volatility implies a higher likelihood of generating a positive return in any given period, but this trend is anything but perfect. Seattle has a lower historical absolute volatility than that of Los Angeles and Phoenix but is more likely to generate a positive return. Additionally, Phoenix is more likely than DC, Chicago, and Houston to generate a return above the national average. Seattle is historically most likely to generate an above-average return.
DC
PHX
LA
SEA
0%
20% 20%
40% 40%
60% 60%8
80% 0%
100% 100%
Likelihood of Return Above National Average (AROR) Likelihood of Positive Return (AROR)
Absolute Volatility Visualized
Is Phoenix a “Boom-and-Bust” Market?
Annualized Returns (Income + Capital)
Most metrics point to Phoenix shifting away from its “boom-and-bust” reputation. Phoenix is historically more likely to generate a positive return than Los Angeles, Chicago, and Houston. Phoenix is also historically less volatile than Los Angeles, on an absolute basis.
75% 65% 55% 45% 35% 25% 15%
Q1 2021
Q2 2020
Q3 2019
Q4 2018
Q1 2018
Q2 2017
Q3 2016
Q4 2015
Q1 2015
Q2 2014
Q3 2013
Q4 2012
Q1 2012
Q2 2011
Q3 2010
Q4 2009
Q1 2009
Q2 2008
Q3 2007
Q4 2006
Q1 2006
Q2 2005
Q3 2004
Q4 2003
Q1 2003
Q2 2002
Q3 2001
Q4 2000
Q1 2000
Q2 1999
Q3 1998
Q4 1997
Q1 1997
Q2 1996
Q3 1995
Q4 1994
Q1 1994
Q2 1993
Q3 1992
Q4 1991
Q1 1991
Q2 1990
-15%
Q3 1989
-5%
Q4 1988
5%
-25% -35% -45% -55% PHX
LA
DC
SEA
HOU
CHI
US
One can quickly identify Phoenix and Los Angeles as markets with high absolute volatilities—returns tend to swing up and down more so than in other markets. One can also visually confirm the U.S. market has the lowest absolute volatility out of the considered MSAs. Seattle tends to stand out to a greater degree than Chicago, Houston, or DC, reflecting the numerical data in the table above.
Revisiting ß: How do macro and micro factors affect ß in each market over time? The below highlights the extent to which six macro and micro factors influence the observed changes in ß in each market over time.
PHOE N IX
LOS ANG E LE S
WASHINGTON D.C.
SE AT TLE
HOUSTON
CH ICAGO
OF THE METRICS CONSIDERED, THE STRONGEST RECENT DRIVER OF ß IN PHOENIX IS THE WILSHIRE 5000 TO GDP RATIO, WHICH MEASURES THE EXTENT TO WHICH THE U.S. STOCK MARKET IS OVERVALUED. OTHER METRICS HELPING TO DRIVE INCREASED ß RISK IN PHOENIX INCLUDE THE NUMBER OF U.S. HOUSEHOLDS, SFH PRICES, AND THE APARTMENT RISK PREMIUM.
LOS ANGELES’ RECENT ß IS MODERATELY DRIVEN BY INFL ATION, SFH PRICES, AND THE APARTMENT RISK PREMIUM, AND STRONGLY DRIVEN BY THE NUMBER OF U.S. HOUSEHOLDS. OTHER DIFFICULT-TO-MEASURE FACTORS MAY INCLUDE LEGISL ATIVE ACTIONS SUCH AS RENT AND EXPENSE CONTROLS.
ß IN DC TENDS TO BE LARGELY DRIVEN BY HOUSEHOLDS, SFH PRICES, AND THE WILSHIRE 5000 TO GDP RATIO, CONSIDERING THE METRICS EXAMINED. OF THE SIX MARKETS CONSIDERED, DC’S HISTORICAL AND RECENT ß VALUES INDICATE THE MARKET’S RETURNS MOVE CLOSELY WITH THE NATION. DC’S ß IS ALSO SECOND-MOST STABLE OUT OF THE SIX.
INFL ATION IS A RATHER STRONG DRIVER OF RECENT ß RISK IN SEATTLE; HOWEVER, INFL ATION’S EFFECT ON SEATTLE’S ß RISK IS HISTORICALLY WEAK. CONTINUING THE TREND IN OTHER MARKETS, SEATTLE’S RECENT ß RISK HAS BEEN DRIVEN BY HOUSEHOLDS, SFH PRICES, AND THE APARTMENT RISK PREMIUM.
HISTORICALLY, THE NUMBER OF U.S. HOUSEHOLDS AND INFL ATION HAVE BEEN MODERATE TO STRONG DRIVERS OF ß IN HOUSTON. SWITCHING PERSPECTIVES FROM HISTORICALLY TO RECENTLY, HOUSTON’S ß IS L ARGELY UNAFFECTED BY ALL FACTORS BUT INFL ATION. ADDING TO THE CHANGING NATURE OF HOUSTON’S ß, THE RECENT REL ATIONSHIP BETWEEN INFL ATION AND ß IS SUCH THAT THEY MOVE IN OPPOSITE DIRECTIONS.
CHICAGO’S RECENT ß IS L ARGELY UNAFFECTED BY ALL SIX FACTORS. COINCIDENTALLY, CHICAGO’S ß IS THE MOST STABLE OF ALL MARKETS CONSIDERED. THERE EXIST NEGATIVE REL ATIONSHIPS BETWEEN CHICAGO’S RECENT ß AND HOUSEHOLDS, SFH PRICES, AND THE WILSHIRE 5000 TO GDP RATIO, BUT THEY ARE WEAK.
Visualizing Drivers of ß Drivers of ß, ByStrength MSA as Driver of ß
Strength as Driver of ß PAST FOUR YE ARS
HISTORICALLY
0.5-1
CHI
CHI
HOU
HOU
SEA
SEA
0-0.5
0.5-0
Strength as Driver of ß DC
1--0.5
DC
CHI
LA
Real Real GDP GDP
Inflation PCE PCE
U.S.HH HH U.S. Estimates Estimates
Case Case Shiller Shiller HPI HPI
Wilshire Wilshire 5000/GDP 5000/GDP
MODER ATE TO STRONG POSITIVE DRIVER 0.5-1 WE AK POSITIVE DRIVER 0-0.5 WE AK NEGATIVE DRIVER -0.5-0 MODER ATE TO STRONG NEGATIVE DRIVER -1--0.5
PHX Cap Cap RateRate -Yield 10YCM 10YCM Yield Spread
Real Real GDP GDP
Inflation PCE PCE
U.S.HH HH HOUU.S.
Estimates Estimates
LA
Case Case Shiller Shiller HPI HPI
• Partially illustrated by the expansive yellow SEA region above, SFH prices and U.S. household growth have become more influential in driving ß risk in Phoenix, Los Angeles, Seattle, and DC. These factors tend to move in the same direction as ß risk in these markets, whereas this relationship did not always exist historically. DCDC’s recent ß is • In stark contrast to the historical relationship, strongly driven by household growth, SFH prices, and the degree to which the U.S. stock market is overvalued.
Dominated by blue and gray, the historical plot makes clear that few of the factors • T he historical plot shows Houston’s ß being moderately driven by considered moderately or strongly drive ß inflation and household growth (top left yellow LA strip). Shifting to risk in each market. Houston is an exception a more recent perspective, the yellow strip vanishes, replaced with to this trend, however, with ß risk in the red indicating inflation’s negative effect on Houston’s ß, and blue market moderately to strongly driven by indicating household growth’s lack of influence over Houston’s ß. inflation and the number of U.S. households. • Inflation has been influencing recent ß risk in Los Angeles and In general, the six considered factors have PHX Seattle, indicated by the yellow intersections of Los Angeles little to do with the historical behavior of ß Real GDP PCE U.S. HH Case Shiller Wilshire Cap Rateand Seattle with inflation (PCE). The historical plot reveals this risk across each market. Estimates relationship HPI did not exist 5000/GDP 10YCM Yield before. Note: The term “driver” refers to a variable’s • Recent ß risk has been moving alongside increases in the apartment association with another variable, and should risk premium, as measured by the Cap Rate - 10YCM Yield Spread, not be construed as establishing or implying although this relationship is hardly uniform across markets. a causal link.
SOURCES: Proprietary statistical model, Federal Reserve Economic Data (FRED), National Council of Real Estate Investment Fiduciaries (NCREIF)
Noah Stone Author, Economic Analyst 415.263.4257 noah.stone@berkadia.com Irvine, California
Jonathan Rappa Senior Director of Business Operations 602.445.1634 jonathan.rappa@berkadia.com Scottsdale, Arizona
Wilshire Wilshire 5000/GDP 5000/GDP
PHX CapCap RateRateYield 10YCM
10YCM Yield Spread
POSITIVE DRIVER:
INCREASES IN METRIC ARE ASSOCIATED WITH INCREASES IN ß . DECREASES IN METRIC ARE ASSOCIATED WITH DECREASES IN ß .
NEGATIVE DRIVER:
INCREASES IN METRIC ARE ASSOCIATED WITH DECREASES IN ß . DECREASES IN METRIC ARE ASSOCIATED WITH INCREASES IN ß .
© 2021 Berkadia Proprietary Holding LLC. Berkadia® is a trademark of Berkadia Proprietary Holding LLC. Commercial mortgage loan banking and servicing businesses are conducted exclusively by Berkadia Commercial Mortgage LLC and Berkadia Commercial Mortgage Inc. This advertisement is not intended to solicit commercial mortgage loan brokerage business in Nevada. Investment sales / real estate brokerage business is conducted exclusively by Berkadia Real Estate Advisors LLC and Berkadia Real Estate Advisors Inc. Tax credit syndication business is conducted exclusively by Berkadia Affordable Tax Credit Solutions. In California, Berkadia Commercial Mortgage LLC conducts business under CA Finance Lender & Broker Lic. #988-0701, Berkadia Commercial Mortgage Inc. under CA Real Estate Broker Lic. #01874116, and Berkadia Real Estate Advisors Inc. under CA Real Estate Broker Lic. #01931050. For state licensing details for the above entities, visit www. berkadia.com/legal/licensing This document and any statements, information, data and content contained therein, and any materials, information, images, links, sounds, graphics or video provided in conjunction with this document (collectively “Materials”) has been prepared for informational purposes or general guidance on matters of interest only, and does not constitute professional advice, advertising or a solicitation. The Materials are of a general nature and not intended to address the circumstances of any particular individual or entity. You should not act upon the information contained in the Materials without obtaining specific professional advice. As such, nothing herein constitutes legal, financial, business, investment or tax advice and you should consult your own legal, financial, tax, investment or other professional advisor(s) before engaging in any activity in connection herewith. The information in the Materials is not a substitute for a thorough due diligence investigation. No representation or warranty (express or implied) is given as to the accuracy or completeness of the information contained in the Materials, and, to the extent permitted by law, Berkadia Commercial Mortgage LLC ( together with its affiliates, the “Company”) neither accept nor assume any liability, responsibility or duty of care for any consequences of you or anyone else acting, or refraining to act, in reliance on the Materials or for any decision based on them. No part of the Materials is to be copied, reproduced, distributed or disseminated in any way without the prior written consent of the Company.