Is Value-Added and Opportunisitc Real Estate Investing Beneficial ...

Is Value-Added and Opportunisitc Real Estate Investing
Beneficial? If So, Why?
James D. ShillingΨ
Charlie Wurtzebach†
DePaul University
DePaul University
April 11, 2010
Abstract
There has been a great deal of interest in the question of whether value-added and oppor-
tunistic real estate investing has resulted in appropriate risk-adjusted returns. However,
a satisfactory answer to this question has not been agreed upon. In this paper, data
from the National Council of Real Estate Investment Fiduciaries (NCREIF) property
database are examined to bring new evidence to bear on the subject. Using these data,
ex post returns are calculated for all sold properties. Then groups are formed based
on these returns. A series of discriminant functions are then estimated to relate mem-
bership in these groups over time to value-added and opportunistic indicator variables
(i.e., risk exposures) and market conditions. Results demonstrate that while value-added
and opportunistic private equity real estate investments have higher returns than core
investments, their superior returns are driven by beleaguered market conditions as well
as by the use of cheap debt.
Keywords: Asset Pricing, Pension Funds, and Other Private Financial Institutions
JEL Classification: G12 and G23
Ψ DePaul University, 1 East Jackson Boulevard, Chicago, IL 60606, Email: shilling@depaul.edu.
† DePaul University, 1 East Jackson Boulevard, Chicago, IL 60604, Email: cwurtzeb@depaul.edu.
Funding for this research was provided in part by a grant from the Real Estate Research
Institute. We wish to thank Doug Poutasse and Elaine Worzala for comments on an
earlier draft and the National Council of Real Estate Investment Fiduciaries who provided
data.

---

1
Introduction
This paper develops evidence that returns on value-added and opportunistic private equity
real estate investments are ultimately driven by beleaguered market conditions as well as by
the use of cheap debt.1 These results are certainly not unique to private equity real estate
investments. Tests, such as Baker and Wurgler (2007), Chan (2003), Fisher and Statman
(2000), and Neal and Wheatley (1998) find that asset classes such as value stocks and small
stocks have unusually high returns when the market and investor sentiment are depressed.
Sweeney and Warga (1986), Flannery and James (1984), Lynge and Zumwalt (1980) and
others find that an interest factor is priced for utility stocks and stocks in the banking and
financial services. Bredin et al. (2007), He et al. (2003), Lizieri and Satchell (1997), and
McCue and Kling (1994) and others find that REIT stocks are highly sensitive to interest
rate risk.2
Few empirical studies have directly examined whether conditions of the market or risk are
the determinants of the returns on value-added and opportunistic private equity real estate
investments. The basic difficulty in testing how value-added and opportunistic private equity
real estate investments are priced is the general lack of (transactions-based) returns data.3
This paper presents an alternative test of the pricing of value-added and opportunistic
1 Piazzesi and Schneider (2009) find that recent financial developments may have increased the potential for
the use of cheap debt to drive real estate prices. In their model, agents who suffer from inflation illusion see
the existence and increased availability of cheap debt as an opportune time to borrow money to invest in real
estate, so property prices will climb. Rising property prices, in turn, imply an increase in the contemporaneous
return, holding rents constant. In contrast, in periods of high nominal interest rates, agents who suffer from
inflation illusion invest in bonds because they see bonds offering high nominal interest rates. However, smart
investors subtract high inflation from nominal rates of interest on bonds and realize that real rates of interest
are low. These investors borrow from the agents who suffer from inflation illusion, using real estate as collateral,
causing property prices to increase. Thus, the Piazzesi and Schneider model is capable of accounting for asset
pricing booms when nominal interest rates – and inflation rates – are both high and low.
2 On the other hand, Mueller and Pauley (1995) find that REIT stocks are insensitive to changes in interest
rates.
3 Geltner (1998), Quan and Quigley (1991), Edelstein and Quan (2006), among others suggest that appraisal
smoothing errors engender an underestimation of both the first and second moments for real estate returns.
Thus, one must either deal with transactions-based data, or somehow unsmoothed the appraisal-based returns.
1

---

private equity real estate investments. The test is based on a property of asset returns that,
to our knowledge, has not previously been exploited. This property is that core, value-added,
and opportunistic private equity real estate investments are normally underwritten to earn
leveraged rates of return of 8 to 12%, 12 to 18%, and 18+%, respectively. Thus, if we were
to sort private equity real estate investments into portfolios based on past performance, and
evaluate the combination of variables that reliably discriminate between these portfolios, we
can conduct a direct test of whether high returns are caused by conditions of the market or
the type of investment strategy followed.4
Normally, value-added and opportunistic private equity real estate investments involve
investments in a wide variety of assets, including residential apartments, office buildings, retail
centers, and industrial properties. Generally, these investments are exposed to a high degree of
risk, as they typically involve a significant amount of “value creation” through redevelopment
or development. In contrast, the investment profile of a core investment is similar to that
of a bond, with stable cash flows over long periods of time. For these reasons, the “greater
the risk associated with a financial decision, the greater the return expected from it” theory
suggests that we should be able to classify each and every property investment into one of
the return groups based on the investment style followed and the risks taken on. However,
from a characteristics-pricing model perspective (Daniel and Titman (1997), and Daniel et
al. (1997)), group membership may also be determined in large part by the conditions of the
market and individual property attributes.
To date, the literature has not reached a consensus on the role of characteristics or risk
in explaining asset returns. Previous research by Lakonishok, Shleifer, and Vishny (1994),
4 The test proposed here is a joint test of rationality and asset pricing which embodies a rationality constraint
that ex ante expectations (as reflected in typical underwriting criteria) are realized in the long run.
2

---

Chan and Lakonishok (2004) among others suggests that the high returns to value and op-
portunistic investing, at least for stocks, is not attributed to risk adjustments. Instead, this
research suggests that cognitive biases underlying investor behavior and the agency costs of
professional investment management are the main cause of the high returns to value-added
and opportunistic investing.
Within the real estate literature, several researchers have studied whether value-added
and opportunistic investment styles ubiquitously generate abnormal returns for real estate
investors. For example, using data on real estate mutual funds and using style descriptors
created for real estate related securities, Lin and Yung (2004), Gallo, Lockwood and Ruther-
ford (2000), Damodaran and Liu (1993), and Kallberg, Liu and Trzcinka (2000) provide some
evidence that value investment strategies provide superior returns for real estate investors.
However, using the same data as Lin and Yung (2004) but over a much shorter time period,
O’Neal and Page (2000) reach the opposite conclusion.
Why re-investigate the issue of what determines the returns on core, value-added, and
opportunistic private equity real estate investments now? First, if the high returns on value-
added and opportunistic real estate investments are indeed based on conditions of the market
rather than risk, the implications for portfolio analysis and performance evaluation are strik-
ing. Value-added and opportunity funds generally sport higher fees to justify the process of
searching for high-return investments.5 Clearly, then, some sort of positive abnormal perfor-
mance (asset selection skills) is needed to justify investing in real estate funds that specialize
in value-added and opportunistic investments rather than core investments.
5 Value-added and opportunity funds typically have targeted return hurdles above which incentive fees
(e.g., 20% of profits) can be earned. Anecdotal evidence suggests that these hurdle rates are set at return
levels between 8 and 12%, since this is basically the rate at which core investments are expected to perform.
Typically, there is also a high watermark to ensure that underperformance is made up before any performance
fee can be charged. Most value-added and opportunity funds will also charge an asset-based management fee
that is 1 to 2% of the assets under management.
3

---

Second, the empirical literature has not been able, as yet, to provide a satisfying answer to
the question, Is there something fundamentally different about core, value-added, and oppor-
tunistic real estate investments, or are the higher returns on value-added and opportunistic
real estate investments simply due to characteristics or conditions of the market? Plainly, if
asymptotic arbitrage is possible, and prices move freely so that real estate markets are always
in equilibrium (or moving to it), we would expect property returns to load, on average, only
on risk.
Third, the history of value-added and opportunistic investing in the US is essentially a
tale of excess capital flows during the mid- to late 1980s and a global real estate crash in the
late 1980s and early 1990s (see Conner and Liang (2003)). If these conditions kept real estate
markets out of equilibrium during the late 1980s and early 1990s, we would generally expect
to find high rates of return on value-added and opportunistic investments. Similarly, as prices
of properties in general begin to rise, one should see these abnormally high returns fall.6
Our study is in the same vein as the recent work by Peyton (2008), who provides a
statistically sound technique for identifying the different return cutoff points that can be
used for separating core investments from value-added investment styles, and value-added
investments from opportunistic investment styles. However, we differ from Peyton (2008)
since we derive and estimate discriminant functions which are used to differentiate properties
into the factor group memberships listed above. We also differ from Peyton (2008) in that,
6 If the cross-sectional variation in expected returns on value-added and opportunistic private equity real
estate investments can be explained by beleaguered market conditions as well as by the use of cheap debt, as
prices of properties in general begin to rise, one should see these abnormally high returns fall, unless, of course,
value-added and opportunistic funds were to begin to operate in riskier markets elsewhere outside of the US
where uncertainty is relatively high, or if interest rates were to remain low and value-added and opportunistic
funds were to take on more debt. Also, as prices of properties in general begin to rise, we would expect high
prices to be bid for core investments, and returns to fall as well producing style-creep and/or style-gaming
(where fund managers leave their particular style or take on more debt in order to boost sagging performance).
Because of this style-creep or style-gaming on the part of core fund managers, the relative returns on core,
value-added, and opportunistic real estate investments could converge, making it difficult to find statistical
significance among the average returns on the different funds.
4

---

in our model, we measure performance by the internal rate of return rather than the de facto
NPI rates of return.
Like us, the aforementioned studies by Lin and Yung (2004), Gallo, Lockwood and Ruther-
ford (2000), Damodaran and Liu (1993), and Kallberg, Liu and Trzcinka (2000) examine
whether (or not) value-added and opportunistic investment styles generate high returns for
real estate investors. But the focus in these studies is much different. These studies use fund-
level data to determine whether (or not) value-added and opportunistic investment styles
generate high returns. We use property-level data to test the same hypothesis. This data
allows us to judge all properties (or more technically, all sold properties) by their returns on
investment and to classify these observations into different factor group membership, which
is important in judging whether core, value-added and opportunistic investment styles span
the space of realized values of the factor loadings.
Our evidence suggests that much of the high return for value-added and opportunistic
investment styles is the result of leverage. However, we also find that market conditions, i.e.,
business cycle expansion and contraction,and the use of cheap debt can discriminate private
equity real estate investments with high returns from those investments with moderate to low
returns. This part of the paper is motivated by the fact that there may be better times for
investing in value-added and opportunistic private equity real estate investments than others,
and that these times are related to beleaguered market conditions i.e., recession periods, and
times when debt is cheap compared to the cost of equity.
We also test for and find some evidence of style-creep among core property managers,
but not among value-added and opportunistic fund managers. We rationalize these results in
terms of manager compensation packages which provide strong incentives for core property
5

---

managers to take on more leverage (within limits) in order to achieve a “target” rate of return,
especially when property prices are rising and yields are low. In contrast, we generally find
that value-added and opportunistic fund managers consistently use a high amount of debt to
fund deals regardless of the specific market conditions.
The paper proceeds as follows. Section 2 contains brief background information on value-
added and opportunistic private equity real estate investments.
In Section 3 we explain
the data used in this study. The main task, addressed in Sections 4 and 5, is to compare
and evaluate the returns on core, value-added, and opportunistic investments, and determine
whether we can classify properties into predetermined groups based on risk characteristics or
market conditions. Section 6 concludes the study and offers areas for further research.
2
Value-Added and Opportunistic Funds in Real Estate
The goal of most value-added real estate funds is to achieve a 12 to 18% return. These funds
raise money through commitments primarily to blind-pool limited partnerships. These limited
partnerships invest in all major property types, plus other retail, hospitality, senior living, and
storage. Value-added funds characteristically require significant capital expenditures to allow
for rent growth. These capital improvements may include tenant improvements, upgrading
facades and signage, and curing deferred maintenance in the areas of roof, parking lot and
HVAC systems to improve the property’s return and marketability. However, there is generally
a lot of uncertainty about future real estate prices, and about the appropriate level of capital
expenditure that will maximize the current value of the investment. Thus, in compensation
for this high risk, value-added investments will in general generate high returns.
Normally, opportunistic investments are exposed to an even higher degree of risk than
6

---

value-added funds, as they typically involve a significant amount of “value creation” through
the development (rather than the redevelopment or rehabilitation) of residential, commercial,
industrial, and other uses (including speculative development that is for sale or rent) that
generate income, which is not known, nor can be known, a priori. Most opportunistic real
estate funds are organized in the same manner as value-added funds. These opportunistic
funds generally seek to generate an 18+% internal rate of return for the fund and its investment
period is typically over a relatively short period of time.
According to Private Equity Real Estate (2009), the 30 largest value-added and oppor-
tunistic real estate funds raised more than $20 billion last year (in 2009) and more than
$210 billion over the past five years. Among these funds, the five largest – The Blackstone
Group, Morgan Stanley Real Estate Investing, Goldman Sachs Real Estate Principal Invest-
ment Area, Colony Capital, and Beacon Capital Partners, – raised more than $80 billion over
the past five years (accounting for almost 40% of the total capital raised). Overall, The Black-
stone Group and Morgan Stanley Real Estate Investing were the most active when measured
by value of a fund’s sales. The two funds transacted in more than $147 billion of property
sales between 2004 and 2009. Further, when measured by the number of transactions, The
Blackstone Group was again the most active, buying and selling 1,892 properties between
2004 and 2009. Altogether, the largest number of transactions over this period were for office
and retail properties. They accounted for more than 52% of all transactions by the 30 largest
real estate funds over this period (Real Estate Analytics (2009)).
Most opportunistic real estate funds will typically employ high leverage, with loan-to-value
ratios in excess of 65%. In contrast, value-added funds tend to use higher leverage ratios than
core property funds, but lower than opportunistic funds. For example, the loan-to-value ratio
7

---

for value-added real estate funds is typically between 50 and 65%. Clearly, the use of high
leverage is a reason why value-added and opportunistic investments should earn high returns,
but is it the only reason? Obviously, value-added and opportunistic investments are also
exposed to a variety of risk characteristics such as exposure to development and significant
leasing risk. It has been suggested by Fama (1991), Fama and French (1993, 1996), and
Davis, Fama, and French (2000), among others that the higher returns on value-added and
opportunistic funds are simply compensation for higher systematic risk, at least for stocks.
The vast majority of investors today appear to have actual allocations to real estate, pri-
vate equity, and hedge funds in excess of target allocations, mainly as a result of a shrinking
total asset base stemming from large stock losses. For example, according to Kingsley As-
sociates (2009), actual allocations to real estate are 56 basis points above target allocations,
levels not seen since 2002. Obviously, as a result, most investors are not too interested in
allocating new capital flows to real estate. However, of the new capital that was allocated
to real estate in 2009, most of this capital was earmarked to value-added and opportunistic
funds. A “windows-of-opportunity” hypothesis would explain this behavior as follows. In ab-
stract terms, suppose that sellers are distressed. Further suppose that buyers are distressed as
well. Under these difficult market conditions, assets are generally sold to industry outsiders
or non-traditional real estate investors including hedge funds. But industry outsiders face
significant costs of acquiring the assets. In addition, these investors generally fear overpaying
for the assets because they cannot value the asset properly. As a result, asset prices fall
well below value in best use. If this model is correct, then in the study of value-added and
opportunistic real estate funds, there literally may be windows of opportunity in which real
estate markets are not integrated, and in which value-added and opportunistic investments
8

---

may lead to higher returns.
3
The Sample
We gathered a sample of core, value-added, and opportunistic real estate investments in order
to make the proposed comparison between the returns on value-added and opportunistic in-
vestments and the returns on core properties. To be classified as a core property, the property
must be fully operational and fully let, or close to fully let, generally involving little capital
expenditure after purchase, and have a loan-to-value ratio between zero and 50%. To be
classified as a value-added property, the property must be characterized by active manage-
ment and substantial value-added expansion or conversion; the latter generally involving a
significant enhancement (greater than 10% of market value) or a change in use of the property
from lower use to a higher and better use (e.g., the conversion of industrial properties into
office, or the conversion of rental apartments into condominiums, etc.). In addition, value-
added properties must have a loan-to-value ratio between 50 and 65%. To be classified as
an opportunistic investment, the property must be a new development opportunity or pre-
development property, or a more speculative investment requiring an initial leasing program
to attract new tenants. In addition, opportunistic investments must have a loan-to-value ratio
in excess of 65%. We used the NCREIF (National Council of Real Estate Investment Fidu-
ciaries) database for sampling core, value-added, and opportunistic real estate investments.
The data start in the fourth quarter 1978 with quarterly updates over time.
We first summarize some basic facts about the NCREIF sample. In this part of the anal-
ysis, the total number of properties includes both sold and unsold properties, and properties
both included and excluded from the NPI (NCREIF Property Index).7 Table 1 shows the
7 In the selection of properties to be included in the NPI, properties must meet three major requirements:
9

---