Differences between diversified and specialized real estate investment trusts in the United States

(1)

Differences between Diversified and Specialized Real Estate

Investment Trusts in the United States

Bachelor’s Thesis Economics & Business Finance & Organization, University of Amsterdam

Supervisor: Prof. dr. Peter van Gool FRICS Philip P. Brouwer, Student number: 10676880

(2)

Abstract

This study examines the total monthly returns of ‘diversified’ and ‘specialized’ Real Estate Investment Trusts (REITs) over the time span of 2002 until 2016 and splits that period up into three sub periods. Namely the period prior to the financial crisis 2002-2006, the period during the peak of the financial crisis 2007-2011 and finally the period during the recovery 2012-2016. The sample fourteen REITs have been selected on the basis of the fact that they only invest and own holdings in the U.S.; have a minimal of 500 million U.S. dollars in assets and are publicly traded on the New York Stock Exchange at least from 2002. Finally the REITs need to have survived the financial crisis. The sample consists of six ‘diversified’ and eight ‘specialized’ REITs.

The used methodology in this study incorporates t-tests, variance tests and multiple regression analyses.

The hypotheses that diversified REITs generate higher total monthly returns when examining the total period compared to specialized REITs cannot significantly be supported. However evidence is found that diversified REITs have significantly higher volatility opposed to the specialized REITs.

Statement of Originality

This document is written by Philip P. Brouwer who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

(3)

Content

Abstract - 2 –

List of tables and graphs - 3 -

Used abbreviations - 4 -

Chapter 1: Introduction - 6 -

Chapter 2: Literature Review - 9 -

Chapter 3: REIT Descriptions - 11 -

Chapter 4: Data - 15 -

Chapter 5: Methodology - 19 -

Chapter 6: Results - 21 -

Chapter 7: Conclusion and future research - 28 -

(4)

List of tables and graphs

Table 1: Sector specifications of used REITs - 11 -

Table 2: Key financial aspects of used REITs - 15 -

Graph 1: Total monthly returns of the All Equity REITs from the NAREIT - 16 - Graph 2: Annual returns of the sectors in the order off: Diversified,

Retail, Residential, Industrial and Office. - 16 -

Table 3: Time periods used in the study - 18 -

Graph 3: the NAREIT All Equity REIT index for the entire period - 18 - Table 4: Summarized total monthly returns over the periods - 21 -

Table 5: Summarized t-test statistics - 21 -

Table 6: Variance ratio test p-values - 21 -

Table 7: Total monthly return results over the first period - 22 - Table 8: Total monthly return results over the second period - 23 - Table 9: Total monthly return results over the third period - 24 - Table 10: Total monthly return results over the entire period - 24 - Table 11: Correlation table of the REITs that are part of the S&P

1500 Index as well as included in the sample - 26 -

Table 12: Correlation table of the total monthly returns on Portfolio,

Diversified, Specialized, All Equity REITs and the S&P 1500 - 26 - Table 13: R-Squared, adjusted R-Squared and p-value of the multiple

regression model - 27 -

(5)

-Used abbreviations

TMR: Total Monthly Return

CRPS: Center for Research in Securities Prices IND: Industrial

NAREIT: National Association of Real Estate Investment Trusts NYSE: New York Stock Exchange

OFF: Office

RER: TMR All Equity REITS RSP: TMR S&P 1500 Compose Index REIT: Real Estate Investment Trust RVNO: TMR Vornado Realty Trust RFCE: TMR Forest City Realty Trust RSLG: TMR SL Green Realty Corporation ROLP: TMR One Liberty Properties RCLI: TMR Mack-Cali Corporation RWRE: TMR Washington REIT RGGP: TMR General Growth Properties RKIM: TMR Kimco Realty Corporation REQR: TMR Equity Residential

RACC: TMR American Campus Communities RFR: TMR First Industrial Realty Trust

RMNR: TMR Monmouth Real Estate Investment Corporation RBXP: TMR Boston Properties

RBND: TMR Brandywine Realty Trust RDIV: TMR Diversified REITs RSPEC: TMR Specialized REITs RPORT: TMR Portfolio

RES: Residential RET: Retail

(6)

1. Introduction

As a result of my long time interest in real estate markets and properties I am writing my thesis in the area of the performance of Real Estate Investment Trusts (REITs) in the United States of America. The field of REITs is interesting because of their relative high liquidity and

transparency in real estate markets, as well as their high cash dividends and avoidance of double taxation. In my opinion though their biggest advantage is the possibility for investors to simply diversify their real estate exposure both by property type as well as by geographical spread.

REITs are companies that own, operate or finance income producing real estate. REIT portfolios can consist of several different types of commercial real estate and can range from residential and office buildings to hotels, hospitals, shopping centers and industrial real estate. REITs allow anyone to invest in portfolios of real estate assets the same way they invest in other industries, namely through the purchase of individual company stock or through a mutual or exchange traded fund. The stockholders of a REIT earn a share of the income produced through real estate investment without actually having to make the effort of having to buy, manage or finance property.

The U.S. Congress set up REITs in 1960 with the purpose of providing a real estate investment structure similar to the format mutual funds provide for investment in stocks. REITs are strong income vehicles because, in order to avoid incurring liability for U.S. Federal income tax, REITs generally must disburse an amount equal to at least 90 percent of their taxable income in the form of dividends to shareholders according to the definition stated on the REIT official website (https://www.reit.com/what-reit).

There are four types of REITs. Equity REITs: consist of the majority of REITs and are publicly traded. Equity REITs own or operate income-producing real estate. The market often refers to equity REITs simply as REITs. Mortgage REITs provide financing for income-producing real estate by purchasing or originating mortgages and mortgage-backed securities and earning income from the interest on these investments. Public Non-listed REITs: are registered with the Securities and Exchange Commission but do not trade on national stock exchanges.

Private REITs: are offerings that are exempt from Securities and Exchange Commission registration and whose shares are not traded on national stock exchanges

(7)

(https://www.reit.com/what-reit/reit-basics). For my thesis I focus on U.S. Equity that are publicly listed on the New York Stock Exchange (NYSE).

The fourteen REITs used for this thesis consist of six ‘diversified’ and eight ‘specialized’ REITs. Diversified REITs own and manage a mix of property types, they might own portfolios made up of both office and industrial properties. Specialized REITs operate only in one real estate sector, for example just residential real estate. The Wharton database Center for Research in Securities Prices (CRSP)/Ziman US Real Estate Data Series provides historical information on the stock prices used. These REITs are selected based on criteria as stated in Chapter 3. In order to test the performance differences I will use the total monthly returns over the periods and execute a two samples t-test.

It is my aim to research if one can determine a difference between the returns on diversified versus specialized REITs that only invest and own property in U.S. markets. The reason to narrow the search for U.S markets only is that it will present the best outcome of how REITs in the U.S. are performing since all their returns are only generated in the U.S. The research covers the time frame of 2002 up until 2016, subdivided into three periods: 2002-2006 can be indicated as the period prior to the ‘financial crisis’; 2007-2011 can be marked as the ‘peak of the

financial crisis’, while 2012-2016 can be described as the ‘recovery of the financial markets’.

The previous study by Benefield, Anderson and Zumpano (2009) shows that diversified REITs are better performers when the overall economy and market conditions performed well, but indicate that there is limited evidence that specialized REITs would have performed better during a recession. This creates an opportunity to research whether the diversified REITs would outperform the specialized REITs in the recession period as mentioned above.

While the research of Ro and Ziobrowski (2011) showed that diversified REITs outperformed specialized REITs, and that specialized REITs had higher market risks than diversified REITs. These studies used equity REITs compiled of REIT indexes whereas I will use fourteen individually selected REITs that fit the researched criteria.

In short there is no research on how a handpicked selected sample of ‘diversified’ and ‘specialized’ REITs, only investing in the U.S., react to a recession, and how they recovered in the years following the recession.

(8)

Hence my research question: Have diversified REITs significantly outperformed specialized

REITs in the period 2002 until 2016? This will be substantiated with the outcome of my sub

question that will research how the circumstances linked to the three time frames affect the performance outcome.

Null Hypotheses: No difference between the performances of ‘diversified’ REITs and ‘specialized’ REITs during the period of 2002 up to 2016.

The alternative Hypotheses will be that: ‘diversified’ REITs outperform the ‘specialized’ REITs during the period 2002 - 2016.

(9)

2. Literature Review

The study of Benefield (2009) utilizes the performance ranking methodologies of Jensen’s Alpha, Treynor Index, and the Sharpe ratio. These measures were used in combination with multifactor analogs, to compare whether the stock market performance of REITs that are diversified by property type, differ from the stock market performance of REITs specialized in their property holdings. When they took the combined results from the tests as mentioned above, it strongly indicated a significant difference in performance between property-type diversified REITs and property-type specialized REITs. However, they found out that the difference seems to depend on overall market conditions during the sample period and that the diversified REITs are actually the better performers when overall markets are performing well. Their results provided limited evidence that specialized REITs perform better when overall market conditions are not as favorable. Benefield classified ‘specialized’ as REITs that have 75% or more of their portfolio invested in a particular property type for more than half of the sample period.

‘Diversified’ is therefore classified as a REIT, which has less than 75%, invested in any one property type over a time span of more than half of the examined period.

Some of the other findings of the study are that ‘diversified’ REITs seem to have a significantly higher percentage of office properties in their portfolios when compared to specialized REITs based on the data they received from the National Council of Real Estate Investment Fiduciaries. That fact could explain their counterintuitive results for the period of 1995 - 2001 if office property outperformed other properties. It also showed that diversified REITs seem to devote significantly less of their investments into residential and retail properties as opposite to specialized REITs.

Ro and Ziobrowski (2011) state that REITs which limit their portfolio to a single property type typically preserve their absence of diversification by claiming the management possesses specific investment expertise in that particular property type. Therefore their paper examines whether specialized REITs outperform diversified REITs, and thus providing proof of superior management judgement associated with their sector specialized REITs. Ro and Ziobrowski compared the performance of specialized versus diversified REIT portfolios during the time span of 1997 up to 2006, investigating abnormal returns using the CAP-Model as well as the Fama-French three-factor model with momentum. In their findings they found no evidence of superior performance associated with REITs specializing in a single property type. Ro and Ziobrowski did find evidence of the fact that specialized REITs contain a higher market risk in

(10)

comparison with diversified REITs. Opposed to their expectations they even found some, but not significant, evidence that diversified REITs slightly outperform specialized REITs.

The fourteen REITs used for this study, are all of different sizes; therefore I think it is important to take earlier research that takes size into account, also because, according to the research conducted by Lin and Yung (2004), size is positively related to risk-adjusted real estate fund performance. Morri and Lee (2009) suggest that larger real estate mutual funds have some advantages over smaller ones, where the economy of scale, for instance, can be expressed in a lower expense for overhead costs. They also mention that larger real estate mutual funds possess better bargaining and negotiation power as opposed to smaller real estate mutual funds. Even though it is a research conducted in Italy, the essential findings are useful as comparative material.

The research of Mueller (1998), which examines the period 1993 - 1997, questions whether a bigger size of a REIT is actually better since the earnings per share growth decline as the number of outstanding shares increases according to his study. In order so solve that problem Mueller suggests that REITs should sell properties that lack growth potential and reinvest in properties that have higher growth protections. Mueller also suggests that large REITs may be forced to split up into smaller regionally focused REITs in order to recapture profitable growth. On the other end Mueller states that smaller REITs that manage less than $500 million have been just as profitable in the examined period as large REITs because they are able to grow more profitably from a smaller base.

(11)

3. REIT Descriptions

Hereunder listed and explained the REITs chosen for this research. The criteria for these REITs are; publicly traded on the New York Stock Exchange (NYSE) since, at least, 2002, with minimum assets of 500 million U.S. dollars and having survived the financial crisis. The final criterion is that these REITs only invest in the U.S. and do not have holdings outside the U.S.

Ticker Primarily Type Sector

Diversified REITs

1. VNO Office/Retail Diversified

2. FCE Office/Residential/Retail Diversified

3. SLG Office/Retail Diversified

4. OLP Retail/Industrial Diversified 5. CLI Office/Residential Diversified 6. WRE Office/Residential/Retail Diversified

Specialized REITs

7. GGP Regional Malls Retail

8. KIM Shopping Centre Retail

9. EQR Apartment Residential

10. ACC Student Housing Residential 11. FR Distribution Centre Ind. 12. MNR Net Leased/Long Term Ind. 13. BXP Class A Offices Office

14. BDN Urban Offices Office

Table 1

1.Vornado Realty Trust (VNO) (Diversified) (NYSE)

Vornado Realty Trust was founded in 1982 and based in New York. Vornado is one of the largest publically traded REITs on the NYSE and mainly invests in office buildings and retail in Manhattan but also has some properties in San Francisco and Chicago. Vornado has been recognized by the NAREIT as the highest ranked diversified REIT.

2.Forest City Realty Trust (FCE) (Diversified) (NYSE)

Forest City Realty Trust, which has its headquarters in Cleveland, Ohio, went public in 1960 as one of the first REITs on the NYSE. Forest City’s investments are diversified among office buildings, residential apartments and shopping centers throughout the areas of Boston, Chicago, Dallas, Denver, Los Angeles, Philadelphia, New York City, San Francisco and Washington.

(12)

3. SL Green Realty Corporation (SLG) (NYSE) (Diversified)

SL Green was founded in 1997 and is listed on the NYSE as well as the S&P 500 index. SL Green’s primarily investments are in the New York City office and retail markets. SL Green also invests in multi-family residential homes in the greater New York City area.

4. One Liberty Properties, Inc. (OLP) (NYSE) (Diversified)

One Liberty Properties was founded in 1982 and acquires, owns and manages a geographically diversified property portfolio spread over 30 states. The portfolio primarily consists of property in the segments of retail, industrial, restaurant, health and fitness, and theater properties.

5. Mack-Cali Realty Corporation (CLI) (NYSE) (Diversified)

Mack-Cali Realty Corporation owns real estate throughout the Northeast of the U.S. in

waterfront and transit-oriented markets. Transit-oriented developments are urban types creation of compact, walk-able, pedestrian-oriented, mixed-use communities centered around high quality train systems according to the transit-oriented development website

(http://www.tod.org/). The major part of Mack-Cali’s holdings are Class A office buildings, and investments in multi-family residential properties. Its headquarters is located in Jersey City, New Jersey. Since 1994 Mack-Cali has a listing on the NYSE and has been a component of the S&P MidCap 400 Index since 2003.

6. Washington Real Estate Investment Trust (WRE) (NYSE) (Diversified)

As the name suggests, Washington REIT focuses exclusively on the Washington metropolitan area. The Washington REIT portfolio is diversified in the segments of offices, retail and multi-family real estate. Washington REIT shares are publicly traded on the NYSE since 1996.

7. General Growth Properties Inc. (GGP) (NYSE) (Retail)

General Growth Properties with its headquarters in Chicago went public on the NYSE in 1970 and is incorporated in the S&P 500 Index. General Growth Properties primarily focuses is on owning, managing, leasing and redeveloping high-quality retail properties such as shopping centers throughout 40 states.

(13)

Kimco Realty Corporation is publically traded on the NYSE since 1991 and is included in the S&P 500 Index. Kimco is headquartered in New York and is one of the largest publicly traded holders and exploiters of open-air shopping centers across the U.S. in metropolitan areas.

9. Equity Residential (EQR) (NYSE) (Residential)

Equity Residential its main focus is on the acquisition, development and management of

apartment units located in Boston, New York City, Washington DC, Seattle, San Francisco, and Southern California. As of 1993 Equity Residential has had a public listing on the NYSE and is listed on the S&P 500 Index.

10. American Campus Communities (ACC) (NYSE) (Residential)

American Campus Communities was founded in 1993 and is the largest private dormitory manager in the U.S. and holder and developer of both on and off-campus dormitories for schools that outsource dormitory development and management. American Campus Communities is also a component of the S&P MidCap 400 Index.

11. First Industrial Realty Trusts (FR) (NYSE) (Industrial)

First Industrial Realty Trust went public in 1994 and is based in Chicago. First Industrial invests in over 20 states across the United States. First Industrial owns, manages and develops

distribution centers, light industrial and alternative industrial facility types.

12. Monmouth Real Estate Investment Corporation (MNR) (NYSE) (Industrial)

Monmouth Real Estate Investment Corporation, based in Freehold New Jersey specializes in industrial real estate and owns single tenant, industrial buildings and leased to investment-grade tenants or their subsidiaries on long-term net leases. Most of its real estate assets are located near airports, transportation hubs and manufacturing plants across 30 states. Monmouth was founded in 1968 and is listed on the NYSE since 1987.

(14)

13. Boston Properties (BXP) (NYSE) (Office)

Boston Properties is listed on the NYSE since 1997 and is based in Boston. Boston Properties is a primarily investor in Class A office space and is one of the largest owners and developers of Class A office properties in the U.S. Its property focus lies on the five markets of Boston, Los Angeles, New York, San Francisco and Washington, DC.

14. Brandywine Realty Trust (BDN) (NYSE) (Office)

Brandywine Realty Trust is based in Philadelphia and mainly invests in office buildings and has its focus on the Austin, Washington, D.C., and Philadelphia markets. Brandywine owns and manages an urban and transit-oriented portfolio and is listed on the NYSE since 1986.

(15)

Table 2

In the above table 2 the key aspects of the fourteen REITs are summarized. The general finances of the revenue, net income, assets, equity and debt are in millions. Next are the debt to equity ratio, earnings per share and the number of employees. The purpose of this table is to provide a clear overview of the main economic situation of all the used REITs in 2016 except for Equity Residential. Please note that for Equity Residential I have used the finances of 2015 because the company had one sale of over four billion US dollars so that their net income was over three billion, this would give the false impression that the company delivers that amount of net income every year.

(16)

Graph 1 represents the total monthly returns of all equity REITs, based on the data of NAREIT.

Graph 2 shows the total annual returns of the examined sectors (Annual in this graph because the

total monthly returns give an unclear visual impression of the period.) The terms in graph 2 are Diversified, Retail, Residential, Industrial and Office respectively.

-40 -30 -20 -10 0 10 20 30 40 200201 200206 20021 1 200304 200309 200402 200407 200412 200505 200510 200603 200608 200701 200706 20071 1 200804 200809 200902 200907 200912 201005 201010 201 103 201 108 201201 201206 20121 1 201304 201309 201402 201407 201412 201505 201510 201603 201608

Total Monthly Return All Equity REITs in

%

-80 -60 -40 -20 0 20 40 60 DIV RET RES IND OFF

(17)

4. Data

The purpose of this chapter is to explain how the data are collected and how the methodology is tested. Data on the fourteen REITs studied for this thesis are clearly accessible and honest. The data time span overlaps the period January 2002 to December 2016. The data collected over that same period is collected on a monthly basis. In order to receive the individual returns for the REITs, I have used the Wharton database Center for Research in Securities Prices (CRSP) / Ziman US Real Estate Data Series. CRSP gives the following definition for the return in their database: ‘A return is the change in the total value of an investment in a common stock over some period of time per dollar of initial investment’

(https://wrds-web.wharton.upenn.edu/wrds/query_forms/variable_documentation.cfm?vendorCode=CRSP&libraryCode=crspa&fileCode=msf&id=ret).

The individual returns for the REITs are calculated in the following way:

t’ = previous trading day (t-1) r(t) = return on REIT for day t

p(t) = last sale price or closing bid/ask average for day t d(t) = cash adjustment for day t

f(t) = price adjustment factor for day t

The returns for the S&P Composite 1500 index are also collected with the use of the Wharton database Center for Research in Securities Prices (CRSP). The Standard & Poor (S&P)

Composite 1500 Index contains three indexes namely the S&P 500, S&P MidCap 400 and the S&P SmallCap 600 and covers around 90 percent of the U.S. market capitalization.

(http://us.spindices.com/indices/equity/sp-composite-1500).

The CRSP provides the following approach on how they calculate the returns for the S&P 1500 Index. SPRTRN is the return on the S&P Composite Index defined as:

(18)

SPINDX is the level of the S&P 1500 Composite at the end of the trading month. The S&P 1500 Index is added in the model as a benchmark to review whether the total monthly returns of the REITs move together with the S&P 1500 index over the examined periods.

Furthermore I use the total monthly return on All Equity REITs in the U.S. Those returns are provided by the National Association of REITs and can be accessed on their website.

(https://www.reit.com/data-research/reit-indexes/monthly-index-values-returns)

The classifications of the REITs into the sectors based on the specifications can be found on the site: (https://www.reit.com/what-reit/reit-sectors). The financial aspects of the fourteen REITs that are

summarized in the REIT description, which is part of this thesis, are all drawn from their 2016 Annual Reports.

All collected data have been split up into sub periods of which the exact time span of the periods are stated below.

Time Periods Start End

Full Period 01 January 2002 31 December 2016 Sub Period one 01 January 2002 31 December 2006 Sub Period two 01 January 2007 31 December 2011 Sub Period three 01 January 2012 31 December 2016

Table 3

Graph 3 represents the time-series of the periods used in order to get a clearer view on the market

movements in terms of Equity REITs. The line shows the NAREIT All Equity REIT index for the entire period used in this study and is collected from Yahoo Finance.

(19)

5. Methodology

In this thesis the null hypothesize indicates that diversified REITs do not significantly

outperform the specialized REITs during the period 2002 - 2016. If this null hypothesis would be rejected it would give evidence that diversified REITs produce higher total monthly returns than specialized REITs during the set period, which includes the financial crisis.

In order to find the returns for the periods and to calculate the correlations I have used the following multiple regression model with the monthly total return on the combined portfolio of all the used REITs as the dependent variable and defined it as RPORT. The total monthly return of the S&P 1500 Index is presented, as RSP and the following variables in the model are the returns for each REIT separately. The letters that follow R for return are defined by the REITs separate company tickers, which are again defined in the REITs description part. To complete the model the error term is included in the form of ε.

With this model I estimate how the explanatory variables of the total monthly returns have overall influence on the dependent variable, namely the total monthly returns of the portfolio. When performing a regression on this model I exclude the explanatory variable that includes the return of BND because this would otherwise lead to a situation close to perfect multicollinearity causing the model to lose its independency and to be considered invalid. Multicollinearity can be described as a situation in which two or more explanatory variables in a multiple regression model have some sort of linear relation that can lead to high correlations in a model and thereby having a negative impact on the validity of the multiple regression model (Farrar, Donald E.; Glauber, Robert R. 1967).

RPORT=β0+β1RER+β2RSP+β3RVNO+β4RFCE+β5RSLG+β6ROLP+β7

RCLI+β8RWRE+β9RGGP+β10RKIM+β11REQR+β12RACC+β13*RFR+

β14RMNR+β15RBXP+β16RBND + ε*

With the use of the model the total monthly returns and the standard deviations of the All Equity, S&P 1500 and fourteen used REITs are estimated in STATA. The outcome of this multiple regression will not be of a significant contribution to this thesis since most of the explanatory variables are part of the dependent variable. They are added to provide an

impression of how the explanatory variables have influence on the total monthly return of the portfolio.

(20)

Because of the fact that six of the used REITs are incorporated in the S&P 1500 Index,

multicollinearity could occur to some extent. Therefore a correlations table is framed to confirm that the models are still reliable.

For this thesis I have chosen to give an equal weighting to all the total monthly returns of the REITs. This way all sectors are given an equal importance and will project fair sector

outcomes, compared to market value weighted, where the bigger REITs are given more importance and causing the sector returns to act as the total monthly returns of REITs with the largest market value.

In order to test whether the ‘diversified’ REITs outperform the ‘specialized’ over the entire and sub periods, I use a two-sample t-test with unequal variances. For this test I have created two new variables namely RDIV and RSPEC, which represent the combined returns of the diversified and specialized REITs respectively.

For that reason I have created a second multiple regression model:

RPORT = β0+β1RDIV+ β2RSPEC + β3RER +β4RSP+ε

With the use of this model I will test whether the difference between the two means differs from zero with the use of STATA that will provide the corresponding t- and p-values for all four periods. As well as calculating the correlations over the full period and test it for significance.

For the purpose of testing whether the variances of the diversified and specialized differ over the periods a Variance Ratio test is performed in STATA that will test if the standard deviations are equal over the entire and sub periods, and also whether the standard deviation of the diversified REITs are larger or smaller than the standard deviation of the specialized REITs.

The used formula for calculating the value is stated bellow. The outcomes of smaller, equal or larger than are presented in the results as < 1, = 1 and >1 respectively.

(21)

6. Results

This chapter discusses the findings of my regression models, test statistics and split up the results in the periods examined. With first the results for total monthly returns in percentages and second the results of the t-test and the corresponding p-value. For all the calculated t-values the critical value for testing is done at a significance level of 5%. Since I am testing on whether the diversified REITs outperform the specialized REITs the test is a one-sided on the right side. In table 4 the total monthly returns (TMR) over the periods are summarized. The used terms are Diversified, Specialized, Retail, Residential, Industrial, Office, All Equity REITs and S&P 1500 total monthly returns. Table 5 displays the summarized t-test statistics and table 6 the variance test p-values. Table 6 represents the p-values of the variance ratio test.

2002-2006 2007-2011 2012-2016 2002-2016

R(Div) 1.9785% R(Div) 0.4991% R(Div) 1.0593% R(Div) 1.1790%

R(Spec) 1.7920% R(Spec) 0.0453% R(Spec) 1.1225% R(Spec) 1.0208%

R(Ret) 2.5759% R(Ret) -2.1220% R(Ret) 1.2705% R(Ret) 0.5747% R(Res) 1.3025% R(Res) 1.1964% R(Res) 0.7688% R(Res) 1.0892% R(Ind) 1.3211% R(Ind) 0.4849% R(Ind) 1.7378% R(Ind) 1.1181% R(Off) 1.9685% R(Off) 0.6221% R(Off) 1.1241% R(Off) 1.2382%

RER 1.8391% RER 0.3610% RER 1.0232% RER 1.0745%

RSP 0.4160% RSP -0.0500% RSP 1.0100% RSP 0.4580%

Table 4

Test Statistics T-Test Value P-Value

Full Period 0.1945 0.423 Period One 0.2562 0.3991 Period Two 0.2079 0.4178 Period Three -0.2016 0.5797 Table 5 P-Value's Variance-test If < 1 If = 1 If > 1 Full period 0.9784 0.0432 0.0216 Period One 0.7429 0.5142 0.2571 Period Two 0.9087 0.1827 0.0913 Period Three 0.6929 0.6142 0.3071 Table 6

(22)

2002-2006 TMR St. Dev.

Return All Equity REITs 1.8391% 4.12% Return S&P 1500 Index 0.4160% 3.57%

Return VNO 2.4020% 5.43% Return FCE 2.0288% 4.81% Return SLG 2.9800% 5.77% Return OLP 1.5894% 5.02% Return CLI 1.5289% 5.45% Return WRE 1.3417% 5.11%

Return Diversified REITs 1.9785% 4.15%

Return GGP 2.9197% 6.09% Return KIM 2.2320% 4.99% Return EQR 1.5405% 5.21% Return ACC 1.0644% 3.81% Return FR 1.4806% 5.58% Return MNR 1.1616% 4.23% Return BXP 2.4737% 5.00% Return BND 1.4633% 5.84%

Return Specialized REITs 1.7920% 3.81%

Table 7

Results in period 1 2002-2006

As shown in the table 7 the diversified REITs delivered a combined total monthly return of 1.9785%. That return is higher than the 1.7920% monthly return of the specialized REITs. The t-test gave the corresponding value of 0.2562 and the matching value is 0.3991. Since the p-value is greater than 0.05 it is not significant. Even though the monthly returns are higher for the first period there is insufficient evidence to claim that this is actually true.

(23)

2007-2011 TMR St. Dev.

Return All Equity REITs 0.3610% 9.64% Return S&P 1500 Index -0.0500% 5.46%

Return VNO 0.2570% 11.62% Return FCE 0.1250% 24.91% Return SLG 0.7780% 18.08% Return OLP 1.5315% 17.09% Return CLI 0.0087% 11.42% Return WRE 0.2161% 8.41% Return Diversified REITs 0.4991% 12.95% Return GGP -4.1398% 17.16% Return KIM -0.1047% 15.47% Return EQR 1.0366% 9.59% Return ACC 1.3561% 7.80% Return FR -0.0264% 19.17% Return MNR 0.9962% 6.41% Return BXP 0.6694% 10.18% Return BND 0.5747% 22.23% Return Specialized REITs 0.0453% 10.87% Table 8 Results in period 2, 2007-2011

In the second period, including the financial crisis the diversified REITs delivered a substantial higher total monthly return as apposed to the specialized REITs. As shown in table 8 its total monthly return is over ten times the amount of the specialized REITs.

When also comparing it with the All Equity REITs and S&P 1500 Index it generated a higher total monthly return during the second period. Nonetheless the t-test results show no significant evidence for supporting that statement with a p-value of 0.4178. The Variance test however did provide some proof that the standard deviation of the diversified is indeed higher than that of the specialized REITs over the second period; namely with a p-value of 0.0913 which is significant at a ten percent level, but not at five percent. The greater total monthly return could to some extent be explained by the higher standard deviation and thereby volatility of the diversified REITs opposed to other total monthly returns.

(24)

Results in period 3, 2012-2016

In the final sub period the results show that total monthly results are on the contrary. As shown in table 9 the specialized REITs produced a total monthly return of 1.2253%, which is above that of the diversified REITs. As a direct result the t-test gives a negative value of -0.2016 and a corresponding p-value of 0.5797, which is not significant. The negative difference that arose in the final period could be explained by the substantial higher performance in the industrial sector as First Industrial and Monmouth generate considerable higher total monthly returns when compared to the other REITs.

2012-2016 TMR St. Dev.

(25)

2002-2016 TMR St. Dev.

Table 10

Results over the total period 2002 – 2016

The results over the entire period show that the total monthly return of the diversified REITs are considerably higher when compared to the total monthly returns of the specialized REITs, All Equity REITs and the S&P 1500 Index. The calculated t-value of 0.1945 is not significant due to the p-value of 0.4230. Thus unfortunately the t-test shows no evidence for supporting the

hypotheses that the diversified REITs show significant higher total monthly returns when compared to specialized REITs over the same period. The variance test does support the fact that the standard deviation of the diversified REITs is higher than the one of the specialized REITs with a p-value of 0.0216 and thus significant at a five percent level. Thereby a higher volatility could be the reason of the total monthly higher returns.

(26)

The final part of the results consists of the calculated correlations and finally the multiple regression outcomes.

Table 11

Correlation RSP RSLG RCLI RGGP RKIM REQR

RSP 1 RSLG 0.6303* 1 RCLI 0.6018* 0.7326* 1 RGGP 0.5605* 0.6140* 0.5067* 1 RKIM 0.5407* 0.8017* 0.7234* 0.5685* 1 REQR 0.5963* 0.6752* 0.6509* 0.4803* 0.6973* 1 RACC 0.3833* 0.4949* 0.5101* 0.3915* 0.6461* 0.6493* * = Significant at 1%

In table 11, the S&P 1500 shows some, but no extreme correlations with the REITs that are part of the S&P 1500 Index as well as included in the sample. It can be concluded that

multicollinearity is not an issue for the used models. All the correlations are significant at a one percent level.

Table 12

Correlation RPORT RDIV RSPEC RER RSP

RPORT 1 RDIV 0.9823* 1 RSPEC 0.9866* 0.9386* 1 RER 0.9684* 0.9362* 0.9685* 1 RSP 0.6884* 0.6977* 0.6604* 0.6815* 1 * = Significant at 1%

Table 12, shows the correlation between the total monthly returns on; the Portfolio, Diversified, Specialized, All Equity REITs and the S&P 1500 index respectively. All the correlations are significant at a one percent significance level. From the correlations it can be implemented that there are very high relations between the portfolio, diversified, specialized and all equity REITS. Since these all operate in the similar areas of the U.S. market this is a plausible outcome. The correlations between the total monthly returns of the S&P 1500 Index and the rest is also logical because one expects there to be a positive relation between those, but not substantially high.

(27)

Table 13 represents the R- Squared, adjusted R-Squared and the p-value of the model.

Table 14 lists the outcome of the multiple regression with the coefficient, standard error, t-value and corresponding p-value respectively over the entire period.

Table 13

R-Squared 0.998 Adj

R-Squared 0.9979

Prob > F 0

RPORT Coefficient Std. Err. T-value P-value

RER 0.06566 0.21239 3.09 0.002 RSP -0.02728 0.00968 -2.82 0.005 RVNO 0.06706 0.00890 7.53 0 RFCE 0.09242 0.00330 27.81 0 RSLG 0.10533 0.00547 19.27 0 ROLP 0.07120 0.00387 18.39 0 RCLI 0.08984 0.00647 13.88 0 RWRE 0.07762 0.00858 9.05 0 RGGP 0.05614 0.00384 14.64 0 RKIM 0.09571 0.00671 14.27 0 REQR 0.04262 0.00780 5.41 0 RACC 0.07546 0.00674 11.2 0 RFR 0.06680 0.00394 16.95 0 RMNR 0.05234 0.00661 7.92 0 RBXP 0.05237 0.00978 5.35 0 Constant -0.00012 0.00028 -0.43 0.671 Table 14

Since this model was constructed in a manner that most explanatory variables are expressed to some extend in the dependant variable, these outcomes are very likely. Therefore I will not use any of these outcomes for my conclusion but they serve as an impression.

(28)

7. Conclusion

In this thesis the performance of six ‘diversified’ and eight ‘specialized’ REITs have been studied over the period of 2002 until 2016.

The aim was to research if a difference can be determined between the returns on

‘diversified’ versus ‘specialized’ REITs investing and holding property in the U.S. markets only. The full research period of 2002 up until 2016, was subdivided into three sub periods: the first period 2002-2006 can be indicated as the period prior to the ‘financial crisis’; the second period 2007-2011 can be marked as the ‘peak of the financial crisis’, while the final period 2012-2016 can be described as the ‘recovery of the financial markets’.

These REITs were selected based on a set of criteria to best fit a fair study.

In the null hypotheses it is stated that there is no difference in performances between

‘diversified’ REITs and ‘specialized’ REITs during the period of 2002 up to 2016. Alternative outcome could be that ‘diversified’ REITs outperformed the ‘specialized’ REITs during the period 2002 - 2016.

From the research the following conclusion can be drawn; when simply examining the total monthly return the diversified REITs indeed produce higher returns. Unfortunately the results of the performed t-test showed no significant result of that claim. Thereby the six diversified REITs did not significantly generate higher total monthly returns when compared to total monthly returns of the eight specialized REITs. The variance test did find significant evidence that the standard deviation of the diversified REITs were greater than those of the specialized REITs that indicated higher volatility in diversified REITs. When examining the first period of 2002-2006 diversified REITs achieved greater but no significant higher returns. The total monthly returns of the diversified REITs were substantially higher during the period of 2007-2011 and thereby to some, but not significant, extent there can be claimed that diversified REITs performed better during the financial crisis. In the period of 2012-2016 the specialized REITs on the contrary showed that their total monthly returns were higher, thanks to the fact that the industrial sector outperformed the other sectors.

This study has taught us that diversified REITs, based on this sample, seem to be a better investment during a period of recession when compared to specialized REITs.

In order to get a better insight on the differences between the specialized and diversified REITs future researchers, for this specific period, might consider the extra implementation of

(29)

8. References

Benefield J.D., Anderson R.I., Zumpano L.V. (2009). ‘Performance differences in property-type diversified versus specialized real estate investment trusts (REITs)’. Review of Financial

Economics vol. 18, 70-79.

Farrar, D.E.; Glauber, R.R. (1967). "Multicollinearity in Regression Analysis: The Problem Revisited". Review of Economics and Statistics vol. 49 (1), 92–107.

Lin, C.Y. & Yung, K., (2004). ‘Real Estate Mutual Funds: Performance and Persistence’.

Journal of Real Estate Research, vol. 26 (1), 69-94.

Morri, G. & Lee, S., (2009). The Performance of Italian Real Estate Mutual Funds. Journal of

European Real Estate Research, 2(2), 170-185.

Mueller, G., (1998). ‘REIT size and earnings growth: is bigger better or a new challenge?’

Journal of Real Estate Portfolio Management 4(2), 149-157.

Ro, S.H. & Ziobrowski, A.J., (2011). Does focus really matter? Specialised vs. Diversified REITs. The Journal of Real Estate Finance and Economics, vol. 42, 68–83.

Annual Reports 2016 of the companies surveyed

• American Campus Communities: http://ir.americancampus.com/sec-filings/Index?KeyGenPage=302584

• Boston Properties: http://ir.bostonproperties.com/phoenix.zhtml?c=120176&p=irol-reports • Brandywine Realty Trust http://investor.brandywinerealty.com/Docs

• Equity Residential : http://www.equityapartments.com/corporate/annualreport.html • First Industrial Realty Trust:

http://www.firstindustrial.com/SecondaryNavigation/investors/annual_reports.php#.Wm9_i5 OdXUI

(30)

• Forest City Realty Trust: http://ir.forestcity.net/phoenix.zhtml?c=88464&p=irol-reportsannual

• General Growth Properties Inc.: http://investor.ggp.com/document-library/2016/ggp-2016-annual-report

• Kimco Realty Corporation:

http://investors.kimcorealty.com/CustomPage/Index?KeyGenPage=205764 • Mack-Cali Realty Corporation

http://investors.mack-cali.com/annual-reports/Index?KeyGenPage=1073752134

• Monmouth Real Estate Investment Corporation: http://investors.mreic.reit/Docs • One Liberty Properties http://1liberty.com/investor_relations/annual_report_form10k • SL Green Realty Corporation http://investor.shareholder.com/slg/annuals-proxies.cfm • Vornado Realty Trust

https://www.vno.com/investor-relations/financial-reports-and-filings/sec-filings?type=2&year=2016&company=3&limit=50 • Washington REIT: https://www.snl.com/IRW/Docs/103036

Supporting websites:

Definition S&P 1500 Compose Index: (http://us.spindices.com/indices/equity/sp-composite-1500).

Definition of REIT Sectors: https://www.reit.com/what-reit/reit-sectors). Definition of REIT: https://www.reit.com/what-reit)

Definition of transit-oriented development: http://www.tod.org/). Description of calculated Returns in the CRSP

Database:(https://wrds-web.wharton.upenn.edu/wrds/query_forms/variable_documentation.cfm?vendorCode=CRSP&li braryCode=crspa&fileCode=msf&id=ret)

Differences between diversified and specialized real estate investment trusts in the United States