The history and performance of Hotel Real Estate Investment Trusts (REITs)

(1)

The history and performance of Hotel Real Estate

Investment Trusts (REITs)

BSc Economics and Business Economics Specialization Finance

Name: Matthew Reek

Student number: 11911581 Supervisor: Dr. N. Edwards Submission date: 29th of June, 2020

(2)

Statement of originality

This document is written by Matthew Reek and I declare 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. The Faculty of Economics and Business is responsible solely for the

supervision of completion of the work, not for the contents.

Abstract

This paper examines the performance of the hotel REIT sector from 2008 to 2019

compared to the broader market of equity REITs and its subsequent property sectors. The history of the equity REIT market will be explored along with the sector-specific history and recent changes within the hospitality industry concerning the management structure of the industry. The performance of hotel REITs will be examined in this paper in the pursuit of providing a clearer view of the recent performance of hotel REITs compared to the broader market and other REIT property sectors.

(3)

Currently, the discussions and news have been on the rise concerning Hotel Real Estate Investment Trusts, or short Hotel REITs, which gained momentum in the 1990s, and have since grown significantly in importance and size (Sarheim, 2006). These investment trusts are currently in the spotlight of financial news because of the downturn amidst the rampant COVID-19

pandemic and have fallen over 40% in the last month (BNP Paribas, 2020). The existence of the REIT market can be traced back to the early 1960s when the U.S. government allowed for the creation of REITs and experienced significant growth in the late 1960s and again in the early 1990s. This was the fundamental start of the investment vehicles and for decades there has been an expansion of REITs into many sectors. (National Association of Real Estate Investment Trusts1, 2020). However, the hotel REITs only gained momentum in the mid-1990s and expanded rapidly in this period gaining popularity and importance within the REIT market (Sarheim, 2006).

The Hotel REITs, also called lodging REITs, are a significant part of this expanding REIT market. At the end of 2019, the total market value of REITs in the U.S. amounted to $1.33 trillion with a total number of 219 REITs in the market. (Nareit, 2020) Focusing on the lodging REITs there are a total of 18 lodging REITs at the end of 2019 with a total market capitalization of roughly $46 billion. This means that the current share of lodging REITs within the total market capitalization is around 3.5%. This is a significant share, but there are many other investment trusts within the real estate market ranging from infrastructure trusts to residential trusts. Within this increasing real estate trusts market a very important question arises with regards to these lodging trusts in comparison to other REITs in this market which will be examined further in this paper. To make viable investment decisions investors are always looking for the answer to which investments are the best allocation within their limited resources. The question that will be examined further in this paper is the following:

- How have hotel REITs performed relative to other REITs property sectors?

1

(5)

This is a broad question with many answers and is fairly difficult yet very important to answer. Hotel REITs are a part of one broader market, in particular, the real estate investment trusts market. De Roos (1997) emphasized the need for further research because of growing hotel capital markets and in this paper, the hotel REITs will be compared to other REITs within the market. Furthermore, it is interesting to compare the hotel REITs to the broader market and compare the performance of the hotel REITs to the S&P 500, which is an alternative to the broader market that is widely used (Han and Liang, 1995). The question of the performance of these REITs does not only depend on the returns that the investment trusts offer, but also on the risks that investors are exposed to. Taking into account all prior notes, the broad question posed before can be narrowed down to the following sub-questions:

- How does the average return of hotel REITs compare to other REITs and the broader

stock market?

- How does the return of hotel REITs compare to the broader market when taking into

account the risk investors are exposed to?

Jackson (2008) has performed a similar study on this for the period of January 1993, when the hotel REITs took off as an investment vehicle, to December 2005 by comparing different REITs to lodging REITs. However, many factors have changed in the industry which asks for further research on the current stance and the period proceeding from the period studied by Jackson (2008). Therefore, in this paper, the focus will mainly be on the period around the financial crisis of 2008 and the decade following this crisis ending at the current stance of the hotel REITs with a focus around the structure within the hospitality industry.

2. Literature review 2.1 Definition and history of REITs

In September of the year 1960, a law was signed that created Real Estate Investment Trusts (REITs) as we know them today. (Nareit, 2020) However, this was not the first foundation of the REITs as we know them today. As Chan et al. (2003) illustrated, the first foundations were laid in the early 1900s when there were examples of Real Estate Investment Trusts, but these businesses needed a REIT-like investment vehicle to operate with real estate being the core

(6)

business of such an investment business and to obtain the benefits associated with such an

investment vehicle. This can be seen as the foundation of an investment vehicle as a REIT, but in this paper, the focus is on the development of the REITs after the Real Estate Investment Trust act of 1960. The REIT act of 1960 was the foundation for REITs we know today and have existed from then on. The main purpose of the creation of real estate investment trusts as investment vehicles was to enable individual investors to profit off of commercial real estate investments, something that was only available to high net worth individuals and institutional investors. (Nareit, 2020). This act was opening up the opportunity for small investors to benefit from ownership in large commercial real estate through these REITs and small investors could receive returns from continuous dividend income.

After the creation of REITs, the current structure as we know them, with the Real Estate Investment Trust act of 1960 and tax provision law that were both signed, several specific requirements were set up for these investment vehicles they had to meet. This was to ensure that REITs were operating as a mutual fund investment vehicle and were financing real estate to generate income for investors. The act that was signed in 1960 gave qualifying REITs a tax-exempt status. (Beals & Arabia, 1998) For a REIT to qualify and be tax-exempted from corporate taxes in the U.S. it should meet many requirements of which the following are the most important ones. (1) At least 90% of the taxable income has to be distributed to shareholders of the REIT (formerly 95% prior to the REIT modernization act of 1995); (2) at least 75% of the assets should consist of mortgages, real estate, cash and government securities; (3) a maximum of 30% of the gross income can be obtained from the sale of stocks or securities sold within six months or real estate within four years; (4) at least 75% of the generated gross income should be gained through rents, mortgages and sale of real estate. (Kim et al., 2002; Jackson 2008) If an investment vehicle meets the prior set requirements it is legally considered a REIT and exempted from corporate taxes consequentially. The changes that the U.S. government has made to its tax laws have affected the REITs in their viability and efficiency. (Semer, 2009)

REITs exist in all sectors and has increasingly expanded throughout the years after its inauguration. A trend that was not seen before was that the REIT industry started segmenting into various property sectors. Today, REITs exist in almost all sectors ranging from office spaces,

(7)

commercial malls to hotels and industrial real estate, and even diversified REITs who hold diversified real estate portfolios across several property sectors. (Kim et al., 2002) This can be mainly attributed to the fact that the market has experienced rapid growth in the 1990s where the market capitalization grew from $8.7 bln at the beginning of the decade to over $136 bln at the end of the decade meaning that the total market capitalization of REITs increased more than fourteen-fold within this decade. This is a trend that is expected by PwC (2018) in the Middle East where the REIT market is currently maturing and property-specific REITs will emerge. When the REIT market matures this property specialization within REITs allow them to

understand the demand within the market and satisfy this demand within their specific sector and extend their knowledge in this field. The segmentation within the REITs is a process that occurs when sector-specific REITs emerge.

This segmentation is a specialization within the property sector, but there is a larger division within REITs that classify them. Firstly, REITs can be classified between publicly traded and private REITs, where the latter is not listed nor traded on a public exchange. (Jackson, 2009) In this paper, the focus is namely on the former because these are the REITs where its stocks are publicly traded and thus priced by a market accessible to the public. Currently, the REIT market accounts for almost $3 trillion in gross assets in real estate of which over two-thirds are held by public REITs, either listed or non-listed (Nareit, 2020). Within the publicly traded REITs, an important distinction is the division of publicly traded REITs into mortgage REITs and equity REITs. Equity REITs are investment trusts that obtain interest from properties and real estate and mainly focus on this segment of the real estate market and mortgage REITs operate by owning mortgage papers which are secured by underlying real estate (Jackson, 2009). Both REITs operate within the real estate and property sector, but where equity REITs acquire and secure property interests, mortgage REITs operate by acquiring interest from mortgages backed by underlying real estate assets. The equity REITs are essential for the fundaments surrounding the creating of REITs as found by Kuhle et al. (2005) This because they found equity REITs to have significant benefits for diversification of the portfolio of small individual investors through real estate, the crucial element in founding the REITs as we currently know them today in 1960, whereas mortgage REITs were seen as too risky for individual investors.

(8)

2.2 History and structure of hotel REITs

One of the property-sector specific REITs within the REIT industry are hotel REITs. Hotel REITs, also called lodging REITs, are equity REITs that acquire interest and capital gains from investing in hotel properties. Lodging REITs have been around for more than half a millennium and can be dated back to 1969 when the first lodging REIT was created but overall, the lodging REITs were not seen as an attractive investment vehicle. However, at the start of 1993 lodging REITs took off as an investment vehicle and had grown from 2 lodging REITs with a market capitalization of $100 million in early 1993 to 18 lodging REITs at the end of 2019 with a market capitalization of over $46 billion. From a management perspective, equity REITs are set up to manage and own income-producing real estate. (Jackson, 2008) In the case of lodging REITs, this has the implication that lodging REITs own hotel and leisure real estate that produce interest and income which can be passed onto shareholders who hold lodging REITs’ shares. Gu & Kim (2003) found that between 1993 and 1996 over $1.4 billion had been acquired within lodging REITs since the reintroduction in 1993.

Gu & Kim (2003) have identified multiple reasons for the growth of the sector in the period of 1993 onwards. First of all, an important act regarding taxation was passed in 1993 in the U.S. called the Revenue Reconciliation Act. The tax legislation in this act removed a major barrier concerning taxation which made equity REITs a more attractive investment not only to individual investors, but mostly to institutional investors. The aforementioned requirements to qualify as a REIT were set up within the Internal Revenue Code. Before the Revenue

Reconciliation was put in place, a fifth requirement was set that was disadvantageous to REITs position to attract institutional investors. This was the requirement that if a REIT had more than 50% of its shares outstanding with five or less shareholders, and these five shareholders were holding a majority, could not qualify for the exemption of corporate tax laws as a REIT. Cash flows from institutional investors now poured into REITs after the Revenue Reconciliation Act was put into place. Secondly, investors had viewed REITs as a viable investment vehicle to pool financial resources and benefit from real estate investments. An argument that had been further extended by Jackson (2008) that had found that hotel investors purchased into lodging REITs with the expectation and intention that a constant appreciation in real estate values along with the expectation that dividends could be reinvested and therefore, increase the value of the shares held

(9)

by investors. A third reason cited by Gu & Kim (2003) was that the hotel sector in the U.S. was experiencing a rebound around 1992 after an economic recession that took place prior from 1990 to 1991. There had been an overcapacity in the construction sector in the 1980s and combined with an economic recession room occupancy rates and the revenue per room (RevPar) had declined and made the hotel sector less profitable overall. Gu and Kim (2003) had found that the occupancy rates increased by 5 percentage points to 66% along with an increase of the daily rate of $20 to $79 from 1991 to 1998. The profitability over this period had increased significantly from depressed occupancy and daily rates prior to this period. This recovery could have

encouraged investors to buy into lodging REITs as the hotel sector recovered and therefore bring large amounts of new capital into the lodging REITs. Jackson (2008) further build upon this argument by obtaining data from the revenue per room showing an increase from the period of 1991 to 2005, with a decrease surrounding the terrorist attacks on 11th of September, 2001 with a consequential decrease in tourism. Lastly, Gu & Kim (2003) had found that there was an overall perception under investors that holding shares in REITs would provide them with the ability to hedge against inflation.

It is clear that the lodging REITs experienced a rapid growth after the year 1993 with many factors attributing to this growth. An interesting notion that can be made to this growth is the fact that Jackson (2008) found the unsystematic risk of hotel REITs to be exceeding the national average by 14 percent. However, as shown before, this did not limit the growth of lodging REITs over the past period. The growth of lodging REITs as an efficient investment vehicle has had an upwards trend since its re-emergence in 1993. This trend had peaked in 2005 when the total number of lodging REITs was 19 with a market capitalization of $18 billion. The next year, the downward trend had started when the number of lodging REITs decreased to 14 in 2006 and 10 in 2007. The market capitalization peaked in 2006 with a total market capitalization of almost $29 billion which decreased to over $19 billion the next year in 2007. This sharp decline could be attributed to the fact that a market correction took place and weak performing REITs exited the market. (Jackson, 2008) Nevertheless, the market continued its expansion in the years that followed and currently, there are 18 lodging REITs with a total market capitalization of over $48 billion. (Nareit, 2020) This continuous growth, with a correction in the years preceding the financial crisis of 2008, has increased cash flows into the hotel industry and changed the

(10)

ownership structure, prior to the expansion of lodging REITs, within the hotel industry. (Jackson, 2009)

2.3 Structure of the hotel industry

The passive component within the requirements to legally qualify as a REIT withheld investors to view REITs as an attractive investment vehicle. However, after Starwood Hotels and Resorts Trust was formed this view had significantly changed. Starwood Hotels and Resorts Trust had created a paired structure and formed their own Real Estate Investment Trust,

something which was not seen before and were considered a pioneer in this segment. Before this possibility was explored, hotel companies were not keen on converting into REITs since they were only able to lease properties to third parties. This structure came along with high operating costs and fixed costs to run the hotels. The hotel industry suffered because of this structure since they were not able to maximize profits. (Jackson, 2009) After this, as shown before, the lodging REIT market took off and expanded. This structure, laid the ground for a different structure within the hotel industry that has evolved over the years.

Before this development, another creative REIT structure was developed and executed that was the so-called paperclip structure. This paperclip structure closely resembles the paired share REIT that Starwood Hotels and Resorts Trusts pioneered in the sense that it combines the operator, also called C-Corporation, and the owner of the real estate, a REIT in this case. In this cooperative agreement, the operating company has the right, which may be exclusive, to operate the REITs assets. The operating company in this structure is closely linked to the REIT to emulate a common interest in their operations. However, when there is too much divergence between the operating C-Corporation and the REIT, different interests can be pursued by the respective companies. (Beals & Arabia, 1998)

The aforementioned structures have one thing in common and this is the objective to prevent leakage of revenues which would be the case under a structure where the lodging REIT would need to rely on a third-party operator to realize value. (Beals & Arabia, 1998) The optimal REIT structure as described by Beals & Arabia (1998) would be a structure that would legally qualify as a REIT to obtain tax benefits under U.S. law and could own non-passive operating

(11)

companies. Furthermore, the leakage of revenue and potential conflicts of interests should be minimized and a maximum tax efficiency should be achieved together with a low risk within the structure adopted. All of the aforementioned reasons come together with a structure that is currently being adopted in the hospitality industry.

In the 1980s the famous American brands of Marriott and Hilton began cutting back on property on their balance sheets and were moving towards a business model that was viewed as less capital intense and would offer higher returns on their capital. This was a shift from

operating and managing the properties themselves, which is considered the typical hospitality structure, to a structure where fees would be collected for operating the properties. An

incremental part of this process is that the sale of these properties are mostly followed by management contracts. This would mean that a major brand such as Marriott would sell the property, but immediately gain a management contract to manage the property, which can be called a sale and manage-back contract. (Sohn et al., 2013; Page, 2007) This movement can be characterized as a shift from the typical structure within the hospitality industry towards an asset-light and fee-oriented, also called ALFO, structure. The shift towards an ALFO structure is a shift that has occurred throughout the industry and major U.S. hotel companies barely own or lease properties. Major U.S. hotel companies such as Starwood, Hilton, Marriott and IHG held less than 5% of leased or owned properties in the year 2013. (Blal & Bianchi, 2019) An import factor that should be taken into account in the case of major U.S. hotel operators that are branded is the fact that the management contracts that can be obtained are of a longer term than those for independent operators. (HVS, 2014)

A typical management contract that is currently constructed and executed consists mainly of three major management fee components: base fees, incentive fees and, other fees. Bases fees can often be attributed to the majority of compensation within a management contract and are mostly a fixed percentage of the revenue. Also, in this case, it holds that the base fees are higher for brand operators than independent operators. On top of these base fees, incentive fees are set up, because base fees are tied to the revenue and not a profit-related fee. Incentive fees are paid to the operator when specific benchmarks are met and can receive an incentive fee on a wide variety of benchmarks such as an incentive fee for the operating cash flow. Lastly, owners can be subject

(12)

to other fees such as accounting fees and project management fees that are set up in a management contract. (HVS, 2014)

The reason as to why there has been a shift from the typical hospitality structure towards the ALFO structure can be attributed to many different factors. First, it has been shown that a fee-income based business model has a lower variance and therefore, translates to lower changes in the revenues of fee-income based businesses (Sohn et al., 2013). Furthermore, Sohn et al. (2013) showed that when companies adopt an ALFO structure that increased firm value is created through stabilized earnings and increased operating incomes. On top of that, it had been found that one of the main advantages of such a sale and manage-back structure is the reduced risk for an operator and reducing the risk also reduces fluctuations in the profitability and creates more stable earnings. (Page, 2007) Another reason identified by Page (2007) is the demand for owners of hotels to shift their capital from real estate properties through the sale of these properties and return capital to shareholders and increase the return on capital of these companies. Moreover, Sarheim (2006) found that enterprises with a higher degree of contracts within their firm structure obtained benefits, for example; higher revenues per room, faster than REITs due to the fee

structure involved in this business model.

It has become clear that the ALFO structure within the hospitality industry has become a dominant strategy with more companies within the hotel industry taking the ALFO structure as its main goal. (Li & Singal, 2019) The pioneers of this strategy, Hilton and Marriott, have both pledged in their latest annual reports that they would bind to an ALFO structure and would rely on a fee-income based strategy to expand their business. (Marriott International, Inc., 2019; Li & Singal, 2019) It is also becoming increasingly clear that this strategy is also being adopted by other big hotel brands in the hospitality industry such as Hyatt. In 2017, Hyatt pledged to sell around $1.5 billion in real estate to move from their “asset-recycling” strategy to the asset-light strategy. (Ting, 2017) Only two years later, Hyatt pledged to sell another $1.5 billion in real estate assets to further focus on its asset-light structure within their business. (Ting, 2019) With current businesses with a significant share within the hospitality industry adopting this strategy, it is clear that the ALFO is the dominating structure for hotel companies to shift towards an

(13)

3. Hypothesis

In a study conducted by Kim et al. (2002) it was found that the Hotel REITs showed similar performance to the market portfolio. This finding was confirmed by Jackson (2008) when performing an assessment of lodging REITs relative to the broader market benchmarks and the overall REIT sector. However, both studies were conducted in a period prior to the expansion of the REIT market and in specific prior to the change within the industry to increasingly adopt an asset-light model in the past decade. The findings by Kim et al. (2002) and Jackson (2008) were in line with other findings by for example Chen and Peiser (1999) who found that REITs, and in specific equity REITs that make up the majority of the REIT market, underperformed stocks when looking at returns on a nominal basis and there was no proof for a better performance when accounting for risks involved.

On top of that, the majority of REITs take part in property development, an activity that significantly increases the systematic risk of REITs and this activity had been on the rise in the period preceding research performed by Brounen, Kanters & Eichholtz (1999). However, the percentage of property developing REITs in the lodging sector was low amongst other REIT sectors and could indicate lower risks for hotel REITs. Overall it has been mostly found that the lodging REITs are performing similar compared to the broader market, a finding that is further confirmed by Sarheim (2006) that examined the fluctuations and returns of the S&P 500 on the one hand, and hotel REITs and C-corporations on the other hand. Based on findings within previous research, it can be hypothesized that lodging REITs will have a relative performance to the market or a better performance. This can be stated as follows:

H1: Hotel REITs will perform similar to, or perform better than, the broader market during the period of January 2008 to December 2019.

When looking at the different REIT property sectors within the market, different findings arise. Kim et al. (2002) and Jackson (2008) had found that the hotel REITs underperformed compared to other REIT property sectors and there was found to be no evidence that they performed significantly better compared to the other REITs in the REIT equity market. The second hypothesis that will be tested can be described as follows:

(14)

H2: Hotel REITs will perform similar to, or perform better than, the other REIT property sectors during the period of January 2008 to December 2019.

4. Data

In this study, time-series data will be used to study the monthly REIT index returns. The monthly stock returns of specific REIT property sectors will be analysed for the period from January 2008 to December 2019. These data will be used to measure the performance of the hotel REITs compared to the other REIT property sectors within the industry. This because existing research does not cover the period after the global financial crisis of 2008. The data in the year of 2008 and the following years are important to assess the performance of the hotel REITs

compared to the other REIT property sectors and the broader market (S&P 500) both during the financial crisis and the decade following the financial crisis. To collect the data on the REIT property sectors, the data per property sector was collected from NAREIT with monthly returns per property sector, which included both distributions and price changes. To assess the

performance of the lodging REITs compared to other property sectors, only property sectors were included that contained observations for every month in the period from January 2008 to

December 2019. These property sectors were as follows: Lodging, Diversified, Health Care, Industrial, Office, Residential, Retail, and Self Storage. For these REIT property sectors, the monthly performance of the listed REITs in a certain property sector was collected.

Furthermore, the performance of lodging REITs needs to be compared to the broader market. To measure this, data has been collected from the NAREIT regarding the performance of the overall REIT market (both equity and mortgage REITs) and only equity REITs, the latter being the category lodging REITs are listed in. The data collected here is also a monthly return, including distributions and price changes, for both cumulative indices. In these measures for the total equity REIT return and the overall REIT return, it has to be taken into account that there might be REITs in the measurement period that exited the market later on or vice versa.

To assess the performance of lodging REITs, data also needs to be obtained from the broader market. For equivalents of the broader market, we will use the CRSP equal-weighted

(15)

index, CRSP value-weighted index and the largest stock exchange in the U.S., the S&P 500. These data files will be obtained through the comprehensive historical stock returns provided by the CRSP database. The value-weighted and index-weighted indices are comprised of data of the NYSE, AMEX and the NASDAQ to provide an index with stocks across different industries and stock indices. In addition, the monthly returns including distributions/dividends were used to assess the monthly performance of the lodging REITs compared to other REIT property sectors. A relevant equivalent for the risk-free rate will be the 90-day treasury bill monthly returns. This alternative for a risk-free rate has been used in many studies by Kim et al. (2002) and Jackson (2008). The monthly 90-day T-bill returns were also obtained through the CRSP database.

5. Methodology

To analyse the performance of a stock portfolio many indices can be used. In the pursuit of measuring the performance of a stock portfolio, financial researchers have generally used a model that is based on the CAPM model to assess the risks and returns of a portfolio. To assess the performance of the lodging REITs compared to the broader market and other equity REIT property sectors, there are three models to use: the Sharp index, the Treynor index and the Jensen’s index. The Sharp index and Treynor index are indices that are both based on the return compared to the risk. A difference between the Sharp index and Treynor index on the one hand and the Jensen’s index on the other hand is that it tries to estimate and measure the performance of a portfolio based on the security market line. This is one of the most commonly used risk-adjusted performance measures to relevant established benchmarks. (Jackson, 2008) The alpha which can be obtained through the model is a measurement of whether the firm is outperforming a certain benchmark or not. The Jensen index can be described as follows:

(Ri,t – Rf,t) = αi + βi * (Rm,t + Rf,t) + εi,t

In the above-described model the following variables are used within the Jensen index: Ri,t being the return on portfolio i at given time t; Rf,t is the risk-free rate which is the 90-day T-bill in this paper; Rm,t is the return of the market portfolio; αi is the performance measure of the portfolio; βi is the systematic risk and lastly εi,t is a random error with an expected mean of zero. In this model, the αi and βi were estimated using a least-squares regression which will regress a

(16)

line that minimizes the squared errors in a model. Where αi will function as the measured excess return or loss on a given REIT and the βi can be considered the systematic risk on the model. Furthermore, as described in the data section, the 90-day treasury bill rate will be used as an equivalent to the risk-free rate. The 90-day T-bill rate will be used to find a return that is closest to the definition of the risk-free rate, being a return that is not subjected to yield curve risks, default risks and inflation risk. The first and last reason being minimized with a treasury bill of short term. Furthermore, the market portfolio used will be the S&P 500, CRSP Value-weighted index and CRSP Equal-weighted value index.

To test whether lodging REITs are outperforming a benchmark or

underperforming compared to a benchmark we need to test whether αi is significantly different

from zero. The βi in this model will be tested to see if the systematic risk of the portfolio is significantly different from 1 (the market portfolio).

But the above-described model solely looks on the systematic risk that is involved and therefore assumes that idiosyncratic risk is diversified away. This might not always hold true and for investors who are only investing in one type of REIT, it is interesting to look at the total risk, both idiosyncratic and systematic, of the REIT portfolio in a property sector. To obtain this, the Sharpe ratio can be used to compare the different REIT property sectors by looking at the excess return per unit of total risk. The Sharpe ratio can be described as follows:

Sharpe Ratio =

!"#!$_%"

Where rp consists of the mean return on the portfolio; rf is considered the risk-free rate for which the 90-day T-bill will be used as an equivalent and σp is the standard deviation of the portfolio. When this ratio is calculated, a clearer view can be obtained on the excess return per unit of total risk for every REIT property sector regardless of its performance to another broader benchmark. All raw data has been obtained through Excel files and converted into Stata to test the Jensen index and Sharpe ratios of the firms.

(17)

6. Results

To answer the first hypothesis, it is first most useful to look at the descriptive statistics of the different REIT property sectors and other benchmarks in Table 1. The hotel REIT sector had an average monthly return of 0.834% over the period from 2008 to 2019. When looking at the overall equity REIT market the average monthly return was 0.864% throughout this period, which was higher than the hotel REITs over the same period. When looking at the overall REIT market (equity and mortgage) the average monthly return was 0.869%, higher than both hotel REITs and the equity REIT market. When comparing the hotel REITs to the overall (equity) REIT market, the volatility of the hotel REITs is higher than the overall (equity) REIT market despite the lower average monthly return. The alternative used for the risk-free rate, the 90-day treasury bill rate, had the lowest volatility at 0.789% along with the lowest average monthly return at 0.056% amongst all categories as expected for the risk-free rate. Compared to the relevant market benchmarks the return on lodging REITs was higher among all three relevant market benchmarks tested. The S&P 500 (0.826%), the equal-weighted CRSP index (0.763%) and the value-weighted CRSP index (0.776%) all had lower average monthly returns than the hotel REITs at an average monthly return of 0.834%. Conversely, the relevant market

benchmarks had standard deviations that were all significantly lower than the standard deviation of hotel REITs at 4.298%, 5.399% and 4.503% respectively against a standard deviation of 10.372% for the hotel REIT sector.

Table 1: Summarized statistics of the different benchmarks and property sectors used.

Sector Mean Std. Dev. Min Max

Office REIT 0.718% 7.224% -31.796% 32.458%

Industrial REIT 1.034% 11.311% -56.188% 70.483%

Retail REIT 0.695% 7.911% -36.779% 43.516%

Residential REIT 1.204% 6.438% -26.656% 22.242%

Diversified REIT 0.718% 7.363% -31.435% 39.687%

Health Care REIT 0.976% 6.904% -25.480% 27.730%

Lodging REIT 0.834% 10.372% -33.433% 67.525%

Self Storage REIT 1.378% 6.339% -22.244% 21.928%

All equity REITs 0.864% 6.754% -31.668% 31.020%

(18)

Equal-weighted CRSP index 0.763% 5.399% -20.600% 20.416% Value-weighted CRSP index 0.776% 4.503% -18.467% 11.533% S&P 500 0.826% 4.298% -16.700% 10.901% 90-day T-bill 0.056% 0.079% -0.013% 0.501%

Furthermore, the hotel REIT sector as a whole had one of the highest standard deviations, and thus volatility, of the REIT property sectors only after the Industrial REIT sector with a standard deviation of 11.311%. This on the contrary to the average monthly return of hotel REITs which was not amongst the highest in the different REIT property sectors and fell short of the return provided by taking a position into the equity REIT market as a whole. Of all REIT

property sectors, the self storage REITs provided the highest monthly return on average with one of the smallest standard deviations. It is interesting to note that all REIT property sectors had standard deviations ranging from 6.326% to 11.311% that were all higher than the relevant market benchmarks.

The descriptive statistics provided in Table 1 only account for the non-risk adjusted performances of the different market benchmarks and REIT property sectors. To test whether there is a significant difference in the performance of one sector compared to a market

benchmark it should be tested. As previously described in the methodology, the Jensen index will be used and the α and β will be estimated using least-squares regression. The α estimated will show whether a firm is outperforming (positive α) or underperforming (negative α) relative to a market benchmark and whether this finding is significant. The estimated β will show whether the firm is riskier (β > 1), less risky (β < 1) or carries equal risk (β = 1) compared to the market portfolio.

(19)

Table 2: Performance of REITs using Jensen index with the S&P 500.

When looking at the performance of the different REIT property sectors and the overall (equity) REIT market, there are significant differences to note. When looking at the significant outperformance or underperformance compared to the market, the only REIT property sector that significantly outperformed the market was the Self Storage REIT sector. All other REIT property sectors, compared to the S&P 500, did not significantly outperform or underperform compared to the S&P 500. From all REIT property sectors, the hotel REIT sector was the sector with the lowest α-coefficient pointing out a weaker performance. However, this finding was not significant enough to determine that the hotel REITs underperformed when compared to the market. Conversely, all β-coefficients were found to be significant and ranged from least risky compared to the market portfolio for the Self Storage REITs to riskiest compared to the market portfolio for the hotel REITs. The hotel REITs were found to have the highest risk compared to the market portfolio, but the lowest performance when compared to the market portfolio. The reverse is true for the Self Storage REITs.

Sector α α (t-test) β β (t-test) R2-adjusted

Lodging REIT -0.00641 -1.13 1.842 14.17*** 0.5829 Office REIT -0.00289 -0.69 1.235 12.98*** 0.5392 Industrial REIT -0.00353 -0.49 1.728 10.43*** 0.4297 Retail REIT -0.00294 -0.58 1.217 10.48*** 0.4322 Residential REIT 0.00440 1.02 0.919 9.31*** 0.3747 Diversified REIT -0.00256 -0.57 1.193 11.63*** 0.4841

Health Care REIT 0.00321 0.63 0.778 6.63*** 0.2309

Self Storage REIT 0.00783 1.65* 0.700 6.45*** 0.2213 All equity REITs -0.00057 -0.14 1.123 12.24*** 0.5101

All REITs -0.00018 -0.05 1.080 12.95*** 0.5382

(20)

Table 3: Performance of REITs using Jensen index with CRSP value-weighted index.

Sector α α (t-test) β β (t-test) R2-adjusted

Health Care REIT 0.00377 0.74 0.754 6.76*** 0.2383

Self Storage REIT 0.00838 1.78* 0.672 6.51*** 0.2242 All equity REITs 0.00022 0.06 1.092 12.74*** 0.5299

All REITs 0.00058 0.16 1.050 13.47*** 0.5578

*,**,*** Indicates the significance at the 10%, 5% and 1% level respectively.

In this case, when using different market indices, the CRSP equal- and value-weighted indices, as a market portfolio the results show great similarities. When estimating the α and β using the CRSP value-weighted index in table 3 the hotel REITs property sector has the lowest α of all REIT property sector and thus the lowest performance. This is also true when estimating the α with the CRSP equal-weighted index in table 4. However, both findings were not

significant to conclude that the hotel REITs underperformed when compared to the relevant benchmarks. The Self Storage REIT property sector was the sole sector that significantly outperformed the relevant market benchmarks and bears the lowest risk compared to the market portfolio with the lowest β of all tested REIT property sectors. Nevertheless, the estimated β for hotel REITs when comparing to all three benchmarks was significantly higher than 1 and for the hotel REITs, the reverse is true. The estimated β for hotel REITs was, in all three market

benchmarks used, the highest of all REIT property sectors and the overall (equity) REIT market. These findings indicate that there is insufficient evidence to state that the hotel REITs

significantly outperformed or underperformed compared to the market, but is significantly riskier than the market portfolio compared to.

(21)

Table 4: Performance of REITs using Jensen index with CRSP equal-weighted index.

Sector α α (t-test) β β (t-test) R2-adjusted

Health Care REIT 0.00488 0.95 0.610 6.50*** 0.2238

Self Storage REIT 0.00959 2.00** 0.513 5.81*** 0.1865 All equity REITs 0.00168 0.43 0.905 12.54*** 0.5220

All REITs 0.00202 0.56 0.865 13.10*** 0.5441

*,**,*** Indicates the significance at the 10%, 5% and 1% level respectively.

When addressing the second hypothesis, the previously discussed findings are relevant to address this hypothesis. By looking at the α coefficients, the over- or underperformance

compared to the relevant market benchmarks can be determined. As previously discussed, the hotel REITs were found to have the lowest α coefficient when comparing to every market benchmark used in this paper. The coefficient was found to be insignificant and an

underperformance compared to the relevant benchmarks cannot be significantly concluded. The only REIT property sector that significantly outperformed the relevant benchmarks was the Self Storage sector. Nevertheless, the hotel REITs had the highest β coefficient of all REIT property sectors meaning that the hotel REITs carried the most risk compared to the market portfolio of all REIT property sectors. There is insufficient evidence to conclude that the hotel REITs

outperformed other REIT property sectors. To further elaborate on this, the Sharpe ratios were calculated for all REIT property sectors and the overall (equity) REIT market in Table 5, sorted from low to high, below.

(22)

Table 5: Annualized Sharpe ratios of REIT property sectors

Sector Sharpe ratio

Lodging REIT 0.2595

Retail REIT 0.2799

Industrial REIT 0.2993

Diversified REIT 0.3118

Office REIT 0.3173

All equity REITs 0.4143

All REITs 0.4451

Health Care REIT 0.4614

Residential REIT 0.5023

Self Storage REIT 0.7223

The Sharpe ratio measures the excess return per total unit of risk and can be described as the reward to volatility. The Sharpe ratio looks at the total excess return compared to the risk-free rate along with the standard deviation (volatility) of the portfolio. The overall equity REIT

market was found to have a Sharpe ratio of 0.4143. An interesting finding is that the hotel REITs were found to have the lowest Sharpe ratio of all REIT property sectors at 0.2595 compared to the Self Storage REITs that were found to have the highest Sharpe ratio at 0.7223 meaning that the hotel REITs had the lowest excess return per unit of risk of all REIT property sectors. This finding is in line with the previously found α coefficients where the hotel REITs were found to have the lowest coefficients and the Self Storage REITs were the only REIT property sector that significantly outperformed the relevant market benchmarks. There is no significant evidence to conclude that the hotel REITs outperformed other REIT property sectors and the reverse is more likely true.

7. Conclusion

This paper examined the recent changes in the hospitality industry along with the history of hotel REITs and the equity REIT market as a whole. The REIT industry has become a viable and common investment for many investors and has grown over time in which the hotel REIT industry has taken its place. It has been shown that many recent innovations and changes took place when hotel REITs first gained popularity and momentum around 1993. After that, the hotel

(23)

REIT industry has experienced significant growth and the hospitality industry as a whole has undergone many changes. Large hotel operating brands have doubled down on their asset-light model in return for management contracts that have redefined the hospitality industry and moved away from the classic model within the hospitality industry.

As previously stated, the hotel industry has undergone many changes and has moved away from its traditional operations. On top of that, the hotel REIT industry has grown significantly over time and has taken a significant stance in the overall REIT market and its market capitalization has grown continuously after hotel REITs have gained popularity in 1993. Therefore, the main purpose of this paper is to examine the performance of the hotel REIT sector in the recent period of 2008 to 2019. It was found that there is insufficient evidence to state that hotel REITs outperformed relevant market benchmarks, the S&P 500 and CRSP equal- and value-weighted indices. The reverse might actually be the case, although the findings were not significant to state that the hotel REIT market underperformed compared to relative market benchmarks. However, whereas there is insufficient evidence to state that the hotel REITs outperformed the relevant market benchmarks, there is sufficient evidence to state that investing in hotel REITs increases the risk exposure of investors as the risk of hotel REITs was

significantly higher than the market portfolio. This also holds true when comparing the different REIT property sectors to hotel REITs when taking into account risk. The hotel REITs were found to be the riskiest property sector out of the REIT property sectors examined and showed no better risk-adjusted performance to other REIT property sectors and was actually found to have worse risk-adjusted performance than other REIT property sectors due to its significantly higher risk.

8. Discussion 8.1 Contribution

The first contribution this paper has made to existing literature and research is concerning the management structure within the hospitality industry. By analysing the impact of the REIT developments along with the changes in the hotel capital structure and management structures a new insight has been provided in the current developments surrounding the hospitality industry. This paper mainly showed the changes that pioneers of the asset-light model made to the

(24)

of the emerging hotel REIT industry in relation to the existing relationship between operators and investors with management contracts. Briefly, the shift from the traditional operations within the hospitality industry towards an asset-light model and the influence of REITs on this shift and the relationship between the REITs and hotel brands.

Moreover, this paper has contributed to existing research by providing insights into the performance of lodging REITs over a recent period which has not been examined yet. By examining the performance of lodging REITs to other property sectors and the overall market useful insights have been gained in the performance of lodging REITs in the period during the financial crisis of 2008 and the recovery period after the financial crisis upon till the current stance of the hotel REITs. Whereas this can be argued as a limitation of the research and need to further study the effects of such a significant event within the measurement period, it also allows us to gain insight regarding the performance of hotel REITs over a prolonged time with

significant macro-economic events within the measurement period. In short, the research in this paper allows to examine the performance of hotel REITs over a period with a significant macro-economic event and assess the performance of hotel REITs compared to other REIT property sectors during the same period. Furthermore, this paper has allowed insights into the relative performance of lodging REITs to other REIT property sectors. This by not only assessing the performance of the lodging REITs on a risk-adjusted basis where the risk is solely considered systematic risk in the Jensen index, but also by looking at the Sharpe ratios and looking at the performance when accounting for both idiosyncratic risk and systematic risk mimicking an investor who solely invests in only one REIT property sector.

8.2 Limitations

In this paper, there were significant limitations to the assessment of the performance of the hotel REITs. First of all, the only source of data collection used for the performance of hotel REITs and all other REIT property sectors was retrieved from NAREIT which did not provide any individual returns on the REITs listed within a certain property sector. Therefore, the results do not account for firms that were only present in the REIT sector from the beginning of the measurement period, January 2008, until the end of the measurement period, December 2019. This could have resulted in distorted results because the underlying composition of the sectors

(25)

has changed in this period. This also holds true for the overall (equity) market where, on this larger scale, there is significantly more distortion because of certain REIT property sectors that have been discontinued during the measurement period or entered the REIT market during the measurement period. On top of that, the alternative used to assess the performance relative to the market portfolio was the S&P 500. A benchmark that is widely used as the market when

assessing the performance of REITs compared to relevant market benchmarks (Han and Liang, 1995). However, using the S&P 500 does not come without its limitations. The S&P 500 is a stock market that does not include any medium-sized enterprises. When examining the

performance of small firms, it was found that small firms performed differently than the larger firms listed in the S&P 500 and outperformed the S&P 500. An issue that can distort the view of how REITs perform against the market when solely taking into account larger firms listed on major stock exchanges (Han and Liang, 1995).

Furthermore, the measurements started at the beginning of the financial crisis which could be a limitation to the accuracy of the average monthly returns by including such a significant event. On the other hand, the purpose was to focus on both the impact of such a significant and the decade following this event to measure the recovery. However, as Payne and Waters (2007) have found in their paper, is the hotel REIT sector as a REIT property subsector the only subsector that showed persistent behaviour that was consistent with collapsing bubbles during their measurement period of 1994 to 2005. This may have influenced the position of the lodging REIT returns negatively compared to the other REIT property sectors and the overall REIT market.

Moreover, this paper focused mainly on the current performance of the REIT property sectors and did not take into account the capital structures of listed REITs within a certain property sector. This could have explained certain factors such as higher volatility and better or worse performance. On top of that, aside from the financial crisis, we did not take into account any sector-specific events that may have affected the performance of certain property sectors. The performance measured in this paper were only for one sector, the Self storage sector, found to be significant to find an outperforming sector relative to the market benchmarks used in this paper. This limited the ability in this paper to draw significant conclusions about the performance

(26)

of lodging REITs, something that has also been found by Kim et al. (2002) and Jackson (2008) regarding the insignificance of findings on the performance of the REIT property sectors.

8.3 Recommendations

There is a contribution by this paper on the one hand, and limitations within this paper that allow for recommendations on future research. Whereas this paper mainly focused on a risk-adjusted basis on both systematic and idiosyncratic risk, future research could elaborate on this by assessing the performance of hotel REITs on a volatility basis rather than a sole systematic basis as examined in this paper. By doing so, further insights will be gained in the situation for investors in lodging REITs who focus their investments in the hospitality industry and do not diversify their portfolios. Moreover, future research could focus on the capital structures of different REIT property sectors and explain why the lodging REITs showed the performance examined in this paper over the course of the examined period.

Furthermore, the findings in this paper regarding the performance were limited. Future research could focus on using other models, such as the Treynor model, to assess the performance of lodging REITs and other REIT property sectors by gaining new insights into the performance by using different methods. The findings in this paper relative to the market benchmarks were found to be mostly insignificant and approaching the topic via a different method could gain new insights.

Lastly, one of the limitations mainly concerns the fact that the use of the S&P 500 as a market benchmark can create a distorted view of the performance of REITs relative to the market. The solution to this problem has been stated as unsolved by Han and Liang (1995), but there might be an additional assessment that can be performed to assess the performance of REITs. Future research could focus on the performance of REITs compared to both a market portfolio consisting of large listed companies (S&P 500, NASDAQ or AMEX) and a market portfolio consisting of smaller listed companies on small-cap indices in the U.S. for example. A performance assessment relative to larger market indices and small-cap indices could illustrate a clearer view of the performance of hotel REITs relative to the broader market.

(27)

References

Beals, P., & Arabia, J. V. (1998). Lodging REITs: Ready for an encore. Cornell Hotel and Restaurant Administration Quarterly, 39(6), 52-59.

Blal, I. & Bianchi, G. (2019). The asset light model: A blind spot in hospitality research. International Journal of Hospitality Management, 76: 39-42.

Block, R. L. (1998). Investing in REITs: Real estate investment trusts. Princeton, NJ: Bloomberg Press.

BNP Paribas S.A. (2020) A perspective on property in a global pandemic. Retrieved 25th of April, 2020 from https://investors-corner.bnpparibas-am.com/investment-themes/a-perspective-on-property-in-a-global-pandemic/

Brounen, D., Kanters, P. M., & Eichholtz, P. M. A. (1999). The effects of property development activities on the performance of REITs. Real Estate Finance, 16(4), 45-52.

Buyhan, R., Kuhle, J., Mohammed Al-Deehani, T., Mahmood, M. (2015). “Portfolio

Diversification Benefits Using Real Estate Investment”. International Journal of Economics and Financial Issues, 5(4), 922-928.

Chan, S. H., Erickson, J., & Wang, K. (2003). Real estate investment trusts: Structure, performance, and investment opportunities (1st ed.). New York: Oxford University Press. Chen, J., & Peiser, R. (1999). The risk and return characteristics of REITs. Real Estate Finance, 16, 61–68

Gu, Z., & Kim, H. (2003). An examination of the determinants of hotel REITs’ unsystematic risk. Journal of Hospitality & Tourism Research, 27(2): 166-184.

Detlefsen, H. & Glodz, M. (2013). Historical Trends Hotel Management Contracts. HVS. Retrieved from: https://www.hospitalitynet.org/file/152005027.pdf

(28)

Han, J. & Liang, Y. (1995) The Historical Performance of Real Estate Investment Trusts, Journal of Real Estate Research, American Real Estate Society, vol. 10(3), pages 235-262.

Jackson, L.A. (2008). The structure and performance of US hotel real estate investment trusts. Journal of Retail & Leisure Property, 7(4): 275-290.

Jackson, L.A. (2009). Lodging REIT performance and comparison with other equity REIT returns. International Journal of Hospitality & Tourism Administration, 10(4): 296-325.

Kim, H., Gu, Z., & Mattila, A.S. (2002). Hotel real estate investment trusts’ risk features and beta determinants. Journal of Hospitality & Tourism Research, 26(2): 138-154.

Kim, H., Gu, Z. & Mattila, A.S. (2002). Performance of hotel real estate investment trusts: a comparative analysis of Jensen indexes. Hospitality Management, 21: 85-97.

Lee, S., & Upneja, A. (2007). Does Wall Street Truly Understand Valuation of Publicly Traded Lodging Stocks? Journal of Hospitality & Tourism Research, 31(2), 168–181.

Li, Y. & Singal, M. (2019). Capital structure in the hospitality industry: The role of the asset- light and fee-oriented strategy. Tourism Management, 70: 124-133.

Marriot International, Inc. (2019). 2018 Annual Report. Retrieved 11th of May, 2020 from https://marriott.gcs-web.com/static-files/8799734e-b9e0-4e53-b194-7bd24a381118

National Association for Real Estate Investment Trusts, Inc. (2019) Monthly Index Constituents. Retrieved 4th of May, 2020 from https://www.reit.com/data-research/reit-indexes/monthly-index-constituents

National Association for Real Estate Investment Trusts, Inc. (2020) History of REITs. Retrieved 4th of May, 2020 from https://www.reit.com/what-reit/history-reits

(29)

National Association for Real Estate Investment Trusts, Inc. (2020) U.S. REIT Industry Equity Market Cap. Retrieved 4th of May, 2020 from https://www.reit.com/data-research/reit-market-data/us-reit-industry-equity-market-cap

Payne, J. E., & Waters, G. A. (2007). Have equity REITs experienced periodically collapsing bubbles? Journal of Real Estate Finance Economics, 34, 207–224.

PricewaterhouseCoopers, (2018). Emergence of Real Estate Investment Trust (REIT) in the Middle East. Retrieved 5th of May, 2020 from

https://www.pwc.com/m1/en/publications/emergence-real-estate-investment-trust.html

Sarheim, G. (2006). REIT or Wrong. Retrieved 26th of June, 2020 from: https://www.hospitalitynet.org/opinion/4027022.html

Semer, S. L. (2009). Brief history of U.S. reits. Canadian Tax Journal, 57(4), 960-971

Sohn, J., Tang, C., & Jang, S. (2013). Does the asset-light and fee-oriented strategy create value? International Journal of Hospitality Management, 32: 270-277.

Ting, D. (2017). Hyatt shifts strategy in plan to sell $1.5 billion in hotel real estate over next 3 years. Retrieved on 11th of May, 2020 from: https://skift.com/2017/11/02/hyatt-shifts-strategy-in-plan-to-sell-1-5-billion-in-hotel-real-estate-over-next-3-years/

Ting, D. (2019). Hyatt doubles down on its asset lighter approach. Retrieved on 11th of May, 2020 from: https://skift.com/2019/03/06/hyatt-doubles-down-on-its-asset-lighter-approach/

The history and performance of Hotel Real Estate Investment Trusts (REITs)